SEPA
United States
Environmental Protection
Agency
EPA 430-R-24-004
Inventory of
U.S. Greenhouse Gas
Emissions and Sinks
1990-2022
-------
Front cover photo credit for cow and digester: Vanguard Renewables.
-------
HOW TO OBTAIN COPIES
You can electronically download this document on the U.S. EPA's homepage at
https://www.epa.gov/ghgemissions/inventory-us-greenhouse-gas-emissions-and-sinks.
All data tables of this document for the full time series 1990 through 2022, inclusive, will be made available within
4-6 weeks after publication of the final report online at the link mentioned in the citation below.
RECOMMENDED CITATION
EPA (2024) Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022. U.S. Environmental Protection
Agency, EPA 430-R-24-004. https://www.epa.gov/ghgemissions/inventory-us-greenhouse-gas-emissions-and-
sinks-1990-2022.
FOR FURTHER INFORMATION
Contact Ms. Mausami Desai, Environmental Protection Agency, (202) 304-8932, desai.mausami@epa.gov,
or Mr. Vincent Camobreco, Environmental Protection Agency, (202) 617-6678, camobreco.vincent@epa.gov.
For more information regarding climate change and greenhouse gas emissions, see the EPA web site at
https://www.epa.gov/ghgemissions.
-------
Acknowledgments
The Environmental Protection Agency would like to acknowledge the many individual and organizational
contributors to this document, without whose efforts this report would not be complete. Although the complete
list of researchers, government employees, and consultants who have provided technical and editorial support is
too long to list here, EPA would like to thank some key contributors and reviewers whose work has significantly
improved this year's report.
Within EPA's Office of Atmospheric Protection (OAP), development and compilation of emissions from fuel
combustion was led by Vincent Camobreco. Sarah Roberts (EPA Office of Transportation and Air Quality (OTAQ))
directed the work to compile estimates of emissions from mobile sources. Work on fugitive methane emissions
from the Energy sector was directed by Julie Powers, Melissa Weitz and Chris Sherry. Development and
compilation of emissions estimates for the Waste sector were led by Lauren Aepli and Mausami Desai. John Steller
and Kenna Rewcastle directed work to compile estimates for the Agriculture and the Land Use, Land-Use Change,
and Forestry chapters with support from Jake Beaulieu and Alex Hall (EPA Office of Research and Development
(ORD) on compiling the inventories for CO2 and Cm associated with flooded lands. Development and compilation
of Industrial Processes and Product Use (IPPU) CO2, Cm, and N2O emissions was directed by Amanda Chiu and
Vincent Camobreco. Development and compilation of emissions of HFCs, PFCs, SF6, and NF3 from the IPPU sector
was directed by Deborah Ottinger, Dave Godwin, and Stephanie Bogle. Cross-cutting work was directed by
Mausami Desai. We thank Bill Irving for general advice, guidance, and cross-cutting review.
Other EPA offices and programs also contributed data, analysis, and technical review for this report. OAP's
Greenhouse Gas Reporting Program staff facilitated aggregation and review of facility-level data for use in the
Inventory, in particular aggregation of confidential business information data. The Office of Air Quality Planning
and Standards (OAQPS) with contributions from OTAQ provided analysis for precursor estimates and review for
several of the source categories (i.e., natural gas and petroleum systems) included in this report. ORD conducted
field research and developed estimates associated with flooded lands. The Office of Land and Emergency
Management (OLEM) also contributed analysis and research.
The Energy Information Administration (EIA) and the Department of Energy (DOE) contributed invaluable data and
analysis on numerous energy-related topics. William Sanchez and Kevin Nakolan at EIA provided annual energy
data that are used in fossil fuel combustion estimates. We also thank Chris Tremper, Soudeh Motamedi, and
Ashley Ruocco at the Department of Energy for providing data and information on emissions of SF6 and PFCs from
Other Product Use. Other government agencies have contributed data as well, including the U.S. Geological Survey
(USGS), the Federal Highway Administration (FHWA), the Department of Transportation (DOT), the Bureau of
Transportation Statistics (BTS), the Department of Commerce, the Mine Safety and Health Administration (MSHA),
and the National Agricultural Statistics Service (NASS).
We thank the Department of Defense (DOD) (David Asiello, DoD and Matthew Cleaver of Leidos) for compiling the
data on military bunker fuel use.
We thank the Federal Aviation Administration (FAA) (Ralph Lovinelli and Jeetendra Upadhyay) for compiling the
inventory of emissions from commercial aircraft jet fuel consumption.
-------
We thank the United States Department of Agriculture's Office of the Chief Economist (USDA-OCE) (Meg Xiarchos)
and Economic Research Service (USDA-ERS) (Jeffrey Hopkins) for providing data on agricultural energy use.
We thank the U.S. Forest Service (USFS) (Grant Domke, Brian Walters, James Smith, and Courtney Giebink) for
compiling the inventories for CO2, Cm, and N2O fluxes associated with forest land.
We thank the Department of Agriculture's Agricultural Research Service (USDA-ARS) (Stephen Del Grosso) and the
Natural Resource Ecology Laboratory and Department of Statistics at Colorado State University (CSU) (Stephen
Ogle, Bill Parton, Shannon Spencer, Alisa Keyser, Lauren Hoskovec, Ram Gurung, Ryan Scheiderer, Veronica
Thompson, Stephen Williams, and Guhan Dheenadayalan Sivakami) for compiling the inventories for CH4
emissions, N2O emissions, and CO2 fluxes associated with soils in croplands, grasslands and settlements.
We thank the National Oceanic and Atmospheric Administration (NOAA) (Nate Herold, Ben DeAngelo and
Meredith Muth), Silvestrum Climate Associates (Stephen Crooks, Lisa Schile Beers, Rebeca Brenes), the
Smithsonian Environmental Research Center (J. Patrick Megonigal, James Holmquist, Jaxine Wolfe and Meng Lu),
and Florida International University (Tiffany Troxler) as well as members of the U.S. Coastal Wetland Carbon
Working Group for compiling inventories of land use change, soil carbon stocks and stock change, Cm emissions,
and N2O emissions from aquaculture in coastal wetlands. We also thank NOAA's Global Monitoring Lab (Stephen
Montzka and Lei Hu) for information on atmospheric measurements and derived emissions of HFCs and SF6.
We thank Marian Martin Van Pelt, Leslie Chinery, Alexander Lataille, Mollie Carroll, and the full Inventory team at
ICF including Diana Pape, Robert Lanza, Mollie Averyt, Larry O'Rourke, Ted Atwood, Skyler Brown, Deborah Harris,
Rebecca Ferenchiak, Fiona Wissell, Bikash Acharya, Sophie Johnson, Kyle Herdegen, Hazelle Tomlin, Lou Browning,
Johanna Garfinkel, Anna Cliche, Valerie Hammer, Mallory Giesie, David Landolfi, Emily Carr, Georgia Kerkezis,
Isabella Scornaienchi, Katie O'Malley, Maris Welch, Emily Adkins, Zeyu Hu, Alex Da Silva, Sneha Balakrishnan,
Kenny Yerardi, Leah Hartung, Molly Rickles, Seth Hartley and Ajo Rabemiarisoa for technical support in compiling
synthesis information across the report and preparing many of the individual analyses for specific report chapters
including fluorinated emissions and fuel combustion.
We thank Eastern Research Group for their analytical support. Deborah Bartram, Kara Edquist and Madison Eaton
support the development of emissions estimates for wastewater. Kara Edquist, Cortney Itle, Amber Allen, Spencer
Sauter, Sarah Wagner, and Madison Eaton support the development of emission estimates for manure
management, enteric fermentation, peatlands (included in wetlands remaining wetlands^, and landfilled yard
trimmings and food scraps (included in settlements remaining settlements). Brandon Long, Gopi Manne, and
Marty Wolf support the development of estimates for natural gas and petroleum systems. Gopi Manne and Tara
Stout support the development of emission estimates for coal mine methane.
Finally, we thank the RTI International team: Kate Bronstein, Emily Thompson, Jeff Coburn, and Keith Weitz for
their analytical support in development of the estimates of emissions from landfills; Melissa Icenhour, Michael
Laney, David Randall, Gabrielle Raymond, Karen Schaffner, Riley Vanek, Ricky Strott, Libby Robinson, Matt Hakos,
and Jeremy Kaelin for their analytical support in development of IPPU CO2, CH4, and N2O emissions; Karen
Schaffner and Haley Key for their analytical support in the development of the estimates of emissions from
fluorochemical production; and Tiffany Moore and Matt Hakos for their analytical support on disaggregating
industrial sector fossil fuel combustion emissions.
-------
Preface
The United States Environmental Protection Agency (EPA) prepares the official U.S. Inventory of Greenhouse Gas
Emissions and Sinks to fulfill annual existing commitments under the Paris Agreement and the United Nations
Framework Convention on Climate Change (UNFCCC). National greenhouse gas (GHG) inventory reports are to be
submitted each year by April 15.
In an effort to engage the public and researchers across the country, the EPA has instituted an annual public
review and comment process for this document. The draft document was announced via Federal Register Notice
and published for a 30-day comment period on the EPA Greenhouse Gas Emissions and Removals web site from
February 14 through March 15, 2024, and comments received during the public review period are posted to the
docket EPA-HQ-QAR-2024-0004. Comments received after the closure of the public comment period will be
considered for the next edition of this annual report. Responses to comments are typically posted to EPA's website
2-4 weeks following publication of the final report in April 2024.
-------
Table of Contents
TABLE OF CONTENTS VI
LIST OF TABLES, FIGURES, BOXES, AND EQUATIONS IX
EXECUTIVE SUMMARY ES-1
ES.l Background Information ES-2
ES.2 Recent Trends in U.S. Greenhouse Gas Emissions and Sinks ES-4
ES.3 Overview of Sector Emissions and Trends ES-16
ES.4 Other Information ES-20
1. INTRODUCTION 1-1
1.1 Background Information 1-2
1.2 National Inventory Arrangements 1-11
1.3 Inventory Preparation Process 1-14
1.4 Methodology and Data Sources 1-18
1.5 Key Categories 1-18
1.6 Quality Assurance and Quality Control (QA/QC) 1-25
1.7 Uncertainty Analysis 1-29
1.8 Completeness 1-32
1.9 Organization of Report 1-33
2. TRENDS IN GREENHOUSE GAS EMISSIONS AND REMOVALS 2-1
2.1 Overview of U.S. Greenhouse Gas Emissions and Sinks Trends 2-1
2.2 Emissions and Sinks by Economic Sector 2-29
2.3 Precursor Greenhouse Gas Emissions (CO, NOx, NMVOCs, and SO2) 2-42
3. ENERGY 3-1
3.1 Fossil Fuel Combustion (CRT Source Category 1A) 3-6
3.2 Carbon Emitted from Non-Energy Uses of Fossil Fuels (CRT Source Category 1A) 3-50
3.3 Incineration of Waste (CRT Source Category 1A) 3-58
3.4 Coal Mining (CRT Source Category lBla) 3-62
vi Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
3.5 Abandoned Underground Coal Mines (CRT Source Category lBla) 3-69
3.6 Petroleum Systems (CRT Source Category lB2a) 3-74
3.7 Natural Gas Systems (CRT Source Category lB2b) 3-94
3.8 Abandoned Oil and Gas Wells (CRT Source Categories lB2a and lB2b) 3-117
3.9 International Bunker Fuels (CRT Source Category 1: Memo Items) 3-121
3.10 Biomass and Biofuels Consumption (CRT Source Category 1A) 3-127
3.11 Energy Sources of Precursor Greenhouse Gases 3-131
4. INDUSTRIAL PROCESSES AND PRODUCT USE 4-1
4.1 Cement Production (CRT Source Category 2A1) 4-10
4.2 Lime Production (CRT Source Category 2A2) 4-15
4.3 Glass Production (CRT Source Category 2A3) 4-21
4.4 Other Process Uses of Carbonates (CRT Source Category 2A4) 4-25
4.5 Ammonia Production (CRT Source Category 2B1) 4-32
4.6 Urea Consumption for Non-Agricultural Purposes (CRT Source Category 2B10) 4-37
4.7 Nitric Acid Production (CRT Source Category 2B2) 4-41
4.8 Adipic Acid Production (CRT Source Category 2B3) 4-45
4.9 Caprolactam, Glyoxal and Glyoxylic Acid Production (CRT Source Category 2B4) 4-49
4.10 Carbide Production and Consumption (CRT Source Category 2B5 & 2B10) 4-53
4.11 Titanium Dioxide Production (CRT Source Category 2B6) 4-57
4.12 Soda Ash Production (CRT Source Category 2B7) 4-61
4.13 Petrochemical Production (CRT Source Category 2B8) 4-64
4.14 HCFC-22 Production (CRT Source Category 2B9a) 4-73
4.15 Production of Fluorochemicals Other Than HCFC-22 (CRT Source Category 2B9b) 4-76
4.16 Carbon Dioxide Consumption (CRT Source Category 2B10) 4-96
4.17 Phosphoric Acid Production (CRT Source Category 2B10) 4-100
4.18 Iron and Steel Production (CRT Source Category 2C1) and Metallurgical Coke Production 4-104
4.19 Ferroalloy Production (CRT Source Category 2C2) 4-116
4.20 Aluminum Production (CRT Source Category 2C3) 4-121
4.21 Magnesium Production and Processing (CRT Source Category 2C4) 4-127
4.22 Lead Production (CRT Source Category 2C5) 4-133
4.23 Zinc Production (CRT Source Category 2C6) 4-137
4.24 Electronics Industry (CRT Source Category 2E) 4-143
4.25 Substitution of Ozone Depleting Substances (CRT Source Category 2F) 4-161
4.26 Electrical Equipment (CRT Source Category 2G1) 4-168
4.27 SFs and PFCs from Other Product Use (CRT Source Category 2G.2) 4-178
vii
-------
4.28 Nitrous Oxide from Product Uses (CRT Source Category 2G3) 4-183
4.29 Industrial Processes and Product Use Sources of Precursor Gases 4-186
5. AGRICULTURE 5-1
5.1 Enteric Fermentation (CRT Source Category 3A) 5-4
5.2 Manure Management (CRT Source Category 3B) 5-11
5.3 Rice Cultivation (CRT Source Category 3C) 5-21
5.4 Agricultural Soil Management (CRT Source Category 3D) 5-28
5.5 Liming (CRT Source Category 3G) 5-47
5.6 Urea Fertilization (CRT Source Category 3H) 5-50
5.7 Field Burning of Agricultural Residues (CRT Source Category 3F) 5-53
6. LAND USE, LAND-USE CHANGE, AND FORESTRY 6-1
6.1 Representation of the U.S. Land Base 6-9
6.2 Forest Land Remaining Forest Land (CRT Category 4A1) 6-25
6.3 Land Converted to Forest Land (CRT Category 4A2) 6-53
6.4 Cropland Remaining Cropland (CRT Category 4B1) 6-61
6.5 Land Converted to Cropland (CRT Category 4B2) 6-74
6.6 Grassland Remaining Grassland (CRT Category 4C1) 6-81
6.7 Land Converted to Grassland (CRT Category 4C2) 6-93
6.8 Wetlands Remaining Wetlands (CRT Category 4D1) 6-101
6.9 Land Converted to Wetlands (CRT Source Category 4D2) 6-144
6.10 Settlements Remaining Settlements (CRT Category 4E1) 6-166
6.11 Land Converted to Settlements (CRT Category 4E2) 6-187
6.12 Other Land Remaining Other Land (CRT Category 4F1) 6-194
6.13 Land Converted to Other Land (CRT Category 4F2) 6-194
7. WASTE 7-1
7.1 Landfills (CRT Source Category 5A1) 7-4
7.2 Wastewater Treatment and Discharge (CRT Source Category 5D) 7-19
7.3 Composting (CRT Source Category 5B1) 7-54
7.4 Anaerobic Digestion at Biogas Facilities (CRT Source Category 5B2) 7-58
7.5 Waste Incineration (CRT Source Category 5C1) 7-64
7.6 Waste Sources of Precursor Greenhouse Gases 7-65
8. OTHER 8-1
9. RECALCULATIONS AND IMPROVEMENTS 9-1
10. REFERENCES AND ABBREVIATIONS 10-1
Executive Summary 10-1
Introduction 10-2
viii Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
Trends in Greenhouse Gas Emissions 10-3
Energy 10-4
Industrial Processes and Product Use 10-24
Agriculture 10-52
Land Use, Land-Use Change, and Forestry 10-72
Waste 10-109
Recalculations and Improvements 10-120
Abbreviations 10-121
List of Tables, Figures, Boxes, and
Equations
Tables
Table ES-1: Global Warming Potentials (100-Year Time Horizon) Used in this Report ES-3
Table ES-2: Recent Trends in U.S. Greenhouse Gas Emissions and Sinks (MMT CO2 Eq.) ES-5
Table ES-3: Recent Trends in U.S. Greenhouse Gas Emissions and Sinks by UNFCCC/IPCC Sector (MMT CO2 Eq.)....
ES-16
Table ES-4: U.S. Greenhouse Gas Emissions and Removals (Net Flux) from Land Use, Land-Use Change, and
Forestry (MMT C02 Eq.) ES-19
Table ES-5: U.S. Greenhouse Gas Emissions Allocated to Economic Sectors (MMT CO2 Eq.) ES-21
Table ES-6: U.S. Greenhouse Gas Emissions with Electricity-Related Emissions Distributed by Economic Sector
(MMT CO2 Eq.) ES-22
Table ES-7: Recent Trends in Various U.S. Data (Index 1990 = 100) ES-24
Table 1-1: Global Atmospheric Concentration, Rate of Concentration Change, and Atmospheric Lifetime of
Selected Greenhouse Gases 1-4
Table 1-2: Global Warming Potentials and Atmospheric Lifetimes (Years) Used in this Report 1-9
Table 1-3: Comparison of 100-Year GWP values 1-11
Table 1-4: Summary of Key Categories for the United States (1990 and 2022) by Sector 1-20
Table 1-5: Estimated Overall Inventory Quantitative Uncertainty for 1990 (MMT CO2 Eq. and Percent) 1-30
ix
-------
Table 1-6: Estimated Overall Inventory Quantitative Uncertainty for 2022 (MMT CO2 Eq. and Percent) 1-30
Table 1-7: Quantitative Assessment of Trend Uncertainty (MMT CO2 Eq. and Percent) 1-32
Table 1-8: CRT/IPCC Sector Descriptions 1-33
Table 1-9: List of Annexes 1-34
Table 2-1: Recent Trends in U.S. Greenhouse Gas Emissions and Sinks by Gas (MMT CO2 Eq.) 2-4
Table 2-2: Recent Trends in U.S. Greenhouse Gas Emissions and Sinks by Gas (kt) 2-6
Table 2-3: Recent Trends in U.S. Greenhouse Gas Emissions and Sinks by UNFCCC/IPCC Sector/Category (MMT CO2
Eq.) 2-9
Table 2-4: Emissions from Energy by Gas (MMT CO2 Eq.) 2-12
Table 2-5: CO2 Emissions from Fossil Fuel Combustion by End-Use Sector (MMT CO2 Eq.) 2-14
Table 2-6: Emissions from Industrial Processes and Product Use (MMT CO2 Eq.) 2-19
Table 2-7: Emissions from Agriculture (MMT CO2 Eq.) 2-23
Table 2-8: U.S. Greenhouse Gas Emissions and Removals (Net Flux) from Land Use, Land-Use Change, and Forestry
(MMT CO2 Eq.) 2-25
Table 2-9: Emissions from Waste (MMT CO2 Eq.) 2-28
Table 2-10: U.S. Greenhouse Gas Emissions Allocated to Economic Sectors (MMT CO2 Eq. and Percent of Total in
2022) 2-30
Table 2-11: Electric Power-Related Greenhouse Gas Emissions (MMT CO2 Eq.) 2-34
Table 2-12: U.S. Greenhouse Gas Emissions by Economic Sector and Gas with Electricity-Related Emissions
Distributed (MMT CO2 Eq.) and Percent of Total in 2022 2-35
Table 2-13: Transportation-Related Greenhouse Gas Emissions (MMT CO2 Eq.) 2-38
Table 2-14: Recent Trends in Various U.S. Data (Index 1990 = 100) 2-41
Table 2-15: Emissions of NOx, CO, NMVOCs, and S02 (kt) 2-43
Table 3-1: CO2, CFU, and N2O Emissions from Energy (MMT CO2 Eq.) 3-3
Table 3-2: CO2, CFU, and N2O Emissions from Energy (kt) 3-4
Table 3-3: CO2, CFU, and N2O Emissions from Fossil Fuel Combustion (MMT CO2 Eq.) 3-7
Table 3-4: CO2, CFU, and N2O Emissions from Fossil Fuel Combustion (kt) 3-7
Table 3-5: CO2 Emissions from Fossil Fuel Combustion by Fuel Type and Sector (MMT CO2 Eq.) 3-8
Table 3-6: Annual Change in CO2 Emissions and Total 2022 CO2 Emissions from Fossil Fuel Combustion for Selected
Fuels and Sectors (MMT CO2 Eq. and Percent) 3-9
Table 3-7: CO2 Emissions from Stationary Fossil Fuel Combustion (MMT CO2 Eq.) 3-13
Table 3-8: CFU Emissions from Stationary Combustion (MMT CO2 Eq.) 3-14
Table 3-9: N2O Emissions from Stationary Combustion (MMT CO2 Eq.) 3-14
Table 3-10: CO2, CFU, and N2O Emissions from Fossil Fuel Combustion by Sector (MMT CO2 Eq.) 3-15
Table 3-11: CO2, CFU, and N2O Emissions from Fossil Fuel Combustion by End-Use Sector with Electricity Emissions
Distributed (MMT C02 Eq.) 3-16
Table 3-12: Electric Power Generation by Fuel Type (Percent) 3-18
x Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
Table 3-13: CO2 Emissions from Fossil Fuel Combustion in Transportation End-Use Sector (MMT CO2 Eq.) 3-27
Table 3-14: CFU Emissions from Mobile Combustion (MMT CO2 Eq.) 3-30
Table 3-15: N2O Emissions from Mobile Combustion (MMT CO2 Eq.) 3-31
Table 3-16: Carbon Intensity from Direct Fossil Fuel Combustion by Sector (MMT CO2 Eq./QBtu) 3-35
Table 3-17: Approach 2 Quantitative Uncertainty Estimates for CO2 Emissions from Energy-Related Fossil Fuel
Combustion by Fuel Type and Sector (MMT CO2 Eq. and Percent) 3-38
Table 3-18: Comparison of Electric Power Sector Emissions (MMT CO2 Eq. and Percent) 3-39
Table 3-19: Comparison of Emissions Factors (MMT Carbon/QBtu) 3-40
Table 3-20: Approach 2 Quantitative Uncertainty Estimates for CFU and N2O Emissions from Energy-Related
Stationary Combustion, Including Biomass (MMT CO2 Eq. and Percent) 3-44
Table 3-21: Approach 2 Quantitative Uncertainty Estimates for CFU and N2O Emissions from Mobile Sources (MMT
CO2 Eq. and Percent) 3-48
Table 3-22: CO2 Emissions from Non-Energy Use Fossil Fuel Consumption (MMT CO2 Eq. and Percent C) 3-51
Table 3-23: Adjusted Consumption of Fossil Fuels for Non-Energy Uses (TBtu) 3-52
Table 3-24: 2022 Adjusted Non-Energy Use Fossil Fuel Consumption, Storage, and Emissions 3-52
Table 3-25: Approach 2 Quantitative Uncertainty Estimates for CO2 Emissions from Non-Energy Uses of Fossil Fuels
(MMT CO2 Eq. and Percent) 3-55
Table 3-26: Approach 2 Quantitative Uncertainty Estimates for Storage Factors of Non-Energy Uses of Fossil Fuels
(Percent) 3-55
Table 3-27: CO2, CFU, and N2O Emissions from the Combustion of Waste (MMT CO2 Eq.) 3-59
Table 3-28: CO2, CFU, and N2O Emissions from the Combustion of Waste (kt) 3-59
Table 3-29: Municipal Solid Waste Combusted (Short Tons) 3-60
Table 3-30: Calculated Fossil CO2 Content per Ton Waste Combusted (kg C02/Short Ton Combusted) 3-60
Table 3-31: CO2 Emissions from Combustion of Tires (MMT CO2 Eq.) 3-60
Table 3-32: Approach 2 Quantitative Uncertainty Estimates for CO2 and N2O from the Incineration of Waste (MMT
CO2 Eq. and Percent) 3-61
Table 3-33: Coal Production (kt) 3-62
Table 3-34: CFU Emissions from Coal Mining (MMT CO2 Eq.) 3-63
Table 3-35: CFU Emissions from Coal Mining (kt) 3-63
Table 3-36: CO2 Emissions from Coal Mining (MMT CO2 Eq.) 3-66
Table 3-37: CO2 Emissions from Coal Mining (kt) 3-66
Table 3-38: Approach 2 Quantitative Uncertainty Estimates for CFU and CO2 Emissions from Coal Mining (MMT CO2
Eq. and Percent) 3-68
Table 3-39: CFU Emissions from Abandoned Coal Mines (MMT CO2 Eq.) 3-70
Table 3-40: CFU Emissions from Abandoned Coal Mines (kt) 3-70
Table 3-41: Number of Gassy Abandoned Mines Present in U.S. Basins in 2022, Grouped by Class According to
Post-Abandonment State 3-72
xi
-------
Table 3-42: Approach 2 Quantitative Uncertainty Estimates for Cm Emissions from Abandoned Underground Coal
Mines (MMT CO2 Eq. and Percent) 3-74
Table 3-43: Total Greenhouse Gas Emissions (CO2, Cm, and N2O) from Petroleum Systems (MMT CO2 Eq.) 3-76
Table 3-44: Cm Emissions from Petroleum Systems (MMT CO2 Eq.) 3-76
Table 3-45: Cm Emissions from Petroleum Systems (kt CH4) 3-77
Table 3-46: CO2 Emissions from Petroleum Systems (MMT CO2) 3-77
Table 3-47: CO2 Emissions from Petroleum Systems (kt CO2) 3-77
Table 3-48: N2O Emissions from Petroleum Systems (Metric Tons CO2 Eq.) 3-77
Table 3-49: N2O Emissions from Petroleum Systems (Metric Tons N2O) 3-78
Table 3-50: Approach 2 Quantitative Uncertainty Estimates for Cm and CO2 Emissions from Petroleum Systems
(MMT CO2 Eq. and Percent) 3-81
Table 3-51: Recalculations of CO2 in Petroleum Systems (MMT CO2) 3-83
Table 3-52: Recalculations of CH4 in Petroleum Systems (MMT CO2 Eq.) 3-84
Table 3-53: HF Completions National Cm Emissions (Metric Tons CH4) 3-85
Table 3-54: HF Completions National CO2 Emissions (kt CO2) 3-85
Table 3-55: HF Workovers National CH4 Emissions (Metric Tons CH4) 3-86
Table 3-56: HF Workovers National CO2 Emissions (Metric Tons CO2) 3-86
Table 3-57: Pneumatic Controllers National CO2 Emissions (Metric Tons CO2) 3-87
Table 3-58: Pneumatic Controllers National CH4 Emissions (Metric Tons CH4) 3-87
Table 3-59: Equipment Leaks National CO2 Emissions (Metric Tons CO2) 3-88
Table 3-60: Chemical Injection Pump National CO2 Emissions (Metric Tons CO2) 3-88
Table 3-61: Storage Tanks National CO2 Emissions (kt CO2) 3-89
Table 3-62: Chemical Injection Pumps National CH4 Emissions (Metric Tons CH4) 3-89
Table 3-63: Produced Water National CH4 Emissions (Metric Tons CH4) 3-89
Table 3-64: Associated Gas Flaring National CH4 Emissions (Metric Tons CH4) 3-89
Table 3-65: Gas Engines National CH4 Emissions (Metric Tons CH4) 3-90
Table 3-66: Miscellaneous Production Flaring National CO2 Emissions (kt CH4) 3-90
Table 3-67: Offshore Production National CH4 Emissions (Metric Tons CH4) 3-90
Table 3-68: Refining National CH4 Emissions (Metric Tons CH4) 3-91
Table 3-69: Refining National CO2 Emissions (kt CO2) 3-91
Table 3-70: Quantity of CO2 Captured and Extracted for EOR Operations (kt CO2) 3-92
Table 3-71: Geologic Sequestration Information Reported Under GHGRP Subpart RR 3-92
Table 3-72: Total Greenhouse Gas Emissions (CH4, CO2, and N2O) from Natural Gas Systems (MMT CO2 Eq.) 3-97
Table 3-73: CH4 Emissions from Natural Gas Systems (MMT CO2 Eq.) 3-97
Table 3-74: CH4 Emissions from Natural Gas Systems (kt) 3-97
Table 3-75: CO2 Emissions from Natural Gas Systems (MMT) 3-98
xii Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
Table 3-76: CO2 Emissions from Natural Gas Systems (kt) 3-98
Table 3-77: N2O Emissions from Natural Gas Systems (Metric Tons CO2 Eq.) 3-98
Table 3-78: N2O Emissions from Natural Gas Systems (Metric Tons N2O) 3-98
Table 3-79: Approach 2 Quantitative Uncertainty Estimates for CH4 and Non-combustion CO2 Emissions from
Natural Gas Systems (MMT CO2 Eq. and Percent) 3-101
Table 3-80: Recalculations of CO2 in Natural Gas Systems (MMT CO2) 3-104
Table 3-81: Recalculations of CH4 in Natural Gas Systems (MMT CO2 Eq.) 3-104
Table 3-82: HF Completions National Cm Emissions (Metric Tons CH4) 3-105
Table 3-83: HF Completions National CO2 Emissions (kt CO2) 3-106
Table 3-84: Non-HF Completions National CH4 Emissions (Metric Tons CH4) 3-107
Table 3-85: Non-HF Completions National CO2 Emissions (Metric Tons CO2) 3-107
Table 3-86: HF Workovers National CH4 Emissions (Metric Tons CH4) 3-108
Table 3-87: HF Workovers National CChEmissions (kt CO2) 3-108
Table 3-88: Non-HF Workovers National CH4 Emissions (Metric Tons CH4) 3-109
Table 3-89: Non-HF Workovers National CO2 Emissions (Metric Tons CO2) 3-109
Table 3-90: Production Equipment Leaks National CO2 Emissions (Metric Tons CO2) 3-110
Table 3-91: Production Equipment Leaks National CH4 Emissions (Metric Tons CH4) 3-110
Table 3-92: Chemical Injection Pumps National CO2 Emissions (MetricTons CO2) 3-110
Table 3-93: Pneumatic Controllers National CO2 Emissions (Metric Tons CO2) 3-111
Table 3-94: Pneumatic Controllers National CH4 Emissions (MetricTons CH4) 3-111
Table 3-95: Storage Tanks National CH4 Emissions (Metric Tons CH4) 3-111
Table 3-96: Storage Tanks National CO2 Emissions (kt CO2) 3-112
Table 3-97: Liquids Unloading National CH4 Emissions (Metric Tons CH4) 3-112
Table 3-98: Production Gas Engines National CH4 Emissions (Metric Tons CH4) 3-112
Table 3-99: Miscellaneous Production Flaring National Emissions (kt CO2) 3-112
Table 3-100: Station Blowdowns National Emissions (Metric Tons CH4) 3-113
Table 3-101: Pneumatic Controllers National Emissions (Metric Tons CH4) 3-113
Table 3-102: Acid Gas Removal Units National Emissions (kt CO2) 3-113
Table 3-103: Yard Piping National Emissions (MetricTons CH4) 3-114
Table 3-104: Processing Segment Flares National CH4 Emissions (Metric Tons CH4) 3-114
Table 3-105: Processing Segment Flares National CO2 Emissions (kt CO2) 3-114
Table 3-106: Transmission Compressor Station National CH4 Emissions (Metric Tons CH4) 3-115
Table 3-107: Transmission Compressor Station National CO2 Emissions (Metric Tons CO2) 3-115
Table 3-108: Pipeline Venting National CH4 Emissions (Metric Tons CH4) 3-116
Table 3-109: CH4 Emissions from Abandoned Oil and Gas Wells (MMT CO2 Eq.) 3-118
Table 3-110: CH4 Emissions from Abandoned Oil and Gas Wells (kt) 3-118
xiii
-------
Table 3-111: CO2 Emissions from Abandoned Oil and Gas Wells (MMT CO2) 3-118
Table 3-112: CO2 Emissions from Abandoned Oil and Gas Wells (kt) 3-118
Table 3-113: Abandoned Oil Wells Activity Data, CH4 and CO2 Emissions (kt) 3-119
Table 3-114: Abandoned Gas Wells Activity Data, CH4 and CO2 Emissions (kt) 3-119
Table 3-115: Approach 2 Quantitative Uncertainty Estimates for CH4 and CO2 Emissions from Petroleum and
Natural Gas Systems (MMT CO2 Eq. and Percent) 3-120
Table 3-116: CO2, CH4, and N2O Emissions from International Bunker Fuels (MMT CO2 Eq.) 3-123
Table 3-117: CO2, CH4, and N2O Emissions from International Bunker Fuels (kt) 3-123
Table 3-118: Aviation Jet Fuel Consumption for International Transport (TBtu) 3-124
Table 3-119: Marine Fuel Consumption for International Transport (Million Gallons) 3-125
Table 3-120: CO2 Emissions from Wood Consumption by End-Use Sector (MMT CO2 Eq.) 3-127
Table 3-121: CO2 Emissions from Wood Consumption by End-Use Sector (kt) 3-127
Table 3-122: CO2 Emissions from Biogenic Components of MSW (MMT CO2 Eq.) 3-128
Table 3-123: CO2 Emissions from Biogenic Components of MSW (kt) 3-128
Table 3-124: CO2 Emissions from Ethanol Consumption (MMT CO2 Eq.) 3-128
Table 3-125: CO2 Emissions from Ethanol Consumption (kt) 3-128
Table 3-126: CO2 Emissions from Biodiesel Consumption (MMT CO2 Eq.) 3-129
Table 3-127: CO2 Emissions from Biodiesel Consumption (kt) 3-129
Table 3-128: Calculated Biogenic CO2 Content per Ton Waste (kg C02/Short Ton Combusted) 3-129
Table 3-129: Woody Biomass Consumption by Sector (Trillion Btu) 3-130
Table 3-130: Ethanol Consumption by Sector (Trillion Btu) 3-130
Table 3-131: Biodiesel Consumption by Sector (Trillion Btu) 3-130
Table 3-132: NOx, CO, NMVOC, and SO2 Emissions from Energy-Related Activities (kt) 3-132
Table 4-1: Emissions from Industrial Processes and Product Use (MMT CO2 Eq.) 4-4
Table 4-2: Emissions from Industrial Processes and Product Use (kt) 4-5
Table 4-3: CO2 Emissions from Cement Production (MMT CO2 Eq.) 4-11
Table 4-4: CO2 Emissions from Cement Production (kt CO2) 4-11
Table 4-5: Clinker Production (kt) 4-12
Table 4-6: Approach 2 Quantitative Uncertainty Estimates for CO2 Emissions from Cement Production (MMT CO2
Eq. and Percent) 4-13
Table 4-7: CO2 Emissions from Lime Production (MMT CO2 Eq.) 4-16
Table 4-8: Gross, Recovered, and Net CO2 Emissions from Lime Production (kt CO2) 4-16
Table 4-9: High-Calcium- and Dolomitic-Quicklime, High-Calcium- and Dolomitic-Hydrated, and Dead-Burned-
Dolomite Lime Production (kt) 4-18
Table 4-10: Adjusted Lime Production (kt) 4-18
Table 4-11: Approach 2 Quantitative Uncertainty Estimates for CO2 Emissions from Lime Production (MMT CO2 Eq.
and Percent) 4-19
xiv Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
Table 4-12: CO2 Emissions from Glass Production (MMT CO2 Eq.) 4-22
Table 4-13: CO2 Emissions from Glass Production (kt CO2) 4-22
Table 4-14: Limestone, Dolomite, Soda Ash, and Other Carbonates Used in Glass Production (kt) and Average
Annual Production Index for Glass and Glass Product Manufacturing 4-23
Table 4-15: Approach 2 Quantitative Uncertainty Estimates for CO2 Emissions from Glass Production (MMT CO2 Eq.
and Percent) 4-24
Table 4-16: CO2 Emissions from Other Process Uses of Carbonates (MMT CO2 Eq.) 4-27
Table 4-17: CO2 Emissions from Other Process Uses of Carbonates (kt CO2) 4-27
Table 4-18: Limestone and Dolomite Consumption from Other Uses of Carbonates (kt) 4-28
Table 4-19: Limestone and Dolomite Consumption from Ceramics Production (kt) 4-29
Table 4-20: Other Uses of Soda Ash Consumption Not Associated with Glass Manufacturing (kt) 4-29
Table 4-21: Magnesite and Limestone Consumption from Non-Metallurgical Magnesia Production (kt) 4-30
Table 4-22: Approach 2 Quantitative Uncertainty Estimates for CO2 Emissions from Other Process Uses of
Carbonates (MMT CO2 Eq. and Percent) 4-31
Table 4-23: CO2 Emissions from Ammonia Production (MMT CO2 Eq.) 4-33
Table 4-24: CO2 Emissions from Ammonia Production (kt CO2) 4-33
Table 4-25: Total Ammonia Production, Total Urea Production, and Recovered CO2 Consumed for Urea Production
(kt) 4-35
Table 4-26: Approach 2 Quantitative Uncertainty Estimates for CO2 Emissions from Ammonia Production (MMT
CO2 Eq. and Percent) 4-36
Table 4-27: CO2 Emissions from Urea Consumption for Non-Agricultural Purposes (MMT CO2 Eq.) 4-38
Table 4-28: CO2 Emissions from Urea Consumption for Non-Agricultural Purposes (kt CO2) 4-38
Table 4-29: Urea Production, Urea Applied as Fertilizer, Urea Imports, and Urea Exports (kt) 4-39
Table 4-30: Approach 2 Quantitative Uncertainty Estimates for CO2 Emissions from Urea Consumption for Non-
Agricultural Purposes (MMT CO2 Eq. and Percent) 4-40
Table 4-31: N2O Emissions from Nitric Acid Production (MMT CO2 Eq.) 4-42
Table 4-32: N2O Emissions from Nitric Acid Production (kt N2O) 4-42
Table 4-33: Nitric Acid Production (kt) 4-44
Table 4-34: Approach 2 Quantitative Uncertainty Estimates for N2O Emissions from Nitric Acid Production (MMT
CO2 Eq. and Percent) 4-44
Table 4-35: N2O Emissions from Adipic Acid Production (MMT CO2 Eq.) 4-46
Table 4-36: N2O Emissions from Adipic Acid Production (kt N2O) 4-46
Table 4-37: Adipic Acid Production (kt) 4-48
Table 4-38: Approach 2 Quantitative Uncertainty Estimates for N2O Emissions from Adipic Acid Production (MMT
CO2 Eq. and Percent) 4-49
Table 4-39: N2O Emissions from Caprolactam Production (MMT CO2 Eq.) 4-51
Table 4-40: N2O Emissions from Caprolactam Production (kt N2O) 4-51
Table 4-41: Caprolactam Production (kt) 4-52
xv
-------
Table 4-42: Approach 2 Quantitative Uncertainty Estimates for N2O Emissions from Caprolactam, Glyoxal and
Glyoxylic Acid Production (MMT CO2 Eq. and Percent) 4-53
Table 4-43: CO2 and Cm Emissions from Silicon Carbide Production and Consumption (MMT CO2 Eq.) 4-54
Table 4-44: CO2 and Cm Emissions from Silicon Carbide Production and Consumption (kt) 4-55
Table 4-45: Production and Consumption of Silicon Carbide (Metric Tons) 4-56
Table 4-46: Approach 2 Quantitative Uncertainty Estimates for Cm and CO2 Emissions from Silicon Carbide
Production and Consumption (MMT CO2 Eq. and Percent) 4-57
Table 4-47: CO2 Emissions from Titanium Dioxide (MMT CO2 Eq.) 4-58
Table 4-48: CO2 Emissions from Titanium Dioxide (kt CO2) 4-58
Table 4-49: Titanium Dioxide Production (kt) 4-59
Table 4-50: Approach 2 Quantitative Uncertainty Estimates for CO2 Emissions from Titanium Dioxide Production
(MMT CO2 Eq. and Percent) 4-60
Table 4-51: CO2 Emissions from Soda Ash Production (MMT CO2 Eq.) 4-62
Table 4-52: CO2 Emissions from Soda Ash Production (kt CO2) 4-62
Table 4-53: Trona Ore Used in Soda Ash Production (kt) 4-63
Table 4-54: Approach 2 Quantitative Uncertainty Estimates for CO2 Emissions from Soda Ash Production (MMT
CO2 Eq. and Percent) 4-63
Table 4-55: CO2 and Cm Emissions from Petrochemical Production (MMT CO2 Eq.) 4-65
Table 4-56: CO2 and Cm Emissions from Petrochemical Production (kt) 4-66
Table 4-57: Production of Selected Petrochemicals (kt) 4-69
Table 4-58: Approach 2 Quantitative Uncertainty Estimates for Cm Emissions from Petrochemical Production and
CO2 Emissions from Petrochemical Production (MMT CO2 Eq. and Percent) 4-70
Table 4-59: HFC-23 Emissions from HCFC-22 Production (MMT C02 Eq.) 4-74
Table 4-60: HFC-23 Emissions from HCFC-22 Production (kt HFC-23) 4-74
Table 4-61: HCFC-22 Production (kt) 4-75
Table 4-62: Approach 2 Quantitative Uncertainty Estimates for HFC-23 Emissions from HCFC-22 Production (MMT
CO2 Eq. and Percent) 4-75
Table 4-63: Emissions of HFCs, PFCs, SF6, and NF3 from Production of Fluorochemicals Other Than HCFC-22 (MMT
CO2 Eq.) 4-79
Table 4-64: Emissions of HFCs, PFCs, SF6, and NF3 from Production of Fluorochemicals Other Than HCFC-22 (Metric
Tons) 4-80
Table 4-65: Emissions of Other Fluorinated GHGs from Production of Fluorochemicals Other Than HCFC-22 (MMT
CO2 Eq.) 4-80
Table 4-66: Emissions of Other Fluorinated GHGs from Production of Fluorochemicals Other Than HCFC-22 (Metric
Tons) 4-80
Table 4-67: Production and Transformation of Fluorinated GHGs (kt)a 4-81
Table 4-68: Fluorinated GHG Groups Under Which Certain Emissions Are Reported Under Subpart L of the GHGRP
and Associated GWPs 4-83
xvi Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
Table 4-69: Destruction Efficiency Range Values Used to Estimate Pre-Abatement Emissions for Production and
Transformation Processes 4-85
Table 4-70: Estimated Starting Years for Emission Controls at Each Fluorinated Gas Production Facility Reporting
under Subpart L of the GHGRP 4-86
Table 4-71: List of Saturated HFCs, Unsaturated HFCs (Hydrofluoroolefins or HFOs), and Unsaturated HCFCs
(Hydrochlorofluoroolefins or HCFOs) whose 1990-2009 Production Was Estimated Using Vintaging Model, Virgin
Manufacturing by Chemical 4-88
Table 4-72: Approach 1 Quantitative Uncertainty Estimates for HFC, PFC, SF6, and NF3 from Production of
Fluorochemicals other than HCFC-22 (MMT CO2 Eq. and Percent) 4-95
Table 4-73: CO2 Emissions from CO2 Consumption (MMT CO2 Eq.) 4-97
Table 4-74: CO2 Emissions from CO2 Consumption (kt CO2) 4-97
Table 4-75: CO2 Production (kt CO2) and the Percent Used for Non-EOR Applications 4-98
Table 4-76: Approach 2 Quantitative Uncertainty Estimates for CO2 Emissions from CO2 Consumption (MMT CO2
Eq. and Percent) 4-99
Table 4-77: CO2 Emissions from Phosphoric Acid Production (MMT CO2 Eq.) 4-101
Table 4-78: CO2 Emissions from Phosphoric Acid Production (kt CO2) 4-101
Table 4-79: Phosphate Rock Domestic Consumption, Exports, and Imports (kt) 4-102
Table 4-80: Chemical Composition of Phosphate Rock (Percent by Weight) 4-102
Table 4-81: Approach 2 Quantitative Uncertainty Estimates for CO2 Emissions from Phosphoric Acid Production
(MMT CO2 Eq. and Percent) 4-103
Table 4-82: CO2 Emissions from Metallurgical Coke Production (MMT CO2 Eq.) 4-106
Table 4-83: CO2 Emissions from Metallurgical Coke Production (kt CO2) 4-106
Table 4-84: CO2 Emissions from Iron and Steel Production (MMT CO2 Eq.) 4-106
Table 4-85: CO2 Emissions from Iron and Steel Production (kt CO2) 4-107
Table 4-86: CFU Emissions from Iron and Steel Production (MMT CO2 Eq.) 4-107
Table 4-87: CFU Emissions from Iron and Steel Production (kt CH4) 4-107
Table 4-88: Material Carbon Contents for Metallurgical Coke Production 4-109
Table 4-89: Production and Consumption Data for the Calculation of CO2 Emissions from Metallurgical Coke
Production (Thousand Metric Tons) 4-110
Table 4-90: Production and Consumption Data for the Calculation of CO2 Emissions from Metallurgical Coke
Production (Million ft3) 4-110
Table 4-91: Material Carbon Contents for Iron and Steel Production 4-111
Table 4-92: CFU Emission Factors for Sinter and Pig Iron Production 4-111
Table 4-93: CO2 Emission Factors for Sinter Production, Direct Reduced Iron Production, and Pellet Production
4-112
Table 4-94: Production and Consumption Data for the Calculation of CO2 and CFU Emissions from Iron and Steel
Production (Thousand Metric Tons) 4-113
Table 4-95: Production and Consumption Data for the Calculation of CO2 Emissions from Iron and Steel Production
(Million ft3 unless otherwise specified) 4-113
xvii
-------
Table 4-96: Approach 2 Quantitative Uncertainty Estimates for CO2 and Cm Emissions from Iron and Steel
Production and Metallurgical Coke Production (MMT CO2 Eq. and Percent) 4-115
Table 4-97: CO2 and Cm Emissions from Ferroalloy Production (MMT CO2 Eq.) 4-117
Table 4-98: CO2 and Cm Emissions from Ferroalloy Production (kt) 4-117
Table 4-99: Production of Ferroalloys (Metric Tons) 4-119
Table 4-100: Approach 2 Quantitative Uncertainty Estimates for CO2 Emissions from Ferroalloy Production (MMT
CO2 Eq. and Percent) 4-120
Table 4-101: CO2 Emissions from Aluminum Production (MMT CO2 Eq.) 4-121
Table 4-102: CO2 Emissions from Aluminum Production (kt CO2) 4-121
Table 4-103: PFC Emissions from Aluminum Production (MMT CO2 Eq.) 4-122
Table 4-104: PFC Emissions from Aluminum Production (kt) 4-122
Table 4-105: Summary of HVAE Emissions (MMT CO2 Eq.) 4-124
Table 4-106: Summary of LVAE Emissions (MMT CO2 Eq.) 4-125
Table 4-107: Production of Primary Aluminum (kt) 4-126
Table 4-108: Approach 2 Quantitative Uncertainty Estimates for CO2 and PFC Emissions from Aluminum
Production (MMT CO2 Eq. and Percent) 4-127
Table 4-109: SF6, HFC-134a, FK 5-1-12 and CO2 Emissions from Magnesium Production and Processing (MMT CO2
Eq.) 4-128
Table 4-110: SF6, HFC-134a, FK 5-1-12 and CO2 Emissions from Magnesium Production and Processing (kt).... 4-128
Table 4-111: SF6 Emission Factors (kg SF6 per metric ton of magnesium) 4-131
Table 4-112: Approach 2 Quantitative Uncertainty Estimates for SF6, HFC-134a and CO2 Emissions from
Magnesium Production and Processing (MMT CO2 Eq. and Percent) 4-132
Table 4-113: CO2 Emissions from Lead Production (MMT CO2 Eq.) 4-134
Table 4-114: CO2 Emissions from Lead Production (kt CO2) 4-134
Table 4-115: Lead Production (Metric Tons) 4-135
Table 4-116: Approach 2 Quantitative Uncertainty Estimates for CO2 Emissions from Lead Production (MMT CO2
Eq. and Percent) 4-136
Table 4-117: CO2 Emissions from Zinc Production (MMT CO2 Eq.) 4-138
Table 4-118: CO2 Emissions from Zinc Production (kt CO2) 4-139
Table 4-119: Zinc Production (Metric Tons) 4-139
Table 4-120: Approach 2 Quantitative Uncertainty Estimates for CO2 Emissions from Zinc Production (MMT CO2 Eq.
and Percent) 4-142
Table 4-121: PFC, HFC, SF6, NF3, and N2O Emissions from Electronics Industry (MMT CO2 Eq.) 4-146
Table 4-122: PFC, HFC, SF6, NF3, and N2O Emissions from Semiconductor Manufacture (Metric Tons) 4-146
Table 4-123: F-HTF Emissions from Electronics Manufacture by Compound Group (kt CO2 Eq.) 4-147
Table 4-124: Approach 2 Quantitative Uncertainty Estimates for HFC, PFC, SF6, NF3 and N2O Emissions from
Electronics Manufacture (MMT CO2 Eq. and Percent) 4-159
Table 4-125: Emissions of HFCs, PFCs, and CO2 from ODS Substitutes (MMT CO2 Eq.) 4-161
xviii Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
Table 4-126: Emissions of HFCs, PFCs, and CO2 from ODS Substitution (Metric Tons) 4-162
Table 4-127: Emissions of HFCs, PFCs, and CO2 from ODS Substitutes (MMT CO2 Eq.) by Sector 4-163
Table 4-128: Approach 2 Quantitative Uncertainty Estimates for HFC and PFC Emissions from ODS Substitutes
(MMT CO2 Eq. and Percent) 4-166
Table 4-129: SF6 and CF4 Emissions from Electric Power Systems and Electrical Equipment Manufacturers (MMT
CO2 Eq.) 4-169
Table 4-130: SF6 and CF4 Emissions from Electric Power Systems and Electrical Equipment Manufacturers (kt) 4-169
Table 4-131: GHGRP-only Average Emission Rate (kg per mile) 4-172
Table 4-132: Categorization of Utilities and Timeseries for Application of Corresponding Emission Estimation
Methodologies 4-172
Table 4-133: California GHGRP and Voluntarily Reported SF6 Emissions Compared to CARB's SF6 Emissions (MMT
CO2 Eq.) 4-173
Table 4-134: Approach 2 Quantitative Uncertainty Estimates for SF6 and CF4 Emissions from Electrical Equipment
(MMT CO2 Eq. and Percent) 4-176
Table 4-135: SF6 and PFC Emissions from Other Product Use (MMT CO2 Eq.) 4-179
Table 4-136: Approach 2 Quantitative Uncertainty Estimates for SF6 and PFC Emissions from Other Product Use
(MMT CO2 Eq. and Percent) 4-182
Table 4-137: N20 Production (kt) 4-183
Table 4-138: N2O Emissions from N2O Product Usage (MMT CO2 Eq.) 4-183
Table 4-139: N2O Emissions from N2O Product Usage (kt N2O) 4-184
Table 4-140: Approach 2 Quantitative Uncertainty Estimates for N2O Emissions from N2O Product Usage (MMT
CO2 Eq. and Percent) 4-185
Table 4-141: NOx, CO, NMVOC, and SO2 Emissions from Industrial Processes and Product Use (kt) 4-187
Table 5-1: Emissions from Agriculture (MMT CO2 Eq.) 5-3
Table 5-2: Emissions from Agriculture (kt) 5-3
Table 5-3: CH4 Emissions from Enteric Fermentation (MMT CO2 Eq.) 5-5
Table 5-4: CH4 Emissions from Enteric Fermentation (kt CH4) 5-5
Table 5-5: Approach 2 Quantitative Uncertainty Estimates for CH4 Emissions from Enteric Fermentation (MMT CO2
Eq. and Percent) 5-9
Table 5-6: CH4 and N2O Emissions from Manure Management (MMT CO2 Eq.) 5-13
Table 5-7: CH4 and N2O Emissions from Manure Management (kt) 5-14
Table 5-8: Approach 2 Quantitative Uncertainty Estimates for CH4 and N2O (Direct and Indirect) Emissions from
Manure Management (MMT CO2 Eq. and Percent) 5-18
Table 5-9: IPCC (2006) Implied Emission Factor Default Values Compared with Calculated Values for CH4 from
Manure Management (kg/head/year) 5-19
Table 5-10
Table 5-11
Table 5-12
Table 5-13
CH4 Emissions from Rice Cultivation (MMT CO2 Eq.) 5-22
CH4 Emissions from Rice Cultivation (kt CH4) 5-23
Rice Area Harvested (1,000 Hectares) 5-25
Average Ratooned Area as Percent of Primary Growth Area (Percent) 5-25
xix
-------
Table 5-14: Approach 2 Quantitative Uncertainty Estimates for Cm Emissions from Rice Cultivation (MMT CO2 Eq.
and Percent) 5-27
Table 5-15: N2O Emissions from Agricultural Soils (MMT CO2 Eq.) 5-30
Table 5-16: N2O Emissions from Agricultural Soils (kt N2O) 5-30
Table 5-17: Direct N2O Emissions from Agricultural Soils by Land Use Type and Nitrogen Input Type (MMT CO2 Eq.)
5-30
Table 5-18: Indirect N2O Emissions from Agricultural Soils (MMT CO2 Eq.) 5-31
Table 5-19: Quantitative Uncertainty Estimates of N2O Emissions from Agricultural Soil Management in 2022
(MMT CO2 Eq. and Percent) 5-46
Table 5-20
Table 5-21
Table 5-22
Table 5-23
Percent)...
Table 5-24
Table 5-25
Table 5-26
Table 5-27
Percent)...
Emissions from Liming (MMT CO2 Eq.) 5-48
Emissions from Liming (MMT C) 5-48
Applied Minerals (MMT) 5-49
Approach 2 Quantitative Uncertainty Estimates for CO2 Emissions from Liming (MMT CO2 Eq. and
5-50
CO2 Emissions from Urea Fertilization (MMT CO2 Eq.) 5-51
CO2 Emissions from Urea Fertilization (MMT C) 5-51
Applied Urea (MMT) 5-51
Quantitative Uncertainty Estimates for CO2 Emissions from Urea Fertilization (MMT CO2 Eq. and
5-52
Table 5-28: CFU and N2O Emissions from Field Burning of Agricultural Residues (MMT CO2 Eq.) 5-53
Table 5-29: CFU, N2O, CO, and NOx Emissions from Field Burning of Agricultural Residues (kt) 5-54
Table 5-30: Agricultural Crop Production (kt of Product) 5-58
Table 5-31: U.S. Average Percent Crop Area Burned by Crop (Percent) 5-59
Table 5-32: Parameters for Estimating Emissions from Field Burning of Agricultural Residues 5-60
Table 5-33: Greenhouse Gas Emission Ratios and Conversion Factors 5-61
Table 5-34: Approach 2 Quantitative Uncertainty Estimates for CFU and N2O Emissions from Field Burning of
Agricultural Residues (MMT CO2 Eq. and Percent) 5-61
Table 6-1: Emissions and Removals (Net Flux) from Land Use, Land-Use Change, and Forestry (MMT CO2 Eq.).... 6-4
Table 6-2: Emissions and Removals from Land Use, Land-Use Change, and Forestry by Gas (MMT CO2 Eq.) 6-6
Table 6-3: Emissions and Removals from Land Use, Land-Use Change, and Forestry by Gas (kt) 6-7
Table 6-4: Managed and Unmanaged Land Area by Land-Use Categories for All 50 States (Thousands of Hectares)
6-11
Table 6-5: Land Use and Land-Use Change for the U.S. Managed Land Base for All 50 States (Thousands of
Hectares) 6-11
Table 6-6: Data Sources Used to Determine Land Use and Land Area for the Conterminous United States, Hawaii,
and Alaska 6-17
Table 6-7: Total Land Area (Hectares) by Land Use Category for U.S. Territories 6-24
Table 6-8: Net CO2 Flux from Forest Ecosystem Pools in Forest Land Remaining Forest Land and Harvested Wood
Pools (MMT CO2 Eq.) 6-29
xx Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
Table 6-9: Net Carbon Flux from Forest Ecosystem Pools in Forest Land Remaining Forest Land and Harvested
Wood Pools (MMTC) 6-30
Table 6-10: Forest Area (1,000 ha) and Carbon Stocks in Forest Land Remaining Forest Land and Harvested Wood
Pools (MMT C) 6-31
Table 6-11: Estimates of CO2 (MMT per Year) Emissions3 from Forest Fires in the Conterminous 48 States, Hawaii,
Puerto Rico, Guam, and Alaska 6-33
Table 6-12: Quantitative Uncertainty Estimates for Net CO2 Flux from Forest Land Remaining Forest Land: Changes
in Forest Carbon Stocks (MMT CO2 Eq. and Percent) 6-38
Table 6-13: Recalculations of Forest Area (1,000 ha) and Carbon Stocks in Forest Land Remaining Forest Land and
Harvested Wood Pools (MMT C) 6-40
Table 6-14: Recalculations of Forest Area (1,000 ha) and Carbon Stocks in Forest Land Remaining Forest Land
(MMT C) in the Conterminous United States 6-40
Table 6-15: Recalculations of Forest Area (1,000 ha) and Carbon Stocks in Forest Land Remaining Forest Land
(MMT C) in Coastal Southeast and Southcentral Alaska 6-41
Table 6-16: Recalculations of Forest Area (1,000 ha) and Carbon Stocks in Forest Land Remaining Forest Land
(MMT C) in Interior Alaska 6-41
Table 6-17: Recalculations of Forest Area (1,000 ha) and Carbon Stocks in Forest Land Remaining Forest Land
(MMT C) in Hawaii and United States Territories 6-41
Table 6-18: Recalculations of Net Carbon Flux from Forest Ecosystem Pools in Forest Land Remaining Forest Land
and Harvested Wood Pools (MMT C) 6-42
Table 6-19: Recalculations of Net Carbon Flux from Forest Ecosystem Pools in Forest Land Remaining Forest Land
(MMT C) in the Conterminous United States 6-43
Table 6-20: Recalculations of Net Carbon Flux from Forest Ecosystem Pools in Forest Land Remaining Forest Land
(MMT C) in Coastal Alaska 6-43
Table 6-21: Recalculations of Net Carbon Flux from Forest Ecosystem Pools in Forest Land Remaining Forest Land
(MMT C) in Interior Alaska 6-43
Table 6-22: Recalculations of Net Carbon Flux from Forest Ecosystem Pools in Forest Land Remaining Forest Land
(MMT C) in Hawaii and United States Territories 6-43
Table 6-23: Non-CC>2 Emissions from Forest Fires (MMT CO2 Eq.)a 6-45
Table 6-24: Non-CC>2 Emissions from Forest Fires (kt)a 6-45
Table 6-25: Quantitative Uncertainty Estimates of Non-C02 Emissions from Forest Fires (MMT CO2 Eq. and
Percent)3 6-46
Table 6-26: N2O Fluxes from Soils in Forest Land Remaining Forest Land and Land Converted to Forest Land (MMT
CO2 Eq. and kt N20) 6-48
Table 6-27: Quantitative Uncertainty Estimates of N2O Fluxes from Soils in Forest Land Remaining Forest Land and
Land Converted to Forest Land (MMT CO2 Eq. and Percent) 6-49
Table 6-28: Non-C02 Emissions from Drained Organic Forest Soilsa b (MMT CO2 Eq.) 6-50
Table 6-29: Non-C02 Emissions from Drained Organic Forest Soilsa b (kt) 6-50
Table 6-30: States identified as having Drained Organic Soils, Area of Forest on Drained Organic Soils, and
Sampling Error 6-51
Table 6-31: Quantitative Uncertainty Estimates for Non-C02 Emissions on Drained Organic Forest Soils (MMT CO2
Eq. and Percent)3 6-52
XXI
-------
Table 6-32: Net CO2 Flux from Forest Carbon Pools in Land Converted to Forest Land by Land Use Change Category
(MMT CO2 Eq.) 6-54
Table 6-33: Net Carbon Flux from Forest Carbon Pools in Land Converted to Forest Land by Land Use Change
Category (MMT C) 6-55
Table 6-34: Quantitative Uncertainty Estimates for Forest Carbon Pool Stock Changes (MMT CO2 Eq. per Year) in
2022 from Land Converted to Forest Land by Land Use Change 6-58
Table 6-35: Recalculations of the Net Carbon Flux from Forest Carbon Pools in Land Converted to Forest Land by
Land Use Change Category (MMT C) 6-60
Table 6-36: Net CO2 Flux from Soil Carbon Stock Changes in Cropland Remaining Cropland (MMT CO2 Eq.) 6-62
Table 6-37: Net CO2 Flux from Soil Carbon Stock Changes in Cropland Remaining Cropland (MMT C) 6-63
Table 6-38: Approach 2 Quantitative Uncertainty Estimates for Soil Carbon Stock Changes occurring within
Cropland Remaining Cropland (MMT CO2 Eq. and Percent) 6-72
Table 6-39: Comparison of Managed Land Area in Cropland Remaining Cropland and Area in the Current Cropland
Remaining Cropland Inventory (Thousand Hectares) 6-73
Table 6-40: Net CO2 Flux from Soil, Dead Organic Matter and Biomass Carbon Stock Changes in Land Converted to
Cropland by Land-Use Change Category (MMT CO2 Eq.) 6-75
Table 6-41: Net CO2 Flux from Soil, Dead Organic Matter and Biomass Carbon Stock Changes in Land Converted to
Cropland (MMT C) 6-76
Table 6-42: Approach 2 Quantitative Uncertainty Estimates for Soil, Dead Organic Matter and Biomass Carbon
Stock Changes occurring within Land Converted to Cropland (MMT CO2 Eq. and Percent) 6-79
Table 6-43: Net CO2 Flux from Soil, Dead Organic Matter and Biomass Carbon Stock Changes in Grassland
Remaining Grassland (MMT CO2 Eq.) 6-82
Table 6-44: Net CO2 Flux from Soil, Dead Organic Matter and Biomass Carbon Stock Changes in Grassland
Remaining Grassland (MMT C) 6-82
Table 6-45: Approach 2 Quantitative Uncertainty Estimates for Carbon Stock Changes Occurring Within Grassland
Remaining Grassland (MMT CO2 Eq. and Percent) 6-88
Table 6-46: Comparison of Managed Land Area in Grassland Remaining Grassland and the Area in the current
Grassland Remaining Grassland Inventory (Thousand Hectares) 6-89
Table 6-47: CH4 and N2O Emissions from Biomass Burning in Grassland (MMT CO2 Eq.) 6-90
Table 6-48: CH4, N2O, CO, and NOx Emissions from Biomass Burning in Grassland (kt) 6-90
Table 6-49: Thousands of Grassland Hectares Burned Annually 6-91
Table 6-50: Uncertainty Estimates for Non-C02 Greenhouse Gas Emissions from Biomass Burning in Grassland
(MMT CO2 Eq. and Percent) 6-92
Table 6-51: Net CO2 Flux from Soil, Dead Organic Matter and Biomass Carbon Stock Changes for Land Converted to
Grassland (MMT C02 Eq.) 6-94
Table 6-52: Net CO2 Flux from Soil, Dead Organic Matter and Biomass Carbon Stock Changes for Land Converted to
Grassland (MMT C) 6-95
Table 6-53: Approach 2 Quantitative Uncertainty Estimates for Soil, Dead Organic Matter and Biomass Carbon
Stock Changes occurring within Land Converted to Grassland (MMT CO2 Eq. and Percent) 6-99
Table 6-54: Comparison of Managed Land Area in Land Converted to Grassland and Area in the current Land
Converted to Grassland Inventory (Thousand Hectares) 6-100
xxii Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
Table 6-55: Emissions from Peatlands Remaining Peatlands (MMT CO2 Eq.) 6-103
Table 6-56: Emissions from Peatlands Remaining Peatlands (kt) 6-103
Table 6-57: Peat Production of Conterminous 48 States (kt) 6-105
Table 6-58: Peat Production of Alaska (Thousand Cubic Meters) 6-105
Table 6-59: Peat Production Area of Conterminous 48 States (Hectares) 6-105
Table 6-60: Peat Production Area of Alaska (Hectares) 6-106
Table 6-61: Peat Production (Hectares) 6-106
Table 6-62: Approach 2 Quantitative Uncertainty Estimates for CO2, CH4, and N2O Emissions from Peatlands
Remaining Peatlands (MMT CO2 Eq. and Percent) 6-107
Table 6-63: Emissions and Removals from Coastal Wetlands Remaining Coastal Wetlands (MMT CO2 Eq.) 6-110
Table 6-64: Net CO2 Flux from Carbon Stock Changes in Vegetated Coastal Wetlands Remaining Vegetated Coastal
Wetlands (MMT C02 Eq.) 6-112
Table 6-65: Net CO2 Flux from Carbon Stock Changes in Vegetated Coastal Wetlands Remaining Vegetated Coastal
Wetlands (MMT C) 6-112
Table 6-66: CH4 Emissions from Vegetated Coastal Wetlands Remaining Vegetated Coastal Wetlands (MMT CO2
Eq. and kt CH4) 6-112
Table 6-67: Area of Vegetated Coastal Wetlands Remaining Vegetated Coastal Wetlands, Vegetated Coastal
Wetlands Converted to Unvegetated Open Water Coastal Wetlands, and Unvegetated Open Water Coastal
Wetlands Converted to Vegetated Coastal Wetlands (ha) 6-113
Table 6-68
Table 6-69
Table 6-70
Table 6-71
Aboveground Biomass Carbon Stocks for Vegetated Coastal Wetlands (t C ha_1) 6-113
Root to Shoot Ratios for Vegetated Coastal Wetlands 6-113
Annual Soil Carbon Accumulation Rates for Vegetated Coastal Wetlands (t C ha 1 yr-1) 6-114
IPCC Approach 1 Quantitative Uncertainty Estimates for Carbon Stock Changes and CH4 Emissions
occurring within Vegetated Coastal Wetlands Remaining Vegetated Coastal Wetlands in 2021 (MMT CO2 Eq. and
Percent) 6-115
Table 6-72: Net CO2 Flux from Carbon Stock Changes in Vegetated Coastal Wetlands Converted to Unvegetated
Open Water Coastal Wetlands (MMT CO2 Eq.) 6-117
Table 6-73: Net CO2 Flux from Carbon Stock Changes in Vegetated Coastal Wetlands Converted to Unvegetated
Open Water Coastal Wetlands (MMT C) 6-117
Table 6-74: Approach 1 Quantitative Uncertainty Estimates for CO2 Flux Occurring within Vegetated Coastal
Wetlands Converted to Unvegetated Open Water Coastal Wetlands in 2022 (MMT CO2 Eq. and Percent) 6-119
Table 6-75: CO2 Flux from Carbon Stock Changes from Unvegetated Open Water Coastal Wetlands Converted to
Vegetated Coastal Wetlands (MMT CO2 Eq.) 6-121
Table 6-76: CO2 Flux from Carbon Stock Changes from Unvegetated Open Water Coastal Wetlands Converted to
Vegetated Coastal Wetlands (MMT C) 6-121
Table 6-77: Approach 1 Quantitative Uncertainty Estimates for Carbon Stock Changes Occurring within
Unvegetated Open Water Coastal Wetlands Converted to Vegetated Coastal Wetlands in 2022 (MMT CO2 Eq. and
Percent) 6-124
Table 6-78: N2O Emissions from Aquaculture in Coastal Wetlands (MMT CO2 Eq. and kt N2O) 6-125
Table 6-79: Approach 1 Quantitative Uncertainty Estimates for N2O Emissions from Aquaculture Production in
Coastal Wetlands in 2022 (MMT CO2 Eq. and Percent) 6-126
xxiii
-------
Table 6-80: Cm Emissions from Flooded Land Remaining Flooded Land—Reservoirs (MMT CO2 Eq.) 6-128
Table 6-81: CFU Emissions from Flooded Land Remaining Flooded Land—Reservoirs (kt CH4) 6-128
Table 6-82: Surface and Downstream CH4 Emissions from Reservoirs in Flooded Land Remaining Flooded Land in
2022 (kt CH4) 6-129
Table 6-83: IPCC (2019) Default CH4 Emission Factors for Surface Emission from Reservoirs in Flooded Land
Remaining Flooded Land 6-130
Table 6-84: National Totals of Reservoir Surface Area in Flooded Land Remaining Flooded Land (millions of ha)
6-132
Table 6-85: State Breakdown of Reservoir Surface Area in Flooded Land Remaining Flooded Land (millions of ha)
6-132
Table 6-86: Approach 2 Quantitative Uncertainty Estimates for CH4 Emissions from Reservoirs in Flooded Land
Remaining Flooded Land 6-133
Table 6-87: CH4 Emissions from Other Constructed Waterbodies in Flooded Land Remaining Flooded Land (MMT
CO2 Eq.) 6-135
Table 6-88: CH4 Emissions from Other Constructed Waterbodies in Flooded Land Remaining Flooded Land (kt CH4)
6-135
Table 6-89: CH4 Emissions from Other Constructed Waterbodies in Flooded Land Remaining Flooded Land in 2022
(kt CH4) 6-135
Table 6-90: IPCC (2019) Default CH4 Emission Factors for Surface Emissions from Other Constructed Waterbodies in
Flooded Land Remaining Flooded Land 6-137
Table 6-91: Predictors used in Decision Tree to Identify Canal/Ditches 6-139
Table 6-92: Validation Results for Ditch/Canal Classification Decision Tree 6-139
Table 6-93: National Surface Area Totals in Flooded Land Remaining Flooded Land - Other Constructed
Waterbodies (ha) 6-140
Table 6-94: State Totals of Surface Area in Flooded Land Remaining Flooded Land— Canals and Ditches (ha). 6-140
Table 6-95: State Totals of Surface Area in Flooded Land Remaining Flooded Land— Freshwater Ponds (ha)... 6-141
Table 6-96: Approach 2 Quantitative Uncertainty Estimates for CH4 Emissions from Other Constructed
Waterbodies in Flooded Land Remaining Flooded Land 6-143
Table 6-97: Net CO2 Flux from Carbon Stock Changes in Land Converted to Vegetated Coastal Wetlands (MMT CO2
Eq.) 6-145
Table 6-98: Net CO2 Flux from Carbon Stock Changes in Land Converted to Vegetated Coastal Wetlands (MMT C)
6-145
Table 6-99: CH4 Emissions from Land Converted to Vegetated Coastal Wetlands (MMT CO2 Eq. and kt CH4).... 6-146
Table 6-100: Approach 1 Quantitative Uncertainty Estimates for Carbon Stock Changes occurring within Land
Converted to Vegetated Coastal Wetlands in 2022 (MMT CO2 Eq. and Percent) 6-149
Table 6-101
Table 6-102
Table 6-103
Table 6-104
CH4 Emissions from Land Converted to Flooded Land—Reservoirs (MMT CO2 Eq.) 6-151
CH4 Emissions from Land Converted to Flooded Land—Reservoirs (kt CH4) 6-151
CO2 Emissions from Land Converted to Flooded Land —Reservoirs (MMT CO2) 6-152
CO2 Emissions from Land Converted to Flooded Land —Reservoirs (MMT C) 6-152
xxiv Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
Table 6-105: Methane and CO2 Emissions from Reservoirs in Land Converted to Flooded Land in 2022 (kt CH4; kt
CO2) 6-152
Table 6-106: IPCC (2019) Default Cm and CO2 Emission Factors for Surface Emissions from Reservoirs in Land
Converted to Flooded Land 6-154
Table 6-107: National Totals of Reservoir Surface Area in Land Converted to Flooded Land (thousands of ha).. 6-155
Table 6-108: State Breakdown of Reservoir Surface Area in Land Converted to Flooded Land (thousands of ha)
.6-156
Table 6-109: Approach 2 Quantitative Uncertainty Estimates for CFU and CO2 Emissions from Reservoirs in Land
Converted to Flooded Land 6-157
Table 6-110: CFU Emissions from Other Constructed Waterbodies in Land Converted to Flooded Land (MMT CO2
Eq.) 6-159
Table 6-111: CFU Emissions from Other Constructed Waterbodies in Land Converted to Flooded Land (kt CH4)6-159
Table 6-112: CO2 Emissions from Other Constructed Waterbodies
Eq.)
Table 6-113: CO2 Emissions from Other Constructed Waterbodies
n Land Converted to Flooded Land (MMT CO2
6-159
n Land Converted to Flooded Land (MMT C)
6-159
Table 6-114: CFU and CO2 Emissions from Other Constructed Waterbodies in Land Converted to Flooded Land in
2022 (MTCO2 Eq.) 6-159
Table 6-115: IPCC Default Methane and CO2 Emission Factors for Other Constructed Waterbodies in Land
Converted to Flooded Land 6-161
Table 6-116: National Surface Area Totals of Other Constructed Waterbodies in Land Converted to Flooded Land
(ha) 6-163
Table 6-117: State Surface Area Totals of Other Constructed Waterbodies in Land Converted to Flooded Land (ha)
6-163
Table 6-118: Approach 2 Quantitative Uncertainty Estimates for CH4 and CO2 Emissions from Other Constructed
Waterbodies in Land Converted to Flooded Land 6-164
Table 6-119: Net CO2 Flux from Soil C Stock Changes in Settlements Remaining Settlements (MMT CO2 Eq.)... 6-166
Table 6-120: Net CO2 Flux from Soil C Stock Changes in Settlements Remaining Settlements (MMT C) 6-167
Table 6-121: Thousands of Hectares of Drained Organic Soils in Settlements Remaining Settlements 6-167
Table 6-122: Uncertainty Estimates for CO2 Emissions from Drained Organic Soils in Settlements Remaining
Settlements (MMT CO2 Eq. and Percent) 6-168
Table 6-123: Area of Managed Land in Settlements Remaining Settlements that is not included in the current
Inventory (Thousand Hectares) 6-169
Table 6-124: Net Flux from Trees in Settlements Remaining Settlements (MMT CO2 Eq. and MMT C)a 6-170
Table 6-125: Carbon Storage (kg C/m2 tree cover), Gross and Net Sequestration (kg C/m2 tree cover/year) and Tree
Cover (percent) among Sampled U.S. Cities (see Nowak et al. 2013) 6-172
Table 6-126: Estimated Annual Carbon Sequestration, Tree Cover, and Annual Carbon Sequestration per Area of
Tree Cover for settlement areas in the United States by State and the District of Columbia (2022) 6-174
Table 6-127: Approach 2 Quantitative Uncertainty Estimates for Net CO2 Flux from Changes in Carbon Stocks in
Settlement Trees (MMT CO2 Eq. and Percent) 6-176
Table 6-128: Recalculations of the Settlement Tree Categories 6-176
xxv
-------
Table 6-129: N2O Emissions from Soils in Settlements Remaining Settlements (MMT CO2 Eq. and kt N2O) 6-177
Table 6-130: Quantitative Uncertainty Estimates of N2O Emissions from Soils in Settlements Remaining
Settlements (MMT CO2 Eq. and Percent) 6-180
Table 6-131: Net Changes in Yard Trimmings and Food Scrap Carbon Stocks in Landfills (MMT CO2 Eq.) 6-181
Table 6-132: Net Changes in Yard Trimmings and Food Scrap Carbon Stocks in Landfills (MMT C) 6-182
Table 6-133: Moisture Contents, Carbon Storage Factors (Proportions of Initial Carbon Sequestered), Initial C
Contents, and Decay Rates for Yard Trimmings and Food Scraps in Landfills 6-185
Table 6-134: Carbon Stocks in Yard Trimmings and Food Scraps in Landfills (MMT C) 6-185
Table 6-135: Approach 2 Quantitative Uncertainty Estimates for CO2 Flux from Yard Trimmings and Food Scraps in
Landfills (MMT CO2 Eq. and Percent) 6-186
Table 6-136: Net CO2 Flux from Soil, Dead Organic Matter and Biomass Carbon Stock Changes for Land Converted
to Settlements (MMT CO2 Eq.) 6-188
Table 6-137: Net CO2 Flux from Soil, Dead Organic Matter and Biomass Carbon Stock Changes for Land Converted
to Settlements (MMT C) 6-188
Table 6-138: Approach 2 Quantitative Uncertainty Estimates for Soil, Dead Organic Matter and Biomass Carbon
Stock Changes occurring within Land Converted to Settlements (MMT CO2 Eq. and Percent) 6-192
Table 6-139: Area of Managed Land in Land Converted to Settlements that is not included in the current Inventory
(Thousand Hectares) 6-193
Table 7-1: Emissions from Waste (MMT CO2 Eq.) 7-2
Table 7-2: Emissions from Waste (kt) 7-2
Table 7-3: CFU Emissions from Landfills (MMT CO2 Eq.) 7-7
Table 7-4: CFU Emissions from Landfills (kt CH4) 7-7
Table 7-5: Approach 2 Quantitative Uncertainty Estimates for CFU Emissions from Landfills (MMT CO2 Eq. and
Percent) 7-13
Table 7-6: Materials Discarded in the Municipal Waste Stream by Waste Type from 1990 to 2018 (Percent) 7-17
Table 7-7: CFU and N2O Emissions from Domestic and Industrial Wastewater Treatment (MMT CO2 Eq.) 7-21
Table 7-8: CFU and N2O Emissions from Domestic and Industrial Wastewater Treatment (kt) 7-21
Table 7-9: Domestic Wastewater CH4 Emissions from Septic and Centralized Systems (2022, kt, MMT CO2 Eq. and
Percent) 7-23
Table 7-10: Variables and Data Sources for CFU Emissions from Septic Systems 7-24
Table 7-11: Variables and Data Sources for Organics in Domestic Wastewater 7-25
Table 7-12: U.S. Population (Millions) and Domestic Wastewater TOW (kt) 7-25
Table 7-13: Variables and Data Sources for Organics in Centralized Domestic Wastewater 7-26
Table 7-14: Variables and Data Sources for CFU Emissions from Centrally Treated Aerobic Systems (Other than
Constructed Wetlands) 7-27
Table 7-15: Variables and Data Sources for CFU Emissions from Centrally Treated Aerobic Systems (Constructed
Wetlands) 7-28
Table 7-16: Variables and Data Sources for CFU Emissions from Centrally Treated Anaerobic Systems 7-29
Table 7-17: Variables and Data Sources for Emissions from Anaerobic Sludge Digesters 7-30
xxvi Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
Table 7-18: Variables and Data Sources for CFU Emissions from Centrally Treated Systems Discharge 7-31
Table 7-19: Total Industrial Wastewater Cm Emissions by Sector (2022, MMT CO2 Eq. and Percent) 7-32
Table 7-20: U.S. Pulp and Paper, Meat, Poultry, Vegetables, Fruits and Juices, Ethanol, Breweries, and Petroleum
Refining Production (MMT) 7-35
Table 7-21: U.S. Industrial Wastewater Characteristics Data (2022) 7-35
Table 7-22: U.S. Industrial Wastewater Treatment Activity Data 7-36
Table 7-23: Sludge Variables for Aerobic Treatment Systems 7-36
Table 7-24: Fraction of TOW Removed During Treatment by Industry 7-37
Table 7-25: Wastewater Outflow (m3/ton) for Pulp, Paper, and Paperboard Mills 7-38
Table 7-26: Wastewater Outflow (m3/ton) and BOD Production (g/L) for U.S. Vegetables, Fruits, and Juices
Production 7-39
Table 7-27: Domestic Wastewater N2O Emissions from Septic and Centralized Systems (2022, kt, MMT CO2 Eq. and
Percent) 7-41
Table 7-28: Variables and Data Sources for Protein Consumed 7-42
Table 7-29: Variables and Data Sources for N2O Emissions from Septic System 7-43
Table 7-30: Variables and Data Sources for Non-Consumed Protein and Nitrogen Entering Centralized Systems 7-44
Table 7-31: Variables and Data Sources for N2O Emissions from Centrally Treated Aerobic Systems (Other than
Constructed Wetlands) 7-45
Table 7-32: Variables and Data Sources for N2O Emissions from Centrally Treated Aerobic Systems (Constructed
Wetlands) 7-46
Table 7-33: Variables and Data Sources for N2O Emissions from Centrally Treated Anaerobic Systems 7-47
Table 7-34: U.S. Population (Millions) Fraction of Population Served by Centralized Wastewater Treatment
(percent), Protein Supply (kg/person-year), and Protein Consumed (kg/person-year) 7-47
Table 7-35: Variables and Data Sources for N2O Emissions from Centrally Treated Systems Discharge 7-48
Table 7-36: Total Industrial Wastewater N2O Emissions by Sector (2022, MMT CO2 Eq. and Percent) 7-49
Table 7-37: U.S. Industrial Wastewater Nitrogen Data 7-50
Table 7-38: Industrial Wastewater Nitrogen Discharged in 2018 by Sector (kg N) 7-51
Table 7-39: Approach 2 Quantitative Uncertainty Estimates for 2022 Emissions from Wastewater Treatment (MMT
CO2 Eq. and Percent) 7-52
Table 7-40: CFU and N2O Emissions from Composting (MMT CO2 Eq.) 7-56
Table 7-41: CFU and N2O Emissions from Composting (kt) 7-56
Table 7-42: U.S. Waste Composted (kt) 7-57
Table 7-43: Tier 1 Quantitative Uncertainty Estimates for Emissions from Composting (MMT CO2 Eq. and Percent)
7-57
Table 7-44: CFU Emissions from Anaerobic Digestion at Biogas Facilities (MT CO2 Eq.) 7-59
Table 7-45: CFU Emissions from Anaerobic Digestion at Biogas Facilities (kt CH4) 7-60
Table 7-46: Estimated U.S. Waste Digested (kt) from 1990-2022 7-62
Table 7-47: Estimated Number of Stand-Alone AD Facilities Operating from 1990-2022 7-62
xxvii
-------
Table 7-48: Approach 1 Quantitative Uncertainty Estimates for Emissions from Anaerobic Digestion (MT CO2 Eq.
and Percent) 7-63
Table 7-49: Emissions of NOx, CO, NMVOC, and SO2 from Waste (kt) 7-65
Table 9-1: Overall Impact of Recalculations by Gas Compared to Previous Inventory 9-2
Table 9-2: Overall Impact of Recalculations by Sector Compared to Previous Inventory 9-3
Table 9-3: Key Recalculations 9-3
Table 9-4: Revisions to U.S. Greenhouse Gas Emissions (MMT CO2 Eq.) 9-5
Table 9-5: Revisions to U.S. Greenhouse Gas Emissions and Removals (Net Flux) from Land Use, Land-Use Change,
and Forestry (MMT C02 Eq.) 9-8
Figures
Figure ES-1: U.S. Greenhouse Gas Emissions and Sinks by Gas ES-5
Figure ES-2: Annual Percent Change in Gross U.S. Greenhouse Gas Emissions and Sinks Relative to the Previous
Year ES-6
Figure ES-3: Impacts of Recalculations on Net Emissions ES-7
Figure ES-4: 2022 Total Gross U.S. Greenhouse Gas Emissions by Gas (Percentages based on MMT CO2 Eq.) ES-8
Figure ES-5: 2022 Sources of CO2 Emissions ES-9
Figure ES-6: 2022 End-Use Sector Emissions of CO2 from Fossil Fuel Combustion ES-10
Figure ES-7: Electric Power Generation and Emissions ES-12
Figure ES-8: 2022 Sources of CH4 Emissions ES-13
Figure ES-9: 2022 Sources of N2O Emissions ES-14
Figure ES-10: 2022 Sources of HFCs, PFCs, SF6, and NF3 Emissions ES-15
Figure ES-11: U.S. Greenhouse Gas Emissions and Sinks by UNFCCC/IPCC Sector ES-16
Figure ES-12: 2022 U.S. Energy Consumption by Energy Source (Percent) ES-17
Figure ES-13: U.S. Greenhouse Gas Emissions Allocated to Economic Sectors ES-21
Figure ES-14: U.S. Greenhouse Gas Emissions with Electricity-Related Emissions Distributed to Economic Sectors
ES-23
Figure ES-15: U.S. Greenhouse Gas Emissions Per Capita and Per Dollar of Gross Domestic Product (GDP) ES-24
Figure ES-16: 2022 Key Categories (Approach 1 including LULUCF)3 ES-26
Figure 1-1: National Inventory Arrangements and Process Diagram 1-12
Figure 1-2: Summary of Key QC Processes from U.S. QA/QC Plan 1-26
Figure 2-1: U.S. Greenhouse Gas Emissions and Sinks by Gas 2-2
Figure 2-2: Annual Percentage Change in Gross U.S. Greenhouse Gas Emissions Relative to the Previous Year.... 2-2
Figure 2-3: 2022 Gross Total U.S. Greenhouse Gas Emissions by Gas (Percentages based on MMT CO2 Eq.) 2-3
Figure 2-4: U.S. Greenhouse Gas Emissions and Sinks by IPCC Sector 2-9
Figure 2-5: Trends in Energy Sector Greenhouse Gas Sources 2-11
Figure 2-6: Trends in CO2 Emissions from Fossil Fuel Combustion by End-Use Sector and Fuel Type 2-15
xxviii Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
Figure 2-7: Trends in End-Use Sector Emissions of CO2 from Fossil Fuel Combustion 2-16
Figure 2-8: Electric Power Generation (Billion kWh) and Emissions (MMT CO2 Eq.) 2-17
Figure 2-9: Trends in Industrial Processes and Product Use Sector Greenhouse Gas Sources 2-19
Figure 2-10: Trends in Agriculture Sector Greenhouse Gas Sources 2-22
Figure 2-11: Trends in Emissions and Removals (Net CO2 Flux) from Land Use, Land-Use Change, and Forestry.. 2-25
Figure 2-12: Trends in Waste Sector Greenhouse Gas Sources 2-28
Figure 2-13: U.S. Greenhouse Gas Emissions Allocated to Economic Sectors 2-30
Figure 2-14: U.S. Greenhouse Gas Emissions with Electricity-Related Emissions Distributed to Economic Sectors....
2-35
Figure 2-15: Trends in Transportation-Related Greenhouse Gas Emissions 2-38
Figure 2-16: U.S. Greenhouse Gas Emissions Per Capita and Per Dollar of Gross Domestic Product 2-41
Figure 3-1: 2022 Energy Sector Greenhouse Gas Sources 3-2
Figure 3-2: Trends in Energy Sector Greenhouse Gas Sources 3-2
Figure 3-3: 2022 U.S. Fossil Carbon Flows (MMT C02 Eq.) 3-3
Figure 3-4: 2022 U.S. Energy Use by Energy Source 3-10
Figure 3-5: Annual U.S. Energy Use 3-10
Figure 3-6: 2022 CO2 Emissions from Fossil Fuel Combustion by Sector and Fuel Type 3-11
Figure 3-7: Annual Deviations from Normal Heating Degree Days for the United States (1970-2022, Index Normal =
100) 3-12
Figure 3-8: Annual Deviations from Normal Cooling Degree Days for the United States (1970-2022, Index Normal =
100) 3-12
Figure 3-9: Fuels Used in Electric Power Generation and Total Electric Power Sector CO2 Emissions 3-19
Figure 3-10: Electric Power Retail Sales by End-Use Sector 3-19
Figure 3-11: Industrial Production Indices (Index 2017=100) 3-21
Figure 3-12: Fuels and Electricity Used in Industrial Sector, Industrial Output, and Total Sector CO2 Emissions
(Including Electricity) 3-21
Figure 3-13: Fuels and Electricity Used in Residential and Commercial Sectors, Heating and Cooling Degree Days,
and Total Sector CO2 Emissions (Including Electricity) 3-23
Figure 3-14: Fuels Used in Transportation Sector, On-road VMT, and Total Sector CO2 Emissions 3-25
Figure 3-15: Sales-Weighted Fuel Economy of New Passenger Cars and Light-Duty Trucks, 1990-2022 3-27
Figure 3-16: Sales of New Passenger Cars and Light-Duty Trucks, 1990-2022 3-27
Figure 3-17: Mobile Source CH4 and N2O Emissions 3-30
Figure 3-18: U.S. Energy Consumption and Energy-Related CO2 Emissions Per Capita and Per Dollar GDP 3-36
Figure 4-1: Industrial Processes and Product Use Sector Greenhouse Gas Sources 4-2
Figure 4-2: Trends in Industrial Processes and Product Use Sector Greenhouse Gas Sources 4-3
Figure 5-1: 2022 Agriculture Sector Greenhouse Gas Emission Sources 5-1
Figure 5-2: Trends in Agriculture Sector Greenhouse Gas Emission Sources 5-2
xx ix
-------
Figure 5-3: Annual Cm Emissions from Rice Cultivation, 2020, Using the Tier 3 DayCent Model 5-23
Figure 5-4: Sources and Pathways of Nitrogen that Result in N2O Emissions from Agricultural Soil Management5-29
Figure 5-5: Croplands, 2020 Annual Direct N2O Emissions Estimated Using the Tier 3 DayCent Model 5-32
Figure 5-6: Grasslands, 2020 Annual Direct N2O Emissions Estimated Using the Tier 3 DayCent Model 5-32
Figure 5-7: Croplands, 2020 Annual Indirect N2O Emissions from Volatilization Using the Tier 3 DayCent Model 5-33
Figure 5-8: Grasslands, 2020 Annual Indirect N2O Emissions from Volatilization Using the Tier 3 DayCent Model
5-34
Figure 5-9: Croplands, 2020 Annual Indirect N2O Emissions from Leaching and Runoff Using the Tier 3 DayCent
Model 5-34
Figure 5-10: Grasslands, 2020 Annual Indirect N2O Emissions from Leaching and Runoff Using the Tier 3 DayCent
Model 5-35
Figure 6-1
Figure 6-2
Figure 6-3
Figure 6-4
2022 LULUCF Chapter Greenhouse Gas Sources and Sinks 6-3
Trends in Emissions and Removals (Net CO2 Flux) from Land Use, Land-Use Change, and Forestry 6-3
Percent of Total Land Area for Each State in the General Land Use Categories for 2022 6-13
Changes in Forest Area by Region for Forest Land Remaining Forest Land in the conterminous United
States and Alaska (1990-2022) 6-28
Figure 6-5: Estimated Net Annual Changes in Carbon Stocks for All Carbon Pools in Forest Land Remaining Forest
Land in the Conterminous United States and Alaska (1990-2022) 6-32
Figure 6-6: Total Net Annual Soil Carbon Stock Changes for Mineral Soils under Agricultural Management within
States, 2020, Cropland Remaining Cropland 6-64
Figure 6-7: Total Net Annual Soil Carbon Stock Changes for Organic Soils under Agricultural Management within
States, 2020, Cropland Remaining Cropland 6-65
Figure 6-8: Total Net Annual Soil Carbon Stock Changes for Mineral Soils under Agricultural Management within
States, 2020, Grassland Remaining Grassland 6-83
Figure 6-9: Total Net Annual Soil Carbon Stock Changes for Organic Soils under Agricultural Management within
States, 2020, Grassland Remaining Grassland 6-84
Figure 6-10: U.S. Reservoirs (black polygons) in the Flooded Land Remaining Flooded Land Category in 2022 . 6-127
Figure 6-11: Total CFU Emissions (Downstream + Surface) from Reservoirs in Flooded Land Remaining Flooded
Land in 2022 (kt CH4) 6-128
Figure 6-12: Selected Features from NWI that Meet Flooded Lands Criteria 6-131
Figure 6-13: 2022 CFU Emissions from A) Ditches and Canals and B) Freshwater Ponds in Flooded Land Remaining
Flooded Land (kt CH4) 6-137
Figure 6-14: Left: NWI Features Identified as Canals/Ditches (pink) by Unique Narrow, Linear/Angular Morphology.
Right: Non-Canal/Ditches with More Natural Morphology (blue) 6-138
Figure 6-15: Structure of Decision Tree Used to Identify Canals/Ditches 6-139
Figure 6-16: 2022 Surface Area of A) Ditches and Canals and B) Freshwater Ponds in Flooded Land Remaining
Flooded Land (ha) 6-140
Figure 6-17
Figure 6-18
Figure 6-19
U.S. Reservoirs (black polygons) in the Land Converted to Flooded Land Category in 2022 6-151
2022 A) Cm and B) CO2 Emissions from U.S. Reservoirs in Land Converted to Flooded Land 6-152
Selected Features from NWI that meet Flooded Lands Criteria 6-155
xxx Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
Figure 6-20: Number of Dams Built per Year from 1990 through 2022 6-156
Figure 6-21: 2022 A) CFU and B) CO2 Emissions from Other Constructed Waterbodies (Freshwater Ponds) in Land
Converted to Flooded Land (MT CO2 Eq.) 6-161
Figure 6-22: Surface Area of Other Constructed Waterbodies in Land Converted to Flooded Land (ha) in 2022 6-163
Figure 7-1: 2022 Waste Sector Greenhouse Gas Sources 7-1
Figure 7-2: Trends in Waste Sector Greenhouse Gas Sources 7-2
Figure 7-3: Methodologies Used Across the Time Series to Compile the U.S. Inventory of Emission Estimates for
MSW Landfills 7-9
Figure 7-4: Management of Municipal Solid Waste in the United States, 2018 7-16
Figure 7-5: MSW Management Trends from 1990 to 2018 7-17
Figure 7-6: Percent of Degradable Materials Diverted from Landfills from 1990 to 2018 (Percent) 7-18
Figure 9-1: Impacts of Recalculations on Net Emissions 9-2
Figure 9-2: Impacts of Recalculations to U.S. Greenhouse Gas Emissions and Sinks by Sector 9-5
Boxes
Box ES-1: Methodological Approach for Estimating and Reporting U.S. Emissions and Removals, including
Relationship to EPA's Greenhouse Gas Reporting Program ES-1
Box ES-2: Trends in Various U.S. Greenhouse Gas Emissions-Related Data ES-23
Box ES-3: Use of Ambient Measurements Systems for Validation of Emission Inventories ES-27
Box 1-1: Methodological Approach for Estimating and Reporting U.S. Emissions and Removals, including
Relationship to EPA's Greenhouse Gas Reporting Program 1-2
Box 1-2: The IPCC Sixth Assessment Report and Global Warming Potentials 1-10
Box 1-3: Examples of Verification Activities 1-27
Box 2-1: Methodology for Aggregating Emissions by Economic Sector 2-32
Box 2-2: Trends in Various U.S. Greenhouse Gas Emissions-Related Data 2-40
Box 3-1: Methodological Approach for Estimating and Reporting U.S. Emissions and Removals, including
Relationship to EPA's Greenhouse Gas Reporting Program 3-5
Box 3-2: Weather and Non-Fossil Energy Effects on CO2 Emissions from Fossil Fuel Combustion Trends 3-11
Box 3-3: Uses of Greenhouse Gas Reporting Program Data and Improvements in Reporting Emissions from
Industrial Sector Fossil Fuel Combustion 3-22
Box 3-4: Carbon Intensity of U.S. Energy Consumption 3-35
Box 3-5: Reporting of Lubricants, Waxes, and Asphalt and Road Oil Product Use in Energy Sector 3-53
Box 3-6: Carbon Dioxide Transport, Injection, and Geological Storage 3-91
Box 4-1: Methodological Approach for Estimating and Reporting U.S. Emissions and Removals 4-8
Box 4-2: Industrial Process and Product Use Data from EPA's Greenhouse Gas Reporting Program 4-9
Box 5-1: Methodological Approach for Estimating and Reporting U.S. Emissions and Removals 5-3
Box 5-2: Surrogate Data Method 5-26
Box 5-3: Tier 1 vs. Tier 3 Approach for Estimating N2O Emissions 5-36
xxxi
-------
Box 5-4: Data Splicing Method 5-38
Box 5-5: Comparison of the Tier 2 U.S. Inventory Approach and IPCC (2006) Default Approach 5-48
Box 5-6: Comparison of Tier 2 U.S. Inventory Approach and IPCC (2006) Default Approach 5-56
Box 6-1: Methodological Approach for Estimating and Reporting U.S. Emissions and Removals 6-9
Box 6-2: Preliminary Estimates of Land Use in U.S. Territories 6-24
Box 6-3: CO2 Emissions from Forest Fires 6-32
Box 6-4: Surrogate Data Method 6-67
Box 6-5: Tier 3 Approach for Soil Carbon Stocks Compared to Tier 1 or 2 Approaches 6-68
Box 6-6: State-Level Case Studies for the Estimation ofGHG Removals in Seagrasses 6-116
Box 7-1: Methodological Approach for Estimating and Reporting U.S. Emissions and Removals, including
Relationship to Greenhouse Gas Reporting Data 7-3
Box 7-2: Description of a Modern, Managed Landfill in the United States 7-4
Box 7-3: Nationwide Municipal Solid Waste Data Sources 7-10
Box 7-4: Overview of U.S. Solid Waste Management Trends 7-16
Equations
Equation 1-1: Calculating CO2 Equivalent Emissions 1-9
Equation 3-1: Estimating Fugitive CO2 Emissions from Underground Mines 3-67
Equation 3-2: Estimating CO2 Emissions from Drained Methane Flared or Catalytically Oxidized 3-67
Equation 3-3: Decline Function to Estimate Venting Abandoned Mine Methane Emissions 3-71
Equation 3-4: Decline Function to Estimate Flooded Abandoned Mine Methane Emissions 3-71
Equation 4-1: 2006 IPCC Guidelines Tier 1 Emission Factor for Clinker (precursor to Equation 2.4) 4-12
Equation 4-2: 2006 IPCC Guidelines Tier 2 Emission Factor for Lime Production, High-Calcium Lime (Equation 2.9)
4-17
Equation 4-3: 2006 IPCC Guidelines Tier 2 Emission Factor for Lime Production, Dolomitic Lime (Equation 2.9).. 4-17
Equation 4-4: 2006 IPCC Guidelines Tier 3: N2O Emissions From Nitric Acid Production (Equation 3.6) 4-43
Equation 4-5: 2006 IPCC Guidelines Tier 2: N2O Emissions From Adipic Acid Production (Equation 3.8) 4-47
Equation 4-6: 2006 IPCC Guidelines Tier 1: N2O Emissions From Caprolactam Production (Equation 3.9) 4-51
Equation 4-78: 2006 IPCC Guidelines Tier 1: Emissions from Carbide Production (Equation 3.11) 4-55
Equation 4-9: 2006 IPCC Guidelines Tier 1: CO2 Emissions from Titanium Production (Equation 3.12) 4-58
Equation 4-10: CO2 Emissions from Phosphoric Acid Production 4-101
Equation 4-11: CO2 Emissions from Coke, Pig Iron, EAF Steel, and BOF Steel Production, based on 2006 IPCC
Guidelines Tier 2 Methodologies 4-107
Equation 4-12: 2006 IPCC Guidelines Tier 1: Emissions from Sinter, Direct Reduced Iron, and Pellet Production
(Equations 4.6, 4.7, and 4.8) 4-108
Equation 4-13: 2006 IPCC Guidelines Tier 1: CO2 Emissions for Ferroalloy Production (Equation 4.15) 4-117
Equation 4-14: 2006 IPCC Guidelines Tier 1: CFU Emissions for Ferroalloy Production (Equation 4.18) 4-118
xxxii Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
Equation 4-15: CF4 Emissions Resulting from Low Voltage Anode Effects 4-125
Equation 4-16: 2006IPCC Guidelines Tier 1: CO2 Emissions From Lead Production (Equation 4.32) 4-135
Equation 4-17: 2006 IPCC Guidelines Tier 1: CO2 Emissions from Zinc Production (Equation 4.33) 4-139
Equation 4-18: Waelz Kiln CO2 Emission Factor for Zinc Produced 4-140
Equation 4-19: Waelz Kiln CO2 Emission Factor for EAF Dust Consumed 4-140
Equation 4-20: Total Emissions from Electronics Industry 4-155
Equation 4-21: Total Emissions from Semiconductor Manufacturing 4-156
Equation 4-22: Total Emissions from MEMS Manufacturing 4-158
Equation 4-23: Total Emissions from PV Manufacturing 4-158
Equation 4-24: Estimation for SF6 Emissions from Electric Power Systems 4-170
Equation 4-25: Regression Equation for Estimating SF6 Emissions of Non-Reporting Facilities in 1999 4-172
Equation 4-26: Regression Equation for Estimating SF6 Emissions of GHGRP-Only Reporters in 2011 4-172
Equation 4-27: Total Emissions from Other Product Use 4-181
Equation 4-28: Total Emissions from Military Applications 4-181
Equation 4-29: Total Emissions from Scientific Applications 4-182
Equation 4-30: N2O Emissions from Product Use 4-184
Equation 5-1: Elemental C or N Released through Oxidation of Crop Residues 5-55
Equation 5-2: Emissions from Crop Residue Burning 5-56
Equation 5-3: Estimation of Greenhouse Gas Emissions from Fire 5-56
Equation 5-4: Estimation of Greenhouse Gas Emissions from Fire 5-57
Equation 6-1: Net State Annual Carbon Sequestration 6-174
Equation 6-2: Total Carbon Stock for Yard Trimmings and Food Scraps in Landfills 6-184
Equation 6-3: Carbon Stock Annual Flux for Yard Trimmings and Food Scraps in Landfills 6-184
Equation 7-1: Landfill Methane Generation 7-8
Equation 7-2: Net Methane Emissions from MSW Landfills 7-8
Equation 7-3: Net Methane Emissions from Industrial Waste Landfills 7-10
Equation 7-4: Total Domestic CFU Emissions from Wastewater Treatment and Discharge 7-23
Equation 7-5: CFU Emissions from Septic Systems 7-24
Equation 7-6: Total Wastewater BOD5 Produced per Capita (U.S.-Specific [ERG 2018a]) 7-24
Equation 7-7: Total Organically Degradable Material in Domestic Wastewater (IPCC 2019 [Eq. 6.3]) 7-25
Equation 7-8: Total Domestic CFU Emissions from Centrally Treated Aerobic Systems 7-26
Equation 7-9: Total Organics in Centralized Wastewater Treatment [IPCC 2019 (Eq. 6.3A)] 7-26
Equation 7-10: Organic Component Removed from Aerobic Wastewater Treatment (IPCC 2019 [Eq. 6.3B]) 7-27
Equation 7-11: CFU Emissions from Centrally Treated Aerobic Systems (other than Constructed Wetlands) (IPCC
2019 [Eq. 6.1]) 7-27
xxxiii
-------
Equation 7-12: Cm Emissions from Centrally Treated Aerobic Systems (Constructed Wetlands Only) [IPCC 2014 (Eq.
6.1)] 7-28
Equation 7-13: Cm Emissions from Centrally Treated Aerobic Systems (Constructed Wetlands used as Tertiary
Treatment) (U.S. Specific) 7-28
Equation 7-14: Cm Emissions from Centrally Treated Anaerobic Systems (IPCC 2019 (Eq. 6.1]) 7-29
Equation 7-15: Cm Emissions from Anaerobic Sludge Digesters (U.S. Specific) 7-30
Equation 7-16: Cm Emissions from Centrally Treated Systems Discharge (U.S.-Specific) 7-31
Equation 7-17: Total Organics in Centralized Treatment Effluent (IPCC 2019 [Eq. 6.3D]) 7-31
Equation 7-18: Total Organics in Effluent Discharged to Reservoirs, Lakes, or Estuaries (U.S.-Specific) 7-31
Equation 7-19: Total Organics in Effluent Discharged to Other Waterbodies (U.S.-Specific) 7-31
Equation 7-20: Total Cm Emissions from Industrial Wastewater 7-33
Equation 7-21: TOW in Industry Wastewater Treatment Systems 7-33
Equation 7-22: Organic Component Removed from Aerobic Wastewater Treatment - Pulp, Paper, and Paperboard
7-34
Equation 7-23: Organic Component Removed from Aerobic Treatment Plants 7-34
Equation 7-24: Raw Sludge Removed from Wastewater Treatment as Dry Mass 7-34
Equation 7-25: Cm Emissions from Industrial Wastewater Treatment Discharge 7-36
Equation 7-26: TOW in Industrial Wastewater Effluent 7-37
Equation 7-27: Cm Emissions from Pulp and Paper Discharge (U.S. Specific) 7-38
Equation 7-28: Total Organics in Pulp and Paper Effluent Discharged to Reservoirs, Lakes, Or Estuaries (U.S.
Specific) 7-38
Equation 7-29: Total Organics in Pulp and Paper Effluent Discharged to Other Waterbodies (U.S. Specific) 7-38
Equation 7-30: Total Domestic N2O Emissions from Wastewater Treatment and Discharge 7-41
Equation 7-31: Annual per Capita Protein Supply (U.S. Specific) 7-42
Equation 7-32: Consumed Protein (IPCC 2019 [Eq. 6.10A]) 7-42
Equation 7-33: Total Nitrogen Entering Septic Systems (IPCC 2019 [Eq. 6.10]) 7-42
Equation 7-34: N2O Emissions from Septic Systems (IPCC 2019 [Eq. 6.9]) 7-43
Equation 7-35: Total Nitrogen Entering Centralized Systems (IPCC 2019 [Eq. 10]) 7-44
Equation 7-36: Total Domestic N2O Emissions from Centrally Treated Aerobic Systems 7-45
Equation 7-37: N2O Emissions from Centrally Treated Aerobic Systems (other than Constructed Wetlands) (IPCC
2019 [Eq. 6.9]) 7-45
Equation 7-38: N2O Emissions from Centrally Treated Aerobic Systems (Constructed Wetlands Only) (IPCC 2014 [Eq.
6.9]) 7-46
Equation 7-39: N2O Emissions from Centrally Treated Aerobic Systems (Constructed Wetlands used as Tertiary
Treatment) (U.S.-Specific) 7-46
Equation 7-40: N2O Emissions from Centrally Treated Anaerobic Systems (IPCC 2019 [Eq. 6.9]) 7-47
Equation 7-41: N2O Emissions from Centrally Treated Systems Discharge (U.S.-Specific) 7-48
Equation 7-42: Total Organics in Centralized Treatment Effluent (IPCC 2019 [Eq. 6.8]) 7-48
xxxiv Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
Equation 7-43: Total Nitrogen in Effluent Discharged to Impaired Waterbodies (U.S.-Specific) 7-48
Equation 7-44: Total Nitrogen in Effluent Discharged to Nonimpaired Waterbodies (U.S.-Specific) 7-48
Equation 7-45: Total Nitrogen in Industrial Wastewater 7-50
Equation 7-46: N2O Emissions from Industrial Wastewater Treatment Plants 7-50
Equation 7-47: N2O Emissions from Industrial Wastewater Treatment Effluent 7-51
Equation 7-48: Greenhouse Gas Emission Calculation for Composting 7-56
Equation 7-49: Methane Emissions Calculation for Anaerobic Digestion 7-60
Equation 7-50: Weighted Average of Waste Processed 7-61
xxxv
-------
Executive Summary
An inventory that identifies and quantifies a country's anthropogenic1 sources and sinks of greenhouse gas
emissions and removals is essential for addressing climate change. This Inventory adheres to both (1) a
comprehensive and detailed set of methodologies for estimating national sources and sinks of anthropogenic
greenhouse gases, and (2) a common and consistent format that enables Parties to the Paris Agreement and
United Nations Framework Convention on Climate Change (UNFCCC) to compare the relative contribution of
different greenhouse gases emissions and removals to climate change.
The United States is party to both the 1992 UNFCCC and the 2015 Paris Agreement. The Paris Agreement set a
global temperature goal - holding the increase in the global average temperature to well below 2°C above pre-
industrial levels and pursuing efforts to limit the increase to 1.5°C - that articulates with greater precision States'
views on what is necessary to meet the UNFCCC's objective of "stabilizing]... greenhouse gas concentrations in
the atmosphere at a level that would prevent dangerous anthropogenic interference with the climate system.2
The United States is committed to submitting a national inventory of anthropogenic emission sources and
removals by sinks of greenhouse gases by April 15 of each year. The United States has prepared this report, in
conjunction with Common Reporting Tables (CRTs) that accompany this report, consistent with its obligations
under those agreements.
This Executive Summary provides the latest information on U.S. anthropogenic greenhouse gas emission trends
from 1990 through 2022. The structure of this report is consistent with requirements under the Paris Agreement
and the UNFCCC on national greenhouse gas inventory reporting, as discussed in Box ES-1. Throughout this report,
emission and sink estimates are grouped into five reporting sectors (i.e., chapters): Energy, IPPU, Agriculture, Land
Use, Land-Use Change, and Forestry (LULUCF), and Waste. In describing trends (Chapter 2), emissions and sinks are
also summarized according to commonly used economic sector categories: residential, commercial, industry,
transportation, electric power, and agriculture. See Box 2-1 for more information on how economic sectors are
defined.
Box ES-1: Methodological Approach for Estimating and Reporting U.S. Emissions and
Removals, including Relationship to EPA's Greenhouse Gas Reporting Program
Consistent with Article 13.7(a) of the Paris Agreement and Article 4.1(a) of the UNFCCC as well as relevant
decisions under those agreements, the emissions and removals presented in this report and this chapter are
organized by source and sink categories and calculated using internationally accepted methods in the 2006IPCC
Guidelines for National Greenhouse Gas Inventories (2006 IPCC Guidelines) and, where appropriate, its
supplements and refinements. Additionally, the calculated emissions and removals in a given year for the United
States are presented in a common manner in line with the reporting guidelines for the reporting of inventories
1 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 See Paris Agreement, Article 2.1(a); UNFCCC, Article 2.
Executive Summary ES-1
-------
under the Paris Agreement and the UNFCCC. The Parties' use of consistent methods to calculate emissions and
removals for their inventories helps to ensure that these reports are comparable. The presentation of emissions
and removals provided in this Inventory does not preclude alternative examinations (e.g., economic sectors).
Rather, this Inventory presents emissions and removals in a common format consistent with how Parties are to
report their national inventories under the Paris Agreement and the UNFCCC. The report itself, and this chapter,
follows this common format, and provides an explanation of the application of methods used to calculate
emissions and removals.
EPA also collects greenhouse gas emissions data from individual facilities and suppliers of certain fossil fuels and
industrial gases through its Greenhouse Gas Reporting Program (GHGRP), which is complementary to the U.S.
Inventory. ; The GHGRP applies to direct greenhouse gas emitters, fossil fuel suppliers, industrial greenhouse gas
suppliers, and facilities that inject carbon dioxide (CO ) underground for sequestration or other reasons and
requires reporting by over 8,000 sources or suppliers in 41 industrial categories.1 Annual 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 CO Eq. per year. Facilities in most source categories subject to
GHGRP began reporting for the 2010 reporting year while additional types of industrial operations began
reporting for reporting year 2011.'' Methodologies used in EPA's GHGRP are consistent with the 2006 IPCC
Guidelines. While the GHGRP does not provide full coverage of total annual U.S. greenhouse gas emissions and
removals (e.g., the GHGRP excludes emissions from the agricultural, land use, and forestry sectors), it is an
important input to the calculations of national-level emissions in this Inventory.
The GHGRP dataset provides not only annual emissions information, but also other annual information such as
activity data and emission factors that can improve and refine national emission estimates over time. GHGRP
data also allow EPA to disaggregate national inventory estimates in new ways that can highlight differences
across regions and subcategories of emissions, along with enhancing the application of QA/QC procedures and
assessment of uncertainties. See Annex 9 for more information on specific uses of GHGRP data in the Inventory
(e.g., use of Subpart W data in compiling estimates for natural gas systems).
ES.l Background Information
Greenhouse gases absorb infrared radiation, trapping heat in the atmosphere and making the planet warmer. The
most important greenhouse gases directly emitted by human activities include carbon dioxide (CO2), methane
(CH4), nitrous oxide (N2O), and several fluorine-containing halogenated substances (HFCs, PFCs, SF6 and NF3).
Although CO2, CH4, and N2O occur naturally in the atmosphere, human activities have changed their atmospheric
concentrations. From the pre-industrial era (i.e., ending about 1750) to 2022, concentrations of these greenhouse
gases have increased globally by 49.5,161.9, and 24.3 percent, respectively (IPCC 2013; NOAA/ESRL 2024a, 2024b,
2024c). This annual report estimates the total national greenhouse gas emissions and removals associated with
human activities across the United States.
3 On October 30, 2009, EPA promulgated a rule requiring annual reporting of greenhouse gas data from large greenhouse gas
emissions sources in the United States. Implementation of the rule, codified at 40 CFR Part 98, is referred to as EPA's
Greenhouse Gas Reporting Program (GHGRP).
4 See http://www.epa.gov/ehereportine and http://ehgdata.epa.gov/ehgp/main.do.
5 See https://eedsupport.eom/eonflu8nee/pages/yi8wpaee.aetion7pag8fe322699300.
ES-2 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
Global Warming Potentials
The IPCC developed the global warming potential (GWP) concept to compare the ability of a greenhouse gas to
trap heat in the atmosphere relative to another gas. A GWP is a quantified measure of the relative globally
averaged radiative forcing impacts of emissions of a particular greenhouse gas over time. The GWP of a
greenhouse gas is defined as the ratio of the accumulated radiative forcing within a specific time horizon caused by
emitting 1 kilogram of the gas, relative to that of the reference gas CO2 (IPCC 2021); therefore, CCh-equivalent
emissions are provided in million metric tons of CO2 equivalent (MMT CO2 Eq.) for non-CC>2 greenhouse gases.6,7
All estimates are provided throughout the main report in both CO2 equivalents and unweighted units, while
estimates for all gases in this Executive Summary are presented in units of MMT CO2 Eq. Emissions by gas in
unweighted mass kilotons are also provided in the Trends chapter and individual sector chapters of this report, and
in the CRTs that are included in the submission to the UNFCCC.
Recent decisions under the Paris Agreement8 and the UNFCCC9 require Parties to use 100-year GWP values from
the IPCC Fifth Assessment Report (AR5) for calculating CC>2-equivalents in their national reporting (IPCC 2013) by
the end of 2024. This reflects updated science and ensures that national greenhouse gas inventories reported by
all nations are comparable. This report reflects CC>2-equivalent greenhouse gas emission totals using 100-year AR5
GWP values. A comparison of emission values with the IPCC Sixth Assessment Report (AR6) (IPCC 2021) values can
be found in Annex 6.1 of this report. The 100-year GWP values used in this report are listed below in Table ES-1.
Table ES-1: Global Warming Potentials (100-Year Time Horizon) Used in this Report
Gas GWP
C02 1
CH4a 28
N20 265
HFCs up to 12,400
PFCs up to 11,100
SF6 23,500
NFs 16,100
Other Fluorinated Gases See Annex 6
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 production of
C02 is not included. See Annex 6 for additional information.
Source: IPCC (2013).
6 Carbon comprises 12/44 of carbon dioxide by weight.
7 One million metric ton is equal to 1012 grams or one teragram.
8 See Annex to decision 18/CMA.l, available online at https://unfccc.int/sites/default/files/resource/CMA2018 03a02E.pdf.
9 See paragraphs 1 and 2 of the decision on common metrics adopted at the 27th UNFCCC Conference of Parties (COP27),
available online at https://unfccc.int/sites/default/files/resource/cp2022 IQaOl E.pdf.
Executive Summary ES-3
-------
ES.2 Recent Trends in U.S. Greenhouse Gas
Emissions and Sinks
In 2022, total gross U.S. greenhouse gas emissions were 6,343.2 million metric tons of carbon dioxide equivalent
(MMT CO2 Eq.).10Total gross U.S. emissions decreased by 3.0 percent from 1990 to 2022, down from a high of 15.2
percent above 1990 levels in 2007. Gross emissions increased from 2021 to 2022 by 0.2 percent (14.4 MMT CO2
Eq.). Net emissions (including sinks) were 5,489.0 MMT CO2 Eq. in 2022. Overall, net emissions increased by 1.3
percent from 2021 to 2022 and decreased by 16.7 percent from 2005 levels as shown in Table ES-2. Between 2021
and 2022, the increase in total greenhouse gas emissions was driven largely by an increase in CO2 emissions from
fossil fuel combustion across most end-use sectors due in part to increased energy use from the continued
rebound of economic activity after the height of the COVID-19 pandemic. In 2022, CO2 emissions from fossil fuel
combustion increased by 1.0 percent relative to the previous year and were 1.1 percent below emissions in 1990.
Carbon dioxide emissions from natural gas use increased by 5.2 percent (84.8 MMT CO2 Eq.) from 2021, while CO2
emissions from coal consumption decreased by 6.1 percent (58.6 MMT CO2 Eq.) from 2021 to 2022. The increase in
natural gas consumption and associated emissions in 2022 is observed across all sectors except U.S. Territories,
while the coal decrease is due to reduced use in the electric power sector. Emissions from petroleum use also
increased by 0.9 percent (19.0 MMT CO2 Eq.) from 2021 to 2022. Carbon sequestration from the Land Use, Land-
Use Change, and Forestry (LULUCF) sector offset 14.5 percent of total emissions in 2022.
Figure ES-1 and Figure ES-2 illustrate the overall trends in total U.S. emissions by gas and annual percent changes,
and Table ES-2 provides information on trends in gross U.S. greenhouse gas emissions and sinks for 1990 through
2022. Unless otherwise stated, all tables and figures provide total gross emissions and exclude the greenhouse gas
fluxes from the LULUCF sector. For more information about the LULUCF sector, see Section ES-3.
10 The gross emissions total presented in this report for the United States excludes emissions and removals from Land Use,
Land-Use Change, and Forestry (LULUCF). The net emissions total presented in this report for the United States includes
emissions and removals from LULUCF.
ES-4 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
Figure ES-1: U.S. Greenhouse Gas Emissions and Sinks by Gas
¦ HFCs, PFCs, SFe and NF3 ~ Net Emissions (including LULUCF sinks)
q nnn
y'uuu ¦ Nitrous Oxide
¦ Methane
8,000 ¦ Carbon Dioxide
¦ Net CO2 Flux from LULUCF*
7,000
6,000 - _ ' -
d- 5,000
LU
(N
8 4,000
2,000
1,000
0
-1,000
o--—irNro^rLnkDr^cocnoHrNroT
CT> CT> CTi C7> CT* CTi CTi CTi CT> CTt O O O O O
cricricri0^cr>cricricriO>i0^ooooo
rH 1—I 1—I rH rH 1—I 1—I 1—I 1—I rH fNj (N CNJ (~\J ("Nj
^rvoocr»Oi-ir\iro^-Lo^p^ooaiOi-irsi
o o o o '—1 i-H t-h t—1 1—1 i-H t—1 i-H y—1 i-H rsj r\i p\J
00000000000000000
fM(N(N(MfNJ(N(Nf\(NfM(N(NrMfNfM(N(N
a The term "flux" is used to describe the exchange of C02 to and from the atmosphere, with net flux being either positive or
negative depending on the overall balance. Removal and long-term storage of C02 from the atmosphere is also referred to as
"carbon sequestration."
Table ES-2: Recent Trends in U.S. Greenhouse Gas Emissions and Sinks (MMT CO2 Eq.)
Percent
Change
Since
Gas/Source
1990
2005
2018
2019
2020
2021
2022
1990
C02
5,131.6
6,126.9
5,362.2
5,234.5
4,689.0
5,017.2
5,053.0
-1.5%
CH4(excludes LULUCF sources)3
871.7
795.4
771.5
754.3
735.3
720.5
702.4
-19.4%
N20 (excludes LULUCF sources)3
408.2
419.2
439.5
416.4
391.2
398.2
389.7
-4.5%
HFCs
47.7
121.7
163.9
168.2
170.3
177.0
182.8
282.9%
PFCs
39.5
10.2
7.4
7.3
6.6
6.3
6.7
-83.1%
sf6
37.9
20.2
7.6
8.4
8.1
8.5
7.6
-80.0%
nf3
0.3
1.0
0.7
1.1
1.3
1.1
1.1
238.3%
Total Gross Emissions (Sources)3
6,536.9
7,494.6
6,752.7
6,590.1
6,001.8
6,328.8
6,343.2
-3.0%
LULUCF Emissions'5
58.0
68.9
62.8
58.0
68.4
72.9
67.6
16.5%
ch4
53.1
58.5
55.5
52.5
59.3
62.1
58.4
10.0%
n2o
4.8
10.3
7.3
5.5
9.1
10.7
9.1
88.3%
LULUCF Carbon Stock Changec
(1034.7)
(976.6)
(978.3)
(921.6)
(972.8)
(983.4)
(921.8)
-10.9%
LULUCF Sector Net Totald
(976.7)
(907.7)
(915.5)
(863.6)
(904.4)
(910.6)
(854.2)
-12.5%
Net Emissions (Sources and Sinks)
5,560.2
6,586.9
5,837.3
5,726.6
5,097.4
5,418.2
5,489.0
-1.3%
a Gross emissions totals do not include CH4 and N20 emissions from LULUCF. LULUCF CH4 and N20 emissions are included in
net emission totals.
Executive Summary ES-5
-------
b LULUCF emissions subtotal of CH4 and N20 are reported separately from gross emissions totals. LULUCF emissions include
the CH4 and N20 emissions reported for peatlands remaining peatlands, forest fires, drained organic soils, grassland fires,
and coastal wetlands remaining coastal wetlands; CH4 emissions from land converted to coastal wetlands, flooded land
remaining flooded land, and land converted to flooded land; and N20 emissions from forest soils and settlement soils.
c LULUCF carbon stock change is the net carbon stock change from the following categories: forest land remaining forest land,
land converted to forest land, cropland remaining cropland, land converted to cropland, grassland remaining grassland, land
converted to grassland, wetlands remaining wetlands, land converted to wetlands, settlements remaining settlements, and
land converted to settlements.
d The LULUCF sector net total is the net sum of all LULUCF CH4 and N20 emissions to the atmosphere plus LULUCF net carbon
stock changes.
Notes: Total (gross) are emissions presented without LULUCF. Net emissions are presented with LULUCF. Totals may not sum
due to independent rounding. Parentheses indicate negative values or sequestration.
Figure ES-2: Annual Percent Change in Gross U.S. Greenhouse Gas Emissions and Sinks
Relative to the Previous Year
Improvements and Recalculations Relative to the Previous
Inventory
Each year, some emission and removal estimates in the Inventory are recalculated and revised to incorporate
improved methods and/or data. The most common reason for recalculating U.S. greenhouse gas emission
estimates is to update recent historical data. Changes in historical data are generally the result of changes in data
supplied by other U.S. government agencies or organizations, as they continue to make refinements and
improvements. These improvements are implemented consistently across the previous Inventory's time series, as
necessary, (i.e., 1990 to 2021) to ensure that the trend is accurate.
Collectively, all methodological changes and historical data updates made in the current Inventory resulted in an
annual average decrease of 114.8 MMT CO2 Eq. (1.9 percent) for net emissions.
ES-6 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
Figure ES-3: Impacts of Recalculations on Net Emissions
Below are categories with methodological and data-related recalculations resulting in an average change of
greater than 2.0 MMT CO2 Eq. over the time series.
• Forest land remaining forest land: changes in forest carbon stocks (CO2)
• Land converted to grassland: changes in all ecosystem carbon stocks (CO2)
• Land converted to cropland: changes in all ecosystem carbon stocks (CO2)
• Grassland remaining grassland: changes in all ecosystem carbon stocks (CO2)
• Non-energy use of fuels (CO2)
• Land converted to settlements: changes in all ecosystem carbon stocks (CO2)
• Fluorochemical production (HFCs)
• Fossil fuel combustion (CO2)
• Cropland remaining cropland: changes in all ecosystem carbon stocks (CO2)
• Agricultural Soil Management (N2O)
• Petroleum Systems (CH4)
• Wetlands Remaining Wetlands: changes in soil carbon stocks in coastal wetlands (CO2)
In addition, the current Inventory includes new categories not included in the previous Inventory that improve
completeness of the national estimates: CO2 emissions from ceramics production and non-metallurgical magnesia
production within other process use of carbonates category, fluorinated gases from fluorochemical production
other than HCFC-22 within the fluorochemical production category, and managed forest land in Hawaii and several
U.S. Territories. This report also now includes SF6 and PFCs from product uses.
In each Inventory, the results of all methodological changes and historical data updates and the inclusion of new
sources and sink estimates are summarized in the Recalculations and Improvements chapter (Chapter 9). For more
detailed descriptions of each recalculation including references for data, please see the respective source or sink
category description(s) within the relevant report chapter (the Energy chapter [Chapter 3], the Industrial Processes
and Product Use [IPPU] chapter [Chapter 4], the Agriculture chapter [Chapter 5], the Land Use, Land-Use Change,
and Forestry [LULUCF] chapter [Chapter 6], and the Waste chapter [Chapter 7]).
Executive Summary ES-7
-------
Emissions and Sinks by Greenhouse Gas
Figure ES-4 illustrates the relative contribution of the greenhouse gases to total gross U.S. emissions in 2022,
weighted by GWP. The primary greenhouse gas emitted by human activities in the United States is CO2,
representing 79.7 percent of total greenhouse gas emissions. The largest source of CO2 and of overall greenhouse
gas emissions is fossil fuel combustion, primarily from transportation and power generation. Methane (CH4)
emissions account for 11.1 percent of emissions. The major sources of methane include enteric fermentation
associated with domestic livestock, natural gas systems, and decomposition of wastes in landfills. Agricultural soil
management, wastewater treatment, stationary sources of fuel combustion, and manure management are the
major sources of N2O emissions. Emissions of substitutes for ozone depleting substances are the primary
contributor to aggregate hydrofluorocarbon (HFC) emissions. Perfluorocarbon (PFC) emissions are primarily
attributable to electronics manufacturing, fluorochemical production, and primary aluminum production. Electrical
equipment systems account for most sulfur hexafluoride (SFs) emissions. The electronics industry and
fluorochemical production are the only sources of nitrogen trifluoride (NF3) emissions. U.S. greenhouse gas
emissions were partly offset by carbon (C) sequestration in forests, trees in urban areas, agricultural soils,
landfilled yard trimmings and food scraps, and coastal wetlands, which together offset 14.5 percent of gross total
emissions in 2022 (as reflected in Figure ES-1). The following sections describe each gas's contribution to total U.S.
greenhouse gas emissions in more detail.
Figure ES-4: 2022 Total Gross U.S. Greenhouse Gas Emissions by Gas (Percentages based on
MMT C02 Eq.)
3.1%
HFCs, PFCs, SFe and NF3
Note: Emissions and removals from Land Use, Land-Use Change, and Forestry are excluded from the figure above.
Carbon Dioxide Emissions
The global carbon cycle is made up of large carbon flows and reservoirs. Billions of tons of carbon in the form of
CO2 are absorbed by oceans and living biomass (i.e., sinks) and are emitted to the atmosphere annually through
natural processes (i.e., sources). When in equilibrium, global carbon fluxes among these various reservoirs are
roughly balanced.11
Since the Industrial Revolution (i.e., about 1750), global atmospheric concentrations of CO2 have risen 49.5 percent
(IPCC 2013; NOAA/ESRL 2024a), principally due to the combustion of fossil fuels for energy. Globally, an estimated
11 The term "flux" is used to describe the exchange of C02 to and from the atmosphere, with net flux being either positive or
negative depending on the overall balance. Removal and long-term storage of C02 from the atmosphere is also referred to as
"carbon sequestration."
ES-8 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
33,423 MMT of CO2 were added to the atmosphere through the combustion of fossil fuels in 2022, of which the
United States accounted for approximately 14.1 percent.12
Overall CO2 emissions have decreased by 1.5 percent since 1990 and increased by 0.7 percent since 2021,
consistent with trends in fuel combustion emissions. Within the United States, fossil fuel combustion accounted
for 93.0 percent of CO2 gross emissions in 2022. Nationally, the transportation sector was the largest emitter of
CO2 in 2022 followed by electric power generation. There are 27 additional sources of CO2 emissions included in
the Inventory (see [Table 2-1 in Trends]). Although not illustrated in Table ES-4, changes in land use and forestry
practices can also lead to net CO2 emissions (e.g., through conversion of forest land to agricultural or urban use) or
to a net sink for CO2 (e.g., through net additions to forest biomass). See more on these emissions and removals in
Table ES-4.
Figure ES-5: 2022 Sources of CO2 Emissions
Fossil Fuel Combustion
Non-Energy Use of Fuels
Cement Production
Iron and Steel Production
Other Industrial Processes
Natural Gas Systems
Petrochemical Production
Petroleum Systems
Ammonia Production
Incineration of Waste
Lime Production
Other Energy
Net Carbon Stock Change from LULUCF
-100 -75 -50 -25 0 25 50 75 100 125 150
Note: "Other Industrial Processes" includes emissions from aluminum production, carbide production and consumption,
carbon dioxide consumption, ferroalloy production, glass production, lead production, magnesium production, other
process uses of carbonates, phosphoric acid production, substitution of ozone depleting substances, soda ash production,
titanium dioxide production, urea consumption for non-agricultural purposes, and zinc production. "Other Energy" includes
emissions from abandoned oil and gas wells and coal mining.
Between 1990 and 2022, CO2 emissions from fossil fuel combustion decreased by 1.1 percent; emissions decreased
by 18.2 percent (1,044.7 MMT CO2 Eq.) from 2005 levels; and from 2021 to 2022, these emissions increased by 1.0
percent (45.1 MMT CO2 Eq.).
Historically, changes in emissions from fossil fuel combustion have been the driving factor affecting U.S. emission
trends. Important drivers include changes in demand for energy and a general decline in the overall carbon
intensity of fuels combusted for energy in recent years by non-transport sectors of the economy. Between 2019
and 2021, changes in economic activity and travel due to the COVID-19 pandemic and the subsequent recovery
had significant impacts on energy use and fossil fuel combustion emissions.
The five major fuel-consuming economic sectors are transportation, electric power, industrial, residential, and
commercial and are described below. Carbon dioxide emissions are produced by the electric power sector as fossil
fuel is consumed to provide electricity to one of the other four economic sectors, or "end-use" sectors. In Figure
12 Global C02 emissions from fossil fuel combustion were taken from International Energy Agency Global energy-related C02
emissions, 1990-2022 - Charts. Available at: https://www.iea.ore/reports/co2-emissions-in-2022 (IEA 2022).
|4,699
CO2 as a Portion of
All Emissions
1 CO2
1 CH4
1 N2O
HFCs, PFCs, SFe and NF3
-922
Executive Summary ES-9
-------
ES-6, electric power emissions have been distributed to each end-use sector on the basis of each sector's share of
aggregate electricity use (i.e., indirect fossil fuel combustion). Greenhouse gas emissions from the commercial and
residential and industry increase substantially when indirect emissions from electricity end-use are distributed,
due to the relatively large share of electricity use by buildings (e.g., heating, ventilation, and air conditioning;
lighting; and appliances) and use of electricity for powering industrial machinery.
Figure ES-6: 2022 End-Use Sector Emissions of CO2 from Fossil Fuel Combustion
Direct Fossil Fuel Combustion
Indirect Fossil Fuel Combustion
23
1,757
U.S. Territories
Commercial
Residential
Industrial
Transportation
Transportation End-Use Sector. Transportation activities accounted for 37.4 percent of U.S. CO2 emissions from
fossil fuel combustion in 2022, with the largest contributors being light-duty trucks (36.8 percent), followed by
medium- and heavy-duty trucks (23.0 percent) and passenger vehicles (20.6 percent). In terms of the overall trend
from 1990 to 2022, total transportation CO2 emissions increased due largely to increased demand for travel, which
was a result of a confluence of factors including population growth, economic growth, urban sprawl, and low fuel
prices during the beginning of this period. While an increased demand for travel has led to generally increasing CO2
emissions since 1990, improvements in average new vehicle fuel economy since 2005 have slowed the rate of
increase of CO2 emissions. In 2022, petroleum-based products supplied 94.3 percent of the energy consumed for
transportation, primarily from gasoline consumption in automobiles and other highway vehicles (51.9 percent).
Industrial End-Use Sector. Industrial CO2 emissions, resulting both directly from the combustion of fossil fuels13 and
indirectly from the generation of electricity that is used by industry, accounted for 26.3 percent of CO2 emissions
from fossil fuel combustion in 2022. Approximately 64.7 percent of these emissions resulted from direct fossil fuel
combustion to produce steam and/or heat for industrial processes. The remaining emissions resulted from the use
of electricity for motors, electric furnaces, ovens, lighting, and other applications. Total direct and indirect
emissions from the industrial sector have declined by 20.8 percent since 1990. This decline is due to structural
changes in the U.S. economy (i.e., shifts from a manufacturing-based to a service-based economy), fuel switching,
and efficiency improvements. From 2021 to 2022, total energy use in the industrial sector increased by 1.8 percent
due to an increase in total industrial production and manufacturing output.
Residential and Commercial End-Use Sectors. The residential and commercial end-use sectors accounted for 19.1
and 16.6 percent, respectively, of CO2 emissions from fossil fuel combustion in 2022 including indirect emissions
from electricity. The residential and commercial sectors relied heavily on electricity for meeting energy demands,
with 62.9 and 66.9 percent, respectively, of their emissions attributable to electricity use for building -related
activities such as lighting, heating, cooling, and operating appliances. The remaining emissions were due to the
consumption of natural gas and petroleum for heating and cooking. Total direct and indirect emissions from the
residential sector have decreased by 3.4 percent since 1990, and total direct and indirect emissions from the
commercial sector have increased by 2.1 percent since 1990. From 2021 to 2022, an increase in heating degree
13 This does not include fossil fuels used as feedstocks and reductants, which are reported under IPPU emissions.
ES-10 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
days (7.9 percent) increased energy demand for heating in the residential and commercial sectors; also, a 4.3
percent increase in cooling degree days compared to 2021 increased demand for air conditioning in the residential
and commercial sectors. Combined, this resulted in a 2.5 percent increase in residential sector energy use. From
2021 to 2022, energy use in the commercial sector increased by 4.7 percent.
Electric Power. The United States relies on electricity to meet a significant portion of its energy demands.
Electricity generators used 30.5 percent of U.S. energy from fossil fuels and emitted 32.6 percent of the CO2 from
fossil fuel combustion in 2022. The type of energy source used to generate electricity, and the mix of electric
generation resources used to meet demand, are the main factors influencing emissions.14 Coal-fired electric
generation (in kilowatt-hours [kWh]) decreased from 54.1 percent of generation in 1990 to 20.3 percent in 2022.15
This corresponded with an increase in natural gas generation and non-fossil fuel renewable energy generation,
largely from wind and solar energy. Natural gas generation (in kWh) represented 10.7 percent of electric power
generation in 1990 and increased over the 33-year period to represent 38.8 percent of electric power generation
in 2022. Wind and solar generation (in kWh) represented 0.1 percent of electric power generation in 1990 and
increased over the 33-year period to represent 14.2 percent of electric power generation in 2022. Between 2021
and 2022, coal electricity generation decreased by 10.2 percent, natural gas generation increased by 4.0 percent,
and renewable energy generation increased by 7.6 percent.
Across the time series, changes in electricity generation and the carbon intensity of fuels used for electric power
have a significant impact on CO2 emissions. While CO2 emissions from fossil fuel combustion in the electric power
sector have decreased by 15.8 percent since 1990, the carbon intensity of the electric power sector, in terms of
CO2 Eq. per QBtu input, decreased by 27.6 percent during that same timeframe. This decoupling of the level of
electric power generation and the resulting CO2 emissions is shown in Figure ES-7.
14 In line with the reporting requirements for inventories submitted under the UNFCCC, C02 emissions from biomass
combustion have been estimated separately from fossil fuel C02 emissions and are not included in the electricity sector totals
and trends discussed in this section. Net carbon fluxes from changes in biogenic carbon reservoirs are accounted for in the
estimates for Land Use, Land-Use Change, and Forestry.
15 Values represent electricity net generation from the electric power sector. See Table 7.2b Electricity Net Generation: Electric
Power Sector of EIA (2024).
Executive Summary ES-11
-------
Figure ES-7
: Electric Power Generation and Emissions
4,500
4,000
¦ 3,500
= 3,000
XI
E 2,500
crio^(jicriCTiffi(7ioicricr>
in n lo r->. co oV o r>j ro > in ud co cr« o
3,500
3,000
2,500
o
u
2,000
1,500 I
1,000
500
oooooooooo
IN (N N
ooooooooooooooooooooooo
r\lrsJrMr\lrMrNrslrMr\lrMrsJrNrN(NfMrJrMrMr\J
-------
Figure ES-8: 2022 Sources of CH4 Emissions
Enteric Fermentation
Natural Gas Systems
Landfills
Manure Management
Flooded Land
Coal Mining
Petroleum Systems
Wastewater Treatment
Rice Cultivation
Other Energy
Other LULUCF
Stationary Combustion
Other Waste
Field Burning of Agricultural Residues
Other Industrial Processes
193
Cl-U as a Portion of All
Emissions
11.1%
< 0.5
I CO2
I CH4
1 N2O
HFCs, PFCs, SFe and NFs
0
20
40
60
80 100 120
MMT CO2 Eq.
140
160
180
200
Note: "Other Energy" includes CH4 emissions from abandoned oil and gas wells, abandoned underground coal mines,
incineration of waste, and mobile combustion. "Other Waste" includes CH4 emissions from anaerobic digestion at biogas
facilities and composting. "Other Industrial Processes" includes CH4 emissions from carbide production and consumption,
ferroalloy production, iron and steel production and metallurgical coke production, and petrochemical production. "Other
LULUCF" includes the CH4 reported for peatlands remaining peatlands, forest fires, drained organic soils, grassland fires,
coastal wetlands remaining coastal wetlands, and land converted to coastal wetlands.
Overall, CH4 emissions in the United States in 2022, including LULUCF CH4 emissions, accounted for 760.8 MMT CO2
Eq., representing a decrease of 17.7 percent (164.0 MMT CO2 Eq.) since 1990 and 2.8 percent (21.8 MMT CO2 Eq.)
since 2021. Significant trends for the largest sources of anthropogenic CH4 emissions include the following:
• Enteric fermentation was the largest anthropogenic source of CH4 emissions in the United States in 2022,
accounting for 27.4 percent of total CH4 emissions and 3.0 percent of total gross emissions. Emissions
have increased by 5.2 percent (9.5 MMT CO2 Eq.) since 1990. This increase in emissions from 1990 to
2022 generally follows the increasing trends in cattle populations.
• Natural gas systems were the second largest anthropogenic source category of CH4 emissions in the
United States in 2022, accounting for 24.6 percent of total CH4 emissions and 2.7 percent of total gross
emissions. Emissions have decreased by 20.9 percent (45.7 MMT CO2 Eq.) since 1990, largely due to
decreases in emissions from distribution, transmission, and storage.
• Landfills were the third largest anthropogenic source of CH4 emissions in the United States in 2022,
accounting for 17.1 percent of total CH4 emissions and 1.9 percent of total gross emissions and
representing a decrease of 39.4 percent (78.0 MMT CO2 Eq.) since 1990, with small year-to-year increases.
This downward trend in emissions coincided with increased landfill gas collection and control systems,
and a reduction of decomposable materials (i.e., paper and paperboard, food scraps, and yard trimmings)
discarded in MSW landfills over the time series.16
Nitrous Oxide Emissions
Nitrous oxide (N2O) is produced by biological processes that occur in soil and water and by a variety of
anthropogenic activities in the agricultural, energy, industrial, and waste management fields. While total N2O
emissions are much lower than CO2 emissions, N2O is 265 times more powerful than CO2 at trapping heat in the
atmosphere over a 100-year time frame (IPCC 2013). Since 1750, the global atmospheric concentration of N2O has
16 Carbon dioxide emissions from landfills are not included specifically in summing waste sector totals. Net carbon fluxes from
changes in biogenic carbon reservoirs and decay of disposed wood products are accounted for in the estimates for LULUCF.
Executive Summary ES-13
-------
risen by 24.3 percent (IPCC 2013; NOAA/ESRL 2024c). The main anthropogenic activities producing N2O in the
United States are agricultural soil management, wastewater treatment, stationary fuel combustion, manure
management, fuel combustion in motor vehicles, and nitric acid production (see Figure ES-9).
Figure ES-9: 2022 Sources of N2O Emissions
Agricultural Soil Management
Stationary Combustion
Wastewater Treatment
Manure Management
Mobile Combustion
LULUCF Emissions
Nitric Acid Production
Other Industrial Processes
Adipic Acid Production
Composting
Other Energy
Field Burning of Agricultural Residues
0 5 10 15 20 25 30 35 40
MMT CO2 Eq.
Note: "Other Industrial Processes" includes N20 emissions from caprolactam, glyoxal, and glyoxylic acid production; the
electronics industry; and product uses. "Other Energy" includes N20 emissions from petroleum systems, natural gas systems,
and incineration of waste. LULUCF emissions include N20 emissions reported for peatlands remaining peatlands, forest fires,
drained organic soils, grassland fires, coastal wetlands remaining coastal wetlands, forest soils, and settlement soils.
Overall, N2O emissions in the United States in 2022, including LULUCF N2O emissions, accounted for 398.8 MMT
CO2 Eq., representing a decrease of 3.4 percent (14.2 MMT CO2 Eq.) since 1990 and a decrease of 2.5 percent (10.1
MMT CO2 Eq.) since 2021. Significant trends for the largest sources of anthropogenic N2O emissions include the
following:
• Agricultural soils were the largest anthropogenic source of N2O emissions in 2022, accounting for 74.6
percent of N2O emissions and 4.6 percent of total gross greenhouse gas emissions in the United States.
These emissions increased by 0.7 percent (2.0 MMT CO2 Eq.) from 1990 to 2022 but fluctuated during that
period due to annual variations in weather patterns, fertilizer use, and crop production.
• Stationary combustion was the second largest source of anthropogenic N2O emissions in 2022, accounting
for 6.3 percent of N2O emissions and 0.4 percent of total gross U.S. greenhouse gas emissions in 2022.
Stationary combustion emissions peaked in 2007 and steadily decreased until 2020. Emissions increased
in 2021 and 2022. Stationary combustion emissions have increased by 10.6 percent (2.4 MMT CO2 Eq.)
since 1990.
• Wastewater treatment, both domestic and industrial, was the third largest anthropogenic source of N2O
emissions in 2022, accounting for 5.6 percent of N2O emissions and 0.3 percent of total gross greenhouse
gas emissions in the United States in 2022. Emissions from wastewater treatment increased by 48.2
percent (7.1 MMT CO2 Eq.) since 1990 as a result of growing U.S. population and protein consumption.
ES-14 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
HFC, PFC, SFe, and NF3 Emissions
Hydrofluorocarbons (HFCs) are synthetic chemicals that are used as alternatives to ozone depleting substances
(ODS), which are being phased out under the Montreal Protocol and Clean Air Act Amendments of 1990.
Hydrofluorocarbons do not deplete the stratospheric ozone layer and therefore have been used as alternatives
under the Montreal Protocol.
Perfluorocarbons (PFCs) are emitted from the production of electronics and aluminum and also (in smaller
quantities) from their use as alternatives to ODS. Sulfur hexafluoride (SFe) is emitted from the manufacturing and
use of electrical equipment as well as the production of electronics and magnesium. NF3 is emitted from
electronics production. HFCs are also emitted during production of HCFC-22 and electronics (see Figure ES-10).
HFCs, PFCs, SFe, and NF3 are potent greenhouse gases. In addition to having very high GWPs, SF6, NF3, and PFCs
have extremely long atmospheric lifetimes, resulting in their essentially irreversible accumulation in the
atmosphere once emitted. Sulfur hexafluoride is the most potent greenhouse gas the IPCC has evaluated (IPCC
2021).
Figure ES-10: 2022 Sources of HFCs, PFCs, SF6, and NF3 Emissions
Substitution of Ozone Depleting Substances 178
Fluorochemical Production 3.1%
HFCs, PFCs, SFe, and NF3 as
Electrical Equipment . .
H K a Portion of All Emissions
Electronics Industry
Magnesium Production and Processing ¦ cq2
Other Product Manufacture and Use "
¦ N2O
Aluminum Production ¦ HFCs, PFCs, SFe and NF3
0 2 4 6 8 10 12 14 16 18 20
MMT CO2 Eq.
Some significant trends for the largest sources of U.S. HFC, PFC, SF6, and NF3 emissions include the following:
• Hydrofluorocarbon and perfluorocarbon emissions resulting from their use as substitutes for ODS (e.g.,
chlorofluorocarbons [CFCs]) are the largest share of fluorinated emissions (89.9 percent) in 2022 and have
been consistently increasing, from small amounts since 1990. This increase over the time series was
largely the result of efforts to phase out CFCs and other ODS in the United States.
• Sulfur hexafluoride emissions from electrical equipment decreased by 79.4 percent (19.6 MMT CO2 Eq.)
from 1990 to 2022. There are two factors contributing to this decrease: (1) a sharp increase in the price of
SFe during the 1990s and (2) a growing awareness of the environmental impact of SF6 emissions through
programs such as EPA's SF6 Emission Reduction Partnership for Electric Power Systems.
• HFC, PFC, SFe, and NF3 emissions from fluorochemical production decreased by 89.0 percent (63.2 MMT
CO2 Eq.) from 1990 to 2022 due to a reduction in the HFC-23 emission rate from HCFC-22 production (kg
HFC-23 emitted/kg HCFC-22 produced), the imposition of emissions controls at production facilities, and a
decrease in SF6 production (due to the cessation of production at the major SF6 production facility in
2010).
• PFC emissions from aluminum production decreased by 96.1 percent (18.5 MMT CO2 Eq.) from 1990 to
2022, due to both industry emission reduction efforts and lower domestic aluminum production.
Executive Summary ES-15
-------
ES.3 Overview of Sector Emissions and Trends
Figure ES-11 and Table ES-3 aggregate emissions and sinks by the sectors defined by the UNFCCC and Paris
Agreement reporting guidelines and methodological framework in the IPCC guidelines to promote comparability
across countries. Over the 33-year period of 1990 to 2022, total emissions from the Energy and Waste sectors
decreased by 3.4 percent (181.2 MMT CO2 Eq.) and 29.3 percent (69.1 MMT CO2 Eq.) respectively. Emissions from
the Industrial Processes and Product Use and Agriculture sectors grew by 3.9 percent (14.4 MMT CO2 Eq.), and 7.7
percent (42.2 MMT CO2 Eq.), respectively. Over the same period, the overall net flux from LULUCF (i.e., the net
sum of all Cm and N2O emissions to the atmosphere plus LULUCF net carbon stock changes in units of MMT CO2
Eq.) decreased by 12.5 percent (122.5 MMT CO2 Eq.) and resulted in a removal of 854.2 MMT CO2 Eq. in 2022.
Figure ES-11: U.S. Greenhouse Gas Emissions and Sinks by UNFCCC/IPCC Sector
¦ LULUCF (emissions) ¦ Agriculture
9,000 ¦ Waste ¦ Energy
¦ Industrial Processes and Product Use ¦ LULUCF (removals)
0 nnn — Net Emissions (including LULUCF sinks)
o,UUU
7,000
6,000
£ 5,000
(N
O
^ 4,000
2
3,000
H CM ro LD KD
(Jl (Ji (J) CTi 01
cn (ji cn ai en
Is-. 00
cr> cn
Ch CT>
) 1" 1 ( 1 O O O O O O rH t~H rH rH rH i-H rH rH rH rH fNj f\J
OOOOOOOOOOOOOOOOOOOOOOO
rNllN(N(N(NfM(NlNfMfM(N(N(NfMMfMfN(M(NfS(NfN(N
Table ES-3: Recent Trends in U.S. Greenhouse Gas Emissions and Sinks by UNFCCC/IPCC
Sector (MMT C02 Eq.)
Percent Change
UNFCCC/IPCC Sector
1990
2000
2018
2019
2020
2021
2022
Since 1990
Energy
5,381.0
6,349.5
5,570.0
5,422.4
4,862.6
5,173.3
5,199.8
-3.4%
Industrial Processes and Product Use
368.8
371.3
367.2
371.9
367.9
381.6
383.2
3.9%
Agriculture
551.1
581.8
642.4
620.1
599.7
604.8
593.4
7.7%
Waste
235.9
192.0
173.2
175.8
171.7
169.2
166.9
-29.3%
Total Gross Emissions3 (Sources)
6,536.9
7,494.6
6,752.7
6,590.1
6,001.8
6,328.8
6,343.2
-3.0%
LULUCF Sector Net Totalb
(976.7)
(907.6)
(915.5)
(863.6)
(904.4)
(910.5)
(854.2)
-12.5%
Net Emissions (Sources and Sinks)c
5,560.2
6,586.9
5,837.3
5,726.6
5,097.4
5,418.2
5,489.0
-1.3%
a Total emissions without LULUCF.
b The LULUCF sector net total is the sum of all LULUCF CH4 and N20 emissions to the atmosphere plus LULUCF net carbon stock
changes in units of MMT C02 Eq.
c Net emissions with LULUCF.
Notes: Totals may not sum due to independent rounding. Parentheses indicate negative values or sequestration.
ES-16 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
Energy
The Energy chapter contains emissions of all greenhouse gases resulting from stationary and mobile energy
activities including fuel combustion and fugitive fuel emissions, and the use of fossil fuels for non-energy purposes.
Energy-related activities, primarily fossil fuel combustion, accounted for the vast majority of U.S. CO2 emissions for
the period of 1990 through 2022. Energy-related activities are also responsible for CFU and N2O emissions (40.2
percent and 10.8 percent of total U.S. emissions of each gas, respectively).17 Overall, emission sources in the
Energy chapter account for a combined 82.0 percent of total gross U.S. greenhouse gas emissions in 2022.
Emissions from energy increased by 0.5 percent (26.5 MMT CO2 Eq.) since 2021, but they have decreased by 3.4
percent (181.2 MMT CO2 Eq.) since 1990.
In 2022, 83.0 percent of the energy used in the United States (on a Btu basis) was produced through the
combustion of fossil fuels. The remaining 17.0 percent came from other energy sources, such as hydropower,
biomass, nuclear, wind, and solar energy (see Figure ES-12).
Figure ES-12: 2022 U.S. Energy Consumption by Energy Source (Percent)
Nuclear Electric Power
Industrial Processes and Product Use
The Industrial Processes and Product Use (IPPU) chapter contains greenhouse gas emissions generated and
emitted as the byproducts of non-energy-related industrial processes, which involve the chemical or physical
transformation of raw materials and can release waste gases such as CO2, Cm, N2O, and fluorinated gases (e.g.,
HFC-23). These processes include iron and steel production and metallurgical coke production, cement production,
petrochemical production, ammonia production, lime production, other process uses of carbonates (e.g., other
uses of carbonates, other uses of soda ash not associated with glass manufacturing, ceramics production, and non-
metallurgical magnesia production), nitric acid production, adipic acid production, urea consumption for non-
agricultural purposes, aluminum production, HCFC-22 production, other fluorochemical production, glass
production, soda ash production, ferroalloy production, titanium dioxide production, caprolactam production, zinc
production, phosphoric acid production, lead production, and silicon carbide production and consumption. Most of
these industries also emit CO2 from fossil fuel combustion which, in line with UNFCCC/IPCC sectoral definitions, is
included in the Energy sector.
17 The contribution of energy non-C02 emissions is based on gross totals and excludes LULUCF methane (CH4) and nitrous oxide
(N20) emissions. The contribution of energy-related CH4 and N20 including LULUCF non-C02 emissions is 37.1 percent and 9.8
percent respectively.
Executive Summary ES-17
-------
This chapter also contains emissions resulting from the release of HFCs, PFCs, SF6, and NF3 and other man-made
compounds used in industrial manufacturing processes and by end-consumers (e.g., residential and mobile air
conditioning). These industries include electronics manufacturing, electric power transmission and distribution,
and magnesium metal production and processing. In addition, N2O is used in and emitted by electronics industry
and anesthetic and aerosol applications, PFCs and SF6 are emitted in other product use, and CO2 is consumed and
emitted through various end-use applications. In 2022, emissions resulting from use of the substitution of ODS
(e.g., chlorofluorocarbons [CFCs]) by end-consumers was the largest source of IPPU emissions and accounted for
46.5 percent of total IPPU emissions.
IPPU activities are responsible for 3.3, less than 0.5, and 4.1 percent of total U.S. CO2, CH4, and N2O emissions
respectively as well as for all U.S. emissions of fluorinated gases including HFCs, PFCs, SF6 and NF3. Overall,
emission sources in the IPPU chapter accounted for 6.0 percent of U.S. greenhouse gas emissions in 2022. IPPU
emissions have increased by 0.4percent (1.6 MMT CO2 Eq.) since 2021 and by 3.9 percent (14.4 MMT CO2 Eq.)
since 1990, mostly due to increased use of ODS substitutes (e.g., HFCs).
Agriculture
The Agriculture chapter contains information on anthropogenic emissions from agricultural activities (except fuel
combustion, which is addressed in the Energy chapter, and some agricultural CO2, CH4, and N2O fluxes, which are
addressed in the Land Use, Land-Use Change, and Forestry chapter).
Several agricultural activities contribute directly to emissions of greenhouse gases including the following sources:
agricultural soil management, enteric fermentation in domestic livestock, livestock manure management, rice
cultivation, urea fertilization, liming, and field burning of agricultural residues.
In 2022, agricultural activities were responsible for 9.4 percent of total gross U.S. greenhouse gas emissions.
Agriculture sector emissions decreased by 11.4 MMT CO2 Eq. (1.9 percent) since 2021 and have increased by 42.2
MMT CO2 Eq. (7.7 percent) since 1990, mostly from trends in enteric fermentation and manure management.
Methane, N2O, and CO2 are greenhouse gases emitted by agricultural activities. Methane emissions from enteric
fermentation and manure management represented 36.6 percent of total CH4 emissions from anthropogenic
activities in 2022. Agricultural soil management activities, such as application of synthetic and organic fertilizers,
deposition of livestock manure, and growing N-fixing plants, were the largest contributors to U.S. N2O emissions in
2022, accounting for 74.6 percent of total N2O emissions. Carbon dioxide emissions from the application of
crushed limestone and dolomite (i.e., soil liming) and urea fertilization represented 0.2 percent of total CO2
emissions from anthropogenic activities.
Land Use, Land-Use Change, and Forestry
The LULUCF chapter contains emissions and removals of CO2 and emissions of CH4 and N2O from managed lands in
the United States. Consistent with the 2006IPCC Guidelines, emissions and removals from managed lands are
considered to be anthropogenic, while emissions and removals from unmanaged lands are considered to be
natural.18 The share of managed land in the United States is approximately 95 percent of total land included in the
Inventory.19 More information on the definition of managed land used in the Inventory is provided in Chapter 6.
Overall, the Inventory results show that managed land is a net sink for CO2 (C sequestration). The primary drivers
of fluxes on managed lands include forest management practices, tree planting in urban areas, the management of
agricultural soils, lands remaining and lands converted to reservoirs and other constructed waterbodies, landfilling
18 See http://www.ipcc-negip.iges.or.ip/public/2006el/pdf/4 Volume4/V4 01 Chi lntroduction.pdf.
19 The current land representation does not include land in U.S. Territories, but there are planned improvements to include
these regions in future Inventories. U.S. Territories represent approximately 0.1 percent of the total land base for the United
States. See Box 6-2 in Chapter 6 of this report.
ES-18 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
of yard trimmings and food scraps, and activities that cause changes in carbon stocks in coastal wetlands. The main
drivers for forest carbon sequestration include forest growth and increasing forest area (i.e., afforestation), as well
as a net accumulation of carbon stocks in harvested wood pools. The net sequestration in settlements remaining
settlements, which occurs predominantly from urban forests (i.e., settlement trees) and landfilled yard trimmings
and food scraps, is a result of net tree growth and increased urban forest area, as well as long-term accumulation
of carbon from yard trimmings and food scraps in landfills.
The LULUCF sector in 2022 resulted in a net increase in carbon stocks (i.e., net CO2 removals) of 921.8 CO2 Eq.20
The removals of carbon offset 14.5 percent of total gross greenhouse gas emissions in 2022. Emissions of CH4 and
N2O from LULUCF activities in 2022 represented 1.2 percent of net greenhouse gas emissions.21 Carbon dioxide
removals from carbon stock changes are presented in Table ES-4 along with CH4 and N2O emissions for LULUCF
source categories.
Between 1990 and 2022, total carbon sequestration in the LULUCF sector decreased by 10.9 percent, primarily due
to a decrease in the rate of net carbon accumulation in forests and in cropland remaining cropland, as well as an
increase in CO2 emissions from land converted to settlements. The overall net flux from LULUCF (i.e., net sum of all
Cm and N2O emissions to the atmosphere plus LULUCF net carbon stock changes in units of MMT CO2 Eq.) resulted
in a removal of 854.2 MMT C02 Eq. in 2022.
Flooded lands were the largest source of CH4 emissions from the LULUCF sector and the fifth largest source overall
net CH4 emissions in 2022. Forest fires were the second largest source of CH4 emissions, followed by coastal
wetlands remaining coastal wetlands. Forest fires were the largest source of N2O emissions from the LULUCF
sector in 2022.
Table ES-4: U.S. Greenhouse Gas Emissions and Removals (Net Flux) from Land Use, Land-Use
Change, and Forestry (MMT CO2 Eq.)
Land-Use Category
1990
2005
2018
2019
2020
2021
2022
Forest Land Remaining Forest Land3
(968.8)
(860.1) 1
(863.4)
(807.0)
(846.3)
(823.9)
(771.7)
Land Converted to Forest Landb
(100.2) 1
(100.2) i
(100.4)
(100.3)
(100.3)
(100.3)
(100.3)
Cropland Remaining Cropland
(5.0)
(31.6)
(17.8)
(19.4)
(8.8)
(32.0)
(31.7)
Land Converted to Cropland0
45.4 1
34.5 1:
31.9
31.4
29.3
34.9
35.1
Grassland Remaining Grasslandd
24.6
24.9
29.7
28.9
17.1
11.5
14.0
Land Converted to Grassland0
35.3 1
21.8 i:
25.2
25.4
28.7
24.5
25.6
Wetlands Remaining Wetlandse
36.8
39.4
38.2
38.1
38.1
38.1
38.1
Land Converted to Wetlandse
7.2
1.8
0.7
0.7
0.7
0.7
0.7
Settlements Remaining Settlements'
(109.1)
(115.2)
(131.0)
(131.5)
(131.8)
(132.3)
(132.3)
Land Converted to Settlements0
57.2 1
77.11
71.4
70.2
68.8
68.2
68.2
LULUCF Carbon Stock Change^
(1,034.7)
(976.6)
(978.3)
(921.6)
(972.8)
(983.4)
(921.8)
LULUCF Emissions'1
58.0
68.9
62.8
58.0
68.4
72.9
67.6
ch4
53.1
58.5
55.5
52.5
59.3
62.1
58.4
n2o
4.8 1
10.3 ¦
7.3
5.5
9.1
10.7
9.1
LULUCF Sector Net Total'
(976.7)
(907.7)
(915.5)
(863.6)
(904.4)
(910.6)
(854.2)
a Includes the net changes to carbon stocks stored in all forest ecosystem pools and harvested wood products,
emissions from fires on both forest land remaining forest land and land converted to forest land, emissions from N
fertilizer additions on both forest land remaining forest land and land converted to forest land, and CH4 and N20
emissions from drained organic soils on both forest land remaining forest land and land converted to forest land.
20 LULUCF carbon stock change is the net C stock change from the following categories: forest land remaining forest land, land
converted to forest land, cropland remaining cropland, land converted to cropland, grassland remaining grassland, land
converted to grassland, wetlands remaining wetlands, land converted to wetlands, settlements remaining settlements, and
land converted to settlements.
21 LULUCF emissions include the CH4 and N20 emissions reported for peatlands remaining peatlands, forest fires, drained
organic soils, grassland fires, and coastal wetlands remaining coastal wetlands; CH4 emissions from land converted to coastal
wetlands; and N20 emissions from forest soils and settlement soils.
Executive Summary ES-19
-------
b Includes the net changes to carbon stocks stored in all forest ecosystem pools.
c Includes changes in mineral and organic soil carbon stocks for all land use conversions to cropland, grassland, and
settlements, respectively. Also includes aboveground/belowground biomass, dead wood, and litter carbon stock
changes for conversion of forest land to cropland, grassland, and settlements, respectively.
d Estimates include CH4 and N20 emissions from fires on both grassland remaining grassland and land converted to
grassland.
e Estimates include CH4 emissions from flooded land remaining flooded land and land converted to flooded land.
f Estimates include N20 emissions from N fertilizer additions on both settlements remaining settlements and land
converted to settlements because it is not possible to separate the activity data at this time,
s LULUCF carbon stock change includes any carbon stock gains and losses from all land use and land use conversion
categories.
h LULUCF emissions subtotal includes the CH4 and N20 emissions reported for peatlands remaining peatlands, forest
fires, drained organic soils, grassland fires, and coastal wetlands remaining coastal wetlands; CH4 emissions from land
converted to coastal wetlands, flooded land remaining flooded land, and land converted to flooded land; and N20
emissions from forest soils and settlement soils. Emissions values are included in land-use category rows.
' The LULUCF sector net total is the net sum of all LULUCF CH4 and N20 emissions to the atmosphere plus LULUCF net
carbon stock changes in units of MMT C02 Eq.
Notes: Totals may not sum due to independent rounding. Parentheses indicate net sequestration.
Waste
The Waste chapter contains emissions from waste management activities (except the incineration of waste, which
is addressed in the Energy chapter). Landfills were the largest source of anthropogenic greenhouse gas emissions
from waste management activities, accounting for 71.8 percent of total greenhouse gas emissions from waste
management activities, and 17.1 percent of total U.S. CH4 emissions.22 Additionally, wastewater treatment
accounted for 25.6 percent of total Waste sector greenhouse gas emissions, 3.0 percent of U.S. CH4 emissions, and
5.6 percent of U.S. N2O emissions in 2022. Emissions of CH4 and N2O from commercial composting are also
included in this chapter, accounting for 1.5 percent (2.6 MMT CO2 Eq.) and 1.1 percent (1.8 MMT CO2 Eq.) of
overall waste sector emissions, respectively. Anaerobic digestion at biogas facilities generated CH4 emissions,
accounting for less than 0.05 percent of emissions from the Waste sector. Overall, emission sources in the Waste
chapter accounted for 2.6 percent of total gross U.S. greenhouse gas emissions in 2022. Waste sector emissions
decreased by 1.4 percent (2.3 MMT CO2 Eq.) since 2021 and by 29.3 percent (69.1 MMT CO2 Eq.) since 1990.
ES.4 Other Information
Emissions and Sinks by Economic Sector
In addition to the Paris Agreement and UNFCCC sectors and methods defined by the IPCC, this report also
characterizes emissions according to commonly used economic sector categories: residential, commercial,
industry, transportation, electric power, and agriculture.23 Emissions from U.S. Territories are reported as their
own end-use sector due to a lack of specific consumption data for the individual end-use sectors within U.S.
Territories. For more information on trends in the Land Use, Land-Use Change, and Forestry sector, see Section
ES.2 Recent Trends in U.S. Greenhouse Gas Emissions and Sinks.
22 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. Also,
the estimated total methane emissions used to estimate contribution excludes methane emissions from the LULUCF sector.
23 The agriculture economic sector includes emissions from fossil fuel combustion and electricity use within the agricultural
sector.
ES-20 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
Figure ES-13 shows the trend in emissions by economic sector from 1990 to 2022, and Table ES-5 summarizes
emissions from each of these economic sectors.
Figure ES-13: U.S. Greenhouse Gas Emissions Allocated to Economic Sectors
Note: Emissions and removals from Land Use, Land-Use Change, and Forestry are excluded from figure above. Excludes U.S.
Territories.
Table ES-5: U.S. Greenhouse Gas Emissions Allocated to Economic Sectors (MMT CO2 Eq.)
Percent
Change
Since
Economic Sectors
1990
2005
2018
2019
2020
2021
2022
1990
Transportation
1,521.4
1,965.9
1,871.6
1,874.6
1,625.3
1,805.5
1,801.5
18.4%
Electric Power Industry
1,880.2
2,457.4
1,799.2
1,650.8
1,482.2
1,584.4
1,577.5
-16.1%
Industry
1,723.3
1,587.3
1,541.9
1,531.8
1,435.9
1,455.8
1,452.5
-15.7%
Agriculture
595.9
634.3
683.5
661.0
640.0
645.9
634.0
6.4%
Commercial
447.0
418.9
453.5
462.6
436.9
443.7
463.7
3.7%
Residential
345.6
371.2
376.8
384.2
358.0
369.6
391.3
13.2%
U.S. Territories
23.4
59.7
26.3
25.1
23.4
23.9
22.7
-3.1%
Total Gross Emissions (Sources)
6,536.9
7,494.6
6,752.7
6,590.1
6,001.8
6,328.8
6,343.2
-3.0%
LULUCF Sector Net Total3
(976.7)
(907.7)
(915.5)
(863.6)
(904.4)
(910.6)
(854.2)
-12.5%
Net Emissions (Sources and Sinks)
5,560.2 |
| 6,586.9 |
| 5,837.3
5,726.6
5,097.4
5,418.2
5,489.0
-1.3%
a The LULUCF sector net total is the net sum of all LULUCF CH4 and N20 emissions to the atmosphere plus LULUCF net carbon
stock changes.
Notes: Total (gross) emissions are presented without LULUCF. Total net emissions are presented with LULUCF. Totals may not
sum due to independent rounding. Parentheses indicate negative values or sequestration.
Using this categorization, emissions from transportation activities accounted for the largest portion (28.4 percent)
of total gross U.S. greenhouse gas emissions in 2022. Electric power accounted for the second largest portion (24.9
percent) of U.S. greenhouse gas emissions in 2022, while emissions from industry accounted for the third largest
portion (22.9 percent). Emissions from industry have in general declined over the past decade, due to a number of
factors, including structural changes in the U.S. economy (i.e., shifts from a manufacturing-based to a service-
based economy), fuel switching, and energy efficiency improvements.
The remaining 23.8 percent of U.S. greenhouse gas emissions were contributed by, in order of magnitude, the
agriculture, commercial, and residential sectors, plus emissions from U.S. Territories. Activities related to
agriculture accounted for 10.0 percent of U.S. emissions; unlike other economic sectors, agricultural sector
Executive Summary ES-21
-------
emissions were dominated by N2O emissions from agricultural soil management and CH4 emissions from enteric
fermentation. An increasing amount of carbon is stored in agricultural soils each year, but this CO2 sequestration is
assigned to the LULUCF sector rather than the agriculture economic sector. The commercial and residential sectors
accounted for 7.3 percent and 6.2 percent of emissions, respectively, and U.S. Territories accounted for 0.4
percent of emissions; emissions from these sectors primarily consisted of CO2 emissions from fossil fuel
combustion. Carbon dioxide was also emitted and sequestered by a variety of activities related to forest
management practices, tree planting in urban areas, the management of agricultural soils, landfilling of yard
trimmings, and changes in carbon stocks in coastal wetlands.
Electricity is ultimately used in the economic sectors described above. Table ES-6 presents greenhouse gas
emissions from economic sectors with emissions related to electric power distributed into end-use categories (i.e.,
emissions from electric power generation are allocated to the economic sectors in which the electricity is used). To
distribute electricity emissions among end-use sectors, emissions from the source categories assigned to electric
power were allocated to the residential, commercial, industry, transportation, and agriculture economic sectors
according to retail sales of electricity for each end-use sector (EIA 2024).24 These source categories include CO2
from fossil fuel combustion and the use of limestone and dolomite for flue gas desulfurization, CO2 and N2O from
incineration of waste, CFU and N2O from stationary sources, and SF6 from electrical equipment systems.
When emissions from electricity use are distributed among these end-use sectors, industrial activities and
transportation account for the largest shares of U.S. greenhouse gas emissions (29.5 percent and 28.5 percent,
respectively) in 2022. The commercial and residential sectors contributed the next largest shares of total gross U.S.
greenhouse gas emissions in 2022 (15.8and 15.3 percent, respectively). Emissions from the commercial and
residential sectors increase substantially when emissions from electricity use are included, due to their relatively
large share of electricity use for energy (e.g., lighting, cooling, appliances). Figure ES-14 shows the trends in these
emissions by sector from 1990 to 2022.
Table ES-6: U.S. Greenhouse Gas Emissions with Electricity-Related Emissions Distributed by
Economic Sector (MMT CO2 Eq.)
Percent
Change
Since
Economic Sectors
1990
2005
2018
2019
2020
2021
2022
1990
Industry
2,397.3
2,302.9
2,017.1
1,974.8
1,823.5
1,877.8
1,872.9
-21.9%
Transportation
1,524.6
1,970.8 I:
1,876.5
1,879.5
1,629.5
1,810.6
1,807.8
18.6%
Commercial
1,002.5
1,241.1
1,074.3
1,030.5
931.5
976.8
1,002.6
0.0%
Residential
958.0 J
1,247.7
1,035.9
984.0
919.5
958.0
973.5
1.6%
Agriculture
631.1
672.6
722.7
696.3
674.4
681.6
663.6
5.2%
U.S. Territories
23.4
59.7
26.3
25.1
23.4
23.9
22.7
-3.1%
Total Gross Emissions (Sources)
6,536.9
7,494.6
6,752.7
6,590.1
6,001.8
6,328.8
6,343.2
-3.0%
LULUCF Sector Net Total3
(976.7)
(907.7)
(915.5)
(863.6)
(904.4)
(910.6)
(854.2)
-12.5%
Net Emissions (Sources and Sinks)
5,560.2
6,586.9
5,837.3
5,726.6
5,097.4
5,418.2
5,489.0
-1.3%
a The LULUCF sector net total is the net sum of all LULUCF CH4 and N20 emissions to the atmosphere plus LULUCF net carbon
stock changes.
Notes: Emissions from electric power are allocated based on aggregate electricity use in each end-use sector. Totals may not
sum due to independent rounding. Parentheses indicate negative values or sequestration.
24 U.S. Territories consumption data that are obtained from EIA are only available at the aggregate level and cannot be broken
out by end-use sector. The distribution of emissions to each end-use sector for the 50 states does not apply to Territories data.
ES-22 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
Figure ES-14: U.S. Greenhouse Gas Emissions with Electricity-Related Emissions Distributed to
Economic Sectors
Note: Emissions and removals from Land Use, Land-Use Change, and Forestry are excluded from figure above. Excludes U.S.
Territories.
Box ES-2: Trends in Various U.S. Greenhouse Gas Emissions-Related Data
Total (gross) greenhouse gas emissions can be compared to other economic and social indices to highlight
changes over time. These comparisons include: (1) aggregate energy use, because energy-related activities are
the largest sources of emissions; (2) energy use per capita as a measure of efficiency; (3) emissions per unit of
total gross domestic product as a measure of national economic activity; and (4) emissions per capita.
Table ES-7 provides data on various statistics related to U.S. greenhouse gas emissions normalized to 1990 as a
baseline year. These values represent the relative change in each statistic since 1990. Greenhouse gas emissions
in the United States have declined at an average annual rate of 0.01 percent since 1990, although changes from
year to year have been significantly larger. This growth rate is slightly slower than that for total energy use and
fossil fuel consumption, and overall gross domestic product (GDP), and national population (see Figure ES-15).
The direction of these trends started to change after 2005, when greenhouse gas emissions, total energy use,
and fossil fuel consumption began to peak. Greenhouse gas emissions in the United States have decreased at an
average annual rate of 0.9 percent since 2005. Since 2005, GDP and national population, generally, continued to
increase while energy has decreased slightly—noting 2020 was impacted by the COVID-19 pandemic.
Executive Summary ES-23
-------
Table ES-7: Recent Trends in Various U.S. Data (Index 1990 = 100)
Avg. Annual
Avg. Annual
Growth Rate
Growth Rate
Variable
1990
2005
2018
2019
2020
2021
2022
Since 1990a
Since 2005a
Greenhouse Gas Emissions'5
100
115
103
101
92
97
97
-0.1%
-0.9%
Energy Usec
100
119
118
117
107
113
115
0.5%
-0.2%
GDPd
100
159
201
206
201
213
217
2.5%
1.9%
Population6
100
118
130
131
132
132
133
0.9%
0.7%
+ Absolute value does not exceed 0.05 percent.
a Average annual growth rate.
b Gross total GWP-weighted values.
c Energy content-weighted values (EIA 2024).
d GDP in chained 2017 dollars (BEA 2024).
e U.S. Census Bureau (2024).
Figure ES-15: U.S. Greenhouse Gas Emissions Per Capita and Per Dollar of Gross Domestic
Product (GDP)
Source: BEA (2024), U.S. Census Bureau (2024), and emission estimates in this report.
Key Categories
The 2006IPCC Guidelines for National Greenhouse Gas Inventories (IPCC 2006) and 2019 Refinement to the 2006
IPCC Guidelines for National Greenhouse Gas Inventories (IPCC 2019) defines key categories as "inventory
categories which individually, or as a group of categories (for which a common method, emission factor and
activity data are applied) are prioritized within the national inventory system because their estimates have a
significant influence on a country's total inventory of greenhouse gases in terms of the absolute level, the trend, or
ES-24 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
the level of uncertainty in emissions or removals."25 A key category analysis identifies priority source or sink
categories for focusing efforts to improve overall Inventory quality. In addition, a qualitative review of key
categories and non-key categories can also help identify additional source and sink categories to consider for
improvement efforts, including reducing uncertainty.
Figure ES-16 presents the 2022 key categories identified by the Approach 1 level assessment, including the LULUCF
sector. A level assessment using Approach 1 identifies all source and sink categories that cumulatively account for
95 percent of total (i.e., gross) emissions in a given year when assessed in descending order of absolute magnitude.
For a complete list of key categories and more information regarding the overall key category analysis, including
approaches accounting for uncertainty and the influence of trends of individual source and sink categories, see the
Introduction chapter, Section 1.5, and Annex 1.
25 See Chapter 4 "Methodological Choice and Identification of Key Categories" in IPCC (2006) and IPCC (2019). See
http://www.ipcc-nggip.iges.or.ip/public/2QQ6gl/voll.html and https://www.ipcc-
nggip.iges.or.jp/public/2019rf/pdf/l Volumel/19R VI CH04 MethodChoice.pdf.
Executive Summary ES-25
-------
Figure ES-16: 2022 Key Categories (Approach 1 including LULUCF)3
CO2 Emissions from Mobile Combustion: Road
CO2 Emissions from Stationary Combustion - Coal - Electricity Generation
Net Carbon Stock Change from Forest Land Remaining Forest Land
CO2 Emissions from Stationary Combustion - Natural Gas - Electricity Generation
CO2 Emissions from Stationary Combustion - Natural Gas - Industrial
CO2 Emissions from Stationary Combustion - Natural Gas - Residential
N2O Emissions from Direct Agricultural Soil Management
CO2 Emissions from Stationary Combustion - Oil - Industrial
CO2 Emissions from Stationary Combustion - Natural Gas - Commercial
Ct-U Emissions Enteric Fermentation: Cattle
CH4 Emissions from Natural Gas Systems
CO2 Emissions from Mobile Combustion: Aviation
Emissions Substitutes for Ozone Depleting Substances: Refrigeration and Air conditioning
Net Carbon Stock Change from Settlements Remaining Settlements
CO2 Emissions from Non-Energy Use of Fuels
CH4 Emissions from Commercial Landfills
Net Carbon Stock Change from Land Converted to Forest Land
CO2 Emissions from Mobile Combustion: Other
Net Carbon Stock Change from Land Converted to Settlements
CO2 Emissions fromStationary Combustion - Oil - Commercial
CO2 Emissions from Stationary Combustion - Oil - Residential
CH4 Emissions from Flooded Lands Remaining Flooded Lands
CH4 Emissions from Fugitive Emissions from Coal Mining
CO2 Emissions from Stationary Combustion - Coal - Industrial
CO2 Emissions from Cement Production
CO2 Emissions from Mobile Combustion: Marine
CO2 Emissions from Iron and Steel Production & Metallurgical Coke Production
CH4 Emissions from Petroleum Systems
CH4 Emissions from Manure Management: Cattle
CO2 Emissions from Natural Gas Systems
Net Carbon Stock Change from Converted to Cropland
CO2 Emissions from Mobile Combustion: Railways
CO2 Emissions from Net Cropland Remaining Cropland
CO2 Emissions from Petrochemical Production
N2O Emissions from Indirect Applied Nitrogen
CH4 Emissions from Manure Management: Other Livestock
CO2 Emissions from Net Land Converted to Grassland
CO2 Emissions fromPetroleum Systems
N2O Emissions from Domestic Wastewater Treatment
CO2 Emissions from Stationary Combustion - Oil - Electricity Generation
CH4 Emissions from Industrial Landfills
CH4 Emissions from Rice Cultivation
N2O Emissions from Stationary Combustion - Coal - Electricity Generation
Emissions from Substitutes for Ozone Depleting Substances: Aerosols
Net Carbon Stock Change from Net Grassland Remaining Grassland
CH4 Emissions from Abandoned Oil and Natural Gas Wells
CH4 Emissions from Stationary Combustion - Residential
Key Categories as a Portion of All
Emissions
96.3%
I Key Categories
I Key Categoreis LULUCF
Other Categories
0 200 400 600 800 1,000 1,200 1,400
2022 Emissions (MMT CO2 Eq.)
Note: For a complete list of key categories and detailed discussion of the underlying key category analysis, see Annex 1. Bars
indicate key categories identified using Approach 1 level assessment including the LULUCF sector. The absolute values of net
C02 emissions from LULUCF are presented in this figure but reported separately from gross emissions totals. Refer to Table
ES-4 for a breakout of emissions and removals for LULUCF by source/sink category.
Quality Assurance and Quality Control (QA/QC)
The United States seeks continuous improvements to the quality, transparency, and usability 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. The procedures followed for the Inventory have been formalized in accordance with the U.S
Inventory QA/QC plan, and the UNFCCC reporting guidelines and 2006IPCC Guidelines for National Greenhouse
Gas inventories. The QA process includes expert and public reviews for the inventory estimates and this report.
ES-26 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
Box ES-3: Use of Ambient Measurements Systems for Validation of Emission Inventories
In following Article 13.7(a) of the Paris Agreement and Article 4.1(a) of the UNFCCC, as well as relevant decisions
under those agreements to develop and submit national greenhouse gas emission inventories, the emissions
and sinks presented in this report are organized by source and sink categories and calculated using
internationally accepted methods in the 2006 IPCC Guidelines for National Greenhouse Gas Inventories (2006
IPCC Guidelines) and, where appropriate, its supplements and refinements.26 Several recent studies have
estimated emissions at the national or regional level with estimated results that sometimes differ from EPA's
estimate of emissions. EPA has engaged with researchers on how remote sensing, ambient measurement, and
inverse modeling techniques for estimating greenhouse gas emissions could assist in improving the
understanding of inventory estimates. In working with the research community to improve national greenhouse
gas inventories, EPA follows guidance from the IPCC on the use of measurements and modeling to validate
emission inventories.27 An area of particular interest in EPA's outreach efforts is how ambient measurement
data can be used to assess estimates or potentially be incorporated into the Inventory in a manner consistent
with this Inventory report's transparency of its calculation methodologies, and the ability of inverse modeling
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.
The 2019 Refinement to the IPCC 2006 Guidelines for National Greenhouse Gas Inventories (IPCC 2019) Volume
1 General Guidance and Reporting, Chapter 6: Quality Assurance, Quality Control and Verification notes that
emission estimates derived from atmospheric concentration measurements can provide independent data sets
as a basis for comparison with inventory estimates. The 2019 Refinement provides guidance on conducting such
comparisons (as summarized in Table 6.2 of IPCC [2019] Volume 1, Chapter 6) and provides guidance on using
such comparisons to identify areas of improvement in national inventories (as summarized in Box 6.5 of IPCC
[2019] Volume 1, Chapter 6). Further, it identified fluorinated gases as particularly suitable for such comparisons
due their limited natural sources, their generally long atmospheric lifetimes, and well-understood loss
mechanisms, which makes it relatively more straightforward to model their emission fluxes from observed mass
quantities. Unlike emissions of CO2, CFU, and N2O, emissions of fluorinated greenhouse gases are almost
exclusively anthropogenic, meaning that the fluorinated greenhouse gas emission sources included in this
Inventory account for the majority of the total U.S. emissions of these gases detectable in the atmosphere. This
evaluation approach is also useful for gases and sources with larger uncertainties in available bottom-up
inventory methods and data, such as emissions of CH4, which are primarily from uncertain biological (e.g.,
enteric fermentation) and fugitive (e.g., natural gas production) activities.
In this Inventory, EPA includes the results from current and previous comparisons between fluorinated gas
emissions inferred from atmospheric measurements and fluorinated gas emissions estimated based on bottom-
up measurements and modeling. These comparisons, performed for HFCs and SF6 respectively, are described
under the QA/QC and Verification discussions in Chapter 4, Sections 4.25 Substitution of Ozone Depleting
Substances and 4.26 Electrical Equipment, in the IPPU chapter of this report.
Consistent with the 2019 Refinement, a key element to facilitate such comparisons is a spatially-explicit (or
gridded inventory as an input to inverse modeling. To improve the ability to compare methane emissions from
the national-level greenhouse gas inventory with observation-based estimates, a team of researchers from U.S.
EPA, SRON Netherlands Institute for Space Research, Harvard University, and Lawrence Berkely National
Laboratory and other coauthors developed a time series of anthropogenic methane emissions maps with 0.1° x
0.1° (10 km x 10 km) spatial resolution and monthly temporal resolution for the contiguous United States.28 The
gridded methane inventory is designed to be consistent with the U.S. EPA Inventory of U.S. Greenhouse Gas
26 See http://www.ipcc-negip.iges.or.jp/public/index.html.
27 See http://www.ipcc-nggip.iges.or.jp/meeting/pdfiles/1003 Uncertaintv%20meeting report.pdf.
28 See https://www.epa.gov/ghgemissions/us-gridded-methane-emissions.
Executive Summary ES-27
-------
Emissions and Sinks estimates, which presents national totals for different source types.29 The development of
this gridded inventory is consistent with the recommendations contained in two National Academies of Science
reports examining greenhouse gas emissions data (National Research Council 2010; National Academies of
Sciences, Engineering, and Medicine 2018).
Finally, in addition to the use of atmospheric concentration measurement data for comparison with Inventory
data, information from top-down studies is directly incorporated in the natural gas systems calculations to
quantify emissions from certain well blowout events.
Uncertainty Analysis of Emission and Sink Estimates
Uncertainty assessment is an essential element of a complete inventory of greenhouse gas emissions and removals
because it helps to inform and prioritize inventory improvements. Recognizing the benefit of conducting an
uncertainty analysis, the UNFCCC reporting guidelines follow the recommendations of the 2006IPCC Guidelines for
National Greenhouse Gas Inventories (IPCC 2006), Volume 1, Chapter 3 and require that countries provide single
estimates of uncertainty for source and sink categories. In addition to quantitative uncertainty assessments, a
qualitative discussion of uncertainty is presented for each source and sink category identifying specific factors
affecting the uncertainty surrounding the estimates provided in accordance with UNFCCC reporting guidelines.
Some of the current estimates, such as those for CO2 emissions from energy-related combustion activities, are
considered to have low uncertainties. This is because the amount of CO2 emitted from energy-related combustion
activities is directly related to the amount of fuel consumed, the fraction of the fuel that is oxidized, and the
carbon content of the fuel, and for the United States, the uncertainties associated with estimating those factors
are relatively small. For some other categories of emissions and sinks, however, inherent variability or a lack of
data increases the uncertainty or systematic error associated with the estimates presented. Finally, an analysis is
conducted to assess uncertainties associated with the overall emissions, sinks, and trends estimates. The overall
uncertainty surrounding total net greenhouse gas emissions is estimated to be -6 to +6 percent in 1990 and -5 to
+6 percent in 2022. When the LULUCF sector is excluded from the analysis the uncertainty is estimated to be -3 to
+4 percent in 1990 and -2 to +4 percent in 2022.
29 See https://www.epa.eov/eheemissions/inventorv-us-ereenhouse-eas-emissions-and-sinks.
ES-28 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
1. Introduction
This report presents an inventory of U.S. anthropogenic greenhouse gas emissions and sinks for the years 1990
through 2022 compiled by the United States government. A summary of source and sink category estimates is
provided in Table 2-1, Table 2-2, andTable 2-4 by gas and IPCC sector in the Trends in Greenhouse Gas Emissions
and Sinks chapter. The emission and sink estimates in these tables are presented throughout the main report in
both CO2 equivalents (CO2 Eq.30 and unweighted units). This report also discusses the methods and data used to
calculate the emission and sink estimates.
The United States is party to both the 1992 UNFCCC and the 2015 Paris Agreement. The Paris Agreement set a
global temperature goal-holding the increase in the global average temperature to well below 2°C above pre-
industrial levels and pursuing efforts to limit the increase to 1.5°C-that articulates with greater precision States'
views on what is necessary to meet the UNFCCC's objective of "stabilizing]... greenhouse gas concentrations in
the atmosphere at a level that would prevent dangerous anthropogenic interference with the climate system.31
The United States is committed to submitting a national inventory of anthropogenic emissions sources and
removals by sinks of greenhouse gases by April 15 of each year. The United States has prepared this report, in
conjunction with Common Reporting Tables (CRTs) that accompany this report, consistent with its obligations
under those agreements.
In 1988, preceding the creation of the UNFCCC, the World Meteorological Organization (WMO) and the United
Nations Environment Programme (UNEP) jointly established the Intergovernmental Panel on Climate Change
(IPCC). The role of the IPCC is to assess on a comprehensive, objective, open and transparent basis the scientific,
technical and socio-economic information relevant to understanding the scientific basis of risk of human-induced
climate change, its potential impacts and options for adaptation and mitigation (IPCC 2021). The Paris Agreement
and the UNFCCC require use of methods from the 2006 IPCC Guidelines for National Greenhouse Gas Inventories
and encourages Parties to use the 2013 Supplement to the 2006 IPCC Guidelines for National Greenhouse Gas
Inventories: Wetlands and the 2019 Refinement to the 2006 IPCC Guidelines for National Greenhouse Gas
Inventories. The recently released 2019 Refinement clarify and elaborate on the existing guidance in the 2006 IPCC
Guidelines, along with providing updates to default values of emission factors and other parameters based on
updated science. This report applies both the 2013 Supplement and updated guidance in the 2019 Refinement to
improve accuracy and completeness of the Inventory. For more information on specific uses, see Section 1.4 of this
chapter on Methodology and Data Sources.
30 More information is provided in the Global Warming Potentials section of this chapter on the use of IPCC Fifth Assessment
Report (AR5) GWP values.
31 See Paris Agreement, Article 2.1(a); UNFCCC, Article 2.
Introduction 1-1
-------
Box 1-1: Methodological Approach for Estimating and Reporting U.S. Emissions and
Removals, including Relationship to EPA's Greenhouse Gas Reporting Program
Consistent with Article 13.7(a) of the Paris Agreement and Article 4.1(a) of the UNFCCC, as well as relevant
decisions under those agreements, the emissions and removals presented in this report and this chapter are
organized by source and sink categories and calculated using internationally accepted methods in the 2006IPCC
Guidelines for National Greenhouse Gas Inventories (2006 IPCC Guidelines) and, where appropriate, its
supplements and refinements. Additionally, the calculated emissions and removals in a given year for the
United States are presented in a common manner in line with the reporting guidelines for the reporting of
inventories under the Paris Agreement and the UNFCCC. The Parties' use of consistent methods to calculate
emissions and removals for their inventories helps to ensure that these reports are comparable. The
presentation of emissions and removals provided in this Inventory does not preclude alternative examinations
(e.g., economic sectors). Rather, this Inventory presents emissions and removals in a common format consistent
with how Parties are to report their national inventories under the Paris Agreement and the UNFCCC. The
report itself, and this chapter, follows this common format, and provides an explanation of the application of
methods used to calculate emissions and removals.
EPA also collects greenhouse gas emissions data from individual facilities and suppliers of certain fossil fuels and
industrial gases through its Greenhouse Gas Reporting Program (GHGRP), which is complementary to the U.S.
Inventory,32 The GHGRP applies to direct greenhouse gas emitters, fossil fuel suppliers, industrial greenhouse
gas suppliers, and facilities that inject carbon dioxide (CO2) underground for sequestration or other reasons and
requires reporting by over 8,000 sources or suppliers in 41 industrial categories.33 Annual 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. Facilities in most source
categories34 subject to GHGRP began reporting for the 2010 reporting year while additional types of industrial
operations began reporting for the 2011 reporting year. Methodologies used in EPA's GHGRP are consistent
with the 2006 IPCC Guidelines. While the GHGRP does not provide full coverage of total annual U.S. greenhouse
gas emissions and removals (e.g., the GHGRP excludes emissions from the Agriculture and Land Use, Land-Use
Change, and Forestry sectors), it is an important input to the calculations of national-level emissions in this
Inventory.
Data presented in this Inventory report and EPA's GHGRP are complementary. The GHGRP dataset continues to
be an important resource for the Inventory, providing not only annual emissions information, but also other
annual information such as activity data and emission factors that can improve and refine national emission
estimates and trends over time. Methodologies used in EPA's GHGRP are consistent with the 2006 IPCC
Guidelines (e.g., higher tier methods). GHGRP data also allow EPA to disaggregate national inventory estimates
in new ways that can highlight differences across regions and sub-categories of emissions, along with enhancing
the application of QA/QC procedures and assessment of uncertainties. EPA uses annual GHGRP data in several
categories to improve the national estimates presented in this Inventory, consistent with IPCC methodological
guidance. See Annex 9 for more information on specific uses of GHGRP data in the Inventory (e.g., natural gas
systems).
1.1 Background Information
Science
For over the past 200 years, the burning of fossil fuels such as coal, oil, and natural gas, along with deforestation,
land-use changes, and other activities have caused the concentrations of heat-trapping "greenhouse gases" to
increase significantly in our atmosphere (IPCC 2021). These gases in the atmosphere absorb some of the energy
1-2 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
being radiated from the surface of the Earth that would otherwise be lost to space, essentially acting like a blanket
that makes the Earth's surface warmer than it would be otherwise.
Greenhouse gases are necessary to life as we know it. Without greenhouse gases to create the natural heat-
trapping properties of the atmosphere, the planet's surface would be about 60 degrees Fahrenheit cooler than
present (USGCRP 2017). Carbon dioxide is also necessary for plant growth. With emissions from biological and
geological sources, there is a natural level of greenhouse gases that is maintained in the atmosphere. Human
emissions of greenhouse gases and subsequent changes in atmospheric concentrations alter the balance of energy
transfers between space and the earth system (IPCC 2021). A gauge of these changes is called radiative forcing,
which is a measure of a substance's total net effect on the global energy balance for which a positive number
represents a warming effect, and a negative number represents a cooling effect (IPCC 2021). IPCC concluded in its
most recent scientific assessment report that it is "unequivocal that human influence has warmed the atmosphere,
ocean and land" (IPCC 2021).
As concentrations of greenhouse gases continue to increase from man-made sources, the Earth's temperature is
climbing above past levels. The Earth's average land and ocean surface temperature has increased by about 2.0
degrees Fahrenheit from the 1850 to 1900 period to the decade of 2011 to 2020 (IPCC 2021). The last four decades
have each been the warmest decade successively at the Earth's surface since at least 1850 (IPCC 2021). Other
aspects of the climate are also changing, such as rainfall patterns, snow and ice cover, and sea level. If greenhouse
gas concentrations continue to increase, climate models predict that the average temperature at the Earth's
surface is likely to increase between 0.27 to 9.99 degrees Fahrenheit (0.15 to 5.55 degrees Celsius) relative to 1995
to 2014 levels by the end of this century, depending on the emissions scenario and the responsiveness of the
climate system (IPCC 2021).
For further information on greenhouse gases, radiative forcing, and implications for climate change, see the recent
scientific assessment reports from the IPCC,35 the U.S. Global Change Research Program (USGCRP),36 and the
National Academies of Sciences, Engineering, and Medicine (NAS).37
Greenhouse Gases
Although the Earth's atmosphere consists mainly of oxygen and nitrogen, neither plays a significant role in
enhancing the greenhouse effect because both are essentially transparent to terrestrial radiation. The greenhouse
effect is primarily a function of the concentration of water vapor, carbon dioxide (CO2), methane (CH4), nitrous
oxide (N2O), and other trace gases in the atmosphere that absorb the terrestrial radiation leaving the surface of
the Earth (IPCC 2021).
Naturally occurring greenhouse gases include water vapor, CO2, CH4, N2O, and ozone (O3). Several classes of
halogenated substances that contain fluorine, chlorine, or bromine are also greenhouse gases, but they are, for the
most part, solely a product of industrial activities. Chlorofluorocarbons (CFCs) and hydrochlorofluorocarbons
(HCFCs) are halocarbons that contain chlorine, while halocarbons that contain bromine are referred to as
bromofluorocarbons (i.e., halons). As stratospheric ozone depleting substances, CFCs, HCFCs, and halons are
covered under the Montreal Protocol on Substances that Deplete the Ozone Layer. The UNFCCC defers to this
earlier international treaty. Consequently, Parties to the UNFCCC are not required to include these gases in
32 On October 30, 2009, EPA promulgated a rule requiring annual reporting of greenhouse gas data from large greenhouse gas
emissions sources in the United States. Implementation of the rule, codified at 40 CFR Part 98, is referred to as EPA's
Greenhouse Gas Reporting Program (GHGRP).
33 See http://www.epa.gov/ghgreporting and http://ghgdata.epa.gov/ghgp/main.do.
34 See https://www.eedsupport.eom/eonfluenee/pages/yiewpage.aetion7pagefe322699300
35 See https://www.ipcc.ch/report/ar6/wgl/.
36 See https://nca2018.globalchange.gov/.
37 See https://www.nationalacademies.org/topics/climate.
Introduction 1-3
-------
national greenhouse gas inventories.38 Some other fluorine-containing halogenated substances—
hydrofluorocarbons (HFCs), perfluorocarbons (PFCs), sulfur hexafluoride (SFs), and nitrogen trifluoride (NF3)—do
not deplete stratospheric ozone but are potent greenhouse gases. These latter substances are addressed by the
UNFCCC and accounted for in national greenhouse gas inventories.
There are also several other substances that influence the global radiation budget but are short-lived and
therefore not well-mixed, leading to spatially variable radiative forcing effects. These substances include carbon
monoxide (CO), nitrogen dioxide (NO2), sulfur dioxide (SO2), and tropospheric (ground level) ozone (O3).
Tropospheric ozone is formed from chemical reactions in the atmosphere of precursor pollutants, which include
volatile organic compounds (VOCs), CFU, and nitrogen oxides (NOx), in the presence of ultraviolet light (sunlight).
Aerosols are extremely small particles or liquid droplets suspended in the Earth's atmosphere that are often
composed of sulfur compounds, carbonaceous combustion products (e.g., black carbon), crustal materials (e.g.,
dust) and other human-induced pollutants. They can affect the absorptive characteristics of the atmosphere (e.g.,
scattering incoming sunlight away from the Earth's surface, or, in the case of black carbon, absorbing sunlight) and
can play a role in affecting cloud formation and lifetime, as well as the radiative forcing of clouds and precipitation
patterns.
CO2, CH4, and N2O are continuously emitted to and removed from the atmosphere by natural processes on Earth.
Anthropogenic activities (such as fossil fuel combustion, cement production, land-use, land-use change, and
forestry, agriculture, or waste management), however, can cause additional quantities of these and other
greenhouse gases to be emitted or sequestered, thereby changing their global average atmospheric
concentrations. Natural activities such as respiration by plants or animals and seasonal cycles of plant growth and
decay are examples of processes that only cycle carbon or nitrogen between the atmosphere and organic biomass.
Such processes, except when directly or indirectly perturbed out of equilibrium by anthropogenic activities,
generally do not alter average atmospheric greenhouse gas concentrations over decadal timeframes. Climatic
changes resulting from anthropogenic activities, however, could have positive or negative feedback effects on
these natural systems. Atmospheric concentrations of these gases, along with their rates of growth and
atmospheric lifetimes, are presented in Table 1-1.
Table 1-1: Global Atmospheric Concentration, Rate of Concentration Change, and
Atmospheric Lifetime of Selected Greenhouse Gases
Atmospheric Variable
CO.
CH.,
NO
SF„
CF,
Pre-industrial atmospheric
concentration
280 ppm
0.730 ppm
0.270 ppm
0.01 ppt
34.1 ppt
Atmospheric concentration
419 ppma
1.912 ppmb
0.336 ppmc
11.02 pptd
85.5 ppte
Rate of concentration change
2.28 ppm/yrf
8.83 ppb/yrf'S
1.01 ppb/yrf
0.32 ppt/yrf
0.81 ppt/yrf
Atmospheric lifetime (years)
See footnote11
11.8
109'
About l,000i
50,000
a The atmospheric C02 concentration is the 2022 annual average at the Mauna Loa, HI station (NOAA/ESRL 2024a). The global
atmospheric C02 concentration, computed using an average of sampling sites across the world, was 417 ppm in 2022.
b The values presented are global 2022 annual average mole fractions (NOAA/ESRL 2024b).
c The values presented are global 2022 annual average mole fractions (NOAA/ESRL 2024c).
d The values presented are global 2022 annual average mole fractions (NOAA/ESRL 2024d).
e The 2019 CF4 global mean atmospheric concentration is from the Advanced Global Atmospheric Gases Experiment (IPCC
2021).
f The rate of concentration change for C02 is an average of the rates from 2007 through 2022 and has fluctuated between 1.5
to 3.0 ppm per year over this period (NOAA/ESRL 2024a). The rate of concentration change for CH4, N20, and SF6, is the
average rate of change between 2007 and 2022 (NOAA/ESRL 2024b; NOAA/ESRL 2024c; NOAA/ESRL 2024d). The rate of
concentration change for CF4 is the average rate of change between 2011 and 2019 (IPCC 2021).
g The growth rate for atmospheric CH4 decreased from over 10 ppb/year in the 1980s to nearly zero in the early 2000s;
recently, the growth rate has been about 13.22 ppb/year (NOAA/ESRL 2024b).
38 Emissions estimates of CFCs, HCFCs, halons and other ozone-depleting substances are included in this document for
informational purposes.
1-4 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
h For a given amount of C02 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.
' This table reports the "perturbation lifetime" for both CH4 and N20, which takes into account the interactions between
emissions of the gas and its own atmospheric residence time.
j The lifetime for SF6 was revised from 3,200 years to about 1,000 years based on recent studies (IPCC 2021).
Source: Pre-industrial atmospheric concentrations and atmospheric lifetimes for CH4, and N20, are from IPCC (2021), pre-
industrial atmospheric concentration for SF6, is from Rigby (2010), and pre-industrial atmospheric concentration for CF4 is
from Meinhausen (2017).
A brief description of each greenhouse gas, its sources, and its role in the atmosphere is given below. The following
section then explains the concept of global warming potentials (GWPs), which are assigned to individual gases as a
measure of their relative average global radiative forcing effect.
Water Vapor (H2O). Water vapor is the largest contributor to the natural greenhouse effect. Water vapor is
fundamentally different from other greenhouse gases in that it can condense and rain out when it reaches high
concentrations, and the total amount of water vapor in the atmosphere is in part a function of the Earth's
temperature. While some human activities such as evaporation from irrigated crops or power plant cooling release
water vapor into the air, these activities have been determined to have a negligible effect on global climate (IPCC
2021). The lifetime of water vapor in the troposphere is on the order of 10 days. Water vapor can also contribute
to cloud formation, and clouds can have both warming and cooling effects by either trapping or reflecting heat.
Because of the relationship between water vapor levels and temperature, water vapor and clouds serve as a
feedback to climate change, such that for any given change in other greenhouse gas concentrations, the total
temperature change is greater than would happen in the absence of water vapor. Aircraft emissions can create
contrails, which may also develop into contrail-induced cirrus clouds, with complex regional and temporal net
radiative forcing effects that currently have a low level of scientific certainty (IPCC 2021).
Carbon Dioxide (CCh). In nature, carbon is cycled between various atmospheric, oceanic, land biotic, marine biotic,
and mineral reservoirs. The largest fluxes occur between the atmosphere and terrestrial biota, and between the
atmosphere and surface water of the oceans. In the atmosphere, carbon predominantly exists in its oxidized form
as CO2. Atmospheric CO2 is part of this global carbon cycle, and therefore its fate is a complex function of
geochemical and biological processes. Carbon dioxide concentrations in the atmosphere increased from
approximately 280 parts per million by volume (ppmv) in pre-industrial times to 419 ppmv in 2022, a 50 percent
increase (IPCC 2021; NOAA/ESRL 2024a).39,40 The IPCC states that "Observed increases in well-mixed greenhouse
gas (GHG) concentrations since around 1750 are unequivocally caused by human activities" (IPCC 2021). The
predominant source of anthropogenic CO2 emissions is the combustion of fossil fuels. Forest clearing, other
biomass burning, and some non-energy production processes (e.g., cement production) also emit notable
quantities of CO2. In its Sixth Assessment Report, the IPCC determined that of the 2.0 degrees of observed
warming, the best estimate is that 1.9 degrees of that are due to human influence, with elevated CO2
concentrations being the most important contributor to that warming (IPCC 2021).
Methane (CHa). Methane is primarily produced through anaerobic decomposition of organic matter in biological
systems. Agricultural processes such as wetland rice cultivation, enteric fermentation in animals, and the
decomposition of animal wastes emit CH4, as does the decomposition of municipal solid wastes and treatment of
wastewater. Methane is also emitted during the production and distribution of natural gas and petroleum, and is
released as a byproduct of coal mining and incomplete fossil fuel combustion. Atmospheric concentrations of Cm
have increased by about 162 percent since 1750, from a pre-industrial value of about 730 ppb to 1,912 ppb in
202241 although the rate of increase decreased to near zero in the early 2000s, and has recently increased again to
39 The pre-industrial period is considered as the time preceding the year 1750 (IPCC 2021).
40 Carbon dioxide concentrations during the last 1,000 years of the pre-industrial period (i.e., 750 to 1750), a time of relative
climate stability, fluctuated by about +10 ppmv around 280 ppmv (IPCC 2021).
41 This value is the global 2022 annual average mole fraction (NOAA/ESRL 2024b).
Introduction 1-5
-------
about 8.83 ppb/year. The IPCC has estimated that about half of the current Cm flux to the atmosphere (and the
entirety of the increase in concentration) is anthropogenic, from human activities such as agriculture, fossil fuel
production and use, and waste disposal (IPCC 2021).
Methane is primarily removed from the atmosphere through a reaction with the hydroxyl radical (OH) and is
ultimately converted to CO2. Minor removal processes also include reaction with chlorine in the marine boundary
layer, a soil sink, and stratospheric reactions. Increasing emissions of Cm reduce the concentration of OH, creating
a feedback that increases the atmospheric lifetime of CH4 (IPCC 2021). Methane's reactions in the atmosphere also
lead to production of tropospheric ozone and stratospheric water vapor, both of which also contribute to climate
change. Tropospheric ozone also has negative effects on human health and plant productivity.
Nitrous Oxide (N2O). Anthropogenic sources of N2O emissions include agricultural soils, especially production of
nitrogen-fixing crops and forages, the use of synthetic and manure fertilizers, and manure deposition by livestock;
fossil fuel combustion, especially from mobile combustion; adipic (nylon) and nitric acid production; wastewater
treatment and waste incineration; and biomass burning. The atmospheric concentration of N20 has increased by
24 percent since 1750, from a pre-industrial value of about 270 ppb to 336 ppb in 2022,42 a concentration that has
not been exceeded during at least the last 800 thousand years. Nitrous oxide is primarily removed from the
atmosphere by the photolytic action of sunlight in the stratosphere (IPCC 2021).
Ozone (O3). Ozone is present in both the upper stratosphere,43 where it shields the Earth from harmful levels of
ultraviolet radiation, and at lower concentrations in the troposphere,44 where it is the main component of
anthropogenic photochemical "smog." During the last two decades, emissions of anthropogenic chlorine and
bromine-containing halocarbons, such as CFCs, have depleted stratospheric ozone concentrations. This loss of
ozone in the stratosphere has resulted in negative radiative forcing, representing an indirect effect of
anthropogenic emissions of chlorine and bromine compounds (IPCC 2021). The depletion of stratospheric ozone
and its radiative forcing remained relatively unchanged since 2000 for the last two decades and is starting to
decline; recovery is expected to occur shortly after the middle of the twenty-first century (WMO/UNEP 2018).
The past increase in tropospheric ozone, which is also a greenhouse gas, is estimated to provide the third largest
increase in direct radiative forcing since the pre-industrial era, behind CO2 and CH4. Tropospheric ozone is
produced from complex chemical reactions of volatile organic compounds and CH4 mixing with NOx in the presence
of sunlight. The tropospheric concentrations of ozone and these other pollutants are short-lived and, therefore,
spatially variable (IPCC 2021).
Halocarbons, Sulfur Hexafluoride (SFs), and Nitrogen Trifluoride (NF3). Halocarbons are, for the most part, man-
made chemicals that have direct radiative forcing effects and could also have an indirect effect. Halocarbons that
contain chlorine (CFCs, HCFCs, methyl chloroform, and carbon tetrachloride) and bromine (halons, methyl
bromide, and hydrobromofluorocarbons) result in stratospheric ozone depletion and are therefore controlled
under the Montreal Protocol on Substances that Deplete the Ozone Layer. Although most CFCs and HCFCs are
potent global warming gases, their net radiative forcing effect on the atmosphere is reduced because they cause
stratospheric ozone depletion, which itself is a greenhouse gas but which also shields the Earth from harmful levels
of ultraviolet radiation. Under the Montreal Protocol, the United States phased out the production and
importation of halons by 1994 and of CFCs by 1996. Under the Copenhagen Amendments to the Protocol, a cap
42 This value is the global 2022 annual average (NOAA/ESRL 2024c).
43 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.
44 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.
1-6 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
was placed on the production and importation of HCFCs by non-Article 5 countries, including the United States,45
beginning in 1996, and then followed by intermediate requirements and a complete phase-out by the year 2030.
While ozone depleting gases covered under the Montreal Protocol and its Amendments are not covered by the
UNFCCC, they are reported in this Inventory under Annex 6.2 for informational purposes.
HFCs, PFCs, SFs, and NF3 are not ozone depleting substances. The most common HFCs are, however, powerful
greenhouse gases. Hydrofluorocarbons are primarily used as replacements for ozone depleting substances but are
also emitted as a byproduct of the HCFC-22 (chlorodifluoromethane) manufacturing process. Other contributing
sources to HFC emissions include the electronics industry and magnesium production and processing. Currently,
these emissions have a small aggregate radiative forcing impact, but it was anticipated that without further
controls their contribution to overall radiative forcing would increase, the ERF (effective radiative forcing) of
halogenated gases increased by 3.5 percent between 2011 and 2019 primarily due to a decrease in atmospheric
mixing-ratios of CFCs and an increase in their replacements (IPCC 2021). On December 27, 2020, the American
Innovation and Manufacturing (AIM) Act was enacted by Congress and which gives EPA authority to phase down
HFC production and consumption (i.e., production plus import, minus export), through an allowance allocation
program, promulgate certain regulations for purposes of maximizing reclamation and minimizing releases of HFCs
and their substitutes from equipment, and facilitate the transition to next-generation technologies through sector-
based restrictions, which will lead to lower HFC emissions over time. On October 31, 2022, the United States also
ratified the Kigali Amendment to the Montreal Protocol, committing to continued phase down of HFCs.
Perfluorocarbons, SF6, and NF3 are predominantly emitted from various industrial processes including aluminum
smelting, semiconductor manufacturing, electric power transmission and distribution, and magnesium casting.
Currently, the radiative forcing impact of PFCs, SF6, and NF3 is also small, but they have a significant growth rate,
extremely long atmospheric lifetimes, and are strong absorbers of infrared radiation, and therefore have the
potential to influence climate far into the future (IPCC 2021).
Carbon Monoxide (CO). Carbon monoxide has an indirect radiative forcing effect by elevating concentrations of CH4
and tropospheric ozone through chemical reactions with other atmospheric constituents (e.g., the hydroxyl radical,
OH) that would otherwise assist in destroying CH4 and tropospheric ozone. Carbon monoxide is created when
carbon-containing fuels are burned incompletely. Through natural processes in the atmosphere, it is eventually
oxidized to CO2. Carbon monoxide concentrations are both short-lived in the atmosphere and spatially variable.
Nitrogen Oxides (NOx). The primary climate change effects of nitrogen oxides (i.e., NO and NO2) are indirect.
Warming effects can occur due to reactions leading to the formation of ozone in the troposphere, but cooling
effects can occur due to the role of NOx as a precursor to nitrate particles (i.e., aerosols) and due to destruction of
stratospheric ozone when emitted from very high-altitude aircraft.46 Additionally, NOx emissions are also likely to
decrease CH4 concentrations, thus having a negative radiative forcing effect (IPCC 2021). Nitrogen oxides are
created from lightning, soil microbial activity, biomass burning (both natural and anthropogenic fires) fuel
combustion, and, in the stratosphere, from the photo-degradation of N20. Concentrations of NOx are both
relatively short-lived in the atmosphere and spatially variable.
Non-methane Volatile Organic Compounds (NMVOCs). Non-methane volatile organic compounds include
substances such as propane, butane, and ethane. These compounds participate, along with NOx, in the formation
of tropospheric ozone and other photochemical oxidants. NMVOCs are emitted primarily from transportation and
industrial processes, as well as biomass burning and non-industrial consumption of organic solvents.
Concentrations of NMVOCs tend to be both short-lived in the atmosphere and spatially variable.
45 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 of ten
additional years in the phase-out of ozone depleting substances.
46 NOx emissions injected higher in the stratosphere, primarily from fuel combustion emissions from high altitude supersonic
aircraft, can lead to stratospheric ozone depletion.
Introduction 1-7
-------
Aerosols. Aerosols are extremely small particles or liquid droplets found in the atmosphere that are either directly
emitted into or are created through chemical reactions in the Earth's atmosphere. Aerosols or their chemical
precursors can be emitted by natural events such as dust storms, biogenic or volcanic activity, or by anthropogenic
processes such as transportation, coal combustion, cement manufacturing, waste incineration, or biomass burning.
Various categories of aerosols exist from both natural and anthropogenic sources, such as soil dust, sea salt,
biogenic aerosols, sulfates, nitrates, volcanic aerosols, industrial dust, and carbonaceous47 aerosols (e.g., black
carbon, organic carbon). Aerosols can be removed from the atmosphere relatively rapidly by precipitation or
through more complex processes under dry conditions.
Aerosols affect radiative forcing differently than greenhouse gases. Their radiative effects occur through direct and
indirect mechanisms: directly by scattering and absorbing solar radiation (and to a lesser extent scattering,
absorption, and emission of terrestrial radiation); and indirectly by increasing cloud droplets and ice crystals that
modify the formation, precipitation efficiency, and radiative properties of clouds (IPCC 2021). Despite advances in
understanding of cloud-aerosol interactions, the contribution of aerosols to radiative forcing are difficult to
quantify because aerosols generally have short atmospheric lifetimes, and have number concentrations, size
distributions, and compositions that vary regionally, spatially, and temporally (IPCC 2021).
The net effect of aerosols on the Earth's radiative forcing is believed to be negative (i.e., net cooling effect on the
climate). In fact, aerosols contributed a cooling influence of up to 1.4 degrees, offsetting a substantial portion of
greenhouse gas warming (IPCC 2021). Because aerosols remain in the atmosphere for only days to weeks, their
concentrations respond rapidly to changes in emissions.48 Not all aerosols have a cooling effect. Current research
suggests that another constituent of aerosols, black carbon, has a positive radiative forcing by heating the Earth's
atmosphere and causing surface warming when deposited on ice and snow (IPCC 2021). Black carbon also
influences cloud development, but the direction and magnitude of this forcing is an area of active research.
Global Warming Potentials
A GWP is a quantified measure of the relative globally averaged radiative forcing impacts of emissions of a
particular greenhouse gas over time (see Table 1-2). It is defined as the accumulated radiative forcing within a
specific time horizon caused by emitting 1 kilogram (kg) of the gas, relative to that of the reference gas CO2 (IPCC
2021). Direct radiative effects occur when the gas itself absorbs radiation. Indirect radiative forcing occurs when
chemical transformations involving the original gas produce a gas or gases that are greenhouse gases, or when a
gas influences other radiatively important processes such as the atmospheric lifetimes of other gases. The
reference gas used is CO2, and therefore GWP-weighted emissions are measured in CO2 equivalent (CO2 Eq.).49 For
example, the relationship between a kg of emissions of a gas and a kg of CO2 Eq. emissions can be expressed as
follows and also adapted to other units (e.g. metric tons, etc.):
47 Carbonaceous aerosols are aerosols that are comprised mainly of carbon and hydrogen. Those carbonaceous aerosols with
more hydrogen are classified as "organic carbon", and are generally reflective, while the aerosols that are nearly pure carbon
are classified as "black carbon" (also referred to as "soot") and can absorb light (IPCC 2021).
48 Volcanic activity can inject significant quantities of aerosol producing sulfur dioxide and other sulfur compounds into the
stratosphere, which can result in a longer lasting negative forcing effect (i.e., a few years) (IPCC 2021).
49 Carbon comprises 12/44ths of carbon dioxide by weight.
1-8 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
Equation 1-1: Calculating CO2 Equivalent Emissions
kg C02 Eq. = (kg emission of gas) x (GWP)
where,
kg CO2 Eq.
kt
GWP
= kilograms of C02 equivalent
= kilograms (equivalent to a thousand metric grams)
= Global warming potential
GWP values allow for a comparison of the impacts of emissions and reductions of different gases. According to the
IPCC, GWPs typically have an uncertainty of ±40 percent.
All estimates are provided throughout the report in both MMT CO2 equivalents and unweighted units. Recent
decisions under the UNFCCC require Parties to use 100-year GWP values from the IPCC Fifth Assessment Report
(AR5) for calculating CC>2-equivalent emissions in their national reporting by the end of 2024.
...Decides that, until it adopts a further decision on the matter, the global warming potential values
used by Parties in their reporting under the Convention to calculate the carbon dioxide equivalence of
anthropogenic greenhouse gas emissions by sources and removals by sinks shall be based on the
effects of greenhouse gases over a 100-year time horizon as listed in table 8.A.1 in appendix 8. A to the
contribution of Working Group I to the Fifth Assessment Report of the Intergovernmental Panel on
Climate Change,50 excluding the value for fossil methane.51
This reflects updated science and ensures that national greenhouse gas inventories reported by all nations are
comparable. In preparation for upcoming UNFCCC requirement,52 this report reflects CC>2-equivalent greenhouse
gas totals using 100-year AR5 GWP values. A comparison of emission values with the previously used 100-year
GWP values from IPCC Fourth Assessment Report (AR4) (IPCC 2007), and the IPCC Sixth Assessment Report (AR6)
(IPCC 2021) values can be found in Annex 6.1 of this report. The 100-year GWP values used in this report are listed
below in Table 1-2.
Greenhouse gases with relatively long atmospheric lifetimes (e.g., CO2, CFU, N2O, HFCs, PFCs, SF6, NF3) tend to be
evenly distributed throughout the atmosphere, and consequently global average concentrations can be
determined. The short-lived gases such as water vapor, carbon monoxide, tropospheric ozone, ozone precursors
(e.g., NOx, and NMVOCs), and tropospheric aerosols (e.g., SO2 products and carbonaceous particles), however, vary
regionally, and consequently it is difficult to quantify their global radiative forcing impacts. Parties to the UNFCCC
have not agreed upon GWP values for these gases that are short-lived and spatially inhomogeneous in the
atmosphere.
Table 1-2: Global Warming Potentials and Atmospheric Lifetimes (Years) Used in this Report
Gas Atmospheric Lifetime GWP-1
C02 See footnote15 1
CH4c 12.4 28
N20 121 265
50 Intergovernmental Panel on Climate Change. 2013. Climate Change 2013: The Physical Science Basis. Contribution of
Working Group I to the Fifth Assessment Report of the Intergovernmental Panel on Climate Change. TF Stocker, D Qin, G-K
Plattner, et al. (eds.). Cambridge and New York: Cambridge University Press. Available at http://www.ipcc.ch/report/ar5/wgl.
51 See paragraphs 1 and 2 of the decision on common metrics adopted at the 27th UNFCCC Conference of Parties (COP27),
available online at https://unfccc.int/sites/default/files/resource/cp2022 IQaOl E.pdf.
52 See Annex to decision 18/CMA.l, available online at https://unfccc.int/sites/default/files/resource/CMA2018 03a02E.pdf
The Paris Agreement reporting guidelines also clarified use of the 100-year GWPs listed in table 8.A.1 in Annex 8.A of Chapter 8
of the Fifth Assessment Report (AR5) of the Intergovernmental Panel on Climate Change, excluding the value for fossil methane.
United Nations Framework Convention on Climate Change, see paragraph 25 of Decision 5/CMA.3 available online at
https://unfccc.int/sites/default/files/resource/CMA2021 L10a2E.pdf.
Introduction 1-9
-------
HFC-23
HFC-32
HFC-41d
HFC-125
HFC-134a
HFC-143a
HFC-152a
HFC-227ea
HFC-236fa
CF4
c2f6
C3Fs
c-C4Fs
sf6
nf3
Other Fluorinated Gases
222
5.2
2.8
28.2
13.4
47.1
1.5
38.9
242
50,000
10,000
2,600
3,200
3,200
500
12,400
677
116
3,170
1,300
4,800
138
3,350
8,060
6,630
11,100
8,900
9,540
23,500
16,100
See Annex 6
a 100-year time horizon.
b For a given amount of C02 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.
c 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 C02 is not included.
d See Table A-l of 40 CFR Part 98
Source: IPCC(2013).
Box 1-2: The IPCC Sixth Assessment Report and Global Warming Potentials
In 2021, the IPCC published its Sixth Assessment Report (AR6), which updated its comprehensive scientific
assessment of climate change. Within the AR6 report, the GWP values of gases were revised relative to
previous IPCC assessment reports, e.g., the IPCC Fifth Assessment Report (AR5) (IPCC 2014). Although the
AR5 GWP values are used throughout this report, consistent with UNFCCC reporting requirements, it is
straight-forward to review the changes to the GWP values and their impact on estimates of the total GWP-
weighted emissions of the United States. In the AR6, the IPCC used more recent estimates of the
atmospheric lifetimes and radiative efficiencies of some gases and updated background concentrations. The
AR6 now includes climate-carbon feedback effects for non-CC>2 gases, improving the consistency between
treatment of CO2 and non-CC>2 gases. Indirect effects of gases on other atmospheric constituents (such as the
effect of methane on ozone) have also been updated to match more recent science.
Table 1-3 presents the new GWP values, relative to those presented in the AR5, using the 100-year time
horizon common to Paris Agreement and UNFCCC reporting.53 Updated reporting guidelines under the Paris
Agreement require the United States and other countries to shift to use of the IPCC Fifth Assessment Report
(AR5) (IPCC 2013) 100-year GWP values (without feedbacks) for national inventory reporting.54 All estimates
provided throughout this report are also presented in unweighted units. For informational purposes,
emission estimates that use 100-year GWPs from other recent IPCC Assessment Reports are presented in
detail in Annex 6.1 of this report.
53 See Decision 7/CP.27 included in https://unfccc.int/sites/default/files/resource/cp2022 lOaOl E.pdf.
54 See https://unfccc.int/process-and-meetines/transparencv-and-reportine/reportine-and-review-under-the-paris-aereement.
1-10 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
Table 1-3: Comparison of 100-Year GWP values
100-Year GWP Values
Comparisons to AR5
AR5 with
AR5 with
Gas
AR5-
feedbacks'*
AR6'-
feedbacks'*
AR61-
C02
1
1
1
NC
NC
CH4d
28
34
27
6
(1)
N20
265
298
273
33
8
HFC-23
12,400
13,856
14,600
1,456
2,200
HFC-32
677
817
771
140
94
HFC-41
116
141
135
25
19
HFC-125
3,170
3,691
3,740
521
570
HFC-134a
1,300
1,549
1,530
249
230
HFC-143a
4,800
5,508
5,810
708
1,010
HFC-152a
138
167
164
29
26
HFC-227ea
3,350
3,860
3,600
510
250
HFC-236fa
8,060
8,998
8,690
938
630
cf4
6,630
7,349
7,380
719
750
c2f6
11,100
12,340
12,400
1,240
1,300
C3Fs
8,900
9,878
9,290
978
390
c-C4Fs
9,540
10,592
10,200
1,052
660
sf6
23,500
26,087
24,300
2,587
800
nf3
16,100
17,885
17,400
1,785
1,300
NC (No Change)
a The GWP values in this column reflect values used in this report from AR5 excluding climate-carbon feedbacks and
the value for fossil methane.
b The GWP values in this column are from the AR5 report but include climate-carbon feedbacks for the non-C02
gases in order to be consistent with the approach used in calculating the C02 lifetime.
c The GWP values in this column are from the AR6 report.
d The GWP of CH4 includes the direct effects and those indirect effects due to the production of tropospheric ozone
and stratospheric water vapor. Including the indirect effect due to the production of C02 resulting from methane
oxidation would lead to an increase in AR5 methane GWP values by 2 for fossil methane and is not shown in this
table.
Note: Parentheses indicate negative values.
Sources: IPCC (2021), IPCC (2013), IPCC (2007), IPCC (2001), IPCC (1996).
1.2 National Inventory Arrangements
The U.S. Environmental Protection Agency (EPA), in cooperation with other U.S. government agencies, prepares
the Inventory of U.S. Greenhouse Gas Emissions and Sinks. A wide range of agencies and individuals are involved in
supplying data to, planning methodological approaches and improvements, reviewing, or preparing portions of the
Inventory—including federal and state government authorities, research and academic institutions, industry
associations, and private consultants.
Within EPA, the Office of Atmospheric Protection (OAP) is the lead office responsible for the emission and removal
calculations provided in the Inventory, as well as the completion of the National Inventory Report including the
Common Reporting Tables (CRTs). EPA's Office of Transportation and Air Quality (OTAQ) and Office of Research
and Development (ORD) are also involved in calculating emissions and removals for the Inventory. The U.S.
Department of State (DOS) serves as the overall national focal point to the Paris Agreement and the UNFCCC, and
EPA's OAP serves as the National Inventory Focal Point for this report, including responding to technical questions
and comments on the U.S. Inventory. EPA staff coordinate the annual methodological choice, activity data
collection, emission and removal calculations, uncertainty assessment, QA/QC processes, and improvement
Introduction 1-11
-------
planning at the individual source and sink category level. EPA's inventory coordinator manages overall compilation
of the entire Inventory into the proper reporting format for submission under the Paris Agreement and the
UNFCCC and is responsible for the synthesis of information along with the consistent application of cross-cutting
IPCC good practice across the Inventory.
Several other government agencies contribute to the collection and analysis of the necessary underlying activity
data used in the Inventory calculations via formal (e.g., interagency agreements) and informal relationships, in
addition to the calculation of estimates integrated in the report (e.g., U.S. Department of Agriculture's U.S. Forest
Service and Agricultural Service, National Oceanic and Atmospheric Administration, Federal Aviation
Administration, and Department of Defense). Other U.S. agencies provide official data for use in the Inventory. The
U.S. Department of Energy's Energy Information Administration provides national fuel consumption data and the
U.S. Department of Defense provides data on military fuel consumption and use of bunker fuels. Other U.S.
agencies providing activity data for use in EPA's emission calculations include: the U.S. Department of Agriculture,
National Oceanic and Atmospheric Administration, the U.S. Geological Survey, the Federal Highway
Administration, the Department of Transportation, the Bureau of Transportation Statistics, the Department of
Commerce, and the Federal Aviation Administration. Academic and research centers also provide activity data and
calculations to EPA, as well as individual companies participating in voluntary outreach efforts with EPA. EPA
engages with agencies regularly on data needs and improvements to ensure sufficient activity collection for annual
compilation of estimates. Finally, EPA as the National Inventory Focal Point, in coordination with the U.S.
Department of State, officially submits the Inventory under the Paris Agreement and the UNFCCC each April by the
reporting deadline.
Figure 1-1: National Inventory Arrangements and Process Diagram
United States Greenhouse Gas Inventory Institutional Arrangements
1. Data Collection
Energy Data Sources
Agriculture and
LULUCF Data Sources
Industrial Processes
and Product Use Data
Sources
Waste Data Sources
2. Emissions and
Removals Calculations
and Uncertainty
Assessment
U.S. Environmental
Protection Agency
Other U.S.
Government Agencies
(USFS, NOAA,
DOD, FA A)
3. Inventory
Compilation
(including overall
uncertainty, report and
reporting table compilation)
U.S. Environmental
Protection Agency
Inventory Compiler
4. Inventory
Submission
U.S. Department
of State
United Nations
Framework
Convention on
Climate Change
QA/QC and Archiving
1-12 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
Overview of Inventory Data Sources by Source and Sink Category
Energy
Agriculture and LULUCF
IPPU
Waste
U.S. Energy Information
Administration
USDA U.S. Forest Service Forest
Inventory and Analysis Program
(FIA)
EPA Greenhouse Gas Reporting EPA Greenhouse Gas
Program (GHGRP) Reporting Program (GHGRP)
U.S. Department of Commerce USDA Natural Resource
- Bureau of the Census Conservation Service (NRCS)
U.S. Geological Survey (USGS)
National Minerals Information
Center
EPA Office of Land and
Emergency Management
(OLEM)
U.S. Department of Defense -
Defense Logistics Agency
USDA National Agricultural
Statistics Service (NASS) and
Agricultural Research Service
(ARS)
American Chemistry Council
(ACC)
EPA Clean Watershed Needs
Survey (CWNS)
U.S. Department of Homeland
Security
EPA Office of Research and
Development (ORD)
American Iron and Steel
Institute (AISI)
American Housing Survey
U.S. Department of
Transportation - Federal
Highway Administration
U.S. Fish and Wildlife Service
U.S. International Trade
Commission (USITC)
Data from research studies,
trade publications, and
industry associations
U.S. Department of
Transportation - Federal
Aviation Administration
U.S. Department of Agriculture
(USDA) Animal and Plant Health
Inspection Service (APHIS)
Air-Conditioning, Heating, and
Refrigeration Institute
U.S. Department of
Transportation & Bureau of
Transportation Statistics
Association of American Plant
Food Control Officials (AAPFCO)
Data from other U.S.
government agencies, research
studies, trade publications, and
industry association
U.S. Department of Labor -
Mine Safety and Health
Administration
National Oceanic and
Atmospheric Administration
(NOAA)
UNEP Technology and
Economic Assessment Panel
(TEAP)
U.S. Department of Energy and EPA Office of Land and Emergency
its National Laboratories Management (OLEM)
EPA Acid Rain Program
USDA Farm Service Agency
EPA MOVES Model
U.S. Geological Survey (USGS)
EPA Greenhouse Gas Reporting U.S. Department of the Interior
Program (GHGRP) (DOI) - Bureau of Land
Management (BLM)
U.S. Department of Labor - EPA Office of Land and Emergency
Mine Safety and Health Management (OLEM)
Administration
American Association of Alaska Department of Natural
Railroads Resources
American Public Transportation U.S. Department of Commerce-
Association Bureau of the Census
U.S. Department of Interior - Data from research studies, trade
Bureau of Ocean Energy publications, and industry
Management associations
Federal Energy Regulatory
Commission
Data from research studies,
trade publications, and industry
associations
Note: This table is not an exhaustive list of all data sources.
Introduction 1-13
-------
1.3 Inventory Preparation Process
This section describes EPA's approach to preparing the annual U.S. Inventory, which includes both the National
Inventory Document (NID) and Common Reporting Tables (CRTs). The inventory coordinator at EPA, with support
from the cross-cutting compilation staff, is responsible for coordinating aggregation of all emission and removal
estimates, conducting the overall uncertainty analysis of Inventory emissions and trends over time, and ensuring
consistency and quality throughout the NID and CRTs. Emission and removal calculations, including associated
uncertainty analysis for individual sources and/or sink categories are the responsibility of individual source and
sink category leads, who are most familiar with each category, underlying data, and the unique national
circumstances relevant to its emissions or removals profile. Using IPCC methodological decision trees and
suggested good practice guidance, the individual leads determine the most appropriate methodology and collect
the relevant activity data to use in the emission and removal calculations, based upon their expertise in the source
or sink category, as well as coordinating with researchers and expert consultants familiar with the sources and
sinks. Each year, the coordinator oversees a multi-stage process for collecting information from each individual
source and sink category lead to compile all information and data for the Inventory.
Methodology Development, Data Collection, and Emissions
and Sinks Estimation
Source and sink category leads at EPA coordinate the collection of input data (e.g., activity data and other
information) and, as necessary, evaluate or develop the estimation methodology for the individual source and/or
sink categories. Because EPA has been leading preparation of the Inventory for many years, for most source and
sink categories, the methodology for the previous year is applied to the new "current" year of the Inventory, and
inventory analysts collect any new data or update data that have changed from the previous year. If estimates for
a new source or sink category are being developed for the first time, or if the methodology is changing for an
existing category (e.g., the United States is implementing improvement efforts to apply a higher tiered approach
for that category), then the source and/or sink category lead will develop and implement the new or refined
methodology, gather the appropriate activity data and other information (e.g., emission factors or in some cases
direct emission measurements) for the entire time series, and conduct any further category-specific review with
involvement of relevant experts from industry, government, and universities (see Chapter 9 and Box ES-3 on EPA's
approach to recalculations).
Once the methodology is in place and the data are collected, the individual source and sink category leads
calculate emission and removal estimates. The individual leads then update or create the relevant national
inventory document text and accompanying annexes for the Inventory. Source and sink category leads are also
responsible for completing the relevant sectoral background tables of the CRTs, conducting quality control (QC)
checks, preparing relevant category materials for QA, or expert reviews, category-level uncertainty assessments,
and reviewing data for publication in EPA's GHG Data Explorer.55
The treatment of confidential business information (CBI) in the Inventory is based on EPA internal guidelines, as
well as regulations56 applicable to the data used. EPA has specific procedures in place to safeguard CBI during the
inventory compilation process. When information derived from CBI data is used for development of inventory
calculations, EPA procedures ensure that these confidential data are sufficiently aggregated to protect
confidentiality while still providing useful information for analysis. For example, within the Energy and Industrial
55 See https://cfpub.epa.gov/ehedata/inventorvexplorer/.
56 40 CFR Part 2, Subpart B titled "Confidentiality of Business Information" which is the regulation establishing rules governing
handling of data entitled to confidentiality treatment. See https://www.ecfr.gov/cgi-bin/text-
idx?SID=a764235c9eadf9afe05fe04c07a28939&mc=true&node=sp40.1.2.b&rgn=div6.
1-14 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
Processes and Product Use (IPPU) sectors, EPA has used aggregated facility-level data from the Greenhouse Gas
Reporting Program (GHGRP) to develop, inform, and/or quality-assure U.S. emission estimates. In 2014, EPA's
GHGRP, with industry engagement, compiled criteria that would be used for aggregating its confidential data to
shield the underlying CBI from public disclosure.57 In the Inventory, EPA is publishing only data values that meet
the GHGRP aggregation criteria.58 Specific uses of aggregated facility-level data are described in the respective
methodological sections within those chapters. In addition, EPA uses historical data reported voluntarily to EPA via
various voluntary initiatives with U.S. industry (e.g., EPA Voluntary Aluminum Industrial Partnership (VAIP)) and
follows guidelines established under the voluntary programs for managing CBI.
Data Compilation and Archiving
The inventory coordinator at EPA with support from the data/document manager collects the source and sink
categories' descriptive text and annexes, and also aggregates the emission and removal estimates into a summary
data file that links the individual source and sink category data files together. This summary data file contains all of
the essential data in one central location, in formats commonly used in the Inventory document. In addition to the
data from each source and sink category, other national trend and related data are also gathered in the summary
sheet for use in the Executive Summary, Introduction, and Trends chapters of the Inventory report (e.g., GDP,
population, energy use). Similarly, the recalculation analysis and key category analysis are completed in a separate
data file based on output from the summary data file. The uncertainty estimates for each source and sink category
are also aggregated into uncertainty summary data files that are used to conduct the overall Inventory uncertainty
analysis (see Section 1.7). A Microsoft SharePoint work site, maintained within EPA's IT infrastructure by the
inventory coordinator, provides a platform for facilitating collaboration on the national inventory report
preparation during each compilation phase, but also the efficient storage and archiving of electronic document and
data files each annual cycle. Previous final published inventories are also maintained on a report archive page on
EPA's Greenhouse Gas Emissions website.59
National Inventory Document (NID) Preparation
The NID is compiled from the sections developed by each individual source or sink category lead. In addition, the
inventory coordinator prepares a brief overview of each chapter that summarizes the emissions and removals from
all sources and sinks discussed in the chapters. Also at this time, the Executive Summary, Introduction, Trends in
Greenhouse Gas Emissions and Removals, and Recalculations and Improvements chapters are drafted, to reflect
the trends and impact from improvements for the time series of the current Inventory. The analysis of trends
necessitates gathering supplemental data, including annual climate, economic activity and gross domestic product,
population, atmospheric conditions, and the annual use of electricity, energy, fossil and non-fossil fuels. Changes in
these data are used to explain the trends observed in greenhouse gas emissions in the United States. Furthermore,
specific factors that affect individual sectors are researched and discussed. Many of the factors that affect
emissions are included in the Inventory document as separate analyses or side discussions in boxes within the text.
Finally, the uncertainty analysis and key category analysis are compiled and updated in the report as part of final
analysis steps. Throughout the report text boxes are also created to provide additional documentation (e.g.,
definitions) and/or examine the data aggregated in different ways than in the remainder of the document, such as
a focus on transportation activities or emissions from electricity generation. The document is prepared to align
with the Paris Agreement and UNFCCC reporting guidelines for National Inventory Reports while also reflecting
national circumstances.
57 Federal Register Notice on "Greenhouse Gas Reporting Program: Publication of Aggregated Greenhouse Gas Data." See pp.
79 and 110 of notice at https://www.EPO.gOv/fdsvs/pkg/FR-2014-06-09/pdf/2 5.pdf.
58 U.S. EPA Greenhouse Gas Reporting Program. Developments on Publication of Aggregated Greenhouse Gas Data, November
25, 2014. See http://www.epa.gov/ghgreporting/confidential-business-information-ghg-reporting.
59 See https://www.epa.gov/ghgemissions/us-greenhouse-gas-inventorv-report-archive.
Introduction 1-15
-------
Common Reporting Tables (CRTs) Compilation
The CRTs are compiled from individual time series input data sheets completed by each individual source or sink
category lead, which contain emissions and/or removals and activity data, estimates, methodological and
completeness notations and associated explanations. The inventory coordinator and cross-cutting compilation
staff import the U.S. category and subcategory background data into the UNFCCC's Enhanced Transparency
Framework Reporting Tools to export CRTs, assuring consistency and completeness across all sectoral tables. The
summary reports for emissions and removals, methods, and emission factors used, the summary tables indicating
completeness of estimates (i.e., notation key NE/IE tables), the recalculation tables, and the emission and removal
trends tables are automatically compiled by the online reporting software and reviewed by the inventory
coordinator with support from the cross-cutting compilation staff. Internal automated quality checks within the
software, as well as checks by the cross-cutting and category leads, are completed for the entire time series of
CRTs before submission.
QA/QC and Uncertainty
Quality assurance and quality control (QA/QC) and uncertainty analyses are guided by the QA/QC and inventory
coordinators, who help maintain the QA/QC plan and the overall uncertainty analysis procedures (see sections on
QA/QC and Uncertainty, below) in collaboration with the broader inventory compilation team. The QA/QC
coordinator works closely with the Inventory coordinator and source and sink category leads to ensure that a
consistent QA/QC plan is implemented across all inventory categories. Similarly, the inventory coordinator ensures
the uncertainty analysis is implemented consistently across all categories. The inventory QA/QC plan, outlined in
Section 1.6 and Annex 8, is consistent with the quality assurance procedures outlined by EPA and IPCC good
practices. The QA/QC and uncertainty findings also inform overall improvement planning, and specific
improvements are noted in the Planned Improvements sections of respective categories. QA processes are
outlined below.
Expert, Public, and UNFCCC/Paris Reviews
The compilation of the Inventory includes a two-stage review or QA process, in addition to international technical
expert review following submission of the report under the UNFCCC and Paris Agreement. EPA publishes responses
to comments received during both expert and public reviews with the publication of the final report on its
website.60 Responses to UNFCCC and Paris reviews are included in Annex 8 of this document.
During the first stage of review, i.e., the 30-day expert review period, a first draft of updated sectoral chapters are
sent to technical experts who are not directly involved in preparing estimates. The purpose of the expert review is
to provide an objective review of the methodological approaches and data sources used in the current Inventory,
especially for sources and sinks which have experienced any changes since the previous Inventory. Expert review
follows good practices from EPA's Peer Review handbook, i.e., the review is organized by sector, and reviewers are
provided a guidance memo and charge questions to facilitate their review.61 Expert reviewers include other
federal agency staff, researchers, industry experts, and others who have technical knowledge of the data, industry,
and methods. EPA reviews and updates expert participation and outreach on an annual basis prior to each expert
review cycle. Experts are identified in various ways; for example, many reach out to EPA with technical feedback
and are added to the expert reviewer list. Reviewers are also identified through direct outreach by inventory staff
based on expertise. Currently, EPA's expert list includes nearly 300 experts across all sectors. Once comments are
received, they are reviewed by the source or sink lead and addressed in several ways. For example, comments
suggesting methodological clarifications may be incorporated into methodological discussions prior to the next
60 See https://www.epa.gov/eheemissions/draft-inventorv-us-greenhouse-gas-emissions-and-sinks-1990-2022.
61 See https://www.epa.gov/osa/peer-review-handbook-4th-edition-2015.
1-16 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
review phase, while comments citing new literature or data will be noted for review as part of planned
improvements.
Following expert review, a second draft of the document, including cross-cutting synthesis chapters, is released for
a 30-day public review through a notice in the U.S. Federal Register. The entire draft Inventory document is
published on the EPA website. The public review period is open to the entire U.S. public. Comments are submitted
and tracked using an online electronic docket that is accessible to the general public as well. Similar to expert
review, some comments received may require further discussion with commenters, other experts and/or
additional research. Specific Inventory improvements requiring further analysis as a result of comments are noted
in the relevant category's Planned Improvement section.
As mentioned above, following completion and submission of the report under the UNFCCC and the Paris
Agreement, the report also undergoes review by an international team of independent experts for adherence to
UNFCCC/Paris reporting guidelines and consistency with IPCC methodological guidance.62 Feedback from all
review processes that contribute to improving inventory quality over time are described within each planned
improvement section and further in Annex 8. See also the Improvement Planning process discussed below.
Final Submittal and Publication under the Paris Agreement
and the UNFCCC
After the final revisions to incorporate any comments from the Expert Review and Public Review periods, EPA
prepares the final NIR, which includes the NID and the accompanying CRTs for electronic reporting. Prior to
submission, EPA's Office of Atmospheric Protection briefs senior leadership on reporting findings and
improvements since the previous report, along with an overview of feedback from the expert and public review
processes.
EPA, as the National Inventory Focal Point, sends the official submission of the U.S. Inventory under the Paris
Agreement and the UNFCCC using the UN's reporting software, coordinating with the U.S. Department of State,
the overall UNFCCC focal point. Concurrently, the report is also published on EPA's website.63 On EPA's website,
users can also visualize and download the current time-series estimates from the GHG Inventory Data Explorer
Tool,64 and also download more detailed data presented in tables within the report and report annex in CSV
format.
Improvement Planning
Each year, several emission and sink estimates in the Inventory of U.S. Greenhouse Gas Emissions and Sinks are
recalculated and revised, through the use of better methods and/or data with the goal of improving inventory
quality and reducing uncertainties, including the transparency, completeness, consistency, and overall usefulness
of the report. In this effort, the United States follows the 2006 IPCC Guidelines (IPCC 2006) and its 2019
Refinement, which state, "Both methodological changes and refinements over time are an essential part of
improving inventory quality. It is good practice to change or refine methods when available data have changed; the
previously used method is not consistent with the IPCC guidelines for that category; a category has become key;
the previously used method is insufficient to reflect mitigation activities in a transparent manner; the capacity for
inventory preparation has increased; improved inventory methods become available; and/or for correction of
errors." The EPA's OAP coordinates improvement planning across all sectors and also cross-cutting analyses based
on annual review and input from the technical teams leading compilation of each sector's estimates, including
62 See https://unfccc.int/process-and-meetings/transparencv-and-reporting/reporting-and-review-under-the-
convention/greenhouse-gas-inventories-annex-i-parties/review-process.
63 See https://www.epa.gov/ghgemissions/inventory-us-greenhouse-gas-emissions-and-sinks.
64 See https://cfpub.epa.gov/ghgdata/inventorvexplorer/.
Introduction 1-17
-------
continuous improvements to the overall data and document compilation and QA/QC processes. Planned
improvements are identified through QA/QC processes (including completeness checks), the key category analysis,
and the uncertainty analysis. The inventory coordinator, with input from EPA source and sink category leads,
maintains a log of all planned improvements, by sector and cross-cutting, tracking the category significance,
specific category improvement, prioritization, anticipated time frame for implementation of each proposed
improvement, and status of progress in implementing improvement. Improvements for significant or key
categories are usually prioritized across all improvements unless effort would require disproportionate levels of
effort and resources relative to improvements for other key categories to address.
1.4 Methodology and Data Sources
Emissions and removals of greenhouse gases from various source and sink categories have been estimated using
methodologies that are consistent with the 2006IPCC Guidelines for National Greenhouse Gas Inventories (IPCC
2006) and its supplements and refinements. To a great extent, this report makes use of published official economic
and physical statistics for activity data, emission factors and other key parameters as inputs to the methods
applied. Depending on the category, activity data can include fuel consumption or deliveries, vehicle-miles
traveled, raw material processed, or commodity produced, etc. Emission factors are factors that relate quantities
of emissions to an activity. For more information on data sources see Section 1.2 above, Box 1-1 on use of GHGRP
data, and categories' methodology sections for more information on other data sources. In addition to official
statistics, the report utilizes findings from academic studies, trade association surveys and statistical reports, along
with expert judgment, consistent with the 2006 IPCC Guidelines.
The methodologies provided in the 2006 IPCC Guidelines represent foundational methodologies for a variety of
source and sink categories, and many of these methodologies continue to be improved and refined as new
research and data become available. This report uses those IPCC methodologies when applicable, and supplements
them with refined guidance, other available country-specific methodologies and data where possible (e.g., EPA's
GHGRP). For example, as noted earlier in this chapter, this report does apply recent supplements and refinements
to 2006 IPCC Guidelines in estimating emissions and removals from coal mining, wastewater treatment and
discharge, low voltage anode effects (LVAE) during aluminum production, drained organic soils, and management
of wetlands, including flooded lands. Choices made regarding the methodologies and data sources used are
provided in the Methodology and Time-Series Consistency discussion of each category within each sectoral chapter
of the report, applying higher tiered methods when feasible, especially for key categories consistent with
methodological decision trees. Where additional detail is helpful and necessary to explain methodologies and data
sources used to estimate emissions, complete documentation is provided in the annexes as indicated in the
methodology sections of those respective source categories (e.g., Annex 3.13 for forest land remaining forest land
and land converted to forest land). Methods used for key categories (discussed below) are summarized in Annex 1.
1.5 Key Categories
The 2006 IPCC Guidelines (IPCC 2006) and 2019 Refinement to the 2006 IPCC Guidelines (IPCC 2019) define key
categories as "inventory categories which individually, or as a group of categories (for which a common method,
emission factor and activity data are applied) are prioritized within the national inventory system because their
estimates have a significant influence on a country's total inventory of greenhouse gases in terms of the absolute
level, the trend, or the level of uncertainty in emissions or removals. Whenever the term key category is used, it
1-18 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
includes both source and sink categories."65 A key category analysis identifies source or sink categories for
focusing efforts to improve overall Inventory quality, including additional review when feasible.
The 2006IPCC Guidelines (IPCC 2006) and its 2019 Refinement (2019) define several approaches, both quantitative
and qualitative, to conduct a key category analysis and identify key categories both in terms of absolute level and
trend, along with consideration of uncertainty. This report employs all approaches to identify key categories for
the United States. The first approach, Approach 1, identifies significant or key categories without considering
uncertainty in its calculations. An Approach 1 level assessment identifies all source and sink categories that
cumulatively account for 95 percent of total level, i.e., total emissions (gross) in a given year when assessed in
descending order of absolute magnitude. The level analysis was performed twice, including and excluding sources
and sinks from the land use, land-use change, and forestry (LULUCF) sector categories. Similarly, an Approach 1
trend analysis can identify categories with trends that significantly influence overall trends by identifying all source
and sink categories that cumulatively account for 95 percent of the sum all the trend assessments (e.g., percent
change relative to national trend) when sorted in descending order of absolute magnitude.
The next method, Approach 2, was then implemented to identify any additional key categories not already
identified from the Approach 1 level and trend assessments by considering uncertainty. The Approach 2 analysis
differs from Approach 1 by incorporating each category's uncertainty assessments in its calculations and was also
performed twice, including and excluding LULUCF categories. An Approach 2 level assessment identifies all sources
and sink categories that cumulatively account for 90 percent of the sum of all level assessments when sorted in
descending order of magnitude. Similarly, an Approach 2 trend analysis can identify categories whose trends
contribute significantly to overall trends weighing the relative trend difference with the category's relative
uncertainty assessment for 2022.
For 2022, based on the key category analysis, excluding the LULUCF sector and uncertainty, 34 categories
accounted for 95 percent of emissions. Four categories account for 55 percent of emissions: CO2 from road
transport-related fuel combustion, CO2 from coal-fired electricity generation, CO2 from gas fired electricity
generation, and CChfrom gas-fired industrial processes. When considering uncertainties, additional categories
such as emissions from substitutes for ozone depleting substances in aerosols were also identified as a key
category. In the trend analysis, 32 categories were identified as key categories, and when considering
uncertainties, 7 additional categories were identified as key. The trend analysis shows that CO2 emissions from
coal-fired electricity generation, in addition to CO2 from gas fired electricity generation, CO2 from road transport
related combustion, and HFC and PFC emissions from substitutes for ozone depleting substances in the
refrigeration and air conditioning sector are also significant with respect to trends over the time series.
When considering the contribution of the LULUCF sector to 2022 emissions and removals, 42 categories accounted
for 95 percent of emissions and sinks, with the most significant category from LULUCF being net CO2 emission from
forest land remaining forest land. When considering uncertainties and the contribution of the LULUCF sector,
additional categories such as net CO2 emissions from grassland remaining grassland were also identified as a key
category. In the trend analysis, 40 categories were identified as key, and when considering uncertainties, 2
additional categories were identified as key.
Finally, in addition to conducting Approach 1 and 2 level and trend assessments as described above, a qualitative
assessment of the source and sinks categories was conducted to capture any additional key categories that were
not identified using the previously described quantitative approaches. For this Inventory, no additional categories
were identified using qualitative criteria recommend by IPCC, but EPA continues to review its qualitative
assessment on an annual basis. Find more information on the key category analysis, including the approach to
disaggregation of inventory estimates, see Annex 1 to this report.
65 See Chapter 4 Volume 1, "Methodological Choice and Identification of Key Categories" in IPCC (2006) and IPCC (2019). See
http://www.ipcc-nggip.iges.or.ip/public/2QQ6gl/index.html.
Introduction 1-19
-------
Table 1-4: Summary of Key Categories for the United States (1990 and 2022) by Sector
Approach 1"
Approach 2 (includes uncertainty)-1
2022
CRT Code and Source/Sink
Greenhouse
Level
Trend
Level
Trend
Level
Trend Level
Trend
Emissions
Category
Gas
Without
Without
With
With
Without
Without With
With
(MMT CO.
LULUCF
LULUCF
LULUCF
LULUCF
LULUCF
LULUCF LULUCF
LULUCF
Eq.)
Energy
l.A.3.b Transportation: Road
C02
•
®
•
®
®
•
1,438.1
l.A.l Stationary Combustion
- Coal - Electricity
co2
•
®
•
®
®
•
851.5
Generation
l.A.l Stationary Combustion
- Natural Gas - Electricity
co2
•
®
•
®
•
•
659.3
Generation
1.A.2 Stationary Combustion
- Natural Gas - Industrial
co2
•
®
•
®
®
•
510.4
l.A.4.b Stationary
Combustion - Natural Gas -
co2
•
®
•
®
®
272.0
Residential
1.A.2 Stationary Combustion
- Oil - Industrial
co2
®
•
®
•
®
®
•
247.6
l.A.4.a Stationary
Combustion - Natural Gas -
co2
®
•
®
•
®
• •
•
192.3
Commercial
l.A.3.a Transportation:
Aviation
co2
®
•
®
•
®
®
165.6
1.A.5 Non-Energy Use of
Fuels
co2
®
®
®
®
102.8
l.A.3.e Transportation:
Other
co2
®
•
®
•
•
69.3
l.A.4.a Stationary
Combustion - Oil -
co2
®
•
®
•
65.1
Commercial
l.A.4.b Stationary
Combustion - Oil -
co2
®
•
®
•
62.1
Residential
1-20 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
Approach 1"
Approach 2 (includes uncertainty)-1
2022
CRT Code and Source/Sink
Greenhouse
Level
Trend
Level
Trend
Level Trend Level Trend
Emissions
Category
Gas
Without
Without
With
With
Without Without With With
(MMT CO.
LULUCF
LULUCF
LULUCF
LULUCF
LULUCF LULUCF LULUCF LULUCF
Eq.)
1.A.2 Stationary Combustion
- Coal - Industrial
C02
®
•
®
•
® • 0 •
43.0
l.A.3.d Transportation:
Domestic Navigation
co2
®
40.9
1.B.2 Natural Gas Systems
co2
®
36.5
1.A.3.C Transportation:
Railways
co2
®
32.5
1.B.2 Petroleum Systems
co2
•
•
•
•
• •
22.0
l.A.l Stationary Combustion
- Oil - Electricity
co2
®
•
®
•
O • •
20.5
Generation
1.A.5 Stationary Combustion
- Oil - U.S. Territories
co2
0
0
17.0
l.A.5.b Transportation:
co2
4.8
Military
l.A.4.a Stationary
Combustion - Coal -
co2
•
•
1.4
Commercial
l.A.4.b Stationary
Combustion - Coal -
co2
•
NO
Residential
1.B.2 Natural Gas Systems
ch4
®
•
®
•
® • ® •
173.1
l.B.l Fugitive Emissions
from Coal Mining
ch4
®
•
®
•
® ®
43.6
1.B.2 Petroleum Systems
ch4
®
•
®
•
® * ®
39.6
1.B.2 Abandoned Oil and
ch4
® ®
8.5
Natural Gas Wells
l.A.4.b Stationary
ch4
® • ® •
4.3
Combustion - Residential
l.A.l Stationary Combustion
- Coal - Electricity
n20
®
18.2
Generation
l.A.3.b Transportation: Road
N20
0
•
0
•
8.9
Introduction 1-21
-------
Approach 1"
Approach 2 (includes uncertainty)-1
2022
CRT Code and Source/Sink
Greenhouse
Level
Trend Level
Trend
Level Trend Level Trend
Emissions
Category
Gas
Without
Without With
With
Without Without With With
(MMT CO.
LULUCF
LULUCF LULUCF
LULUCF
LULUCF LULUCF LULUCF LULUCF
Eq.)
l.A.l Stationary Combustion
- Natural Gas - Electricity
N20
•
3.4
Generation
Industrial Processes and Product Use
2.A.1 Cement Production
co2
®
•
41.9
2.C.1 Iron and Steel
Production & Metallurgical
co2
®
•
® • 0 •
40.7
Coke Production
2.B.8 Petrochemical
Production
co2
®
•
28.8
2.B.3 Adipic Acid Production
n2o
•
•
2.1
2.F.1 Substitutes for Ozone
Depleting Substances:
Refrigeration and Air
HFCs, PFCs
•
• •
•
• • • •
144.6
Conditioning
2.F.4 Substitutes for Ozone
Depleting Substances:
HFCs, PFCs
•
•
• • • •
17.0
Aerosols
2.F.2 Substitutes for Ozone
Depleting Substances:
HFCs, PFCs
•
•
11.7
Foam Blowing Agents
2.B.9 Fluorochemical
Production
PFCs, HFCs,
SF6, NFb
0
• 0
•
0 • 0 •
7.8
2.G Electrical Equipment
PFCs, SF6
0
• 0
•
• •
5.1
2.C.3 Aluminum Production
PFCs
0
• 0
•
0.8
Agriculture
3.A.1 Enteric Fermentation:
Cattle
ch4
®
•
® • ® •
185.9
3.B.1 Manure Management:
ch4
37.7
Cattle
3.B.4 Manure Management:
Other Livestock
ch4
®
®
•
27.0
3.C Rice Cultivation
ch4
®
•
® ®
18.9
1-22 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
Approach 1"
Approach 2 (includes uncertainty)-1
2022
CRT Code and Source/Sink
Greenhouse
Level
Trend
Level
Trend
Level
Trend
Level
Trend
Emissions
Category
Gas
Without
Without
With
With
Without
Without
With
With
(MMT CO.
LULUCF
LULUCF
LULUCF
LULUCF
LULUCF
LULUCF
LULUCF
LULUCF
Eq.)
3.D.1 Direct Emissions from
Agricultural Soil
N20
®
®
®
®
262.5
Management
3.D.2 Indirect Emissions
from Applied Nitrogen
n2o
®
®
•
®
28.3
Waste
5.A Commercial Landfills
ch4
®
•
®
•
®
•
®
•
100.9
5.A Industrial Landfills
ch4
•
•
•
•
18.9
5.D Domestic Wastewater
Treatment
ch4
o
13.6
5.D Domestic Wastewater
Treatment
n2o
•
•
®
•
®
•
21.4
Land Use, Land-Use Change, and Forestry
4.E.2 Net Emissions from
Land Converted to
C02
®
®
68.2
Settlements
4.B.2 Net Emissions from
Land Converted to
C02
®
®
35.1
Cropland
4.C.2 Net Emissions from
Land Converted to
C02
®
®
25.6
Grassland
4.C.1 Net Emissions from
Grassland Remaining
C02
o •
®
13.4
Grassland
4.B.1 Net Removals from
Cropland Remaining
C02
• •
®
(31.7)
Cropland
4.A.2 Net Removals from
Land Converted to Forest
C02
®
®
(100.3)
Land
4.E.1 Net Removals from
Settlements Remaining
C02
®
®
(134.8)
Settlements
Introduction 1-23
-------
CRT Code and Source/Sink
Category
Greenhouse
Gas
Approach 1"
Approach 2 (includes uncertainty)-1
2022
Emissions
(MMT CO.
Eq.)
Level Trend Level Trend
Without Without With With
LULUCF LULUCF LULUCF LULUCF
Level Trend Level Trend
Without Without With With
LULUCF LULUCF LULUCF LULUCF
4.A.1 Net Removals from
Forest Land Remaining
Forest Land
4.D.1 Flooded Lands
Remaining Flooded Lands
C02
ch4
®
®
(787.0)
44.2
Subtotal of Key Categories Without LULUCF1,
Total Gross Emissions Without LULUCF
Percent of Gross Total Without LULUCF
6,169.7
6,343.2
97%
Subtotal of Key Categories With LULUCF'
Total Net Emissions With LULUCF
Percent of Net Total With LULUCF
5,285.4
5,488.9
96%
NO (Not Occurring)
a Symbols correspond to the year(s) in which a category is key: 1990 = o; 2022 = •; 1990 and 2022 = ®.
b Subtotal includes key categories from Level Approach 1 Without LULUCF, Level Approach 2 Without LULUCF, Trend Approach
Without LULUCF.
c Subtotal includes key categories from Level Approach 1 With LULUCF, Level Approach 2 With LULUCF, Trend Approach 1 With
Note: Parentheses indicate negative values (or sequestration).
1 Without LULUCF, and Trend Approach 2
LULUCF, and Trend Approach 2 With LULUCF.
1-24 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
As part of efforts to achieve its stated goals for inventory quality, transparency, and credibility, the United States
has developed a quality assurance and quality control plan designed to check, document, and improve the quality
of its inventory over time. QA/QC activities on the Inventory are undertaken within the framework of the U.S.
Quality Assurance/Quality Control and Uncertainty Management Plan (QA/QC plan)/or the U.S. Greenhouse Gas
Inventory: Procedures Manual for QA/QC and Uncertainty Analysis.
Key attributes of the QA/QC plan are summarized in Figure 1-2. These attributes include:
• Procedures and Forms: detailed and specific systems that serve to standardize the process of
documenting and archiving QA/QC implementation and related information, as well as to guide the
implementation of QA/QC and the analysis of uncertainty
• Implementation of Procedures: guidance on application of QA/QC procedures throughout the whole
Inventory development process from initial data collection, through preparation of the emission and
removal estimates, to publication of the Inventory, consistent with the 2006IPCC Guidelines.
• Quality Assurance (QA): process for implementing expert and public reviews for both the inventory
estimates and the Inventory report (which is the primary vehicle for disseminating the results of the
inventory development process). The expert technical review conducted by the UNFCCC supplements
these QA processes, consistent with the QA good practice and the 2006 IPCC Guidelines (IPCC 2006). See
Section 1.3 for more details on these QA processes.
• Quality Control (QC): application of General (Tier 1) and Category-specific (Tier 2) quality controls and
checks, as recommended by 2006 IPCC Guidelines (IPCC 2006), along with consideration of secondary data
and category-specific checks (additional Tier 2 QC) in parallel and coordination with the uncertainty
assessment; the development of protocols and templates, which provides for more structured
communication and integration with the suppliers of secondary information
• General (Tier 1) and Category-specific (Tier 2) Checks: quality controls and checks, as recommended by
IPCC Good Practice Guidance and 2006 IPCC Guidelines (IPCC 2006)
• Record Keeping: provisions to track which procedures have been followed, the results of the QA/QC,
uncertainty analysis, and feedback mechanisms for corrective action based on the results of the
investigations which provide for continual data quality improvement and guided research efforts moving
forward.
• Multi-Year Implementation: tracking the application of more involved QA/QC procedures which may take
more than one cycle to fully implement, especially for category-specific QC, prioritizing key categories in
conjunction with improvement planning (see Section 1.3).
• Interaction and Coordination: promoting communication within the EPA, across federal agencies and
departments, state government programs, and research institutions and consulting firms involved in
supplying data or preparing estimates for the Inventory. The QA/QC Management Plan itself is intended to
be revised and reflect new information that becomes available as the program develops, methods are
improved, or additional supporting documents become necessary.
Introduction 1-25
-------
Figure 1-2: Summary of Key QC Processes from U.S. QA/QC Plan
Data
Data
Calculating
Gathering
*
Documentation
Emissions
• Obtain data in
•
Contact reports
• Clearly label
electronic
for non-electronic
parameters, units,
format (if
communications
and conversion
possible)
•
Provide cell
factors
• Review data
references for
• Review data
input/calculation
primary data
input/calculation
workbooks
elements
workbooks
o Avoid
•
Obtain copies of
integrity
hardwiring
all data sources
o Equations
o Use data
•
List and location
o Units
+J
l/l
validation
of any
o Inputs and
"rc
o Protect cells
working/external
outputs
c
<
• Develop
data or
• Develop
>-
i
automatic
input/calculation
automated
o
4—1
checkers for:
workbooks
checkers for:
c
d)
o Outliers,
•
Document
o Input ranges
>
c
negative
assumptions
o Calculations
values, or
missing data
•
Complete QA/QC
checklists
o Emission
aggregation
o Variable types
•
CRF and summary
o Trend and IEF
match values
tab links
checks
o Time series
consistency
• Maintain
tracking tab for
status of
gathering efforts
• Check input data
•
Check citations in
• Reproduce
for transcription
data
calculations
errors
input/calculation
• Review time
• Inspect
workbooks and
series
automatic
text for accuracy
consistency
checkers
and style
• Review changes
4-»
IA
• Identify data
•
Check reference
in
TO
input/calculation
docket for new
data/consistency
c
<
workbooks
citations
with IPCC
o
modifications
•
Review
methodology
a
that could
documentation
§
provide
for any data /
additional
QA/QC checks
•
•
methodology
changes
Complete QA/QC
checklists
CRF and summary
tab links
Cross-Cutting
Coordination
• Common starting
versions for each
inventory year
• Utilize
unalterable
summary and
CRF tab for each
source data
input/calculation
workbook for
linking to a
master summary
workbook
• Follow strict
version control
procedures
• Document
QA/QC
procedures
1-26 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
Box 1-3: Examples of Verification Activities
Consistent with IPCC guidance for national greenhouse gas inventories, verification activities include
comparisons with emission or removal estimates prepared by other bodies and comparisons with estimates
derived from fully independent assessments, e.g., atmospheric concentration measurements. Verification
activities provide information to improve inventories and are part of the overall QA/QC system.
Use of Lower Tier Methods. The Paris Agreement/UNFCCC reporting guidelines require countries to complete a
"top-down" reference approach for estimating CO2 emissions from fossil fuel combustion in addition to their
"bottom-up" sectoral methodology for purposes of verification. This estimation method uses alternative
methodologies and different data sources than those contained in that section of the Energy chapter. The
reference approach estimates fossil fuel consumption by adjusting national aggregate fuel production data for
imports, exports, and stock changes rather than relying on end-user consumption surveys (see Annex 4 of this
report). The reference approach assumes that once carbon-based fuels are brought into a national economy,
they are either saved in some way (e.g., stored in products, kept in fuel stocks, or left unoxidized in ash) or
combusted, and therefore the carbon in them is oxidized and released into the atmosphere. Accounting for
actual consumption of fuels at the sectoral or sub-national level is not required.
Use of Ambient Measurements Systems for Validation of Emission Inventories. In following the Paris
Agreement and UNFCCC reporting requirements to develop and submit national greenhouse gas emission
inventories, the emissions and sinks presented in this report are organized by source and sink categories and
calculated using internationally accepted methods provided by the IPCC.66 Several recent studies have
estimated emissions at the national or regional level with estimated results that sometimes differ from EPA's
estimate of emissions. EPA has engaged with researchers on how remote sensing, ambient measurement, and
inverse modeling techniques for estimating greenhouse gas emissions could assist in improving the
understanding of inventory estimates. In working with the research community to improve national greenhouse
gas inventories, EPA follows guidance from the IPCC on the use of measurements and modeling to validate
emission inventories.67 An area of particular interest in EPA's outreach efforts is how ambient measurement
data can be used to assess estimates or potentially be incorporated into the Inventory in a manner consistent
with this Inventory report's transparency of its calculation methodologies, and the ability of inverse modeling to
attribute emissions and removals from remote sensing to anthropogenic sources, as defined by the IPCC for this
report, versus natural sources and sinks.
The 2019 Refinement to the IPCC 2006 Guidelines for National Greenhouse Gas Inventories (IPCC 2019) Volume
1 General Guidance and Reporting, Chapter 6: Quality Assurance, Quality Control and Verification notes that
emission estimates derived from atmospheric concentration measurements can provide independent data sets
as a basis for comparison with inventory estimates. The 2019 Refinement provides guidance on conducting such
comparisons (as summarized in Table 6.2 of IPCC [2019] Volume 1, Chapter 6) and provides guidance on using
such comparisons to identify areas of improvement in national inventories (as summarized in Box 6.5 of IPCC
[2019] Volume 1, Chapter 6). Further, it identified fluorinated gases as particularly suitable for such comparisons
due their limited natural sources, their generally long atmospheric lifetimes, and well-understood loss
mechanisms, which makes it relatively more straightforward to model their emission fluxes from observed mass
quantities. Unlike emissions of CO2, CFU, and N2O, emissions of fluorinated greenhouse gases are almost
exclusively anthropogenic, meaning that the fluorinated greenhouse gas emission sources included in this
Inventory account for the majority of the total U.S. emissions of these gases detectable in the atmosphere. This
evaluation approach is also useful for gases and sources with larger uncertainties in available bottom-up
inventory methods and data, such as emissions of CH4, which are primarily from uncertain biological (e.g.,
enteric fermentation) and fugitive (e.g., natural gas production) activities.
In this Inventory, EPA includes the results from current and previous comparisons between fluorinated gas
66 See http://www.ipcc-nggip.iges.or.jp/public/index.html.
67 See http://www.ipcc-nggip.iges.or.jp/meeting/pdfiles/1003 Uncertaintv%20meeting report.pdf.
Introduction 1-27
-------
emissions inferred from atmospheric measurements and fluorinated gas emissions estimated based on bottom-
up measurements and modeling. These comparisons, performed for HFCs and SF6 respectively, are described
under the QA/QC and Verification discussions in Chapter 4, Sections 4.25 Substitution of Ozone Depleting
Substances and 4.26 Electrical Equipment in the IPPU chapter of this report.
Consistent with the 2019 Refinement, a key element to facilitate such comparisons is a spatially-explicit (or
gridded) emissions inventory as an input to inverse modeling. To improve the ability to compare methane
emissions from the national-level greenhouse gas inventory with observation-based emission estimates, a team
of researchers from U.S. EPA, SRON Netherlands Institute for Space Research, Harvard University, and Lawrence
Berkely National Laboratory developed a time series of annual anthropogenic methane emissions maps with
0.1° x 0.1° (~10km x 10km) spatial resolution and monthly temporal resolution for the contiguous United
States.68 The gridded methane inventory is designed to be consistent with the U.S. EPA Inventory of U.S.
Greenhouse Gas Emissions and Sinks estimates, which presents national totals for different source types.69 The
development of this gridded inventory is consistent with the recommendations contained in two National
Academies of Science reports examining greenhouse gas emissions data (National Research Council 2010;
National Academies of Sciences, Engineering, and Medicine 2018).
Finally, in addition to use of atmospheric concentration measurement data for comparison with Inventory data,
information from top-down studies is directly incorporated in the Natural Gas Systems calculations to quantify
emissions from certain well blowout events.
In addition, based on the national QA/QC plan for the Inventory, some sector, subsector and category-specific
QA/QC and verification checks have been applied. These checks follow the procedures outlined in the national
QA/QC plan, tailoring the procedures to the specific documentation and data files associated with individual
sources. For each greenhouse gas emissions source or sink category included in this Inventory, a minimum of
general or Tier 1 QC analysis has been undertaken. Where QC activities for a particular category go beyond the
minimum general checks and include category-specific checks (Tier 2) or include verification, further explanation is
provided within the respective source or sink category text. Similarly, responses or updates based on comments
from the expert, public and the international technical expert reviews (e.g., UNFCCC) are also addressed within the
respective source or sink category sections in each sectoral chapter and Annex 8.
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 during the expert review and the public review, in addition to the UNFCCC expert technical
review. While all phases significantly contribute to improving inventory quality, the public review phase is also
essential for promoting the openness of the inventory development process and the transparency of the inventory
methods and underlying input data sources.
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 and removal estimates. The QA/QC procedures also include
feedback loops and provide for corrective actions that are designed to improve the inventory estimates over time.
68 See https://www.epa.gov/ghgemissions/us-gridded-methane-emissions.
69 See https://www.epa.gov/ghgemissions/inventorv-us-greenhouse-gas-emissions-and-sinks.
1-28 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
1.7 Uncertainty Analysis
Emissions and removals calculated for the U.S. Inventory reflect best estimates for greenhouse gas source and sink
categories in the United States and are continuously revised and improved as new information becomes available.
Uncertainty assessment is an essential element of a complete and transparent emissions inventory because it
helps inform and prioritize Inventory improvements. For the U.S. Inventory, uncertainty analyses are conducted for
each source and sink category as well as for the uncertainties associated with the overall emission (current and
base year) and trends estimates. These analyses reflect the quantitative uncertainty in the emission (and removal)
estimates associated with uncertainties in their input parameters (e.g., activity data and EFs) and serve to evaluate
the relative contribution of individual input parameter uncertainties to the overall Inventory, its trends, and each
source and sink category.
The overall level and trend uncertainty estimates for total U.S. greenhouse gas emissions was developed using the
IPCC Approach 2 uncertainty estimation methodology (assuming a Normal distribution for Approach 1 estimates),
which employs a Monte Carlo stochastic simulation technique. The IPCC provides good practice guidance on two
approaches—Approach 1 and Approach 2—to estimating uncertainty for both individual and combined source
categories. Approach 2 quantifies uncertainties based on a distribution of emissions (or removals), built-up from
repeated calculations of emission estimation models and the underlying input parameters, randomly selected
according to their known distributions. Approach 2 methodology is applied to each individual source and sink
category wherever data and resources are permitted and is also used to quantify the uncertainty in the overall
Inventory and its Trends. Source and sink chapters in this report provide additional details on the uncertainty
analysis conducted for each source and sink category. See Annex 7 of this report for further details on the U.S.
process for estimating uncertainty associated with the overall emission (base and current year) and trends
estimates. Consistent with IPCC (IPCC 2006), the United States has ongoing efforts to continue to improve the
overall Inventory uncertainty estimates presented in this report.
The United States has also implemented many improvements over the last several years to reduce uncertainties
across the source and sink categories and improve Inventory estimates. These improvements largely result from
new data sources that provide more accurate data and/or increased data coverage, as well as methodological
improvements. Following IPCC good practice, additional efforts to reduce Inventory uncertainties can occur
through efforts to incorporate excluded emission and sink categories (see Annex 5), improve estimation methods,
and collect more detailed, measured, and representative data. Individual category chapters and Annex 7 both
describe current ongoing and planned Inventory and uncertainty analysis improvements. Consistent with IPCC
(2006), the United States has ongoing efforts to continue to improve the category-specific uncertainty estimates
presented in this report, largely prioritized by considering improvements categories identified as significant by the
Key Category Analysis.
Estimates of quantitative uncertainty for the total U.S. greenhouse gas emissions in 1990 (base year) and 2022 are
shown below in Table 1-5 and Table 1-6, respectively. The overall uncertainty surrounding the Total Net Emissions
is estimated to be -6 to +6 percent in 1990 and -5 to +6 percent in 2022. When the LULUCF sector is excluded from
the analysis the uncertainty is estimated to be -3 to +4 percent in 1990 and -2 to +4 percent in 2022.
Introduction 1-29
-------
Table 1-5: Estimated Overall Inventory Quantitative Uncertainty for 1990 (MMT CO2 Eq. and
Percent)
1990
Emission Uncertainty Range Relative to Greenhouse Gas Standard
Estimate
Estimate-1
Mean1, Deviation1,
Gas
(MMT CO.
Eq.)
(MMT CO
Eq.)
(%)
(MMT CO.
Eq.)
Lower
Upper
Lower
Upper
Bound'
Bound'
Bound
Bound
C02
5,131.6
5,008.2
5,348.2
-2%
4%
5,098.2
88.0
CH4d
871.7
731.3
948.4
-16%
9%
701.5
56.3
N2Od
408.2
349.7
513.0
-14%
26%
434.8
41.6
PFC, HFC, SF6, and NF3d
125.5
108.6
152.9
-13%
22%
207.3
11.6
Total Gross Emissions
6,536.9
6,354.3
6,792.8
-3%
4%
6,441.8
113.3
LULUCF Emissions6
57.9
55.2
61.9
-5%
7%
68.7
1.7
LULUCF Carbon Stock Change
Fluxf
(1,034.7)
(1,296.1)
(845.3)
25%
-18%
(957.3)
116.7
LULUCF Sector Net Total -
(976.7)
(1,237.7)
(787.8)
27%
-19%
(888.6)
116.7
Net Emissions (Sources and
Sinks)
5,560.2
5,247.0
5,882.2
-6%
6%
5,553.3
161.4
a 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.
b 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.
c 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.
d The overall uncertainty estimates did not take into account the uncertainty in the GWP values for CH4, N20 and high GWP
gases used in the Inventory emission calculations for 1990.
e LULUCF emissions include the CH4 and N20 emissions reported for peatlands remaining peatlands, forest fires, drained
organic soils, grassland fires, and coastal wetlands remaining coastal wetlands; CH4 emissions from land converted to coastal
wetlands, land converted to flooded land, and flooded land remaining flooded land; and N20 emissions from forest soils and
settlement soils.
f LULUCF carbon stock change is the net C stock change from the following categories: forest land remaining forest land, land
converted to forest land, cropland remaining cropland, land converted to cropland, grassland remaining grassland, land
converted to grassland, wetlands remaining wetlands, land converted to wetlands, settlements remaining settlements, and
land converted to settlements. Since the resulting flux is negative the signs of the resulting lower and upper bounds are
reversed.
s The LULUCF sector net total is the net sum of all CH4 and N20 emissions to the atmosphere plus net carbon stock changes.
Notes: Total emissions (excluding emissions for which uncertainty was not quantified) are presented without LULUCF. Net
emissions are presented with LULUCF. Totals may not sum due to independent rounding. Parentheses indicate net
sequestration.
Table 1-6: Estimated Overall Inventory Quantitative Uncertainty for 2022 (MMT CO2 Eq. and
Percent)
2022 Emission
Uncertainty Range Relative to Greenhouse
Standard
Estimate
Gas Estimate-1
Mean1, Deviation1,
Gas
(MMTCO. Eq.)
(MMT CO
Eq.)
(%)
(MMT CO . Eq.)
Lower
Upper
Lower
Upper
Bound'
Bound'
Bound
Bound
C02
5,053.0
4,937.3
5,257.7
-2%
4%
5,095.2 81.9
CH4d
702.4
604.3
803.1
-14%
14%
703.8 52.0
N2Od
389.7
324.6
490.2
-17%
26%
399.5 42.3
PFC, HFC, SF6, and NF3d
198.1
182.8
217.5
-8%
10%
199.5 9.0
1-30 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
Total Gross Emissions
6,343.2
6,190.3
6,604.8
-2%
4%
6,397.9
106.3
LULUCF Emissions6
67.5
64.3
73.2
-5%
8%
68.6
2.3
LULUCF Carbon Stock Change Fluxf
(921.8)
(1,158.6)
(748.7)
26%
-19%
(957.5)
105.3
LULUCF Sector Net Totals
(854.3)
(1,090.3)
(680.5)
28%
-20%
(888.8)
105.3
Net Emissions (Sources and Sinks)
5,488.9
5,216.2
5,801.9
-5%
6%
5,509.0
150.6
a 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.
b 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.
c 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.
d The overall uncertainty estimates did not take into account the uncertainty in the GWP values for CH4, N20 and high GWP
gases used in the Inventory emission calculations for 2022.
e LULUCF emissions include the CH4 and N20 emissions reported for peatlands remaining peatlands, forest fires, drained
organic soils, grassland fires, and coastal wetlands remaining coastal wetlands; CH4 emissions from land converted to coastal
wetlands, land converted to flooded land, and flooded land remaining flooded land; and N20 emissions from forest soils and
settlement soils.
f LULUCF carbon stock change is the net C stock change from the following categories: forest land remaining forest land, land
converted to forest land, cropland remaining cropland, land converted to cropland, grassland remaining grassland, land
converted to grassland, wetlands remaining wetlands, land converted to wetlands, settlements remaining settlements, and
land converted to settlements. Since the resulting flux is negative the signs of the resulting lower and upper bounds are
reversed.
g The LULUCF sector net total is the net sum of all CH4 and N20 emissions to the atmosphere plus net carbon stock changes.
Notes: Total emissions (excluding emissions for which uncertainty was not quantified) are presented without LULUCF. Net
emissions are presented with LULUCF. Totals may not sum due to independent rounding. Parentheses indicate net
sequestration.
The uncertainty for 2022 is similar to the uncertainty for 1990, though slightly lower. There have been some
improvements in significant categories which do not necessarily reduce uncertainties as also acknowledged in the
IPCC 2006 IPCC GL, p. 3.13 (e.g., improvements to estimates for Agricultural Soil Management over time,
implications of methodological choice). For example, the 95 percent uncertainty bounds for nitrous oxide emissions
from Agricultural Soil Management were increased from -25 percent to 29 percent in 1990 to -30 percent to 72 percent
in 2022. Methodological and data quality improvements were also made for HFCs, PFCs, SF6 and NF3 this year but
the uncertainties for these sources also slightly increased, better representing the limitations of existing emissions
estimates. The methods and data for fossil fuel combustion categories, the most significant source, has not
changed significantly and neither have uncertainties. It is also worth noting that some of the improvements to shift
to use of GHGRP have been in less significant categories within the inventory (e.g., for IPPU). So, the overall
uncertainty for latest year reflects these offsetting effects and trends within the uncertainty assessment.
In addition to the estimates of uncertainty associated with the current and base year estimates, Table 1-7 presents
the estimates of inventory trend uncertainty. The 2006 IPCC Guidelines defines trend as the difference in emissions
between the base year (i.e., 1990) and the current year (i.e., 2022) Inventory estimates. However, for purposes of
understanding the concept of trend uncertainty, the trend is defined in this Inventory as the percentage change in
the gross emissions (or net emissions) estimated for the current year, relative to the gross emission (or net
emissions) estimated for the base year. The uncertainty associated with this trend is referred to as trend
uncertainty and is reported as between -8 and 8 percent at the 95 percent confidence level between 1990 and
2022. This indicates a range of approximately -8 percent below and 8 percent above the trend estimate of -1.3
percent. See Annex 7 for trend uncertainty estimates for individual source and sink categories by gas.
Introduction 1-31
-------
Table 1-7: Quantitative Assessment of Trend Uncertainty (MMT CO2 Eq. and Percent)
Base Year
2022
Emissions
Gas
Emissions-1
Emissions
Trend
Trend Range1,
(MMT CO
Eq.)
(%)
(%)
Lower
Upper
Bound
Bound
C02
5,131.6
5,053.0
-2%
-6%
3%
ch4
871.7
702.4
-19%
-32%
2%
n2o
408.2
389.7
-5%
-30%
51%
HFCs, PFCs, SF6, andNFs
125.5
198.1
58%
32%
95%
Total Gross Emissionsc
6,536.9
6,343.2
-3%
-7%
3%
LULUCF Emissions'1
57.9
67.5
17%
6%
30%
LULUCF Carbon Stock Change Fluxe
(1,034.7)
(921.8)
-11%
-35%
21%
LULUCF Sector Net Total1
(976.7)
(854.3)
-13%
-37%
21%
Net Emissions (Sources and Sinks)'
5,560.2
5,488.9
-1.3%
-8%
8%
a Base Year is 1990 for all sources.
bThe trend range represents a 95 percent confidence interval for the emission trend, with the lower bound corresponding to
2.5th percentile value and the upper bound corresponding to 97.5th percentile value.
c Totals exclude emissions for which uncertainty was not quantified.
d LULUCF emissions include the CH4 and N20 emissions reported for peatlands remaining peatlands, forest fires, drained organic
soils, grassland fires, and coastal wetlands remaining coastal wetlands; CH4 emissions from land converted to coastal wetlands,
land converted to flooded land, and flooded land remaining flooded land; and N20 emissions from forest soils and settlement
soils.
e LULUCF carbon stock change is the net C stock change from the following categories: forest land remaining forest land, land
converted to forest land, cropland remaining cropland, land converted to cropland, grassland remaining grassland, land
converted to grassland, wetlands remaining wetlands, land converted to wetlands, settlements remaining settlements, and
land converted to settlements.
f The LULUCF sector net total is the net sum of all CH4 and N20 emissions to the atmosphere plus net carbon stock changes.
Notes: Totals may not sum due to independent rounding. Parentheses indicate net sequestration. Total emissions (excluding
emissions for which uncertainty was not quantified) are presented without LULUCF. Net emissions are presented with LULUCF.
1.8 Completeness
This report, along with its accompanying CRTs, serves as a thorough assessment of the anthropogenic sources and
sinks of greenhouse gas emissions for the United States for the time series 1990 through 2022. This report is
intended to be comprehensive and includes the vast majority of emissions and removals identified as
anthropogenic, consistent with IPCC methods and the Paris Agreement and UNFCCC reporting guidelines. In
general, sources or sink categories not accounted for in this Inventory are excluded because they are not occurring
in the United States and its territories, or because data are unavailable to develop an estimate and/or the
categories were determined to be insignificant70 in terms of overall national emissions per the Paris Agreement
and UNFCCC reporting guidelines.
The United States is continually working to improve upon the understanding of such sources and sinks currently
not included and seeking to find the data required to estimate related emissions and removals, focusing on
categories that are anticipated to be significant. See Chapter 9 on Improvements and Recalculations for more
70 See paragraph 32 in the Annex to Decisionl8/CMA.l of the Paris Agreement reporting guidelines on national inventories
that state "...emissions from a category should only be considered insignificant if the likely level of emissions is below 0.05 per
cent of the national total GHG emissions, excluding LULUCF, or 500 kilotonnes of carbon dioxide equivalent (kt C02 eq),
whichever is lower. The total national aggregate of estimated emissions for all gases from categories considered insignificant
shall remain below 0.1 per cent of the national total GHG emissions, excluding LULUCF. Parties should use approximated
activity data and default IPCC emission factors to derive a likely level of emissions for the respective category."
1-32 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
information on completeness and improvements implemented this cycle. As such improvements are implemented,
new emission and removal estimates are quantified and included in the Inventory, improving completeness of
national estimates. For a list of sources and sink categories not included and more information on significance of
these categories, see Annex 5 and the respective category sections in each sectoral chapter of this report.
1.9 Organization of Report
In accordance with the Paris Agreement and UNFCCC reporting guidelines, this Inventory is grouped into five
sector-specific chapters consistent with the Paris Agreement Common Reporting Tables (CRT),71 listed below in
Table 1-8. In addition, the U.S. Inventory submission includes chapters on Trends in Greenhouse Gas Emissions,
Other information, and Recalculations and Improvements to be considered consistent with the suggested outline
or national inventory documents submitted under the Paris Agreement and UNFCCC.
Table 1-8: CRT/IPCC Sector Descriptions
Chapter (CRT and
Activities Included
UNFCCC/IPCC Sector)
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.
Agriculture
Emissions from agricultural activities except fuel combustion, which is
addressed under Energy.
Land Use, Land-Use
Emissions and removals of CO2, and emissions of CH4, and N20 from land use,
Change, and Forestry
land-use change, and forestry.
Waste
Emissions from waste management activities.
Within each chapter, emissions are identified by the anthropogenic activity that is the source or sink of the
greenhouse gas emissions being estimated (e.g., coal mining). Overall, the following organizational structure is
consistently applied throughout this report:
Chapter/CRT/IPCC Sector: Overview of emissions and trends for each CRT/IPCC defined sector.
CRT Source or Sink Category: Description of category pathway and emission/removal trends based on IPCC
methodologies, consistent with the Paris Agreement and UNFCCC reporting guidelines.
Methodology and Time-Series Consistency: Description of analytical methods (e.g., from 2006 IPCC Guidelines, or
country-specific methods) employed to produce emission estimates and identification of data references, primarily
for activity data and emission factors, and a discussion of time-series consistency.
Uncertainty: A discussion and quantification of the uncertainty in emission estimates.
QA/QC and Verification: A discussion on steps taken to QA/QC and verify the emission estimates, consistent with
the U.S. QA/QC plan, and any key QC findings.
Recalculations Discussion: A discussion of any data or methodological changes that necessitate a recalculation of
previous years' emission estimates, and the impact of the recalculation on the emission estimates, if applicable.
Planned Improvements: A discussion on any category-specific planned improvements, if applicable.
71 See paragraph 50 in the Annex to Decision 18/CMA.l
Introduction 1-33
-------
Special attention is given to C02 from fossil fuel combustion relative to other sources because of its share of
emissions and its dominant influence on emission trends. For example, each energy consuming end-use sector
(i.e., residential, commercial, industrial, and transportation), as well as the electricity generation sector, is
described individually. Additional information for certain source categories and other topics is also provided in
several Annexes listed in Table 1-9.
Table 1-9: List of Annexes
ANNEX 1
Key Category Analysis
ANNEX 2
Methodology and Data for Estimating C02 Emissions from Fossil Fuel Combustion
2.1.
Methodology for Estimating Emissions of C02 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 CH4, N20, and Indirect Greenhouse Gases from Stationary
Combustion
3.2.
Methodology for Estimating Emissions of CH4, N20, 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
3.5.
Methodology for Estimating CH4 and C02 Emissions from Petroleum Systems
3.6.
Methodology for Estimating CH4 Emissions from Natural Gas Systems
3.7.
Methodology for Estimating C02 and N20 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 CH4 Emissions from Enteric Fermentation
3.11.
Methodology for Estimating CH4 and N20 Emissions from Manure Management
3.12.
Methodology for Estimating N20 Emissions, CH4 Emissions and Soil Organic C Stock Changes from
Agricultural Lands (Cropland and Grassland)
3.13.
Methodology for Estimating Net Carbon Stock Changes in Forest Land Remaining Forest Land and Land
Converted to Forest Land
3.14.
Methodology for Estimating CH4 Emissions from Landfills
ANNEX 4
IPCC Reference Approach for Estimating C02 Emissions from Fossil Fuel Combustion
ANNEX 5
Assessment of the Sources and Sinks of Greenhouse Gas Emissions Not Included
ANNEX 6
Additional Information
6.1.
Global Warming Potential Values
6.2.
Ozone Depleting Substance Emissions
6.3.
Greenhouse Gas Precursors: Mapping of NEI categories to the Inventory
6.4.
Constants, Units, and Conversions
6.5.
Chemical Formulas
ANNEX 7
Uncertainty
7.1.
Overview
7.2.
Methodology and Results
7.3.
Reducing Uncertainty
7.4.
Planned Improvements
7.5.
Additional Information on Uncertainty Analyses by Source
ANNEX 8
QA/QC Procedures
8.1.
Background
8.2.
Purpose
8.3.
Assessment Factors
8.4.
Responses During the Review Process
ANNEX 9
Use of Greenhouse Gas Reporting Program (GHGRP) in Inventory
1-34 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
2. Trends in Greenhouse Gas Emissions
and Removals
2.1 Overview of U.S. Greenhouse Gas
Emissions and Sinks Trends
In 2022, total gross U.S. greenhouse gas emissions were 6,343.2 million metric tons of carbon dioxide equivalent
(MMT CO2 Eq.).1 Total gross U.S. emissions decreased by 3.0 percent from 1990 to 2022, down from a high of 15.2
percent above 1990 levels in 2007. Gross emissions increased from 2021 to 2022 by 0.2 percent (14.4 MMT CO2
Eq.). Net emissions (i.e., including sinks) were 5,489.0 MMT CO2 Eq. in 2022. Overall, net emissions increased by
1.3 percent from 2021 to 2022 and decreased by 16.7 percent from 2005 levels, as shown in Table 2-1. Between
2021 and 2022, the increase in total greenhouse gas emissions was driven largely by an increase in CO2 emissions
from fossil fuel combustion across most end-use sectors due in part to increased energy use from the continued
rebound of economic activity after the height of the COVID-19 pandemic. The CO2 emissions from fossil fuel
combustion increased by 1.0 percent from 2021 to 2022, including a 5.0 percent increase in residential sector
emissions, 8.9 percent increase in commercial sector emissions, 0.1 percent decrease in transportation sector
emissions, 2.6 percent increase in industrial emissions, and 0.6 percent decrease in electric power sector
emissions. Carbon sequestration in the Land Use, Land-Use Change, and Forestry (LULUCF) sector offset 14.5
percent of total emissions in 2022.
Figure 2-1 and Figure 2-2 illustrate the overall trend in total U.S. emissions and sinks since 1990, by gas and by
annual percentage changes relative to the previous year.
1 The gross emissions total presented in this report for the United States excludes emissions and sinks from removals from
LULUCF. The net emissions total presented in this report for the United States includes emissions and sinks from removals from
LULUCF.
Trends 2-1
-------
Figure 2-1: U.S. Greenhouse Gas Emissions and Sinks by Gas
9,000
8,000
7,000
6,000
5,000
8 4,000
3,000
2,000
1,000
0
-1,000
HFCs, PFCs, SFe and NF3 ~ Net Emissions (including LULUCF sinks)
Nitrous Oxide
Methane
Carbon Dioxide
Net CO2 Flux from LULUCF3
O H f\J n I- L/1 UD
cr>cric^cr>cr>a>CT»cr>a>cr>
CTi CT> CTi CTi CTt CF> CF* CT> CT\ CT»
r^cocnoi-Hrsim^-LnvDr^oocriOi-HfNro^rLnvDr^oocTiOT-ifN
OOOOOOOOOOi—ii—Ii-h*—I*—li-Hi—l*—irMfMfM
00000000000000000000000
fMfMNfMrMfNfNfMrMlNfMrvlfNfNfNfNrMfMrMfNfNOJlN
3 The term "flux" is used to describe the exchange of C02 to and from the atmosphere, with net flux being either positive or
negative depending on the overall balance. Removal and long-term storage of C02 from the atmosphere is also referred to as
"carbon sequestration."
Figure 2-2: Annual Percentage Change in Gross U.S. Greenhouse Gas Emissions Relative to
the Previous Year
5.4%
1.7%
1.4% 2.8% 1.0% 2.8%
1.6%
1.4%
2.9%
3.1%
-10%
--irMroTLnuDr^ooch
oio^^^o^oioiaioi
cr>cr»cr>cricr>cr»
INfNrMrMOJrMfNNlM(NIN(N(N(NfNIOJrMlNfN(NrMfNlN
2-2 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
Emissions and Sinks by Gas
Figure 2-3 illustrates the relative contribution of each gas to total gross U.S. greenhouse gas emissions in 2022, in
CO2 equivalents (i.e., weighted by global warming potential). The primary greenhouse gas emitted by human
activities in the United States is CO2, representing 79.7 percent of total greenhouse gas emissions. The largest
source of CO2—and of overall greenhouse gas emissions—is fossil fuel combustion, primarily from transportation
and power generation. Methane (CH4) emissions account for 11.1 percent of emissions. The major sources of
methane include enteric fermentation associated with domestic livestock, natural gas systems, and decomposition
of waste in landfills. Agricultural soil management, wastewater treatment, stationary sources of fuel combustion,
and manure management are the major sources of N2O emissions. Ozone depleting substance (ODS) substitute
emissions were the primary contributor to aggregate hydrofluorocarbon (HFC) emissions. Perfluorocarbon (PFC)
emissions were attributable primarily to fluorochemical production and electronics manufacturing. Electrical
equipment accounted for most sulfur hexafluoride (SFs) emissions. Nitrogen trifluoride (NF3) emissions were
approximately evenly split between electronics manufacturing and fluorochemical production.
Figure 2-3: 2022 Gross Total U.S. Greenhouse Gas Emissions by Gas (Percentages based on
MMT C02 Eq.)
3.1%
HFCs, PFCs, SFe and NFs
Note: Emissions and removals from LULUCF are excluded from the figure above.
From 1990 to 2022, total emissions of CO2decreased by 1.5 percent (78.6 MMT CO2 Eq.), total emissions of
methane (CH4) decreased by 19.4 percent (169.3 MMT CO2 Eq.), and total emissions of nitrous oxide (N2O)
decreased by 4.5 percent (18.4 MMT CChEq.). During the same period, emissions of fluorinated gases including
HFCs, PFCs, SFs, and NF3 rose by 57.9 percent (72.7 MMT CO2 Eq.). Rapidly growing emissions of HFCs drove this
trend, overwhelming decreases in emissions of PFCs and SF6. Despite being emitted in smaller quantities relative to
the other principal greenhouse gases, emissions of HFCs, PFCs, SF6, and NF3 are significant because many of them
have extremely high global warming potentials (GWPs), and, in the cases of PFCs, SF6, and NF3, very long
atmospheric lifetimes. U.S. greenhouse gas emissions were partly offset by carbon sequestration in managed
forests, trees in urban areas, agricultural soils, landfilled yard trimmings, and coastal wetlands. These were
estimated to offset 14.5 percent (921.8 MMT CO2 Eq.) of total gross emissions in 2022.
Table 2-1 provides information on trends in emissions and sinks from all U.S. anthropogenic sources and sinks in
weighted units of MMT CO2 Eq., while unweighted gas emissions and sinks in kilotons (kt) are provided in Table
2-2.
Trends 2-3
-------
Table 2-1: Recent Trends in U.S. Greenhouse Gas Emissions and Sinks by Gas (MMT CO2 Eq.)
Gas/Source
1990
2005
2018
2019
2020
2021
2022
C02
5,131.6
6,126.9
5,362.2
5,234.5
4,689.0
5,017.2
5,053.0
Fossil Fuel Combustion
A,752.2 ¦
5,744.1
4,988.2
4,852.6
4,341.7
4,654.3
4,699.4
Transportation
1,468.9
1,858.6
1,813.1
1,816.6
1,572.8
1,753.5
1,751.3
Electric Power Sector
1,820.0 III:
2,400.1 1
1,753.4
1,606.7
1,439.6
1,540.9
1,531.7
Industrial
876.5
847.6
810.5
809.8
762.0
780.5
801.1
Residential
338.6
358.9
338.9
342.9
314.8
318.0
334.1
Commercial
228.3
227.1
246.3
251.7
229.3
237.5
258.7
U.S. Territories
20.0 1
51.9:
25.9
24.8
23.3
23.8
22.6
Non-Energy Use of Fuels
99.1
125.0
118.4
106.5
97.8
111.6
102.8
Cement Production
33.5
46.2 ;
39.0
40.9
40.7
41.3
41.9
Iron and Steel Production &
Metallurgical Coke Production
104.7
70.1
42.9
43.1
37.7
41.9
40.7
Natural Gas Systems
32.4
26.3 ;
32.8
38.5
36.7
35.8
36.5
Petrochemical Production
20.1
26.9
27.2
28.5
27.9
30.7
28.8
Petroleum Systems
9-6:
10.2 j,
34.8
45.5
28.9
24.1
22.0
Ammonia Production
14.4
10.2
12.7
12.4
13.0
12.2
12.6
Incineration of Waste
12.9
13-3!
13.3
12.9
12.9
12.5
12.4
Lime Production
11.7
14.6
13.1
12.1
11.3
11.9
12.2
Other Process Uses of Carbonates
7.11
8-5 '
7.9
9.0
9.0
8.6
10.4
Urea Consumption for Non-Agricultural
Purposes
3.8
3.7
6.1
6.2
5.8
6.6
7.1
Urea Fertilization
2.4
3.5
4.9
5.0
5.1
5.2
5.3
Carbon Dioxide Consumption
1.5
1.4
4.1
4.9
5.0
5.0
5.0
Liming
4.7 1
4.4 ;
2.2
2.2
2.9
2.4
3.3
Coal Mining
4.6
4.2
3.1
3.0
2.2
2.5
2.5
Glass Production
2-3
2.4 III;
2.0
1.9
1.9
2.0
2.0
Soda Ash Production
1.4
1.7
1.7
1.8
1.5
1.7
1.7
Titanium Dioxide Production
I.2!
1.8
1.5
1.3
1.3
1.5
1.5
Aluminum Production
6.8
4.1
1.5
1.9
1.7
1.5
1.4
Ferroalloy Production
2.2 ¦
1.4|
2.1
1.6
1.4
1.6
1.3
Zinc Production
0.6
1.0
1.0
1.0
1.0
1.0
0.9
Phosphoric Acid Production
1.5 1
i-3::
0.9
0.9
0.9
0.9
0.8
Lead Production
0.5
0.6
0.5
0.5
0.5
0.4
0.4
Carbide Production and Consumption
0.2
0.2 s
0.2
0.2
0.2
0.2
0.2
Abandoned Oil and Gas Wells
+
+
+
+
+
+
+
Substitution of Ozone Depleting
1
Substances
+ £
+
+
+
+
+
+
Magnesium Production and Processing
0.1
+
+
+
+
+
+
Biomass and Biodiesel Consumptiona
237.9
245.4
336.0
333.1
295.7
303.0
305.4
International Bunker Fuelsb
103.6
113.3
124.3
113.6
69.6
80.2
98.2
CH4c
871.7
795.4
771.5
754.3
735.3
720.5
702A
Enteric Fermentation
183.1
188.2
196.8
197.3
196.3
196.5
192.6
Natural Gas Systems
218.8 £
210.1 I
190.3
188.7
180.3
174.6
173.1
Landfills
197.8
147.7
126.3
128.7
124.1
122.0
119.8
Manure Management
39.1 •
55.0 lllll;
67.7
66.7
66.9
66.4
64.7
Coal Mining
108.1
71.5
59.1
53.0
46.2
44.7
43.6
Petroleum Systems
49.4 1
48.2
59.0
52.2
53.3
48.6
39.6
Wastewater T reatment
22.7
22.7
21.4
21.1
21.0
20.7
20.8
Rice Cultivation
18.9
20.6 ::
19.9
15.6
18.6
18.3
18.9
Stationary Combustion
9.7
8.8
9.6
9.8
8.0
8.0
8.6
Abandoned Oil and Gas Wells
CO
¦Sill
8.2 I
8.4
8.5
8.5
8.6
8.5
Abandoned Underground Coal Mines
8.1
7.4
6.9
6.6
6.5
6.3
6.3
Mobile Combustion
7.2
4-3 i
2.8
2.9
2.5
2.6
2.6
Composting
0.4
2.1
2.5
2.5
2.6
2.6
2.6
Field Burning of Agricultural Residues
0.51
0.6 1
0.6
0.7
0.6
0.6
0.6
Anaerobic Digestion at Biogas Facilities
+
+
+
+
+
+
+
2-4 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
Carbide Production and Consumption
+:
+
+
+
+
+
Ferroalloy Production
+
+
+
+
+
+
+
Iron and Steel Production &
1
lllllli
Metallurgical Coke Production
+=
¦
+
+
+
+
+
Petrochemical Production
+
+
+
+
+
+
+
Incineration of Waste
+:
+
+
+
+
+
International Bunker Fuelsb
0.2
0.1
0.1
0.1
0.1
0.1
0.1
N2Oc
408.2
419.2
439.5
416.4
391.2
398.2
389.7
Agricultural Soil Management
288.8
294.1
333.4
315.6
292.1
298.0
290.8
Stationary Combustion
22.3
30.5
25.1
22.2
20.5
22.0
24.7
Wastewater T reatment
14.8
18.1
21.2
21.6
22.3
22.1
21.9
Manure Management
13.4
15.2 •
16.6
16.8
16.9
17.1
17.0
Mobile Combustion
38.4
37.0
17.7
19.1
16.1
16.8
16.7
Nitric Acid Production
10.8 i
10.1 ,
8.5
8.9
8.3
7.9
8.6
N20 from Product Uses
3.8
3.8
3.8
3.8
3.8
3.8
3.8
Adipic Acid Production
13.5
6.3 1
9.3
4.7
7.4
6.6
2.1
Composting
0.3
1.5
1.8
1.8
1.8
1.8
1.8
Caprolactam, Glyoxal, and Glyoxylic
!!!!!!!
I
Acid Production
1-9
1.3
1.2
1.1
1.2
1.3
Incineration of Waste
0.4
0.3
0.4
0.4
0.3
0.4
0.3
Electronics Industry
+
o-11
0.2
0.2
0.3
0.3
0.3
Field Burning of Agricultural Residues
0.2
0.2
0.2
0.2
0.2
0.2
0.2
Natural Gas Systems
-
+
+
+
+
+
0.2
Petroleum Systems
+
+
+
+
+
+
+
International Bunker Fuelsb
0.8 1
0.9
1.0
0.9
0.5
0.6
0.8
HFCs
47.7
121.7
163.9
168.2
170.3
177.0
182.8
Substitution of Ozone Depleting
;;;;;;;
1
1
Substances
0.3
99.5 1
157.9
162.1
166.2
172.6
178.1
Fluorochemical Production
47.3
22.1
5.7
5.7
3.8
4.0
4.3
Electronics Industry
0.2
I
0.3
0.3
0.3
0.4
0.3
Magnesium Production and Processing
0.0
0.0
0.1
0.1
0.1
+
+
PFCs
39.5
10.2
7.4
7.3
6.6
6.3
6.7
Fluorochemical Production
17.5
4.0
2.9
3.0
2.5
2.6
3.0
Electronics Industry
2.5
3.0
2.9
2.6
2.5
2.6
2.7
Aluminum Production
19.3
3.1
1.4
1.4
1.4
0.9
0.8
SF6 and PFCs from Other Product Use
0.1 •
0.15
0.2
0.2
0.2
0.1
0.2
Substitution of Ozone Depleting
Substances
0.0
+
+
+
+
+
+
Electrical Equipment
o.o
+ iii
0.0
+
+
+
+
SF,
37.9
20.2
7.6
8.4
8.1
8.5
7.6
Electrical Equipment
24.7
1.1.8 (
5.0
6.1
5.9
6.0
5.1
Magnesium Production and Processing
5.6
3.0
1.1
0.9
0.9
1.2
1.1
Electronics Industry
o
Ln
II11B1
0.8
0.8
0.8
0.8
0.9
0.8
SF6 and PFCs from Other Product Use
1.3
1.3
0.8
0.6
0.5
0.4
0.6
Fluorochemical Production
5-8 1
3.3
+
+
+
+
+
NF,
0.3
1.0
0.7
1.1
1.3
1.1
1.1
Electronics Industry
+ 1
0.4!!!
0.5
0.5
0.6
0.6
0.6
Fluorochemical Production
0.3
0.6
0.1
0.6
0.7
0.5
0.5
Total Gross Emissions (Sources)
6,536.9
7,494.6
6,752.7
6,590.1
6,001.8
6,328.8
6,343.2
LULUCF Emissions'1
58.0
68.9
62.8
58.0
68.4
72.9
67.6
ch4
53.1
58.5
55.5
52.5
59.3
62.1
58.4
n2o
4.8
10.3
7.3
5.5
9.1
10.7
9.1
LULUCF Carbon Stock Change8
(1,034.7)
(976.6)
(978.3)
(921.6)
(972.8)
(983.4)
(921.8)
LULUCF Sector Net Total'
(976.7)
(907.7)
(915.5)
(863.6)
(904.4)
(910.6)
(854.2)
Net Emissions (Sources and Sinks)
5,560.2
6,586.9
5,837.3
5,726.6
5,097.4
5,418.2
5,489.0
+ Does not exceed 0.05 MMT C02 Eq.
NO (Not Occurring)
Trends 2-5
-------
a Emissions from biomass and biofuel consumption are not included specifically in Energy sector totals. Net carbon fluxes from
changes in biogenic carbon reservoirs are accounted for in the estimates for LULUCF.
b Emissions from international bunker fuels are not included in totals.
c LULUCF emissions of CH4 and N20 are reported separately from gross emissions totals. LULUCF emissions include the CH4
and N20 emissions reported for peatlands remaining peatlands, forest fires, drained organic soils, grassland fires, and coastal
wetlands remaining coastal wetlands; CH4 emissions from land converted to coastal wetlands, flooded land remaining
flooded land, and land converted to flooded land; and N20 emissions from forest soils and settlement soils. Refer to Table
2-8 for a breakout of emissions and removals for LULUCF by gas and source category.
d Small amounts of PFC emissions from this source are included under HFCs due to confidential business information.
e LULUCF carbon stock change is the net carbon stock change from the following categories: forest land remaining forest land,
land converted to forest land, cropland remaining cropland, land converted to cropland, grassland remaining grassland, land
converted to grassland, wetlands remaining wetlands, land converted to wetlands, settlements remaining settlements, and
land converted to settlements. Refer to Table 2-8 for a breakout of emissions and removals for LULUCF by gas and source
category.
f The LULUCF sector net total is the net sum of all LULUCF CH4 and N20 emissions to the atmosphere plus LULUCF net carbon
stock changes.
Notes: Total (gross) emissions are presented without LULUCF. Net emissions are presented with LULUCF. Totals may not sum
due to independent rounding. Parentheses indicate negative values or sequestration.
Table 2-2: Recent Trends in U.S. Greenhouse Gas Emissions and Sinks by Gas (kt)
Gas/Source
1990
2005
2018
2019
2020
2021
2022
co2
5,131,650
6,126,864 5
,362,191
5,234,488
4,688,969
5,017,202
5,053,019
Fossil Fuel Combustion
4,752,232
5,744,134 4
111: 7 7 m
,988,198
4,852,631
4,341,710
4,654,265
4,699,403
Transportation
1,468,944
1,858,552
,813,135
1,816,636
1,572,820
1,753,546
1,751,286
Electric Power Sector
1,819,951
2,400,057
,753,432
1,606,721
1,439,566
1,540,933
1,531,680
Industrial
876,468
847,640
810,472
809,807
761,986
780,453
801,064
Residential
338,568
I 358,898 1
338,940
342,905
314,795
318,034
334,065
Commercial
228,293
227,130
246,297
251,749
229,264
237,528
258,733
U.S. Territories
20,010
2 51,8571,
25,923
24,813
23,279
23,772
22,575
Non-Energy Use of Fuels
99,104
124,988
118,382
106,474
97,757
111,624
102,808
Cement Production
33,484
46,194 1
38,971
40,896
40,688
41,312
41,884
Iron and Steel Production &
1 1
Metallurgical Coke
Production
104,740
70,082
42,863
43,095
37,724
41,873
40,672
Natural Gas Systems
32,427
26,312
32,768
38,525
36,719
35,780
36,470
Petrochemical Production
20,075
26,882 1
27,200
28,483
27,926
30,656
28,788
Petroleum Systems
9,585
10,210
34,777
45,498
28,937
24,140
21,967
Ammonia Production
14,404
10,234
12,669
12,401
13,006
12,192
12,610
Incineration of Waste
12,900
13,254
13,339
12,948
12,921
12,476
12,357
Lime Production
11,700
14,552
13,106
12,112
11,299
11,870
12,208
Other Process Uses of
::
1 1
Carbonates
7,103
8,472
7,938
8,973
9,012
8,583
10,384
Urea Consumption for Non-
Agricultural Purposes
3,784
3,653
6,113
6,150
5,805
6,600
7,053
Urea Fertilization
2,417
3,504 g
4,936
5,034
5,132
5,229
5,327
Carbon Dioxide Consumption
1,472
1,375
4,130
4,870
4,970
4,990
5,000
Liming
4,690
4,351.1
2,240
2,203
2,887
2,387
3,268
Coal Mining
4,606
4,169
3,139
2,992
2,197
2,455
2,474
Glass Production
2,263
2,402 |
1,989
1,940
1,858
1,969
1,956
Soda Ash Production
1,431
1,655
1,714
1,792
1,461
1,714
1,704
Titanium Dioxide Production
1,195
¦ 1,755 1
1,541
1,340
1,340
1,474
1,474
Aluminum Production
6,831
4,142
1,455
1,880
1,748
1,541
1,446
Ferroalloy Production
2,152
1,392 1
2,063
1,598
1,377
1,567
1,327
Zinc Production
632
1,030
999
1,026
977
1,007
947
Phosphoric Acid Production
1,529
1'342 1
937
909
901
874
840
2-6 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
Gas/Source
1990
2005
2018
2019
2020
2021
2022
Lead Production
516
553 1
527
531
450
439
428
Carbide Production and
I
Consumption
243
213
184
175
154
172
210
Abandoned Oil and Gas Wells
7
7
8
8
8
8
8
Substitution of Ozone
iiiiiii
1
1
Depleting Substances
+
11
3
3
4
4
4
Magnesium Production and
1
Processing
129
4
2
2
3
3
3
Biomass and Biodiesel
I
1
1
1
Consumptiona
237,946 I!
245,421 !!
335,971
333,057
295,695
303,014
305,417
International Bunker Fuelsb
103,634
113,328
124,279
113,632
69,638
80,180
98,241
CH4c
31,131
28,408
'
27,553
26,941
26,262
25,731
25,084
Enteric Fermentation
6,539
6,722
7,028
7,045
7,010
7,017
6,878
Natural Gas Systems
7,813
7,505
6,795
6,741
6,439
6,235
6,183
Landfills
7,063
5,275
4,512
4,595
4,431
4,359
4,277
Manure Management
1,398 i
1,964
2,418
2,382
2,390
2,373
2,312
Coal Mining
3,860
2,552
2,110
1,892
1,648
1,595
1,558
Petroleum Systems
1,765
1,723
2,108
1,865
1,904
1,737
1,415
Wastewater Treatment
811
809
763
755
748
738
743
Rice Cultivation
677 iiiiii
735 ;;
711
558
664
653
674
Stationary Combustion
345
313
344
351
285
286
307
Abandoned Oil and Gas Wells
279 '
294 B
301
302
303
306
303
Abandoned Underground Coal
Mines
288
264
247
237
232
224
225
Mobile Combustion
258
154 1
101
102
91
92
93
Composting
15
75
90
91
92
92
92
Field Burning of Agricultural
1
iiiiiii
1
Residues
19
23 ;
22
23
22
22
22
Anaerobic Digestion at Biogas
Facilities
+
+
+
1
+
+
+
Carbide Production and
I
I
Consumption
+ 1
+
+
+
+
+
Ferroalloy Production
1
+
1
+
+
+
+
Iron and Steel Production &
I
I
Metallurgical Coke
si
Production
l III!
1 111
+
+
+
+
+
Petrochemical Production
+
+
+
+
+
+
+
Incineration of Waste
+
+ !!!!!!;
¦
+
+
+
+
+
International Bunker Fuelsb
7
5
4
4
3
3
3
N2Oc
1,540
1,582
1,658
1,571
1,476
1,503
1,471
Agricultural Soil Management
1,090
1,110
1,258
1,191
1,102
1,124
1,097
Stationary Combustion
84
115.
95
84
78
83
93
Wastewater Treatment
56
68
80
81
84
83
83
Manure Management
50 5
57 =
63
63
64
65
64
Mobile Combustion
145
140
67
72
61
63
63
Nitric Acid Production
41
¦
38:
32
34
31
30
33
N20 from Product Uses
14
14
14
14
14
14
14
Adipic Acid Production
51
24 I-
35
18
28
25
8
Composting
1
6
7
7
7
7
7
Caprolactam, Glyoxal, and
!!!!!!!
!!!!!!!
1
Glyoxylic Acid Production
6
7
5
5
4
5
5
Incineration of Waste
2
1 1
1
1
1
1
1
Electronics Industry
+
¦
+ 5
1
1
1
1
1
Trends 2-7
-------
Gas/Source
1990
2005
2018
2019
2020
2021
2022
Field Burning of Agricultural
1
Residues
1
1
1
1
1
1
1
Natural Gas Systems
+
+
+
+
+
+
1
Petroleum Systems
+
+
+
+
+
+
+
International Bunker Fuelsb
3 !!!!!!!
3
4
3
2
2
3
HFCs
M
M
M
M
M
M
M
Substitution of Ozone
mm;
1
Depleting Substances
M |
M 1
M
M
M
M
M
Fluorochemical Production
M
M
M
M
M
M
M
Electronics Industry
M 1
M 1
M
M
M
M
M
Magnesium Production and
Processing
0
0
+
+
+
+
+
PFCs
M
M llll
M
M
M
M
M
Fluorochemical Production
M
M
M
M
M
M
M
Electronics Industry
M ¦
M
M
M
M
M
M
Aluminum Production
M
M
M
M
M
M
M
SF6 and PFCs from Other
iiiiiii
1
I
Product Use
1011
83
178
173
167
138
172
Substitution of Ozone
I
Depleting Substances
+
+
+
+
+
+
+
Electrical Equipment
+ ;
+ 1
USE
0
+
+
+
+
sf6
2
1
+
+
+
+
+
Electrical Equipment
1 III
1
+
+
+
+
+
Magnesium Production and
Processing
+
+
+
+
+
+
+
Electronics Industry
+
+
+
+
+
+
+
SF6 and PFCs from Other
Product Use
+ "
+ -
+
+
+
+
+
Fluorochemical Production
+
+ 1
!!!!!!!
+
+
+
+
+
nf3
+
+
+
+
+
+
+
Electronics Industry
+
+
+
+
+
+
+
Fluorochemical Production
+
+
+
+
+
+
+
+ Does not exceed 0.5 kt.
M (Mixture of multiple gases)
NO (Not Occurring)
a Emissions from biomass and biofuel consumption are not included specifically in Energy sector totals. Net carbon fluxes
from changes in biogenic carbon reservoirs are accounted for in the estimates for LULUCF.
b Emissions from international bunker fuels are not included in totals.
c LULUCF emissions of LULUCF CH4 and N20 are reported separately from gross emissions totals. Refer to Table 2-8 for a
breakout of emissions and removals for LULUCF by gas and source category.
d Small amounts of PFC emissions from this source are included under HFCs due to confidential business information.
Notes: Totals by gas may not sum due to independent rounding. Parentheses indicate negative values or sequestration.
Emissions and Sinks by UNFCCC/IPCC Sector
Emissions and removals of all gases can be summed from each source and sink category into a set of five sectors
defined by the Paris Agreement and UNFCCC reporting guidelines and methodological framework provided by the
Intergovernmental Panel on Climate Change (IPCC). Figure 2-4 and Table 2-3 illustrate that over the 33-year period
of 1990 to 2022, total emissions from the Energy and Waste sectors decreased by 3.4 percent (181.2 MMT CO2 Eq.)
and 29.3 percent (69.1 MMT CO2 Eq.), respectively. Emissions from Industrial Processes and Product Use and
Agriculture grew by 3.9 percent (14.4 MMT CO2 Eq.) and 7.7 percent (42.2 MMT CO2 Eq.), respectively. Over the
same period, total carbon sequestration in the LULUCF sector decreased by 12.5 percent (122.5 MMT CO2), and
emissions from the LULUCF sector increased by 16.5 percent (9.6 MMT CO2 Eq.).
2-8 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
Figure 2-4: U.S. Greenhouse Gas Emissions and Sinks by IPCC Sector
¦ LULUCF (emissions)
9,000 ¦ Waste
¦ Industrial Processes and Product Use
o nnn — Net Emissions (including LULUCF sinks)
o,UUU
7,000
6,000
S 5,000
N
O
^ 4,000
E
3,000
2,000
1,000
0
-1,000
I Agriculture
I Energy
I LULUCF (removals)
Table 2-3: Recent Trends in U.S. Greenhouse Gas Emissions and Sinks by UNFCCC/IPCC
Sector/Category (MMT CO2 Eq.)
UNFCCC/IPCC Sector/Category
1990
2005
2018
2019
2020
2021
2022
Energy
5,381.0
6,349.5
5,570.0
5,422.4
4,862.6
5,173.3
5,199.8
Fossil Fuel Combustion
4,752.2
5,744.1
4,988.2
4,852.6
4,341.7
4,654.3
4,699.4
Natural Gas Systems
251.2
236.5
223.0
227.3
217.0
210.4
209.7
Non-Energy Use of Fuels
99.1
125.0
118.4
106.5
97.8
111.6
102.8
Petroleum Systems
59.0
58.5
93.8
97.8
82.3
72.8
61.6
Coal Mining
112.7
75.6
62.2
56.0
48.3
47.1
46.1
Stationary Combustion3
32.0
39.3
34.7
32.0
28.5
30.1
33.3
Mobile Combustion
45.6
41.3
20.5
21.9
18.7
19.4
19.3
Incineration of Waste
13.3
13.6
13.7
13.3
13.3
12.8
12.7
Abandoned Oil and Gas Wells
7.8
8.2
8.4
8.5
8.5
8.6
8.5
Abandoned Underground Coal Mines
8.1
7.4
6.9
6.6
6.5
6.3
6.3
Biomass and Biodiesel Consumption
237.9
245.4
336.0
333.1
295.7
303.0
305.4
International Bunker Fuelsc
104.6
114.3
125.3
114.6
70.3
80.9
99.1
Industrial Processes and Product Use
368.8
371.3
367.2
371.9
367.9
381.6
383.2
Substitution of Ozone Depleting
Substances
0.3
99.5
157.9
162.1
166.2
172.7
178.1
Cement Production
33.5
46.2
39.0
40.9
40.7
41.3
41.9
Iron and Steel Production &
Metallurgical Coke Production
104.8
70.1
42.9
43.1
37.7
41.9
40.7
Petrochemical Production
20.1
26.9
27.2
28.5
27.9
30.7
28.8
Ammonia Production
14.4
10.2
12.7
12.4
13.0
12.2
12.6
Lime Production
11.7
14.6
13.1
12.1
11.3
11.9
12.2
Other Process Uses of Carbonates
7.1
8.5
7.9
9.0
9.0
8.6
10.4
Nitric Acid Production
10.8
10.1
8.5
8.9
8.3
7.9
8.6
Fluorochemical Production
70.9
30.0
8.7
9.3
6.9
7.1
7.8
Trends 2-9
-------
Urea Consumption for Non-Agricultural
1
1
Purposes
3.8
3.7
6.1
6.2
5.8
6.6
7.1
Electrical Equipment
24.7
11.9
5.0
6.1
5.9
6.0
5.1
Carbon Dioxide Consumption
^ ^ llllll!
4.1
4.9
5.0
5.0
5.0
Electronics Industry
3.3
4.5
4.8
4.5
4.5
4.8
4.7
N20 from Product Uses
3.8
3.8
;;;;;;;
3.8
3.8
3.8
3.8
3.8
Aluminum Production
26.1
7.2
2.9
3.3
3.2
2.5
2.2
Adipic Acid Production
13.5
6.3
:::::::
9.3
4.7
7.4
6.6
2.1
Glass Production
2.3
2.4
2.0
1.9
1.9
2.0
2.0
Soda Ash Production
1.4
1.7 i
1.7
1.8
1.5
1.7
1.7
Titanium Dioxide Production
1.2
1.8
1.5
1.3
1.3
1.5
1.5
Ferroalloy Production
2.2
!!!!!!!
1.4 ®
2.1
1.6
1.4
1.6
1.3
Caprolactam, Glyoxal, and Glyoxylic
Acid Production
1.5
1.9
1.3
1.2
1.1
1.2
1.3
Magnesium Production and Processing
5.7
3.0
1.1
1.0
0.9
1.2
1.2
Zinc Production
0.6
1.0
1.0
1.0
1.0
1.0
0.9
Phosphoric Acid Production
i-s:
1,3 '
0.9
0.9
0.9
0.9
0.8
SF6 and PFCs from Other Product Use
1.4
1.5
0.9
0.8
0.7
0.5
0.8
Lead Production
o
Ln
o.6:
0.5
0.5
0.5
0.4
0.4
Carbide Production and Consumption
0.3
0.2
0.2
0.2
0.2
0.2
0.2
Agriculture
551.1
581.8
642.4
620.1
599.7
604.8
593.4
Agricultural Soil Management
288.8
294.1
333.4
315.6
292.1
298.0
290.8
Enteric Fermentation
183.1
188.2
196.8
197.3
196.3
196.5
192.6
Manure Management
52.5
70.2
84.3
83.5
83.8
83.6
81.7
Rice Cultivation
18-9 1
20.6;;;
19.9
15.6
18.6
18.3
18.9
Urea Fertilization
2.4
3.5
4.9
5.0
5.1
5.2
5.3
Liming
4.7 S
4.4
2.2
2.2
2.9
2.4
3.3
Field Burning of Agricultural Residues
0.7
0.8
0.8
0.9
0.8
0.8
0.8
Waste
235.9
192.0
173.2
175.8
171.7
169.2
166.9
Landfills
197.8
147.7
126.3
128.7
124.1
122.0
119.8
Wastewater Treatment
37.5 I
40.7 1
42.5
42.7
43.2
42.7
42.7
Composting
0.7
3.6
4.3
4.3
4.4
4.4
4.4
Anaerobic Digestion at Biogas Facilities
+ ¦
+
+
+
+
+
+
Total Gross Emissions'1 (Sources)
6,536.9
7,494.6
6,752.7
6,590.1
6,001.8
6,328.8
6,343.2
LULUCF Sector Net Totale
(976.7)
(907.7)
(915.5)
(863.6)
(904.4)
(910.6)
(854.2)
Forest Land
(1,069.0)
(960.2)
(963.8)
(907.3)
(946.6)
(924.2)
(872.0)
Cropland
40.4 -
2.9
14.2
12.0
20.5
2.9
3.4
Grassland
59.8
46.7
54.9
54.3
45.8
36.0
39.6
Wetlands
44.0
41.2
38.9
38.9
38.8
38.8
38.8
Settlements
(51.9)
(38.1)
(59.7)
(61.4)
(63.0)
(64.1)
(64.1)
Net Emission (Sources and Sinks)f
5,560.2
6,586.9
5,837.3
5,726.6
5,097.4
5,418.2
5,489.0
+ Does not exceed 0.05 MMT C02 Eq.
a Includes CH4 and N20 emissions from fuel combustion.
b Emissions from biomass and biofuel 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 LULUCF.
c Emissions from international bunker fuels are not included in totals.
d Total emissions without LULUCF.
e LULUCF emissions of CH4 and N20 are reported separately from gross emissions totals. LULUCF emissions include the CH4
and N20 emissions reported for peatlands remaining peatlands, forest fires, drained organic soils, grassland fires, and
coastal wetlands remaining coastal wetlands; CH4 emissions from land converted to coastal wetlands, flooded land
remaining flooded land, and land converted to flooded land; and N20 emissions from forest soils and settlement soils. Refer
to Table 2-8 for a breakout of emissions and removals for LULUCF by gas and source category.
f Net emissions with LULUCF.
Notes: Total (gross) emissions are presented without LULUCF. Net emissions are presented with LULUCF. Totals may not sum
due to independent rounding. Parentheses indicate negative values or sequestration.
2-10 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
Energy
Emissions from energy-related activities come from two main categories: 1) direct emissions associated with fuel
use (i.e., fossil fuel combustion, non-energy use of fossil fuels and waste combustion), and 2) fugitive emissions
mainly from coal, natural gas, and oil production. Energy emissions also include some categories that are not
added to Energy sector totals but are instead presented as memo items, including international bunker fuels and
biomass emissions. Energy-related activities, primarily fossil fuel combustion, accounted for the vast majority of
U.S. CO2 emissions from 1990 through 2022. Fossil fuel combustion is the largest source of energy-related
emissions, with CO2 being the primary gas emitted (see Figure 2-5). Due to their relative importance, fossil fuel
combustion-related CO2 emissions are considered in detail in the Energy chapter (see Chapter 3).
In 2022, 83.0 percent of the energy used in the United States on a Btu basis was produced through the combustion
of fossil fuels. The remaining 17.0 percent came from other energy sources such as hydropower, biomass, nuclear,
wind, and solar energy. A discussion of specific trends related to CO2 and other greenhouse gas emissions from
energy use is presented here with more detail in the Energy chapter. Energy-related activities are also responsible
for Cm and N2O emissions (40.2 percent and 10.8 percent of gross total U.S. emissions of each gas, respectively).2
Table 2-4 presents greenhouse gas emissions from the Energy chapter by source and gas.
Figure 2-5: Trends in Energy Sector Greenhouse Gas Sources
I Incineration of Waste
I U.S Territories Fossil Fuel Combustion
J Non-Energy Use of Fuels
Commerical Fossil Fuel Combustion
Residential Fossil Fuel Combustion
I Fugitive Emissions
I Industrial Fossil Fuel Combustion
I Transportation Fossil Fuel Combustion
I Electric Power Fossil Fuel Combustion
m ro m
i. (N
ID
¦*" CO
R
ltT
3,000
2,000
1,000
o^Hrsiro,3-Lnt£ir^oo<7>o^HrNrQ'3-Lnu3rvoo!^o^-o^-oooooooooo^HT-i*HT-i*-i^H!-iT-i*H»-CT1C^CT>CT>CT1C^CTICT>00000000000000000000000
HHHHHHHHHHNNNNfMMMNNNfMNfMNNMNNNfMtMNN
2 The contribution of energy non-C02 emissions is based on gross totals so excludes LULUCF methane (CH4) and nitrous oxide
(N20) emissions. The contribution of energy-related CH4 and N20 including LULUCF non-C02 emissions, is 37.1 percent and 9.8
percent, respectively.
Trends 2-11
-------
Table 2-4: Emissions from Energy by Gas (MMT CO2 Eq.)3
Percent
Change
Since
Gas/Source
1990
2005
2018
2019
2020
2021
2022
1990
co2
4,910.9
5,923.1
5,190.6
5,059.1
4,520.2
4,840.7
4,875.5
-0.7%
Fossil Fuel Combustion
4,752.2 :
5,744.1 :
4,988.2
4,852.6
4,341.7
4,654.3
4,699.4
-1.1%
Transportation
1,468.9
1,858.6
1,813.1
1,816.6
1,572.8
1,753.5
1,751.3
19.2%
Electricity Generation
1,820.0 ;
2,400.1
847.6
1,753.4
1,606.7
1,439.6
1,540.9
1,531.7
-15.8%
Industrial
876.5
810.5
809.8
762.0
780.5
801.1
-8.6%
Residential
338.6
358.9
338.9
342.9
314.8
318.0
334.1
-1.3%
Commercial
228.3
227.1
246.3
251.7
229.3
237.5
258.7
13.3%
U.S. Territories
20.0
5i.9 ;
25.9
24.8
23.3
23.8
22.6
12.8%
Non-Energy Use of Fuels
99.1
125.0
118.4
106.5
97.8
111.6
102.8
3.7%
Natural Gas Systems
32.4
26.3
32.8
38.5
36.7
35.8
36.5
12.5%
Petroleum Systems
9.6
10.2
34.8
45.5
28.9
24.1
22.0
129.2%
Incineration of Waste
12.9
13.3
13.3
12.9
12.9
12.5
12.4
-4.2%
Coal Mining
4.6
4.2
3.1
3.0
2.2
2.5
2.5
-46.3%
Abandoned Oil and Gas Wells
¦
+ B
+
+
+
+
+
+
13.0%
Biomass-Wood"
215.2
206.9
220.0
217.7
190.6
192.5
195.3
-9.2%
International Bunker Fuelsb
103.6
113.3
124.3
113.6
69.6
80.2
98.2
-5.2%
Biofuels-Ethanola
4.2
22.9
81.9
82.6
71.8
79.1
79.6
1783.2%
Biofuels-Biodiesela
0.0 :
0.9
17.9
17.1
17.7
16.1
15.6
100.0%
Biomass-MSWa
18.5
14.7
16.1
15.7
15.6
15.3
14.9
-19.8%
ch4
409.0
358.5
336.2
321.7
305.3
293.3
282.4
-31.0%
Natural Gas Systems
218.8
210.1
190.3
188.7
180.3
174.6
173.1
-20.9%
Coal Mining
108.1 -
71.5
59.1
53.0
46.2
44.7
43.6
-59.6%
Petroleum Systems
49.4
48.2
59.0
52.2
53.3
48.6
39.6
-19.8%
Stationary Combustion
9.7
—i
00
00
9.6
9.8
8.0
8.0
8.6
-10.8%
Abandoned Oil and Gas Wells
7.8
8.2
8.4
8.5
8.5
8.6
8.5
8.8%
Abandoned Underground
8.1
7.4
6.9
6.6
6.5
6.3
6.3
-21.8%
Coal Mines
1
1
Mobile Combustion
7.2
4.3
2.8
2.9
2.5
2.6
2.6
-63.8%
Incineration of Waste
+ ?!
ill!
+
+
+
+
+
+
-17.9%
International Bunker Fuelsb
0.2
0.1
0.1
0.1
0.1
0.1
0.1
-51.8%
n2o
61.2
67.9
43.2
41.6
37.1
39.2
41.9
-31.5%
Stationary Combustion
22.3
30.5
25.1
22.2
20.5
22.0
24.7
10.6%
Mobile Combustion
38.4
37.0 §
17.7
19.1
16.1
16.8
16.7
-56.5%
Incineration of Waste
0.4
0.3
0.4
0.4
0.3
0.4
0.3
-17.9%
Natural Gas Systems
+ j
+ 1
+
+
+
+
0.2
3,204.7%
Petroleum Systems
+
+
+
+
+
+
+
282.0%
International Bunker Fuelsb
0.8 1
0.9 1
1.0
0.9
0.5
0.6
0.8
1.4%
Total
5,381.0
'
6,349.5 a
5,570.0
5,422.4
4,862.6
5,173.3
5,199.8
-3.4%
+ Does not exceed 0.05 MMT C02 Eq.
a Emissions from biomass and biofuel consumption are not included specifically in Energy sector totals. Net carbon fluxes
from changes in biogenic carbon reservoirs are accounted for in the estimates for LULUCF.
b Emissions from international bunker fuels are not included in totals. These values are presented for informational purposes
only, in line with the 2006IPCC Guidelines, the Paris Agreement and the UNFCCC reporting obligations.
Note: Totals may not sum due to independent rounding.
3 The full time series data is available in Common Reporting Tables (CRTs) included in the U.S. Paris Agreement and UNFCCC
submission and in CSV format available at https://www.epa.Eov/ghgemissions/inventorv-us-greenhouse-gas-emissions-and-
sinks.
2-12 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
Fossil Fuel Combustion CO2 Emissions
As the largest contributor to U.S. greenhouse gas emissions, CO2 from fossil fuel combustion has accounted for
approximately 74.6 percent of CCh-equivalent total gross emissions on average across the time series. Within the
United States, fossil fuel combustion accounted for 93.0 percent of CO2 emissions in 2022. Emissions from this
source category include CO2 associated with the combustion of fossil fuels (coal, natural gas, and petroleum) for
energy use. Fossil fuel combustion CO2 emissions decreased by 1.1 percent (52.8 MMT CO2 Eq.) from 1990 to 2022
and were responsible for most of the decrease in national emissions during this period. Similarly, CO2 emissions
from fossil fuel combustion have decreased by 18.2 percent (1,044.7 MMT CO2 Eq.) since 2005. From 2021 to
2022, these emissions increased by 1.0 percent (45.1 MMT CO2 Eq.).
Historically, changes in emissions from fossil fuel combustion have been the main factor influencing U.S. emission
trends. Changes in CO2 emissions from fossil fuel combustion since 1990 are affected by many long-term and
short-term factors, including population and economic growth, energy price fluctuations and market trends,
technological changes, carbon intensity of energy fuel choices, and seasonal temperatures. On an annual basis, the
overall consumption and mix of fossil fuels in the United States fluctuates in response to changes in general
economic conditions, overall energy prices, the relative price of different fuels, weather, and the availability of
non-fossil alternatives. For example, coal consumption for electric power is influenced by factors such as the
relative price of coal and alternative sources, the ability to switch fuels, and longer-term trends in coal markets.
Fossil fuel combustion CO2 emissions also depend on the type of fuel consumed or energy used and its carbon
intensity. Producing a unit of heat or electricity using natural gas instead of coal, for example, reduces CO2
emissions because of the lower carbon content of natural gas (see Table 3-12 in Chapter 3 for more detail on
electricity generation by source and see Table A-19 in Annex 2.1 for more detail on the carbon content coefficient
of different fossil fuels).
Overall CO2 emissions from electric power generation decreased by 36.2 percent from 2005 to 2022 (see Figure
2-7), reflecting the continued shift in the share of electric power generation from coal to natural gas and
renewables since 2005. Carbon dioxide emissions from coal combustion for electric power generation gradually
increased between 1990 and 2007, then began to decrease at a faster rate from 2008 to 2022. Carbon dioxide
emissions from natural gas combustion for electric power generation remained relatively constant, with a slight
increase between 1990 and 2009, then began to consistently increase between 2010 and 2022.
Petroleum use is another major driver of CO2 emissions from fossil fuel combustion, particularly in the
transportation sector, which has represented the largest source of CO2 emissions from fossil fuel combustion since
2017. Emissions from petroleum consumption for transportation (including bunker fuels) decreased by 0.1 percent
from 2021 to 2022. Fuel economy of light-duty vehicles is an important factor in transportation sector CO2
emissions trends. The decline in new light-duty vehicle fuel economy between 1990 and 2004 reflected the
increasing market share of light-duty trucks, which grew from about 29.6 percent of new vehicle sales in 1990 to
48.0 percent in 2004. Starting in 2005, average new vehicle fuel economy began to increase while light-duty VMT
grew only modestly for much of the period and has slowed the rate of increase of CO2 emissions.
Trends in CO2 emissions from fossil fuel combustion by end-use sector are presented in Table 2-5 and Figure 2-6
based on the underlying U.S. energy consumer data collected by the U.S. Energy Information Administration (EIA).
Figure 2-7 further describes trends in direct and indirect CO2 emissions from fossil fuel combustion by end-use
sector. Estimates of CO2 emissions from fossil fuel combustion are calculated from these EIA "end-use sectors"
based on total fuel consumption and appropriate fuel properties described below.4
• Transportation. ElA's fuel consumption data for the transportation sector consists of all vehicles whose
primary purpose is transporting people and/or goods from one physical location to another.
• Electric Power. ElA's fuel consumption data for the electric power sector are composed of electricity-only
and combined-heat-and-power (CHP) plants within the North American Industry Classification System
(NAICS) 22 category whose primary business is to sell electricity, or electricity and heat, to the public.
4 Additional analysis and refinement of the EIA data is further explained in the Energy chapter of this report.
Trends 2-13
-------
(Non-utility power producers are included in this sector as long as they meet the electric power sector
definition.)
• Industry. EIA statistics for the industrial sector include fossil fuel consumption that occurs in the fields of
manufacturing, agriculture, mining, and construction. ElA'sfuel consumption data for the industrial sector
consist of all facilities and equipment used for producing, processing, or assembling goods. (EIA includes
generators that produce electricity and/or useful thermal output primarily to support on-site industrial
activities in this sector.)
• Residential. ElA's fuel consumption data for the residential sector consist of living quarters for private
households.
• Commercial. ElA's fuel consumption data for the commercial sector consist of service-providing facilities
and equipment from private and public organizations and businesses. (EIA includes generators that
produce electricity and/or useful thermal output primarily to support the activities at commercial
establishments in this sector.)
Table 2-5: CO2 Emissions from Fossil Fuel Combustion by End-Use Sector (MMT CO2 Eq.)
End-Use Sector
1990
2005
2018
2019
2020
2021
2022
Transportation
1,472.0
1,863.3
1,817.9
1,821.4
1,576.9
1,758.6
1,757.4
Combustion
1,468.9 !!!!
1,858.6 1
1,813.1
1,816.6
1,572.8
1,753.5
1,751.3
Electricity
3.0
4.7
4.8
4.8
4.1
5.0
6.1
Industrial
1,562.9
1,584.0
1,311.8
1,275.3
1,171.8
1,225.6
1,238.0
Combustion
876.5
847.6
810.5
809.8
762.0
780.5
801.1
Electricity
686-4 I
736.3
501.3
465.5
409.8
445.1
437.0
Residential
931.3
1,214.9
981.2
926.7
860.1
890.3
899.4
Combustion
338.6 1
358.9 2
338.9
342.9
314.8
318.0
334.1
Electricity
592.7
856.0
642.3
583.7
545.3
572.2
565.3
Commercial
766.0
1,030.1
851.3
804.4
709.6
756.1
782.0
Combustion
228.3
227.1
246.3
251.7
229.3
237.5
258.7
Electricity
537.7 1
803.0 1
605.0
552.7
480.3
518.5
523.3
U.S. Territories3
20.0
51.9
25.9
24.8
23.3
23.8
22.6
Total
4,752.2
5,744.1
4,988.2
4,852.6
4,341.7
4,654.3
4,699.4
Electric Power
1,820.0
2,400.1
1,753.4
1,606.7
1,439.6
1,540.9
1,531.7
a Fuel consumption by U.S. Territories (i.e., American Samoa, Guam, Puerto Rico, U.S. Virgin Islands, Wake Island, and other
outlying U.S. Pacific Islands) is included in this report.
Notes: Combustion-related emissions from electric power are allocated based on aggregate national electricity use by each
end-use sector. Totals may not sum due to independent rounding.
2-14 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
Figure 2-6: Trends in CO2 Emissions from Fossil Fuel Combustion by End-Use Sector and Fuel
Type
Coal ¦ Geothermal
U.S. Territories
Commercial
Residential
1992 1998 2004 2010 2016 2022
Natural Gas
_ 2,000
1,000
o
o
2,000
cr
O
u
1,000
2,000
O
u
1,000
Petroleum
Industrial
Electric Power
Transportation
1992 1998 2004 2010 2016 2022
Note: Fossil fuel combustion for electric power also includes emissions of less than 0.5 MMT C02 Eq. from geothermal-based
generation. Although not technically a fossil fuel, geothermal energy-related C02 emissions are included for reporting
purposes. The source of C02 is non-condensable gases in subterranean heated water.
Trends 2-15
-------
Figure 2-7: Trends in End-Use Sector Emissions of CO2 from Fossil Fuel Combustion
I Direct Fossil Fuel Combustion
Indirect Fossil Fuel Combustion
U.S. Territories
Industrial
1990 1994 1998 2002 2006 2010 2014 2018 2022
Electric power was the second largest end-use emitter of CO2 in 2022 (surpassed by transportation in 2017);
electric power generators used 30.5 percent of U.S. energy from fossil fuels and emitted 32.6 percent of the CO2
from fossil fuel combustion in 2022. CO2 emissions from the fossil fuel combustion in the electric power sector
decreased by 0.6 percent between 2021 and 2022 due to changes in the mix of electric generation resources.
Between 2021 and 2022 overall electricity generation increased by 3 percent, coal electricity generation decreased
by 10.2 percent, natural gas generation increased by 4.0 percent, and renewable energy generation increased by
7.6 percent. Changes in electricity demand and the carbon intensity of fuels used for electric power generation
have a significant impact on CO2 emissions. Carbon dioxide emissions from fossil fuel combustion from the electric
power sector have decreased by 15.8 percent since 1990, and the carbon intensity of the electric power sector, in
terms of CO2 Eq. per QBtu input, has decreased by 27.6 percent during that same timeframe. This decoupling of
electric power generation and the resulting CO2 emissions is shown below in Figure 2-8.
2-16 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
Figure 2-8: Electric Power Generation (Billion kWh) and Emissions (MMT CO2 Eq.)
4,500
4,000
' 3,500
C
1 3,000
In
c
o
E 2,500
aj
c
a^cri
-------
• Emissions (Cm, CO2, and N2O) from oil and gas systems decreased by 13.0 percent (38.8 MMT CO2 Eq.)
since 1990 and decreased by 4.2 percent (11.8 MMT CO2 Eq.) from 2021 to 2022. Natural gas systems CH4
emissions have decreased by 20.9 percent (45.7 MMT CO2 Eq.) since 1990, due to a decrease in emissions
from the distribution, transmission and storage, processing, and exploration segments. The decrease in
distribution emissions is due mainly to reduced emissions from pipeline and distribution station leaks, and
the decrease in transmission and storage emissions is due mainly to reduced compressor station
emissions (including emissions from compressors and leaks). Over the same time period (i.e., since 1990),
methane emissions from the natural gas production segment increased due to increased gathering and
boosting emissions. Between 2021 and 2022, methane emissions from natural gas systems decreased 0.8
percent, due to a decrease in emissions from production segment pneumatic controllers. Petroleum
systems CH4 emissions decreased by 19.8 percent (9.8 MMT CO2 Eq.) since 1990 and 18.5 percent
between 2021 and 2022. This decrease is due primarily to decreases in emissions from offshore platforms,
tanks, and pneumatic controllers. Carbon dioxide emissions from natural gas and petroleum systems
increased by 39.1 percent (16.4 MMT CO2) from 1990 to 2022 and decreased by 2.5 percent between
2021 and 2022. This increase since 1990 is due primarily to increases in the production segment, where
emissions from associated gas flaring, tanks, and miscellaneous production flaring have increased over
time. The decrease in emissions between 2021 and 2022 and is also due primarily to the production
segment, where flaring emissions decreased for associated gas and tanks.
• Methane emissions from coal mining decreased by 59.6 percent (64.4 MMT CO2 Eq.) from 1990 through
2022 and by 2.3 percent between 2021 and 2022 primarily due to a decrease in the number of active
mines and annual coal production over the 1990 to 2022 time period. Between 2021 and 2022, the
number of mines and coal production increased.
• Nitrous oxide emissions from mobile combustion decreased by 56.5 percent (21.7 MMT CO2 Eq.) from
1990 through 2022 and by 0.7 percent (0.1 MMT CO2 Eq.) between 2021 and 2022, primarily as a result of
national vehicle criteria pollutant emissions standards and emission control technologies for on-road
vehicles.
• Nitrous oxide emissions from stationary combustion were the third largest source of anthropogenic N2O
emissions in 2022, accounting for 6.3 percent of N2O emissions and 0.4 percent of total gross U.S.
greenhouse gas emissions in 2022. Stationary combustion emissions peaked in 2007 and have steadily
decreased since then.
• Carbon dioxide emissions from non-energy uses of fossil fuels increased by 3.7 percent (3.7 MMT CO2 Eq.)
from 1990 through 2022 but decreased by 7.9 percent (8.8 MMT CO2 Eq.) between 2021 and 2022.
Emissions from non-energy uses of fossil fuels were 102.8 MMT CO2 Eq. in 2022, which constituted 2.0
percent of total national CO2 emissions, approximately the same proportion as in 1990.
• Carbon dioxide emissions from incineration of waste decreased slightly by 4.2 percent (0.5 MMT CO2 Eq.)
from 1990 through 2022, as the volume of scrap tires and other fossil carbon-containing materials in
waste decreased. Emissions decreased 0.9 percent (0.1 MMT CO2 Eq.) between 2021 and 2022, consistent
with trends across the time series.
Industrial Processes and Product Use
Greenhouse gases can be generated and emitted by industry in two different ways. First, they are generated and
emitted as the byproducts of many non-energy-related industrial activities. For example, industrial processes can
chemically or physically transform raw materials, which often release waste gases such as CO2, CFU, N2O, and
fluorinated gases (e.g., HFC-23). In the case of byproduct emissions, the emissions are generated by an industrial
process itself, and are not directly a result of energy consumed during the process.
Second, industrial manufacturing processes and use by end-consumers also release HFCs, PFCs, SF6, and NF3 and
other man-made compounds. In addition to the use of HFCs and some PFCs as substitutes for ozone depleting
substances (ODS), fluorinated compounds such as HFCs, PFCs, SF6, NF3, and others are also emitted through use by
2-18 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
a number of other industrial sources in the United States. These industries include the electronics industry,
electrical equipment, and magnesium metal production and processing. In addition, N2O is used in and emitted by
the electronics industry and anesthetic and aerosol applications, PFCs and SF6 are emitted from other product use,
and CO2 is consumed and emitted through various end-use applications.
Emission sources in the Industrial Processes and Product Use (IPPU) chapter accounted for 6.0 percent of U.S.
greenhouse gas emissions in 2022. Emissions from the IPPU sector increased by 3.9 percent from 1990 to 2022.
The use of HFCs as substitutes for ODS is the largest source of emissions in this sector, contributing 46.5 percent of
IPPU emissions in 2022 and driving growth since 1990. From 2021 to 2022, total emissions from IPPU increased 0.4
percent between 2021 and 2022. Despite the sectoral increase in emissions, emissions from adipic acid production
decreased by almost 70 percent, emissions from ferroalloy production decreased by over 10 percent, and
emissions from aluminum production, petrochemical production and zinc production decreased by between 5 and
10 percent. Figure 2-9 presents greenhouse gas emissions from IPPU by source category.
Figure 2-9: Trends in Industrial Processes and Product Use Sector Greenhouse Gas Sources
500
450
400
350
w 300
I Electronics Industry
Other Product Manufacture and Use
I Mineral Industry
I Metal Industry ^
I Chemical Industry
Substitution of Ozone Depleting Substances
co rv
m u-i
ro n
rv,
oj r\j
<3-
o
u
250
200
150
100
50
0
rv £
m
10
rv
vo
Oi
CO LO
rv.
ro
LO
ro
LO LO
rv y) 0 is
fO fO
ro ro
-------
Lime Production
11.7 5
14.6 j
13.1
12.1
11.3
11.9
12.2
4.3%
Other Process Uses of
Carbonates
7.1
8.5
7.9
9.0
9.0
8.6
10.4
46.2%
Urea Consumption for Non-
1
r--.
CO
Agricultural Purposes
3-8 i
6.1
6.2
5.8
6.6
7.1
86.4%
Carbon Dioxide Consumption
1.5
1.4
4.1
4.9
5.0
5.0
5.0
239.7%
Glass Production
2.3 S
2-4 1
2.0
1.9
1.9
2.0
2.0
-13.6%
Soda Ash Production
1.4
1.7
1.7
1.8
1.5
1.7
1.7
19.0%
Titanium Dioxide Production
1-2 ¦
1.8 1
1.5
1.3
1.3
1.5
1.5
23.3%
Aluminum Production
6.8
4.1
1.5
1.9
1.7
1.5
1.4
-78.8%
Ferroalloy Production
2.2
1.4
2.1
1.6
1.4
1.6
1.3
-38.3%
Zinc Production
0.6
1.0
1.0
1.0
1.0
1.0
0.9
49.9%
Phosphoric Acid Production
1-5 1
1.3 iiiiii;
0.9
0.9
0.9
0.9
0.8
-45.0%
Lead Production
0.5
0.6
0.5
0.5
0.5
0.4
0.4
-17.1%
Carbide Production and
1
I
Consumption
0-2:
0.2 E
0.2
0.2
0.2
0.2
0.2
-13.7%
Substitution of Ozone Depleting
I
Substances3
+
+
+
+
+
+
+
28,664.2%
Magnesium Production and
miiii
!!!!!!!
111111!
I
Processing
0.1 III
;
+
+
+
+
+
-97.7%
ch4
0.1
+
+
+
+
+
+
-48.9%
Carbide Production and
1
Consumption
+ '
+
+
+
+
+
-38.5%
Ferroalloy Production
+
+
+
+
+
+
+
-45.2%
Iron and Steel Production &
lllllll
iiiiiii
1
Metallurgical Coke Production
+
+
+
+
+
+
-67.8%
Petrochemical Production
+
+
+
+
+
+
+
-21.7%
n2o
29.6
22.2
23.1
18.7
20.8
19.7
16.1
-45.7%
Nitric Acid Production
10.8
10.1
8.5
8.9
8.3
7.9
8.6
-20.2%
N20 from Product Uses
3.8
3.8 |
3.8
3.8
3.8
3.8
3.8
-0.4%
AdipicAcid Production
13.5
6.3
9.3
4.7
7.4
6.6
2.1
-84.5%
Caprolactam, Glyoxal, and
1
1
Glyoxylic Acid Production
1.5 =
I-9
1.3
1.2
1.1
1.2
1.3
-10.5%
Electronics Industry
+
0.1
0.2
0.2
0.3
0.3
0.3
720.0%
HFCs
47.7
121.7
163.9
168.2
170.3
177.0
182.8
282.9%
Substitution of Ozone Depleting
Substances3
0.3
99.5
157.9
162.1
166.2
172.6
178.1
70,357.0%
Fluorochemical Production
47.3
22.11=
5.7
5.7
3.8
4.0
4.3
-90.9%
Electronics Industry
0.2
0.2
0.3
0.3
0.3
0.4
0.3
74.0%
Magnesium Production and
iiiiiii
mm;
mm!
Processing
NO 1
NO 1
0.1
0.1
0.1
+
+
100.0%
PFCs
39.5
10.2
7.4
7.3
6.6
6.3
6.7
-83.1%
Fluorochemical Production
17.5 =
4.0 j
2.9
3.0
2.5
2.6
3.0
-83.1%
Electronics Industry
2.5
3.0
2.9
2.6
2.5
2.6
2.7
7.8%
Aluminum Production
19-3
3-11
1.4
1.4
1.4
0.9
0.8
-96.1%
SF6 and PFCs from Other Product
Use
0.1
0.1
0.2
0.2
0.2
0.1
0.2
39.4%
Substitution of Ozone Depleting
IIIIIH
lllllll
1
Substances3
NO =
+
+
+
+
+
+
100.0%
Electrical Equipment
+
+
NO
+
+
+
+
-99.2%
SF6
37.9
20.2
7.6
8.4
8.1
8.5
7.6
-80.0%
Electrical Equipment
24.7
11.8
5.0
6.1
5.9
6.0
5.1
-79.4%
Magnesium Production and
IIIIIH
iiiiii;
I
Processing
5.6 |
3.0 1
1.1
0.9
0.9
1.2
1.1
-80.0%
Electronics Industry
0.5
0.8
0.8
0.8
0.8
0.9
0.8
47.0%
2-20 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
SFe and PFCs from Other Product
Use
1.3
1.3
1
0.8
0.6
0.5
0.4
0.6
-53.0%
Fluorochemical Production
5.8
3.3
+
+
+
+
+
-100.0%
nf3
0.3
1.0
0.7
1.1
1.3
1.1
1.1
238.3%
Electronics Industry
+
0.4
0.5
0.5
0.6
0.6
0.6
1313.8%
Fluorochemical Production
0.3 1
0.6
0.1
0.6
0.7
0.5
0.5
72.7%
Total
368.8
371.3
367.2
371.9
367.9
381.6
383.2
3.9%
+ Does not exceed 0.05 MMT C02 Eq.
NO (Not Occurring)
a Small amounts of PFC emissions from this source are included under HFCs due to confidential business information.
Note: Totals may not sum due to independent rounding.
IPPU sector emissions increased 0.4 percent since 2021 and 3.9 percent since 1990. Some significant trends in U.S.
emissions from IPPU source categories over the 33-year period from 1990 through 2022 included the following:
• HFC and PFC emissions resulting from the substitution of ODS (e.g., chlorofluorocarbons [CFCs]) increased
from small amounts in 1990 to 178.1 MMT CO2 Eq. in 2022 (an increase of 70,357 percent).
• Combined CO2 and Cm emissions from iron and steel production and metallurgical coke production
decreased by 2.9 percent from 2021 to 2022 to 40.7 MMT CO2 Eq. and declined overall by 61.2 percent
(64.1 MMT CO2 Eq.) from 1990 through 2022, due to restructuring of the industry. The trend in the United
States has been a shift toward fewer integrated steel mills and more electric arc furnaces (EAFs). EAFs use
scrap steel as their main input and generally have lower on-site emissions.
• Carbon dioxide emissions from petrochemical production increased by 43.4 percent between 1990 and
2022, from 20.1 MMT CO2 Eq. to 28.8 MMT CO2 Eq. The increase in emissions is largely driven by the
production of ethylene more than doubling over that period.
• Carbon dioxide emissions from ammonia production have decreased by 12.5 percent (1.8 MMT CO2 Eq.)
since 1990. Ammonia production relies on natural gas as both a feedstock and a fuel, and as such, market
fluctuations and volatility in natural gas prices affect the production of ammonia from year to year.
Emissions from ammonia production have increased since 2016, due to the addition of new ammonia
production facilities and new production units at existing facilities. Agricultural demands continue to drive
demand for nitrogen fertilizers and the need for new ammonia production capacity.
• Carbon dioxide emissions from cement production increased by 25.1 percent (8.4 MMT CO2 Eq.) from
1990 through 2022. Emissions rose from 1990 through 2006 and then fell until 2009, due to a decrease in
demand for construction materials during the economic recession. Since 2010, CO2 emissions from
cement production have risen by 33.2 percent.
• HFC, PFC, SFe, and NF3 emissions from fluorochemical production decreased by 89.0 percent (63.2 MMT
CO2 Eq.) from 1990 to 2022 due to a reduction in the HFC-23 emission rate from HCFC-22 production (kg
HFC-23 emitted/kg HCFC-22 produced), the imposition of emissions controls at production facilities, and a
decrease in SF6 production due to the cessation of production at the major SF6 production facility in 2010.
• PFC emissions from aluminum production decreased by 96.1 percent (18.5 MMT CO2 Eq.) from 1990 to
2022, due to both industry emission reduction efforts and lower domestic aluminum production.
• SFe emissions from electrical equipment decreased by 79.4 percent (19.6 MMT CO2 Eq.) from 1990 to
2022 due to a sharp increase in the price of SF6 during the 1990s and industry emission reduction efforts.
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, liming, urea fertilization, and field burning of
agricultural residues. Methane and N2O are the primary greenhouse gases emitted by agricultural activities, with
Trends 2-21
-------
small amounts of CO2 also emitted.5 Carbon stock changes from agricultural soils are included in the LULUCF
sector.
In 2022, agricultural activities were responsible for emissions of 593.4 MMT CO2 Eq., or 9.4 percent of total U.S.
greenhouse gas emissions. Agricultural soil management activities, such as the application of synthetic and organic
fertilizers, deposition of livestock manure, and growing N-fixing plants, were the largest contributors to
agricultural-related emissions (49.0 percent) and were the largest source of U.S. N2O emissions in 2022, accounting
for 74.6 percent. Methane emissions from enteric fermentation and manure management represented 27.4
percent and 9.2 percent of total Cm emissions from anthropogenic activities, respectively, in 2022. Carbon dioxide
emissions from the application of crushed limestone and dolomite (i.e., soil liming) and urea fertilization
represented 0.2 percent of total CO2 emissions from anthropogenic activities. Figure 2-10 and Table 2-7 illustrate
agricultural greenhouse gas emissions by source and gas.
Figure 2-10: Trends in Agriculture Sector Greenhouse Gas Sources
Field Burning of Agricultural Residues
I Urea Fertilization
I Liming
Rice Cultivation
I Manure Management
I Enteric Fermentation
I Agricultural Soil Management
^ in
vo ID
S ft ft s _ffi J2. 10 " 10
LO *£> u ' LP) UJ
"
o J (N n _
— ® —»L R ¦¦¦¦
¦ ¦£.
s
H S
450
S 400
o
v 350
2:
51 300
O
CT>
i-HfNro^-LnvDr^cocn
0^0^0^0~>0^0^0^0^CTi
OT-nrNro^-LouDr^cocnoT-irsjm^-Lo
OOOOOOOOOO'-H'-H'-H'-H'-H'-l
0000000000000000
r\ir\ir\ir\ir\ir\ir\irsjrsjrsjr\jc\icsjrsjcsjr\i
00 O T-H CM
*—• 1 t—1 t—1 r\i rvi rsi
0000000
fN fN fN fN fN fN rvj
5 The contribution of agriculture non-C02 emissions is based on gross totals and excludes LULUCF methane (CH4) and nitrous
oxide (N20) emissions. The contribution of agriculture CH4 and N20 including LULUCF non-C02 emissions, is 40.5 percent and
48.3 percent, respectively.
2-22 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
Table 2-7: Emissions from Agriculture (MMT CO2 Eq.)
Percent
Change
Since
Gas/Source
1990
2005
2018
2019
2020
2021
2022
1990
C02
7.1
7.9
7.2
7.2
8.0
7.6
8.6
21.0%
Urea Fertilization
2.4 •
3-5 1
4.9
5.0
5.1
5.2
5.3
120.4%
Liming
4.7
4.4
2.2
2.2
2.9
2.4
3.3
-30.3%
ch4
241.7
264.4
285.0
280.2
282.4
281.8
276.8
14.5%
Enteric Fermentation
183.1
188.2
196.8
197.3
196.3
196.5
192.6
5.2%
Manure Management
39.1 ;
55-° 1
67.7
66.7
66.9
66.4
64.7
65.3%
Rice Cultivation
18.9
20.6
19.9
15.6
18.6
18.3
18.9
-0.4%
Field Burning of Agricultural
Residues
°-5
0-6 1
0.6
0.7
0.6
0.6
0.6
14.4%
n2o
302.3
309.5
350.2
332.6
309.2
315.3
308.0
1.9%
Agricultural Soil Management
288.8
294.1 1
333.4
315.6
292.1
298.0
290.8
0.7%
Manure Management
13.4
15.2
16.6
16.8
16.9
17.1
17.0
27.2%
Field Burning of Agricultural
IS
I
1
Residues
0.2
0.2 1
0.2
0.2
0.2
0.2
0.2
15.5%
Total
551.1
581.8
642.4
620.1
599.7
604.8
593.4
7.7%
Note: Totals may not sum due to independent rounding.
Agriculture sector emissions decreased by 1.9 percent since 2021 and increased by 7.7 percent since 1990. Some
significant trends in U.S. emissions from Agriculture source categories (Figure 2-10) over the 33-year time series
from 1990 through 2022 included the following:
• Agricultural soils are the largest anthropogenic source of agriculture-related emissions and of N2O
emissions in the United States, accounting for 74.6 percent of N2O emissions and 4.6 percent of total
emissions in the United States in 2022. Annual N2O emissions from agricultural soils fluctuated between
1990 and 2022, and overall emissions were 0.7 percent (2.0 MMT CO2 Eq.) higher in 2022 than in 1990.
Year-to-year fluctuations are largely a reflection of annual variation in weather patterns, synthetic
fertilizer use, and crop production.
• Enteric fermentation is the largest anthropogenic source of CH4 emissions in the United States. In 2022,
enteric fermentation CFU emissions were 27.4 percent of total CFU emissions, which represents an
increase of 5.2 percent (9.5 MMT CO2 Eq.) since 1990. This increase in emissions from enteric
fermentation from 1990 to 2022 generally follows the increasing trends in cattle populations. For
example, from 1990 to 1995, emissions increased and then generally decreased from 1996 to 2004,
mainly due to fluctuations in beef cattle populations and increased digestibility of feed for feedlot cattle.
Emissions increased from 2005 to 2007, as both dairy and beef populations increased. Research indicates
that the feed digestibility of dairy cow diets decreased during this period. Emissions decreased again from
2008 to 2014 as beef cattle populations again decreased. Emissions increased from 2014 to 2022,
consistent with an increase in beef cattle population over those same years. CFU emissions from enteric
fermentation decreased by 2.0 percent (3.9 MMT CO2 Eq.) from 2021 to 2022, however, largely driven by
a decrease in beef cattle populations.
• Manure management is the fourth largest anthropogenic source of CH4 and N2O emissions in the United
States and accounted for 13.8 percent of Agriculture sector emissions in 2022. Emissions from manure
management increased by 55.6 percent between 1990 and 2022. This includes an increase of 65.3 percent
(25.6 MMT CO2 Eq.) for CFU and an increase of 27.2 percent (3.6 MMT CO2 Eq.) for N2O. The majority of
the increase observed in CFU emissions resulted from swine and dairy cattle manure, where emissions
increased by 37.2 and 108.7 percent, respectively, from 1990 to 2022. From 2021 to 2022, CFU emissions
from manure management decreased by 2.6 percent, mainly due to minor shifts in the animal populations
and the resultant effects on manure management system allocations.
Trends 2-23
-------
• Liming and urea fertilization are the only sources of CO2 emissions reported in the Agriculture sector. All
other CO2 emissions and removals (e.g., carbon stock changes from the management of croplands) are
included in the LULUCF sector. Liming emissions increased by 36.9 percent relative to 2021 and decreased
by 30.3 percent (1.4 MMT CO2 Eq.) relative to 1990, while urea fertilization emissions increased by 1.9
percent relative to 2021 and 120.4 percent (2.9 MMT CO2 Eq.) relative to 1990.
Land Use, Land-Use Change, and Forestry
When humans alter the terrestrial biosphere through land use, changes in land use, and land management
practices, they also influence the carbon stock fluxes on these lands and cause emissions of CH4 and N2O. Overall,
managed land is a net sink for CO2 (carbon sequestration) in the United States. The primary driver of fluxes on
managed lands is from management of forest lands, but also includes trees in settlements (i.e., urban areas),
afforestation, conversion of forest lands to settlements and croplands, the management of croplands and
grasslands, flooded lands, and the landfilling of yard trimmings and food scraps. The main drivers for net forest
sequestration include net forest growth, increasing forest area, and a net accumulation of carbon stocks in
harvested wood pools. The net sequestration in settlements remaining settlements is driven primarily by carbon
stock gains in urban forests (i.e., settlement trees) through net tree growth and increased urban area, as well as
long-term accumulation of carbon in landfills from additions of yard trimmings and food scraps.
The LULUCF sector in 2022 resulted in a net increase in carbon stocks (i.e., net CO2 removals) of 921.8 MMT CO2
Eq. (Table 2-8).6 This represents an offset of 14.5 percent of total (i.e., gross) greenhouse gas emissions in 2022.
Emissions of CFU and N2O from LULUCF activities in 2022 were 67.6 MMT CO2 Eq. and represented 1.2 percent of
net greenhouse gas emissions.7 Between 1990 and 2022, total net carbon sequestration in the LULUCF sector
decreased by 10.9 percent, primarily due to a decrease in the rate of net carbon accumulation in forests and
cropland remaining cropland, as well as an increase in CO2 emissions from land converted to settlements.
Flooded land remaining flooded land was the largest source of CH4 emissions from LULUCF and the fifth largest
source overall of net CFU emissions in 2022, totaling 44.2 MMT CO2 Eq. (1,579 kt of CH4). Forest fires were the
second largest source of CH4 emissions from LULUCF in 2022, totaling 9.1 MMT CO2 Eq. (327 kt of CH4). Forest fires
were the largest source of N2O emissions from LULUCF in 2022, totaling 5.7 MMT CO2 Eq. (22 kt of N2O). Figure
2-11 and Table 2-8 illustrate LULUCF emissions and removals by land-use category and gas.
6 LULUCF carbon stock change is the net carbon stock change from the following categories: forest land remaining forest land,
land converted to forest land, cropland remaining cropland, land converted to cropland, grassland remaining grassland, land
converted to grassland, wetlands remaining wetlands, land converted to wetlands, settlements remaining settlements, and
land converted to settlements.
7 LULUCF emissions include the CH4 and N20 emissions reported for peatlands remaining peatlands, forest fires, drained
organic soils, grassland fires, and coastal wetlands remaining coastal wetlands; CH4 emissions from flooded land remaining
flooded land, land converted to flooded land, and land converted to coastal wetlands; and N20 emissions from forest soils and
settlement soils.
2-24 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
Figure 2-11: Trends in Emissions and Removals (Net CO2 Flux) from Land Use, Land-Use
Change, and Forestry
600
400
I Forest Land Remaining Forest Land
I Land Converted to Cropland
I Land Converted to Settlements
Land Converted to Wetlands
I Settlements Remaining Settlements
Cropland Remaining Cropland
Wetlands Remaining Wetlands
Grassland Remaining Grassland
Land Converted to Forest Land
Land Converted to Grassland
¦ Net Emissions (Sources and Sinks)
» £ 2? £
200
m 00 o
fN m ^
CO
K P\
co D! 01
CO
a o> a
-200
-400
-600
-800
-1,000
1
"
-1,200
OTHfNro^LnvjDrvCOChO'-HrgroTLn
cncr.CT\cricr>cricr>cno^cr>oooooo
o^o^0">cr»cho>0'»^oooooo
vorvOOCTiOi-irsirO'^-Ln^DrvCoaiO'rHrNj
OOOOOOOOOOOOOOOOO
fNfNlN(NrMfNrMPJrMrM(N(NfNr\JlN(NrM
Table 2-8: U.S. Greenhouse Gas Emissions and Removals (Net Flux) from Land Use, Land-Use
Change, and Forestry (MMT CO2 Eq.)
Land-Use Category
1990
2005
2018
2019
2020
2021
2022
Percent
Change
Since
1990
Forest Land Remaining Forest Land
(968.8)
(860.1)
(863.4)
(807.0)
(846.3)
(823.9)
(771.7)
-20.3%
Changes in Forest Carbon Stocks3
(974.8)
(876.0)
(873.5)
(813.2)
(862.0)
(844.2)
(787.0)
-19.3%
Non-C02 Emissions from Forest
Firesb
5.8
15.4
9.7
5.7
15.3
19.9
14.8
153.9%
N20 Emissions from Forest Soilsc
0.1
0.4
0.4
0.4
0.4
0.4
0.4
455.1%
Non-C02 Emissions from Drained
Organic Soilsd
0.1
0.1
0.1
0.1
0.1
0,1
0.1
0.0%
Land Converted to Forest Land
(100.2)
(100.2)
(100.4)
(100.3)
(100.3)
(100.3)
(100.3)
0.1%
Changes in Forest Carbon Stockse
(100.2)
(100.2)
(100.4)
(100.3)
(100.3)
(100.3)
(100.3)
0.1%
Cropland Remaining Cropland
(5.0)
(31.6)
(17.8)
(19.4)
(8.8)
(32.0)
(31.7)
529.0%
Changes in Mineral and Organic Soil
Carbon Stocks
(5.0)
(31.6)
(17.8)
(19.4)
(8.8)
(32.0)
(31.7)
529.0%
Land Converted to Cropland
45.4
34.5
31.9
31.4
29.3
34.9
35.1
-22.7%
Changes in all Ecosystem Carbon
Stocks'
45.4
34.5
31.9
31.4
29.3
34.9
35.1
-22.7%
Grassland Remaining Grassland
24.6
24.9
29.7
28.9
17.1
11.5
14.0
-43.1%
Trends 2-25
-------
Changes in Mineral and Organic Soil
1
1
Carbon Stocks
24.4
24.1
28.6
28.5
16.1
10.6
13.4
-45.2%
l\lon-C02 Emissions from Grassland
1
0.2 j
1
Firess
0-8 1
1.1
0.3
1.1
0.9
0.6
184.1%
Land Converted to Grassland
35.3
21.8
25.2
25.4
28.7
24.5
25.6
-27.3%
Changes in all Ecosystem Carbon
!!!!!#
35.3 \
1
Stocks'
21.8 1
25.2
25.4
28.7
24.5
25.6
-27.3%
Wetlands Remaining Wetlands
36.8
39.4
38.2
38.1
38.1
38.1
38.1
3.6%
Changes in Organic Soil Carbon
1
1
Stocks in Peatlands
1.1 i
1.11
0.7
0.6
0.6
0.5
0.6
-45.7%
Non-C02 Emissions from Peatlands
Remaining Peatlands
+
+
+
+
+
+
+
-47.0%
Changes in Biomass, DOM, and Soil
inns
I
iiiiiii
;;;;;;;
!!!!!E
Carbon Stocks in Coastal Wetlands
00
o
T—1
(10.1) 1
(11.1)
(11.1)
(11.1)
(11.1)
(11.1)
2.8%
CH4 Emissions from Coastal
I
1
Wetlands Remaining Coastal
Wetlands
4.2
4.2
4.3
4.3
4.3
4.3
4.3
3.4%
N20 Emissions from Coastal
1
Wetlands Remaining Coastal
I
1
Wetlands
IIIIIII
-------
d Estimates include CH4 and N20 emissions from drained organic soils on both forest land remaining forest land and land
converted to forest land. Carbon stock changes from drained organic soils are included with the forest land remaining forest
land forest ecosystem pools.
e Includes the net changes to carbon stocks stored in all forest ecosystem pools.
f Includes changes in mineral and organic soil carbon stocks for all land use conversions to cropland, grassland, and
settlements. Also includes aboveground/belowground biomass, dead wood, and litter carbon stock changes for conversion
of forest land to cropland, grassland, and settlements.
g Estimates include CH4 and N20 emissions from fires on both grassland remaining grassland and land converted to grassland.
h Estimates include N20 emissions from N fertilizer additions on both settlements remaining settlements and land converted
to settlements because it is not possible to separate the activity data at this time.
' LULUCF carbon stock change includes any carbon stock gains and losses from all land use and land-use conversion
categories.
' LULUCF emissions subtotal includes the CH4 and N20 emissions reported for peatlands remaining peatlands, forest fires,
drained organic soils, grassland fires, and coastal wetlands remaining coastal wetlands; CH4 emissions from flooded land
remaining flooded land, and land converted to flooded land, and land converted to coastal wetlands; and N20 emissions
from forest soils and settlement soils. Emissions values are included in land-use category rows.
k The LULUCF sector net total is the net sum of all LULUCF CH4 and N20 emissions to the atmosphere plus LULUCF net carbon
stock changes in units of MMT C02 Eq.
Notes: Totals may not sum due to independent rounding. Parentheses indicate net sequestration.
Overall CH4 and N2O emissions from LULUCF decreased by 7.3 percent from 2021 and increased by 16.5 percent
since 1990 while total net sequestration decreased by 10.9 percent since 1990 and 6.3 percent from 2021. Other
trends from 1990 to 2022 in emissions from LULUCF categories (Figure 2-11) over the 33-year period included the
following:
• Annual carbon sequestration by forest land (i.e., annual carbon stock accumulation in the five ecosystem
carbon pools and harvested wood products for forest land remaining forest land and land converted to
forest land) has decreased by 17.5 percent since 1990. This is primarily due to decreased carbon stock
gains in land converted to forest land and the harvested wood products pools within forest land
remaining forest land.
• Annual carbon sequestration from settlements remaining settlements (which includes organic soils,
settlement trees, and landfilled yard trimmings and food scraps) has increased by 21.2 percent over the
period from 1990 to 2022. This is primarily due to an increase in urbanized land area in the United States
with trees growing on it.
• Annual emissions from land converted to settlements increased by 19.1 percent from 1990 to 2022 due
primarily to carbon stock losses from forest land converted to settlements and mineral soils carbon stocks
from grassland converted to settlements.
Waste
Waste management and treatment activities are sources of CH4 and N2O emissions (see Figure 2-12 and Table 2-9).
Overall, emission sources accounted for in the Waste chapter generated 166.9 MMT CO2 Eq., or 2.6 percent of total
U.S. greenhouse gas emissions in 2022. In 2022, landfills were the largest source of waste emissions, accounting
for 71.8 percent of waste-related emissions. Landfills are also the third-largest source of U.S. anthropogenic CH4
emissions, generating 119.8 MMT CO2 Eq. and accounting for 17.1 percent of total U.S. CH4 emissions in 2022.8
Additionally, wastewater treatment generated emissions of 42.7 MMT CO2 Eq. and accounted for 25.6 percent of
waste emissions, 3.0 percent of U.S. CH4 emissions, and 5.6 percent of U.S. N2O emissions in 2022. Emissions of
CH4 and N2O from composting are also accounted for in this chapter, generating emissions of 2.6 MMT CO2 Eq. and
1.8 MMT CO2 Eq., accounting for 1.5 and 1.1 percent of Waste sector emissions, respectively. Anaerobic digestion
8 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 2-27
-------
at biogas facilities generated Cm emissions of less than 0.05 MMT CO2 Eq., accounting for less than 0.05 percent of
emissions from the Waste sector.
Figure 2-12: Trends in Waste Sector Greenhouse Gas Sources
250
200
O
O
150
100
.n Ch O CO
n fN N pj fN n °
rsj m ^
rsj o
CM
fN o
1 Anaerobic Digestion at Biogas Facilities
I Composting
I Wastewater Treatment
1 Landfills
o 8
r\l CTi CTi
rN ro
CO CO ^ i\
S 15 R R R 3 §
fP rv a»(r>ooooooooooT-H^HT-(Tiooooooooooooooooooooooo
HHHHHHHHHHfN(N(NNfNrMlN(N(N(NfMfNrNrM(NlNN(NfNI(N(N(N(N
Table 2-9: Emissions from Waste (MMT CO2 Eq.)
Gas/Source
1990
2005
2018
2019
2020
2021
2022
Percent
Change
Since
1990
ch4
220.9
172.4
150.2
152.4
147.6
145.3
143.2
(35.2%)
Landfills
197.8
147.7
126.3
128.7
124.1
122.0
119.8
(39.4%)
Wastewater T reatment
22.7
22.7
21.4
21.1
21.0
20.7
20.8
(8.4%)
Composting
0.4
2.1
2.5
2.5
2.6
2.6
2.6
504.8%
Anaerobic Digestion at
Biogas Facilities
+
+
+
+
+
+
+
1,109.3%
n2o
15.1
19.5
23.0
23.4
24.1
23.9
23.7
57.4%
Wastewater T reatment
14.8
18.1
21.2
21.6
22.3
22.1
21.9
48.2%
Composting
0.3
1.5
1.8
1.8
1.8
1.8
1.8
504.8%
Total
235.9
192.0
173.2
175.8
171.7
169.2
166.9
(29.3%)
+ Does not exceed 0.05 MMT C02 Eq.
Note: Totals may not sum due to independent rounding.
Waste sector emissions decreased by 1.4 percent since 2021 and 29.3 percent since 1990. Some notable trends in
U.S. emissions from Waste source categories (Figure 2-12) over the 33-year period from 1990 through 2022
included the following:
• Net Cm emissions from landfills decreased by 78.0 MMT CO2 Eq. (39.4 percent), with small increases
occurring in interim years. This downward trend in emissions coincided with increased landfill gas
2-28 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
collection and control systems, and a reduction of decomposable materials (i.e., paper and paperboard,
food scraps, and yard trimmings) discarded in municipal solid waste (MSW) landfills over the time series.
• Cm and N2O emissions from wastewater treatment decreased by 8.4 percent (1.9 MMT CO2 Eq.) and
increased by 48.2 percent (7.1 MMT CO2 Eq.), respectively. Methane emissions from domestic wastewater
treatment have decreased since 1999 due to decreasing percentages of wastewater being treated in
anaerobic systems, including reduced use of on-site septic systems and central anaerobic treatment
systems. N2O emissions from wastewater treatment processes gradually increased across the time series
as a result of increasing U.S. population and protein consumption.
• Combined CH4 and N2O emissions from composting have increased by 504.8 percent (3.7 MMT CO2 Eq.)
since 1990. The growth in composting since the 1990s is attributable to primarily four factors: (1) the
enactment of legislation by state and local governments that discouraged the disposal of yard trimmings
and food waste in landfills; (2) an increase in yard trimming collection and yard trimming drop-off sites
provided by local solid waste management districts; (3) an increased awareness of the environmental
benefits of composting; and (4) loans or grant programs to establish or expand composting infrastructure.
2.2 Emissions and Sinks by Economic
Sector
In addition to the Paris Agreement and UNFCCC reporting sectors and methods defined by the IPCC, this report
also characterizes gross emissions according to commonly used economic sector categories: residential,
commercial, industry, transportation, electric power, and agriculture. Emissions from U.S. Territories are reported
as their own end-use sector due to a lack of specific consumption data for the individual end-use sectors within
U.S. Territories. See Box 2-1 for more information on how economic sectors are defined. For more information on
trends in the LULUCF sector, see Section 2.1.
Using this categorization, transportation activities accounted for the largest portion (28.4 percent) of total U.S.
greenhouse gas emissions in 2022. Emissions from electric power accounted for the second largest portion (24.9
percent), while emissions from industry accounted for the third-largest portion (22.9 percent) of total U.S.
greenhouse gas emissions in 2022. Emissions from industry have in general declined over the past decade due to a
number of factors, including structural changes in the U.S. economy (i.e., shifts from a manufacturing-based to a
service-based economy), fuel switching, and efficiency improvements.
The remaining 23.8 percent of U.S. greenhouse gas emissions were contributed by, in order of magnitude, the
agriculture, commercial, and residential sectors, plus emissions from U.S. Territories. Activities related to
agriculture accounted for 10.0 percent of emissions; unlike other economic sectors, agricultural sector emissions
were dominated by N2O emissions from agricultural soil management and CH4 emissions from enteric
fermentation, rather than CO2 from fossil fuel combustion. An increasing amount of carbon is stored in agricultural
soils each year, but this carbon sequestration is assigned to the LULUCF sector rather than the agriculture
economic sector. The commercial and residential sectors accounted for roughly 7.3 percent and 6.2 percent of
greenhouse gas emissions, respectively, and U.S. Territories accounted for 0.4 percent of emissions; emissions
from these sectors primarily consisted of CO2 emissions from fossil fuel combustion. Carbon dioxide was also
emitted and sequestered (in the form of carbon) by a variety of activities related to forest management practices,
tree planting in urban areas, the management of agricultural soils, landfilling of yard trimmings, and changes in
carbon stocks in coastal wetlands. Table 2-10 presents a detailed breakdown of emissions from each of these
economic sectors by source category, as they are defined in this report. Figure 2-13 shows the trend in emissions
by sector from 1990 to 2022.
Trends 2-29
-------
Figure 2-13:
U.S. Greenhouse Gas Emissions Allocated to Economic Sectors
Note: Emissions and removals from Land Use, Land-Use Change, and Forestry are excluded from figure above. Excludes U.S.
Territories.
Table 2-10: U.S. Greenhouse Gas Emissions Allocated to Economic Sectors (MMT CO2 Eq. and
Percent of Total in 2022)
Percent of
Total
Sector/Source
1990
2005
2018
2019
2020
2021
2022
Emissions3
Transportation
1,521.4
1,965.9
1,871.6
1,874.6
1,625.3
1,805.5
1,801.5
28.4%
C02from Fossil Fuel
Combustion
1,468.9
1,858.6
1,813.1
1,816.6
1,572.8
1,753.5
1,751.3
27.6%
Substitution of Ozone
Depleting Substances
+
63.1
35.5
34.0
32.5
31.2
29.6
0.5%
Mobile Combustion15
40.6
34.1
13.7
15.1
12.2
12.7
12.2
0.2%
Non-Energy Use of Fuels
11.8
10.2
9.2
8.8
7.8
8.0
8.4
0.1%
Electric Power Industry
1,880.2
2,457.4
1,799.2
1,650.8
1,482.2
1,584.4
1,577.5
24.9%
C02 from Fossil Fuel
Combustion
1,820.0
2,400.1
1,753.4
1,606.7
1,439.6
1,540.9
1,531.7
24.1%
Stationary Combustion15
18.7
27.7
23.1
20.2
18.9
20.4
22.9
0.4%
Incineration of Waste
13.3
13.6
13.7
13.3
13.3
12.8
12.7
0.2%
Other Process Uses of
Carbonates
3.6
4.2
4.0
4.5
4.5
4.3
5.2
0.1%
Electrical Equipment
24.7
11.9
5.0
6.1
5.9
6.0
5.1
0.1%
Industry
1,723.3
1,587.3
1,541.9
1,531.8
1,435.9
1,455.8
1,452.5
22.9%
C02 from Fossil Fuel
Combustion
833.1
796.8
770.7
770.1
722.9
740.7
761.8
12.0%
Natural Gas Systems
251.2
236.5
223.0
227.3
217.0
210.4
209.7
3.3%
Non-Energy Use of Fuels
83.9
107.2
108.9
97.4
89.9
103.5
94.3
1.5%
Petroleum Systems
59.0
58.5
93.8
97.8
82.3
72.8
61.6
1.0%
Coal Mining
112.7
75.6
62.2
56.0
48.3
47.1
46.1
0.7%
Cement Production
33.5
46.2
39.0
40.9
40.7
41.3
41.9
0.7%
Iron and Steel
Production
104.8
70.1
42.9
43.1
37.7
41.9
40.7
0.6%
Substitution of Ozone
Depleting Substances
+
8.0
31.9
33.1
33.9
32.2
33.4
0.5%
2-30 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
Petrochemical
I
1
Production
20.1
26.9
27.2
28.5
27.9
30.7
28.8
0.5%
Landfills (Industrial)
12.2
16.1
18.7
18.8
18.9
18.9
18.9
0.3%
Ammonia Production
14.4
10.2
12.7
12.4
13.0
12.2
12.6
0.2%
Lime Production
11.7
14.6 !
13.1
12.1
11.3
11.9
12.2
0.2%
Nitric Acid Production
10.8
10.1
8.5
8.9
8.3
7.9
8.6
0.1%
Abandoned Oil and Gas
1
1
Wells
7.8 jjjjjj
8.2
8.4
8.5
8.5
8.6
8.5
0.1%
Fluorochemical
Production
70.9
30.0
8.7
9.3
6.9
7.1
7.8
+%
Wastewater T reatment
cn
iiiii
7.1
7.5
7.6
7.6
7.6
7.7
+%
Urea Consumption for
Non-Agricultural
1
I
Purposes
3.8
3.7
6.1
6.2
5.8
6.6
7.1
+%
Abandoned
!!!!!!:
I
Underground Coal
Illlll'
!!!!!!:
iiiiiii
1
Mines
8-i::
7.4
6.9
6.6
6.5
6.3
6.3
+%
Mobile Combustion15
3.6
5.6
5.5
5.6
5.3
5.5
5.9
+%
Other Process Uses of
mm:
I
mm:
Carbonates
3.6
4-2 I
4.0
4.5
4.5
4.3
5.2
+%
Carbon Dioxide
Consumption
1.5
1.4
4.1
4.9
5.0
5.0
5.0
+%
Electronics Industry
3.3
4^
Ln
111111
4.8
4.5
4.5
4.8
4.7
+%
N20 from Product Uses
3.8
3.8
3.8
3.8
3.8
3.8
3.8
+%
Stationary Combustion15
¦ps.
CO
MMM
4.5 i
¦
3.9
3.8
3.6
3.6
3.5
+%
Aluminum Production
26.1
7.2
2.9
3.3
3.2
2.5
2.2
+%
AdipicAcid Production
13.5
6-31:
9.3
4.7
7.4
6.6
2.1
+%
Glass Production
2.3
2.4
2.0
1.9
1.9
2.0
2.0
+%
Soda Ash Production
1 4 i
1-7 I
1.7
1.8
1.5
1.7
1.7
+%
Titanium Dioxide
Production
1.2
1.8
1.5
1.3
1.3
1.5
1.5
+%
Ferroalloy Production
2.21
mm:
1..4 =
2.1
1.6
1.4
1.6
1.3
+%
Caprolactam, Glyoxal,
and Glyoxylic Acid
Production
1.5
1.9
1.3
1.2
1.1
1.2
1.3
+%
Magnesium Production
!!!!!!!
:mm
lllllli
and Processing
5-7 HI1
3.° :
1.1
1.0
0.9
1.2
1.2
+%
Zinc Production
0.6
1.0
1.0
1.0
1.0
1.0
0.9
+%
Phosphoric Acid
III
mm:
Production
1.5
!-3 \
0.9
0.9
0.9
0.9
0.8
+%
SF6 and PFCs from Other
Product Use
1.4
1.5
0.9
0.8
0.7
0.5
0.8
+%
Lead Production
o-5:
0.6
0.5
0.5
0.5
0.4
0.4
+%
Carbide Production and
Consumption
0.3
0.2
0.2
0.2
0.2
0.2
0.2
+%
Agriculture
595.9
634.3
683.5
661.0
640.0
645.9
634.0
10.0%
N20 from Agricultural
Soil Management
288.8
294.1
333.4
315.6
292.1
298.0
290.8
4.6%
Enteric Fermentation
183-1 i
188.2 =
196.8
197.3
196.3
196.5
192.6
3.0%
Manure Management
52.5
70.2
84.3
83.5
83.8
83.6
81.7
1.3%
CO2 from Fossil Fuel
I
1
Combustion
43.4 1
50.8 1
39.8
39.7
39.1
39.8
39.3
0.6%
Rice Cultivation
18.9
20.6
19.9
15.6
18.6
18.3
18.9
0.3%
Urea Fertilization
2.4
3.5 1
4.9
5.0
5.1
5.2
5.3
0.1%
Liming
4.7
4.4
2.2
2.2
2.9
2.4
3.3
0.1%
Trends 2-31
-------
Mobile Combustion15
1.4
1.6
1.2
1.2
1.2
1.2
1.2
+%
Field Burning of
Agricultural Residues
0.7
0.8
0.8
0.9
0.8
0.8
0.8
+%
Stationary Combustion15
0.1
+
0.1
0.1
0.1
0.1
0.1
+%
Commercial
447.0
418.9
453.5
462.6
436.9
443.7
463.7
7.3%
C02 from Fossil Fuel
Combustion
228.3
227.1
246.3
251.7
229.3
237.5
258.7
4.1%
Landfills (Municipal)
185.5
131.6
107.7
109.9
105.2
103.1
100.9
1.6%
Substitution of Ozone
Depleting Substances
+
21.4
58.5
59.8
60.8
61.9
62.9
1.0%
Wastewater T reatment
30.9
33.6
35.0
35.1
35.6
35.1
35.0
0.6%
Composting
0.7
3.6
4.3
4.3
4.4
4.4
4.4
0.1%
Stationary Combustion15
1.5
1.5
1.7
1.7
1.6
1.6
1.7
0.0%
Anaerobic Digestion at
Biogas Facilities
+
+
+
+
+
+%
Residential
345.6
371.2
376.8
384.2
358.0
369.6
391.3
6.2%
C02 from Fossil Fuel
Combustion
338.6
358.9
338.9
342.9
314.8
318.0
334.1
5.3%
Substitution of Ozone
Depleting Substances
0.2
7.0
31.9
35.1
39.0
47.3
52.2
0.8%
Stationary Combustion15
6.8
5.3
6.0
6.2
4.2
4.2
5.0
0.1%
U.S. Territories
23.4
59.7
26.3
25.1
23.4
23.9
22.7
0.4%
C02 from Fossil Fuel
Combustion
20.0
51.9
25.9
24.8
23.3
23.8
22.6
0.4%
Non-Energy Use of Fuels
3.4
7.6
0.2
0.2
0.1
0.1
0.1
+%
Stationary Combustion15
0.1
0.2
0.1
0.1
0.1
0.1
0.1
+%
Total Gross Emissions
(Sources)
6,536.9
7,494.6
6,752.7
6,590.1
6,001.8
6,328.8
6,343.2
100.0%
LULUCF Sector Net Total0
(976.7)
(907.7)
(915.5)
(863.6)
(904.4)
(910.6)
(854.2)
(13.5%)
Net Emissions (Sources
and Sinks)
5,560.2
6,586.9
5,837.3
5,726.6
5,097.4
5,418.2
5,489.0
86.5%
+ Does not exceed 0.05 MMT C02 Eq. or 0.05 percent.
a Percent of total (gross) emissions excluding emissions from LULUCF for 2022.
b Includes CH4 and N20 emissions from fuel combustion.
c The LULUCF sector net total is the net sum of all LULUCF CH4 and N20 emissions to the atmosphere plus LULUCF net carbon
stock changes.
Notes: Total gross emissions presented are without LULUCF. Total net emissions are presented with LULUCF. Totals may not
sum due to independent rounding. Parentheses indicate negative values or sequestration.
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 sectors common for reporting under the Paris Agreement and the
UNFCCC. Discussing greenhouse gas emissions relevant to U.S.-specific economic sectors improves
communication of the report's findings.
The electric power economic sector includes CO2, Cm and N2O emissions from the combustion of fossil fuels that
are included in the EIA electric power sector. Carbon dioxide, CH4, and N2O emissions from waste incineration
are included in the electric power economic sector, as the majority of MSW is combusted in plants that produce
electricity. The electric power economic sector also includes SF6 from electrical equipment, and a portion of CO2
from other process uses of carbonates (from pollution control equipment installed in electric power plants).
The transportation economic sector includes CO2 emissions from the combustion of fossil fuels that are included
in the EIA transportation fuel-consuming sector. (Additional analyses and refinement of the EIA data are further
explained in the Energy chapter of this report.) Emissions of CH4 and N2O from mobile combustion are also
2-32 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
apportioned to the transportation economic sector based on the EIA transportation fuel-consuming sector.
Emissions of ODS substitutes are apportioned to the transportation economic sector based on emissions from
refrigerated transport and motor vehicle air-conditioning systems. Finally, CO2 emissions from non-energy uses
of fossil fuels identified as lubricants for transportation vehicles are included in the transportation economic
sector.
The industry economic sector includes CO2 emissions from the combustion of fossil fuels that are included in the
EIA industrial fuel-consuming sector, minus the agricultural use of fuel explained below. The CH4 and N2O
emissions from stationary and mobile combustion are also apportioned to the industry economic sector based
on the EIA industrial fuel-consuming sector, minus emissions apportioned to the agriculture economic sector.
Emissions of ODS substitutes are apportioned based on their specific end-uses within the source category, with
most emissions falling within the industry economic sector. Finally, CFU emissions from industrial landfills and
Cm and N2O from industrial wastewater treatment are included in the industry economic sector.
Additionally, all process-related emissions from sources with methods considered within the IPCC IPPU sector
are apportioned to the industry economic sector. This includes the process-related emissions (i.e., emissions
from the actual process to make the material, not from fuels to power the plant) from activities such as cement
production, iron and steel production and metallurgical coke production, and ammonia production.
Additionally, fugitive emissions from energy production sources, such as natural gas systems, coal mining, and
petroleum systems are included in the industry economic sector. A portion of CO2 from other process uses of
carbonates (from pollution control equipment installed in large industrial facilities) is also included in the
industry economic sector. Finally, all remaining CO2 emissions from non-energy uses of fossil fuels are assumed
to be industrial in nature (besides the lubricants for transportation vehicles specified above) and are attributed
to the industry economic sector.
The agriculture economic sector includes CO2 emissions from the combustion of fossil fuels that are based on
supplementary sources of agriculture fuel use data, because EIA includes agriculture equipment in the industrial
fuel-consuming sector. Agriculture fuel use estimates are obtained from U.S. Department of Agriculture survey
data, in combination with EIA Fuel Oil and Kerosene Sales (FOKS) data (EIA 1991 through 2022). Agricultural
operations are based on annual energy expense data from the Agricultural Resource Management Survey
(ARMS) conducted by the National Agricultural Statistics Service (NASS) of the USDA. NASS collects information
on farm production expenditures including expenditures on diesel fuel, gasoline, LP gas, natural gas, and
electricity use on the farm with the annual ARMS. A USDA publication (USDA/NASS 2023) shows national totals,
as well as selected States and ARMS production regions. These supplementary data are subtracted from the
industrial fuel use reported by EIA to obtain agriculture fuel use. Carbon dioxide emissions from fossil fuel
combustion, and CH4 and N2O emissions from stationary and mobile combustion, are then apportioned to the
agriculture economic sector based on agricultural fuel use.
The other IPCC Agriculture emission source categories apportioned to the agriculture economic sector include
N2O emissions from agricultural soils, CH4 from enteric fermentation, CH4 and N2O from manure management,
CH4 from rice cultivation, CO2 emissions from liming and urea application, and CH4 and N2O from field burning of
agricultural residues.
The residential economic sector includes CO2 emissions from the combustion of fossil fuels that are included in
the EIA residential fuel-consuming sector. Stationary combustion emissions of CH4 and N2O are also based on
the EIA residential fuel-consuming sector. Emissions of ODS substitutes are apportioned to the residential
economic sector based on emissions from residential air-conditioning systems. N2O emissions from the
application of fertilizers to developed land (termed "settlements" by the IPCC) are also included in the
residential economic sector.
The commercial economic sector includes CO2 emissions from the combustion of fossil fuels that are included in
the EIA commercial fuel-consuming sector. Emissions of CH4 and N2O from mobile combustion are also
apportioned to the commercial economic sector based on the EIA commercial fuel-consuming sector. Emissions
of ODS substitutes are apportioned to the commercial economic sector based on emissions from commercial
refrigeration/air-conditioning systems. Public works sources, including direct CH4 from municipal landfills, CH4
Trends 2-33
-------
from anaerobic digestion at biogas facilities, CH and N O from domestic wastewater treatment, and
composting, are also included in the commercial economic sector.
Emissions with Electricity Distributed to Economic Sectors
It is also useful to view greenhouse gas emissions from economic sectors with emissions related to electric power
distributed into end-use categories (i.e., emissions from the electric power sector are allocated to the economic
end-use sectors in which the electricity is used). For example, greenhouse gas emissions from some economic
sectors, i.e., commercial and residential and industry, increase substantially when indirect emissions from
electricity end-use are included, due to the relatively large share of electricity use by buildings (75 percent of the
electricity generated in the United States for heating, ventilation, and air conditioning; lighting; and appliances,
etc.)9 and use of electricity for powering industrial machinery.
The generation, transmission, and distribution of electricity directly accounted for 24.9 percent of total U.S.
greenhouse gas emissions in 2022. Electric power-related emissions decreased by 16.1 percent since 1990 mainly
due to fuel switching in the electric power sector. From 2021 to 2022, electric power-related emissions decreased
by 0.4 percent. Between 2021 and 2022, the consumption of natural gas and petroleum for electric power
generation increased by 7.6 percent and 18.9 percent, respectively, while the consumption of coal decreased by
6.4 percent. Electric power-related emissions are still lower than pre-pandemic 2019 levels. Table 2-11 provides a
detailed summary of emissions from electric power-related activities.
From 2021 to 2022, electricity sales to the residential end-use sector increased by 2.6 percent. Electricity sales to
the commercial end-use and industrial sectors increased by 4.7 percent and 2.0 percent, respectively. Overall,
from 2021 to 2022, the amount of electricity retail sales (in kWh) increased by 3.2 percent.
Table 2-11: Electric Power-Related Greenhouse Gas Emissions (MMT CO2 Eq.)
Gas/Fuel Type or Source
1990
2005
2018
2019
2020
2021
2022
C02
1,836.4
2,417.5
1,770.7
1,624.2
1,457.0
1,557.7
1,549.2
Fossil Fuel Combustion
1,820.0
1
I
2,400.1
1,753.4
1,606.7
1,439.6
1,540.9
1,531.7
Coal
1,546.5
1,982.8
1,152.9
973.5
788.2
910.1
851.5
Natural Gas
175.4
1
I
318.9
577.9
616.6
634.8
612.8
659.3
Petroleum
97.5
98.0
22.2
16.2
16.2
17.7
20.5
Geothermal
0.5
I
1
0-5 :
0.4
0.4
0.4
0.4
0.4
Incineration of Waste
12.9
13.3
13.3
12.9
12.9
12.5
12.4
Other Process Uses of Carbonates
3.6
|
4.2 I
4.0
4.5
4.5
4.3
5.2
ch4
0.5
1.0
1.4
1.4
1.4
1.4
1.3
Stationary Sources3
0.5
1
1.0
1.4
1.4
1.4
1.4
1.3
Incineration of Waste
+
+
+
+
+
+
+
l\l20
18.6
1
1
27.1
22.1
19.1
17.9
19.4
21.9
Stationary Sources3
18.2
26.7
21.7
18.8
17.5
19.0
21.6
Incineration of Waste
0.4
I
0.3
0.4
0.4
0.3
0.4
0.3
sf6
24.7
11.8
5.0
6.1
5.9
6.0
5.1
Electrical Equipment
24.7
1
11-8 1
5.0
6.1
5.9
6.0
5.1
cf4
+
+
+
+
+
+
+
Electrical Equipment
+
+
+
+
+
+
+
Total
1,880.2
2,457.4
1,799.2
1,650.8
1,482.2
1,584.4
1,577.5
+ Does not exceed 0.05 MMT C02 Eq.
9 See https://www.nrel.gov/news/features/2023/nrel-researcher-reveal-how-buildines-across-the-united-states-do-and-could-
use-energv.html#:~:text=Buildings%20are%20responsible%20for%2Q4Q,buildings%20stock%20is%20also%20essential.
2-34 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
a Includes only stationary combustion emissions related to the generation of electricity.
Note: Totals may not sum due to independent rounding.
To distribute electricity emissions among economic end-use sectors, emissions from the source categories
assigned to the electric power sector were allocated to the residential, commercial, industry, transportation, and
agriculture economic sectors according to each economic sector's share of retail sales of electricity (EIA 2020;
USDA/NASS 2023). These source categories include CO2 from fossil fuel combustion, CH4 and N2O from stationary
combustion, incineration of waste, other process uses of carbonates, and SF6 from electrical equipment. Note that
only 50 percent of the emissions from other process uses of carbonates were associated with electric power and
distributed as described; the remaining emissions from other process uses of carbonates were attributed to the
industry economic end-use sector.10
When emissions from electricity use are distributed among these economic end-use sectors, 2022 emissions from
industrial activities account for the largest share of total U.S. greenhouse gas emissions (29.5 percent), followed
closely by emissions from transportation (28.5 percent). Emissions from the commercial and residential sectors
also increase substantially when emissions from electricity are included (15.8 and 15.3 percent, respectively). In all
economic end-use sectors except agriculture, CO2 accounts for more than 78 percent of greenhouse gas emissions,
primarily from the combustion of fossil fuels. Table 2-12 presents a detailed breakdown of emissions from each of
these economic sectors, with emissions from electric power distributed to them. Figure 2-14 shows the trend in
these emissions by sector from 1990 to 2022.
Figure 2-14: U.S. Greenhouse Gas Emissions with Electricity-Related Emissions Distributed to
Economic Sectors
Note: Emissions and removals from Land Use, Land-Use Change, and Forestry are excluded from figure above. Excludes U.S.
Territories.
Table 2-12: U.S. Greenhouse Gas Emissions by Economic Sector and Gas with Electricity-
Related Emissions Distributed (MMT CO2 Eq.) and Percent of Total in 2022
Emissions by Gas
1990 20051 2018
2019
2020
2021
2022
Percent3
Industry
2,397.3 2,302.91 2,017.1
1,974.8
1,823.5
1,877.8
1,872.9
29.5%
Direct Emissions
1,723.3 1,587.31 1,541.9
1,531.8
1,435.9
1,455.8
1,452.5
22.9%
10 Emissions were not distributed to U.S. Territories, since the electric power sector only includes emissions related to the
generation of electricity in the 50 states and the District of Columbia.
Trends 2-35
-------
co2
1,173.8
1,136.4
1,110.7
1,118.3
1,036.9
1,071.1
1,080.8
17.0%
ch4
413.3
371.2
352.2
337.7
323.3
311.3
299.8
4.7%
n2o
35.7
29.8
30.3
26.0
27.6
26.8
23.6
0.4%
HFCs, PFCs, SF6 and NF3
1°°.6 g
49.8
48.6
49.8
48.1
46.5
48.3
0.8%
Electricity-Related
674.0
715.6
475.3
443.0
387.6
422.0
420.3
6.6%
C02
658.3
704.0
467.7
435.9
381.0
414.9
412.8
6.5%
ch4
0.2
0.3
0.4
0.4
0.4
0.4
0.3
+%
n2o
6-7 1
7-9
5.8
5.1
4.7
5.2
5.8
0.1%
sf6
8.8
3.5
1.3
1.6
1.5
1.6
1.4
+%
Transportation
1,524.6
1,970.8
1,876.5
1,879.5
1,629.5
1,810.6
1,807.8
28.5%
Direct Emissions
1,521.4
1,965.9
1,871.6
1,874.6
1,625.3
1,805.5
1,801.5
28.4%
C02
1,480.8 :
1,868.7 |
1,822.3
1,825.5
1,580.6
1,761.6
1,759.7
27.7%
ch4
6.4
3.1
1.7
1.7
1.4
1.5
1.5
+%
n2o
34.3 ;
31.0 1:
12.1
13.4
10.7
11.2
10.7
0.2%
HFCsb
+
63.1
35.5
34.0
32.5
31.2
29.6
0.5%
Electricity-Related
3.1
4.8
4.9
4.9
4.2
5.2
6.3
0.1%
C02
3.1
4.8
4.8
4.9
4.1
5.1
6.2
0.1%
ch4
+ IIIII:
+ ""
+
+
+
+
+
+%
n2o
+
0.1
0.1
0.1
0.1
0.1
0.1
+%
sf6
+
+
+
+
+
+%
Residential
958.0
1,247.7
1,035.9
984.0
919.5
958.0
973.5
15.3%
Direct Emissions
345.6
371.2
376.8
384.2
358.0
369.6
391.3
6.2%
C02
338.6
358.9
338.9
342.9
314.8
318.0
334.1
5.3%
ch4
5.9
4-5 1
5.1
5.3
3.6
3.6
4.3
0.1%
n2o
0.9
0.8
0.8
0.8
0.6
0.6
0.7
+%
sf6
0.2
7.0 I
31.9
35.1
39.0
47.3
52.2
0.8%
Electricity-Related
612.4
876.5
659.1
599.7
561.5
588.4
582.2
9.2%
C02
598.1 1
862.2 1
648.6
590.1
551.9
578.5
571.8
9.0%
ch4
0.2
0.3
0.5
0.5
0.5
0.5
0.5
+%
n2o
6.1!!!;
9.7 "
8.1
6.9
6.8
7.2
8.1
0.1%
sf6
8.0
4.2
1.8
2.2
2.2
2.2
1.9
+%
Commercial
1,002.5
1,241.1
1,074.3
1,030.5
931.5
976.8
1,002.6
15.8%
Direct Emissions
447.0
418.9
453.5
462.6
436.9
443.7
463.7
7.3%
C02
228.3
227.1 1
246.3
251.7
229.3
237.5
258.7
4.1%
ch4
203.6
150.9
125.9
127.8
122.9
120.5
118.5
1.9%
n2o
15.1
19.4
22.8
23.2
23.9
23.7
23.6
0.4%
HFCs
+
21.4
58.5
59.8
60.8
61.9
62.9
1.0%
Electricity-Related
555.5
822.2
620.8
567.8
494.6
533.2
539.0
8.5%
C02
542.6
808.9
611.0
558.7
486.2
524.2
529.3
8.3%
ch4
o-i 1
0.3 i
0.5
0.5
0.5
0.5
0.4
+%
n2o
5.5
9.1
7.6
6.6
6.0
6.5
7.5
0.1%
sf6
7.3
4.0
¦
1.7
2.1
2.0
2.0
1.7
+%
Agriculture
631.1
672.6
722.7
696.3
674.4
681.6
663.6
10.5%
Direct Emissions
595.9
634.3
683.5
661.0
640.0
645.9
634.0
10.0%
C02
50.5
58.7
47.0
46.9
47.1
47.4
47.9
0.8%
ch4
241.9 ;
264.6
285.2
280.4
282.6
282.0
277.0
4.4%
n2o
303.5
311.0
351.3
333.7
310.3
316.4
309.1
4.9%
Electricity-Related
35.2
38.3
39.2
35.2
34.4
35.7
29.7
0.5%
C02
34.3
37.7
38.6
34.7
33.8
35.1
29.2
0.5%
ch4
+
+
+
+
+
+
+
+%
n2o
0.3
0.4
0.5
0.4
0.4
0.4
0.4
+%
sf6
0.5
0.2
0.1
0.1
0.1
0.1
0.1
+%
U.S. Territories
23.4
59.7
26.3
25.1
23.4
23.9
22.7
0.4%
Total Gross Emissions
(Sources)
6,536.9
7,494.6
6,752.7
6,590.1
6,001.8
6,328.8
6,343.2
100.0%
LULUCF Sector Net Totalc
(976.7)
(907.7)
(915.5)
(863.6)
(904.4)
(910.6)
(854.2)
(13.5%)
Net Emissions (Sources
and Sinks)
5,560.2
6,586.9
5,837.3
5,726.6
5,097.4
5,418.2
5,489.0
86.5%
2-36 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
+ Does not exceed 0.05 MMT C02 Eq. or 0.05 percent.
a Percent of total (gross) emissions excluding emissions from LULUCF for the year 2022.
b Includes primarily HFC-134a.
c The LULUCF sector net total is the net sum of all LULUCF CH4 and N20 emissions to the atmosphere plus LULUCF net carbon
stock changes.
Notes: Total gross emissions are presented without LULUCF. Net emissions are presented with LULUCF. Emissions from
electric power are allocated based on aggregate electricity use in each end-use sector. Totals may not sum due to
independent rounding.
Industry
The industry economic sector includes CO2 emissions from fossil fuel combustion from all manufacturing facilities,
in aggregate, and with the distribution of electricity-related emissions (e.g., powering industrial machinery),
accounted for 29.5 percent of U.S. greenhouse gas emissions in 2022. This end-use sector also includes emissions
that are produced as a byproduct of the non-energy-related industrial process activities. The variety of activities
producing these non-energy-related emissions includes CH4 emissions from petroleum and natural gas systems,
fugitive Cm and CO2 emissions from coal mining, byproduct CO2 emissions from cement production, and HFC, PFC,
SFs, and NF3 byproduct emissions from the electronics industry, to name a few.
Since 1990, industry sector emissions have declined by 21.9 percent. The decline has occurred both in direct
emissions and indirect emissions associated with electricity use. Structural changes within the U.S. economy that
led to shifts in industrial output away from energy-intensive manufacturing products to less energy-intensive
products (e.g., shifts from producing steel to computer equipment) have had a significant effect on industrial
emissions.
Transportation
When electricity-related emissions are distributed to economic end-use sectors, transportation activities
accounted for 28.5 percent of U.S. greenhouse gas emissions in 2022. The largest sources of transportation
greenhouse gas emissions in 2022 were light-duty trucks, which include sport utility vehicles, pickup trucks, and
minivans (36.5 percent); medium- and heavy-duty trucks (22.9 percent); passenger cars (20.4 percent); commercial
aircraft (7.2 percent); pipelines (3.8 percent); ships and boats (2.8 percent); other aircraft (2.0 percent), and rail
(2.0 percent). These figures include direct CO2, CH4, and N2O emissions from fossil fuel combustion used in
transportation, indirect emissions from electricity use, and emissions from non-energy use (i.e., lubricants) used in
transportation, as well as HFC emissions from mobile air conditioners and refrigerated transport allocated to these
vehicle types.
From 1990 to 2022, total transportation emissions from fossil fuel combustion increased by 19.4 percent due, in
large part, to increased demand for travel. From 2021 to 2022, emissions decreased by 0.1 percent. Vehicle miles
traveled (VMT) by light-duty motor vehicles (passenger cars and light-duty trucks) increased by 46.7 percent from
1990 to 2022 as a result of a confluence of factors including population growth, economic growth, urban sprawl,
and periods of low fuel prices. The primary driver of transportation-related emissions was CO2 from fossil fuel
combustion, which increased by 19.4 percent from 1990 to 2022 when including electricity. This rise in CO2
emissions, combined with an increase in HFCs from close to zero emissions in 1990 to 29.6 MMT CO2 Eq. in 2022,
led to an increase in overall greenhouse gas emissions from transportation activities of 18.6 percent.
The decline in new light-duty vehicle fuel economy between 1990 and 2004 reflected the increasing market share
of light-duty trucks, which grew from approximately 29.6 percent of new vehicle sales in 1990 to 48.0 percent in
2004. Starting in 2005, average new vehicle fuel economy began to increase while light-duty VMT grew only
modestly between 2005 and 2013. Light-duty VMT grew by less than one percent or declined each year between
2005 and 2013, then grew at a faster rate until 2016 (2.6 percent from 2014 to 2015, and 2.5 percent from 2015 to
2016). Since 2016, the rate of light-duty VMT growth has slowed to one percent or less each year. Average new
vehicle fuel economy has increased almost every year since 2005, while light-duty truck market share decreased to
Trends 2-37
-------
33.0 percent in 2009 and has since varied from year to year between 35.6 and 63.1 percent. Light-duty truck
market share was about 63.1 percent of new vehicles in model year 2022 (EPA 2023b).
Table 2-13 provides a detailed summary of greenhouse gas emissions from transportation-related activities with
electricity-related emissions included in the totals. Historically, the majority of electricity use in the transportation
sector was for rail transport. However, more recently there has been increased electricity use in on-road electric
and plug-in hybrid vehicles. Despite this increase, almost all of the energy used for transportation was supplied by
petroleum-based products, with more than half being related to gasoline consumption in automobiles and other
highway vehicles. Other fuel uses, especially diesel fuel for freight trucks and jet fuel for aircraft, accounted for the
remainder. Indirect emissions from electricity are less than 1 percent of direct emissions in the transportation
sector. For a more detailed breakout of emissions by fuel type by vehicle see Table A-93 in Annex 3.
Figure 2-15: Trends in Transportation-Related Greenhouse Gas Emissions
Lubricants ¦ Ships and Boats
Motorcycles ¦ Aircraft
Buses ¦ Medium- and Heavy-Duty Trucks
Pipelines Light-Duty Trucks
Rail ¦ Passenger Cars
CNi cp
LTl
N"
^ m
m ld
LT)
rsj
— vo
CN] KD
& -
CO
,sS
¦r rv ¦-!
rv ^ J" _
o 01 °
_J , T-I CM
CO S N ^ ^
JN cr> cn °l °l
_r r *-« h
CO
CO
vd ...
°° s
Is
CO h CO
3^ Ln
rv rv
— 01
3: o-i 3:
CO CO
KD CTi
co co
^ CO
y-i o
CO CO
-----
1,000
800
600
400
200
0
o>-irsiro^rLniDr-vOOCT\o^-irsiro^rLnu3r~ooCTiOT-irNinrrLnuDrvooc7iOT-irN
oicnavCTi-i'-H—iT-i^H^H^HT-HvHrNrNirN
CTicricriCTicricncnaicricriooooooooooooooooooooooo
HHHHHHHHHHtNrJNNINrJININnJnJfMOJNNnJfNtMINNfNNNPJ
Table 2-13: Transportation-Related Greenhouse Gas Emissions (MMT CO2 Eq.)
Gas/Vehicle Type
1990
2005
2018
2019
2020
2021
2022
Passenger Cars
648.4
564.4
398.7
395.5
341,7
374.2
369.5
C02
622.2
521.1
386.5
384.2
331.9
365.0
361.0
ch4
3.8
1.2
0.3
0.3
0.3
0.3
0.2
n2o
22.5
13.3
2.5
2.6
2.0
1.9
1.7
HFCs
0.0
28.8
9.4
8.4
7.6
7.0
6.6
Light-Duty Trucks
302.4
659.3
720.6
711.7
615.3
671.7
660.2
C02
292.1
614.0
699.0
690.1
596.2
654.0
644.5
ch4
1.5
1.0
0.6
0,6
0.5
0.5
0.5
n2o
8.7
14.0
4.6
5.6
4.4
4.2
3.8
HFCs
0.0
30.2
16.4
15.4
14.2
13.0
11.4
2-38 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
Medium- and Heavy-Duty Trucks
234.5 1
391.0
406.5
409.3
386.7
417.0
413.1
C02
233.1 -
386.2 *
397.9
400.3
377.8
407.7
403.6
ch4
0.5
0.2
0.1
0.1
0.1
0.1
0.1
n2o
1.01
1.5:
2.8
3.0
2.7
3.0
3.1
HFCs
0.0
3.2
5.6
5.8
6.1
6.3
6.3
Buses
13.3
17.8
24.9
25.3
24.0
26.1
26.3
C02
13.2
17.5
24.3
24.7
23.4
25.6
25.7
ch4
+ 1
+ 1
+
+
+
+
+
n2o
0.1
0.1
0.2
0.2
0.2
0.2
0.2
HFCs
o.o 1
0.2
0.4
0.4
0.4
0.4
0.4
Motorcycles
3.4
5.0
7.4
7.5
6.7
7.5
7.6
C02
4.9
7.3
7.4
6.6
7.4
7.4
ch4
+
+
+
+
+
+
+
n2o
¦
+
+
0.1
0.1
0.1
0.1
0.1
Commercial Aircraft3
110.8
133.8
130.7
137.8
92.0
120.0
130.8
C02
109-9 1
132.7
129.6
136.7
91.3
119.0
129.7
ch4
0.0
0.0
0.0
0.0
0.0
0.0
0.0
n2o
0.9 i
iiiiiii
1.1
1.1
1.1
0.7
1.0
1.1
Other Aircraftb
78.0
59.5
44.6
45.6
31.0
35.5
37.0
C02
77.3
59-0 :
44.2
45.2
30.7
35.1
36.7
ch4
0.1
0.1
+
+
+
+
+
n2o
0.6 1
0.5
0.4
0.4
0.2
0.3
0.3
Ships and Boatsc
47.0
45.5
41.1
40.0
32.2
50.7
49.9
C02
46.3 1
44.3
36.9
35.5
27.5
45.4
44.4
ch4
0.4
0.5
0.5
0.4
0.4
0.5
0.5
n2o
0.2 I
0.2 III
0.2
0.2
0.1
0.3
0.3
HFCs
0.0
0.5
3.6
3.9
4.2
4.5
4.8
Rail
39.0
51.4
42.5
39.7
34.2
35.5
35.6
C02
38.5
50.8
41.9
39.1
33.7
34.9
35.0
ch4
0-1 9
0.1
0.1
0.1
0.1
0.1
0.1
n2o
0.3
0.4
0.3
0.3
0.3
0.3
0.3
HFCs
0.0 3
0.1:::
0.1
0.1
0.1
0.1
0.1
Other Emissions from Electric
Powerd
0.1
0.1
0.1
0.1
0.1
0.1
0.1
Pipelines8
36.0
32.8
50.3
58.3
58.0
64.4
69.3
C02
36.0
32.8
50.3
58.3
58.0
64.4
69.3
Lubricants
11.8
10.2
9.2
8.8
7.8
8.0
8.4
C02
11.8
10.2
9.2
8.8
7.8
8.0
8.4
Total Transportation
1,524.6
1,970.8
1,876.5
1,879.5
1,629.5
1,810.6
1,807.8
International Bunker Fuelsf
54.7
44.6
32.4
26.2
22.7
22.7
25.3
Ethanol C029
1
21.6 Z
78.6
78.7
68.1
75.4
75.0
Biodiesel C02g
0.0
0.9
17.9
17.1
17.7
16.1
15.6
+ Does not exceed 0.05 MMT C02 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 electric power are a result of waste incineration (as the majority of MSW is combusted in "trash-to-
steam" electric power plants), electrical equipment, and a portion of other process uses of carbonates (from pollution
control equipment installed in electric power plants).
e C02 estimates reflect natural gas used to power pipelines, but not electricity. While the operation of pipelines produces CH4
and N20, these emissions are not directly attributed to pipelines in the Inventory.
f Emissions from International Bunker Fuels include emissions from both civilian and military activities; these emissions are
not included in the transportation totals.
Trends 2-39
-------
g Ethanol and biodiesel C02 estimates are presented for informational purposes only. See Section 3.11 and the estimates in
LULUCF (see Chapter 6), in line with IPCC methodological guidance and reporting obligations under the Paris Agreement and
the UNFCCC, for more information on ethanol and biodiesel.
Notes: Passenger cars and light-duty trucks include vehicles typically used for personal travel and less than 8,500 lbs;
medium- and heavy-duty trucks include vehicles larger than 8,500 lbs. HFC emissions primarily reflect HFC-134a. Totals may
not sum due to independent rounding.
Residential
The residential end-use sector, including electricity-related emissions, accounted for 15.3 percent of U.S.
greenhouse gas emissions in 2022. This sector is heavily reliant on electricity for meeting energy needs, with
electricity use for building-related activities like 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 sector have generally been increasing since 1990,
and annual variations are often correlated with short-term fluctuations in energy use caused by weather
conditions, rather than prevailing economic conditions. In the long term, the residential sector is also affected by
population growth, migration trends toward warmer areas, and changes in housing and building attributes (e.g.,
larger sizes and improved insulation). A shift toward energy-efficient products and more stringent energy
efficiency standards for household equipment has also contributed to recent trends in energy demand in
households.
Commercial
The commercial end-use sector, including electricity-related emissions, accounted for 15.8 percent of U.S.
greenhouse gas emissions in 2022. Like the residential sector it is heavily reliant on electricity for meeting energy
needs, with electricity use for building-related activities like 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 commercial sector have
generally been increasing since 1990, and annual variations are often correlated with short-term fluctuations in
energy use caused by weather conditions, rather than prevailing economic conditions. Decreases in energy-related
emissions in the commercial sector in recent years can be largely attributed to an overall reduction in energy use
driven by a reduction in heating degree days and increases in energy efficiency.
Municipal landfills and wastewater treatment are included in the commercial sector, with landfill emissions
decreasing since 1990 and wastewater treatment emissions increasing slightly.
Agriculture
The agriculture end-use sector accounted for 10.5 percent of U.S. greenhouse gas emissions in 2022 when
electricity-related emissions are distributed, and includes a variety of processes, including enteric fermentation in
domestic livestock, livestock manure management, and agricultural soil management. In 2022, agricultural soil
management was the largest source of N2O emissions, and enteric fermentation was the largest source of CH4
emissions in the United States. This sector also includes small amounts of CO2 emissions from fossil fuel
combustion by motorized farm equipment such as tractors. Indirect emissions from electricity use in agricultural
activities (e.g., powering buildings and equipment) are about 5 percent of direct emissions.
Box 2-2: Trends in Various U.S. Greenhouse Gas Emissions-Related Data
Total (gross) greenhouse gas emissions can be compared to other economic and social indices to highlight
changes over time. These comparisons include: (1) aggregate energy use, because energy-related activities are
the largest sources of emissions; (2) energy use per capita as a measure of efficiency; (3) emissions per unit of
total gross domestic product as a measure of national economic activity; and (4) emissions per capita.
2-40 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
Table 2-14 provides data on various statistics related to U.S. greenhouse gas emissions normalized to 1990 as a
baseline year. These values represent the relative change in each statistic since 1990. Greenhouse gas emissions
in the United States have decreased at an average annual rate of 0.1 percent since 1990, although changes from
year to year have been significantly larger. This growth rate is slightly slower than that for total energy use,
overall gross domestic product (GDP) and national population (see Table 2-14 and Figure 2-16). The direction of
these trends started to change after 2005, when greenhouse gas emissions, total energy use and associated
fossil fuel consumption began to peak. Greenhouse gas emissions in the United States have decreased at an
average annual rate of 0.9 percent since 2005. Since 2005, GDP, and national population, generally continued to
increase, and energy use has decreased slightly, noting 2020 was impacted by the COVID-19 pandemic.
Table 2-14: Recent Trends in Various U.S. Data (Index 1990 = 100)
Variable
1990
2005
2018
2019
2020
2021
2022
Avg. Annual
Change
Since 1990a
Avg. Annual
Change
Since 2005a
Greenhouse Gas
Emissions'5
100
115
103
101
92
97
97
-0.1%
-0.9%
Energy Usec
100
119
118
117
107
113
115
0.5%
-0.2%
GDPd
100
159
201
206
201
213
217
2.5%
1.9%
Population6
100
118
130
131
132
132
133
0.9%
0.7%
a Average annual growth rate.
b Gross total GWP-weighted values.
c Energy-content-weighted values (EIA 2024).
d GDP in chained 2017 dollars (BEA 2024).
e U.S. Census Bureau (2024).
Figure 2-16: U.S. Greenhouse Gas Emissions Per Capita and Per Dollar of Gross Domestic
Product
Source: BEA (2024), U.S. Census Bureau (2024), and gross emission estimates in this report.
Trends 2-41
-------
2.3 Precursor Greenhouse Gas Emissions
(CO, NOx, NMVOCs, and S02)
The reporting requirements of the Paris Agreement and the UNFCCC11 request that information be provided on
emissions of compounds that are precursors to greenhouse gases, which include carbon monoxide (CO), nitrogen
oxides (NOx), non-methane volatile organic compounds (NMVOCs), and sulfur dioxide (SO2). These gases are not
direct greenhouse gases, but can indirectly impact Earth's radiative balance, by altering the concentrations of
other greenhouse gases (e.g., tropospheric ozone) and atmospheric aerosol (e.g., particulate sulfate). Carbon
monoxide is produced when carbon-containing fuels are combusted incompletely in energy, transportation, and
industrial processes, and is also emitted from practices such as agricultural burning and waste disposal and
treatment. Anthropogenic sources of nitrogen oxides (i.e., NO and NO2) are primarily fossil fuel combustion (for
energy, transportation, industrial process) and agricultural burning. Anthropogenic sources of NMVOCs, which
include hundreds of organic compounds that participate in atmospheric chemical reactions (propane, butane,
xylene, toluene, ethane, and many others)—are emitted primarily from transportation, industrial processes, oil
and natural gas production, waste practices, agricultural burning, 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.
As noted above and summarized in Chapter 6 of IPCC (2021), these compounds can have important indirect effects
on Earth's radiative balance. For example, reactions between NMVOCs and NOx in the presence of sunlight lead to
formation of tropospheric ozone, a greenhouse gas. Concentrations of NMVOCs, NOx, and CO can also impact the
abundance and lifetime of primary greenhouse gases. This largely occurs by altering the atmospheric
concentrations of the hydroxyl radical (OH), which is the main sink for atmospheric Cm. For example, NOx
emissions can lead to increases in O3 concentrations and subsequent OH production, which will increase the
amount of OH molecules that are available to destroy CH4. In contrast, NMVOCs and CO can both react directly
with OH, leading to lower OH concentrations, a longer atmospheric lifetime of CH4, and a decrease in CO2
production (i.e., CO+OH-> CO2). Changes in atmospheric CH4 can also feedback on background concentrations of
tropospheric O3. Other indirect impacts include the formation of sulfate and nitrate aerosol from emissions of NOx
and SO2, both of which have a net negative impact on radiative forcing.
Since 1970, the United States has published triennial estimates of emissions of CO, NOx, NMVOCs, and SO2 (EPA
2023a), which are regulated under the Clean Air Act. Emissions of each of these precursor greenhouse gases has
decreased significantly since 1990 as a result of implementation of Clean Air Act programs, as well as technological
improvements.12 Precursor emission estimates for this report for 1990 through 2022 were obtained from data
published on EPA's National Emissions Inventory (NEI) Air Pollutants Emissions Trends Data website (EPA 2023a).
For Table 2-15, NEI-reported emissions of CO, NOx, SO2, and NMVOCs are recategorized from NEI Emissions
Inventory System (EIS) source categories to those more closely aligned with reporting sectors and categories under
the Paris Agreement and the UNFCCC, based on the crosswalk detailed in Annex 6.3. Table 2-15 shows that fuel
combustion accounts for the majority of emissions of these precursors. 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. Precursor emissions from Agriculture and LULUCF categories are
estimated separately and therefore are not taken from EPA (2023a).
11 See paragraph 51 of Annex to 18/CMA.l available online at:
https://unfccc.int/sites/default/files/resource/CMA2018 03a02E.pdf.
12 More information is available online at: https://www.epa.gov/clean-air-act-overview/progress~cleaning-air-and-improving-
peoples-health and https://gispub.epa.gov/neireport/2017/.
2-42 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
Table 2-15: Emissions of NOx, CO, NMVOCs, and SO2 (kt)
Gas/Activity
1990
2005
2018
2019
2020
2021
2022
NOx
22,896
19,754
8,064
7,639
6,879
6,996
6,589
Energy
21,966 1
18,863 :
7,384
7,048
6,235
6,308
5,964
IPPU
774
672
461
440
393
403
389
Agriculture
16 i
18 S
17
18
17
18
19
LULUCF
56
149
130
61
158
190
142
Waste
84
51
73
73
76
76
75
CO
133,549
76,691
38,656
36,234
38,911
41,677
38,853
Energy
124,713 1
64,455;
30,760
30,349
28,427
28,845
28,173
IPPU
4,099
1,701
1,022
1,011
855
902
897
Agriculture
407 ¦
480
433
468
446
480
501
LULUCF
3,301
8,877
5,259
3,224
7,841
10,107
7,939
Waste
l—H«
00
1,182
1,182
1,342
1,343
1,343
NMVOCs
20,918
12,708
8,987
8,804
9,040
9,454
9,325
Energy
13,067 =
' *
8,694::
5,506
5,444
5,306
5,568
5,442
IPPU
6,982
3,668
3,119
2,996
3,366
3,508
3,505
Agriculture
194::
206
208
196
206
206
LULUCF
NA
NA
NA
NA
NA
NA
NA
Waste
llllli
00
152 1
156
157
173
172
172
so2
20,924
13,108
2,001
1,676
1,471
1,621
1,522
Energy
19,400 1
12,312 1
1,643
1,344
1,173
1,315
1,229
IPPU
1,488
776
335
309
266
274
261
Agriculture
+
o!
0
0
+
+
+
LULUCF
NA
NA
NA
NA
NA
NA
NA
Waste
36 I
20 1
23
23
33
32
31
+ Does not exceed 0.5 kt.
NA (Not Available)
Note: Totals by gas may not sum due to independent rounding.
Source: (EPA 2023a) except for estimates from forest fires, grassland fires, and field burning of agricultural residues. Emission
categories from EPA (2023a) are aggregated into sectors and categories reported under the Paris Agreement and the
UNFCCC as shown in Table ES-3.
Trends 2-43
-------
3^ s
Energy-related activities were the primary sources of U.S. anthropogenic greenhouse gas emissions, accounting for
82.0 percent of total gross greenhouse gas emissions on a carbon dioxide (CO2) equivalent basis in 2022.1 This
included 96.5, 40.2, and 10.8 percent of the nation's CO2, methane (CH4), and nitrous oxide (N2O) emissions,
respectively.2 Energy-related CO2 emissions alone constituted 76.9 percent of total gross U.S. greenhouse gas
emissions from all sources on a CC>2-equivalent basis, while the non-CC>2 emissions from energy-related activities
represented a much smaller portion of total gross national emissions (5.1 percent collectively).
Emissions from fossil fuel combustion contribute the vast majority of energy-related emissions, with CO2 being the
primary gas emitted (see Figure 3-1 and Figure 3-2). Globally, approximately 33,500 million metric tons (MMT) of
CO2 were added to the atmosphere through the combustion of fossil fuels in 2022, of which the United States
accounted for approximately 14 percent.3 Due to their relative importance over time (see Figure 3-2), fossil fuel
combustion-related CO2 emissions are considered in more detail than other energy-related emissions in this report
(see Figure 3-3).
Fossil fuel combustion also emits CFU 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 fifth largest source. Energy-related
activities other than fuel combustion, such as the production, transmission, storage, and distribution of fossil fuels,
also emit greenhouse gases. These emissions consist primarily of fugitive CH4 emissions from natural gas systems,
coal mining, and petroleum systems.
1 Estimates are presented in units of million metric tons of carbon dioxide equivalent (MMT C02 Eq.), which weight each gas by
its global warming potential, or GWP, value. See section on global warming potentials in the Executive Summary.
2 The contribution of energy non-C02 emissions is based on gross totals so excludes LULUCF methane (CH4) and nitrous oxide
(N20) emissions. The contribution of energy-related methane (CH4) and (N20) including LULUCF non-C02 emissions, is 37.1
percent and 9.8 percent respectively.
3 Global C02 emissions from fossil fuel combustion were taken from International Energy Agency Global energy-related C02
emissions, 2022. Available at: https://www.iea.org/reports/co2-emissions-in-2022 (IEA 2022).
Energy 3-1
-------
Figure 3-1: 2022 Energy Sector Greenhouse Gas Sources
MMT COz Eq.
Figure 3-2: Trends in Energy Sector Greenhouse Gas Sources
3-2 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
Figure 3-3: 2022 U.S. Fossil Carbon Flows (MMT CO2 Eq.)
International
Table 3-1 summarizes emissions from the Energy sector in units of MMT CO2 Eq., while unweighted gas emissions
in kilotons (kt) are provided in Table 3-2. Overall, emissions due to energy-related activities were 5,199.8 MMT CO2
Eq. in 2022,4 a decrease of 3.4 percent since 1990 and an increase of 0.5 percent since 2021. The increase in
emissions in 2021 and 2022 was due to continued rebounding activity levels after the coronavirus (COVID-19)
pandemic reduced overall demand for fossil fuels across all sectors in 2020. Longer term trends are driven by a
number of factors including a shift from coal to natural gas and renewables in the electric power sector.
Table 3-1: CO2, CH4, and N2O Emissions from Energy (MMT CO2 Eq.)
Gas/Source
1990
2005
2018
2019
2020
2021
2022
co2
4,910.9
5,923.1
5,190.6
5,059.1
4,520.2
4,840.7
4,875.5
Fossil Fuel Combustion
4,752.2
5,744.1
4,988.2
4,852.6
4,341.7
4,654.3
4,699.4
Transportation
1,468.9
1,858.6
1,813.1
1,816.6
1,572.8
1,753.5
1,751.3
Electricity Generation
1,820.0
2,400.1
1,753.4
1,606.7
1,439.6
1,540.9
1,531.7
Industrial
876.5
847.6
810.5
809.8
762.0
780.5
801.1
Residential
338.6
358.9
338.9
342.9
314.8
318.0
334.1
Commercial
228.3
227.1
246.3
251.7
229.3
237.5
258.7
U.S. Territories
20.0
51.9
25.9
24.8
23.3
23.8
22.6
Non-Energy Use of Fuels
99.1
125.0
118.4
106.5
97.8
111.6
102.8
Natural Gas Systems
32.4
26.3
32.8
38.5
36.7
35.8
36.5
Petroleum Systems
9.6
10.2
34.8
45.5
28.9
24.1
22.0
Incineration of Waste
12.9
13.3
13.3
12.9
12.9
12.5
12.4
Coal Mining
4.6
4.2
3.1
3.0
2.2
2.5
2.5
Abandoned Oil and Gas Wells
+
+
+
+
+
+
+
Biomass-Wooda
215.2
206.9
220.0
217.7
190.6
192.5
195.3
International Bunker Fuelsb
103.6
113.3
124.3
113.6
69.6
80.2
98.2
Biofuels-Ethanola
4.2
22.9
81.9
82.6
71.8
79.1
79.6
4 Following the current reporting requirements under the UNFCCC, this Inventory report presents C02 equivalent values based
on the IPCC Fifth Assessment Report (AR5) GWP values. See Chapter 1, Introduction for more information.
Energy 3-3
-------
Gas/Source
1990
2005
2018
2019
2020
2021
2022
Biofuels-Biodiesel"
0.0
111111!
0.9 ¦
17.9
17.1
17.7
16.1
15.6
Biomass-MSW"
18.5
14.7
16.1
15.7
15.6
15.3
14.9
ch4
409.0
358.5
336.2
321.7
305.3
293.3
282.4
Natural Gas Systems
218.8
210.1
190.3
188.7
180.3
174.6
173.1
Coal Mining
108.1
71.5
59.1
53.0
46.2
44.7
43.6
Petroleum Systems
49.4
48.2
59.0
52.2
53.3
48.6
39.6
Stationary Combustion
9-7 :
8.8 I
9.6
9.8
8.0
8.0
8.6
Abandoned Oil and Gas Wells
7.8
8.2
8.4
8.5
8.5
8.6
8.5
Abandoned Underground Coal
CO
i-*
liiiiii
7.4
6.9
6.6
6.5
6.3
6.3
Mines
1
=:
1
!!!!!!'
Mobile Combustion
7.2
4.3
2.8
2.9
2.5
2.6
2.6
Incineration of Waste
+ is
+
+
+
+
+
International Bunker Fuelsb
0.2
0.1
0.1
0.1
0.1
0.1
0.1
N20
61.2
67.9
43.2
41.6
37.1
39.2
41.9
Stationary Combustion
22.3
30.5
25.1
22.2
20.5
22.0
24.7
Mobile Combustion
38.4
37.O if
17.7
19.1
16.1
16.8
16.7
Incineration of Waste
0.4
0.3
0.4
0.4
0.3
0.4
0.3
Natural Gas Systems
+
~ I
+
+
+
+
0.2
Petroleum Systems
+
+
+
+
+
+
+
International Bunker Fuelsb
08
0.9 '
1.0
0.9
0.5
0.6
0.8
Total
5,381.0
6,349.5
5,570.0
5,422.4
4,862.6
5,173.3
5,199.8
+ Does not exceed 0.05 MMT C02 Eq.
a Emissions from biomass and biofuel 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 LULUCF.
b Emissions from international bunker fuels are not included in totals. These values are presented for informational
purposes only, in line with the 2006IPCC Guidelines and Paris Agreement and UNFCCC reporting obligations.
Note: Totals may not sum due to independent rounding.
Table 3-2: CO2, CH4, and N2O Emissions from Energy (kt)
Gas/Source
1990
2005
2018
2019
2020
2021
2022
C02
4,910,861
5,923,075
5,190,611
5,059,075
4,520,249
4,840,748
4,875,487
Fossil Fuel Combustion
4,752,232 |
5,744,134
4,988,198
4,852,631
4,341,710
4,654,265
4,699,403
Non-Energy Use of Fuels
99,104
124,988
118,382
106,474
97,757
111,624
102,808
Natural Gas Systems
32,427 |
26,312 :
32,768
38,525
36,719
35,780
36,470
Petroleum Systems
9,585
10,210
34,777
45,498
28,937
24,140
21,967
Incineration of Waste
12,900
1.3,254
13,339
12,948
12,921
12,476
12,357
Coal Mining
4,606
4,169
3,139
2,992
2,197
2,455
2,474
Abandoned Oil and Gas Wells
7 3
7
8
8
8
8
8
Biomass-Wood"
215,186 1
206,901
220,003
217,690
190,554
192,509
195,338
International Bunker Fuelsb
103,634 1
113,328
124,279
113,632
69,638
80,180
98,241
Biofuels-Ethanola
4,227
22,943
81,917
82,578
71,848
79,064
79,593
Biofuels-Biodiesela
0
856 E
17,936
17,080
17,678
16,112
15,622
Biomass-MSWa
18,534
14,722
16,115
15,709
15,614
15,329
14,864
ch4
14,607
12,804
12,007
11,490
10,903
10,476
10,084
Natural Gas Systems
7,813
7,505
6,795
6,741
6,439
6,235
6,183
Coal Mining
3,860 :
2,552
2,110
1,892
1,648
1,595
1,558
Petroleum Systems
1,765
1,723
2,108
1,865
1,904
1,737
1,415
Stationary Combustion
345 5
313
344
351
285
286
307
Abandoned Oil and Gas Wells
279
294
301
302
303
306
303
Abandoned Underground
288 5
I
264 |5
247
237
232
224
225
Coal Mines
Mobile Combustion
258
154
101
102
91
92
93
Incineration of Waste
¦
+
+
+
+
+
+
+
3-4 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
Gas/Source
1990
2005
2018
2019
2020
2021
2022
International Bunker Fuelsb
7
5
4
4
3
3
3
N20
231
256
163
157
140
148
158
Stationary Combustion
84
115
95
84
78
83
93
Mobile Combustion
145
140
67
72
61
63
63
Incineration of Waste
2
1
1
1
1
1
1
Natural Gas Systems
+
+
+
+
+
+
0.6
Petroleum Systems
+
+
+
+
+
+
+
International Bunker Fuelsb
3
3
4
3
2
2
3
+ Does not exceed 0.5 kt.
a Emissions from biomass and biofuel 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 LULUCF.
b Emissions from international bunker fuels are not included in totals. These values are presented for informational
purposes only, in line with the 2006IPCC Guidelines and UNFCCC reporting obligations.
Note: Totals by gas may not sum due to independent rounding.
Emissions estimates reported in the Energy chapter from fossil fuel combustion and fugitive sources include those
from all 50 states, including Hawaii and Alaska, and the District of Columbia. Emissions are also included from U.S.
Territories to the extent they are known to occur (e.g., coal mining does not occur in U.S. Territories). For some
sources there is a lack of detailed information on U.S. Territories including some non-CC>2 emissions from biomass
combustion. As part of continuous improvement efforts, EPA reviews this on an ongoing basis to ensure emission
sources are included across all geographic areas including U.S. Territories if they are occurring. See Annex 5 for
more information on EPA's assessment of the sources not included in this Inventory.
Each year, some emission and sink estimates in the Inventory are recalculated and revised with improved methods
and/or data. In general, recalculations are made to the U.S. greenhouse gas emission estimates either to
incorporate new methodologies or, most commonly, to update recent historical data. These improvements are
implemented consistently across the previous Inventory's time series (i.e., 1990 to 2021) to ensure that the trend
is accurate. Key updates in this year's Inventory include, updated methodologies for completion and workover
emissions estimates and transmission compressor station activity from both natural gas systems and petroleum
systems, a shift of all product supplied of natural gasoline and unfinished oils to crude oil transfers for the time
series and changes to the non-energy use of fossil fuel methodology (e.g., updates to some of the data and
updated methodology for the amount of NEU HGLs). The impact of these recalculations averaged a decrease of 0.2
MMT CO2 Eq. (less than 0.1 percent) per year across the time series. For more information on specific
methodological updates, please see the Recalculations Discussion section for each category in this chapter.
Box 3-1: Methodological Approach for Estimating and Reporting U.S. Emissions and
Removals, including Relationship to EPA's Greenhouse Gas Reporting Program
Consistent with Article 13.7(a) of the Paris Agreement and Article 4.1(a) of the UNFCCC as well as relevant
decisions under those agreements, the emissions and removals presented in this report and this chapter are
organized by source and sink categories and calculated using internationally accepted methods in the 2006 IPCC
Guidelines for National Greenhouse Gas Inventories (2006 IPCC Guidelines). Additionally, the calculated
emissions and removals in a given year for the United States are presented in a common format in line with the
reporting guidelines for the reporting of inventories under the Paris Agreement and the UNFCCC. The Parties'
use of consistent methods to calculate emissions and removals for their inventories helps to ensure that these
reports are comparable. The presentation of emissions and removals provided in the Energy chapter do not
preclude alternative examinations (e.g., economic sectors). Rather, this chapter presents emissions and
removals in a common format consistent with how Parties are to report their national inventories under the
Paris Agreement and the UNFCCC. The report itself, and this chapter, follows this common format, and provides
an explanation of the application of methods used to calculate emissions and removals from energy-related
activities.
Energy Data from EPA's Greenhouse Gas Reporting Program
Energy 3-5
-------
EPA's Greenhouse Gas Reporting Program (GHGRP)5 dataset and the data presented in this Inventory are
complementary. The Inventory was used to guide the development of the GHGRP, particularly in terms of scope
and coverage of both sources and gases. The GHGRP dataset continues to be an important resource for the
Inventory, providing not only annual emissions information, but also other annual information, such as activity
data and emission factors that can improve and refine national emission estimates and trends over time. GHGRP
data also allow EPA to disaggregate national inventory estimates in new ways that can highlight differences
across regions and sub-categories of emissions, along with enhancing application of QA/QC procedures and
assessment of uncertainties.
EPA uses annual GHGRP data in a number of Energy sector categories to improve the national estimates
presented in this Inventory consistent with IPCC guidelines (see Box 3-3 of this chapter, and Sections 3.3
Incineration of Waste, 3.4 Coal Mining, 3.6 Petroleum Systems, and 3.7 Natural Gas Systems).6 Methodologies
used in EPA's GHGRP are consistent with IPCC guidelines, including higher tier methods. Under EPA's GHGRP,
facilities collect detailed information specific to their operations according to detailed measurement standards.
It should be noted that the definitions and provisions for reporting fuel types in EPA's GHGRP may differ from
those used in the Inventory in meeting the Paris Agreement and UNFCCC reporting guidelines. In line with the
Paris Agreement and UNFCCC reporting guidelines, the Inventory report is a comprehensive accounting of all
emissions from fuel types identified in the IPCC guidelines and provides a separate reporting of emissions from
biomass.
In addition to using GHGRP data to estimate emissions (Sections 3.3 Incineration of Waste, 3.4 Coal Mining, 3.6
Petroleum Systems, and 3.7 Natural Gas Systems), EPA also uses the GHGRP fuel consumption activity data in
the Energy sector to disaggregate industrial end-use sector emissions in the category of CO2 emissions from
fossil fuel combustion, for use in reporting emissions in Common Reporting Tables (CRTs) (see Box 3-3). The
industrial end-use sector activity data collected for the Inventory (EIA 2024) represent aggregated data for the
industrial end-use sector. EPA's GHGRP collects industrial fuel consumption activity data by individual categories
within the industrial end-use sector. Therefore, GHGRP data are used to provide a more detailed breakout of
total emissions in the industrial end-use sector within that source category.
As indicated in the respective Planned Improvements sections for source categories in this chapter, EPA
continues to examine the uses of facility-level GHGRP data to improve the national estimates presented in this
Inventory. See Annex 9 for more information on use of EPA's GHGRP in the Inventory.
3.1 Fossil Fuel Combustion (CRT Source
Category 1A)
Emissions from the combustion of fossil fuels for energy include the greenhouse gases CO2, CH4, and N2O. Given
that CO2 is the primary gas emitted from fossil fuel combustion and represents the largest share of U.S. total
emissions, CO2 emissions from fossil fuel combustion are discussed at the beginning of this section. An overview of
CH4 and N2O emissions from the combustion of fuels in stationary sources is then presented, followed by fossil fuel
5 On October 30, 2009, the U.S. Environmental Protection Agency (EPA) published a rule requiring annual reporting of
greenhouse gas data from large greenhouse gas emission sources in the United States. Implementation of the rule, codified at
40 CFR Part 98, is referred to as EPA's Greenhouse Gas Reporting Program (GHGRP).
6 See http://www.ipcc-negip.iges.or.jp/public/tb/TFI Technical Bulletin l.pdf.
3-6 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
combustion emissions for all three gases by sector: electric power, industrial, residential, commercial, U.S.
Territories, and transportation.
Methodologies for estimating CO2 emissions from fossil fuel combustion differ from the estimation of CH4 and N2O
emissions from stationary combustion and mobile combustion. Thus, three separate descriptions of
methodologies, uncertainties, recalculations, and planned improvements are provided at the end of this section.
Total CO2, Cm, and N2O emissions from fossil fuel combustion are presented in Table 3-3 and Table 3-4.
Table 3-3: CO2, CH4, and N2O Emissions from Fossil Fuel Combustion (MMT CO2 Eq.)
Gas 1990
2005
2018
2019
2020
2021
2022
C02 4,752.2 1
5,744.1 1
4,988.2
4,852.6
4,341.7
4,654.3
4,699.4
CH4 16.9
13.1
12.5
12.7
10.5
10.6
11.2
N;0 60.8
67.6
42.8
41.2
36.7
38.9
41.4
Total 4,829.9
5,824.8
5,043.4
4,906.6
4,388.9
4,703.7
4,752.0
Note: Totals may not sum due to independent rounding.
Table 3-4: CO2, CH4, and N2O Emissions from Fossil Fuel Combustion (kt)
Gas 1990
2005
2018
2019
2020
2021
2022
C02 4,752,232
5,744,134
4,988,198
4,852,631
4,341,710
4,654,265
4,699,403
CH4 602
467;;;
445
453
376
379
401
N20 229
255
161
156
138
147
156
CO2 from Fossil Fuel Combustion
Carbon dioxide is the primary gas emitted from fossil fuel combustion and represents the largest share of U.S. total
greenhouse gas emissions. Carbon dioxide emissions from fossil fuel combustion are presented in Table 3-5. In
2022, CO2 emissions from fossil fuel combustion increased by 1.0 percent relative to the previous year (as shown in
Table 3-6). The increase in CO2 emissions from fossil fuel combustion was a result of a 1.8 percent increase in fossil
fuel energy use. This increase in fossil fuel energy use was due primarily to the continued rebound in economic
activity after the COVID-19 pandemic. Carbon dioxide emissions from natural gas increased by 84.8 MMT CO2 Eq.,
a 5.2 percent increase from 2021. In a shift from last year's trend, CO2 emissions from coal consumption decreased
by 58.6 MMT CO2 Eq., a 6.1 percent decrease from 2021. Both the increase in natural gas and decrease in coal
consumption and emissions in 2022 are observed across all sectors. Emissions from petroleum use also increased
19.0 MMT CO2 Eq. (0.9 percent) from 2021 to 2022. In 2022, CO2 emissions from fossil fuel combustion were
4,699,4 MMT CO2 Eq., or 1.1 percent below emissions in 1990 (see Table 3-5).7
7 An additional discussion of fossil fuel emission trends is presented in the Trends in U.S. Greenhouse Gas Emissions chapter.
Energy 3-7
-------
Table 3-5: CO2 Emissions from Fossil Fuel Combustion by Fuel Type and Sector (MMT CO2 Eq.)
Fuel/Sector
1990
2005
2018
2019
2020
2021
2022
Coal
1,719.8
2,113.7
1,211.6
1,028.1
835.6
957.4
898.8
Residential
3.0
0.8
NO
NO
NO
NO
NO
Commercial
12.0
9.3
1.8
1.6
1.4
1.4
1.4
Industrial
157.8 I
117.8
54.4
49.4
43.0
43.0
43.0
Transportation
NO
NO
NO
NO
NO
NO
NO
Electric Power
1,546.5 "
0.5
i'982-8 S
1,152.9
973.5
788.2
910.1
851.5
U.S. Territories
3.0
2.6
3.6
3.1
2.9
2.9
Natural Gas
998.6
1,166.2
1,592.0
1,649.2
1,615.7
1,622.1
1,706.8
Residential
237.8
262.2
273.8
275.5
256.4
258.6
272.0
Commercial
142.0
162-9 :
192.5
192.9
173.5
180.4
192.3
Industrial
407.4
387.8
493.5
501.5
489.7
501.2
510.4
Transportation
36.0
33.1:
50.9
58.9
58.7
65.2
70.2
Electric Power
175.4
318.9
577.9
616.6
634.8
612.8
659.3
U.S. Territories
no ;
1-3 1
3.3
3.8
2.6
3.9
2.7
Petroleum
2,033.3
2,463.8
2,184.2
2,174.9
1,890.0
2,074.4
2,093.4
Residential
97.8
95.9 '
65.1
67.4
58.4
59.4
62.1
Commercial
74.3
54.9
52.0
57.2
54.4
55.7
65.1
Industrial
3H.2 1
342.0 jjjjjj:
262.6
258.9
229.3
236.3
247.6
Transportation
1,432.9
1,825.5
1,762.2
1,757.7
1,514.2
1,688.4
1,681.1
Electric Power
97.5 E
98.0
22.2
16.2
16.2
17.7
20.5
U.S. Territories
19.5
47.6
20.1
17.5
17.5
17.0
17.0
Geothermal3
0.5
0.5
0.4
0.4
0.4
0.4
0.4
Electric Power
0.5
0.5
0.4
0.4
0.4
0.4
0.4
Total
4,752.2
5,744.1
4,988.2
4,852.6
4,341.7
4,654.3
4,699.4
NO (Not Occurring)
a Although not technically a fossil fuel, geothermal energy-related C02 emissions are included for reporting
purposes. The source of C02 is non-condensable gases in subterranean heated water.
Note: Totals may not sum due to independent rounding.
Trends in CO2 emissions from fossil fuel combustion are influenced by many long-term and short-term factors. On
a year-to-year basis, the overall demand for fossil fuels in the United States and other countries generally
fluctuates 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. The 2021 to 2022 trends reflect ongoing impacts of the COVID-19 pandemic which generally led to a
reduction in demand for fossil fuels in 2020, but an increase in demand as activities continued to rebound in 2022.
Longer-term changes in energy usage patterns, however, tend to be more a function of aggregate societal trends
that affect the scale of energy use (e.g., population, number of cars, size of houses, and number of houses), the
efficiency with which energy is used in equipment (e.g., cars, HVAC systems, power plants, steel mills, and light
bulbs), and social planning and consumer behavior (e.g., walking, bicycling, or telecommuting to work instead of
driving).
Carbon dioxide emissions also depend on the source of energy and its carbon intensity. The amount of carbon in
fuels varies significantly by fuel type. For example, coal contains the highest amount of carbon per unit of useful
energy. Petroleum has roughly 75 percent of the carbon per unit of energy as coal, and natural gas has only about
3-8 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
55 percent.8 Table 3-6 shows annual changes in emissions during the last five years for coal, petroleum, and
natural gas in selected sectors.
Table 3-6: Annual Change in CO2 Emissions and Total 2022 CO2 Emissions from Fossil Fuel
Combustion for Selected Fuels and Sectors (MMT CO2 Eq. and Percent)
Sector
Fuel Type
2018 to 2019
2019 to 2020
2020 to 2021
2021 to 2022
Total 2022
Transportation
Petroleum
-4.5
-0.3%
-243.5
-13.9%
174.2
11.5%
-7.2
-0.4%
1,681.1
Electric Power
Coal
-179.3
-15.6%
-185.4
-19.0%
121.9
15.5%
-58.6
-6.4%
851.5
Electric Power
Natural Gas
38.7
6.7%
18.2
3.0%
-22.1
-3.5%
46.5
7.6%
659.3
Industrial
Natural Gas
8.0
1.6%
-11.8
-2.4%
11.4
2.3%
9.2
1.8%
510.4
Residential
Natural Gas
1.7
0.6%
-19.1
-6.9%
2.3
0.9%
13.3
5.2%
272.0
Commercial
Natural Gas
0.4
0.2%
-19.5
-10.1%
6.9
4.0%
11.9
6.6%
192.3
Transportation
All Fuels3
3.5
0.2%
-243.8
-13.4%
180.7
11.5%
-2.3
-0.1%
1,751.3
Electric Power
All Fuels3
-146.7
-8.4%
-167.2
-10.4%
101.4
7.0%
-9.3
-0.6%
1,531.7
Industrial
All Fuels3
-0.7
-0.1%
-47.8
-5.9%
18.5
2.4%
20.6
2.6%
801.1
Residential
All Fuels3
4.0
1.2%
-28.1
-8.2%
3.2
1.0%
16.0
5.0%
334.1
Commercial
All Fuels3
5.5
2.2%
-22.5
-8.9%
8.3
3.6%
21.2
8.9%
258.7
All Sectorsa b
All Fuels3
-135.6
-2.7%
-510.9
-10.5%
312.6
7.2%
45.1
1.0%
4,699.4
a Includes sector and fuel combinations not shown in this table.
b Includes U.S. Territories.
Note: Totals may not sum due to independent rounding.
As shown in Table 3-6, recent trends in CO2 emissions from fossil fuel combustion show a 2.7 percent decrease
from 2018 to 2019, a 10.5 percent decrease from 2019 to 2020, a 7.2 percent increase from 2020 to 2021, and a
1.0 percent increase from 2021 to 2022. These changes contributed to an overall 5.8 percent decrease in CO2
emissions from fossil fuel combustion from 2018 to 2022.
The overall 2021 to 2022 trends were largely driven by the gradual recovery from the COVID-19 pandemic, which
saw reduced economic activity in 2020 and caused changes in energy demand and supply patterns across different
sectors. The continued recovery from the COVID-19 pandemic has generally led to increased energy use and
emissions across all economic sectors except electric power and transportation from 2021 to 2022. The decrease in
emissions from 2021 to 2022 from electric power was due to the reduction in coal consumption for electricity
generation, in a return to a pre-pandemic trend in declining coal-fired power generation.
Recent trends in CO2 emissions from fossil fuel combustion are largely driven by the electric power sector, which
until 2017 has accounted for the largest portion of these emissions. The types of fuels consumed to produce
electricity have changed in recent years. Electric power sector consumption of natural gas primarily increased due
to increased production capacity as natural gas-fired plants replaced coal-fired plants and increased electricity
demand related to heating and cooling needs (EIA 2018; EIA 2023a). Total net electric power generation from all
fossil and non-fossil sources decreased by 1.3 percent from 2018 to 2019, decreased by 2.9 percent from 2019 to
2020, increased by 2.7 percent from 2020 to 2021, and increased by 3.0 percent from 2021 to 2022 (EIA 2024a).
Carbon dioxide emissions from the electric power sector decreased from 2021 to 2022 by 0.6 percent due to
increased production and use of natural gas and decreased production and use of coal for electric power
generation. Carbon dioxide emissions from coal consumption for electric power generation decreased by 26.1
percent overall since 2018, including a 6.4 percent decrease from 2021 to 2022.
Petroleum use in the transportation sector is another major driver of emissions, representing the largest source of
CO2 emissions from fossil fuel combustion in 2022. Emissions from petroleum consumption for transportation have
decreased by 4.6 percent since 2018 and are primarily attributed to a 1.4 percent decrease in VMT over the same
8 Based on national aggregate carbon content of all coal, natural gas, and petroleum fuels combusted in the United States. See
Annex 2.2 for more details on fuel carbon contents.
Energy 3-9
-------
time period. As of 2017, the transportation sector is the largest source of national CO2 emissions-whereas in prior
years, electric power was the largest source sector.
In the United States, 83.0 percent of the energy used in 2022 was produced through the combustion of fossil fuels
such as petroleum, natural gas, and coal (see Figure 3-4 and Figure 3-5). Specifically, petroleum supplied the
largest share of domestic energy demands, accounting for 37 percent of total U.S. energy used in 2022. Natural gas
and coal followed in order of fossil fuel energy demand significance, accounting for approximately 35 percent and
10 percent of total U.S. energy used, respectively. Petroleum was consumed primarily in the transportation end-
use sector and the majority of coal was used in the electric power sector. Natural gas was broadly consumed in all
end-use sectors except transportation (see Figure 3-6) (EIA 2024a). The remaining portion of energy used in 2022
was supplied by nuclear electric power (8 percent) and by a variety of renewable energy sources (9 percent),
primarily wind energy, hydroelectric power, solar, geothermal and biomass (EIA 2024a).9
Figure 3-4: 2022 U.S. Energy Use by Energy Source
Nuclear Electric Power
Note: Totals may not sum due to independent rounding.
Figure 3-5: Annual U.S. Energy Use
120
g, 100
80
60
40
20
Total Energy
Fossil Fuels
Renewable & Nuclear
Oi-irNcOTiritDr^cocTtOi-irNPi^rLritDr^oocTtOT-HrNPi^rLn^r^oocnOi-irvj
CT^O^O^O^O^O^O^O^O^O^OOOOOOOOOOi-Hi-HT-Hi-HT-HT-Hi-HT-Hi-Hi-H(NrsJ(N
cr>cr»cr>cr»cr>cr>cr»cr»cr»cr»ooooooooooooooooooooooo
i-HT-Hi-HT-Hi-Hi-HT-Hi-Hi-Hi-H(NOJrMrMfNrslCNfNrMfN(NfNfNfNfNfNfNfNfNfN(NfN(N
9 Renewable energy, as defined in ElA's energy statistics, includes the following energy sources: hydroelectric power,
geothermal energy, biomass, solar energy, and wind energy.
3-10 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
Figure 3-6: 2022 CO2 Emissions from Fossil Fuel Combustion by Sector and Fuel Type
2,500 Relative Contribution by Fuel Type
<0.05%
(Geothermal)
2,000
1,500
o
u
1,000
500
1,751
U.S. Territories
Commercial
Residential
Industrial
Electricity Generation Transportation
a Although not technically a fossil fuel, geothermal energy-related C02 emissions are included for reporting purposes. The
source of C02 is non-condensable gases in subterranean heated water.
Fossil fuels are generally combusted for the purpose of producing energy for useful heat and work. During the
combustion process, the carbon stored in the fuels is oxidized and emitted as CO2 and smaller amounts of other
gases, including Cm, carbon monoxide (CO), and non-methane volatile organic compounds (NMVOCs).10 These
other C-containing non-CC>2 gases are emitted as a byproduct of incomplete fuel combustion, but are, for the most
part, eventually oxidized to CO2 in the atmosphere. Therefore, as per IPCC guidelines, it is assumed that all of the
carbon in fossil fuels used to produce energy is eventually converted to atmospheric CO2.
Box 3-2: Weather and Non-Fossil Energy Effects on CO2 Emissions from Fossil Fuel
Combustion Trends
The United States in 2022 experienced a colder winter overall compared to 2021, with a 7.9 percent increase in
heating degree days, although 2022 heating degree days were 2.3 percent below normal (see Figure 3-7). Along
with a colder winter, 2022 experienced a warmer summer, with cooling degree days 16.9 percent above normal
and 4.3 percent higher compared to 2021 (see Figure 3-8) (EIA 2024a).11 Warmer summers and colder winters
can lead to increased energy use to heat and cool building spaces in the residential and commercial sectors. The
combination of colder winter and warmer summer conditions in 2022 as compared to 2021 led to an overall
increase in direct emissions from fossil fuel combustion in the residential and commercial sectors of 5.0 and 8.9
percent, respectively.
10 See the sections entitled Stationary Combustion and Mobile Combustion in this chapter for information on non-C02 gas
emissions from fossil fuel combustion.
11 Degree days are relative measurements of outdoor air temperature. Heating degree days are deviations of the mean daily
temperature below 65 degrees Fahrenheit, while cooling degree days are deviations of the mean daily temperature above 65
degrees Fahrenheit. 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 1991 through 2020. The variation in these
normals during this time period was +16 percent and +27 percent for heating and cooling degree days, respectively (99 percent
confidence interval).
Energy 3-11
-------
Figure 3-7: Annual Deviations from Normal Heating Degree Days for the United States
(1970-2022, Index Normal = 100)
Normal
30 (4,243 Heating Degree Days)
-20
Note: Climatological normal data are highlighted in dark red. Statistical confidence interval for
-30 "normal" climatology period of 1991 through 2020.
OfM^-VDOOOrsl^t-VOOOOrNTl-lOCIOOrNTj-VOOOOIN'rvDCOOOl
r-*.r-.r-.r>-.r-.aoceioocococricricric^cric>oooo^'-i-H'-i--cr>cr>cr>cr»cncricr>cr>cr>cr>cr>cr>oooooooooooo
t-Ht—It—It—I t-H t—I t-H t—It—It—It-Hi—It—It—It—IfNfNfNfNfNfNrvlfNfNfNfNfN
Figure 3-8: Annual Deviations from Normal Cooling Degree Days for the United States
(1970-2022, Index Normal = 100)
40
30
20
fD
I „
E
o
c 0
o
4—>
ooorNicoorsisr(£>ooor\i
rv.i\r^.r^r^cocococococricncricr\criOOOOOT-HT-iT-(T-HT-HrNjrNi
c7»cT»c7»cT»c^cr>cT»cr»cr»aicr>cT>oooooooooooo
t-it—ii—it-ht—it—it—it-ht-ht-ht—i-t-ht—irsirNJrMr\ir\irMr\irsJfNr\irMrsi
-40
3-12 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
The carbon intensity of the electric power sector is impacted by the amount of non-fossil energy sources of
electricity. The utilization (i.e., capacity factors)12 of nuclear power plants in 2022 remained high at 93 percent. In
2022, nuclear power represented 19 percent of total electricity generation. Since 1990, the wind and solar power
sectors have shown strong growth and have become relatively important sources of electricity. Between 1990 and
2022, renewable energy generation (in kWh) from solar and wind energy have increased from 0.1 percent in 1990
to 14 percent of total electricity generation in 2022, which helped drive the decrease in the carbon intensity of the
electricity supply in the United States.
Stationary Combustion
The direct combustion of fuels by stationary sources in the electric power, industrial, commercial, and residential
sectors represent the greatest share of U.S. greenhouse gas emissions. Table 3-7 presents CO2 emissions from
fossil fuel combustion by stationary sources. The CO2 emitted is closely linked to the type of fuel being combusted
in each sector (see the Methodology section of CO2 from Fossil Fuel Combustion). In addition to CO2 emissions, CFU
and N2O are emitted from fossil fuel combustion as well. Table 3-8 and Table 3-9 present CFU and N2O emissions
from the combustion of fuels in stationary sources. The CFU and N2O emissions are linked to the type of fuel being
combusted as well as the combustion technology (see the Methodology section for CFU and N2O from Stationary
Combustion).
Table 3-7: CO2 Emissions from Stationary Fossil Fuel Combustion (MMT CO2 Eq.)
Sector/Fuel Type
1990
2005
2018
2019
2020
2021
2022
Electric Power
1,820.0
2,400.1
1,753.4
1,606.7
1,439.6
1,540.9
1,531.7
Coal
a.,546.5;;
1,982.8
1,152.9
973.5
788.2
910.1
851.5
Natural Gas
175.4
318.9
577.9
616.6
634.8
612.8
659.3
Fuel Oil
97.5:
98.0
22.2
16.2
16.2
17.7
20.5
Geothermal
0.5
0.5
0.4
0.4
0.4
0.4
0.4
Industrial
876.5
847.6
810.5
809.8
762.0
780.5
801.1
Coal
157.8
117.8
54.4
49.4
43.0
43.0
43.0
Natural Gas
407.4 1
387.8 =
493.5
501.5
489.7
501.2
510.4
Fuel Oil
311.2
342.0
262.6
258.9
229.3
236.3
247.6
Residential
338.6
358.9
338.9
342.9
314.8
318.0
334.1
Coal
3.0
0.8
NO
NO
NO
NO
NO
Natural Gas
237.8
262.2
273.8
275.5
256.4
258.6
272.0
Fuel Oil
97.8
95.9
65.1
67.4
58.4
59.4
62.1
Commercial
228.3
227.1
9.3
246.3
251.7
229.3
237.5
258.7
Coal
12.0
1.8
1.6
1.4
1.4
1.4
Natural Gas
142.0 1
162.9 1
192.5
192.9
173.5
180.4
192.3
Fuel Oil
74.3
54.9
52.0
57.2
54.4
55.7
65.1
U.S. Territories
20.0
51.9
25.9
24.8
23.3
23.8
22.6
Coal
0.5
3.0
2.6
3.6
3.1
2.9
2.9
Natural Gas
no
!-3 =;;
3.3
3.8
2.6
3.9
2.7
Fuel Oil
19.5
47.6
20.1
17.5
17.5
17.0
17.0
Total
3,283.3
3,885.6
3,175.1
3,036.0
2,768.9
2,900.7
2,948.1
NO (Not Occurring)
Note: Totals may not sum due to independent rounding.
12 The capacity factor equals generation divided by net summer capacity. Summer capacity is defined as "The maximum output
that generating equipment can supply to system load, as demonstrated by a multi-hour test, at the time of summer peak
demand (period of June 1 through September 30)" (EIA 2020a). Data for both the generation and net summer capacity are from
EIA (2024a).
Energy 3-13
-------
Table 3-8: CH4 Emissions from Stationary Combustion (MMT CO2 Eq.)
Sector/Fuel Type
1990
2005
2018
2019
2020
2021
2022
Electric Power
0.5
1.0
1.4
1.4
1.4
1.4
1.3
Coal
0-4
0.4:
0.3
0.2
0.2
0.2
0.2
Fuel Oil
+
+
+
+
+
+
+
Natural gas
o-i ;=
0.5"
1.1
1.2
1.2
1.2
1.0
Wood
+
+
+
+
+
+
+
Industrial
2.1
1.9
1.7
1.7
1.6
1.6
1.6
Coal
0.5
0.3
0.2
0.1
0.1
0.1
0.1
Fuel Oil
0
k>
iiiiii
0.2
!!!!!!!
0.2
0.2
0.1
0.1
0.2
Natural gas
0.2
0.2
0.2
0.3
0.2
0.3
0.3
Wood
1.2 :
i-2:
1.1
1.1
1.1
1.1
1.0
Commercial
1.2
1.2
1.4
1.4
1.3
1.3
1.4
Coal
+ i
+
+
+
+
+
Fuel Oil
0.3
0.2
0.2
0.2
0.2
0.2
0.3
Natural gas
0.4
¦
0.4
0.5
0.5
0.4
0.5
0.5
Wood
0.5
0.6
0.7
0.7
0.7
0.7
0.7
Residential
5.9
4.5
5.1
5.3
3.6
3.6
4.3
Coal
0.3
0.1
NO
NO
NO
NO
NO
Fuel Oil
0.4"
¦
0.4
0.3
0.3
0.2
0.2
0.3
Natural Gas
0.6
0.7
0.7
0.7
0.6
0.6
0.7
Wood
4.6 *
1—¦
CO
4.2
4.4
2.8
2.7
3.4
U.S. Territories
0.1
+
+
+
+
+
Coal
+ 3
+ S
+
+
+
+
+
Fuel Oil
+
0.1
+
+
+
+
+
Natural Gas
NO """
+
+
+
+
+
Wood
NE
NE
NE
NE
NE
NE
NE
Total
9.7
8.8
9.6
9.8
8.0
8.0
8.6
+ Does not exceed 0.05 MMT C02 Eq.
NO (Not Occurring)
NE (Not Estimated)
Note: Totals may not sum due to independent rounding.
Table 3-9: N2O Emissions from Stationary Combustion (MMT CO2 Eq.)
Sector/Fuel Type
1990
2005
2018
2019
2020
2021
2022
Electric Power
18.2
26.7
21.7
18.8
17.5
19.0
21.6
Coal
17.9
24.9 I
18.1
14.8
13.5
15.1
18.2
Fuel Oil
0.1
0.1
+
+
+
+
+
Natural Gas
O
i-7!!!!!!
3.6
3.9
4.0
3.9
3.4
Wood
+
+
+
+
+
+
+
Industrial
2.8
2.6
2.2
2.2
2.0
2.1
2.0
Coal
0.7
0.5
0.2
0.2
0.2
0.2
0.2
Fuel Oil
0-5 =
O
Ln
!!¦!¦!
0.3
0.3
0.3
0.3
0.3
Natural Gas
0.2
0.2
0.2
0.2
0.2
0.2
0.2
Wood
13 1
1.5
1.4
1.4
1.4
1.4
1.3
Commercial
0.3
0.3
0.3
0.3
0.3
0.3
0.3
Coal
+
+i
+
+
+
+
+
Fuel Oil
0.2
0.1
0.1
0.1
0.1
0.1
0.1
Natural Gas
0.1 i
0.1
0.1
0.1
0.1
0.1
0.1
Wood
0.1
0.1
0.1
0.1
0.1
0.1
0.1
Residential
0.9
0.8
0.8
0.8
0.6
0.6
0.7
Coal
+
+
NO
NO
NO
NO
NO
Fuel Oil
0.2 I
0.2
0.2
0.2
0.1
0.1
0.1
3-14 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
Natural Gas
0.1
0.1
0.1
0.1
0.1
0.1
0.1
Wood
°-6
0.4 3
0.5
0.5
0.3
0.3
0.4
U.S. Territories
0.1
0.1
0.1
0.1
0.1
0.1
Coal
+ |
+
+
+
+
+
Fuel Oil
+
0.1
+
+
+
+
+
Natural Gas
NO |
+ §1
+
+
+
+
+
Wood
NE
NE
NE
NE
NE
NE
NE
Total
22.3
30.5
25.1
22.2
20.5
22.0
24.7
+ Does not exceed 0.05 MMT C02 Eq.
NO (Not Occurring)
NE (Not Estimated)
Note: Totals may not sum due to independent rounding.
Fossil Fuel Combustion Emissions by Sector
Table 3-10 provides an overview of the CO2, CH4, and N2O emissions from fossil fuel combustion by sector,
including transportation, electric power, industrial, residential, commercial, and U.S. Territories.
Table 3-10: CO2, CH4, and N2O Emissions from Fossil Fuel Combustion by Sector (MMT CO2
Eq.)
End-Use Sector
1990
2005
2018
2019
2020
2021
2022
Transportation
1,514.6
1,899.9
1,833.6
1,838.6
1,591.5
1,772.9
1,770.6
C02
1,468.9
1,858.6
1,813.1
1,816.6
1,572.8
1,753.5
1,751.3
ch4
7.2
4.3
2.8
2.9
2.5
2.6
2.6
n2o
38.41
37.0 1
17.7
19.1
16.1
16.8
16.7
Electric Power
1,838.7
2,427.8
1,776.5
1,626.9
1,458.5
1,561.3
1,554.5
C02
1,820.0
2,400.11
1,753.4
1,606.7
1,439.6
1,540.9
1,531.7
ch4
0.5
1.0
1.4
1.4
1.4
1.4
1.3
n2o
1.8.2 1
26.7 "
21.7
18.8
17.5
19.0
21.6
Industrial
881.3
852.2
814.4
813.7
765.6
784.1
804.7
C02
876.5
847.6
810.5
809.8
762.0
780.5
801.1
ch4
2.1
1.9
1.7
1.7
1.6
1.6
1.6
n2o
2.8 II
2.6
2.2
2.2
2.0
2.1
2.0
Residential
345.4
364.2
344.9
349.1
319.0
322.3
339.1
C02
338.6 =:
358.9
338.9
342.9
314.8
318.0
334.1
ch4
5.9
4.5
5.1
5.3
3.6
3.6
4.3
n2o
0.9
0.8
0.8
0.8
0.6
0.6
0.7
Commercial
229.8
228.6
248.0
253.5
230.9
239.2
260.5
C02
228.3 ;
227.1
246.3
251.7
229.3
237.5
258.7
ch4
1.2
1.2
1.4
1.4
1.3
1.3
1.4
n2o
0.3
0.3
0.3
0.3
0.3
0.3
0.3
U.S. Territories3
20.1
52.1
26.0
24.9
23.4
23.9
22.7
Total
4,829.9
5,824.8
5,043.4
4,906.6
4,388.9
4,703.7
4,752.0
a U.S. Territories are not apportioned by sector, and emissions shown in the table are total greenhouse gas emissions from all
fuel combustion sources.
Note: Totals may not sum due to independent rounding.
Other than the greenhouse gases CO2, Cm, and N2O, gases emitted from stationary combustion include the
greenhouse gas precursors nitrogen oxides (NOx), CO, NMVOCs, and sulfur dioxide (SO2). Methane and N2O
emissions from stationary combustion sources depend upon fuel characteristics and the size and vintage of
combustion device, along with combustion technology, pollution control equipment, ambient environmental
conditions, and operation and maintenance practices. Nitrous oxide emissions from stationary combustion are
closely related to air-fuel mixes and combustion temperatures, as well as the characteristics of any pollution
control equipment that is employed. Methane emissions from stationary combustion are primarily a function of
the Cm content of the fuel and combustion efficiency.
Energy 3-15
-------
Mobile combustion also produces emissions of Cm, N2O, and greenhouse gas precursors including NOx, CO, and
NMVOCs. As with stationary combustion, N2O and NOx emissions from mobile combustion are closely related to
fuel characteristics, air-fuel mixes, combustion temperatures, and the use of pollution control equipment. Nitrous
oxide from mobile sources, in particular, can be formed by the catalytic processes used to control NOx, CO, and
hydrocarbon emissions. Carbon monoxide emissions from mobile combustion are significantly affected by
combustion efficiency and the presence of post-combustion emission controls. Carbon monoxide emissions are
highest when air-fuel mixtures have less oxygen than required for complete combustion. These emissions occur
especially in vehicle idle, low speed, and cold start conditions. Methane and NMVOC emissions from motor
vehicles are a function of the CH4 content of the motor fuel, the amount of hydrocarbons passing uncombusted
through the engine, and any post-combustion control of hydrocarbon emissions (such as catalytic converters).
An alternative method of presenting combustion emissions is to allocate emissions associated with electric power
to the sectors in which it is used. Four end-use sectors are defined: transportation, industrial, residential, and
commercial. In Table 3-11 below, electric power emissions have been distributed to each end-use sector based
upon the sector's share of national electricity use, with the exception of CH4 and N2O from transportation
electricity use.13 Emissions from U.S. Territories are also calculated separately due to a lack of end-use-specific
consumption data.14 This method assumes that emissions from combustion sources are distributed across the four
end-use sectors based on the ratio of electricity use in that sector. The results of this alternative method are
presented in Table 3-11.
Table 3-11: CO2, CH4, and N2O Emissions from Fossil Fuel Combustion by End-Use Sector with
Electricity Emissions Distributed (MMT CO2 Eq.)
End-Use Sector
1990
2005
2018
2019
2020
2021
2022
Transportation
1,517.6
1,904.6
1,838.4
1,843.4
1,595.6
1,778.0
1,776.7
C02
1,472.0 I
1,863.3
1,817.9
1,821.4
1,576.9
1,758.6
1,757.4
ch4
7.2
4.3
2.8
2.9
2.5
2.6
2.6
n2o
38-4 1
37-°
17.7
19.1
16.1
16.8
16.7
Industrial
1,574.8
1,597.0
1,322.4
1,285.0
1,180.8
1,235.2
1,248.2
C02
1,562.9 1
1,584.0
1,311.8
1,275.3
1,171.8
1,225.6
1,238.0
ch4
2.2
2.2
2.1
2.1
2.0
2.0
1.9
n2o
9.7 1
10.8
8.5
7.6
7.1
7.6
8.2
Residential
944.2
1,230.1
995.7
940.2
871.6
902.1
912.9
C02
931.3 ...
1,214.9
981.2
926.7
860.1
890.3
899.4
ch4
6.0
4.9
5.6
5.8
4.2
4.2
4.8
n2o
6.9 1
10-3
8.8
7.7
7.3
7.7
8.7
Commercial
773.1
1,040.9
861.0
813.1
717.6
764.6
791.6
co2
766.0
1,030.1 :
851.3
804.4
709.6
756.1
782.0
ch4
1.3
1.5
1.8
1.9
1.8
1.8
1.8
n2o
5.7
9.3
7.8
6.8
6.2
6.7
7.7
U.S. Territories3
20.1
52.1
26.0
24.9
23.4
23.9
22.7
Total
4,829.9
5,824.8
5,043.4
4,906.6
4,388.9
4,703.7
4,752.0
a U.S. Territories are not apportioned by sector, and emissions are total greenhouse gas emissions from all fuel combustion
sources.
Notes: Totals may not sum due to independent rounding. Emissions from fossil fuel combustion by electric power are
allocated based on aggregate national electricity use by each end-use sector.
13 Separate calculations are performed for transportation-related CH4 and N20. The methodology used to calculate these
emissions is discussed in the Mobile Combustion section.
14 U.S. Territories (including American Samoa, Guam, Puerto Rico, U.S. Virgin Islands, Wake Island, and other outlying U.S.
Pacific Islands) consumption data obtained from EIA are only available at the aggregate level and cannot be broken out by end-
use sector. The distribution of emissions to each end-use sector for the 50 states does not apply to territories data.
3-16 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
Electric Power Sector
The process of generating electricity is the largest stationary source of CO2 emissions in the United States,
representing 30.3 percent of total CO2 emissions from all CO2 emissions sources across the United States. Methane
and N2O accounted for a small portion of total greenhouse gas emissions from electric power, representing 0.1
percent and 1.4 percent, respectively. Electric power also accounted for 32.6 percent of CO2 emissions from fossil
fuel combustion in 2022. Methane and N2O from electric power represented 11.4 and 52.2 percent of total CH4
and N2O emissions from fossil fuel combustion in 2022, respectively.
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 energy use and emissions associated with the electric power sector are included here. The
electric power sector consists of electric utilities and independent power producers whose primary business is the
production of electricity. This includes both regulated utilities and non-utilities (e.g., independent power
producers, qualifying co-generators, and other small power producers). Energy use and emissions associated with
electric generation in the commercial and industrial sectors is reported in those other sectors where the producer
of the power indicates that its primary business is something other than the production of electricity.15
Total greenhouse gas emissions from the electric power sector have decreased by 15.5 percent since 1990. From
1990 to 2007, electric power sector emissions increased by 33 percent, driven by a significant increase in electricity
demand (39 percent) while the carbon intensity of electricity generated showed a modest decline (2.1 percent).
From 2008 to 2022, as electricity demand increased by 5.2 percent, electric power sector emissions decreased by
35 percent, driven by a significant drop (25 percent) in the carbon intensity of electricity generated. Overall, the
carbon intensity of the electric power sector, in terms of CO2 Eq. per QBtu, decreased by 28 percent from 1990 to
2022 with additional trends detailed in Box 3-4. This decoupling of electric power generation and the resulting CO2
emissions is shown in Figure 3-9. This recent decarbonization of the electric power sector is a result of several key
drivers.
Coal-fired electric generation (in kilowatt-hours [kWh]) decreased from 54 percent of generation in 1990 to 20
percent in 2022.16 This corresponded with an increase in natural gas generation and renewable energy generation,
largely from wind and solar energy. Natural gas generation (in kWh) represented 11 percent of electric power
generation in 1990 and increased over the 33-year period to represent 39 percent of electric power sector
generation in 2022 (see Table 3-12). Natural gas has a much lower carbon content than coal and is generated in
power plants that are generally more efficient in terms of kWh produced per Btu of fuel combusted, which has led
to lower emissions as natural gas replaces coal-powered electricity generation. Natural gas and coal used in the
United States in 2022 had an average carbon content of 14.43 MMT C/Qbtu and 26.13 MMT C/Qbtu respectively.
15 Utilities primarily generate power for the U.S. electric grid for sale to retail customers. Non-utilities typically generate
electricity for sale on the wholesale electricity market (e.g., to utilities for distribution and resale to retail customers). Where
electricity generation occurs outside the ElA-defined electric power sector, it is typically for the entity's own use.
16 Values represent electricity net generation from the electric power sector (EIA 2024a).
Energy 3-17
-------
Table 3-12: Electric Power Generation by Fuel Type (Percent)
Fuel Type
1990
2005
2018
2019
2020
2021
2022
Coal
54.1% j
51.1%
28.4%
24.2%
19.9%
22.6%
20.3%
Natural Gas
10.7%
17.5%
34.0%
37.3%
39.5%
37.3%
38.8%
Nuclear
19.9%
20.0%
20.1%
20.4%
20.5%
19.7%
18.9%
Renewables
11.3% I
8.3%
16.8%
17.6%
19.5%
19.8%
21.4%
Petroleum
4.1% 1
3.0%
0.6%
0.4%
0.4%
0.5%
0.5%
Other Gases3
+
0.1%
0.1%
0.1%
0.1%
0.1%
0.1%
Net Electricity Generation
(Billion kWh)b
ii:
2,905 J
3,902
4,020
3,966
3,851
3,955
4,076
+ Does not exceed 0.05 percent.
a Other gases include blast furnace gas, propane gas, and other manufactured and waste gases derived from fossil fuels.
b Represents net electricity generation from the electric power sector. Excludes net electricity generation from
commercial and industrial combined-heat-and-power and electricity-only plants. Does not include electricity generation
from purchased steam as the fuel used to generate the steam cannot be determined.
In 2022, CO2 emissions from the electric power sector decreased by 0.6 percent relative to 2021. This decrease in
CO2 emissions was primarily driven by a decrease in coal consumed to produce electricity in the electric power
sector. Consumption of coal for electric power decreased by 6.4 percent while consumption of natural gas
increased 7.6 percent from 2021 to 2022, leading to an overall decrease in emissions. There has also been a rapid
increase in renewable energy electricity generation in the electric power sector in recent years. Electricity
generation from renewable sources increased by 11 percent from 2021 to 2022 (see Table 3-12). A decrease in
coal-fired electricity generation and increases in natural gas and renewable energy sources for electricity
generation contributed to a decoupling of emissions trends from electric power generation trends over the recent
time series (see Figure 3-9).
Decreases in natural gas prices and the associated increase in natural gas generation, particularly between 2005
and 2019, was a primary driver of the fuel switching from using coal to using natural gas for electricity generation,
which led to a significant decrease in CO2 emissions from electricity generation. During this time period, the cost of
natural gas (in $/MMBtu) decreased by 56 percent while the cost of coal (in $/MMBtu) increased by 74 percent
(EIA 2024a). However, from 2020 to 2022, natural gas prices increased 200 percent and are now 9 percent higher
than 2005 levels due to the COVID-19 pandemic and other factors disrupting the domestic and global natural gas
markets. While the increase in natural gas prices led to a temporary trend reversal, with coal consumption
increasing and natural gas consumption decreasing from 2020 to 2021, the broader trend of declining coal
consumption for electricity generation continues. From 2021 to 2022, coal consumption decreased 6 percent while
natural gas consumption increased 8 percent.
Moving forward, the shift away from coal—and increasingly towards renewable energy sources in addition to
natural gas—for electricity generation will further contribute to reductions in power sector emissions. Renewable
energy generation (in kWh) from wind and solar energy increased from 0.1 percent of total generation in 1990 to 5
percent in 2015 and increased at a faster pace to 14 percent of total generation in 2022. The decrease in carbon
intensity occurred even as total electricity retail sales increased 45 percent, from 2,713 billion kWh in 1990 to
3,927 billion kWh in 2022.
3-18 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
Figure 3-9: Fuels Used in Electric Power Generation and Total Electric Power Sector CO2
Emissions
50,000
40,000
30,000
1 Nuclear (TBtu)
Renewable Energy Sources (TBtu)
1 Petroleum (TBtu)
Natural Gas (TBtu)
Coal (TBtu)
1 Net Generation (Index from 1990) [Right Axis]
1 Sector CO2 Emissions (Index from 1990) [Right Axis]
160
140
120
100
80
60
40
20
0
aioi^o^oioi^oicnoiooooooo
o->cr>cr>chc^cr>cr>cT»ooooooo
COO^OHOJrOtmiDNCOOiOHfN
O O rH tH t—I i-H t-H t-H t—t 1—< rH tH fNI fN ("Nl
OOOOOOOOOOOOOOO
(N(NfMfMrMfNfM(N(NfNfNrMfM(N(N
Electricity was used primarily in the residential, commercial, and industrial end-use sectors for lighting, heating,
electric motors, appliances, electronics, and air conditioning (see Figure 3-10). Note that transportation is an end-
use sector as well but is not shown in Figure 3-10 due to the sector's relatively low percentage of electricity use.
The Transportation Sector and Mobile Combustion section provides a break-out of CO2 emissions from electricity
use in the transportation end-use sector.
Figure 3-10: Electric Power Retail Sales by End-Use Sector
In 2022, electricity sales to the residential and commercial end-use sectors, as presented in Figure 3-10, increased
by 2.6 percent and 4.7 percent relative to 2021, respectively. Electricity sales to the industrial sector in 2022
Energy 3-19
-------
increased by approximately 2.0 percent relative to 2021. The sections below describe end-use sector energy use in
more detail. Overall, in 2022, the amount of electricity retail sales (in kWh) increased by 3.2 percent relative to
2021.
Industrial Sector
Industrial sector CO2, Cm, and N2O emissions accounted for 17,14, and 5 percent of CO2, Cm, and N2O emissions
from fossil fuel combustion, respectively, in 2022. Carbon dioxide, Cm, and N2O emissions resulted from the direct
consumption of fossil fuels for steam and process heat production.
The industrial end-use sector, per the underlying energy use data from EIA, includes activities such as
manufacturing, construction, mining, and agriculture. The largest of these activities in terms of energy use is
manufacturing, of which six industries—petroleum refineries, chemicals, paper, primary metals, food, and
nonmetallic mineral products—represent the majority of the energy use (EIA 2024a; EIA 2009b).
There are many dynamics that impact emissions from the industrial sector including economic activity, changes in
the make-up of the industrial sector, changes in the emissions intensity of industrial processes, and weather-
related impacts on heating and cooling of industrial buildings.17 Structural changes within the U.S. economy that
lead to shifts in industrial output away from energy-intensive manufacturing products to less energy-intensive
products (e.g., from steel to computer equipment) have had a significant effect on industrial emissions.
From 2021 to 2022, total industrial production and manufacturing output increased by 3.4 percent (FRB 2022).
Over this period, output increased slightly across production indices for Food, Nonmetallic Mineral Products,
Paper, Petroleum Refineries, and Primary Metals. Production of chemicals declined slightly between 2021 and
2022 (see Figure 3-11). From 2021 to 2022, total energy use in the industrial sector increased by 2.0 percent,
driven mainly by a 2.6 percent increase in fossil fuel consumption in the industrial sector. Consumption of
renewables decreased 1.6 percent from 2021 to 2022. Due to the relative increases and decreases of individual
indices there was an increase in natural gas and an increase in electricity used by this sector (see Figure 3-12). In
2022, CO2, Cm, and N2O emissions from fossil fuel combustion and electricity use within the industrial end-use
sector totaled 1,248.2 MMT CO2 Eq., a 1.1 percent increase from 2021 emissions.
Through EPA's Greenhouse Gas Reporting Program (GHGRP), specific industrial sector trends can be discerned
from the overall total EIA industrial fuel consumption data used for these calculations. For example, from 2021 to
2022, the underlying EIA data showed increased consumption of natural gas and petroleum and decreased
consumption of coal in the industrial sector. The GHGRP data highlights that several industries contributed to
these trends, including chemical manufacturing; pulp, paper and print; food processing, beverages and tobacco;
minerals manufacturing; and agriculture-forest-fisheries.18
17 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.
18 Further details on industrial sector combustion emissions are provided by EPA's GHGRP. See
http://ehedata.epa.gov/ehep/main.do.
3-20 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
140
120
100
80
60
Primary Metals
Petroleum Refineries
O
CT^
o>
i-irNc">*rLn^Df^coc^o*-Hrsjro^-Ln^or^cocr>OT-Hr\iro^-Ln^or>*cocr>Oi-Hr\i
o^cr»o>»a^o^cTia»cricr>cncr>cno^ooooooooooooooooooooooo
HHHHrHHHHHfNrMf\fNfNrMNfMfNfNNfNfvJrMf\fMfNfM(N(NfMfNl(N
Figure 3-12: Fuels and Electricity Used in Industrial Sector, Industrial Output, and Total Sector
CO2 Emissions (Including Electricity)
30,000
25,000
180
160
140
9i
20,000
15,000
10,000
5,000
120
100
80
60
40
20
0
(NnfiflONcoffiOrtMntmioNowOHN
000 00000000000000000000
(\rMWN{N(NfNNry«NINNfNn)IN(NNNNfNNM
Figure 3-11: Industrial Production Indices (Index 2017=100)
Total Industrial excluding Computers, Communications
Equipment, and Semiconductors
Total Industrial
¦ Renewable Energy Sources (TBtu)
¦ Coal (TBtu)
¦ Petroleum (TBtu)
¦ Natural Gas
¦ Electricity Use (TBtu)
™ Industrial Output (Index vs. 1990) [Right Axis]
~ Sector CO2 Emissions (Index vs. 1990) [Right Axis]
Energy 3-21
-------
Despite the growth in industrial output (65 percent) and the overall U.S. economy (114 percent) from 1990 to
2022, direct CO2 emissions from fossil fuel combustion in the industrial sector decreased by 8.6 percent over the
same time series. A number of factors are assumed to result in decoupling of growth in industrial output from
industrial greenhouse gas emissions, for example: (1) more rapid growth in output from less energy-intensive
industries relative to traditional manufacturing industries, and (2) energy-intensive industries such as steel are
employing new methods, such as electric arc furnaces, that are less carbon-intensive than the older methods.
Box 3-3: Uses of Greenhouse Gas Reporting Program Data and Improvements in Reporting
Emissions from Industrial Sector Fossil Fuel Combustion
As described in the calculation methodology, total fossil fuel consumption for each year is based on aggregated
end-use sector consumption published by the EIA. The availability of facility-level combustion emissions through
EPA's GHGRP has provided an opportunity to better characterize the industrial sector's energy consumption and
emissions in the United States, through a disaggregation of ElA's industrial sector fuel consumption data from
select industries.
For GHGRP 2010 through 2022 reporting years, facility-level fossil fuel combustion emissions reported through
EPA's GHGRP were categorized and distributed to specific industry types by utilizing facility-reported NAICS
codes (as published by the U.S. Census Bureau). As noted previously in this report, the definitions and provisions
for reporting fuel types in EPA's GHGRP include some differences from the Inventory's use of EIA national fuel
statistics to meet Paris Agreement and UNFCCC reporting guidelines. The IPCC has provided guidance on
aligning facility-level reported fuels and fuel types published in national energy statistics, which guided this
exercise.19
As with previous Inventory reports, the current effort represents an attempt to align, reconcile, and coordinate
the facility-level reporting of fossil fuel combustion emissions under EPA's GHGRP with the national-level
approach presented in this report. Consistent with recommendations for reporting the Inventory to the
UNFCCC, progress was made on certain fuel types for specific industries and has been included in the Common
Reporting Tables (CRTs) that are submitted to the UNFCCC along with this report.20 The efforts in reconciling
fuels focus on standard, common fuel types (e.g., natural gas, distillate fuel oil) where the fuels in ElA's national
statistics aligned well with facility-level GHGRP data. For these reasons, the current information presented in the
CRTs should be viewed as an initial attempt at this exercise. Additional efforts will be made for future Inventory
reports to improve the mapping of fuel types and examine ways to reconcile and coordinate any differences
between facility-1 eve I data and national statistics. The current analysis includes the full time series presented in
the CRTs. Analyses were conducted linking GHGRP facility-level reporting with the information published by EIA
in its MECS data in order to disaggregate the full 1990 through 2022 time period in the CRTs. It is believed that
the current analysis has led to improvements in the presentation of data in the Inventory, but further work will
be conducted, and future improvements will be realized in subsequent Inventory reports. This includes
incorporating the latest MECS data as it becomes available.
Residential and Commercial Sectors
Emissions from the residential and commercial sectors have generally decreased since 2005. Short-term trends are
often correlated with seasonal fluctuations in energy use caused by weather conditions, rather than prevailing
economic conditions. Population growth and a trend towards larger houses has led to increasing energy use over
19 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 http://www.ipcc-
nggip.iges.or.jp/public/tb/TFI Technical Bulletin l.pdf.
20 See https://www.epa.gov/ghgemissions/inventorv-us-greenhouse-gas-emissions-and-sinks.
3-22 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
the time series, while population migration to warmer areas and improved energy efficiency and building
insulation have slowed the increase in energy use in recent years. Starting in around 2014, energy use and
emissions begin to decouple due to decarbonization of the electric power sector (see Figure 3-13).
Figure 3-13: Fuels and Electricity Used in Residential and Commercial Sectors, Heating and
Cooling Degree Days, and Total Sector CO2 Emissions (Including Electricity)
25,000
20,000
15,000
ai
1/1
z>
10,000
5,000
Coal (TBtu) — Sector CO2 Emissions (Index vs. 1990) [Right Axis]
Renewable Energy Sources (TBtu) — Heating and Cooling Degree Days (Index vs. 1990) [Right Axis]
I Petroleum (TBtu)
Natural Gas
I Electricity Use (TBtu)
180
o»-)rsifn^-Loior^coa>o^Hrsiro^j-LotDP^cocTiOi-HfNf
cricricr)Cr>cr>cricricricricr>ooooooooooi—it—it—it
a\cr>a^cr>cr»a\cria^a^c^ooooooooooooo<
HTHHHrtHHrtHHf\J(NfNN(NrMfN(N(N|(N(M(N(N(NfN(N(MN(NN(NOI(N
In 2022, excluding indirect emissions from electricity use, the residential and commercial sectors accounted for 7
and 6 percent of CO2 emissions from fossil fuel combustion, respectively; 39 and 13 percent of CFU emissions from
fossil fuel combustion, respectively; and 2 and 1 percent of N2O emissions from fossil fuel combustion,
respectively. Emissions from these sectors are primarily attributable to building -related activities such as the
direct consumption of natural gas and petroleum products, primarily for heating and cooking needs. Coal
consumption was a minor component of energy use in the commercial sector and did not contribute to any energy
use in the residential sector. Greenhouse gas emissions from commercial and residential buildings also increase
substantially when emissions from electricity end-use are included, because the building sector uses 75 percent of
the electricity generated in the United States (e.g., for heating, ventilation, and air conditioning; lighting; and
appliances) (NREL 2023). In 2022, total emissions (CO2, Cm, and N2O) from fossil fuel combustion and electricity
use within the residential and commercial end-use sectors were 912.9 MMT CO2 Eq. and 791.6 MMT CO2 Eq.,
respectively. Total CO2, Cm, and N2O emissions from combined fossil fuel combustion and electricity use within the
residential and commercial end-use sectors increased by 1.2 and 3.5 percent from 2021 to 2022, respectively. An
increase in heating degree days (7.9 percent) and cooling degree days (4.3 percent) increased energy demand for
heating and cooling in the residential and commercial sectors. This resulted in a 2.6 percent increase in residential
sector electricity use. From 2021 to 2022 there was a 8.4 percent higher direct energy use in the commercial
sector. In addition, a shift toward energy efficient products and more stringent energy efficiency standards for
household equipment has contributed to a decrease in energy demand in households (EIA 2022), resulting in a
decrease in energy-related emissions in the residential sector since 1990. In the long term, the residential sector is
also affected by population growth, migration trends toward warmer areas, and changes in total housing units and
building attributes (e.g., larger sizes and improved insulation).
Energy 3-23
-------
In 2022, combustion emissions from natural gas consumption represented 81 and 74 percent of the direct fossil
fuel CO2 emissions from the residential and commercial sectors, respectively. Carbon dioxide emissions from
natural gas combustion in the residential and commercial sectors in 2022 increased by 5.2 percent and 6.6 percent
from 2021, respectively.
U.S. Territories
Emissions from U.S. Territories are based on the fuel consumption in American Samoa, Guam, Puerto Rico, U.S.
Virgin Islands, Wake Island, and other outlying U.S. Pacific Islands. As described in the Methodology section of CO2
from Fossil Fuel Combustion, this data is collected separately from the sectoral-level data available for the general
calculations. As sectoral information is not available for U.S. Territories, CO2, CFU, and N2O emissions are not
presented for U.S. Territories in the tables above by sector, though the emissions will occur across all sectors and
sources including stationary, transportation and mobile combustion sources.
Transportation Sector and Mobile Combustion
This discussion of transportation emissions follows the alternative method of presenting combustion emissions by
allocating emissions associated with electricity generation to the transportation end-use sector, as presented in
Table 1-9. Table 1-8 presents direct CO2, CFU, and N2O emissions from all transportation sources (i.e., excluding
emissions allocated to electricity consumption in the transportation end-use sector).
The transportation end-use sector and other mobile combustion accounted for 1,776.7 MMT CO2 Eq. in 2022,
which represented 37 percent of CO2 emissions, 23 percent of CFU emissions, and 41 percent of N2O emissions
from fossil fuel combustion, respectively.21 Fuel purchased in the U.S. for international aircraft and marine travel
accounted for an additional 98.9 MMT CO2 Eq. in 2022; these emissions are recorded as international bunkers and
are not included in U.S. totals according to Paris Agreement and UNFCCC reporting protocols.
Transportation End-Use Sector
From 1990 to 2019, transportation emissions from fossil fuel combustion rose by 21 percent, followed by a
reduction of 13 percent from 2019 to 2020, followed by an increase of 12 percent from 2020 to 2022. Overall,
from 1990 to 2022, transportation emissions from fossil fuel combustion increased by 17 percent. The increase in
transportation emissions from fossil fuel combustion from 1990 to 2022 was due, in large part, to increased
demand for travel (see Figure 3-14). The number of vehicle miles traveled by light-duty motor vehicles (passenger
cars and light-duty trucks) increased 47 percent from 1990 to 2022, as a result of a confluence of factors including
population growth, economic growth, urban sprawl, and low fuel prices over much of this period. Between 2019
and 2020, emissions from light-duty vehicles fell by 11 percent, primarily the result of the COVID-19 pandemic and
associated restrictions, such as people working from home and traveling less. Light-duty vehicle VMT rebounded in
2022 but is still estimated to be 2 percent below 2019 levels.
Commercial aircraft emissions decreased by 5 percent between 2019 and 2022 but have decreased 7 percent since
2007 (FAA 2024 and DOT 1991 through 2023).22 Decreases in jet fuel emissions (excluding bunkers) started in 2007
due in part to improved operational efficiency that results in more direct flight routing, improvements in aircraft
and engine technologies to reduce fuel burn and emissions, and the accelerated retirement of older, less fuel-
efficient aircraft; however, the sharp decline in commercial aircraft emissions from 2019 to 2020 and their gradual
recovery since is primarily due to COVID-19 impacts on scheduled passenger air travel.
Almost all of the energy consumed for transportation was supplied by petroleum-based products, with more than
half being related to gasoline consumption in automobiles and other highway vehicles. Other fuel uses, especially
21 Note that these totals include C02, CH4 and N20 emissions from some sources in the U.S. Territories (ships and boats,
recreational boats, non-transportation mobile sources) and CH4 and N20 emissions from transportation rail electricity.
22 Commercial aircraft consists of passenger aircraft, cargo, and other chartered flights.
3-24 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
diesel fuel for freight trucks and jet fuel for aircraft, accounted for the remainder. The primary driver of
transportation-related emissions was CO2 from fossil fuel combustion, which increased by 19 percent from 1990 to
2022. Annex 3.2 presents the total emissions from all transportation and mobile sources, including CO2, N2O, Cm,
and HFCs.
Figure 3-14: Fuels Used in Transportation Sector, On-road VMT, and Total Sector CO2
Emissions
40,000
35,000
30,000
£ 25,000
-
o>
C
20,000
15,000
10,000
5,000
0
I Other Fuels (TBtu)
I Residual Fuel (TBtu)
I Natural Gas (TBtu)
Renewable Energy (TBtu)
Jet Fuel (TBtu)
I Distillate Fuel (TBtu)
I Motor Gasoline (TBtu)
On road VMT (Index vs. 1990) [Left Axis]
1 Sector CO2 Emissions (Index vs. 1990) [Right Axis]
200
180
160
140
120
100
OHNnTUllONCOOl
Oi-irsiro^rLnvDrvoocrio^fNro^-LnvDr^ooa^Oi-irM
OOOOOOOOOOi—It—It-It—It—I t—I t—It-Hi—It—IfNfNfN
^^Cr>00000000000000000000000
fMCNJ(NfM(NOJ(NfVJrM(NrNjrM(N(NfNJ(N(NfNJfM(NrNj(N(N
o
cn
cr>
a;
¦a
c
Notes: Distillate fuel, residual fuel, and jet fuel include adjustments for international bunker fuels. Distillate fuel and motor
gasoline include adjustments for the sectoral allocation of these fuels. Other Fuels includes aviation gasoline and propane.
Source: Information on fuel consumption was obtained from EIA (2022).
Transportation Fossil Fuel Combustion CO2 Emissions
Domestic transportation CO2 emissions increased by 19 percent (285.4 MMT CO2 Eq.) between 1990 and 2022, an
annualized increase of 0.6 percent. This includes a 24 percent increase in CO2 emissions between 1990 and 2019,
followed by a 13 percent decrease in 2020. Carbon dioxide emissions then increased by 11 percent between 2020
and 2022. Among domestic transportation sources, light-duty vehicles (including passenger cars and light-duty
trucks) represented 57 percent of CO2 emissions, medium- and heavy-duty trucks and buses 24 percent,
commercial aircraft 7 percent, and other sources 11 percent. See Table 3-13 for a detailed breakdown of
transportation CO2 emissions by mode and fuel type.
Almost all of the energy consumed by the transportation sector is petroleum-based, including motor gasoline,
diesel fuel, jet fuel, and residual oil. Carbon dioxide emissions from the combustion of ethanol and biodiesel for
transportation purposes, along with the emissions associated with the agricultural and industrial processes
involved in the production of biofuel, are captured in other Inventory sectors.23 Ethanol consumption from the
transportation sector has increased from 0.7 billion gallons in 1990 to 12.9 billion gallons in 2022, while biodiesel
23 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 biofuels are available at EPA's Renewable Fuels Standards website. See https://www.epa.gov/renewable-fuel-
standard-proeram.
Energy 3-25
-------
consumption has increased from 0.01 billion gallons in 2001 to 1.6 billion gallons in 2022." For additional
information, see Section 3.10 on biofuel consumption at the end of this chapter and Table A-74 in Annex 3.2.
Carbon dioxide emissions from passenger cars and light-duty trucks totaled 1,005.5 MMT CO2 in 2022, an increase
of 10 percent (91.2 MMT CO2) from 1990. The increase in CO2 emissions from passenger cars and light-duty trucks
from 1990 to 2022 was due, in large part, to increased demand for travel as fleet-wide light-duty vehicle fuel
economy was relatively stable (average new vehicle fuel economy declined slowly from 1990 through 2004 and
then increased more rapidly from 2005 through 2022). Carbon dioxide emissions from passenger cars and light-
duty trucks peaked at 1,145.5 MMT in 2004, and since then have declined about 12 percent. The decline in new
light-duty vehicle fuel economy between 1990 and 2004 (see Figure 3-15) reflects the increasing market share of
light-duty trucks, which grew from about 30 percent of new vehicle sales in 1990 to 48 percent in 2004. Starting in
2005, average new vehicle fuel economy began to increase while light-duty vehicle VMT grew only modestly for
much of the period. Light-duty vehicle VMT grew by less than one percent or declined each year between 2005
and 2013, and again between 2017 and 2019. VMT grew at faster rates of 1.6 percent from 2014 to 2015, and 1.6
percent from 2015-2016. From 2019 to 2020, light-duty vehicle VMT declined by 11.0 percent due to COVID-19
pandemic; from 2020 to 2022 light-duty vehicle VMT rebounded as a part of the ongoing recovery from the
pandemic, increasing by 9.8 percent.
Average new vehicle fuel economy has improved almost every year since 2005 while the light-duty truck share of
new vehicle sales decreased to about 33 percent of new vehicles in 2009 and has since varied from year to year
between 36 and 63 percent. Since 2014, the light-duty truck share has steadily increased, reaching 62 percent of
new vehicle sales in model year 2022. See Annex 3.2 for data by vehicle mode and information on VMT and the
share of new vehicles (in VMT).
Medium- and heavy-duty truck CO2 emissions increased by 73 percent from 1990 to 2022. This increase was largely
due to a substantial growth in medium- and heavy-duty truck VMT, which increased by 72 percent between 1990
and 2022.
Carbon dioxide from the domestic operation of commercial aircraft increased by 18 percent (19.8 MMT CO2) from
1990 to 2022. Across all categories of aviation, excluding international bunkers, CO2 emissions decreased by 11
percent (20.8 MMT CO2) between 1990 and 2022.24 Carbon dioxide emissions from military aircraft decreased 65
percent between 1990 and 2022. Commercial aircraft CO2 emissions increased 27 percent between 1990 and
2007, dropped 2 percent from 2007 to 2019, dropped another 33 percent from 2019 to 2020, then increased by 30
percent from 2020 to 2022. Overall, this represents a change of approximately 18 percent between 1990 and
2022. Transportation sources also produce CH4 and N2O; these emissions are included in Figure 3-14 and Table
3-15 and in the CH4 and N2O from Mobile Combustion section. Annex 3.2 presents total emissions from all
transportation and mobile sources, including CO2, CFU, N2O, and HFCs.
24 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.
3-26 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
Figure 3-15: Sales-Weighted Fuel Economy of Mew Passenger Cars and Light-Duty Trucks,
1990-2022
Source: EPA (2023).
Figure 3-16: Sales of New Passenger Cars and Light-Duty Trucks, 1990-2022
Source: EPA (2023).
Table 3-13: CO2 Emissions from Fossil Fuel Combustion in Transportation End-Use Sector
(MMT C02 Eq.)
Fuel/Vehicle Type
1990
2005
2018
2019
2020
2021
2022
Gasoline3
958.9
1,150,1
1,097.0
1,086.5
936.9
1,028.7
1,014.5
Passenger Cars
612.8
518.9
382.5
380.0
328.0
360.5
356.0
Ligh t-Duty Trucks
283.6
583.4
667.6
658.6
565.7
619.9
609.5
Medium- and Heavy-Duty
Trucksb
42.8
28.1
26.2
27.0
24.1
27.4
27.9
Buses
2.1
1.1
2.7
2.8
2.5
2.9
2.9
Motorcycles
3.4
4.9
7.3
7.4
6.6
7.4
7.4
Recreational Boatsc
14.3
13.7
10.7
10.7
9.9
10.6
10.8
Energy 3-27
-------
Distillate Fuel Oil (Diesel)3
274.6
472.1
486.6
484.1
455.2
488.1
483.9
Passenger Cars
9.4
2.2
2.8
2.7
2.5
2.7
2.7
Light-Duty Trucks
^r
00
30.4
31.2
31.2
30.2
33.3
33.8
Medium- and Heavy-Duty
1
I
Trucks'5
189.0
357.2
371.5
373.0
353.5
380.1
375.6
Buses
11.11
15.5
20.4
20.7
19.8
21.4
21.4
Rail
35.5
46.1
38.5
36.0
31.2
32.5
32.5
Recreational Boats0
2.7
2.9
2.8
2.9
2.6
2.8
3.0
Ships and Non-Recreational
Boatsd
6.8
8.4
9.3
7.5
7.6
7.8
7.8
International Bunker Fuelse
11.71
9.5
10.0
10.1
7.8
7.4
7.2
Jet Fuel
222.3
249.5
255.2
258.5
160.4
203.5
231.5
Commercial Aircraft?
109.9 1
132.7
129.6
136.7
91.3
119
129.7
Military Aircraft
35.7
19.8
12.1
12.2
11.7
12.5
12.4
General Aviation Aircraft
38.5 S
36.8
30.6
31.4
17.6
21.1
22.7
International Bunker Fuelse
38.2
60.2
83.0
78.3
39.8
50.8
66.6
International Bunker Fuels
I
mm!
"II"!
from Commercial Aviation
30.0 ¦
55.6
79.8
75.1
36.7
47.6
63.5
Aviation Gasoline
3.1
2.4
1.5
1.6
1.4
1.5
1.5
General Aviation Aircraft
3-11
2.4
1.5
1.6
1.4
1.5
1.5
Residual Fuel Oil
76.3
62.9
45.4
39.7
29.4
46.2
47.3
Ships and Non-Recreational
I
mm!
1
Boatse
22.6 :
19-3:
14.0
14.5
7.3
24.2
22.9
International Bunker Fuelse
53.7
43.6
31.4
25.2
22.1
21.9
24.4
Natural Gas'
36.0
33.1
50.9
58.9
58.7
65.2
70.2
Passenger Cars
+
+
+
+
+
+
+
Light-Duty Trucks
+ 1
+ I
+
+
+
+
+
Medium- and Heavy-Duty
Trucks
+
+
0.1
0.1
0.1
0.1
0.1
Buses
+
0.2
0.5
0.5
0.6
0.6
0.7
Pipeline®
36.0
32.8
50.3
58.3
58.0
64.4
69.3
LPG'
1.4
1.8
0.8
0.8
0.5
0.6
0.6
Passenger Cars
+
+
+
+
+
+
+
Light-Duty Trucks
o.i!
o.i
+
+
+
+
+
Medium- and Heavy-Duty
Trucks'5
1.3
0.9
0.1
0.1
0.1
+
+
Buses
+
0.7 s
0.7
0.7
0.5
0.6
0.6
Electricity1'
3.0
4.7
4.8
4.8
4.1
5.0
6.1
Passenger Cars
+ :
+ !
*
1.2
1.4
1.3
1.8
2.4
Light-Duty Trucks
+
+
0.2
0.2
0.3
0.7
1.1
Buses
+
+
+
0.1
0.1
0.1
Rail
3.0
4.7
3.4
3.1
2.4
2.5
2.5
Total
1,472.0
1,863.3
1,817.9
1,821.4
1,576.9
1758.6
1,757.4
International Bunker Fuels
103.6
113.3
124.3
113.6
69.6
80.2
98.2
Biofuels-Ethanolh
4.1 I
216 j
78.6
78.7
68.1
75.4
75.0
Biofuels-Biodieselh
0.0
0.9
17.9
17.1
17.7
16.1
15.6
+ Does not exceed 0.05 MMT C02 Eq.
a On-road fuel consumption data from FHWA Table MF-21 and MF-27 were used to determine total on-road use of motor
gasoline and diesel fuel (FHWA 1996 through 2023). Ratios developed from MOVES3 output are used to apportion FHWA
fuel consumption data to vehicle type and fuel type (see Annex 3.2 for information about the MOVES model).
b Includes medium- and heavy-duty trucks over 8,500 lbs.
c In 2014, EPA incorporated the NONROAD2008 model into the MOVES model framework. The current Inventory uses the
Nonroad component of MOVES3 for years 1999 through 2022. See Annex 3.2 for information about the MOVES model.
d Note that large year over year fluctuations in emission estimates partially reflect nature of data collection for these sources.
e Official estimates exclude emissions from the combustion of both aviation and marine international bunker fuels; however,
estimates of international bunker fuel-related emissions are presented for informational purposes.
f Commercial aircraft, as modeled in FAA's Aviation Environmental Design Tool (AEDT), consists of passenger aircraft, cargo,
and other chartered flights.
3-28 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
g Pipelines reflect C02 emissions from natural gas-powered pipelines transporting natural gas.
h Ethanol and biodiesel estimates are presented for informational purposes only. See Section 3.10 of this chapter and the
estimates in Land Use, Land-Use Change, and Forestry (see Chapter 6), in line with IPCC methodological guidance and Paris
Agreement and UNFCCC reporting obligations, for more information on ethanol and biodiesel.
' Transportation sector natural gas and LPG consumption are based on data from EIA (2023b). Prior to the 1990 to 2015
Inventory, data from DOE TEDB were used to estimate each vehicle class's share of the total natural gas and LPG
consumption. Since TEDB does not include estimates for natural gas use by medium and heavy-duty trucks or LPG use by
passenger cars, EIA Alternative Fuel Vehicle Data (Browning 2017) is now used to determine each vehicle class's share of the
total natural gas and LPG consumption. These changes were first incorporated in the 1990 to 2016 Inventory and apply to
the 1990 to 2022 time period.
j Includes emissions from rail electricity.
k Electricity consumption by passenger cars, light-duty trucks (SUVs), and buses is based on plug-in electric vehicle sales and
engine efficiency data, as outlined in Browning (2018). In prior Inventory years, C02 emissions from electric vehicle charging
were allocated to the residential and commercial sectors. They are now allocated to the transportation sector. These
changes apply to the 2010 through 2022 time period.
Notes: This table does not include emissions from non-transportation mobile sources, such as agricultural equipment and
construction/mining equipment; it also does not include emissions associated with electricity consumption by pipelines or
lubricants used in transportation. In addition, this table does not include C02 emissions from U.S. Territories, since these are
covered in a separate chapter of the Inventory. Totals may not sum due to independent rounding.
Mobile Fossil Fuel Combustion CH4 and N2O Emissions
Mobile combustion includes emissions of Cm and N2O from all transportation sources identified in the U.S.
Inventory with the exception of pipelines and electric locomotives;25 mobile sources also include non-
transportation sources such as construction/mining equipment, agricultural equipment, vehicles used off-road,
and other sources (e.g., snowmobiles, lawnmowers, etc.).26 Annex 3.2 includes a summary of all emissions from
both transportation and mobile sources. Table 3-14 and Table 3-15 provide mobile fossil fuel Cm and N2O emission
estimates in MMT CO2 Eq.27
Mobile combustion was responsible for a small portion of national CH4 emissions (0.4 percent) and was the fifth
largest source of national N2O emissions (4.3 percent) in 2022. From 1990 to 2022, mobile source Cm emissions
declined by 64 percent, to 2.6 MMT CO2 Eq. (93 kt), due largely to emissions control technologies employed in on-
road vehicles since the mid-1990s to reduce CO, NOx, NMVOC, and CH4 emissions. Mobile source emissions of N2O
decreased by 57 percent, to 16.7 MMT CO2 Eq. (63 kt) in 2022. Earlier generation control technologies initially
resulted in elevated N2O emissions, causing a 31 percent increase in N2O emissions from mobile sources between
1990 and 1997. Improvements in later-generation emission control technologies have reduced N2O output,
resulting in a 67 percent decrease in mobile source N2O emissions from 1997 to 2022 (see Figure 3-17). Overall,
Cm and N2O emissions were predominantly from gasoline-fueled passenger cars, light-duty trucks, and non-
highway sources. See Annex 3.2 for data by vehicle mode and information on VMT and the share of new vehicles.
25 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.
26 See the methodology sub-sections of the C02 from Fossil Fuel Combustion and CH4 and N20 from Mobile Combustion
sections of this chapter. Note that N20 and CH4 emissions are reported using different categories than C02. C02 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. C02 emissions from non-transportation mobile sources (e.g., lawn and garden
equipment, farm equipment, construction equipment) are allocated to their respective end-use sector (i.e., construction
equipment C02 emissions are included in the Industrial end-use sector instead of the transportation end-use sector). CH4 and
N20 emissions are reported using the "mobile combustion" category, which includes non-transportation mobile sources. CH4
and N20 emission 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, C02
emissions from non-transportation mobile sources are presented separately from their overall end-use sector in Annex 3.2.
27 See Annex 3.2 for a complete time series of emission estimates for 1990 through 2022.
Energy 3-29
-------
Figure 3-17: Mobile Source CH4 and N2O Emissions
N20
CH4
HfNMTLniDNCOO^OHfNrOtLniDNOOO^OHtNfOt^iDNCO^OHtN
OICiO^O^O^O^OiOI^OOOOOOOOOOt—It—It—It—It—I ¦*—I ¦*—It—It—It—I fN f\l CN
cr»cr>c^c^cr»c^cr>cr*cr>ooooooooooooooooooooooo
HHHHHHHHHOJ(N(NrNrMrNrNrMrMrMfNrN(NrMrMr\(NrM(N(NrMr\irN
Table 3-14: CH4 Emissions from Mobile Combustion (MMT CO2 Eq.)
Fuel Type/Vehicle Type3
1990
2005
2018
2019
2020
2021
2022
Gasoline On-Roadb
5.8
2.4
0.9
1.0
0.8
0.8
0.7
Passenger Cars
3.8
1.2
0.3
0.3
0.2
0.2
0.2
Light-Duty Trucks
1.5
1.0
0.5
0.6
0.5
0.5
0.5
Medium- and Heavy-Duty Trucks
and Buses
0.5
0.1
+
+
+
+
+
Motorcycles
+
+
+
+
+
+
+
Diesel On-Roadb
+
+
0.1
0.1
0.1
0.1
0.1
Passenger Cars
+
+
+
+
+
+
+
Light-Duty Trucks
+
+
+
+
+
+
+
Medium- and Heavy-Duty Trucks
+
+
0.1
0.1
0.1
0.1
0.1
Medium- and Heavy-Duty Buses
+
+
+
+
+
+
+
Alternative Fuel On-Road
+
+
0.1
+
+
+
+
Non-Roadc
1.4
1.8
1.7
1.7
1.6
1.6
1.7
Ships and Boats
0.4
0.5
0.5
0.4
0.4
0.5
0.5
Raild
0.1
0.1
0.1
0.1
0.1
0.1
0.1
Aircraft
0.1
0.1
+
+
+
+
+
Agricultural Equipment6
0.2
0.2
0.1
0.1
0.1
0.1
0.1
Construction/Mining Equipment'
0.2
0.3
0.2
0.2
0.2
0.2
0.2
Others
0.5
0.7
0.8
0.8
0.8
0.7
0.8
Total
7.2
4.3
2.8
2.9
2.5
2.6
2.6
+ Does not exceed 0.05 MMT C02 Eq.
a See Annex 3.2 for definitions of on-road vehicle types.
b Gasoline and diesel highway vehicle mileage estimates are based on data from FHWA Highway Statistics Table VM-1.
VMT estimates from FHWA are allocated to vehicle type using ratios of VMT per vehicle type to total VMT, derived
from EPA's MOVES3 model (see Annex 3.2 for information about the MOVES model).
c Nonroad fuel consumption estimates for 2020 are adjusted to account for the COVID-19 pandemic and associated
restrictions. For agricultural equipment and airport equipment, sector specific adjustment factors were applied to
the 2019 data. For all other sectors, a 7.7 percent reduction factor is used, based on transportation diesel use (EIA
2023b).
d Rail emissions do not include emissions from electric powered locomotives. Class II and Class III diesel consumption
data for 2014 to 2021 is estimated by applying the historical average fuel usage per carload factor to the annual
number of carloads.
e Includes equipment, such as tractors and combines, as well as fuel consumption from trucks that are used off-road in
agriculture.
f Includes equipment, such as cranes, dumpers, and excavators, as well as fuel consumption from trucks that are used
off-road in construction.
3-30 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
g "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.
Table 3-15: N2O Emissions from Mobile Combustion (MMT CO2 Eq.)
Fuel Type/Vehicle Type3
1990
2005 !!!!!!!
2018
2019
2020
2021
2022
Gasoline On-Roadb
32.0
28.5
7.0
8.1
6.3
6.0
5.3
Passenger Cars
22-4 1
13.3 1
2.5
2.5
2.0
1.9
1.6
Light-Duty Trucks
8.7
14.0
4.3
5.4
4.2
3.9
3.5
Medium- and Heavy-Duty Trucks
!!!!!!!
mm!
and Buses
0.8 i
1.2;;;;
0.2
0.2
0.1
0.1
0.1
Motorcycles
+
+
0.1
0.1
0.1
0.1
0.1
Diesel On-Roadb
0.2
0.4
3.0
3.2
3.0
3.3
3.5
Passenger Cars
+
+
+
+
+
+
+
Light-Duty Trucks
+ 1
+ i
0.3
0.2
0.2
0.3
0.3
Medium- and Heavy-Duty Trucks
0.2
0.3
2.4
2.7
2.5
2.7
2.9
Medium- and Heavy-Duty Buses
+
¦i
+ ::
0.3
0.3
0.2
0.3
0.3
Alternative Fuel On-Road
+
+
0.2
0.1
0.1
0.1
0.1
Non-Roadc
6.2
8.1
7.5
7.6
6.7
7.4
7.8
Ships and Boats
0.2
0.2
0.2
0.2
0.1
0.3
0.3
Raild
0.2 1
0.3 ¦
0.3
0.2
0.2
0.2
0.2
Aircraft
1.5
1.6
1.4
1.5
1.0
1.3
1.4
Agricultural Equipment6
!-2
1-4 =
1.1
1.1
1.1
1.1
1.1
Construction/Mining Equipment'
1.2
1.9
1.6
1.7
1.6
1.7
1.7
Other5
1.8 I
2.8 i
2.9
2.9
2.7
2.9
3.2
Total
38.4
37.0
17.7
19.1
16.1
16.8
16.7
+ Does not exceed 0.05 MMT C02 Eq.
a See Annex 3.2 for definitions of on-road vehicle types.
b Gasoline and diesel highway vehicle mileage estimates are based on data from FHWA Highway Statistics Table VM-1.
VMT estimates from FHWA are allocated to vehicle type using ratios of VMT per vehicle type to total VMT, derived
from EPA's MOVES3 model (see Annex 3.2 for information about the MOVES model).
c Nonroad fuel consumption estimates for 2020 are adjusted to account for the COVID-19 pandemic and associated
restrictions. For agricultural equipment and airport equipment, sector specific adjustment factors were applied to
the 2019 data. For all other sectors, a 7.7 percent reduction factor is used, based on transportation diesel use (EIA
2023a).
d Rail emissions do not include emissions from electric powered locomotives. Class II and Class III diesel consumption
data for 2014 through 2021 is estimated by applying the historical average fuel usage per carload factor to the
annual number of carloads.
e Includes equipment, such as tractors and combines, as well as fuel consumption from trucks that are used off-road in
agriculture.
f Includes equipment, such as cranes, dumpers, and excavators, as well as fuel consumption from trucks that are used
off-road in construction.
s "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.
Note: Totals may not sum due to independent rounding.
C02 from Fossil Fuel Combustion
Methodology and Time-Series Consistency
CO2 emissions from fossil fuel combustion are estimated in line with a Tier 2 method described by the IPCC in the
2006 IPCC Guidelines for National Greenhouse Gas Inventories (IPCC 2006) Chapter 2, Figure 2.1 decision tree and
Energy 3-31
-------
available data on energy use and country specific fuel carbon contents with some exceptions as discussed below.28
A detailed description of the U.S. methodology is presented in Annex 2.1, and is characterized by the following
steps:
1. Determine total fuel consumption by fuel type and sector. Total fossil fuel consumption for each year is
estimated by aggregating consumption data by end-use sector (e.g., commercial, industrial), primary fuel
type (e.g., coal, petroleum, gas), and secondary fuel category (e.g., motor gasoline, distillate fuel oil). Fuel
consumption data for the United States were obtained directly from the EIA of the U.S. Department of
Energy (DOE), primarily from the Monthly Energy Review (EIA 2024a). EIA data include fuel consumption
statistics from the 50 U.S. states and the District of Columbia, including tribal lands. The EIA does not
include territories in its national energy statistics, so fuel consumption data for territories were collected
separately from ElA's International Energy Statistics (EIA 2024b).29
For consistency of reporting, the IPCC has recommended that countries report energy data using the
International Energy Agency (IEA) reporting convention and/or IEA data. Data in the IEA format are
presented "top down"—that is, energy consumption for fuel types and categories are estimated from
energy production data (accounting for imports, exports, stock changes, and losses). The resulting
quantities are referred to as "apparent consumption." The data collected in the United States by EIA on
an annual basis and used in this Inventory are predominantly from mid-stream or conversion energy
consumers such as refiners and electric power generators. These annual surveys are supplemented with
end-use energy consumption surveys, such as the Manufacturing Energy Consumption Survey, that are
conducted on a periodic basis (every four years). These consumption datasets help inform the annual
surveys to arrive at the national total and sectoral breakdowns for that total.30
Also, note that U.S. fossil fuel energy statistics are generally presented using gross calorific values (GCV)
(i.e., higher heating values). Fuel consumption activity data presented here have not been adjusted to
correspond to international standards, which are to report energy statistics in terms of net calorific values
(NCV) (i.e., lower heating values).31
2. Subtract uses accounted for in the Industrial Processes and Product Use chapter. Portions of the fuel
consumption data for seven fuel categories—coking coal, distillate fuel, industrial other coal, petroleum
coke, natural gas, residual fuel oil, and other oil—were reallocated to the Industrial Processes and Product
Use chapter, as they were consumed during non-energy-related industrial activity. To make these
adjustments, additional data were collected from AISI (2004 through 2021), Coffeyville (2012), U.S. Census
Bureau (2001 through 2011), EIA (2024a, 2023c, 2023d), USAA (2008 through 2021), USGS (1991 through
2020), (USGS 2019), USGS (2014 through 2021a), USGS (2014 through 2021b), USGS (1995 through 2013),
USGS (1995,1998, 2000, 2001, 2002, 2007), USGS (2021a), USGS (1991 through 2015a), USGS (1991
through 2020), USGS (2014 through 2021a), USGS (1991 through 2015b), USGS (2021b), USGS (1991
through 2020).32
3. Adjust for biofuels and petroleum denaturant. Fossil fuel consumption estimates are adjusted downward
to exclude fuels with biogenic origins and avoid double counting in petroleum data statistics. Carbon
28 The IPCC Tier 3B methodology is used for estimating emissions from commercial aircraft.
29 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 22.6 MMT C02 Eq. in 2022.
30 See IPCC Reference Approach for Estimating C02 Emissions from Fossil Fuel Combustion in Annex 4 for a comparison of U.S.
estimates using top-down and bottom-up approaches.
31 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.
32 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.
3-32 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
dioxide emissions from ethanol added to motor gasoline and biodiesel added to diesel fuel 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 LULUCF, therefore, fuel consumption estimates are
adjusted to remove ethanol and biodiesel.33 For the years 1993 through 2008, petroleum denaturant is
currently included in EIA statistics for both natural gasoline and finished motor gasoline. To avoid double
counting, petroleum denaturant is subtracted from finished motor gasoline for these years.34
4. Adjust for exports ofCC>2. Since October 2000, the Dakota Gasification Plant has been exporting CO2
produced in the coal gasification process to Canada by pipeline. Because this CO2 is not emitted to the
atmosphere in the United States, the associated fossil fuel (lignite coal) that is gasified to create the
exported CO2 is subtracted from EIA (2023d) coal consumption statistics that are used to calculate
greenhouse gas emissions from the Energy Sector. The associated fossil fuel is the total fossil fuel burned
at the plant with the CO2 capture system multiplied by the fraction of the plant's total site-generated CO2
that is recovered by the capture system. To make these adjustments, data for CO2 exports were collected
from Environment and Climate Change Canada (2022). A discussion of the methodology used to estimate
the amount of CO2 captured and exported by pipeline is presented in Annex 2.1.
5. Adjust sectoral allocation of distillate fuel oil and motor gasoline. EPA conducted a separate bottom-up
analysis of transportation fuel consumption based on data from the Federal Highway Administration that
indicated that the amount of distillate and motor gasoline consumption allocated to the transportation
sector in the EIA statistics should be adjusted. Therefore, for these estimates, the transportation sector's
distillate fuel and motor gasoline consumption were adjusted to match the value obtained from the
bottom-up analysis. As the total distillate and motor gasoline consumption estimate from EIA are
considered to be accurate at the national level, the distillate and motor gasoline consumption totals for
the residential, commercial, and industrial sectors were adjusted proportionately. The data sources used
in the bottom-up analysis of transportation fuel consumption include AAR (2008 through 2022), Benson
(2002 through 2004), DOE (1993 through 2020), EIA (2007), EIA (2024a), EPA (2022), and FHWA (1996
through 2023).35
6. Adjust for fuels consumed for non-energy uses. U.S. aggregate energy statistics include consumption of
fossil fuels for non-energy purposes. These are fossil fuels that are manufactured into plastics, asphalt,
lubricants, or other products. Depending on the end-use, this can result in storage of some or all of the
carbon contained in the fuel for a period of time. As the emission pathways of carbon used for non-energy
purposes are vastly different than fuel combustion (since the carbon in these fuels ends up in products
instead of being combusted), these emissions are estimated separately in Section3.2. Therefore, the
amount of fuels used for non-energy purposes was subtracted from total fuel consumption. Data on non-
fuel consumption were provided by EIA (2023c).
7. Subtract consumption of international bunker fuels. According to the Paris Agreement and UNFCCC
reporting guidelines emissions from international transport activities, or bunker fuels, should not be
included in national totals. U.S. energy consumption statistics include these bunker fuels (e.g., distillate
fuel oil, residual fuel oil, and jet fuel) as part of consumption by the transportation end-use sector,
however, so emissions from international transport activities were calculated separately following the
same procedures used to calculate emissions from consumption of all fossil fuels (i.e., estimation of
consumption, and determination of carbon content).36 The Office of the Under Secretary of Defense
(Installations and Environment) and the Defense Logistics Agency Energy (DLA Energy) of the U.S.
Department of Defense (DoD) (DLA Energy 2022) supplied data on military jet fuel and marine fuel use.
33 Natural gas energy statistics from EIA (2023a) are already adjusted downward to account for biogas in natural gas.
34 These adjustments are explained in greater detail in Annex 2.1.
35 Bottom-up gasoline and diesel highway vehicle fuel consumption estimates are based on data from FHWA Highway Statistics
Table MF-21, MF-27, and VM-1 (FHWA 1996 through 2021).
36 See International Bunker Fuels section in this chapter for a more detailed discussion.
Energy 3-33
-------
Commercial jet fuel use was estimated based on data from FAA (2024) and DOT (1991 through 2022);
residual and distillate fuel use for civilian marine bunkers was obtained from DOC (1991 through 2022) for
1990 through 2001 and 2007 through 2020, and DHS (2008) for 2003 through 2006.37 Consumption of
these fuels was subtracted from the corresponding fuels totals in the transportation end-use sector.
Estimates of international bunker fuel emissions for the United States are discussed in detail in Section
3.9.
8. Determine the total carbon content of fuels consumed. Total carbon was estimated by multiplying the
amount of fuel consumed by the amount of carbon in each fuel. This total carbon estimate defines the
maximum amount of carbon that could potentially be released to the atmosphere if all of the carbon in
each fuel was converted to CO2. A discussion of the methodology and sources used to develop the carbon
content coefficients are presented in Annexes 2.1 and 2.2.
9. Estimate CO2 Emissions. Total CO2 emissions are the product of the adjusted energy consumption (from
the previous methodology steps 1 through 7), the carbon content of the fuels consumed, and the fraction
of carbon that is oxidized. The fraction oxidized was assumed to be 100 percent for petroleum, coal, and
natural gas based on guidance in IPCC (2006) (see Annex 2.1). Carbon emissions were multiplied by the
molecular-to-atomic weight ratio of CO2 to carbon (44/12) to obtain total CO2 emitted from fossil fuel
combustion in million metric tons (MMT).
10. Allocate transportation emissions by vehicle type. This report provides a more detailed accounting of
emissions from transportation because it is such a large consumer of fossil fuels in the United States. For
fuel types other than jet fuel, fuel consumption data by vehicle type and transportation mode were used
to allocate emissions by fuel type calculated for the transportation end-use sector. Heat contents and
densities were obtained from EIA (2023c) and USAF (1998).38
• For on-road vehicles, annual estimates of combined motor gasoline and diesel fuel consumption by
vehicle category were obtained from FHWA (1996 through 2023); for each vehicle category, the
percent gasoline, diesel, and other (e.g., CNG, LPG) fuel consumption are estimated using data from
EPA's MOVES model and DOE (1993 through 2022).39'40
• For non-road vehicles, activity data were obtained from AAR (2008 through 2022), APTA (2007
through 2022), APTA (2006), BEA (1991 through 2015), Benson (2002 through 2004), BTS (2019
through 2022), DLA Energy (2022), DOC (1991 through 2022), DOE (1993 through 2022), DOT (1991
through 2022), EIA (2009a), EIA (2023a), EIA (2002), EIA (1991 through 2022), EPA (2022),41 and
Gaffney (2007).
37 Data for 2002 were interpolated due to inconsistencies in reported fuel consumption data.
38 For a more detailed description of the data sources used for the analysis of the transportation end use sector see the Mobile
Combustion (excluding C02) and International Bunker Fuels sections of the Energy chapter, Annex 3.2, and Annex 3.8,
respectively.
39 On-road fuel consumption data from FHWA Table MF-21 and MF-27 were used to determine total on-road use of motor
gasoline and diesel fuel (FHWA 1996 through 2020). Data for 2021 is proxied using FHWA Traffic Volume Travel Trends. Ratios
developed from MOVES3 output are used to apportion FHWA fuel consumption data to vehicle type and fuel type (see Annex
3.2 for information about the MOVES model).
40 Transportation sector natural gas and LPG consumption are based on data from EIA (2024a). In previous Inventory years,
data from DOE (1993 through 2022) TEDB was used to estimate each vehicle class's share of the total natural gas and LPG
consumption. Since TEDB does not include estimates for natural gas use by medium- and heavy-duty trucks or LPG use by
passenger cars, EIA Alternative Fuel Vehicle Data (Browning 2017) is now used to determine each vehicle class's share of the
total natural gas and LPG consumption. These changes were first incorporated in the 1990 through 2015 Inventory and apply to
the time period from 1990 to 2015.
41 In 2014, EPA incorporated the NONROAD2008 model into the MOVES model framework (EPA 2022b). The current Inventory
uses the Nonroad component of MOVES3 for years 1999 through 2022.
3-34 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
• For jet fuel used by aircraft, CO2 emissions from commercial aircraft were developed by the U.S.
Federal Aviation Administration (FAA) using a Tier 3B methodology, consistent IPCC (2006) (see
Annex 3.3). Carbon dioxide emissions from other aircraft were calculated directly based on reported
consumption of fuel as reported by EIA. Allocation to domestic military uses was made using DoD
data (see Annex 3.8). General aviation jet fuel consumption is calculated as the remainder of total jet
fuel use (as determined by EIA) nets all other jet fuel use as determined by FAA and DoD. For more
information, see Annex 3.2.
Methodological recalculations were applied to the entire time series to ensure time-series consistency from 1990
through 2022. Due to data availability and sources, some adjustments outlined in the methodology above are not
applied consistently across the full 1990 to 2022 time series. As described in greater detail in Annex 2.1, to align
with ElA's methodology for calculating motor gasoline consumption, petroleum denaturant adjustments are
applied to motor gasoline consumption only for the period 1993 through 2008. In addition to ensuring time-series
consistency, to ensure consistency in reporting between the Inventory and the Canadian National Greenhouse Gas
Inventory, the amount of associated fossil fuel (lignite coal) that is gasified to create the exported CO2 from the
Dakota Gasification Plant is adjusted to align with the Canadian National Greenhouse Gas Inventory (Environment
and Climate Change Canada 2022). This adjustment is explained in greater detail in Annex 2.1. As discussed in
Annex 5, data are unavailable to include estimates of CO2 emissions from any liquid fuel used in pipeline transport
or non-hazardous industrial waste incineration, but those emissions are assumed to be insignificant.
Box 3-4: Carbon Intensity of U.S. Energy Consumption
The amount of carbon emitted from the combustion of fossil fuels is dependent upon the carbon content of the
fuel and the fraction of that carbon that is oxidized. Fossil fuels vary in their average carbon content, ranging
from about 53 MMT CO2 Eq./QBtu for natural gas to upwards of 95 MMT CO2 Eq./QBtu for coal and petroleum
coke (see Tables A-42 and A-43 in Annex 2.1 for carbon contents of all fuels). In general, the carbon content per
unit of energy of fossil fuels is the highest for coal products, followed by petroleum, and then natural gas. The
overall carbon intensity of the U.S. economy is thus dependent upon the quantity and combination of fuels and
other energy sources employed to meet demand.
Table 3-16 provides a time series of the carbon intensity of direct emissions for each sector of the U.S. economy.
The time series incorporates only the energy from the direct combustion of fossil fuels in each sector. For
example, the carbon intensity for the residential sector does not include the energy from or emissions related to
the use of electricity for lighting, as it is instead allocated to the electric power sector. For the purposes of
maintaining the focus of this section, renewable energy and nuclear energy are not included in the energy totals
used in Table 3-16 in order to focus attention on fossil fuel combustion as detailed in this chapter. Looking only
at this direct consumption of fossil fuels, the residential sector exhibited the lowest carbon intensity, which is
related to the large percentage of its energy derived from natural gas for heating. The carbon intensity of the
commercial sector has predominantly declined since 1990 as commercial businesses shift away from petroleum
to natural gas. The industrial sector was more dependent on petroleum and coal than either the residential or
commercial sectors, and thus had higher carbon intensities over this period. The carbon intensity of the
transportation sector was closely related to the carbon content of petroleum products (e.g., motor gasoline and
jet fuel, both around 70 MMT CO2 Eq./QBtu), which were the primary sources of energy. Lastly, the electric
power sector had the highest carbon intensity due to its heavy reliance on coal for generating electricity.
Table 3-16: Carbon Intensity from Direct Fossil Fuel Combustion by Sector (MMT CO2
Eq./QBtu)
Sector
1990
2005
2018
2019
2020
2021
2022
Residential3
57.4
56.8
55.3
55.3
55.1
55.2
55.2
Commercial3
59.7
57.8
56.0
56.2
56.3
56.2
56.5
Industrial3
64.8
64.6
60.5
60.2
59.6
59.6
59.6
Transportation3
71.1
71.5
71.0
70.9
70.8
70.9
70.8
Electric Powerb
87.3
85.8 |
75.5
72.9
70.5
72.4
70.9
Energy 3-35
-------
U.S. Territories': 73.1 73.4 70.4 70.8 71.7 70.1 71.3
All Sectors0 73.1 73.6 68.3 67.3 66.6 67.0 66.5
a 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. Totals may
not sum due to independent rounding.
For the time period of 1990 through about 2008, the carbon intensity of U.S. energy consumption was fairly
constant, as the proportion of fossil fuels used by the individual sectors did not change significantly over that
time. Starting in 2008 the carbon intensity of U.S. energy consumption has decreased, reflecting the shift from
coal to natural gas in the electric power sector during that time period. Per capita energy consumption
fluctuated little from 1990 to 2007, but then started decreasing after 2007 and, in 2022, was approximately 13.2
percent below levels in 1990 (see Figure 3-18). To differentiate these estimates from those of Table 3-16, the
carbon intensity trend shown in Figure 3-18 and described below includes nuclear and renewable energy EIA
data to provide a comprehensive economy-wide picture of energy consumption. 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 2024).
Figure 3-18: U.S. Energy Consumption and Energy-Related CO2 Emissions Per Capita and
Per Dollar GDP
Carbon intensity estimates were developed using nuclear and renewable energy data from EIA (2023c), EPA
(2010), and fossil fuel consumption data as discussed above and presented in Annex 2.1.
Uncertainty
For estimates of CO2 from fossil fuel combustion, the amount of CO2 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 of CO2
emissions.
Nevertheless, there are uncertainties in the consumption data, carbon content of fuels and products, and carbon
oxidation efficiencies. For example, given the same primary fuel type (e.g., coal, petroleum, or natural gas), the
3-36 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
amount of carbon contained in the fuel per unit of useful energy can vary. For the United States, however, the
impact of these uncertainties on overall CO2 emission estimates is believed to be relatively small. See, for example,
Marland and Pippin (1990). See also Annex 2.2 for a discussion of uncertainties associated with fuel carbon
contents. Recent updates to carbon factors for natural gas and coal utilized the same approach as previous
Inventories with updated recent data, therefore, the uncertainty estimates around carbon contents of the
different fuels as outlined in Annex 2.2 were not impacted and the historic uncertainty ranges still apply.
Although national statistics of total fossil fuel and other energy consumption are relatively accurate, the allocation
of this consumption to individual end-use sectors (i.e., residential, commercial, industrial, and transportation) is
less certain. For example, for some fuels the sectoral allocations are based on price rates (i.e., tariffs), but a
commercial establishment may be able to negotiate an industrial rate or a small industrial establishment may end
up paying an industrial rate, leading to a misallocation of emissions. Also, the deregulation of the natural gas
industry and the more recent deregulation of the electric power industry have likely led to some minor challenges
in collecting accurate energy statistics as firms in these industries have undergone significant restructuring.
To calculate the total CO2 emission estimate from energy-related fossil fuel combustion, the amount of fuel used in
non-energy production processes were subtracted from the total fossil fuel consumption. The amount of CO2
emissions resulting from non-energy related fossil fuel use has been calculated separately and reported in the
Carbon Emitted from Non-Energy Uses of Fossil Fuels section of this report (Section 3.2). These factors all
contribute to the uncertainty in the CO2 estimates. Detailed discussions on the uncertainties associated with
carbon emitted from non-energy uses of fossil fuels can be found within that section of this chapter.
Various sources of uncertainty surround the estimation of emissions from international bunker fuels, which are
subtracted from the U.S. totals (see the detailed discussions on these uncertainties provided in Section 3.9).
Another source of uncertainty is fuel consumption by U.S. Territories. The United States does not collect energy
statistics for its territories at the same level of detail as for the fifty states and the District of Columbia. Therefore,
estimating both emissions and bunker fuel consumption by these territories is difficult.
Uncertainties in the emission estimates presented above also result from the data used to allocate CO2 emissions
from the transportation end-use sector to individual vehicle types and transport modes. In many cases, bottom-up
estimates of fuel consumption by vehicle type do not match aggregate fuel-type estimates from EIA. Further
research is planned to improve the allocation into detailed transportation end-use sector emissions.
The uncertainty analysis was performed by primary fuel type for each end-use sector, using the IPCC-
recommended Approach 2 uncertainty estimation methodology, Monte Carlo stochastic simulation technique,
with @ RISK software. For this uncertainty estimation, the inventory estimation model for CO2 from fossil fuel
combustion was integrated with the relevant variables from the inventory estimation model for International
Bunker Fuels, to realistically characterize the interaction (or endogenous correlation) between the variables of
these two models. About 170 input variables were modeled for CO2 from energy-related fossil fuel combustion
(including about 20 for non-energy fuel consumption and about 20 for International Bunker Fuels).
In developing the uncertainty estimation model, uniform distributions were assumed for all activity-related input
variables and emission factors, based on the SAIC/EIA (2001) report.42 Triangular distributions were assigned for
the oxidization factors (or combustion efficiencies). The uncertainty ranges were assigned to the input variables
based on the data reported in SAIC/EIA (2001) and on conversations with various agency personnel.43
42 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.
43 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.
Energy 3-37
-------
The uncertainty ranges for the activity-related input variables were typically asymmetric around their inventory
estimates; the uncertainty ranges for the emissions factors were symmetric. Bias (or systematic uncertainties)
associated with these variables accounted for much of the uncertainties associated with these variables (SAIC/EIA
2001).44 For purposes of this uncertainty analysis, each input variable was simulated 10,000 times through Monte
Carlo sampling.
The results of the Approach 2 quantitative uncertainty analysis are summarized in Table 3-17. Fossil fuel
combustion CO2 emissions in 2022 were estimated to be between 4,603.1 and 4,905.2 MMT CO2 Eq. at a 95
percent confidence level. This indicates a range of 2 percent below to 4 percent above the 2022 emission estimate
of 4,699.4 MMTCCh Eq.
Table 3-17: Approach 2 Quantitative Uncertainty Estimates for CO2 Emissions from Energy-
Related Fossil Fuel Combustion by Fuel Type and Sector (MMT CO2 Eq. and Percent)
2022 Emission Estimate Uncertainty Range Relative to Emission Estimate-'
(MMT CO ¦ Eq.) (MMT CO.. Eq.) (%)
Lower
Bound
Upper
Bound
Lower
Bound
Upper
Bound
Coalb
898.8
868.6
983.9
-3%
9%
Residential
NO
NO
NO
NO
NO
Commercial
1.4
1.3
1.6
-5%
15%
Industrial
43.0
41.0
49.9
-5%
16%
Transportation
NO
NO
NO
NO
NO
Electric Power
851.5
818.9
933.8
-4%
10%
U.S. Territories
2.9
2.5
3.4
-12%
19%
Natural Gasb
1,706.8
1,687.3
1,784.4
-1%
5%
Residential
272.0
264.4
291.1
-3%
7%
Commercial
192.3
186.9
205.8
-3%
7%
Industrial
510.4
494.9
548.0
-3%
7%
Transportation
70.2
68.2
75.1
-3%
7%
Electric Power
659.3
640.1
693.0
-3%
5%
U.S. Territories
2.7
2.4
3.2
-12%
17%
Petroleumb
2,093.4
1,966.4
2,215.9
-6%
6%
Residential
62.1
58.6
65.4
-6%
5%
Commercial
65.1
61.5
68.5
-5%
5%
Industrial
247.6
193.3
302.0
-22%
22%
Transportation
1,681.1
1,573.2
1,785.6
-6%
6%
Electric Power
20.5
19.7
22.0
-4%
7%
U.S. Territories
17.0
15.7
18.7
-7%
10%
Geothermal
0.4
0.2
1.0
-48%
173%
Electric Power
0.4
0.2
1.0
-48%
173%
Total (including Geothermal)b
4,699.4
4,603.1
4,905.2
-2%
4%
NO (Not Occurring)
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.
Note: Totals may not sum due to independent rounding.
44 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.
3-38 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
QA/QC and Verification
In order to ensure the quality of the CO2 emission estimates from fossil fuel combustion, general (IPCC Tier 1) and
category-specific (Tier 2) Quality Assurance/Quality Control (QA/QC) procedures were implemented consistent
with the U.S. Inventory QA/QC plan outlined in Annex 8. The Tier 2 procedures that were implemented involved
checks specifically focusing on the activity data and methodology used for estimating CO2 emissions from fossil fuel
combustion in the United States. Emission totals for the different sectors and fuels were compared and trends
were investigated to determine whether any corrective actions were needed. Minor corrective actions were taken.
One area of QA/QC and verification is to compare the estimates and emission factors used in the Inventory with
other sources of CO2 emissions reporting. Two main areas and sources of data were considered. The first is a
comparison with the EPA GHGRP combustion data (Subpart C) for stationary combustion sources excluding the
electric power sector. This mainly focused on considering carbon factors for natural gas. The second comparison is
with the EPA Air Markets Program data for electric power production. This considered carbon factors for coal and
natural gas used in electric power production.
The EPA GHGRP collects greenhouse gas emissions data from large emitters including information on fuel
combustion. This excludes emissions from mobile sources and smaller residential and commercial sources, those
emissions are covered under supplier reporting (Subparts MM and NN) and are areas for further research. Fuel
combustion CO2 data reported in 2022 was 2,082.6 MMT CO2. Of that, 1,577.7 MMT CO2 was from electricity
production. Therefore, the non-electric power production fuel combustion reporting was a fraction of the total
covered by the Inventory under fossil fuel combustion. Furthermore, reporters under the GHGRP can use multiple
methods of calculating emissions; one method is to use the default emission factors provided in the rule, while
another is based on a tier 3 approach using their own defined emission factors. Based on data from reporters on
approach used, it was determined that only about 10 percent of natural gas combustion emissions were based on
a tier 3 approach. Given the small sample size compared to the overall Inventory calculations for natural gas
combustion EPA determined it was not reasonable to consider the GHGRP tier 3 natural gas factors at this time. A
more detailed analysis was done on upstream oil and gas natural gas combustion emissions using the GHGRP data
as discussed in Annex 2.2.
EPA collects detailed sulfur dioxide (SO2), nitrogen oxides (NOx), and carbon dioxide (CO2) emissions data and other
information from power plants across the country as part of the Acid Rain Program (ARP), the Cross-State Air
Pollution Rule (CSAPR), the CSAPR Update, and the Revised CSAPR Update (RCU). The CO2 data from these Air
Market Programs (AMP) can be compared to the electric power sector emissions calculated from the Inventory as
shown in Table 3-18 for the three most recent years of data.
Table 3-18: Comparison of Electric Power Sector Emissions (MMT CO2 Eq. and Percent)
CO' Emissions (MMTCO.' Eq.)
% Change
Fuel/Sector
2020
2021
2022
20-21
21-22
Inventory Electric Power Sector
1,439.6
1,540.9
1,531.7
7.0%
-0.6%
Coal
788.2
910.1
851.5
15.5%
-6.4%
Natural Gas
634.8
612.8
659.3
-3.5%
7.6%
Petroleum
16.2
17.7
20.5
9.6%
15.9%
AMP Electric Power Sector
1,430.8
1,531.7
1,520.1
7.1%
-0.8%
Coal
792.7
913.4
858.5
15.2%
-6.0%
Natural Gas
629.4
609.6
652.7
-3.1%
7.1%
Petroleum
8.8
8.7
8.9
-0.8%
2.7%
Note: Totals may not sum due to independent rounding.
In general the emissions and trends from the two sources line up well. There are differences expected based on
coverage and scope of each source. The Inventory covers all emissions from the electric power sector as defined
above. The EPA AMP data covers emissions from electricity generating units of a certain size so in some respects it
could cover more sources (like electric power units at industrial facilities that would be covered under the
industrial sector in the Inventory) and not as many sources (since smaller units are excluded). The EPA AMP data
also includes heat input for different fuel types. That data can be combined with emissions to calculate implied
Energy 3-39
-------
emission factors.45 The following Table 3-19 shows the implied emissions factors for coal and natural gas from the
EPA AMP data compared to the factors used in the Inventory for the three most recent years of data.
Table 3-19: Comparison of Emissions Factors (MMT Carbon/QBtu)
Fuel Type
2020 2021 2022
EPA AMP
Coal
25.67
14.56
25.66
14.60
25.53
14.61
Natural Gas
EPA Inventory
Electric Power Coal
Natural Gas
26.12
14.43
26.13
14.43
26.13
14.43
The factors for natural gas line up reasonably well, the EPA factors are roughly 1 percent lower than those
calculated from the EPA AMP data. For coal the EPA emissions factors are roughly 2 percent higher than those
calculated from the EPA AMP data. One possible reason for the difference is that the EPA Inventory factors are
based on all coal used in electric power production while the factors from the EPA AMP data are based on units
where coal is the source of fuel used. There are units that use coal and other fuel sources but emissions for each
fuel type could not be calculated. This is an area of further research but given current data available the approach
to develop carbon factors as outlined in Annex 2 is still felt to be the most appropriate to represent total fuel
combustion in the United States.
The Paris Agreement and the UNFCCC reporting guidelines also require countries to complete a "top-down"
reference approach for estimating CO2 emissions from fossil fuel combustion in addition to their "bottom-up"
sectoral methodology. The reference approach (detailed in Annex 4) uses alternative methodologies and different
data sources than those contained in this section of the report. The reference approach estimates fossil fuel
consumption by adjusting national aggregate fuel production data for imports, exports, and stock changes rather
than relying on end-user consumption surveys. The reference approach assumes that once carbon-based fuels are
brought into a national economy, they are either saved in some way (e.g., stored in products, kept in fuel stocks, or
left unoxidized in ash) or combusted, and therefore the carbon in them is oxidized and released into the
atmosphere. In the reference approach, accounting for actual consumption of fuels at the sectoral or sub-national
level is not required. One difference between the two approaches is that emissions from carbon that was not
stored during non-energy use of fuels are subtracted from the sectoral approach and reported separately (see
Section 3.2). These emissions, however, are not subtracted in the reference approach. As a result, the reference
approach emission estimates are comparable to those of the sectoral approach, with the exception that the non-
energy use (NEU) source category emissions are included in the reference approach (see Annex 4 for more details).
Recalculations Discussion
EIA (2024a) updated distillate fuel oil consumed by the transportation sector for 2010 and on. This caused
transportation petroleum CO2 emissions to increase by an average annual amount of 0.2 MMT CO2 Eq. (less than
half a percent) for the years 2010 through 2021.
EIA (2024a) updated propane consumed by the industrial sector for 2010 and on, which is used to calculate HGL
(Energy Use) annually variable carbon content coefficients. In addition, EIA (2023b) shifted all 2022 product
supplied of natural gasoline and unfinished oils to crude oil transfers. This change was made to reflect that natural
gasoline and unfinished oils are used as feedstocks in crude oil production instead of directly consumed as an end-
use fuel. EPA made the same adjustment across the timeseries. This change impacted industrial energy
45 These emission factors can be converted from MMT Carbon/QBtu to MMT C02 Eq./QBtu by multiplying the emission factor
by 44/12, the molecular-to-atomic weight ratio of C02 to C. This would assume the fraction oxidized to be 100 percent, which is
the guidance in IPCC (2006) (see Annex 2.1).
3-40 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
consumption across the timeseries as well as non-energy use consumption, which impacts industrial energy
consumption values. This change also impacted the HGL carbon content coefficient used to calculate emissions.
To better align with EIA methodology, the non-energy use consumption of HGLs is now calculated for the entire
timeseries by assuming that 100 percent of ethane, ethylene, and propylene consumption is for non-combustion
use and 85 percent of normal butane, butylene, isobutane, and isobutylene is for non-combustion use. Non-energy
use consumption of propane is calculated by subtracting the non-energy consumption of all other HGLs from the
total non-combustion consumption of HGLs as published by the EIA. Non-energy use consumption is subtracted
from energy consumption, therefore this methodology change impacts industrial petroleum consumption values.
Additionally, the energy HGL carbon contents are now calculated following the above methodology and have
therefore increased across the timeseries, impacting U.S. Territories petroleum and industrial petroleum CO2
emissions.
Overall, these four updates to EIA (2024a) data and methodology caused U.S. Territories petroleum CO2 emissions
to decrease by an average annual amount of less than 0.1 MMT CO2 Eq. (less than half a percent) for the
timeseries, and industrial petroleum CO2 emissions to increase by an average annual amount of 5.0 MMT CO2 Eq.
(1.8 percent) for the timeseries.
EPA revised its calculation of change in total energy use in the industrial sector to include renewable energy and
electricity. The value previously included only fossil fuel energy consumption. Additionally, EPA revised power
sector carbon intensity data to correct for an error and ensure total renewable energy consumed from ElA's
Monthly Energy Review (EIA 2024a) was being used. There were also very minor updates associated with changes
in residential and commercial petroleum use due to MER updates changes in industrial coal use due to updated
data on CO2 exports.
Overall, these changes resulted in an average annual increase of 5.8 MMT CO2 Eq. (0.1 percent) in CO2 emissions
from fossil fuel combustion for the period 1990 through 2021, relative to the previous Inventory.
Planned Improvements
To reduce the uncertainty of CO2 from fossil fuel combustion estimates for U.S. Territories, further expert
elicitation may be conducted to better quantify the total uncertainty associated with emissions from U.S.
Territories. Additionally, although not technically a fossil fuel, since geothermal energy-related CO2 emissions are
included for reporting purposes, further expert elicitation may be conducted to better quantify the total
uncertainty associated with CO2 emissions from geothermal energy use.
EPA will continue to examine the availability of facility-level combustion emissions through EPA's GHGRP to help
better characterize the industrial sector's energy consumption in the United States and further classify total
industrial sector fossil fuel combustion emissions by business establishments according to industrial economic
activity type. Most methodologies used in EPA's GHGRP are consistent with IPCC methodologies, although for
EPA's GHGRP, facilities collect detailed information specific to their operations according to detailed measurement
standards, which may differ with the more aggregated data collected for the Inventory to estimate total national
U.S. emissions. In addition, and unlike the reporting requirements for this chapter under the Paris Agreement and
UNFCCC reporting guidelines, some facility-level fuel combustion emissions reported under the GHGRP may also
include industrial process emissions.46 In line with the Paris Agreement and UNFCCC reporting guidelines, fuel
combustion emissions are included in this chapter, while process emissions are included in the Industrial Processes
and Product Use chapter of this report. In examining data from EPA's GHGRP that would be useful to improve the
emission estimates for the CO2 from fossil fuel combustion category, particular attention will also be made to
ensure time-series consistency, as the facility-level reporting data from EPA's GHGRP are not available for all
inventory years as reported in this Inventory.
Additional analyses will be conducted to align reported facility-level fuel types and IPCC fuel types per the national
energy statistics. For example, additional work will look at CO2 emissions from biomass to ensure they are
46 See https://unfccc.int/resource/docs/2013/copl9/ene/10a03.pdf#paee=2.
Energy 3-41
-------
separated in the facility-level reported data and maintaining consistency with national energy statistics provided
by EIA. In 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 continue to be relied upon.47
EPA is also evaluating the methods used to adjust for conversion of fuels and exports of CO2. EPA is exploring the
approach used to account for CO2 transport, injection, and geologic storage; as part of this there may be changes
made to the accounting for CO2 exports.
Finally, another ongoing planned improvement is to evaluate data availability to update the carbon and heat
content of more fuel types accounted for in this Inventory. This update will impact consumption and emissions
across all sectors and will improve consistency with EIA data as carbon and heat contents of fuels will be accounted
for as annually variable and therefore improve accuracy across the time series. Some of the fuels considered in this
effort include petroleum coke, residual fuel, and woody biomass.
CH4and N20 from Stationary Combustion
Methodology and Time-Series Consistency
Methane and N2O emissions from stationary combustion were estimated by multiplying fossil fuel and wood
consumption data by emission factors (by sector and fuel type for industrial, residential, commercial, and U.S.
Territories; and by fuel and technology type for the electric power sector). The electric power sector utilizes a Tier
2 methodology, whereas all other sectors utilize a Tier 1 methodology in accordance with IPCC methodological
decision tree Figure 2.1 in the 2006 IPCC Guidelines for National Greenhouse Gas Inventories (IPCC 2006) and
available data. The activity data and emission factors used are described in the following subsections.
More detailed information on the methodology for calculating emissions from stationary combustion, including
emission factors and activity data, is provided in Annex 3.1.
Industrial, Residential, Commercial, and U.S. Territories
National coal, natural gas, fuel oil, and wood consumption data were grouped by sector: industrial, commercial,
residential, and U.S. Territories. For the CH4 and N2O emission estimates, consumption data for each fuel were
obtained from ElA's Monthly Energy Review (EIA 2024a). Because the United States does not include territories in
its national energy statistics, fuel consumption data for territories were provided separately by ElA's International
Energy Statistics (EIA 2024b).48 Fuel consumption for the industrial sector was adjusted to subtract out mobile
source construction and agricultural use, which is reported under mobile sources. Construction and agricultural
mobile source fuel use was obtained from EPA (2022) and FHWA (1996 through 2023). Estimates for wood biomass
consumption for fuel combustion do not include municipal solid waste, tires, etc., that are reported as biomass by
EIA. Non-CC>2 emissions from combustion of the biogenic portion of municipal solid waste and tires is included
under waste incineration (Section 3.2). Estimates for natural gas combustion do not include biogas, and therefore
non-CC>2 emissions from biogas are not included (see the Planned Improvements section, below). Tier 1 default
emission factors for the industrial, commercial, and residential end-use sectors were provided by the 2006 IPCC
Guidelines for National Greenhouse Gas Inventories (IPCC 2006). U.S. Territories' emission factors were estimated
using the U.S. emission factors for the primary sector in which each fuel was combusted.
Electric Power Sector
The electric power sector uses a Tier 2 emission estimation methodology as fuel consumption for the electric
47 See http://www.ipcc-nggip.iges.or.jp/public/tb/TFl Technical Bulletin l.pdf.
48 U.S. Territories data also include combustion from mobile activities because data to allocate territories' energy use were
unavailable. For this reason, CH4 and N20 emissions from combustion by U.S. Territories are only included in the stationary
combustion totals.
3-42 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
power sector by control-technology type is based on EPA's Acid Rain Program Dataset (EPA 2024). Total fuel
consumption in the electric power sector from EIA (2024a) was apportioned to each combustion technology type
and fuel combination using a ratio of fuel consumption by technology type derived from EPA (2024) data. The
combustion technology and fuel use data by facility obtained from EPA (2024) were only available from 1996 to
2022 so the consumption estimates from 1990 to 1995 were estimated by applying the 1996 consumption ratio by
combustion technology type from EPA (2024) to the total EIA (2024a) consumption for each year from 1990 to
1995.
Emissions were estimated by multiplying fossil fuel and wood consumption by technology-, fuel-, and country-
specific Tier 2 emission factors. The Tier 2 emission factors used are based in part on emission factors published by
EPA, and EPA's Compilation of Air Pollutant Emission Factors, AP-42 (EPA 1997) for coal wall-fired boilers, residual
fuel oil, diesel oil and wood boilers, natural gas-fired turbines, and combined cycle natural gas units.49
Methodological recalculations were applied to the entire time series to ensure time-series consistency from 1990
through 2022 as discussed below. As discussed in Annex 5, data are unavailable to include estimates of Cm and
N2O emissions from biomass use in Territories, but those emissions are assumed to be insignificant.
Uncertainty
Methane emission estimates from stationary sources exhibit high uncertainty, primarily due to difficulties in
calculating emissions from wood combustion (i.e., fireplaces and wood stoves). The estimates of CH4 and N2O
emissions presented are based on broad indicators of emissions (i.e., fuel use multiplied by an aggregate emission
factor for different sectors), rather than specific emission processes (i.e., by combustion technology and type of
emission control).
An uncertainty analysis was performed by primary fuel type for each end-use sector, using the IPCC-recommended
Approach 2 uncertainty estimation methodology, Monte Carlo stochastic simulation technique, with @RISK
software.
The uncertainty estimation model for this source category was developed by integrating the CH4 and N2O
stationary source inventory estimation models with the model for CO2 from fossil fuel combustion to realistically
characterize the interaction (or endogenous correlation) between the variables of these three models. About 55
input variables were simulated for the uncertainty analysis of this source category (about 20 from the CO2
emissions from fossil fuel combustion inventory estimation model and about 35 from the stationary source
inventory models).
In developing the uncertainty estimation model, uniform distribution was assumed for all activity-related input
variables and N2O emission factors, based on the SAIC/EIA (2001) report.50 For these variables, the uncertainty
ranges were assigned to the input variables based on the data reported in SAIC/EIA (2001).51 However, the CFU
emission factors differ from those used by EIA. These factors and uncertainty ranges are based on IPCC default
uncertainty estimates (IPCC 2006).
49 Several of the U.S. Tier 2 emission factors were used in IPCC (2006) as Tier 1 emission factors. See Table A-67 in Annex 3.1 for
emission factors by technology type and fuel type for the electric power sector.
50 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.
51 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.
Energy 3-43
-------
The results of the Approach 2 quantitative uncertainty analysis are summarized in Table 3-20. Stationary
combustion Cm emissions in 2022 (including biomass) were estimated to be between 5.9 and 19.1 MMT CO2 Eq. at
a 95 percent confidence level. This indicates a range of 31 percent below to 122 percent above the 2022 emission
estimate of 8.6 MMT CO2 Eq.52 Stationary combustion N2O emissions in 2022 (including biomass) were estimated
to be between 16.5 and 33.4 MMT CO2 Eq. at a 95 percent confidence level. This indicates a range of 33 percent
below to 35 percent above the 2022 emission estimate of 24.7 MMT CO2 Eq.
Table 3-20: Approach 2 Quantitative Uncertainty Estimates for CH4 and N2O Emissions from
Energy-Related Stationary Combustion, Including Biomass (MMT CO2 Eq. and Percent)
2022 Emission Estimate
Uncertainty Range Relative to Emission Estimate-1
Source
Gas
(MMT CO . Eq.)
(MMT CO
Eq.)
(%)
Lower
Upper
Lower Upper
Bound
Bound
Bound Bound
Stationary Combustion
ch4
8.6
5.9
19.1
-31% 122%
Stationary Combustion
n2o
24.7
16.5
33.4
-33% 35%
a Range of emission estimates predicted by Monte Carlo stochastic simulation for a 95 percent confidence interval.
The uncertainties associated with the emission estimates of CH4 and N2O are greater than those associated with
estimates of CO2 from fossil fuel combustion, which mainly rely on the carbon content of the fuel combusted.
Uncertainties in both Cm and N2O estimates are due to the fact that emissions are estimated based on emission
factors representing only a limited subset of combustion conditions. For the indirect greenhouse gases,
uncertainties are partly due to assumptions concerning combustion technology types, age of equipment, emission
factors used, and activity data projections.
QA/QC and Verification
In order to ensure the quality of the non-CC>2 emission estimates from stationary combustion, general (IPCC Tier 1)
and category-specific (Tier 2) Quality Assurance/Quality Control (QA/QC) procedures were implemented consistent
with the U.S. Inventory QA/QC plan outlined in Annex 8. The Tier 2 procedures that were implemented involved
checks specifically focusing on the activity data and emission factor sources and methodology used for estimating
Cm, N2O, and the greenhouse gas precursors from stationary combustion in the United States. Emission totals for
the different sectors and fuels were compared and trends were investigated.
Recalculations Discussion
EIA (2024a) updated industrial HGL statistics, which caused CH4 and N20 emissions from industrial fuel oil to
decrease slightly for the years 2010 through 2021.
In addition, EIA (2023b) shifted all 2022 product supplied of natural gasoline and unfinished oils to crude oil
transfers. This change was made to reflect the fact that these fuels are used as feedstocks in crude oil production
instead of directly consumed as end-use fuels. EPA made the change across the time series. This change impacted
energy consumption across the timeseries as well as non-energy use consumption, which impacts energy
consumption values.
To better align with EIA methodology, the non-energy use consumption of HGLs is now calculated for the entire
timeseries by assuming that 100 percent of ethane, ethylene, and propylene consumption is for non-combustion
use and 85 percent of normal butane, butylene, isobutane, and isobutylene is for non-combustion use. Non-energy
use consumption of propane is calculated by subtracting the non-energy consumption of all other HGLs from the
52 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.
3-44 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
total non-combustion consumption of HGLs as published by the EIA. Non-energy use consumption is subtracted
from energy consumption, therefore this methodology change impacts industrial fuel oil consumption values.
Other small updates included changes in residential and commercial/institutional fuel oil are use and changes in
industrial coal due to updated CO2 export data.
Overall, these updates to EIA data and methodology (2024a) caused Cm and N2O emissions from industrial fuel oil
to increase by an average annual amount of 0.01 MMT CO2 Eq. and 0.01 MMT CO2 Eq. (3 percent and 3 percent),
respectively, for the time series. EIA (2024a) updated 2020 and 2021 wood energy consumed by the residential
sector due to new underlying data collected by the Residential Energy Consumption Survey (RECS), which collects
data about once every 5 years and uses Annual Energy Outlook growth rates to estimate data for other years. This
caused Cm and N2O emissions from residential wood consumption to decrease by 0.76 MMT CO2 Eq. and 0.10
MMT CO2 Eq. (22 percent and 13 percent) in 2020, respectively, and 0.94 MMT CO2 Eq. and 0.12 MMT CO2 Eq. (26
percent and 15 percent) in 2021, respectively.
Planned Improvements
Several items are being evaluated to improve the CH4 and N2O emission estimates from stationary combustion and
to reduce uncertainty for U.S. Territories. Efforts will be taken to work with EIA and other agencies to improve the
quality of the U.S. Territories data. Because these data are not broken out by stationary and mobile uses, further
research will be aimed at trying to allocate consumption appropriately. In addition, the uncertainty of biomass
emissions will be further investigated because it was expected that the exclusion of biomass from the estimates
would reduce the uncertainty; and in actuality the exclusion of biomass increases the uncertainty. These
improvements are not all-inclusive but are part of an ongoing analysis and efforts to continually improve these
stationary combustion estimates from U.S. Territories.
Other forms of biomass-based gas consumption include biogas. As an additional planned improvement, EPA will
examine EIA and GHGRP data on biogas collected and burned for energy use and determine if CH4 and N2O
emissions from biogas can be included in future Inventories. EIA (2024a) natural gas data already deducts biogas
used in the natural gas supply, so no adjustments are needed to the natural gas fuel consumption data to account
for biogas.
CH4 and N20 from Mobile Combustion
Methodology and Time-Series Consistency
Estimates of Cm and N2O emissions from mobile combustion were calculated by multiplying emission factors by
measures of activity for each fuel and vehicle type (e.g., light-duty gasoline trucks). Activity data included vehicle
miles traveled (VMT) for on-road vehicles and fuel consumption for non-road mobile sources. The activity data and
emission factors used in the calculations are described in the subsections that follow. A complete discussion of the
methodology used to estimate CH4 and N2O emissions from mobile combustion and the emission factors used in the
calculations is provided in Annex 3.2.
On-Road Vehicles
Estimates of Cm and N2O emissions from gasoline and diesel on-road vehicles are based on VMT and emission
factors (in grams of Cm and N2O per mile) by vehicle type, fuel type, model year, and emission control technology.
Emission estimates for alternative fuel vehicles (AFVs) are based on VMT and emission factors (in grams of CH4 and
N2O per mile) by vehicle and fuel type.53
53 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.
Energy 3-45
-------
CH4 and N2O emissions factors by vehicle type and emission tier for newer (starting with model year 2004) on-road
gasoline vehicles were calculated by Browning (2019) from annual vehicle certification data compiled by EPA. CH4
and N2O emissions factors for older (model year 2003 and earlier) on-road gasoline vehicles were developed by ICF
(2004). These earlier emission factors were derived from EPA, California Air Resources Board (CARB) and
Environment and Climate Change Canada (ECCC) laboratory test results of different vehicle and control technology
types. The EPA, CARB and ECCC tests were designed following the Federal Test Procedure (FTP). The procedure
covers three separate driving segments, since vehicles emit varying amounts of greenhouse gases depending on
the driving segment. These driving segments are: (1) a transient driving cycle that includes cold start and running
emissions, (2) a cycle that represents running emissions only, and (3) a transient driving cycle that includes hot
start and running emissions. For each test run, a bag was affixed to the tailpipe of the vehicle and the exhaust was
collected; the content of this bag was then analyzed to determine quantities of gases present. The emissions
characteristics of driving segment 2 tests were used to define running emissions. Running emissions were
subtracted from the total FTP emissions to determine start emissions. These were then recombined to
approximate average driving characteristics, based upon the ratio of start to running emissions for each vehicle
class from MOBILE6.2, an EPA emission factor model that predicts grams per mile emissions of CO2, CO, HC, NOx,
and PM from vehicles under various conditions.54
Diesel on-road vehicle emission factors were developed by ICF (2006). CH4 and N2O emissions factors for newer
(starting with model year 2007) on-road diesel vehicles (those using engine aftertreatment systems) were
calculated from annual vehicle certification data compiled by EPA.
CH4 and N2O emission factors for AFVs were developed based on the 2022 Greenhouse gases, Regulated
Emissions, and Energy use in Transportation (GREET) model (ANL 2022). For light-duty trucks, EPA used travel
fractions for LDT1 and LDT2 (MOVES Source Type 31 for LDT1 and MOVES Source Type 32 for LDT2; see Annex 3.2
for information about the MOVES model) to determine emission factors. For medium-duty vehicles, EPA used
emission factors for light heavy-duty vocational trucks. For heavy-duty vehicles, EPA used emission factors for long-
haul combination trucks. For buses, EPA used emission factors for transit buses. These values represent vehicle
operations only (tank-to-wheels); upstream well-to-tank emissions are calculated elsewhere in the Inventory.
Biodiesel CH4 emission factors were corrected from GREET values to be the same as CH4 emission factors for diesel
vehicles. GREET overestimated biodiesel CH4 emission factors based upon an incorrect CH4-to-THC ratio for diesel
vehicles with aftertreatment technology.
Annual VMT data for 1990 through 2022 were obtained from the Federal Highway Administration's (FHWA)
Highway Performance Monitoring System database as reported in Highway Statistics (FHWA 1996 through 2023).
VMT estimates were then allocated to vehicle type using ratios of VMT per vehicle type to total VMT, derived from
EPA's MOVES3 model (see Annex 3.2 for information about the MOVES model). This corrects time series
inconsistencies in FHWA definitions of vehicle types (Browning 2022a). VMT for alternative fuel vehicles (AFVs)
were estimated based on Browning (2022b). The age distributions of the U.S. vehicle fleet were obtained from EPA
(2004, 2022), and the average annual age-specific vehicle mileage accumulation of U.S. vehicles were obtained
from EPA (2022).
Control technology and standards data for on-road vehicles were obtained from EPA's Office of Transportation and
Air Quality (EPA 1998, 2022b, and 2023) and Browning (2005). These technologies and standards are defined in
Annex 3.2, and were compiled from EPA (1994a, 1994b, 1998,1999) and IPCC (2006) sources.
Non-Road Mobile Sources
The nonroad mobile category for CH4 and N2O includes ships and boats, aircraft, locomotives, and other mobile
non-road sources (e.g., construction or agricultural equipment). For locomotives, aircraft, ships, and non-
recreational boats, fuel-based emission factors are applied to data on fuel consumption, following the IPCC Tier 1
approach. The Tier 2 approach for these sources would require separate fuel-based emissions factors by
54 Additional information regarding the MOBILE model can be found at https://www.epa.gov/rnoves/deseription-and-historv-
mobile-highwav-vehicle-emission-factor-model.
3-46 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
technology, for which data are not currently available. For other non-road sources, EPA uses the Nonroad
component of the MOVES model to estimate fuel use. Emission factors by horsepower bin are estimated from EPA
engine certification data. Because separate emission factors are applied to specific engine technologies; these non-
road sources utilize a Tier 2 approach.
To estimate CFU and N2O emissions from non-road mobile sources, fuel consumption data were employed as a
measure of activity and multiplied by fuel-specific emission factors (in grams of N2O and CFU per kilogram of fuel
consumed).55 Activity data were obtained from AAR (2008 through 2023), APTA (2007 through 2023), Rail Inc
(2014 through 2022), APTA (2006), BEA (1991 through 2015), Benson (2002 through 2004), Bureau of
Transportation Statistics (BTS; 2023), DLA Energy (2022), DOC (1991 through 2022), DOE (1993 through 2022), DOT
(1991 through 2023), EIA (2002, 2007, 2023a, 2023b), EIA (1991 through 2022), EPA (2022), Esser (2003 through
2004), FAA (2024), FHWA (1996 through 2023),56 Gaffney (2007), FTA (2023), and Whorton (2006 through 2014).
Fuel consumption data for boats and vessels in U.S. Territories data and vessel domestic vessel bunkering is
proxied from 2021 proxy data. Emission factors for non-road modes were taken from IPCC (2006) and Browning
(2020 and 2018).
Uncertainty
A quantitative uncertainty analysis was conducted for the mobile source sector using the IPCC-recommended
Approach 2 uncertainty estimation methodology, Monte Carlo stochastic simulation technique, using @ RISK
software. The uncertainty analysis was performed on 2022 estimates of CFU and N2O emissions, incorporating
probability distribution functions associated with the major input variables. For the purposes of this analysis, the
uncertainty was modeled for the following four major sets of input variables: (1) VMT data, by on-road vehicle and
fuel type, (2) emission factor data, by on-road vehicle, fuel, and control technology type, (3) fuel consumption,
data, by non-road vehicle and equipment type, and (4) emission factor data, by non-road vehicle and equipment
type.
Uncertainty analyses were not conducted for NOx, CO, or NMVOC emissions. Emission factors for these gases have
been extensively researched because emissions of these gases from motor vehicles are regulated in the United
States, and the uncertainty in these emission estimates is believed to be relatively low. For more information, see
Section 3.11. However, a much higher level of uncertainty is associated with CH4 and N2O emission factors due to
limited emission test data, and because, unlike CO2 emissions, the emission pathways of CFU and N2O are highly
complex.
Based on the uncertainty analysis, mobile combustion CH4 emissions from all mobile sources in 2022 were
estimated to be between 2.5 and 3.4 MMT CO2 Eq. at a 95 percent confidence level. This indicates a range of 4
percent below to 30 percent above the corresponding 2022 emission estimate of 2.6 MMT CO2 Eq. Mobile
combustion N2O emissions from mobile sources in 2022 were estimated to be between 15.3 and 20.1MMT CO2 Eq.
at a 95 percent confidence level. This indicates a range of 8 percent below to 20 percent above the corresponding
2022 emission estimate of 16.7 MMT CO2 Eq.
55 The consumption of international bunker fuels is not included in these activity data, but emissions related to the
consumption of international bunker fuels are estimated separately under the International Bunker Fuels source category.
56 This Inventory uses FHWA's Agriculture, Construction, and Commercial/Industrial MF-24 fuel volumes along with the MOVES
model gasoline volumes to estimate non-road mobile source CH4 and N20 emissions for these categories. For agriculture, the
MF-24 gasoline volume is used directly because it includes both non-road trucks and equipment. For construction and
commercial/industrial category gasoline estimates, the 2014 and older MF-24 volumes represented non-road trucks only;
therefore, the MOVES gasoline volumes for construction and commercial/industrial categories are added to the respective
categories in the Inventory. Beginning in 2015, this addition is no longer necessary since the FHWA updated its methods for
estimating on-road and non-road gasoline consumption. Among the method updates, FHWA now incorporates MOVES
equipment gasoline volumes in the construction and commercial/industrial categories.
Energy 3-47
-------
Table 3-21: Approach 2 Quantitative Uncertainty Estimates for CH4 and N2O Emissions from
Mobile Sources (MMT CO2 Eq. and Percent)
2022 Emission
Source Gas Estimate Uncertainty Range Relative to Emission Estimate-'
(MMTCO. Eq.) (MMTCQ. Eq.) (%)
Lower Upper Lower Upper
Bound Bound Bound Bound
Mobile Sources CH4 2.6 2.5 3.4 -4% +30%
Mobile Sources N20 16.7 15.3 20.1 -8% +20%
a Range of emission estimates predicted by Monte Carlo stochastic simulation for a 95 percent confidence
interval.
This uncertainty analysis is a continuation of a multi-year process for developing quantitative uncertainty estimates
for this source category using the IPCC Approach 2 uncertainty estimation methodology. As a result, as new
information becomes available, uncertainty characterization of input variables may be improved and revised. For
additional information regarding uncertainty in emission estimates for Cm and N2O please refer to the Uncertainty
Annex. As discussed in Annex 5, data are unavailable to include estimates of CH4 and N2O emissions from any liquid
fuel used in pipeline transport or some biomass used in transportation sources, but those emissions are assumed
to be insignificant.
QA/QC and Verification
In order to ensure the quality of the emission estimates from mobile combustion, general (IPCC Tier 1) and
category-specific (Tier 2) Quality Assurance/Quality Control (QA/QC) procedures were implemented consistent
with the U.S. Inventory QA/QC plan outlined in Annex 8. The specific plan used for mobile combustion was
updated prior to collection and analysis of this current year of data. The Tier 2 procedures focused on the emission
factor and activity data sources, as well as the methodology used for estimating emissions. These procedures
included a qualitative assessment of the emission estimates to determine whether they appear consistent with the
most recent activity data and emission factors available. A comparison of historical emissions between the current
Inventory and the previous Inventory was also conducted to ensure that the changes in estimates were consistent
with the changes in activity data and emission factors.
Recalculations Discussion
In previous Inventories (1990 through 2020 Inventory and before), on-highway greenhouse gas emissions were
calculated using FHWA fuel consumption and vehicle miles traveled (VMT) data delineated by FHWA vehicle
classes. These fuel consumption estimates were then combined with estimates of fuel shares by vehicle type from
Oak Ridge National Laboratory's Transportation Energy Data Book (TEDB), to develop an estimate of fuel
consumption for each vehicle type in the Inventory (i.e., passenger cars, light-duty trucks, buses, medium- and
heavy-duty trucks, motorcycles). However, in 2011, FHWA changed its methods for estimating 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 in the 1990 through 2008 Inventory
and applied to the time series beginning in 2007. The FHWA methodology update resulted in large changes in VMT
and fuel consumption by vehicle class, leading to a shift in emissions among vehicle classes. For example, FHWA
replaced the vehicle category "Passenger Cars" with "Light-duty Vehicles-Short Wheelbase" and the "Other 2 axle-
4 Tire Vehicles" category was replaced by "Light-duty Vehicles, Long Wheelbase." FHWA changed the definition of
light-duty vehicles to less than 10,000 lbs. GVWR instead of 8,500 lbs. GVWR category updates pushed some
single-unit heavy-duty trucks to the light-duty class. This change in vehicle classification also moved some smaller
trucks and sport utility vehicles from the light truck category to the passenger cars category in this Inventory. These
updates resulted in a disconnect in FHWA VMT and fuel consumption data in the 2006 to 2007 timeframe,
generating a large drop in the light-duty truck VMT and fuel consumption trend lines between 2006 and 2007, and
a corresponding increase in the passenger cars trend lines.
3-48 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
To address this inconsistency in the time series, EPA updated the methodology (starting with the 1990 through
2021 Inventory) to divide FHWA VMT data into vehicle classes and fuel type using distributions from EPA's MOtor
Vehicle Emission Simulator, MOVES. The MOVES model is a nationally recognized model based on vehicle
registration, travel activity, and emission rates that are updated with each model release. MOVES uses forecast
growth factors which provide EPA's best estimate of likely future activity based on historical data (see Annex 3.2
for more information about the MOVES model). Thus, dividing FHWA total VMT data into vehicle class and fuel
type using MOVES ratios provides a more consistent estimate of vehicle activity over the Inventory time series.
MOVES ratios are also used to reallocate FHWA gasoline and diesel fuel use data (Browning 2022a). For this
update, the MOVES3 model was run for calendar years 1990 and 1999 through 2022 for all vehicle types. Calendar
years 1991 through 1998 were linearly interpolated from 1990 and 1999 calendar year MOVES3 outputs. Model
outputs of VMT and fuel consumption were binned by calendar year, MOVES vehicle type, and fuel type; MOVES
vehicle types were then mapped to the vehicle types used in the Inventory. Only outputs of gasoline and diesel fuel
consumption from MOVES3 were used; alternative fuel VMT and fuel consumption outputs are ignored because
they are calculated for the Inventory under a separate methodology. Total gasoline and diesel fuel consumption
values from FHWA were then allocated to Inventory vehicle types using gasoline and diesel fuel consumption ratios
by vehicle type from MOVES3. Similarly, VMT by vehicle type and fuel type was calculated by multiplying the total
VMT from FHWA by VMT ratios by vehicle and fuel type generated by MOVES3. Overall, because total fuel
consumption and VMT values are conserved, the changes in total emissions are small, within 0.1 percent.
Observed differences in total emissions are due to changes in CH4 and N2O emissions, as the methodology for
calculating these non-C02 emissions utilizes more detailed activity data and is therefore sensitive to the re-
allocation of activity data. While total emissions estimates are not significantly impacted by this methodology
update, there are significant changes in the allocation of emissions by vehicle type. The share of emissions
allocated to passenger cars now generally decline through the time series while the share of emissions allocated to
light-duty trucks increase over time.
In addition, the latest version of Argonne National Laboratory's Greenhouse Gas, Regulated Emissions, and Energy
Use in Transportation Model (GREET2022^ provided updated emission factors for all alternative fuel vehicle classes
(ANL 2022). Updated emission factors from GREET2022 were implemented in this Inventory, across the entire time
series.
Additionally, new data from BTS on Amtrak fuel consumption for the time period 2019 through 2022 were
included in this Inventory (BTS 2023).
Planned Improvements
While the data used for this report represent the most accurate information available, several areas for
improvement have been identified.
• Update emission factors for ships and non-recreational boats using residual fuel and distillate fuel.
Develop emission factors for locomotives using ultra-low sulfur diesel and emission factors for aircraft
using jet fuel. The Inventory currently uses IPCC default values for these emission factors.
• Continue to explore potential improvements to estimates of domestic waterborne fuel consumption for
future Inventories. The Inventory estimates for residual and distillate fuel used by ships and boats is based
in part on data on bunker fuel use from the U.S. Department of Commerce. Domestic fuel consumption is
estimated by subtracting fuel sold for international use from the total sold in the United States. Since
2015, all ships travelling within 200 nautical miles of the U.S. coastlines must use distillate fuels, thereby
overestimating the residual fuel used by U.S. vessels and underestimating distillate fuel use in these ships.
Additionally, the EIA has stopped publishing the Fuel Oil and Kerosene Sales report, which reported data
on distillate marine fuel use in the U.S. and the territories. This affects the volume of fuel and emissions
that are allocated to the domestic ships and boats source, although top-down data is still available from
the Monthly Energy Review that will be used to estimate total domestic emission from diesel fuel use.
New data and methods are being explored to improve the diesel ships and boats emissions estimates
going forward.
• Update the analyses to use a forthcoming version of MOVES when it becomes available.
Energy 3-49
-------
3.2 Carbon Emitted from Non-Energy Uses
of Fossil Fuels (CRT Source Category 1A)
In addition to being combusted for energy, fossil fuels are also consumed for non-energy uses (NEU) in the United
States. The fuels used for these purposes are diverse, including natural gas, hydrocarbon gas liquids (HGL),57
asphalt (a viscous liquid mixture of heavy crude oil distillates), petroleum coke (manufactured from heavy oil), and
coal (metallurgical) coke (manufactured from coking coal). The non-energy applications of these fuels are equally
diverse, including feedstocks for the manufacture of plastics, rubber, synthetic fibers and other materials; reducing
agents for the production of various metals and inorganic products; and products such as lubricants, waxes, and
asphalt (IPCC 2006). Emissions from non-energy use of lubricants, paraffin waxes, bitumen/asphalt, and solvents
are reported in the Energy sector, as opposed to the Industrial Processes and Product Use (IPPU) sector, to reflect
national circumstances in its choice of methodology and to increase transparency of this source category's unique
country-specific data sources and methodology (see Box 3-5). In addition, estimates of non-energy use emissions
included here do not include emissions already reflected in the IPPU sector, e.g., fuels used as reducing agents. To
avoid double counting, the "raw" non-energy fuel consumption data reported by EIA are reduced to account for
these emissions already included under IPPU.
Carbon dioxide emissions arise from non-energy uses via several pathways. Emissions may occur during the
manufacture of a product, as is the case in producing plastics or rubber from fuel-derived feedstocks. Additionally,
emissions may occur during the product's lifetime, such as during solvent use. Overall, throughout the time series
and across all uses, about 64 percent of the total carbon consumed for non-energy purposes was stored in
products (e.g., plastics), and not released to the atmosphere; the remaining 36 percent was emitted.
There are several areas in which non-energy uses of fossil fuels are closely related to other parts of this Inventory.
For example, some of the non-energy use products release CO2 at the end of their commercial life when they are
combusted after disposal; these emissions are reported separately within the Energy chapter in the Incineration of
Waste source category. There are also net exports of petrochemical intermediate products that are not completely
accounted for in the EIA data, and the Inventory calculations adjust for the effect of net exports on the mass of
carbon in non-energy applications.
As shown in Table 3-22, fossil fuel emissions in 2022 from the non-energy uses of fossil fuels were 102.8 MMT CO2
Eq., which constituted approximately 2.0 percent of overall fossil fuel emissions. In 2022, the consumption of fuels
for non-energy uses (after the adjustments described above) was 5,428.2 TBtu (see Table 3-23). A portion of the
carbon in the 5,428.2 TBtu of fuels was stored (236.7 MMT CO2 Eq.), while the remaining portion was emitted
(102.8 MMT CO2 Eq.). Non-energy use emissions decreased by 7.9 percent from 2021 to 2022, primarily due to
decreases in industrial coal, natural gas, and HGL fuel consumption. See Annex 2.3 for more details.
57 HGL (formerly referred to as liquefied petroleum gas, or LPG) are hydrocarbons that occur as gases at atmospheric pressure
and as liquids under higher pressures. HGLs include paraffins, such as ethane, propane, butanes, isobutane, and natural
gasoline (formerly referred to as pentanes plus), and HGLs include olefins, such as ethylene, propylene, butylene and
isobutylene.
3-50 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
Table 3-22: CO2 Emissions from Non-Energy Use Fossil Fuel Consumption (MMT CO2 Eq. and
Percent C)
Year
1990
2005
2018
2019
2020
2021
2022
Potential Emissions
292.5 I
357.7
341.5
334.7
328.6
345.0
339.5
C Stored
193.4
232.7
223.2
228.2
230.8
233.4
236.7
Emissions as a % of Potential
34% I
35%
35%
32%
30%
32%
30%
C Emitted
gg ^ a
125.0
118.4
106.5
97.8
111.6
102.8
Note: NEU emissions presented in this table differ from the NEU emissions presented in CRT Table l.A(a)s4 as the CRT NEU
emissions do not include NEU of lubricants and other petroleum in U.S. Territories. NEU emissions from U.S. Territories are
reported under U.S. Territories in the CRT Table l.A(a)s4.
Methodology and Time-Series Consistency
As per discussion of methodology for estimating CO2 emissions from fossil fuel combustion, NEU emissions are
estimated in line with a Tier 2 method described by the IPCC in the 2006IPCC Guidelines for National Greenhouse
Gas Inventories (IPCC 2006) Chapter 2, Figure 2.1 decision tree and available data on energy use and country
specific fuel carbon contents. The first step in estimating carbon stored in products was to determine the
aggregate quantity of fossil fuels consumed for non-energy uses. The carbon content of these feedstock fuels is
equivalent to potential emissions, or the product of consumption and the fuel-specific carbon content values. Both
the non-energy fuel consumption and carbon content data were supplied by the EIA (2023) (see Annex 2.1).
Consumption values for industrial coking coal, petroleum coke, other oils, and natural gas in Table 3-23 and Table
3-24 have been adjusted to subtract non-energy uses that are included in the source categories of the Industrial
Processes and Product Use chapter.58 Consumption of natural gas, HGL, naphthas, other oils, and special naphtha
were adjusted to subtract out net exports of these products that are not reflected in the raw data from EIA.
Consumption values were also adjusted to subtract net exports of HGL components (e.g., propylene, ethane).
For the remaining non-energy uses, the quantity of carbon stored was estimated by multiplying the potential
emissions by a storage factor.
• For several fuel types—petrochemical feedstocks (including natural gas for non-fertilizer uses, HGL,
naphthas, other oils, still gas, special naphtha, and industrial other coal), asphalt and road oil, lubricants,
and waxes—U.S. data on carbon stocks and flows were used to develop carbon storage factors, calculated
as the ratio of (a) the carbon stored by the fuel's non-energy products to (b) the total carbon content of
the fuel consumed. A lifecycle approach was used in the development of these factors in order to account
for losses in the production process and during use. Because losses associated with municipal solid waste
management are handled separately in the Energy sector under the Incineration of Waste source
category, the storage factors do not account for losses at the disposal end of the life cycle.
• For industrial coking coal and distillate fuel oil, storage factors were taken from Marland and Rotty (1984).
• For the remaining fuel types (petroleum coke, miscellaneous products and other petroleum), IPCC (2006)
does not provide guidance on storage factors, and assumptions were made based on the potential fate of
carbon in the respective non-energy use products. Carbon dioxide emissions from carbide production are
implicitly accounted for in the storage factor calculation for the non-energy use of petroleum coke.
58 These source categories include iron and steel production, lead production, zinc production, ammonia manufacture, carbon
black manufacture (included in petrochemical production), titanium dioxide production, ferroalloy production, silicon carbide
production, and aluminum production.
Energy 3-51
-------
Table 3-23: Adjusted Consumption of Fossil Fuels for Non-Energy Uses (TBtu)
Year
1990
2005
2018
2019
2020
2021
2022
Industry
4,110.2
4,961.2
5,261.0
5,143.8
5,096.7
5,343.1
5,301.8
Industrial Coking Coal
NO
80.4 :
124.7
112.8
79.9
77.9
46.7
Industrial Other Coal
7.6
11.0
9.5
9.5
9.5
9.5
9.5
Natural Gas to Chemical Plants
280.6 1
260.7
675.1
663.4
660.5
663.7
654.2
Asphalt & Road Oil
1,170.2
1,323.2
792.8
843.9
832.3
898.1
916.1
HGLa
1'135-0 1
1,554.3 '
2,427.6
2,372.8
2,469.5
2,639.0
2,758.8
Lubricants
186.3
160.2
122.0
118.3
111.1
113.5
119.5
Natural Gasolineb
0.0 I
0-0 1
0.0
0.0
0.0
0.0
0.0
Naphtha (<401 °F)
325.4
679.2
420.6
367.7
327.8
329.3
244.2
Other Oil (>401 °F)
660.4 i
499.2
218.8
211.1
194.7
195.3
111.1
Still Gas
36.7
67.7
166.9
158.7
145.4
152.8
157.1
Petroleum Coke
29-11
104.2 1
0.0
0.0
0.0
0.0
0.0
Special Naphtha
100.6
60.9
86.9
89.1
80.4
75.7
82.4
Distillate Fuel Oil
7.01
16.0
5.8
5.8
5.8
5.8
5.8
Waxes
33.3
31.4
12.4
10.4
9.2
11.8
13.0
Miscellaneous Products
137.8 g
112.8
198.0
180.2
170.7
170.8
183.4
Transportation
176.0
151.3
137.0
131.3
115.6
119.0
125.4
Lubricants
176.0 |
151.3 ;
137.0
131.3
115.6
119.0
125.4
U.S. Territories
50.8
114.9
3.6
3.6
1.0
1.0
1.0
Lubricants
0.7 1
4-6 ]
1.0
1.0
1.0
1.0
1.0
Other Petroleum (Misc. Prod.)
50.1
110.3
2.5
2.6
0.0
0.0
0.0
Total
4,337.1
5,227.5
5,401.6
5,278.8
5,213.4
5,463.2
5,428.2
NO (Not Occurring)
a Excludes natural gasoline.
b Formerly referred to as "pentanes plus." This source has been adjusted and is reported separately from HGL to align with
historic data and revised EIA terminology.
Table 3-24: 2022 Adjusted Non-Energy Use Fossil Fuel Consumption, Storage, and Emissions
Sector/Fuel Type
Adjusted
Non-Energy
Use-1
(TBtu)
Carbon Content
Coefficient
(MMT C/QBtu)
Potential
Carbon
(MMT C)
Storage
Factor
(MMT C)
Carbon
Stored
(MMT C)
Carbon
Emissions
(MMT C)
Carbon
Emissions
(MMT COz Eq.)
Industry
5,301.8
NA
90.0
NA
64.3
25.7
94.3
Industrial Coking Coal
46.7
25.61
1.2
0.10
0.1
1.1
3.9
Industrial Other Coal
9.5
26.10
0.2
0.67
0.2
0.1
0.3
Natural Gas to
Chemical Plants
654.2
14.47
9.4
0.67
6.3
3.1
11.4
Asphalt & Road Oil
916.1
20.55
18.8
1.00
18.7
0.1
0.3
HGLb
2,758.8
16.82
46.4
0.67
31.1
15.3
56.2
Lubricants
119.5
20.20
2.4
0.09
0.2
2.2
8.0
Natural Gasolinec
0.0
18.24
0.0
0.67
0.0
0.0
0.0
Naphtha (<401° F)
244.2
18.55
4.5
0.67
3.0
1.5
5.5
Other Oil (>401° F)
111.1
20.17
2.2
0.67
1.5
0.7
2.7
Still Gas
157.1
17.51
2.8
0.67
1.8
0.9
3.3
Petroleum Coke
0.0
27.85
0.0
0.30
0.0
0.0
0.0
Special Naphtha
82.4
19.74
1.6
0.67
1.1
0.5
2.0
Distillate Fuel Oil
5.8
20.22
0.1
0.50
0.1
0.1
0.2
Waxes
13.0
19.80
0.3
0.58
0.1
0.1
0.4
Miscellaneous
Products
183.4
0.00
0.0
0.00
0.0
0.0
0.0
Transportation
125.4
NA
2.5
NA
0.2
2.3
8.4
Lubricants
125.4
20.20
2.5
0.09
0.2
2.3
8.4
U.S. Territories
1.0
NA
+
NA
+
+
0.1
Lubricants
1.0
20.20
+
0.09
+
+
0.1
3-52 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
Other Petroleum
(Misc. Prod.) + 20.00 + 0.10 + + +_
Total 5,428.2 9Z6 6^6 28^ 102.8
+ Does not exceed 0.05 TBtu, MMT C, or MMT C02 Eq.
NA (Not Applicable)
NO (Not Occurring)
a To avoid double counting, net exports have been deducted.
b Excludes natural gasoline.
c Formerly referred to as "pentanes plus." this source has been adjusted and is reported separately from HGL to align with
historic data and revised EIA terminology.
Note: Totals may not sum due to independent rounding.
Lastly, emissions were estimated by subtracting the carbon stored from the potential emissions (see Table 3-22).
More detail on the methodology for calculating storage and emissions from each of these sources is provided in
Annex 2.3.
Where storage factors were calculated specifically for the United States, data were obtained on (1) products such
as asphalt, plastics, synthetic rubber, synthetic fibers, cleansers (soaps and detergents), pesticides, food additives,
antifreeze and deicers (glycols), and silicones; and (2) industrial releases including energy recovery (waste gas from
chemicals), Toxics Release Inventory (TRI) releases, hazardous waste incineration, and volatile organic compound,
solvent, and non-combustion CO emissions. Data were taken from a variety of industry sources, government
reports, and expert communications. Sources include EPA reports and databases such as compilations of air
emission factors (EPA 2001), EPA's Emissions Inventory System (ElS)to National Inventory Report (NIR) Mapping
file (EPA 2023), Toxics Release Inventory, 1998 (EPA 2000b), Biennial Reporting System (EPA 2000a, 2009),
Resource Conservation and Recovery Act Information System (EPA 2013b, 2015, 2016b, 2018b, 2021), pesticide
sales and use estimates (EPA 1998,1999, 2002, 2004, 2011, 2017), and the Chemical Data Access Tool (EPA
2014b); the EIA Manufacturer's Energy Consumption Survey (MECS) (EIA 1994,1997, 2001, 2005, 2010, 2013,
2017, 2021); the National Petrochemical & Refiners Association (NPRA 2002); the U.S. Census Bureau (1999, 2004,
2009, 2014, 2021); Bank of Canada (2012, 2013, 2014, 2016, 2017, 2018, 2019, 2020, 2021, 2022, 2023); Financial
Planning Association (2006); INEGI (2006); the United States International Trade Commission (2023); Gosselin,
Smith, and Hodge (1984); EPA's Municipal Solid Waste (MSW) Facts and Figures (EPA 2013, 2014a, 2016a, 2018a,
2019); the U.S. Tire Manufacturers Association (USTMA 2012, 2013, 2014, 2016, 2018, 2020, 2022); the
International Institute of Synthetic Rubber Products (IISRP 2000, 2003); the Fiber Economics Bureau (FEB 2001,
2003, 2005, 2007, 2009, 2010, 2011, 2012, 2013); the Independent Chemical Information Service (ICIS 2008, 2016);
the EPA Chemical Data Access Tool (CDAT) (EPA 2014b); the American Chemistry Council (ACC 2003 through 2011,
2013, 2014, 2015, 2016, 2017, 2018, 2019, 2020, 2021, 2022, 2023a); the Guide to the Business of Chemistry (ACC
2023b); and the Chemistry Industry Association of Canada (CIAC 2023). Specific data sources are listed in full detail
in Annex 2.3.
Methodological recalculations were applied to the entire time series to ensure time-series consistency from 1990
through 2022 as discussed below.
Box 3-5: Reporting of Lubricants, Waxes, and Asphalt and Road Oil Product Use in Energy
Sector
IPCC (2006) provides methodological guidance to estimate emissions from the first use of fossil fuels as a
product for primary purposes other than combustion for energy purposes (including lubricants, paraffin waxes,
bitumen / asphalt, and solvents) under the IPPU sector.59 In this Inventory, carbon storage and carbon
59 See for example Volume 3: Industrial Processes and Product Use, and Chapter 5: Non-Energy Products from Fuels and
Solvent Use of the 2006 IPCC Guidelines for National Greenhouse Gas Inventories (IPCC 2006).
Energy 3-53
-------
emissions from product use of lubricants, waxes, and asphalt and road oil are reported under the Energy sector
in the Carbon Emitted from Non-Energy Uses of Fossil Fuels source category (CRT Source Category 1A5).60
The emissions are reported in the Energy sector, as opposed to the IPPU sector, to reflect national
circumstances in its choice of methodology and to increase transparency of this source category's unique
country-specific data sources and methodology. Although emissions from these non-energy uses are reported in
the Energy chapter the methodologies used to determine emissions are compatible with the 2006IPCC
Guidelines. The country-specific methodology used for the Carbon Emitted from Non-Energy Uses of Fossil Fuels
source category is based on a carbon balance (i.e., carbon inputs-outputs) calculation of the aggregate amount
of fossil fuels used for non-energy uses, including inputs of lubricants, waxes, asphalt and road oil (see Table
3-24).
For those inputs, U.S. country-specific data on carbon stocks and flows are used to develop carbon storage
factors, which are calculated as the ratio of the carbon stored by the fossil fuel non-energy products to the total
carbon content of the fuel consumed, taking into account losses in the production process and during product
use.61 The country-specific methodology to reflect national circumstances starts with the aggregate amount of
fossil fuels used for non-energy uses and applies a carbon balance calculation, breaking out the carbon
emissions from non-energy use of lubricants, waxes, and asphalt and road oil. The emissions are reported under
the Energy chapter to improve transparency, report a more complete carbon balance and to avoid double
counting. Due to U.S. national circumstances, reporting these carbon emissions separately under IPPU would
involve making artificial adjustments to allocate both the carbon inputs and carbon outputs of the non-energy
use carbon balance. For example, only the emissions from the first use of lubricants and waxes are to be
reported under the IPPU sector, emissions from use of lubricants in 2-stroke engines and emissions from
secondary use of lubricants and waxes in waste incineration with energy recovery are to be reported under the
Energy sector. Reporting these non-energy use emissions from only first use of lubricants and waxes under IPPU
would involve making artificial adjustments to the non-energy use carbon balance and could potentially result in
double counting of emissions. These artificial adjustments would also be required for asphalt and road oil and
solvents (which are captured as part of petrochemical feedstock emissions) and could also potentially result in
double counting of emissions. To avoid presenting an incomplete carbon balance and a less transparent
approach for the Carbon Emitted from Non-Energy Uses of Fossil Fuels source category calculation, the entire
calculation of carbon storage and carbon emissions is therefore conducted in the Non-Energy Uses of Fossil
Fuels category calculation methodology, and both the carbon storage and carbon emissions for lubricants,
waxes, and asphalt and road oil are reported under the Energy sector.
However, emissions from non-energy uses of fossil fuels as feedstocks or reducing agents (e.g., petrochemical
production, aluminum production, titanium dioxide, and zinc production) are reported in the IPPU chapter,
unless otherwise noted due to specific national circumstances.
Uncertainty
An uncertainty analysis was conducted to quantify the uncertainty surrounding the estimates of emissions and
storage factors from non-energy uses. This analysis, performed using @RISK software and the IPCC-recommended
Approach 2 methodology (Monte Carlo stochastic simulation technique), provides for the specification of
probability density functions for key variables within a computational structure that mirrors the calculation of the
60 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.
61 Data and calculations for lubricants and waxes and asphalt and road oil are in Annex 2.3 - Methodology for Estimating
Carbon Emitted from Non-Energy Uses of Fossil Fuels.
3-54 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
inventory estimate. The results presented below provide the 95 percent confidence interval, the range of values
within which emissions are likely to fall, for this source category.
As noted above, the non-energy use analysis is based on U.S.-specific storage factors for (1) feedstock materials
(natural gas, HGL, natural gasoline, naphthas, other oils, still gas, special naphthas, and other industrial coal), (2)
asphalt, (3) lubricants, and (4) waxes. For the remaining fuel types (the "other" category in Table 3-23 and Table
3-24) the storage factors were taken directly from IPCC (2006), where available, and otherwise assumptions were
made based on the potential fate of carbon in the respective NEU products. To characterize uncertainty, five
separate analyses were conducted, corresponding to each of the five categories. In all cases, statistical analyses or
expert judgments of uncertainty were not available directly from the information sources for all the activity
variables; thus, uncertainty estimates were determined using assumptions based on source category knowledge.
The results of the Approach 2 quantitative uncertainty analysis are summarized in Table 3-25 (emissions) and Table
3-26 (storage factors). Carbon emitted from non-energy uses of fossil fuels in 2022 was estimated to be between
71.0 and 166.6 MMT CO2 Eq. at a 95 percent confidence level. This indicates a range of 31 percent below to 62
percent above the 2022 emission estimate of 102.8 MMT CO2 Eq. The uncertainty in the emission estimates is a
function of uncertainty in both the quantity of fuel used for non-energy purposes and the storage factor.
Table 3-25: Approach 2 Quantitative Uncertainty Estimates for CO2 Emissions from Non-
Energy Uses of Fossil Fuels (MMT CO2 Eq. and Percent)
2022 Emission Estimate Uncertainty Range Relative to Emission Estimate-1
(MMT CO ¦ Eq.) (MMT CQ . Eq.) (%)
Lower
Bound
Upper
Bound
Lower
Bound
Upper
Bound
Feedstocks
C02
81.4
51.7
146.6
-37%
+80%
Asphalt
C02
0.3
0.1
0.7
-58%
+119%
Lubricants
C02
16.6
13.7
19.2
-17%
+16%
Waxes
C02
0.4
0.3
0.7
-23%
+77%
Other
C02
4.2
0.8
4.9
-81%
+17%
Total
C02
102.8
71.0
166.6
-31%
+62%
a Range of emission estimates predicted by Monte Carlo stochastic simulation for a 95 percent confidence
interval.
Note: Totals may not sum due to independent rounding.
Table 3-26: Approach 2 Quantitative Uncertainty Estimates for Storage Factors of Non-Energy
Uses of Fossil Fuels (Percent)
Source
Gas
2022 Storage Factor
(%)
Uncertainty Range Relative to Emission Estimate''
(%) (%, Relative)
Lower
Bound
Upper
Bound
Lower
Bound
Upper
Bound
Feedstocks
C02
67.0%
52.3%
73.5%
-22%
+10%
Asphalt
C02
99.6%
99.1%
99.8%
0.5%
0.3%
Lubricants
C02
9.2%
3.9%
17.5%
-57%
+91%
Waxes
C02
57.8%
47.4%
67.5%
-18%
+17%
Other
C02
13.6%
8.1%
83.0%
-41%
+511%
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).
As shown in Table 3-26, waxes 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—also appears to
have relatively 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 carbon that is not stored is emitted. As the
Energy 3-55
-------
production statistics that drive the storage values are relatively well-characterized, this approach yields a result
that is probably biased toward understating uncertainty.
As is the case with the other uncertainty analyses discussed throughout this document, the uncertainty results
above address only those factors that can be readily quantified. More details on the uncertainty analysis are
provided in Annex 2.3.
/erification
In order to ensure the quality of the emission estimates from non-energy uses of fossil fuels, general (IPCC Tier 1)
and category-specific (Tier 2) Quality Assurance/Quality Control (QA/QC) procedures were implemented consistent
with the U.S. Inventory QA/QC plan outlined in Annex 8. This effort included a general analysis, as well as portions
of a category specific analysis for non-energy uses involving petrochemical feedstocks and for imports and exports.
The Tier 2 procedures that were implemented involved checks specifically focusing on the activity data and
methodology for estimating the fate of carbon (in terms of storage and emissions) across the various end-uses of
fossil C. Emission and storage totals for the different subcategories were compared, and trends across the time
series were analyzed to determine whether any corrective actions were needed. Corrective actions were taken to
rectify minor errors and to improve the transparency of the calculations, facilitating future QA/QC.
For petrochemical import and export data, special attention was paid to NAICS numbers and titles to verify that
none had changed or been removed. Import and export totals were compared with 2021 totals as well as their
trends across the time series.
It is important to ensure no double counting of emissions between fuel combustion, non-energy use of fuels and
industrial process emissions. For petrochemical feedstock production, our review of the categories suggests this is
not a significant issue since the non-energy use industrial release data includes different categories of sources and
sectors than those included in the Industrial Processes and Product Use (IPPU) emissions category for
petrochemicals. Further data integration is not available at this time because feedstock data from the EIA used to
estimate non-energy uses of fuels are aggregated by fuel type, rather than disaggregated by both fuel type and
particular industries. Also, GHGRP-reported data on quantities of fuel consumed as feedstocks by petrochemical
producers are unable to be used due to the data failing GHGRP CBI aggregation criteria. This country-specific
approach taken is better able to reflect the national situation because it is accounting for secondary product
imports and exports that are not included directly in the national energy statistics. Furthermore, it is compatible
with the 2006 IPCC Guidelines as discussed in Box 3-5 above, but also as the NEU emissions are here represent
different emissions from those covered in the IPPU petrochemical production category.
Recalculations Discussion
Several updates to activity data factors lead to recalculations of previous year results. The major updates are as
follows:
• U.S. International Trade Commission (2023) made changes to the classification of certain cleanser types,
which doubled the historic emissions data for cleanser imports while exports remained constant.
• ACC (2023b) updated polyester, polyolefin and nylon fiber, ethylene glycol, maleic anhydride, adipic acid,
and acetic acid production in 2021 which resulted in a slight decrease in emissions relative to the previous
Inventory.
• U.S. International Trade Commission (2023) updated historical import and export data from 1996 to 2021,
resulting in greater net exports relative to the previous Inventory.
• EIA (2024) shifted all 2022 product supplied of natural gasoline and unfinished oils to crude oil transfers,
reflecting that, in actuality, nearly the full volume of these fuels is used as a feedstock in crude oil
production, instead of directly consumed as an end-use fuel. Under ElA's guidance, EPA shifted all product
supplied of natural gasoline to crude oil transfers for the time series. Natural gasoline was entirely
3-56 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
recategorized, which resulted in zero emissions for the time series from 1990 to 2022. Natural gasoline
previously made up 1.7 percent of total emissions on average across the time series for non-energy uses
of fossil fuels.
• To better align with EIA methodology, the non-energy use consumption of HGLs is now calculated for the
entire timeseries by assuming that 100 percent of ethane, ethylene, and propylene consumption is for
non-combustion use and 85 percent of normal butane, butylene, isobutane, and isobutylene is for non-
combustion use. Non-energy use consumption of propane is calculated by subtracting the non-energy
consumption of all other HGLs from the total non-combustion consumption of HGLs as published by the
EIA. A further adjustment is made to natural gas, HGL, naphtha, other oil, and special naphtha
consumption to account for exports of organic chemicals, cleansers, and pesticides. Because this
adjustment is apportioned based on the relative ratios of each fuel, the emissions from these fuels have
also changed slightly across the time series. Additionally, to better align with EIA methodology, the non-
combustion and energy HGL carbon contents are now calculated for the entire timeseries following the
above methodology. Overall, this update caused emissions from the non-energy use of natural gas to
decrease an average of 27 MMT CO2 Eq. annually, or 23 percent.
Overall, these changes resulted in an average annual decrease of 10.1 MMT CO2 Eq. (8.2 percent) in carbon
emissions from non-energy uses of fossil fuels for the period 1990 through 2021, relative to the previous Inventory.
Planned Improvements
There are several future improvements planned:
• More accurate accounting of carbon in petrochemical feedstocks. EPA has worked with EIA to determine
the cause of input/output discrepancies in the carbon mass balance contained within the NEU model. In
the future, two strategies to reduce or eliminate this discrepancy will continue to be pursued as part of
quality control procedures. First, accounting of carbon in imports and exports will be improved. The
import/export adjustment methodology will be examined to ensure that net exports of intermediaries
such as ethylene and propylene are fully accounted for. Second, the use of top-down carbon input
calculation in estimating emissions will be reconsidered. Alternative approaches that rely more
substantially on the bottom-up carbon output calculation will be considered instead.
• Improving the uncertainty analysis. Most of the input parameter distributions are based on professional
judgment rather than rigorous statistical characterizations of uncertainty.
• Better characterizing flows of fossil carbon. Additional fates may be researched, including the fossil
carbon load in organic chemical wastewaters, plasticizers, adhesives, films, paints, and coatings. There is
also a need to further clarify the treatment of fuel additives and backflows (especially methyl tert-butyl
ether, MTBE).
• Reviewing the trends in fossil fuel consumption for non-energy uses. Annual consumption for several fuel
types is highly variable across the time series, including industrial coking coal and other petroleum. A
better understanding of these trends will be pursued to identify any mischaracterized or misreported fuel
consumption for non-energy uses.
• Updating the average carbon content of solvents was researched, since the entire time series depends on
one year's worth of solvent composition data. The data on carbon emissions from solvents that were
readily available do not provide composition data for all categories of solvent emissions and also have
conflicting definitions for volatile organic compounds, the source of emissive carbon in solvents.
Additional sources of solvents data will be investigated in order to update the carbon content
assumptions.
• Updating the average carbon content of cleansers (soaps and detergents) was researched; although
production and consumption data for cleansers are published every 5 years by the Census Bureau, the
composition (C content) of cleansers has not been recently updated. Recently available composition data
Energy 3-57
-------
sources may facilitate updating the average carbon content for this category.
• Revising the methodology for consumption, production, and carbon content of plastics was researched;
because of recent changes to the type of data publicly available for plastics, the NEU model for plastics
applies data obtained from personal communications. Potential revisions to the plastics methodology to
account for the recent changes in published data will be investigated.
• Although U.S.-specific storage factors have been developed for feedstocks, asphalt, lubricants, and waxes,
default values from IPCC are still used for two of the non-energy fuel types (industrial coking coal,
distillate oil), and broad assumptions are being used for miscellaneous products and other petroleum.
Over the long term, there are plans to improve these storage factors by analyzing carbon fate similar to
those described in Annex 2.3 or deferring to more updated default storage factors from IPCC where
available.
• Reviewing the storage of carbon black across various sectors in the Inventory, in particular, the carbon
black abraded and stored in tires.
• Assess the current method and/or identify new data sources (e.g., EIA) for estimating emissions from
ammonia/fertilizer use of natural gas.
• Investigate EIA NEU and MECS data to update, as needed, adjustments made for ammonia production
and "natural gas to chemical plants, other uses" and "natural gas to other" non-energy uses, including
iron and steel production, in energy uses and IPPU.
3.3 Incineration of Waste (CRT Source
Category 1A)
Combustion is used to manage about 7 to 19 percent of the solid wastes generated in the United States,
depending on the source of the estimate and the scope of materials included in the definition of solid waste (EPA
2000; EPA 2020; Goldstein and Madtes 2001; Kaufman et al. 2004; Simmons et al. 2006; van Haaren et al. 2010). In
the context of this section, waste includes all municipal solid waste (MSW) as well as scrap tires. In the United
States, combustion of MSW tends to occur at waste-to-energy facilities or industrial facilities where useful energy
is recovered, and thus emissions from waste combustion are accounted for in the Energy chapter. Similarly, scrap
tires are combusted for energy recovery in industrial and utility boilers, pulp and paper mills, and cement kilns.
Combustion of waste results in conversion of the organic inputs to CO2. According to the 2006 IPCC Guidelines,
when the CO2 emitted is of fossil origin, it is counted as a net anthropogenic emission of CO2 to the atmosphere.
Thus, the emissions from waste combustion are calculated by estimating the quantity of waste combusted and the
fraction of the waste that is carbon derived from fossil sources.
Most of the organic materials in MSW are of biogenic origin (e.g., paper, yard trimmings), and have their net
carbon flows accounted for under the Land Use, Land-Use Change, and Forestry chapter. However, some
components of MSW and scrap tires—plastics, synthetic rubber, synthetic fibers, and carbon black—are of fossil
origin. Plastics in the U.S. waste stream are primarily in the form of containers, packaging, and durable goods.
Rubber is found in durable goods, such as carpets, and in non-durable goods, such as clothing and footwear. Fibers
in MSW are predominantly from clothing and home furnishings. As noted above, scrap tires (which contain
synthetic rubber and carbon black) are also considered a "non-hazardous" waste and are included in the waste
combustion estimate, though waste disposal practices for tires differ from MSW. Estimates on emissions from
hazardous waste combustion can be found in Annex 2.3 and are accounted for as part of the carbon mass balance
for non-energy uses of fossil fuels.
3-58 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
Approximately 26.3 million metric tons of MSW were combusted in 2022 (EPA 2022). Carbon dioxide emissions
from combustion of waste decreased 4.2 percent since 1990, to an estimated 12.4 MMT CO2 (12,357 kt) in 2022.
Emissions across the time series are shown in Table 3-27 and Table 3-28.
Waste combustion is also a source of CH4 and N2O emissions (De Soete 1993; IPCC 2006). Methane emissions from
the combustion of waste were estimated to be less than 0.05 MMT CO2 Eq. (less than 0.05 kt CH4) in 2022 and
have remained steady since 1990. Nitrous oxide emissions from the combustion of waste were estimated to be 0.3
MMT CO2 Eq. (1.3 kt N2O) in 2022 and have decreased by 18 percent since 1990. This decrease is driven by the
decrease in total MSW combusted.
Table 3-27: CO2, CH4, and N2O Emissions from the Combustion of Waste (MMT CO2 Eq.)
Gas
1990
2005
2018
2019
2020
2021
2022
C02
12.9 I
13.3 1
13.3
12.9
12.9
12.5
12.4
ch4
+
+
+
+
+
+
+
n2o
0.4
0.3 [
0.4
0.4
0.3
0.4
0.3
Total
13.3
13.6
13.7
13.3
13.3
12.8
12.7
+ Does not exceed 0.05 MMT C02 Eq.
Note: Totals may not sum due to independent rounding.
Table 3-28: CO2, CH4, and N2O Emissions from the Combustion of Waste (kt)
Gas
1990
2005
2018
2019
2020
2021
2022
C02
12,900
13,254
13,339
12,948
12,921
12,476
12,357
ch4
-
+
+
+
+
+
+
n2o
2
1 ¦
1
1
1
1
1
+ Does not exceed 0.05 kt.
Methodology and Time-Series Consistency
Municipal Solid Waste Combustion
To determine both CO2 and non-CC>2 emissions from the combustion of waste, the tonnage of waste combusted
and an estimated emissions factor are needed. Emission estimates from the combustion of tires are discussed
separately. Data for total waste combusted was derived from BioCycle (van Haaren et al. 2010), EPA Facts and
Figures Report, Energy Recovery Council (ERC), EPA's Greenhouse Gas Reporting Program (GHGRP), and the U.S.
Energy Information Administration (EIA). Multiple sources were used to ensure a complete, quality dataset, as
each source encompasses a different timeframe.
EPA determined the MSW tonnages based on data availability and accuracy throughout the time series.
• 1990-2006: MSW combustion tonnages are from Biocycle combustion data. Tire combustion data from
the U.S. Tire Manufacturers Association (USTMA) are removed to arrive at MSW combusted without tires.
• 2006-2010: MSW combustion tonnages are an average of Biocycle (with USTMA tire data tonnage
removed), U.S. EPA Facts and Figures, EIA, and Energy Recovery Council data (with USTMA tire data
tonnage removed).
• 2011-2022: MSW combustion tonnages are from EPA's GHGRP data.
Energy 3-59
-------
Table 3-29 provides the estimated tons of MSW combusted including and excluding tires.
Table 3-29: Municipal Solid Waste Combusted (Short Tons)
1990
2005
2018
2019
2020
2021
2022
Waste Combusted
(excluding tires)
Waste Combusted
(including tires)
33,344,839
33,766,239
26,486,414
28,631,054
I
I
I
1
29,162,364
30,853,949
28,174,311
29,821,141
27,586,271
29,106,686
27,867,446
29,261,446
26,338,130
27,732,130
Sources: BioCycle, EPA Facts and Figures, ERC, GHGRP, EIA, USTMA.
CO2 Emissions from MSW Excluding Scrap Tires
Fossil CO2 emission factors were calculated from EPA's GHGRP data for non-biogenic sources. Using GHGRP-
reported emissions for CHUand N2O and assumed emission factors, the tonnage of waste combusted, excluding
tires, was derived. Methane and N2O emissions and assumed emission factors were used to estimate the amount
of MSW combusted in terms of energy content. The energy content of MSW combusted was then converted into
tonnage based on assumed MSW heating value. Two estimates were generated (one for Cm and one for N2O) and
the two were averaged together. Dividing fossil CO2 emissions from GHGRP FLIGHT data for MSW combustors by
this estimated tonnage yielded an annual CO2 emission factor. As this data was only available following 2011, all
years prior use an average of the emission factors from 2011 through 2015. See Annex 3.7 for more detail on how
MSW carbon factors were calculated.
Finally, CO2 emissions were calculated by multiplying the annual tonnage estimates, excluding tires, by the
calculated emissions factor. Calculated fossil CO2 emission factors are shown in Table 3-30.
Table 3-30: Calculated Fossil CO2 Content per Ton Waste Combusted (kg CCh/Short Ton
Combusted)
Year
1990
2005
2018
2019
2020
2021
2022
C02 Emission Factors
366 a
366 a
361
363
377
365
382
CO2 Emissions from Scrap Tires
Scrap tires contain several types of synthetic rubber, carbon black, and synthetic fibers. Each type of synthetic
rubber has a discrete carbon content, and carbon black is 100 percent C. For synthetic rubber and carbon black in
scrap tires, information on average weight, disposal percentage, and total tires incinerated for energy was
obtained biannually from U.S. Scrap Tire Management Summary for 2005 through 2022 data (USTMA 2022).
Information about scrap tire composition was taken from the Rubber Manufacturers' Association internet site
(USTMA 2012a). Emissions of CO2 were calculated based on the amount of scrap tires used for fuel and the
synthetic rubber and carbon black content of scrap tires. The mass of combusted material is multiplied by its
carbon content to calculate the total amount of carbon stored. 2022 values are proxied from 2021 data. More
detail on the methodology for calculating emissions from each of these waste combustion sources is provided in
Annex 3.7. Table 3-31 provides CO2 emissions from combustion of waste tires.
Table 3-31: CO2 Emissions from Combustion of Tires (MMT CO2 Eq.)
Year
1990
2005
2018
2019
2020
2021
2022
Synthetic Rubber
0.3
1.6 1
1.3
1.2
1.1
1.0
1.0
C Black
0.4
2.0 J
1.5
1.5
1.4
1.3
1.3
Total
0.7
3.6
2.8
2.7
2.5
2.3
2.3
3-60 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
Non-C02 Emissions
Combustion of waste also results in emissions of Cl-Uand N2O. These emissions were calculated by multiplying the
total estimated mass of waste combusted, including tires, by the respective emission factors. The emission factors
for Cm and N2O emissions per quantity of MSW combusted are default emission factors for the default
continuously-fed stoker unit MSW combustion technology type and were taken from IPCC (2006).
Uncertainty
An Approach 2 Monte Carlo analysis was performed to determine the level of uncertainty surrounding the
estimates of CO2 emissions and N2O emissions from the incineration of waste (given the very low emissions for
CH4, no uncertainty estimate was derived). IPCC Approach 2 analysis allows the specification of probability density
functions for key variables within a computational structure that mirrors the calculation of the Inventory estimate.
Statistical analyses or expert judgments of uncertainty were not available directly from the information sources for
most variables; thus, uncertainty estimates for these variables were determined using assumptions based on
source category knowledge and the known uncertainty estimates for the waste generation variables.
The uncertainties in the waste incineration emission estimates arise from both the assumptions applied to the data
and from the quality of the data. Key factors include reported CO2 emissions; N2O and CH4emissions factors, and
tire synthetic rubber and black carbon contents. The highest levels of uncertainty surround the reported emissions
from GHGRP; the lowest levels of uncertainty surround variables that were determined by quantitative
measurements (e.g., combustion efficiency, carbon content of carbon black).
The results of the Approach 2 quantitative uncertainty analysis are summarized in Table 3-32. Waste incineration
CO2 emissions in 2022 were estimated to be between 10.3 and 14.4 MMT CO2 Eq. at a 95 percent confidence level.
This indicates a range of 17 percent below to 16 percent above the 2022 emission estimate of 12.4 MMT CO2 Eq.
Also at a 95 percent confidence level, waste incineration N2O emissions in 2022 were estimated to be between 0.2
and 0.9 MMT CO2 Eq. This indicates a range of 54 percent below to 164 percent above the 2022 emission estimate
of 0.3 MMT CO2 Eq.
Table 3-32: Approach 2 Quantitative Uncertainty Estimates for CO2 and N2O from the
Incineration of Waste (MMT CO2 Eq. and Percent)
Source
Gas
2022 Emission Estimate
(MMT CO . Eq.)
Uncertainty Range Relative to Emission Estimate-1
(MMTCO' Eq.) (%)
Lower
Upper
Lower Upper
Bound
Bound
Bound Bound
Incineration of Waste
C02
12.4
10.3
14.4
-17% 16%
Incineration of Waste
N20
0.3
0.2
0.9
-54% 164%
a Range of emission estimates predicted by Monte Carlo Simulation for a 95 percent confidence interval.
QA/QC and Verification
In order to ensure the quality of the emission estimates from waste combustion, general (IPCC Tier 1) and
category-specific (Tier 2) Quality Assurance/Quality Control (QA/QC) procedures were implemented consistent
with the U.S. Inventory QA/QC plan outlined in Annex 8. The Tier 2 procedures that were implemented involved
checks specifically focusing on the activity data and specifically focused on the emission factor and activity data
sources and methodology used for estimating emissions from combustion of waste. Trends across the time series
were analyzed to determine whether any corrective actions were needed. Corrective actions were taken to rectify
minor errors in the use of activity data.
Energy 3-61
-------
Recalculations Discussion
No recalculations were performed for the current Inventory.
Planned Improvements
Research was conducted to review the composition of carbon black abraded and stored in tires. No definitive
sources were found that support updating the current factor of 28 percent carbon black for commercial and light
duty tires. This factor was not updated, but additional research can be completed in future Inventory cycles.
3.4 Coal Mining (CRT Source Category
lBla)
Three types of coal mining-related activities release Cm and CO2 to the atmosphere: underground mining, surface
mining, and post-mining (i.e., coal-handling) activities. While surface coal mines account for the majority of U.S.
coal production, underground coal mines contribute the largest share of fugitive CH4 emissions (see Table 3-34 and
Table 3-35) due to the higher Cm content of coal in the deeper underground coal seams. In 2022,185
underground coal mines and 354 surface mines were operating in the United States (EIA 2023). In recent years, the
total number of active coal mines in the United States has declined. In 2022, the United States was the fourth
largest coal producer in the world, after China, India, and Indonesia (IEA 2022).
Table 3-33: Coal Production (kt)
Year
1990
2005
2018
2019
2020
2021
2022
Underground
586 i
Number of Mines
1,683 =
236
226
196
174
185
Production
384,244
334,399
249,804
242,557
177,380
200,122
201,525
Surface
1
mm!
1
Number of Mines
1,656
789
430
432
350
332
354
Production
546,808 =
691,447 ¦
435,521
397,750
307,944
323,142
336,990
Total
is
!E
Number of Mines
3,339
1,398
666
658
546
506
539
Production
931,052
1,025,846
685,325
640,307
485,324
523,264
538,515
Fugitive CH4 Emissions
Underground coal mines liberate Cm from ventilation systems and from degasification systems. Ventilation
systems pump air through the mine workings to dilute noxious gases and ensure worker safety; these systems can
exhaust significant amounts of Cm to the atmosphere in low concentrations. Degasification systems are wells
drilled from the surface or boreholes drilled inside the mine that remove large, often highly concentrated volumes
of Cm before, during, or after mining. Some mines recover and use CH4 generated from ventilation and
degasification systems, thereby reducing emissions to the atmosphere.
Surface coal mines liberate CH4 as the overburden is removed and the coal is exposed to the atmosphere. Methane
emissions are normally a function of coal rank (a classification related to the percentage of carbon in the coal) and
depth. Surface coal mines typically produce lower-rank coals and remove less than 250 feet of overburden, so their
level of emissions is much lower than from underground mines.
In addition, CH4 is released during post-mining activities, as the coal is processed, transported, and stored for use.
3-62 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
Total Cm emissions in 2022 were estimated to be 1,558 kt (43.6 MMT CO2 Eq.), a decline of approximately 60
percent since 1990 (see Table 3-34 and Table 3-35). In 2022, underground mines accounted for approximately 72
percent of total emissions, surface mines accounted for 14 percent, and post-mining activities accounted for 14
percent. In 2022, total CH4 emissions from coal mining decreased by approximately 2 percent relative to the
previous year. Total coal production in 2022 increased by 3 percent compared to 2021. This resulted in an increase
of 4 percent in CH4 emissions from surface mining and post-mining activities in 2022. However, surface mining and
post-mining activities have a lower impact on total CH4 compared to underground mining (72 percent of total
emissions in 2022). The number of operating underground mines increased in 2022 and the amount of CH4
recovered and used in 2022 increased by 25 percent compared to 2021. This resulted in a slight decrease in overall
CH4 emissions in 2022 (2 percent), compared to 2021.
Table 3-34: CH4 Emissions from Coal Mining (MMT CO2 Eq.)
Activity
1990
2005
2018
2019
2020
2021
2022
Underground (UG) Mining
83.1
46.7
43.6
38.5
35.2
32.9
31.5
Liberated
90-6 1
66.9 3
66.7
56.6
53.7
52.3
55.7
Recovered & Used
(7.5)
(20.1)
(23.1)
(18.1)
(18.5)
(19.4)
(24.2)
Surface Mining
12.0
13.3
7.8
7.2
5.4
5.7
6.0
Post-Mining (UG)
10.3
8.6
5.9
5.8
4.3
4.8
4.8
Post-Mining (Surface)
2.6
2.9
1.7
1.5
1.2
1.2
1.3
Total
108.1
71.5
59.1
53.0
46.2
44.7
43.6
Notes: Parentheses indicate negative values. Totals may not sum due to independent rounding.
Table 3-35: CH4 Emissions from Coal Mining (kt)
Activity
1990
2005
2018
2019
2020
2021
2022
Underground (UG) Mining
2,968
1,669
1,557
1,375
1,257
1,176
1,124
Liberated
3,237 II
2,388 1
2,382
2,022
1,917
1,868
1,989
Recovered & Used
(269)
(720)
(825)
(646)
(660)
(692)
(865)
Surface Mining
430
475
280
255
194
205
215
Post-Mining (UG)
368
306
212
206
155
170
173
Post-Mining (Surface)
93
103
61
55
42
44
47
Total
3,860
2,552
2,110
1,892
1,648
1,595
1,558
Note: Parentheses indicate negative values. Totals may not sum due to independent rounding.
Methodology and Time-Series Consistency
EPA uses an IPCC Tier 3 method for estimating CH4 emissions from underground coal mining and an IPCC Tier 2
method for estimating CH4 emissions from surface mining and post-mining activities (for coal production from both
underground mines and surface mines) in accordance with methodological decisions trees in IPCC guidelines
(Volume 2, Chapter 4, Figure 4.1.1 and 4.1.2) and available data (IPCC 2006). The methodology for estimating CH4
emissions from coal mining consists of two steps:
• Estimate CH4 emissions from underground mines. These emissions have two sources: ventilation systems
and degasification systems. They are estimated using mine-specific data, then summed to determine total
CH4 liberated. The CH4 recovered and used is then subtracted from this total, resulting in an estimate of
net emissions to the atmosphere.
• Estimate CH4 emissions from surface mines and post-mining activities. Unlike the methodology for
underground mines, which uses mine-specific data, the methodology for estimating emissions from
surface mines and post-mining activities consists of multiplying basin-specific coal production by basin-
specific gas content and an emission factor.
Step 1: Estimate CH4 Liberated and CH4 Emitted from Underground Mines
Underground mines generate CH4 from ventilation systems and degasification systems. Some mines recover and
Energy 3-63
-------
use the liberated CH4, thereby reducing emissions to the atmosphere. Total Cm emitted from underground mines
equals the Cm liberated from ventilation systems, plus the Cm liberated from degasification systems, minus the
Cm recovered and used.
Step 1.1: Estimate CH4 Liberated from Ventilation Systems
To estimate CH4 liberated from ventilation systems, EPA uses data collected through its Greenhouse Gas Reporting
Program (GHGRP)62 (Subpart FF, "Underground Coal Mines"), data provided by the U.S. Mine Safety and Health
Administration (MSHA) (MSHA 2023), and occasionally data collected from other sources on a site-specific level
(e.g., state gas production databases). Since 2011, the nation's "gassiest" underground coal mines—those that
liberate more than 36,500,000 actual cubic feet of CH4 per year (about 17,525 MT CO2 Eq.)—have been required to
report to EPA's GHGRP (EPA 20 23).63 Mines that report to EPA's GHGRP must report quarterly measurements of
CH4 emissions from ventilation systems; they have the option of recording and reporting their own measurements,
or using the measurements taken by MSHA as part of that agency's quarterly safety inspections of all mines in the
United States with detectable CH4 concentrations.64
Since 2013, ventilation CH4 emission estimates have been calculated based on both quarterly GHGRP data
submitted by underground mines and on quarterly measurement data obtained directly from MSHA. Because not
all mines report under EPA's GHGRP, the emissions of the mines that do not report must be calculated using MSHA
data. The MSHA data also serves as a quality assurance tool for validating GHGRP data. For GHGRP data, reported
quarterly ventilation methane emissions (metric tons) are summed for each mine to develop mine-specific annual
ventilation emissions. For MSHA data, the average daily CH4 emission rate for each mine is determined using the
CH4 total for all data measurement events conducted during the calendar year and total duration of all data
measurement events (in days). The calculated average daily CH4 emission rate is then multiplied by 365 days to
estimate annual ventilation CH4 emissions for the MSHA dataset.
Step 1.2: Estimate CH4 Liberated from Degasification Systems
Particularly gassy underground mines also use degasification systems (e.g., wells or boreholes) to remove CH4
before, during, or after mining. This CH4 can then be collected for use or vented to the atmosphere. Nineteen
mines used degasification systems in 2022 and all of these mines reported the CH4 removed through these systems
to EPA's GHGRP under Subpart FF (EPA 2023). Based on the weekly measurements reported to EPA's GHGRP,
degasification data summaries for each mine are added to estimate the CH4 liberated from degasification systems.
Twelve of the 19 mines with degasification systems had operational CH4 recovery and use projects, including two
mines with two recovery and use projects each (see step 1.3 below).65
Degasification data reported to EPA's GHGRP by underground coal mines is the primary source of data used to
develop estimates of CH4 liberated from degasification systems. Data reported to EPA's GHGRP were used
exclusively to estimate CH4 liberated from degasification systems at 13 of the 19 mines that used degasification
systems in 2022. Data from state gas well production databases were used to supplement GHGRP degasification
data for the remaining six mines (DMME 2023; GSA 2023; WVGES 2023; McElroy OVS 2013).
62 In implementing improvements and integrating data from EPA's GHGRP, EPA followed the latest guidance from the IPCC on
the use of facility-level data in national inventories (IPCC 2011).
63 Underground coal mines report to EPA under Subpart FF of the GHGRP (40 CFR Part 98). In 2022, 61 underground coal mines
reported to the program.
64 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.
65 Several of the mines venting CH4from degasification systems use a small portion of the gas to fuel gob well blowers in
remote locations where electricity is not available. However, this CH4 use is not considered to be a formal recovery and use
project.
3-64 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
For pre-mining wells, cumulative degasification volumes that occur prior to the well being mined through are
attributed to the mine in the inventory year in which the well is mined through.66 EPA's GHGRP does not require
gas production from virgin coal seams (coalbed methane) to be reported by coal mines under Subpart FF.67 Most
pre-mining wells drilled from the surface are considered coalbed methane wells prior to mine-through and
associated CFU emissions are reported under another subpart of the GHGRP (Subpart W, "Petroleum and Natural
Gas Systems"). As a result, GHGRP data must be supplemented to estimate cumulative degasification volumes that
occurred prior to well mine-through. There were four mines with degasification systems that include pre-mining
wells that were mined through in 2022. For all of these mines, GHGRP data were supplemented with historical
data from state gas well production databases (ERG 2023; GSA 2023), as well as with mine-specific information
regarding the locations and dates on which the pre-mining wells were mined through (JWR 2010; El Paso 2009;
ERG 2023).
Step 1.3: Estimate CH4 Recovered from Ventilation and Degasification Systems, and Utilized or
Destroyed (Emissions Avoided)
Twelve mines had a total of fourteen CH4 recovery and use projects in place in 2022, including two mines that each
have two recovery and use projects. Thirteen of these projects involved degasification systems with one mine
having a ventilation air methane abatement project (VAM). Eleven of these mines sold the recovered CH4 to a
pipeline, including one that also used CH4 to fuel a thermal coal dryer. One mine destroyed the recovered CH4
(VAM) using regenerative thermal oxidation (RTO) without energy recovery and using enclosed flares.
The CH4 recovered and used (or destroyed) at the twelve mines described above are estimated using the following
methods:
• EPA's GHGRP data was exclusively used to estimate the CH4 recovered and used from six of the 12 mines
that deployed degasification systems in 2022. Based on quarterly measurements, the GHGRP
degasification destruction data summaries for each mine are added together to estimate the CH4
recovered and used from degasification systems.
• State sales data were used to supplement GHGRP data to estimate CH4 recovered and used from five
mines that deployed degasification systems in 2022 (DMME 2023, ERG 2023, GSA 2023, and WVGES
2023). Four of these mines intersected pre-mining wells in 2022. Supplemental information is used for
these mines because estimating CH4 recovery and use from pre-mining wells requires additional data not
reported under Subpart FF of EPA's GHGRP (see discussion in step 1.2 above) to account for the emissions
avoided prior to the well being mined through. The supplemental data is obtained from state gas
production databases as well as mine-specific information on the location and timing of mined-through
pre-mining wells.
• For the single mine that employed VAM for CH4 recovery and use, the estimates of CH4 recovered and
used were obtained from the mine's offset verification statement (OVS) submitted to the California Air
Resources Board (CARB) (McElroy OVS 2023). This mine also reported CH4 reductions from flaring. GHGRP
data were used to estimate CH4 recovered and flared in 2022.
Step 2: Estimate CH4 Emitted from Surface Mines and Post-Mining Activities
Mine-specific data are not available for estimating CH4 emissions from surface coal mines or for post-mining
activities. For surface mines, basin-specific coal production obtained from the Energy Information Administration's
Annual Coal Report (EIA 2023) is multiplied by basin-specific CH4 contents (EPA 1996, 2005) and a 150 percent
emission factor (to account for CH4from over- and under-burden) to estimate CH4 emissions (King 1994; Saghafi
66 A well is "mined through" when coal mining development or the working face intersects the borehole or well.
67 This applies for pre-drainage in years prior to the well being mined through. Beginning with the year the well is mined
through, the annual volume of CH4 liberated from a pre-drainage well is reported under Subpart FF of EPA's GHGRP.
Energy 3-65
-------
2013). For post-mining activities, basin-specific coal production is multiplied by basin-specific Cm contents and a
mid-range 32.5 percent emission factor for Cm desorption during coal transportation and storage (Creedy 1993).
Basin-specific in situ gas content data were compiled from AAPG (1984) and USBM (1986).
Fugitive C02 Emissions
Methane and CO2 are naturally occurring in coal seams and are collectively referred to as coal seam gas. These
gases remain trapped in the coal seam until coal is mined (i.e., coal seam is exposed and fractured during mining
operations). Fugitive CO2 emissions occur during underground coal mining, surface coal mining, and post-mining
activities. Methods and data to estimate fugitive CO2 emissions from underground and surface coal mining are
presented in the sections below. Fugitive CO2 emissions from post-mining activities were not estimated due to the
lack of an IPCC method and unavailability of data.
Total fugitive CO2 emissions in 2022 were estimated to be 2,474 kt (2.5 MMT CO2 Eq.), a decline of approximately
46 percent since 1990. In 2022, underground mines accounted for approximately 89 percent of total fugitive CO2
emissions. In 2022, total fugitive CO2 emissions from coal mining increased by approximately 1 percent relative to
the previous year. This increase was due to an increase in annual coal production.
Table 3-36: CO2 Emissions from Coal Mining (MMT CO2 Eq.)
Activity
1990
2005
2018
2019
2020
2021
2022
Underground (UG) Mining
4.2
3.6
2.8
2.7
1.9
2.2
2.2
Liberated
4-2:
3.6 iii!
2.7
2.6
1.9
2.2
2.2
Recovered & Used
(+)
(+)
(+)
(+)
(+)
(+)
(+)
Flaring
NO
NO 1
0.1
0.1
+
+
+
Surface Mining
0.4
0.6
0.4
0.3
0.2
0.3
0.3
Total
4.6
4.2
3.1
3.0
2.2
2.5
2.5
+ Does not exceed 0.05 MMT C02
Eq.
NO (Not Occurring)
Notes: Parentheses indicate negative values. Totals may not sum due to independent rounding.
Table 3-37: CO2 Emissions from Coal Mining (kt)
Activity
1990
2005
2018
2019
2020
2021
2022
Underground (UG) Mining
4,164
3,610
2,787
2,670
1,948
2,193
2,201
Liberated
4,171
3,630 1
2,712
2,633
1,926
2,173
2,188
Recovered & Used
(8)
(21)
(23)
(18)
(19)
(19)
(24)
Flaring
NO 1
NO 1
97
55
41
40
38
Surface Mining
443
560
353
322
249
262
273
Total
4,606
4,169
3,139
2,992
2,197
2,455
2,474
NO (Not Occurring)
Notes: Parentheses indicate negative values. Totals may not sum due to independent rounding.
Methodology and Time-Series Consistency
EPA uses an IPCC Tier 1 method for estimating fugitive CO2 emissions from underground coal mining and surface
mining in accordance with methodological decisions trees in IPCC guidelines (Volume 2, Chapter 4, Figure 4.1.1a)
and available data (IPCC 2019). IPCC methods and data to estimate fugitive CO2 emissions from post-mining
activities (for both underground and surface coal mining) are currently not available.
Step 1: Underground Mining
EPA used the following overarching IPCC equation to estimate fugitive CO2 emissions from underground coal mines
(IPCC 2019):
3-66 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
Equation 3-1: Estimating Fugitive CO2 Emissions from Underground Mines
Total C02 from Underground Mines
= C02 from underground mining — Amount of C02 in gas recovered
+ C02 from methane flaring
Step 1.1: Estimate Fugitive CO2 Emissions from Underground Mining
EPA estimated fugitive CO2 emissions from underground mining using the IPCC Tier 1 emission factor (5.9
m3/metric ton) and annual coal production from underground mines (EIA 2023). The underground mining default
emission factor accounts for all the fugitive CO2 likely to be emitted from underground coal mining. Therefore, the
amount of CO2 from coal seam gas recovered and utilized for energy is subtracted from underground mining
estimates in Step 2, below. Under IPCC methods, the CO2 emissions from gas recovered and utilized for energy use
(e.g., injected into a natural gas pipeline) are reported under other sectors of the Inventory (e.g., stationary
combustion of fossil fuel or oil and natural gas systems) and not under the coal mining sector.
Step 1.2: Estimate Amount of CO2 In Coal Seam Gas Recovered for Energy Purposes
EPA estimated fugitive CO2 emissions from coal seam gas recovered and utilized for energy purposes by using the
IPCC Tier 1 default emission factor (19.57 metric tons CCh/million cubic meters of coal bed methane (CBM)
produced) and quantity of coal seam gas recovered and utilized. Data on annual quantity of coal seam gas
recovered and utilized are available from GHGRP and state sales data (GHGRP 2023; DMME 2023; ERG 2023; GSA
2023; WVGES 2023). The quantity of coal seam gas recovered and destroyed without energy recovery (e.g., VAM
projects with RTO) is deducted from the total coal seam gas recovered quantity (McElroy OVS 2023).
Step 1.3: Estimate Fugitive CO2 Emissions from Flaring
The IPCC method includes combustion CO2 emissions from gas recovered for non-energy uses (i.e., flaring, or
catalytic oxidation) under fugitive CO2 emission estimates for underground coal mining. In effect, these emissions,
though occurring through stationary combustion, are categorized as fugitive emissions in the Inventory. EPA
estimated CO2 emissions from methane flaring using the following equation:
Equation 3-2: Estimating CO2 Emissions from Drained Methane Flared or Catalytically
Oxidized
C02 from flaring
= 0.98 x Volume of methane flared x Conversion Factor
x Stoichiometric Mass Factor
Currently there are three mines that report destruction of recovered methane through flaring without energy use.
Annual data for 2022 were obtained from one mine's offset verification statement (OVS) submitted to the
California Air Resources Board (CARB) and the GHGRP for the remaining two mines (McElroy OVS 2023; GHGRP
2023).
Step 2: Surface Mining
EPA estimated fugitive CO2 emissions from surface mining using the IPCC Tier 1 emission factor (0.44 m3/metric
ton) and annual coal production from surface mines (EIA 2023).
Uncertainty
A quantitative uncertainty analysis was conducted for the coal mining source category using the IPCC-
recommended Approach 2 uncertainty estimation methodology. Because emission estimates of CH4 from
underground ventilation systems were based on actual measurement data from EPA's GHGRP or from MSHA,
uncertainty is relatively low. A degree of imprecision was introduced because the ventilation air measurements
Energy 3-67
-------
used were not continuous but rather quarterly instantaneous readings that were used to determine the average
annual emission rates. Additionally, the measurement equipment used can be expected to have resulted in an
average of 10 percent overestimation of annual Cm emissions (Mutmansky & Wang 2000). Equipment
measurement uncertainty is applied to GHGRP data.
Estimates of Cm liberated and recovered by degasification systems are relatively certain for utilized Cm because of
the availability of EPA's GHGRP data and state gas sales information. Many of the liberation and recovery
estimates use data on wells within 100 feet of a mined area. However, uncertainty exists concerning the radius of
influence of each well. The number of wells counted, and thus the liberated CH4 and avoided emissions, may vary if
the drainage area is found to be larger or smaller than estimated.
EPA's GHGRP requires weekly CH4 monitoring of mines that report degasification systems, and continuous CH4
monitoring is required for CH4 utilized on- or off-site. Since 2012, GHGRP data have been used to estimate CH4
emissions from vented degasification wells, reducing the uncertainty associated with prior MSHA estimates used
for this sub-source. Beginning in 2013, GHGRP data were also used for determining CH4 recovery and use at mines
without publicly available gas usage or sales records, which has reduced the uncertainty from previous estimation
methods that were based on information from coal industry contacts.
Surface mining and post-mining emissions are associated with considerably more uncertainty than underground
mines, because of the difficulty in developing accurate emission factors from field measurements. However, since
underground coal mining, as a general matter, results in significantly larger CH4 emissions due to production of
higher-rank coal and greater depth, and estimated emissions from underground mining constitute the majority of
estimated total coal mining CH4 emissions, the uncertainty associated with underground emissions is the primary
factor that determines overall uncertainty.
The major sources of uncertainty for estimates of fugitive CO2 emissions are the Tier 1IPCC default emission
factors used for underground mining (-50 percent to +100 percent) and surface mining (-67 percent to +200
percent) (IPCC 2019). Additional sources of uncertainty for fugitive CO2 emission estimates include ElA's annual
coal production data and data used for gas recovery projects, such as GHGRP data, state gas sales data, and VAM
estimates for the single mine that operates an active VAM project. Uncertainty ranges for these additional data
sources are already available, as these are the same data sources used for CH4 emission estimates.
The results of the Approach 2 quantitative uncertainty analysis are summarized in Table 3-38. Coal mining CH4
emissions in 2022 were estimated to be between 34.8 and 47.8 MMT CO2 Eq. at a 95 percent confidence level. This
indicates a range of 20.3 percent below to 9.5 percent above the 2022 emission estimate of 43.6 MMT CO2 Eq.
Coal mining fugitive CO2 emissions in 2022 were estimated to be between 0.8 and 4.3 MMT CO2 Eq. at a 95 percent
confidence level. This indicates a range of 68.5 percent below to 75.4 percent above the 2022 emission estimate of
2.5 MMT CO2 Eq.
Table 3-38: Approach 2 Quantitative Uncertainty Estimates for CH4 and CO2 Emissions from
Coal Mining (MMT CO2 Eq. and Percent)
Source
Gas
2022 Emission Estimate
(MMTCOz Eq.)
Uncertainty Range Relative to Emission Estimate'
(MMT C02 Eq.) (%)
3
Lower Upper
Lower
Upper
Bound Bound
Bound
Bound
Coal Mining
ch4
43.6
34.8 47.8
-20%
+9%
Coal Mining
C02
2.5
CO
00
0
-69%
+75%
a Range of emission estimates predicted by Monte Carlo stochastic simulation for a 95 percent confidence interval.
QA/QC and Verification
In order to ensure the quality of the emission estimates for coal mining, general (IPCC Tier 1) and category-specific
(Tier 2) Quality Assurance/Quality Control (QA/QC) procedures were implemented consistent with the U.S.
Inventory QA/QC plan outlined in Annex 8. The Tier 2 procedures that were implemented involved checks
specifically focusing on the activity data and reported emissions data used for estimating fugitive emissions from
3-68 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
coal mining. Trends across the time series were analyzed to determine whether any corrective actions were
needed.
Emission estimates for coal mining rely in large part on data reported by coal mines to EPA's GHGRP. EPA verifies
annual facility-level reports through a multi-step process to identify potential errors and ensure that data
submitted to EPA are accurate, complete, and consistent. All reports submitted to EPA are evaluated by electronic
validation and verification checks. If potential errors are identified, EPA will notify the reporter, who can resolve
the issue either by providing an acceptable response describing why the flagged issue is not an error or by
correcting the flagged issue and resubmitting their annual report. Additional QA/QC and verification procedures
occur for each GHGRP subpart. No QA/QC issues or errors were identified in the 2022 Subpart FF data.
Recalculations Discussion
Time series recalculations were performed due to revised historical data from state natural gas sales databases for
five mines, which are used to estimate avoided Cm emissions. Additionally, calculation errors were identified and
corrected for CFU emissions avoided from two mines. As a result of recalculations, Cm emissions decreased by an
average of 0.03 percent across the time series, compared to the previous Inventory. The biggest increase in CFU
emissions was in 1991 where emissions increased by 0.14 percent, compared to the previous Inventory. The
biggest decrease in Cm emissions was in 2006 (0.6 percent). As a result of recalculations, there was a very minor
decrease in CFU emissions in 2021 (less than 0.005 percent), compared to the previous Inventory.
Planned Improvements
EPA is assessing planned improvements for future reports, but at this time has no specific planned improvements
for estimating Cm and CO2 emissions from underground and surface mining and CH4 emissions from post-mining.
3.5 Abandoned Underground Coal Mines
(CRT Source Category lBla)
Underground coal mines contribute the largest share of coal mine methane (CMM) emissions, with active
underground mines the leading source of underground emissions. However, mines also continue to release CH4
after closure. As mines mature and coal seams are mined through, mines are closed and abandoned. Many are
sealed and some flood through intrusion of groundwater or surface water into the void. Shafts or portals are
generally filled with gravel and capped with a concrete seal, while vent pipes and boreholes are plugged in a
manner similar to oil and gas wells. Some abandoned mines are vented to the atmosphere to prevent the buildup
of CH4 that may find its way to surface structures through overburden fractures. As work stops within the mines,
CH4 liberation decreases but it does not stop completely. Following an initial decline, abandoned mines can
liberate CH4 at a near-steady rate over an extended period of time, or if flooded, produce gas for only a few years.
The gas can migrate to the surface through the conduits described above, particularly if they have not been sealed
adequately. In addition, diffuse emissions can occur when CH4 migrates to the surface through cracks and fissures
in the strata overlying the coal mine. The following factors influence abandoned mine emissions:
• Time since abandonment;
• Gas content and adsorption characteristics of coal;
• CH4 flow capacity of the mine;
• Mine flooding;
• Presence of vent holes; and
• Mine seals.
Energy 3-69
-------
Annual gross abandoned mine CFU emissions ranged from 8.1 to 12.1 MMT CO2 Eq. from 1990 to 2022, varying, in
general, by less than 1 percent to approximately 19 percent from year to year. Fluctuations were due mainly to the
number of mines closed during a given year as well as the magnitude of the emissions from those mines when
active. Gross abandoned mine emissions peaked in 1996 (12.1 MMT CO2 Eq.) due to the large number of gassy
mine68 closures from 1994 to 1996 (72 gassy mines closed during the three-year period). In spite of this rapid rise,
abandoned mine emissions have been generally on the decline since 1996. Since 2002, there have been fewer than
twelve gassy mine closures each year. In 2022 there was one gassy mine closure. Gross abandoned mine emissions
decreased slightly from 9.2 MMT C02 Eq. (330 kt CH4) in 2021 to 9.1 (324 kt CH4) MMT C02 Eq. in 2022 (see Table
3-39 and Table 3-40). Gross emissions are reduced by CH4 recovered and used at 51 mines, resulting in net
emissions in 2022 of 6.3 MMT CO2 Eq. (225 kt CH4).
Table 3-39: CH4 Emissions from Abandoned Coal Mines (MMT CO2 Eq.)
Activity 1990 2005
2018
2019
2020
2021
2022
Abandoned Underground Mines 8.1 ¦ 9.3 1 9.9
Recovered & Used NO ¦ (2.0) ¦ (3.0)
9.6
(2.9)
9.4
(2.9)
9.2
(3.0)
9.1
(2.8)
Total 8.1 7.4
6.9
6.6
6.5
6.3
6.3
NO (Not Occurring)
Notes: Parentheses indicate negative values. Totals may not sum due to independent rounding.
Table 3-40: CH4 Emissions from Abandoned Coal Mines (kt)
Activity 1990 2005
2018
2019
2020
2021
2022
Abandoned Underground Mines 288 334
Recovered & Used NO 11 (70)
I 355
(107)
341
(104)
335
(103)
330
(106)
324
(100)
Total 288 264
247
237
232
224
225
NO (Not Occurring)
Note: Parentheses indicate negative values. Totals may not sum due to independent rounding.
Methodology and Time-Series Consistency
Estimating CFU emissions from an abandoned coal mine requires predicting the emissions of a mine from the time
of abandonment through the inventory year of interest. The flow of CH4 from the coal to the mine void is primarily
dependent on the mine's emissions when active and the extent to which the mine is flooded or sealed. The CH4
emission rate before abandonment reflects the gas content of the coal, the rate of coal mining, and the flow
capacity of the mine in much the same way as the initial rate of a water-free conventional gas well reflects the gas
content of the producing formation and the flow capacity of the well. A well or a mine that produces gas from a
coal seam and the surrounding strata will produce less gas through time as the reservoir of gas is depleted.
Depletion of a reservoir will follow a predictable pattern depending on the interplay of a variety of natural physical
conditions imposed on the reservoir. The depletion of a reservoir is commonly modeled by mathematical
equations and mapped as a type curve. Type curves, which are referred to as decline curves, have been developed
for abandoned coal mines. Existing data on abandoned mine emissions through time, although sparse, appear to
fit the hyperbolic type of decline curve used in forecasting production from natural gas wells.
There are sufficient mine level data available to establish decline curves for individual gassy mines abandoned
since 1972. For mines abandoned prior to 1972, county level data are available. Mine status information (i.e.,
whether a mine is sealed, venting, or flooded) is not available for all the abandoned gassy mines. Therefore, a
hybrid Tier 2/Tier 3 method was developed to model abandoned gassy mine emissions using Monte Carlo
simulations. Tier 3 calculations are used for mines with known status information where decline curves can be
used to directly estimate abandoned mine emissions. For mines with unknown status, a Tier 2 approach that
estimates basin level emissions is used. This Tier 2 approach relies on data from other mines with known status
68 A mine is considered a "gassy" mine if it emits more than 100 thousand cubic feet of CH4 per day (100 Mcfd).
3-70 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
and located within the same basin as the unknown status mines. This approach is consistent with the IPCC 2006
Guidelines as underground mines can be considered point sources and measurement methods are available.
To estimate Cm emissions over time for a given abandoned mine, it is necessary to apply a decline function,
initiated upon abandonment, to that mine. In the analysis, mines were grouped by coal basin with the assumption
that they will generally have the same initial pressures, permeability, and isotherm. As Cm leaves the system, the
reservoir pressure (Pr) declines as described by the isotherm's characteristics. The emission rate declines because
the mine pressure (Pw) is essentially constant at atmospheric pressure for a vented mine, and the productivity
index (PI), which is expressed as the flow rate per unit of pressure change, is essentially constant at the pressures
of interest (atmospheric to 30 psia). The Cm flow rate is determined by the laws of gas flow through porous media,
such as Darcy's Law. A rate-time equation can be generated that can be used to predict future emissions. This
decline through time is hyperbolic in nature and can be empirically expressed as:
Equation 3-3: Decline Function to Estimate Venting Abandoned Mine Methane Emissions
q = qt (1 + bDit^'b)
where,
q = Gas flow rate at time t in million cubic feet per day (mmcfd)
q, = Initial gas flow rate at time zero (t0), mmcfd
b = The hyperbolic exponent, dimensionless
Di = Initial decline rate, 1/year
t = Elapsed time from t0 (years)
This equation is applied to mines of various initial emission rates that have similar initial pressures, permeability,
and adsorption isotherms (EPA 2004).
The decline curves created to model the gas emission rate of coal mines must account for factors that decrease the
rate of emissions after mining activities cease, such as sealing and flooding. Based on field measurement data, it
was assumed that most U.S. mines prone to flooding will become completely flooded within eight years and
therefore will no longer have any measurable Cm emissions. Based on this assumption, an average decline rate for
flooded mines was established by fitting a decline curve to emissions from field measurements. An exponential
equation was developed from emissions data measured at eight abandoned mines known to be filling with water
located in two of the five basins. Using a least squares, curve-fitting algorithm, emissions data were matched to
the exponential equation shown below. For this analysis of flooded abandoned mines, there was not enough data
to establish basin-specific equations, as was done with the vented, non-flooding mines (EPA 2004). This decline
through time can be empirically expressed as:
Equation 3-4: Decline Function to Estimate Flooded Abandoned Mine Methane Emissions
q = qie{~Dt)
q = Gas flow rate at time t in mmcfd
q, = Initial gas flow rate at time zero (t0), mmcfd
D = Decline rate, 1/year
t = Elapsed time from t0 (years)
Seals have an inhibiting effect on the rate of flow of Cm into the atmosphere compared to the flow rate that
would exist if the mine had an open vent. The total volume emitted will be the same, but emissions will occur over
a longer period of time. The methodology, therefore, treats the emissions prediction from a sealed mine similarly
to the emissions prediction from a vented mine, but uses a lower initial rate depending on the degree of sealing. A
where,
Energy 3-71
-------
computational fluid dynamics simulator was used with the conceptual abandoned mine model to predict the
decline curve for inhibited flow. The percent sealed is defined as 100 x (1 - [initial emissions from sealed mine /
emission rate at abandonment prior to sealing]). Significant differences are seen between 50 percent, 80 percent,
and 95 percent closure. These decline curves were therefore used as the high, middle, and low values for
emissions from sealed mines (EPA 2004).
For active coal mines, those mines producing over 100 thousand cubic feet per day (Mcfd) of Cm account for
about 98 percent of all CFU emissions. This same relationship is assumed for abandoned mines. It was determined
that the 531 abandoned mines closed since 1972 produced Cm emissions greater than 100 Mcfd when active.
Further, the status of 307 of the 531 mines (or 58 percent) is known to be either: 1) vented to the atmosphere; 2)
sealed to some degree (either earthen or concrete seals); or 3) flooded (enough to inhibit Cm flow to the
atmosphere). The remaining 42 percent of the mines whose status is unknown were placed in one of these three
categories by applying a probability distribution analysis based on the known status of other mines located in the
same coal basin (EPA 2004). Table 3-41 presents the count of mines by post-abandonment state, based on EPA's
probability distribution analysis.
Table 3-41: Number of Gassy Abandoned Mines Present in U.S. Basins in 2022, Grouped by
Class According to Post-Abandonment State
Basin
Sealed
Vented
Flooded
Total Known
Unknown
Total Mines
Central Appl.
43
25
50
118
144
262
Illinois
35
3
14
52
31
83
Northern Appl.
49
23
15
87
39
126
Warrior Basin
0
0
16
16
0
16
Western Basins
28
4
2
34
10
44
Total
155
55
97
307
224
531
Note: Totals may not sum due to independent rounding.
Inputs to the decline equation require the average Cm emission rate prior to abandonment and the date of
abandonment. Generally, these data are available for mines abandoned after 1971; however, such data are largely
unknown for mines closed before 1972. Information that is readily available, such as coal production by state and
county, is helpful but does not provide enough data to directly employ the methodology used to calculate
emissions from mines abandoned before 1972. It is assumed that pre-1972 mines are governed by the same
physical, geologic, and hydrologic constraints that apply to post-1971 mines; thus, their emissions may be
characterized by the same decline curves.
During the 1970s, 78 percent of Cm emissions from coal mining came from seventeen counties in seven states.
Mine closure dates were obtained for two states, Colorado and Illinois, for the hundred-year period extending
from 1900 through 1999. The data were used to establish a frequency of mine closure histogram (by decade) and
applied to the other five states with gassy mine closures. As a result, basin-specific decline curve equations were
applied to the 145 gassy coal mines estimated to have closed between 1920 and 1971 in the United States,
representing 78 percent of the emissions. State-specific, initial emission rates were used based on average coal
mine CFU emission rates during the 1970s (EPA 2004).
Abandoned mine emission estimates are based on all closed mines known to have active mine Cm ventilation
emission rates greater than 100 Mcfd at the time of abandonment. For example, for 1990 the analysis included
145 mines closed before 1972 and 258 mines closed between 1972 and 1990. Initial emission rates based on MSHA
reports, time of abandonment, and basin-specific decline curves influenced by a number of factors were used to
calculate annual emissions for each mine in the database (MSHA 2023). Coal mine degasification data are not
available for years prior to 1990, thus the initial emission rates used reflect only ventilation emissions for pre-1990
closures. Methane degasification amounts were added to the quantity of Cm vented to determine the total CFU
liberation rate for all mines that closed between 1992 and 2022. Since the sample of gassy mines described above
is assumed to account for 78 percent of the pre-1972 and 98 percent of the post-1971 abandoned mine emissions,
the modeled results were multiplied by 1.22 and 1.02, respectively, to account for all U.S. abandoned mine
emissions.
3-72 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
From 1993 through 2022, emission totals were downwardly adjusted to reflect Cm emissions avoided from
abandoned mines with Cm recovery and use or destruction systems. Currently, there are 51 abandoned mines
with recovery projects, including 11 projects at mines abandoned before 1972 (pre-1972 mines) (EPA 2004, CMOP
2022). Because Cm recovered by these projects is expected to decline with the age of the mine, Cm recovery is
assumed to be the total estimated Cm liberated based on the mine's decline function except for three recovery
projects where additional data are available (COGIS 2018, MSHA 2023).69
The Inventory totals were not adjusted for abandoned mine Cm emissions avoided from 1990 through 1992 due to
unavailability of data. Avoided Cm emissions from pre-1972 abandoned mines are estimated by multiplying the
total estimated emissions from these mines in each decade by the fraction of mines with recovery projects in that
decade. For recovery projects at pre-1972 abandoned mines, four projects are at mines abandoned in the 1920s,
three in the 1930s, two in the 1950s, and two in the 1960s (EPA 2004).
Reviewing Coalbed Methane Outreach Program data (CMOP 2023) revealed four additional recovery projects
starting in 2021. In addition to reviewing CMOP data, the recovery project list was checked against the
International Coal Mine Methane Database (GMI 2021). Of the 24 operational recovery projects for U.S.
abandoned coal mines currently available in the GMI dataset, 18 are already included in the AMM model. The
remaining six projects in the GMI dataset are for mines that are not yet abandoned according to MSHA records
(MSHA 2023). Therefore, no new recovery projects were added from the GMI database for the 1990 through 2022
Inventory.
Uncertainty
A quantitative uncertainty analysis was conducted for the abandoned coal mine source category using the IPCC-
recommended Approach 2 uncertainty estimation methodology. The uncertainty analysis provides for the
specification of probability density functions for key variables within a computational structure that mirrors the
calculation of the Inventory estimate. The results provide the range within which, with 95 percent certainty,
emissions from this source category are likely to fall.
As discussed above, the parameters for which values must be estimated for each mine to predict its decline curve
are: 1) the coal's adsorption isotherm; 2) CH4 flow capacity as expressed by permeability; and 3) pressure at
abandonment. Because these parameters are not available for each mine, a methodological approach to
estimating emissions was used that generates a probability distribution of potential outcomes based on the most
likely value and the probable range of values for each parameter. The range of values is not meant to capture the
extreme values, but rather values that represent the highest and lowest quartile of the cumulative probability
density function of each parameter. Once the low, mid, and high values are selected, they are applied to a
probability density function.
The results of the Approach 2 quantitative uncertainty analysis are summarized in Table 3-42. Annual abandoned
coal mine CH4 emissions in 2022 were estimated to be between 5.0 and 7.5 MMT CO2 Eq. at a 95 percent
confidence level. This indicates a range of 21 percent below to 20 percent above the 2022 emission estimate of 6.3
MMT CO2 Eq. One of the reasons for the relatively narrow range is that mine-specific data is available for use in the
methodology for mines closed in 1972 and later years. Emissions from mines closed prior to 1972 have the largest
degree of uncertainty because no mine-specific CH4 liberation rates at the time of abandonment exist.
69 Data from a state oil and gas database (COGIS) is used for one project and the mine status information from MSHA for two
mines (sealed and flooded) indicate zero recovery emissions for these projects.
Energy 3-73
-------
Table 3-42: Approach 2 Quantitative Uncertainty Estimates for CH4 Emissions from
Abandoned Underground Coal Mines (MMT CO2 Eq. and Percent)
Source
Gas
2022 Emission Estimate
(MMT CO . Eq.)
Uncertainty Range Relative to Emission Estimate-1
(MMTCO' Eq.) (%)
Lower
Bound
Upper
Bound
Lower Upper
Bound Bound
Abandoned Underground
Coal Mines
ch4
6.3
5.0
7.5
-21% +20%
a Range of emission estimates predicted by Monte Carlo Simulation for a 95 percent confidence interval.
QA/QC and Verification
In order to ensure the quality of the emission estimates for abandoned coal mines, general (IPCC Tier 1) and
category-specific (Tier 2) Quality Assurance/Quality Control (QA/QC) procedures were implemented consistent
with the U.S. Inventory QA/QC plan outlined in Annex 8. The Tier 2 procedures that were implemented involved
checks specifically focusing on the activity data and reported emissions data used for estimating emissions from
abandoned coal mines. Trends across the time series were analyzed to determine whether any corrective actions
were needed.
Recalculations Discussion
Four new abandoned mine methane recovery projects were added to the AMM model during the current
Inventory (CMOP 2023). CMOP data indicate these recovery projects were started in 2021. Time series
recalculations were performed for 2021. As a result of recalculations, CH4 emissions decreased by 2 percent in
2021, compared to the previous Inventory.
3.6 Petroleum Systems (CRT Source
Category lB2a)
This category (lB2a) is defined in the IPCC methodological guidance as fugitive emissions from petroleum systems,
which per IPCC guidelines include emissions from leaks, venting, and flaring. Methane emissions from petroleum
systems are primarily associated with onshore and offshore crude oil exploration, production, transportation, and
refining operations. During these activities, Cm is released to the atmosphere as emissions from leaks, venting
(including emissions from operational upsets), and flaring. Carbon dioxide emissions from petroleum systems are
primarily associated with onshore and offshore crude oil production and refining operations. Note, CO2 emissions
in petroleum systems exclude all combustion emissions (e.g., engine combustion) except for flaring CO2 emissions.
All combustion CO2 emissions (except for flaring) are accounted for in the fossil fuel combustion chapter (see
Section 3.1). Emissions of N2O from petroleum systems are primarily associated with flaring. Total greenhouse gas
emissions (Cm, CO2, and N2O) from petroleum systems in 2022 were 61.6 MMT CO2 Eq., an increase of 4 percent
from 1990, primarily due to increases in CO2 emissions. Total emissions decreased by 6 percent from 2010 levels
and have decreased by 15 percent since 2021. Total CO2 emissions from petroleum systems in 2022 were 21.97
MMT CO2 (21,967 kt CO2), 2.3 times higher than in 1990. Total CO2 emissions in 2022 were 1.6 times higher than in
2010 and 9 percent lower than in 2021. Total Cm emissions from petroleum systems in 2022 were 39.6 MMT CO2
Eq. (1,415 kt CH4), a decrease of 20 percent from 1990. Since 2010, total CH4 emissions decreased by 24 percent;
and since 2021, CH4 emissions decreased by 19 percent. Total N2O emissions from petroleum systems in 2022
were 0.048 MMT CO2 Eq. (0.179 kt N2O), 3.8 times higher than in 1990, 2.8 times higher than in 2010, and 142
percent higher than in 2021. Since 1990, U.S. oil production has increased by 56 percent. In 2022, U.S. oil
production was 163 percent higher than in 2010 and 7 percent higher than in 2021.
3-74 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
Each year, some estimates in the Inventory are recalculated with improved methods and/or data. These
improvements are implemented consistently across the entire Inventory's time series (i.e., 1990 to 2022) to ensure
that the trend is representative of changes in emissions levels. Recalculations in petroleum systems in this year's
Inventory include:
• Updates to oil and gas production volumes, produced water production volumes, and well counts using
the most recent data from Enverus.
• Updates to oil and gas production volumes using the most recent data from the United States Energy
Information Administration (EIA)
• Recalculations due to Greenhouse Gas Reporting Program (GHGRP) submission revisions
• Recalculations due to methodological updates to completions and workovers.
The Recalculations Discussion section below provides more details on the updated methods.
Exploration. Exploration includes well drilling, testing, and completions. Exploration accounts for less than 0.5
percent of total Cm emissions (including leaks, vents, and flaring) from petroleum systems in 2022. The
predominant sources of Cm emissions from exploration are hydraulically fractured oil well completions. Other
sources include well testing, well drilling, and well completions without hydraulic fracturing. Since 1990,
exploration Cm emissions have decreased 96 percent, and while the number of hydraulically fractured wells
completed increased 64 percent, there were decreases in the fraction of such completions without reduced
emissions completions (RECs) or flaring. Emissions of Cm from exploration were highest in 2008, over 70 times
higher than in 2022; and lowest in 2022. Emissions of Cm from exploration decreased 39 percent from 2021 to
2022, due to a decrease in emissions from hydraulically fractured oil well completions with RECs. Exploration
accounts for 1 percent of total CO2 emissions (including leaks, vents, and flaring) from petroleum systems in 2022.
Emissions of CO2 from exploration in 2022 were 25 percent lower than in 1990, and decreased by 50 percent from
2021, largely due to a decrease in emissions from hydraulically fractured oil well completions with REC and flaring
(by 58 percent from 2021). Emissions of CO2 from exploration were highest in 2014, over 13 times higher than in
2022. Exploration accounts for less than 0.5 percent of total N2O emissions from petroleum systems in 2022.
Emissions of N2O from exploration in 2022 are 29 percent lower than in 1990, and 56 percent lower than in 2021,
due to hydraulically fractured oil well completions with flaring.
Production. Production accounts for 97 percent of total CH4 emissions (including leaks, vents, and flaring) from
petroleum systems in 2022. The predominant sources of emissions from production field operations are pneumatic
controllers, offshore oil platforms, equipment leaks, produced water, gas engines, chemical injection pumps, and
associated gas flaring. In 2022, these seven sources together accounted for 93 percent of the CH4 emissions from
production. Since 1990, CH4 emissions from production have decreased by 15 percent primarily due to decreases in
emissions from offshore production. Overall, production segment CH4 emissions decreased by 19 percent from
2021 levels due primarily to lower pneumatic controller emissions. Production emissions account for 86 percent of
the total CO2 emissions (including leaks, vents, and flaring) from petroleum systems in 2022. The principal sources
of CO2 emissions are associated gas flaring, miscellaneous production flaring, and oil tanks with flares. In 2022,
these three sources together accounted for 96 percent of the CO2 emissions from production. In 2022, CO2
emissions from production were 3.1 times higher than in 1990, due to increases in flaring emissions from
associated gas flaring, miscellaneous production flaring, and tanks. Overall, in 2022, production segment CO2
emissions decreased by 8 percent from 2021 levels primarily due to decreases in associated gas flaring in the
Williston Basin and oil tanks with flares. Production emissions accounted for 84 percent of the total N2O emissions
from petroleum systems in 2022. The principal sources of N2O emissions are oil tanks with flares and associated
gas flaring, accounting for 90% of N2O emissions from the production segment in 2022. In 2022, N2O emissions
from production were 8.0 times higher than in 1990 and were 3.5 times higher than in 2021.
Crude Oil Transportation. Emissions from crude oil transportation account for a very small percentage of the total
emissions (including leaks, vents, and flaring) from petroleum systems. Crude oil transportation activities account
for 0.6 percent of total CH4 emissions from petroleum systems. Emissions from tanks, marine loading, and truck
loading operations accounted for 81 percent of Cm emissions from crude oil transportation in 2022. Since 1990,
Energy 3-75
-------
CH4 emissions from transportation have increased by 27 percent. In 2022, CH4 emissions from transportation
increased by 6 percent from 2021 levels. Crude oil transportation activities account for less than 0.01 percent of
total CO2 emissions from petroleum systems. Emissions from tanks, marine loading, and truck loading operations
account for 81 percent of CO2 emissions from crude oil transportation.
Crude Oil Refining. Crude oil refining processes and systems account for 2 percent of total fugitive (including leaks,
vents, and flaring) CH4 emissions from petroleum systems in 2022. This low share is because most of the CH4 in
crude oil is removed or escapes before the crude oil is delivered to the refineries. There is a negligible amount of
CH4 in all refined products. Within refineries, flaring accounts for 42 percent of the CH4 emissions, while delayed
cokers, uncontrolled blowdowns, and equipment leaks account for 19,14 and 11 percent, respectively. Fugitive
CH4 emissions from refining of crude oil have decreased by 4 percent since 1990, and decreased by 3 percent from
2021; however, like the transportation subcategory, this increase has had little effect on the overall emissions of
CH4 from petroleum systems. Crude oil refining processes and systems account for 13 percent of total fugitive
(including leaks, vents, and flaring) CO2 emissions from petroleum systems. Of the total fugitive CO2 emissions
from refining, almost all (about 99 percent) of it comes from flaring.70 Since 1990, refinery fugitive CO2 emissions
decreased by 10 percent and have decreased by 5 percent from 2021 levels, due to a decrease in flaring. Flaring
occurring at crude oil refining processes and systems accounts for 16 percent of total fugitive N2O emissions from
petroleum systems. In 2022, refinery fugitive N2O emissions increased by 4 percent since 1990 and decreased by 5
percent from 2021 levels.
Table 3-43: Total Greenhouse Gas Emissions (CO2, CH4, and N2O) from Petroleum Systems
(MMT C02 Eq.)
Activity
1990
2005
2018
2019
2020
2021
2022
Exploration
3.4
5.7 1
3.2
2.5
1.1
0.8
0.4
Production
51-6
48.1 1
86.7
90.6
77.3
68.0
57.4
Transportation
0.2
0.1 1
0.2
0.3
0.2
0.2
0.2
Crude Refining
3.9:
11111
-------
Table 3-45: CH4 Emissions from Petroleum Systems (kt CH4)
Activity
1990
2005
2018
2019
2020
2021
2022
Exploration
106
189
19
16
12
7
4
Production
1,626
1,500
2,052
1,809
1,858
1,696
1,377
Pneumatic Controllers
761
811
1,240
881
1,119
1,003
694
Offshore Production
353
" 259 2
197
193
183
182
182
Equipment Leaks
82
101
132
138
115
114
112
Gas Engines
81
70 5!
91
93
89
87
88
Produced Water
92
64
95
99
90
92
95
Chemical Injection Pumps
47 ¦! 80 j
108
123
96
85
80
Assoc Gas Flaring
20
14
66
91
47
35
30
Other Sources
189
100
124
190
119
97
98
Crude Oil Transportation
7
5
8
9
8
8
8
Refining
Hi 29 |,
28
31
26
25
25
Total
1,765 1,723
2,108
1,865
1,904
1,737
1,415
Note: Totals may not sum due to independent rounding.
Table 3-46: CO2 Emissions from Petroleum Systems (MMT CO2)
Activity
1990
2005
2018
2019
2020
2021
2022
Exploration
0.4
I 0.51
2.7
2.0
0.8
0.6
0.3
Production
6.0 6.1 1
29.2
39.9
25.2
20.5
18.8
Transportation
+
1 + a
+
+
+
+
+
Crude Refining
3.2 ¦ 3.6 ¦
2 Q
3.6
2.9
3.0
2 Q
Total
9.6
10.2
34.8
45.5
28.9
24.1
22.0
+ Does not exceed 0.05 MMT C02
Note: Totals may not sum due to independent rounding.
Table 3-47: CO2 Emissions from Petroleum Systems (kt CO2)
Activity
1990
2005
2018
2019
2020
2021
2022
Exploration
398
465 1
2,684
2,044
798
601
300
Production
6,012
6,143 |
29,215
39,882
25,244
20,516
18,793
Transportation
0.9
0.7
1.2
1.3
1.2
1.1
1.2
Crude Refining
3,174
3,602 ¦
2,877
3,571
2,893
3,021
2,872
Total
9,585
1! i
10,210
34,777
45,498
28,937
24,140
21,967
Table 3-48: N2O Emissions from Petroleum Systems (Metric Tons CO2 Eq.)
Activity
1990
2005
2018
2019
2020
2021
2022
Exploration
180
209
1,161
820
353
290
127
Production
4,996:::
4,588
30,822
28,047
13,614
11,414
39,859
Transportation
NE
NE
NE
NE
NE
NE
NE
Crude Refining
7,262 i
8,243 1
7,405
9,312
7,575
7,920
7,523
Total
12,438
13,040
39,387
38,180
21,542
19,624
47,510
NE (Not Estimated)
Note: Totals may not sum due to independent rounding.
Energy 3-77
-------
Table 3-49: N2O Emissions from Petroleum Systems (Metric Tons N2O)
Activity
1990
2005
2018
2019
2020
2021
2022
Exploration
Production
Transportation
Crude Refining
0.7
18.9
NE
27.4 [
°'8l
17.3 1
ne[
31-11
4.4
116.3
NE
27.9
3.1
105.8
NE
35.1
1.3
51.4
NE
28.6
1.1
43.1
NE
29.9
0.5
150.4
NE
28.4
Total
46.9
49.2
148.6
144.1
81.3
74.1
179.3
NE (Not Estimated)
Note: Totals may not sum due to independent rounding.
Methodology and Time-Series Consistency
See Annex 3.5 for the full time series of emissions data, activity data, emission factors, and additional information
on methods and data sources.
Petroleum systems includes emission estimates for activities occurring in petroleum systems from the oil wellhead
through crude oil refining, including activities for crude oil exploration, production field operations, crude oil
transportation activities, and refining operations. Generally, emissions are estimated for each activity by
multiplying emission factors (e.g., emission rate per equipment or per activity) by corresponding activity data (e.g.,
equipment count or frequency of activity). Certain sources within petroleum refineries are developed using an
IPCC Tier 3 approach (i.e., all refineries in the nation report facility-level emissions data to the GHGRP, which are
included directly in the national emissions estimates here). Other estimates are developed with a Tier 2 approach.
Tier 1 approaches are not used.
EPA did not receive stakeholder feedback on updates in the Inventory through EPA's stakeholder process on oil
and gas in the Inventory. More information on the stakeholder process can be found online.71
Emission Factors. Key references for emission factors include Methane Emissions from the Natural Gas Industry by
the Gas Research Institute and EPA (GRI/EPA 1996), Estimates of Methane Emissions from the U.S. Oil Industry (EPA
1999), Compilation of Air Pollutant Emission Factors, AP-42 (EPA 1997), Global Emissions of Methane from
Petroleum Sources (API 1992), consensus of industry peer review panels, Bureau of Ocean Energy Management
(BOEM) reports, Nonpoint Oil and Gas Emission Estimation Tool (EPA 2017), and analysis of GHGRP data (EPA
2023).
Emission factors for hydraulically fractured (HF) oil well completions and workovers (in four control categories)
were developed at the basin level using EPA's GHGRP data; year-specific data were used to calculate basin-specific
emission factors from 2016-forward and the year 2016 emission factors were applied to all prior years in the time
series. For basins not reporting to the GHGRP, Subpart W average emission factors were used. For more
information, please see the 2023 memoranda available online.72
The emission factors for well testing and associated gas venting and flaring were developed using year-specific
GHGRP data for years 2015 forward; earlier years in the time series use 2015 emission factors. For miscellaneous
production flaring, year-specific emission factors were developed for years 2015 forward from GHGRP data, an
emission factor of 0 (assumption of no flaring) was assumed for 1990 through 1992, and linear interpolation was
applied to develop emission factors for 1993 through 2014. For more information, please see memoranda
available online.73 For offshore oil production, emission factors were calculated using BOEM data for offshore
facilities in federal waters of the Gulf of Mexico (and these data were also applied to facilities located in state
waters of the Gulf of Mexico) and GHGRP data for offshore facilities off the coasts of California and Alaska. For
71 See https://www.epa.gov/ghgemissions/natural-gas-and-petroleum-systems.
72 See https://www.epa.gov/ghgemissions/natural-gas-and-petroleum-systems.
73 See https://www.epa.gov/ghgemissions/natural-gas-and-petroleum-systems.
3-78 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
many other sources, emission factors were held constant for the period 1990 through 2022, and trends in
emissions reflect changes in activity levels. Emission factors from EPA 1999 are used for all other production and
transportation activities.
For associated gas venting and flaring and miscellaneous production flaring, emission factors were developed on a
production basis (i.e., emissions per unit oil produced). Additionally, for these two sources, basin-specific activity
and emission factors were developed for each basin that in any year from 2011 forward contributed at least 10
percent of total source emissions (on a CO2 Eq. basis) in the GHGRP. For associated gas venting and flaring, basin-
specific factors were developed for four basins: Williston, Permian, Gulf Coast, and Anadarko. For miscellaneous
production flaring, basin-specific factors were developed for three basins: Williston, Permian, and Gulf Coast. For
each source, data from all other basins were combined, and activity and emission factors were developed for the
other basins as a single group.
For pneumatic controllers and tanks, basin-specific emission factors were calculated for all the basins reporting to
the GHGRP. These emission factors were calculated for all the years with applicable GHGRP data (i.e., 2011 - 2022
or 2015 - 2022). For the remaining basins (i.e., basins not reporting to the GHGRP), Subpart W average emission
factors were used. For more information, please see memoranda available online.3
For the exploration and production segments, in general, CO2 emissions for each source were estimated with
GHGRP data or by multiplying CO2 content factors by the corresponding CH4 data, as the CO2 content of gas relates
to its CH4 content. Sources with CO2 emission estimates calculated using GHGRP data include HF completions and
workovers, associated gas venting and flaring, tanks, well testing, pneumatic controllers, chemical injection pumps,
miscellaneous production flaring, and certain offshore production facilities (those located off the coasts of
California and Alaska). For these sources, CO2 was calculated using the same methods as used for CH4. Carbon
dioxide emission factors for offshore oil production in the Gulf of Mexico were derived using data from BOEM,
following the same methods as used for CH4 estimates. For other sources, the production field operations emission
factors for CO2 are generally estimated by multiplying the CH4 emission factors by a conversion factor, which is the
ratio of CO2 content and CH4 content in produced associated gas.
For the exploration and production segments, N2O emissions were estimated for flaring sources using GHGRP or
BOEM OGOR-B data and the same method used for CO2. Sources with N2O emissions in the exploration segment
include well testing and HF completions with flaring. Sources with N2O emissions in the production segment
include associated gas flaring, tank flaring, miscellaneous production flaring, HF workovers with flaring, and flaring
from offshore production sources.
For crude oil transportation, emission factors for CH4 were largely developed using data from EPA (1997), API
(1992), and EPA (1999). Emission factors for CO2 were estimated by multiplying the CH4 emission factors by a
conversion factor, which is the ratio of CO2 content and CH4 content in whole crude post-separator.
For petroleum refining activities, year-specific emissions from 2010 forward were directly obtained from EPA's
GHGRP. All U.S. refineries have been required to report CH4, CO2, and N2O emissions for all major activities starting
with emissions that occurred in 2010. The reported total CH4, CO2, and N2O emissions for each activity was used
for the emissions in each year from 2010 forward. To estimate emissions for 1990 to 2009, the 2010 to 2013
emissions data from GHGRP along with the refinery feed data for 2010 to 2013 were used to derive CH4 and CO2
emission factors (i.e., sum of activity emissions/sum of refinery feed) and 2010 to 2017 data were used to derive
N2O emission factors; these emission factors were then applied to the annual refinery feed in years 1990 to 2009.
GHGRP delayed coker CH4 emissions for 2010 through 2017 were increased using the ratio of certain reported
emissions for 2018 to 2017, to account for a more accurate GHGRP calculation methodology that was
implemented starting in reporting year 2018.
A complete list of references for emission factors and activity data by emission source is provided in Annex 3.5.
Activity Data. References for activity data include Enverus data (Enverus 2023), Energy Information Administration
(EIA) reports, Methane Emissions from the Natural Gas Industry by the Gas Research Institute and EPA (EPA/GRI
1996), Estimates of Methane Emissions from the U.S. Oil Industry (EPA 1999), consensus of industry peer review
panels, BOEM reports, the Oil & Gas Journal, the Interstate Oil and Gas Compact Commission, the United States
Army Corps of Engineers, and analysis of GHGRP data (EPA 2023).
Energy 3-79
-------
For HF oil well completions and workovers, pneumatic controllers, equipment leaks, chemical injection pumps, and
tanks, basin-specific activity factors were calculated for all the basins reporting to the GHGRP. These factors were
calculated for all the years with applicable GHGRP data (i.e., 2011 through 2022, 2016 through 2022, or 2015
through 2022). For the remaining basins (i.e., basins not reporting to the GHGRP), GHGRP average activity factors
were used. For more information, please see memoranda available online.74
For many sources, complete activity data were not available for all years of the time series. In such cases, one of
three approaches was employed to estimate values, consistent with IPCC good practice. Where appropriate, the
activity data were calculated from related statistics using ratios developed based on EPA/GRI (1996) and/or GHGRP
data. In some cases, activity data are developed by interpolating between recent data points (such as from GHGRP)
and earlier data points, such as from EPA/GRI (1996). Lastly, in limited instances the previous year's data were
used if current year data were not yet available.
A complete list of references for emission factors and activity data by emission source is provided in Annex 3.5. The
United States reports data to the UNFCCC using this Inventory report along with Common Reporting Tables (CRTs).
This note is provided for those reviewing the CRTs: The notation key "IE" is used for CO2 and CH4 emissions from
venting and flaring in CRT l.B.2. Disaggregating flaring and venting estimates across the Inventory would involve
the application of assumptions and could result in inconsistent reporting and, potentially, decreased transparency.
Data availability varies across segments within oil and gas activities systems, and emission factor data available for
activities that include flaring can include emissions from multiple sources (flaring, venting and leaks).
As noted above, EPA's GHGRP data, available starting in 2010 for refineries and in 2011 for other sources, have
improved estimates of emissions from petroleum systems. Many of the previously available datasets were
collected in the 1990s. To develop a consistent time series for sources with new data, EPA reviewed available
information on factors that may have resulted in changes over the time series (e.g., regulations, voluntary actions)
and requested stakeholder feedback on trends as well. For most sources, EPA developed annual data for 1993
through 2009 or 2014 by interpolating activity data or emission factors or both between 1992 (when GRI/EPA data
are available) and 2010 or 2015 data points. Information on time-series consistency for sources updated in this
year's Inventory can be found in the Recalculations Discussion below, with additional detail provided in supporting
memos (relevant memos are cited in the Recalculations Discussion). For information on other sources, please see
the Methodology and Time-Series Consistency discussion above and Annex 3.5.
Methodological recalculations were applied to the entire time series to ensure time-series consistency from 1990
through 2022.
Uncertainty
EPA conducted a quantitative uncertainty analysis using the IPCC Approach 2 methodology (Monte Carlo
Simulation technique) to characterize uncertainty for petroleum systems. For more information on the approach,
please see the memoranda Inventory of U.S. Greenhouse Gas Emissions and Sinks 1990-2016: Natural Gas and
Petroleum Systems Uncertainty Estimates (2018 uncertainty memo) and Inventory of U.S. Greenhouse Gas
Emissions and Sinks 1990-2019: Update for Natural Gas and Petroleum Systems CO2 Uncertainty Estimates (2021
uncertainty memo).75
EPA used Microsoft Excel's @ RISK add-in tool to estimate the 95 percent confidence bound around CH4 and CO2
emissions from petroleum systems for the current Inventory. For the CH4 uncertainty analysis, EPA focused on the
eight highest methane-emitting sources for the year 2022, which together emitted 75 percent of methane from
petroleum systems in 2022, and extrapolated the estimated uncertainty for the remaining sources. For the CO2
uncertainty analysis, EPA focused on the five highest-emitting sources for the year 2022 which together emitted 81
percent of CO2 from petroleum systems in 2022, and extrapolated the estimated uncertainty for the remaining
74 See https://www.epa.gov/ghgemissions/natural-gas-and-petroleum-systems.
75 See https://www.epa.gov/ghgemissions/natural-gas-and-petroleum-svstems.
3-80 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
sources. The @RISK add-in provides for the specification of probability density functions (PDFs) for key variables
within a computational structure that mirrors the calculation of the inventory estimate. For emission factors that
are derived from methane emissions measurement studies, the PDFs are commonly determined to be lognormally
distributed (GRI/EPA 1996; EPA 1999). For activity data that are derived from national datasets, the PDFs are set to
a uniform distribution (see 2018 and 2021 uncertainty memos). Many emission factors and activity factors are
calculated using subpart W data, and for these, the @RISK add-in determines the best fitting PDF (e.g., lognormal,
gaussian), based on bootstrapping of the underlying data (see 2018 and 2021 uncertainty memos). The IPCC
guidance notes that in using this Approach 2 method, "some uncertainties that are not addressed by statistical
means may exist, including those arising from omissions or double counting, or other conceptual errors, or from
incomplete understanding of the processes that may lead to inaccuracies in estimates developed from models." As
a result, the understanding of the uncertainty of emission estimates for this category evolves and improves as the
underlying methodologies and datasets improve. The uncertainty bounds reported below only reflect those
uncertainties that EPA has been able to quantify and do not incorporate considerations such as modeling
uncertainty, data representativeness, measurement errors, misreporting or misclassification. To estimate
uncertainty for N2O, EPA applied the uncertainty bounds calculated for CO2. EPA will seek to refine this estimate in
future Inventories.
The results presented below provide the 95 percent confidence bound within which actual emissions from this
source category are likely to fall for the year 2022, using the recommended IPCC methodology. The results of the
Approach 2 uncertainty analysis are summarized in Table 3-50. Petroleum systems CFU emissions in 2022 were
estimated to be between 32.7 and 48.6 MMT CO2 Eq., while CO2 emissions were estimated to be between 17.9
and 27.4 MMT CO2 Eq. at a 95 percent confidence level. Petroleum systems N2O emissions in 2022 were estimated
to be between 0.039 and 0.059 MMT CO2 Eq. at a 95 percent confidence level.
Uncertainty bounds for other years of the time series have not been calculated, but uncertainty is expected to vary
over the time series. For example, years where many emission sources are calculated with interpolated data would
likely have higher uncertainty than years with predominantly year-specific data. In addition, the emission sources
that contribute the most to CFU and CO2 emissions are different over the time series, particularly when comparing
recent years to early years in the time series. For example, associated gas venting emissions were higher and
flaring emissions were lower in early years of the time series, compared to recent years. Technologies also
changed over the time series (e.g., reduced emissions completions were not used early in the time series).
Table 3-50: Approach 2 Quantitative Uncertainty Estimates for CH4 and CO2 Emissions from
Petroleum Systems (MMT CO2 Eq. and Percent)
Source
Gas
2022 Emission Estimate
(MMTCO. Eq.)1,
Uncertainty Range Relative to Emission Estimate-1
(MMTCO' Eq.) (%)
Lower Upper
Lower
Upper
Bound Bound
Bound
Bound
Petroleum Systems
ch4
39.6
32.7 48.6
-18%
+23%
Petroleum Systems
C02
22.0
17.9 27.4
-19%
+25%
Petroleum Systems
n2o
0.048
0.039 0.059
-19%
+25%
a Range of emission estimates estimated by applying the 95 percent confidence intervals obtained from the Monte Carlo
Simulation analysis conducted for the year 2022 CH4 and C02 emissions.
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.
QA/QC and Verification Discussion
In order to ensure the quality of the emission estimates for petroleum systems, general (IPCC Tier 1) Quality
Assurance/Quality Control (QA/QC) procedures were implemented consistent with the U.S. Inventory QA/QC plan
outlined in Annex 8.
Energy 3-81
-------
The petroleum systems emission estimates in the Inventory are continually being reviewed and assessed to
determine whether emission factors and activity factors accurately reflect current industry practices. A QA/QC
analysis was performed for data gathering and input, documentation, and calculation. QA/QC checks are
consistently conducted to minimize human error in the emission calculations. EPA performs a thorough review of
information associated with new studies, GHGRP data, regulations, public webcasts, and the Natural Gas STAR
Program to assess whether the assumptions in the Inventory are consistent with current industry practices. EPA
has a multi-step data verification process for GHGRP data, including automatic checks during data-entry, statistical
analyses on completed reports, and staff review of the reported data. Based on the results of the verification
process, EPA follows up with facilities to resolve mistakes that may have occurred.76
As in previous years, EPA conducted early engagement and communication with stakeholders on updates prior to
public review of the current Inventory. EPA held a stakeholder webinar on greenhouse gas data for oil and gas in
October of 2023. EPA released memos detailing updates under consideration and requesting stakeholder
feedback. EPA did not receive stakeholder feedback for the updates under consideration for the current Inventory.
In recent years, several studies have measured emissions at the source level and at the national or regional level
and calculated emission estimates that may differ from the Inventory. There are a variety of potential uses of data
from new studies, including replacing a previous estimate or factor, verifying or QA of an existing estimate or
factor, and identifying areas for updates. In general, there are two major types of studies related to oil and gas
greenhouse gas data: studies that focus on measurement or quantification of emissions from specific activities,
processes, and equipment, and studies that use tools such as inverse modeling to estimate the level of overall
emissions needed to account for measured atmospheric concentrations of greenhouse gases at various scales. The
first type of study can lead to direct improvements to or verification of Inventory estimates. In the past few years,
EPA has reviewed, and in many cases, incorporated data from these data sources. The second type of study can
provide general indications on potential over- and under-estimates.
A key challenge in using these types of studies to assess Inventory results is having a relevant basis for comparison
(e.g., the two data sets should have comparable time frames and geographic coverage, and the independent study
should assess data from the Inventory and not another data set, such as the Emissions Database for Global
Atmospheric Research, or "EDGAR"). In an effort to improve the ability to compare the national-level Inventory
with measurement results that may be at other spatial and temporal scales, EPA has developed a gridded
inventory of U.S. anthropogenic methane emissions with 0.1 degree x 0.1 degree spatial resolution, monthly
temporal resolution, and detailed scale-dependent error characterization.77 The most recent version of the
gridded methane inventory is designed to be consistent with the U.S. EPA's Inventory of U.S. Greenhouse Gas
Emissions and Sinks: 1990-2018 estimates for the years 2012 through 2018. The gridded inventory improves efforts
to compare results of this Inventory with atmospheric studies.
As discussed above, refinery emissions are quantified by using the total emissions reported to GHGRP for the
refinery emission categories included in petroleum systems. Subpart Y has provisions that refineries are not
required to report under Subpart Y if their emissions fall below certain thresholds. Each year, a review is conducted
to determine whether an adjustment is needed to the Inventory emissions to include emissions from refineries
that stopped reporting to the GHGRP. Based on the review of the most recent GHGRP data, EPA did not identify
any additional refineries that would require gap filling. There are a total of 7 refineries that EPA previously
identified (i.e., during the 1990 through 2021 Inventory and prior versions) as not reporting to the GHGRP and
continued to gap fill annual emissions for these refineries. EPA used the last reported emissions (by source) for
these refineries as proxy to gap fill annual emissions.
76 See https://www.epa.gov/sites/production/files/2015-07/documents/ghgrp verification factsheet.pdf.
77 See https://www.epa.gov/eheemissions/us-eridded-methane-emissions.
3-82 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
Recalculations Discussion
EPA received information and data related to the emission estimates through GHGRP reporting and presented
information to stakeholders regarding the updates under consideration. In November 2023, EPA released draft
memoranda that discussed changes under consideration and requested stakeholder feedback on those changes.
EPA then released final memoranda documenting the methodology implemented in the current Inventory.78 The
memorandum cited in the Recalculations Discussion below is: Inventory of U.S. Greenhouse Gas Emissions and
Sinks 1990-2022: Updates for Completion and Workover Emissions (Completions and Workovers memo).
EPA evaluated relevant information available and made an updates to the Inventory for hydraulically fractured (HF)
oil well completions and workovers. General information for these source specific recalculations are presented
below and details are available in the Completions and Workovers memo.
In addition to the updates to the sources mentioned above, for certain sources, CFU and/or CO2 emissions changed
by greater than 0.05 MMT CO2 Eq., comparing the previous estimate for 2021 to the current (recalculated)
estimate for 2021. The emissions changes were mostly due to GHGRP data submission revisions. These sources are
discussed below and include associated gas flaring, miscellaneous production flaring, pneumatic controllers, oil
tanks, chemical injection pumps, produced water, offshore production (in Gulf of Mexico federal waters), gas
engines, and refinery flaring.
The combined impact of revisions to 2021 petroleum systems CFU emission estimates on a CCh-equivalent basis,
compared to the previous Inventory, is a decrease from 50.2 to 48.6 MMT CO2 Eq. (1.5 MMT CO2 Eq., or 3 percent).
The recalculations resulted in lower CFU emission estimates on average across the 1990 through 2021 time series,
compared to the previous Inventory, by 2.5 MMT CO2 Eq., or 5 percent.
The combined impact of revisions to 2021 petroleum systems CO2 emission estimates, compared to the previous
Inventory, is a decrease from 24.7 to 24.1 MMT CO2 (0.5 MMT CO2, or 2 percent). The recalculations resulted in
lower emission estimates on average across the 1990 through 2021 time series, compared to the previous
Inventory, by 0.1 MMT CO2 Eq., or 0.2 percent.
The combined impact of revisions to 2021 petroleum systems N2O emission estimates on a CC>2-equivalent basis,
compared to the previous Inventory, is a decrease of 0.002 MMT CO2, Eq. or 10.2 percent. The recalculations
resulted in an average decrease in emission estimates across the 1990 through 2021 time series, compared to the
previous Inventory, of 0.001 MMT CO2 Eq., or 7.4 percent.
In Table 3-51 and Table 3-52 below are categories in petroleum systems with updated methodologies or with
recalculations resulting in a change of greater than 0.05 MMT CO2 Eq., comparing the previous estimate for 2021
to the current (recalculated) estimate for 2021. For more information, please see the discussion below.
Table 3-51: Recalculations of CO2 in Petroleum Systems (MMT CO2)
Previous Estimate Year
Current Estimate Year
Current Estimate
2021,
2021,
Year 2022,
Segment/Source
2023 Inventory
2024 Inventory
2024 Inventory
Exploration
0.5
0.6
0.3
HF Completions
0.5
0.6
0.3
Production
20.0
20.5
18.8
Tanks
5.4
5.6
4.5
HF Workovers
0.2
+
+
Pneumatic Controllers
0.1
0.1
0.1
Equipment Leaks
+
+
+
Chemical Injection Pumps
+
+
+
78 Stakeholder materials including draft and final memoranda for the current (i.e., 1990 to 2022) Inventory are available at
https://www.epa.gov/eheemissions/natural-eas-and-petroleum-svstems.
Energy 3-83
-------
Miscellaneous Production Flaring
4.2
4.6
5.0
Transportation
+
+
+
Refining
4.2
3.0
2.9
Flares
4.2
3.0
2.8
Petroleum Systems Total
24.7
24.1
22.0
+ Does not exceed 0.05 MMT C02 Eq.
Note: Totals may not sum due to independent rounding.
Table 3-52: Recalculations of CH4 in Petroleum Systems (MMT CO2 Eq.)
Previous Estimate Year
Current Estimate Year
Current Estimate
2021,
2021,
Year 2022,
Segment/Source
2023 Inventory
2024 Inventory
2024 Inventory
Exploration
0.2
0.2
0.1
HF Completions
0.1
0.2
0.1
Production
48.9
47.5
38.6
Pneumatic Controllers
28.4
28.1
19.4
Chemical Injection Pumps
3.2
2.4
2.2
Produced Water
2.5
2.6
2.7
Offshore Production from GOM Federal
A 7
A A
A 3
Waters (vented and leaks)
'f.O
HF Workovers
0.1
+
+
Gas Engines
2.5
2.4
2.5
Associated Gas Flaring
0.8
1.0
0.8
Transportation
0.2
0.2
0.2
Refining
0.8
0.7
0.7
Petroleum Systems Total
50.2
48.6
39.6
+ Does not exceed 0.05 MMT C02 Eq.
Note: Totals may not sum due to independent rounding.
Exploration
HF Completions (Methodological Update)
EPA updated the calculation methodology for HF completions to use basin-level HF completion counts from
Enverus and basin-specific activity factors and emission factors calculated from Subpart W data for each control
category (i.e., non-reduced emission completion (REC) with venting, non-REC with flaring, REC with venting, REC
with flaring). Previously, national annual average activity and emission factors calculated using Subpart W data
were applied to national activity data counts to estimate HF gas well completion emissions. In this update, EPA
developed national emission estimates by summing calculated basin-level total emission estimates. The
Completions and Workovers memo presents additional information and considerations for this update.
EPA calculated basin-specific activity factors and CH4 and CO2 emission factors for all basins that reported Subpart
W data. The factors were year-specific for reporting year (RY) 2011 (first year of GHGRP data for this source)
through RY2022. For basin-level HF completion event counts, EPA used Enverus data for 1990 to 2010 and Subpart
W for 2011 forward. For the fraction of completions in each control subcategory, EPA retained the previous
Inventory's assumption that all HF gas well completions were non-REC for 1990 to 2000. The previous Inventory
also assumed that 10 percent of HF completions were non-REC with flaring from 1990 to 2010 (based on national
Subpart W data for RY2011 and RY2012); EPA updated this value using basin-specific Subpart W data for RY2011
and RY2012. For 2011 to 2022, EPA determined the percent contribution of each control category directly from
Subpart W data and used linear interpolation between 2000 and 2011 to determine the percent of gas wells with
RECs. EPA developed year- and basin-specific Subpart W EFs for 2011 forward. Year 2011 emission factors were
applied to all prior years for each basin. For basins without Subpart W data available, EPA applied national average
activity and emission factors (unweighted average of all Subpart W reported data).
3-84 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
Comparing the final completion emissions and those presented at the October 2023 webinar and in the November
2023 Completions and Workovers memo, the final estimates are higher for certain completion categories. These
emissions increases are due not to the basin-level methodology changes discussed here but rather to changes in
the Enverus dataset. EPA applied the same data processing steps to Enverus data in the fall of 2023 as it did for the
previous Enverus data analysis (conducted in 2021) and data changes led to many more completions being
classified as HF completions.
As a result of this methodological update, Cm emissions estimates are on average 10 percent lower across the
time series than in the previous Inventory. The 2021 Cm emissions estimate is 41 percent higher than in the
previous Inventory. The largest increase in the Cm emissions estimates compared to the previous Inventory is 42
percent in 2016, and the largest decrease is 35 percent in 1999. The decrease in Cm emissions is predominantly
due to HF completions that were non-REC with venting. Basins such as the Permian basin (basin 430), Williston
basin (basin 395), and Denver basin (basin 540) had a high number of HF oil well completion events over the time
series. However, they had low EFs for non-REC with venting, decreasing the overall CH4 emissions. The update
resulted in CO2 emissions estimates that are on average 33 percent higher across the time series than in the
previous Inventory. The 2021CO2 emissions estimate is 29 percent higher than in the previous Inventory. The
largest increase in CO2 emissions estimates compared to the previous Inventory is 80 percent in 2000, and the
largest decrease is 17 percent in 2018. The increase in CO2 emissions is due to HF completions that were non-REC
with flaring and REC with flaring. The Permian basin (basin 430) had the highest emissions across the time series
for completions that were non-REC with flaring and REC with flaring. The Permian basin had the highest number of
HF oil well completions and high EFs for both control categories.
Table 3-53: HF Completions National CH4 Emissions (Metric Tons CH4)
Source
1990
2005
2018
2019
2020
2021
2022
HF Completions -
Non-REC with Venting
95,717
179,290 1
380
1,126
845
244
509
HF Completions -
¦ Non-REC with Flaring
898 1
1,190
2,797
2,801
1,989
1,272
1,269
HF Completions -
REC with Venting
NO I
NO
5,478
5,466
6,212
1,114
912
HF Completions -
REC with Flaring
NO I
NO
9,637
6,188
1,945
3,620
685
Total Emissions
96,615
180,480
18,292
15,581
10,992
6,250
3,375
Previous Estimate
143,304 |
202,773
18,090
14,864
10,568
4,430
NA
NO (Not Occurring)
NA (Not Applicable)
Note: Totals may not sum due to independent rounding.
Table 3-54: HF Completions National CO2 Emissions (kt CO2)
Source
1990
2005
2018
2019
2020
2021
2022
HF Completions - Non-REC with Venting
3 1
41
+
+
+
+
+
HF Completions - Non-REC with Flaring
155 1
240 III
485
762
355
262
131
HF Completions - REC with Venting
NO 1
NO 1
+
+
+
+
+
HF Completions - REC with Flaring
NO ¦
NO ¦
2,165
1,278
441
338
141
Total Emissions
157
244
2,651
2,041
797
601
272
Previous Estimate
119
168
3,174
2,431
836
466
NA
+ Does not exceed 0.5 kt.
NO (Not Occurring)
NA (Not Applicable)
Note: Totals may not sum due to independent rounding.
Energy 3-85
-------
Production
HF Workovers (Methodological Update)
EPA updated the activity data source and calculation methodology for HF workovers to use basin-specific activity
factors and emission factors, calculated from Subpart W data for each control category (i.e., non-reduced emission
completion (REC) with venting, non-REC with flaring, REC with venting, REC with flaring). Previously, national HF
workover counts calculated using analyses for NSPS OOOO (i.e., 1 percent of HF oil wells were worked over
annually) and national annual average emission factors calculated using Subpart W data were applied to estimate
HF oil well workover emissions. In this update, EPA developed national emission estimates by summing calculated
basin-level total emission estimates. The Completions and Workovers memo presents additional information and
considerations for this update.
EPA calculated basin-specific activity factors (AFs) and CH4 and CO2 emission factors for all basins that reported
Subpart W data. For basin-level workover counts, instead of applying a 1 percent workover rate to HF oil wells, EPA
developed year- and basin-specific Subpart W AFs for 2016 (the first year of GHGRP data for this source) forward
that represent the number of HF workovers per oil well. Year 2016 Subpart W AFs were applied to all prior years
for each basin. For the fraction of workovers in each control subcategory, EPA retained the previous Inventory's
assumption that all HF oil well workovers were non-REC for 1990 to 2007 and 10 percent flaring from 1990 to
2007. For 2016 to 2022, EPA determined the percent contribution of each control category directly from Subpart
W data at the basin level and used linear interpolation between 2008 and 2015 to determine the percent of oil
wells with RECs and the percent flaring. EPA developed year- and basin-specific Subpart W EFs for 2016 forward.
Year 2016 emission factors were applied to all prior years for each basin. For basins without Subpart W data
available, EPA applied national average activity and emission factors (unweighted average of all Subpart W
reported data).
As a result of this methodological update, CH4 emissions estimates are on average 62 percent lower across the
time series than in the previous Inventory. The largest decrease in CH4 emissions estimates compared to the
previous Inventory is 97 percent in 2021 and the smallest decrease is 39 percent in 2018. The 2021CH4 emissions
estimate is 97 percent lower than in the previous Inventory. The update resulted in CO2 emissions estimates that
are on average 51 percent lower than in the previous Inventory. The 2021CO2 emissions estimate is 97 percent
lower than in the previous Inventory. The largest decrease in CO2 emissions estimates compared to the previous
Inventory is 97 percent in 2021 and the smallest decrease is 26 percent in 1990. The decrease in emissions for both
CH4 and CO2 was primarily due to the change in calculation method for workover counts. HF oil well workover
counts decreased by an average of 52 percent across the 1990 to 2021 time series compared to the previous
Inventory.
Table 3-55: HF Workovers National CH4 Emissions (Metric Tons CH4)
Source
1990
2005
2018
2019
2020
2021
2022
HF Workovers - Non-REC with Venting
17,639 I
16,244 1
87
1,339
5
16
35
HF Workovers-Non-REC with Flaring
104
96
11
7
11
20
NO
HF Workovers - REC with Venting
NO 1
NO 1
1,304
331
130
5
63
HF Workovers - REC with Flaring
NOi
NO ¦
222
75
14
19
33
Total Emissions
17,744
16,340
1,623
1,753
160
60
130
Previous Estimate
37,696 1
41,993
2,670
3,679
3,873
2,151
NA
NO (Not Occurring)
NA (Not Applicable)
Note: Totals may not sum due to independent rounding.
Table 3-56: HF Workovers National CO2 Emissions (Metric Tons CO2)
Source
1990
2005
2018
2019
2020
2021
2022
HF Workovers - Non-REC with Venting
4311
408
5
63
+
1
6
HF Workovers-Non-REC with Flaring
22,654
21,076
2,544
2,093
1,947
2,532
NO
3-86 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
HF Workovers - REC with Venting
NO
NO
17
14
4
+
3
HF Workovers - REC with Flaring
NO
NO
51,135
18,285
5,987
2,768
6,095
Total Emissions
23,085
21,484
53,701
20,456
7,939
5,301
6,105
Previous Estimate
31,219
34,778
89,049
97,515
97,766
205,160
NA
+ Does not exceed 0.5 MT.
NO (Not Occurring)
NA (Not Applicable)
Note: Totals may not sum due to independent rounding.
Pneumatic Controllers (Methodological Update)
In the previous Inventory, EPA updated the Cm emissions calculation methodology for pneumatic controllers to
use basin-specific activity factors and emission factors by bleed type (i.e., low, high, intermittent bleed) from
GHGRP data. However, the CO2 emissions calculation methodology was not updated and instead the previous
Inventory still relied on a national-level methodology to estimate CO2 emissions. For this year's Inventory, EPA
calculated pneumatic controller CO2 emissions using basin-specific emissions data such that the CO2 emissions
reflect the unique CO2 composition of the gas in a basin.
The update for pneumatic controller CO2 emission estimates resulted in an average increase of 61 percent across
the time series and an increase of 57 percent in 2021, compared to the previous Inventory.
In addition, methane emissions for pneumatic controllers were impacted due to recalculations with updated data.
Methane emissions from onshore production pneumatic controllers are an average of 2 percent lower across the
time series and 1 percent lower in 2021, compared to the previous Inventory. The emission changes were due to
GHGRP data submission revisions.
Table 3-57: Pneumatic Controllers National CO2 Emissions (Metric Tons CO2)
Source
1990
2005
2018
2019
2020
2021
2022
High Bleed Controllers
94,495
39,873
5,056
5,369
5,778
4,260
2,403
Low Bleed Controllers
7,109
4,132
3,740
4,819
3,704
4,561
3,992
Intermittent Bleed Controllers
NO
26,370
75,579
80,855
82,493
80,063
70,328
Total Emissions
101,604
70,374
84,374
91,044
91,975
88,884
76,723
Previous Estimate
42,406
46,477
70,322
49,460
63,104
56,641
NA
NO (Not Occurring)
NA (Not Applicable)
Note: Totals may not sum due to independent rounding.
Table 3-58: Pneumatic Controllers National CH4 Emissions (Metric Tons CH4)
Source
1990
2005 2018
2019
2020
2021
2022
High Bleed Controllers
Low Bleed Controllers
Intermittent Bleed Controllers
709,796
51.129
NO
481,760 71,824
62.162 32,106
267,220 1,135,995
72,432
50,770
758,001
86,363
36,740
996,250
44,611
46,060
912,391
23,712
35,439
634,400
Total Emissions
760,925
811,142 1,239,924
881,203
1,119,352
1,003,063
693,551
Previous Estimate
759,970
832,929 1,260,259
886,382
1,130,899
1,015,080
NA
NO (Not Occurring)
NA (Not Applicable)
Note: Totals may not sum due to independent rounding.
Equipment Leaks (Methodological Update)
In the previous Inventory, EPA updated the CH4 emissions calculation methodology for onshore production
equipment leaks to use basin-specific equipment-level activity factors (e.g., separators per well) from GHGRP data.
However, the CO2 emissions calculation methodology was not updated and instead the previous Inventory still
relied on a national-level methodology to estimate CO2 emissions. For this year's Inventory, EPA calculated
Energy 3-87
-------
equipment leak CO2 emissions in the same manner as CH4 emissions. EPA calculated CO2 estimates using the basin-
specific equipment-level activity factors for RY2015 through RY2022 from GHGRP, consistent with the
methodology used to calculate the Cm activity factors, and the CO2 emissions factors for onshore production
segment equipment leaks. Note, this methodological update applies only for activity factors. The previous
Inventory's CO2 emission factors for onshore production segment equipment leaks (by equipment type) were
retained and used to develop CO2 estimates.
The update for CO2 emission estimates resulted in an average increase of 7 percent across the time series and an
increase of 18 percent in 2021, compared to the previous Inventory. Years 2015 to 2021 were impacted more by
the update, with an average increase of 28 percent compared to the previous Inventory.
Table 3-59: Equipment Leaks National CO2 Emissions (Metric Tons CO2)
Source
1990
2005
2018
2019
2020
2021
2022
Oil Wellheads (heavy crude)
2
2
2
1
1
1
1
Oil Wellheads (light crude)
3,159
2,857 •
3,330
3,416
3,280
3,272
3,195
Separators (heavy crude)
1
1
1
1
+
+
+
Separators (light crude)
613
868 a
2,280
2,114
1,601
1,443
1,411
Heater/Treaters (light crude)
508
474
786
1,028
846
951
905
Headers (heavy crude)
+ £
+
+
+
+
+
+
Headers (light crude)
185
458
712
843
432
430
485
Total Emissions
4,468
4,660
7,111
7,403
6,161
6,098
5,997
Previous Estimate
4,453
4,681 I
5,396
5,351
5,159
5,159
NA
+ Does not exceed 0.05 MT C02.
NA (Not Applicable)
Chemical Injection Pumps (Methodological Update)
In the previous Inventory, EPA updated the Cm emissions calculation methodology for chemical injection pumps to
use basin-specific equipment-level activity factors (e.g., pumps per well) from GHGRP data. However, the CO2
emissions calculation methodology was not updated and instead the previous Inventory still relied on a national-
level methodology to estimate CO2 emissions. For this year's Inventory, EPA calculated chemical injection pump
CO2 emissions in the same manner as CH4 emissions. EPA calculated CO2 estimates using the basin-specific
equipment-level activity factors for RY2015 through RY2022 from GHGRP, consistent with the methodology used
to calculate the CH4 activity factors, and the CO2 emission factor. Note, this methodological update applies only for
activity factors. The previous Inventory's chemical injection pumps CO2 emission factor was retained and used to
develop CO2 estimates. The update for CO2 emission estimates resulted in an average decrease of 29 percent
across the time series and a decrease of 35 percent in 2021, compared to the previous Inventory.
Table 3-60: Chemical Injection Pump National CO2 Emissions (Metric Tons CO2)
Source
1990
2005
2018
2019
2020
2021
2022
Chemical Injection Pumps
2,646 |
4,464 I
5,955
6,784
5,338
4,749
4,449
Previous Estimate
4,506 1
6,522
7,689
7,625
7,351
7,351
NA
NA (Not Applicable)
Storage Tanks (Recalculation with Updated Data)
Carbon dioxide emissions from production storage tanks are on average 0.9 percent higher across the time series
compared to the previous Inventory. Emissions estimates for 2021 are 4 percent higher than in the previous
Inventory, which is primarily due to large tanks with flares. The emission changes were due to GHGRP data
submission revisions.
3-88 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
Table 3-61: Storage Tanks National CO2 Emissions (kt CO2)
Source
1990
2005
2018
2019
2020
2021
2022
Large Tanks w/Flares
NO j
716 §
5,336
6,251
5,829
5,594
4,513
Large Tanks w/VRU
NO 1
3
3
9
2
1
1
Large Tanks w/o Control
24
8
4
9
5
4
2
Small Tanks w/Flares
NO I
3
7
9
11
10
11
Small Tanks w/o Flares
12
5
5
4
4
5
5
Malfunctioning Separator Dump
1
iii
Valves
121
13 I
30
26
21
34
8
Total Emissions
48
748
5,386
6,309
5,871
5,649
4,539
Previous Estimate
47
74S |
5,398
6,024
5,255
5439
NA
NO (Not Occurring)
NA (Not Applicable)
Chemical injection Pumps (Recalculation with Updated Data)
Chemical injection pump Cm estimates decreased by an average of 19 percent across the time series and
decreased by 26 percent in 2021, compared to the previous Inventory. The emission changes were due to GHGRP
data submission revisions.
Table 3-62: Chemical Injection Pumps National CH4 Emissions (Metric Tons CH4)
Source
1990
2005
2018
2019
2020
2021
2022
Chemical Injection Pumps
47,425
79,968 1
108,147
122,967
96,186
85,494
79,712
Previous Estimate
47,401
105,458 j
138,866
387,416
116,080
115,678
NA
NA (Not Applicable)
Produced Water (Recalculation with Updated Data)
Methane estimates from produced water increased by an average of 2 percent across the time series and
increased by 4 percent in 2021, compared to the previous Inventory. The emission changes were due to Enverus
data updates.
Table 3-63: Produced Water National CH4 Emissions (Metric Tons CH4)
Source
1990
2005
2018
2019
2020
2021
2022
Produced Water - Regular Pressure
Wells
71,854
49,840
73,727
77,370
70,374
71,749
73,665
Produced Water - Low Pressure Wells
20,482
14,207
1 21,016
22,055
20,061
20,452
20,998
Total Emissions
92,336
64,047
94,743
99,425
90,435
92,201
94,663
Previous Estimate
91,391
62,458
92,863
97,735
88,622
88,622
NA
NA (Not Applicable)
Associated Gas Flaring (Recalculation with Updated Data)
Associated gas flaring Cm emission estimates increased by an average of 2 percent across the time series and
increased by 23 percent in 2021, compared to the previous Inventory. The emission changes were due to GHGRP
data submission revisions.
Table 3-64: Associated Gas Flaring National CH4 Emissions (Metric Tons CH4)
Source
1990
2005
2018
2019
2020
2021
2022
220 - Gulf Coast Basin (LA, TX)
901
480 |
2,379
2,907
3,643
2,136
1,900
360 - Anadarko Basin
452
274 1
350
90
21
35
30
395 - Williston Basin
2,666
3,405 1
36,108
58,138
28,176
21,676
18,341
Energy 3-89
-------
430 - Permian Basin
11,662 5
7,992 11
25,286
26,063
12,695
8,178
7,546
"Other" Basins
4,314
2,335
2,089
3,760
2,353
2,790
2,072
Total Emissions
19,995
14,486
66,211
90,958
46,888
34,814
29,889
220 - Gulf Coast Basin (LA, TX)
886
490
2,440
2,991
3,692
1,864
NA
360 - Anadarko Basin
4471=
274
348
88
21
41
NA
395 - Williston Basin
2,665
3,419
36,120
48,019
23,556
18,734
NA
430 - Permian Basin
11,263
7,805 1
25,198
27,484
13,086
5,852
NA
"Other" Basins
4,369
2,347
1,992
3,563
2,295
1,802
NA
Previous Estimate
19,630 5
14,335 mill;
66,096
82,146
42,649
28,293
NA
NA (Not Applicable)
Gas Engines (Recalculation with Updated Data)
Methane estimates from gas engines decreased by an average of 2 percent across the time series and decreased
by 2 percent in 2021, compared to the previous Inventory. The emission changes were due to Enverus data
updates.
Table 3-65: Gas Engines National CH4 Emissions (Metric Tons CH4)
Source
1990
2005
2018
2019
2020
2021
2022
Gas Engines
81,271
69,973
90,773
92,909
88,6I9
87,040
87,546
Previous Estimate
81,916
71,348
91,719
93,608
89,497
89,233
NA
NA (Not Applicable)
Miscellaneous Production Flaring (Recalculation with Updated Data)
Miscellaneous production flaring CO2 emission estimates are on average 1 percent higher across the time series
compared to the previous Inventory. Carbon dioxide emissions estimates for 2021 increased by 12 percent
compared to the previous Inventory. The emission changes were due to GHGRP data submission revisions.
Table 3-66: Miscellaneous Production Flaring National CO2 Emissions (kt CH4)
Source
1990
2005
2018
2019
2020
2021
2022
220 - Gulf Coast Basin (LA, TX)
395 - Williston Basin
430 - Permian Basin
"Other" Basins
OOOO
103
71 =
214
398 :
567
1,701
1,463
639
609
3,049
4,312
707
654
1,307
2,723
427
802
1,312
2,156
368
649
1,241
2,709
429
Total Emissions
0
786
4,370
8,678
5,110
4,638
5,028
Previous Estimate
0
3,008
4,307
8,225
4,679
4,154
NA
NA (Not Applicable)
Offshore Production - GOM Federal Waters (Recalculation with Updated Data)
Vented and leak Cm emission estimates from offshore production in GOM federal waters decreased by an average
of 0.4 percent across the time series and decreased by 8 percent in 2021, compared to the previous Inventory. The
emission changes were due to updated offshore complex counts from BOEM.
Table 3-67: Offshore Production National CH4 Emissions (Metric Tons CH4)
Source
1990
2005
2018
2019
2020
2021
2022
GOM Federal Waters-
-Vented
196,769 I
101,585
112,786
110,263
103,116
102,365
101,546
GOM Federal Waters -
¦ Leaks
96,575 1
103,712
1
58,938
57,577
53,860
53 395
52,961
Total Emissions
293,344
205,298
171,724
167,840
156,976
155,760
154,507
Previous Estimate
293,204 I
205,207
171,910
170,190
162,543
168,798
NA
NA (Not Applicable)
3-90 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
Transportation
Recalculations for the transportation segment have resulted in calculated Cm and CO2 emissions over the time
series from this segment that are lower (by less than 0.05 percent) than in the previous Inventory.
Refining
Recalculations due to resubmitted GHGRP data in the refining segment have resulted in a decrease in calculated
Cm emissions by an average of 4.5 percent across the time series and a decrease of 14 percent in 2021, compared
to the previous Inventory.
Refining CO2 emission estimates decreased by an average of 8 percent across the time series and decreased by 28
percent in 2021, compared to the previous Inventory. This change in emissions is due to GHGRP resubmissions and
was largely due to a change in reported flaring CO2 emissions.
Table 3-68: Refining National CH4 Emissions (Metric Tons CH4)
Source
1990
2005
2018
2019
2020
2021
2022
Refining
25,742
29,218
27,804
30,814
25,861
25,366
24,685
Previous Estimate
26,398
29,963
30,313
35,516
31,023
29,551
NA
NA (Not Applicable)
Table 3-69: Refining National CO2 Emissions (kt CO2)
Source
1990
2005
2018
2019
2020
2021
2022
Flares
3,023
3,431
2,814
3,523
2,859
2,989
2,836
Total Refining
3,174
3,602
2,877
3,571
2,893
3,021
2,872
Previous Estimate
3,284
3,728 \
3,706
5,009
4,242
4,216
NA
NA (Not Applicable)
Planned Improvements
Planned Improvements for 2025 Inventory
EPA updated the Enverus data and there were notable increases in the number of wells and completions identified
as being hydraulically fractured compared with previous versions of the database. EPA will assess the underlying
Enverus data to determine the cause of these changes.
Upcoming Data, and Additional Data that Could Inform the Inventory
EPA will assess new data received by the Greenhouse Gas Reporting Program, the Methane Challenge Program,
and other relevant programs on an ongoing basis, which may be used to confirm or improve existing estimates and
assumptions.
EPA continues to track studies that contain data that may be used to update the Inventory. EPA will also continue
to assess studies that include and compare both top-down and bottom-up estimates, and which could lead to
improved understanding of unassigned high emitters (e.g., identification of emission sources and information on
frequency of high emitters) as recommended in previous stakeholder comments.
Box 3-6: Carbon Dioxide Transport, Injection, and Geological Storage
Carbon dioxide is produced, captured, transported, and used for Enhanced Oil Recovery (EOR) as well as
commercial and non-EOR industrial applications, or is stored geologically. This CO2 is produced from both
naturally-occurring CO2 reservoirs and from industrial sources such as natural gas processing plants and
ammonia plants. In the Inventory, emissions of CO2 from naturally-occurring CO2 reservoirs are estimated based
Energy 3-91
-------
on the specific application.
In the Inventory, CO2 that is used in non-EOR industrial and commercial applications (e.g., food processing,
chemical production) is assumed to be emitted to the atmosphere during its industrial use. These emissions are
discussed in the Carbon Dioxide Consumption section, 4.15.
For EOR CO2, as noted in the 2006IPCC Guidelines, "At the Tier 1 or 2 methodology levels [EOR CO2 is]
indistinguishable from fugitive greenhouse gas emissions by the associated oil and gas activities." In the U.S.
estimates for oil and gas fugitive emissions, the Tier 2 emission factors for CO2 include CO2 that was originally
injected and is emitted along with other gas from leak, venting, and flaring pathways, as measurement data
used to develop those factors would not be able to distinguish between CO2 from EOR and CO2 occurring in the
produced natural gas. Therefore, EOR CO2 emitted through those pathways is included in CO2 estimates in 1B2.
IPCC includes methodological guidance to estimate emissions from the capture, transport, injection, and
geological storage of CO2. The methodology is based on the principle that the carbon capture and storage
system should be handled in a complete and consistent manner across the entire Energy sector. The approach
accounts for CO2 captured at natural and industrial sites as well as emissions from capture, transport, and use.
For storage specifically, a Tier 3 methodology is outlined for estimating and reporting emissions based on site-
specific evaluations. However, IPCC (IPCC 2006) notes that if a national regulatory process exists, emissions
information available through that process may support development of CO2 emission estimates for geologic
storage.
In the United States, facilities that produce CO2 for various end-use applications (including capture facilities such
as acid gas removal plants and ammonia plants), importers of CO2, exporters of CO2, facilities that conduct
geologic sequestration of CO2, and facilities that inject CO2 underground, are required to report greenhouse gas
data annually to EPA through its GHGRP. Facilities reporting geologic sequestration of CO2 to the GHGRP
develop and implement an EPA-approved site-specific monitoring, reporting and verification plan, and report
the amount of CO2 sequestered using a mass balance approach.
GHGRP data relevant for this Inventory estimate consists of national-level annual quantities of CO2 captured and
extracted for EOR applications for 2010 to 2022 and data reported for geologic sequestration from 2016 to
2022.
The amount of CO2 captured and extracted from natural and industrial sites for EOR applications in 2022 is
36,680 kt (36.7 MMT CO2 Eq.) 6. The quantity of CO2 captured and extracted is noted here for information
purposes only; CO2 captured and extracted from industrial and commercial processes is generally assumed to be
emitted and included in emissions totals from those processes.
Table 3-70: Quantity of CO2 Captured and Extracted for EOR Operations (kt CO2)
Stage
2018
2019
2020
2021
2022
Quantity of C02 Captured and
Extracted for EOR Operations
48,400
52,100
35,210
35,090
36,680
Several facilities are reporting under GHGRP Subpart RR (Geologic Sequestration of Carbon Dioxide). See Table
3-71 for the number of facilities reporting under Subpart RR, the reported CO2 sequestered in subsurface
geologic formations in each year, and of the quantity of CO2 emitted from equipment leaks in each year. The
quantity of CO2 sequestered and emitted is noted here for information purposes only; EPA is considering
updates to its approach for this source and is seeking feedback as part of this public review draft on potential
updates that could be incorporated in future Inventories.
Table 3-71: Geologic Sequestration Information Reported Under GHGRP Subpart RR
Stage
2018
2019
2020
2021
2022
Number of Reporting Facilities
5
5
6
9
13
Reported Annual C02Sequestered (kt)
7,662
8,332
6,802
6,947
7,953
3-92 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
Reported Annual C02 Emissions from
Equipment Leaks (kt) 11 16 13 37 27
Energy 3-93
-------
3.7 Natural Gas Systems (CRT Source
Category lB2b)
The U.S. natural gas system encompasses hundreds of thousands of wells, hundreds of processing facilities, and
over a million miles of transmission and distribution pipelines. This category (lB2b) as defined in the IPCC
methodological guidance is for fugitive emissions from natural gas systems, which per IPCC guidelines include
emissions from leaks, venting, and flaring. Total greenhouse gas emissions (Cm, CO2, and N2O) from natural gas
systems in 2022 were 209.7 MMT CO2 Eq., a decrease of 17 percent from 1990 and a decrease of 0.3 percent from
2021, both primarily due to decreases in CH4 emissions. From 2011, emissions decreased by 5 percent, primarily
due to decreases in CH4 emissions. National total dry gas production in the United States increased by 104 percent
from 1990 to 2022, increased by 5 percent from 2021 to 2022, and increased by 59 percent from 2011 to 2022. Of
the overall greenhouse gas emissions (209.7 MMT CO2 Eq.), 83 percent are CH4 emissions (173.1 MMT CO2 Eq.), 17
percent are CO2 emissions (36.5 MMT), and less than 0.1 percent are N2O emissions (0.15 MMT CO2 Eq.).
Overall, natural gas systems emitted 173.1 MMT CO2 Eq. (6,183 kt CH4) of Cm in 2022, a 21 percent decrease
compared to 1990 emissions, and 1 percent decrease compared to 2021 emissions (see Table 3-72 and Table 3-73).
For non-combustion CO2, a total of 36.5 MMT CO2 Eq. (36,470 kt) was emitted in 2022, a 12 percent increase
compared to 1990 emissions, and a 2 percent increase compared to 2021 levels. The 2022 N2O emissions were
estimated to be 0.15 MMT CO2 Eq. (0.57 kt N2O), a 3205 percent increase compared to 1990 emissions, and a 1104
percent increase compared to 2021 levels.
The 1990 to 2022 emissions trend is not consistent across segments or gases. Overall, the 1990 to 2022 decrease in
Cm emissions is due primarily to the decrease in emissions from the following segments: distribution (70 percent
decrease), transmission and storage (38 percent decrease), processing (37 percent decrease), and exploration (97
percent decrease). Over the same time period, the production segment saw increased CH4 emissions of 38 percent
(with onshore production emissions increasing 16 percent, offshore production emissions decreasing 86 percent,
and gathering and boosting [G&B] emissions increasing 108 percent), and post-meter emissions increasing by 65
percent. The 1990 to 2022 increase in CO2 emissions is primarily due to an increase in CO2 emissions in the
production segment, where emissions from flaring have increased over time.
Methane and CO2 emissions from natural gas systems include those resulting from normal operations, routine
maintenance, and system upsets. Emissions from normal operations include natural gas engine and turbine
uncombusted exhaust, flaring, and leak emissions from system components. Routine maintenance emissions
originate from pipelines, equipment, and wells during repair and maintenance activities. Pressure surge relief
systems and accidents can lead to system upset emissions. Emissions of N2O from flaring activities are included in
the Inventory, with most of the emissions occurring in the processing and production segments. Note, CO2
emissions exclude all combustion emissions (e.g., engine combustion) except for flaring CO2 emissions. All
combustion CO2 emissions (except for flaring) are accounted for in Section 3.1.
Each year, some estimates in the Inventory are recalculated with improved methods and/or data. These
improvements are implemented consistently across the previous Inventory's time series (i.e., 1990 to 2022) to
ensure that the trend is representative of changes in emissions. Recalculations in natural gas systems in this year's
Inventory include:
• Methodological updates to transmission compressor station activity data, completions and workovers,
and underground natural gas storage well events.
• Recalculations due to Greenhouse Gas Reporting Program (GHGRP) submission revisions.
• Recalculations due to updated well counts and production data from Enverus.
The Recalculations Discussion section below provides more details on the updated methods.
3-94 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
Below is a characterization of the six emission subcategories of natural gas systems: exploration, production
(including gathering and boosting), processing, transmission and storage, distribution, and post-meter. Each of the
segments is described and the different factors affecting Cm, CO2, and N2O emissions are discussed.
Exploration. Exploration includes well drilling, testing, and completion. Emissions from exploration accounted for
0.1 percent of Cm emissions and 0.1% of CO2 emissions from natural gas systems in 2022. Well completions
accounted for approximately 90 percent of Cm emissions from the exploration segment in 2022, with the rest
resulting from well testing and drilling. Well completion flaring emissions account for most of the CO2 emissions.
Methane emissions from exploration decreased by 97 percent from 1990 to 2022, with the largest decreases
coming from hydraulically fractured gas well completions without reduced emissions completions (RECs). Methane
emissions from exploration increased 58 percent from 2021 to 2022 due to increases in emissions from
hydraulically fractured well completions (both non-REC with flaring and REC with venting). Methane emissions
from exploration were highest from 2006 to 2008. Carbon dioxide emissions from exploration decreased by 94
percent from 1990 to 2022 primarily due to decreases in hydraulically fractured gas well completions. Carbon
dioxide emissions from exploration increased by 14 percent from 2020 to 2021 due to increases in emissions from
hydraulically fractured gas well completions (REC with flaring). Carbon dioxide emissions from exploration were
highest from 2006 to 2008. Nitrous oxide emissions from exploration decreased 95 percent from 1990 to 2022 and
increased 74 percent from 2021 to 2022.
Production (including gathering and boosting). In the production segment, wells are used to withdraw raw gas
from underground formations. Emissions arise from the wells themselves, and from well-site equipment and
activities such as pneumatic controllers, tanks and separators, and liquids unloading. Gathering and boosting
emission sources are included within the production sector. The gathering and boosting sources include gathering
and boosting stations (with multiple emission sources on site) and gathering pipelines. The gathering and boosting
stations receive natural gas from production sites and transfer it, via gathering pipelines, to transmission pipelines
or processing facilities (custody transfer points are typically used to segregate sources between each segment).
Boosting processes include compression, dehydration, and transport of gas to a processing facility or pipeline.
Emissions from production (including gathering and boosting) accounted for 52 percent of CH4 emissions and 23
percent of CO2 emissions from natural gas systems in 2022. Emissions from gathering and boosting and pneumatic
controllers in onshore production accounted for most of the production segment CH4 emissions in 2022. Within
gathering and boosting, the largest sources of CH4 are compressor exhaust slip, compressor venting and leaks, and
tanks. Flaring emissions account for most of the CO2 emissions from production, with the highest emissions coming
from flare stacks at gathering stations, miscellaneous onshore production flaring, and tank flaring. Methane
emissions from production increased by 38 percent from 1990 to 2022, due primarily to increases in emissions
from pneumatic controllers (due to an increase in the number of controllers, particularly in the number of
intermittent bleed controllers) and increases in emissions from compressor exhaust slip in gathering and boosting.
Methane emissions from production decreased 3 percent from 2021 to 2022 due to decreases in emissions from
pneumatic controllers and liquids unloading. Carbon dioxide emissions from production increased by
approximately a factor of 2.6 from 1990 to 2022 due to increases in emissions at flare stacks in gathering and
boosting and miscellaneous onshore production flaring and decreased 8 percent from 2021 to 2022 due primarily
to decreases in emissions at flare stacks in miscellaneous onshore production flaring and tank venting. Nitrous
oxide emissions from production were 36.9 times higher in 2022 than in 1990 and 17.5 times higher in 2022 than
in 2021. The increase in N2O emissions from 1990 to 2022 and from 2021 to 2022 is primarily due to increases in
emissions from condensate tank flaring.
Processing. In the processing segment, natural gas liquids and various other constituents from the raw gas are
removed, resulting in "pipeline quality" gas, which is injected into the transmission system. Methane emissions
from compressors, including compressor seals, are the primary emission source from this stage. Most of the CO2
emissions come from acid gas removal (AGR) units, which are designed to remove CO2 from natural gas. Processing
plants accounted for 9 percent of Cm emissions and 73 percent of CO2 emissions from natural gas systems.
Methane emissions from processing decreased by 37 percent from 1990 to 2022 as emissions from compressors
(leaks and venting) and equipment leaks decreased; and increased 7 percent from 2021 to 2022 due to increased
emissions from gas engines. Carbon dioxide emissions from processing decreased by 6 percent from 1990 to 2022,
Energy 3-95
-------
due to a decrease in AGR, and increased 4 percent from 2021 to 2022 due to increased emissions from flaring
emissions at processing plants. Nitrous oxide emissions increased 116 percent from 2021 to 2022.
Transmission and Storage. Natural gas transmission involves high pressure, large diameter pipelines that transport
gas long distances from field production and processing areas to distribution systems or large volume customers
such as power plants or chemical plants. Compressor station facilities are used to move the gas throughout the
U.S. transmission system. Leak Cm emissions from these compressor stations and venting from pneumatic
controllers account for most of the emissions from this stage. Uncombusted compressor engine exhaust and
pipeline venting are also sources of Cm emissions from transmission. Natural gas is also injected and stored in
underground formations, or liquefied and stored in above ground tanks, during periods of low demand (e.g.,
summer), and withdrawn, processed, and distributed during periods of high demand (e.g., winter). Leak and
venting emissions from compressors are the primary contributors to Cm emissions from storage. Emissions from
liquefied natural gas (LNG) stations and terminals are also calculated under the transmission and storage segment.
Methane emissions from the transmission and storage segment accounted for approximately 23 percent of
methane emissions from natural gas systems, while CO2 emissions from transmission and storage accounted for 3
percent of the CO2 emissions from natural gas systems. CH4emissions from this source decreased by 38 percent
from 1990 to 2022 due to reduced pneumatic device and compressor station emissions (including emissions from
compressors and leaks) and decreased 1 percent from 2021 to 2022 due to decreased emissions from pipeline
venting transmission compressors. CO2 emissions from transmission and storage were 6.6 times higher in 2022
than in 1990, due to increased emissions from LNG export terminals, and increased by 36 percent from 2021 to
2022, also due to LNG export terminals and flaring (both transmission and storage). The quantity of LNG exported
from the United States increased by a factor of 74 from 1990 to 2022, and by 9 percent from 2021 to 2022. LNG
emissions are about 1 percent of CH4 and 86 percent of CO2 emissions from transmission and storage in year 2022.
Nitrous oxide emissions from transmission and storage increased by 405 percent from 1990 to 2022 and increased
by 177 percent from 2021 to 2022.
Distribution. Distribution pipelines take the high-pressure gas from the transmission system at "city gate" stations,
reduce the pressure and distribute the gas through primarily underground mains and service lines to individual end
users. There were 1,352,384 miles of distribution mains in 2022, an increase of 408,227 miles since 1990 (PHMSA
2022). Distribution system emissions, which accounted for 9 percent of Cm emissions from natural gas systems
and less than 0.1 percent of CO2 emissions from natural gas systems, result mainly from leak emissions from
pipelines and stations. An increased use of plastic piping, which has lower emissions than other pipe materials, has
reduced both Cm and CO2 emissions from this stage, as have station upgrades at metering and regulating (M&R)
stations. Distribution system Cm emissions in 2022 were 70 percent lower than 1990 levels and 1 percent lower
than 2021 emissions. Distribution system CO2 emissions in 2022 were 70 percent lower than 1990 levels and 1
percent lower than 2021 emissions. Annual CO2 emissions from this segment are less than 0.1 MMT CO2 Eq. across
the time series.
Post-Meter. Post-meter includes leak emissions from residential and commercial appliances, industrial facilities
and power plants, and natural gas fueled vehicles. Leak emissions from residential appliances and industrial
facilities and power plants account for the majority of post-meter Cm emissions. Methane emissions from the
post-meter segment accounted for approximately 8 percent of emissions from natural gas systems in 2022. Post-
meter Cm emissions increased by 65 percent from 1990 to 2022 and increased by 3 percent from 2021 to 2022,
due to increases in the number of residential houses using natural gas and increased natural gas consumption at
industrial facilities and power plants. CO2 emissions from post-meter account for less than 0.01 percent of total
CO2 emissions from natural gas systems.
Total greenhouse gas emissions from the six subcategories within natural gas systems are shown in MMT CO2 Eq.
in Table 3-72. Total CH4 emissions for these same segments of natural gas systems are shown in MMT CO2 Eq.
(Table 3-73) and kt (Table 3-74). Most emission estimates are calculated using a net emission approach. However,
a few sources are still calculated with a potential emission approach. Reductions data are applied to those sources.
In 2022, 2.6 MMT CO2 Eq. CH4 is subtracted from production segment emissions, 4.3 MMT CO2 Eq. CH4 is
subtracted from the transmission and storage segment, and 0.1 MMT CO2 Eq. CH4 is subtracted from the
distribution segment to calculate net emissions. More disaggregated information on potential emissions, net
3-96 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
emissions, and reductions data is available in Annex 3.6, Methodology for Estimating Cm and CO2 Emissions from
Natural Gas Systems.
Table 3-72: Total Greenhouse Gas Emissions (CH4, CO2, and N2O) from Natural Gas Systems
(MMT C02 Eq.)
Segment
1990
2005
2018
2019
2020
2021
2022
Exploration
7.3
22.3 1
2.9
2.3
0.3
0.1
0.2
Production
68.4
97.9 1
114.1
114.5
105.8
101.5
98.3
Processing
52.2
31.8
36.3
40.4
39.3
39.7
41.8
Transmission and Storage
64.2
46.2
41.7
41.8
43.1
40.7
40.7
Distribution
51.0
28.5
15.6
15.5
15.5
15.3
15.3
Post-Meter
8.11
12.5
12.8
13.0
13.0
13.4
Total
IB
251.2
236.5
223.0
227.3
217.0
210.4
209.7
Note: Totals may not sum due to independent rounding.
Table 3-73: CH4 Emissions from Natural Gas Systems (MMT CO2 Eq.)
Segment
1990
2005
2018
2019
2020
2021
2022
Exploration
6.7
19.6
2.6
2.1
0.2
0.1
0.2
Production
65.2
93.4
104.9
103.6
96.7
92.2
89.7
Onshore Production
39.9
64.4
60.5
58.0
53.1
48.3
46.2
Gathering and Boosting
20.5
27.0
43.6
44.8
42.7
43.3
42.8
Offshore Production
4.8
2.0
0.9
0.8
0.9
0.6
0.6
Processing
23 9 ¦
13.0
13.5
14.2
13.8
14.2
15.1
Transmission and Storage
64.0
46.0
41.2
40.5
41.1
39.8
39.6
Distribution
50.9
28.5 1
15.6
15.5
15.5
15.3
15.2
Post-Meter
8.1
9.6
12.5
12.8
13.0
13.0
13.4
Total
218.8
210.1
190.3
188.7
180.3
174.6
173.1
Note: Totals may not sum due to independent rounding.
Table 3-74: CH4 Emissions from Natural Gas Systems (kt)
Segment
1990 ¦
2005
2018
2019
2020
2021
2022
Exploration
238
700
93
75
7
4
6
Production
2,328
3,335 •
3,748
3,701
3,453
3,293
3,202
Onshore Production
1,424
2,299
2,162
2,073
1,895
1,726
1,650
Gathering and Boosting
733
963 II
1,556
1,601
1,527
1,545
1,528
Offshore Production
170
73
31
28
32
22
23
Processing
853 =
463 i
483
507
495
507
541
Transmission and Storage
2,285
1,645
1,470
1,448
1,468
1,421
1,413
Distribution
1,819:
1,018 1
556
554
553
547
544
Post-Meter
290
344
445
457
463
464
477
Total
7.813
7.505
6.795
6.741
6.439
6.235
6.183
Note: Totals may not sum due to independent rounding.
Energy 3-97
-------
Table 3-75: CO2 Emissions from Natural Gas Systems (MMT)
Segment
1990
2005
2018
2019
2020
2021
2022
Exploration
0.6
2.7
0.3
0.2
0.1
+
+
Production
-------
Methodology and Time-Series Consistency
See Annex 3.6 for the full time series of emissions data, activity data, and emission factors, and additional
information on methods and data sources—for example, the specific years of reporting data from EPA's GHGRP
that are used to develop certain factors.
This section provides a general overview of the methodology for natural gas system emission estimates in the
Inventory, which involves the calculation of Cm, CO2, and N2O emissions for over 100 emissions sources (i.e.,
equipment types or processes), and then the summation of emissions for each natural gas segment.
The approach for calculating emissions for natural gas systems generally involves the application of emission
factors to activity data. For most sources, the approach uses technology-specific emission factors or emission
factors that vary over time and take into account changes to technologies and practices, which are used to
calculate net emissions directly. For others, the approach uses what are considered "potential methane factors"
and emission reduction data to calculate net emissions. The estimates are developed with an IPCC Tier 2 approach.
Tier 1 approaches are not used.
Emission Factors. Key references for emission factors for CH4 and CO2 emissions from the U.S. natural gas industry
include a 1996 study published by the Gas Research Institute (GRI) and EPA (GRI/EPA 1996), EPA's GHGRP (EPA
2023b), and others.
The 1996 GRI/EPA study developed over 80 CH4 emission factors to characterize emissions from the various
components within the operating segments of the U.S. natural gas system. The GRI/EPA study was based on a
combination of process engineering studies, collection of activity data, and measurements at representative
natural gas facilities conducted in the early 1990s. Year-specific natural gas CFU compositions are calculated using
U.S. Department of Energy's Energy Information Administration (EIA) annual gross production data for National
Energy Modeling System (NEMS) oil and gas supply module regions in conjunction with data from the Gas
Technology Institute (GTI, formerly GRI) Unconventional Natural Gas and Gas Composition Databases (GTI 2001).
These year-specific CFU compositions are applied to emission factors, which therefore may vary from year to year
due to slight changes in the CFU composition of natural gas for each NEMS region.
GHGRP Subpart W data were used to develop CH4, CO2, and N2O emission factors for many sources in the
Inventory. In the exploration and production segments, GHGRP data were used to develop emission factors used
for all years of the time series for well testing, gas well completions and workovers with and without hydraulic
fracturing, pneumatic controllers and chemical injection pumps, condensate tanks, liquids unloading,
miscellaneous flaring, gathering and boosting pipelines, and certain sources at gathering and boosting stations. In
the processing segment, for recent years of the time series, GHGRP data were used to develop emission factors for
leaks, compressors, flares, dehydrators, and blowdowns/venting. In the transmission and storage segment, GHGRP
data were used to develop factors for all years of the time series for LNG stations and terminals and transmission
pipeline blowdowns, and for pneumatic controllers for recent years of the time series.
Other data sources used for CH4 emission factors include Zimmerle et al. (2015) for transmission and storage
station leaks and compressors, GTI (2009 and 2019) for commercial and industrial meters, Lamb et al. (2015) for
recent years for distribution pipelines and meter/regulator stations, Zimmerle et al. (2019) for gathering and
boosting stations, Bureau of Ocean Energy Management (BOEM) reports, and Fischer et al. (2019) and IPCC (2019)
for post-meter emissions.
For CO2 emissions from sources in the exploration, production, and processing segments that use emission factors
not directly calculated from GHGRP data, data from the 1996 GRI/EPA study and a 2001 GTI publication were used
to adapt the CH4 emission factors into related CO2 emission factors. For sources in the transmission and storage
segment that use emission factors not directly calculated from GHGRP data, and for sources in the distribution
segment, data from the 1996 GRI/EPA study and a 1993 GTI publication were used to adapt the CH4 emission
factors into non-combustion related CO2 emission factors. CO2 emissions from post-meter sources (commercial,
industrial and vehicles) were estimated using default emission factors from IPCC (2019). Carbon dioxide emissions
from post-meter residential sources are included in fossil fuel combustion data.
Energy 3-99
-------
Flaring N2O emissions were estimated for flaring sources using GHGRP data.
See Annex 3.6 for more detailed information on the methodology and data used to calculate CH4, CO2, and N2O
emissions from natural gas systems.
Activity Data. Activity data were taken from various published data sets, as detailed in Annex 3.6. Key activity data
sources include data sets developed and maintained by EPA's GHGRP (EPA 2023b); Enverus (Enverus 2023); BOEM;
Federal Energy Regulatory Commission (FERC); EIA; the Natural Gas STAR and Methane Challenge Programs annual
data; Oil and Gas Journal; and PHMSA.
For a few sources, recent direct activity data are not available. For these sources, either 2021 data were used as a
proxy for 2022 data, or a set of industry activity data drivers was developed and used to calculate activity data over
the time series. Drivers include statistics on gas production, number of wells, system throughput, miles of various
kinds of pipe, and other statistics that characterize the changes in the U.S. natural gas system infrastructure and
operations. More information on activity data and drivers is available in Annex 3.6.
A complete list of references for emission factors and activity data by emission source is provided in Annex 3.6.
Calculating Net Emissions. For most sources, net emissions are calculated directly by applying emission factors to
activity data. Emission factors used in net emission approaches reflect technology-specific information and take
into account regulatory and voluntary reductions. However, for production, transmission and storage, and
distribution, some sources are calculated using potential emission factors, and CH4 that is not emitted is deducted
from the total CH4 potential estimates. To account for use of such technologies and practices that result in lower
emissions but are not reflected in "potential" emission factors, data are collected on both regulatory and voluntary
reductions. Regulatory actions addressed using this method include EPA National Emission Standards for
Hazardous Air Pollutants (NESHAP) regulations for dehydrator vents. Voluntary reductions included in the
Inventory are those reported to Natural Gas STAR and Methane Challenge for certain sources.
Methodological recalculations were applied to the entire time series to ensure time-series consistency from 1990
through 2022. GHGRP data available (starting in 2011) and other recent data sources have improved estimates of
emissions from natural gas systems. To develop a consistent time series, for sources with new data, EPA reviewed
available information on factors that may have resulted in changes over the time series (e.g., regulations, voluntary
actions) and requested stakeholder feedback on trends as well. For most sources, EPA developed annual data for
1993 through 2010 by interpolating activity data or emission factors or both between 1992 and 2011 data points.
Information on time-series consistency for sources updated in this year's Inventory can be found in the
Recalculations Discussion below, with additional detail provided in supporting memos (relevant memos are cited in
the Recalculations Discussion). For detailed documentation of methodologies, please see Annex 3.5.
The United States reports data to the UNFCCC using this Inventory report along with Common Reporting Tables
(CRTs). This note is provided for those reviewing the CRTs: The notation key "IE" is used for CO2 and CH4 emissions
from venting and flaring in CRT l.B.2. Disaggregating flaring and venting estimates across the Inventory would
involve the application of assumptions and could result in inconsistent reporting and, potentially, decreased
transparency. Data availability varies across segments within oil and gas activities systems, and emission factor
data available for activities that include flaring can include emissions from multiple sources (flaring, venting and
leaks).
Uncertainty
EPA has conducted a quantitative uncertainty analysis using the IPCC Approach 2 methodology (Monte Carlo
Simulation technique) to characterize the uncertainty for natural gas systems. For more information on the
approach, please see the memoranda Inventory of U.S. Greenhouse Gas Emissions and Sinks 1990-2016: Natural
Gas and Petroleum Systems Uncertainty Estimates (2018 uncertainty memo) and Inventory of U.S. Greenhouse Gas
Emissions and Sinks 1990-2019: Update for Natural Gas and Petroleum Systems CO2 Uncertainty Estimates (2021
3-100 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
uncertainty memo).79
EPA used Microsoft Excel's @RISK add-in tool to estimate the 95 percent confidence bound around Cm and CO2
emissions from natural gas systems for the current Inventory. For the CFU uncertainty analysis, EPA focused on the
17 highest-emitting sources for the year 2022, which together emitted 75 percent of methane from natural gas
systems in 2022, and extrapolated the estimated uncertainty for the remaining sources. For the CO2 uncertainty
analysis, EPA focused on the three highest-emitting sources for the year 2022, which together emitted 81 percent
of CO2 from natural gas systems in 2022, and extrapolated the estimated uncertainty for the remaining sources. To
estimate uncertainty for N2O, EPA applied the uncertainty bounds calculated for CO2. EPA will seek to refine this
estimate in future Inventories. The @ RISK add-in provides for the specification of probability density functions
(PDFs) for key variables within a computational structure that mirrors the calculation of the inventory estimate. For
emission factors that are derived from methane emissions measurement studies, the PDFs are commonly
determined to be lognormally distributed (GRI/EPA 1996; GTI 2001; GTI 2009; Lamb et al. 2015; Zimmerle et al.
2015; Fischer et al. 2018; GTI 2019). For activity data that are derived from national datasets, the PDFs are set to a
uniform distribution (see 2018 and 2021 uncertainty memos). Many emission factors and activity factors are
calculated using Subpart W data, and for these, the @ RISK add-in determines the best fitting PDF (e.g., lognormal,
gaussian), based on bootstrapping of the underlying data (see 2018 and 2021 uncertainty memos). The IPCC
guidance notes that in using this Approach 2 method, "some uncertainties that are not addressed by statistical
means may exist, including those arising from omissions or double counting, or other conceptual errors, or from
incomplete understanding of the processes that may lead to inaccuracies in estimates developed from models."
The uncertainty bounds reported below only reflect those uncertainties that EPA has been able to quantify and do
not incorporate considerations such as modeling uncertainty, data representativeness, measurement errors,
misreporting or misclassification. The understanding of the uncertainty of emission estimates for this category
evolves and improves as the underlying methodologies and datasets improve.
The results presented below provide the 95 percent confidence bound within which actual emissions from this
source category are likely to fall for the year 2022, using the IPCC methodology. The results of the Approach 2
uncertainty analysis are summarized in Table 3-79. Natural gas systems CH4 emissions in 2022 were estimated to
be between 141.2 and 203.0 MMT CO2 Eq. at a 95 percent confidence level. Natural gas systems CO2 emissions in
2022 were estimated to be between 31.9 and 42.1 MMT CO2 Eq. at a 95 percent confidence level. Natural gas
systems N2O emissions in 2022 were estimated to be between 0.13 and 0.18 MMT CO2 Eq. at a 95 percent
confidence level.
Uncertainty bounds for other years of the time series have not been calculated, but uncertainty is expected to vary
over the time series. For example, years where many emission sources are calculated with interpolated data would
likely have higher uncertainty than years with predominantly year-specific data. In addition, the emission sources
that contribute the most to CFU and CO2 emissions are different over the time series, particularly when comparing
recent years to early years in the time series. For example, venting emissions were higher and flaring emissions
were lower in early years of the time series, compared to recent years. Technologies also changed over the time
series (e.g., liquids unloading with plunger lifts and reduced emissions completions were not used early in the time
series and cast iron distribution mains were more prevalent than plastic mains in early years). Transmission and
gas processing compressor leak and vent emissions were also higher in the early years of the time series.
Table 3-79: Approach 2 Quantitative Uncertainty Estimates for CH4 and Non-combustion CO2
Emissions from Natural Gas Systems (MMT CO2 Eq. and Percent)
Source
Gas
2022 Emission Estimate
(MMT CO . Eq.)1,
Uncertainty Range Relative to Emission Estimate-1
(MMTCO' Eq.) (%)
Lower Upper
Bound1, Bound1,
Lower
Bound1,
Upper
Bound1,
Natural Gas Systems
ch4
173.1
141.2 203.0
-18%
+17%
79 See https://www.epa.gov/eheemissions/natural-eas-and-petroleum-svstems.
Energy 3-101
-------
Natural Gas Systems C02 36.5 31.9 42.1 -12% +15%
Natural Gas Systems N20 0.15 0.13 0.18 -12% +15%
a Range of emission estimates estimated by applying the 95 percent confidence intervals obtained from the Monte Carlo
simulation analysis conducted for the year 2022 CH4 and C02 emissions.
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 3-73 and Table 3-74.
QA/QC and Verification Discussion
In order to ensure the quality of the emission estimates for natural gas systems, general (IPCC Tier 1) Quality
Assurance/Quality Control (QA/QC) procedures were implemented consistent with the U.S. Inventory QA/QC plan
outlined in Annex 8.
The natural gas systems emission estimates in the Inventory are continually being reviewed and assessed to
determine whether emission factors and activity factors accurately reflect current industry practices. A QA/QC
analysis was performed for data gathering and input, documentation, and calculation. QA/QC checks are
consistently conducted to minimize human error in the model calculations. EPA performs a thorough review of
information associated with new studies, GHGRP data, regulations, public webcasts, and the Natural Gas STAR
Program to assess whether the assumptions in the Inventory are consistent with current industry practices. The
EPA has a multi-step data verification process for GHGRP data, including automatic checks during data-entry,
statistical analyses on completed reports, and staff review of the reported data. Based on the results of the
verification process, the EPA follows up with facilities to resolve mistakes that may have occurred.80
As in previous years, EPA conducted early engagement and communication with stakeholders on updates prior to
public review of the current Inventory. EPA held a stakeholder webinar in October of 2023. EPA released memos
detailing updates under consideration and requesting stakeholder feedback.
In recent years, several studies have measured emissions at the source level and at the national or regional level
and calculated emission estimates that may differ from the Inventory. There are a variety of potential uses of data
from new studies, including replacing a previous estimate or factor, verifying or QA of an existing estimate or
factor, and identifying areas for updates. In general, there are two major types of studies related to oil and gas
greenhouse gas data: studies that focus on measurement or quantification of emissions from specific activities,
processes and equipment, and studies that use tools such as inverse modeling to estimate the level of overall
emissions needed to account for measured atmospheric concentrations of greenhouse gases at various scales. The
first type of study can lead to direct improvements to or verification of Inventory estimates. In the past few years,
EPA has reviewed and, in many cases, incorporated data from these data sources. The second type of study can
provide general indications of potential over- and under-estimates. In addition, in recent years information from
top-down studies has been directly incorporated to quantify emissions from well blowouts.
A key challenge in using these types of studies to assess Inventory results is having a relevant basis for comparison
(e.g., the two data sets should have comparable time frames and geographic coverage, and the independent study
should assess data from the Inventory and not another data set, such as the Emissions Database for Global
Atmospheric Research, or "EDGAR"). In an effort to improve the ability to compare the national-level Inventory
with measurement results that may be at other spatial and temporal scales, EPA has developed a gridded
inventory of U.S. anthropogenic methane emissions with 0.1 degree x 0.1 degree spatial resolution, monthly
temporal resolution, and detailed scale-dependent error characterization.81 The most recent version of the
gridded methane inventory is designed to be consistent with the U.S. EPA's Inventory of U.S. Greenhouse Gas
Emissions and Sinks: 1990-2018 estimates for the years 2012 to 2018. The gridded inventory improves efforts to
compare results of this Inventory with atmospheric studies.
80 See https://www.epa.gov/sites/production/files/2015-07/documents/ghgrp verification factsheet.pdf.
81 See https://www.epa.gov/eheemissions/us-eridded-methane-emissions.
3-102 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
Recalculations Discussion
EPA received information and data related to the emission estimates through GHGRP reporting and presented
information to stakeholders regarding the updates under consideration. In November 2023, EPA released draft
memoranda that discussed changes under consideration and requested stakeholder feedback on those changes.
EPA then released final memoranda documenting the methodology implemented in the current Inventory.82
Memoranda cited in the Recalculations Discussion below are: Inventory of U.S. Greenhouse Gas Emissions and
Sinks 1990-2022: Updates for Transmission Compressor Station Activity (Transmission Station Activity memo),
Inventory of U.S. Greenhouse Gas Emissions and Sinks 1990-2022: Updates for Completion and Workover
Emissions (Completions and Workovers memo), and Inventory of U.S. Greenhouse Gas Emissions and Sinks 1990-
2022: Updates for Underground Natural Gas Storage Well Emission Events (Storage Well Events memo).
EPA evaluated relevant information available and made several updates to the Inventory, including for
transmission compressor stations, completions and workovers, and underground natural gas storage wells.
General information for these source specific recalculations are presented below and details are available in the
Transmission Station Activity, Completions and Workovers, and Storage Well Events memos, including additional
considerations for the updates.
In addition to the updates to the sources mentioned above, for certain sources, Cm and/or CO2 emissions changed
by greater than 0.05 MMT CO2 Eq., comparing the previous estimate for 2021 to the current (recalculated)
estimate for 2021. The emissions changes were mostly due to GHGRP data submission revisions. These sources are
discussed below and include pneumatic controllers, well pad equipment leaks, condensate tanks, liquids
unloading, gas engines (in production segment), miscellaneous production flaring, gathering and boosting (G&B)
station blowdowns, G&B pneumatic controllers, G&B yard piping, G&B acid gas removal units (AGRU), natural gas
processing flares, and transmission pipeline venting.
The combined impact of revisions to 2021 natural gas systems CH4 emissions, compared to the previous Inventory,
is a decrease from 181.4 to 174.6 MMT CO2 Eq. (6.8 MMT CO2 Eq., or 4 percent). The recalculations resulted in an
average increase in the annual Cm emission estimates across the 1990 through 2021 time series, compared to the
previous Inventory, of 1.9 MMT CO2 Eq., or 0.8 percent.
The combined impact of revisions to 2021 natural gas systems CO2 emissions, compared to the previous Inventory,
is a decrease from 36.2 MMT to 35.8 MMT, or 1.1 percent. The recalculations resulted in an average increase in
emission estimates across the 1990 through 2021 time series, compared to the previous Inventory, of 0.4 MMT
CO2 Eq., or 1.5 percent.
The combined impact of revisions to 2021 natural gas systems N2O emissions, compared to the previous Inventory,
is an increase from 7.6 kt CO2 Eq. to 12.6 kt CO2 Eq., or 65 percent. The recalculations resulted in an average
increase in emission estimates across the 1990 through 2021 time series, compared to the previous Inventory, of
10.4 percent.
In Table 3-80 and Table 3-81 below are categories in natural gas systems with recalculations resulting in a change
of greater than 0.05 MMT CO2 Eq., comparing the previous estimate for 2021 to the current (recalculated)
estimate for 2021. No changes made to N2O estimates resulted in a change greater than 0.05 MMT CO2 Eq. For
more information, please see the Recalculations Discussion below.
82 Stakeholder materials including draft and final memoranda for the current (i.e., 1990 to 2022) Inventory are available at
https://www.epa.gov/eheemissions/natural-eas-and-petroleum-svstems.
Energy 3-103
-------
Table 3-80: Recalculations of CO2 in Natural Gas Systems (MMT CO2)
Segment and Emission Sources with Previous Estimate Year
Current Estimate
Current Estimate
Changes of Greater than 0.05 MMT CO ¦
2021,
Year 2021,
Year 2022,
due to Recalculations
2023 Inventory
2024 Inventory
2024 Inventory
Exploration
+
+
+
Gas Well Completions
+
+
+
Production
9.1
9.3
8.6
Gas Well Workovers
+
+
+
Pneumatic Controllers
0.1
+
+
Well Pad Equipment Leaks
+
0.1
0.1
Chemical Injection Pumps
+
+
+
Misc. Onshore Production Flaring
1.0
1.1
0.7
Condensate Tanks
0.9
1.1
0.6
G&B Station - AGRU
2.3
2.2
2.0
Processing
26.1
25.5
26.7
Flares
7.4
6.9
8.5
Transmission and Storage
0.9
0.9
1.2
Transmission Compressor Station Leaks
and Venting
0.1
0.1
0.1
Storage Wells
+
+
+
Distribution
+
+
+
Post-Meter
+
+
+
Total
36.2
35.8
36.5
+ Does not exceed 0.05 MMT C02.
Note: Totals may not sum due to independent rounding.
Table 3-81: Recalculations of CH4 in Natural Gas Systems (MMT CO2 Eq.)
Segment and Emission Sources with Changes
Previous Estimate
Current Estimate Year
Current Estimate
of Greater than 0.05 MMT CO ¦ due to
Year 2021,
2021,
Year 2022,
Recalculations
2023 Inventory
2024 Inventory
2024 Inventory
Exploration
0.2
0.1
0.2
Gas Well Completions
0.2
0.1
0.2
Production
94.1
92.2
89.7
Gas Well Workovers
0.2
0.04
0.05
Well Pad Equipment Leaks
9.6
9.4
10.8
Pneumatic Controllers
21.3
20.9
18.0
Condensate Tanks
1.2
1.2
1.1
Liquids Unloading
3.4
2.8
2.4
Gas Engines
5.5
5.3
5.3
G&B Stations - Station Blowdowns
1.2
1.0
0.9
G&B Stations - Pneumatic Controllers
0.7
0.6
0.5
G&B Station - Yard Piping
2.6
2.7
2.8
Processing
14.3
14.2
15.1
Flares
0.8
0.8
0.9
Transmission and Storage
44.5
39.8
39.6
Transmission Compressor Station Leaks and
Venting
25.9
21.3
21.5
Storage Wells
0.3
0.3
0.3
Pipeline Venting
4.8
4.6
3.7
Distribution
15.3
15.3
15.2
Post-Meter
13.0
13.0
13.4
Total
181.4
174.6
173.1
Note: Totals may not sum due to independent rounding.
3-104 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
Exploration
HF Completions (Methodological Update)
EPA updated the calculation methodology for HF completions to use basin-level HF completion counts from
Enverus and basin-specific activity factors and emission factors calculated from Subpart W data for each control
category (i.e., non-reduced emission completion (REC) with venting, non-REC with flaring, REC with venting, REC
with flaring). Previously, national annual average activity and emission factors calculated using Subpart W data
were applied to national completion counts to estimate HF gas well completion emissions. In this update, EPA
developed national emission estimates by summing calculated basin-level total emission estimates. The
Completions and Workovers memo presents additional information and considerations for this update.
EPA calculated basin-specific activity factors and CH4 and CO2 emission factors for all basins that reported Subpart
W data. The factors were year-specific for reporting year (RY) 2011 through RY2022. For basin-level HF completion
event counts, EPA used Enverus data for 1990 to 2010 and Subpart W for 2011 forward. For the fraction of
completions in each control subcategory, EPA retained the previous Inventory's assumption that all HF gas well
completions were non-REC for 1990 to 2000. The previous Inventory also assumed that 10 percent of HF
completions were non-REC with flaring from 1990 to 2010 (based on national Subpart W data for RY2011 and
RY2012); EPA updated this value using basin-specific Subpart W data for RY2011 and RY2012. For 2011 to 2022,
EPA determined the percent contribution of each control category directly from Subpart W data and used linear
interpolation between 2000 and 2011 to determine the percent of gas wells with RECs. EPA developed year- and
basin-specific Subpart W EFs for 2011 forward. Year 2011 emission factors were applied to all prior years for each
basin. For basins without Subpart W data available, EPA applied national average activity and emission factors
(unweighted average of all Subpart W reported data).
Comparing the final completion emissions and those presented at the October 2023 webinar and in the November
2023 Completions and Workovers memo, the final estimates are higher for certain completion categories. These
emissions increases are due not to the basin-level methodology changes discussed here but rather to changes in
the Enverus dataset. EPA applied the same data processing steps to Enverus data in the fall of 2023 as it did for the
previous Enverus data analysis (conducted in 2021) and data changes led to many more completions being
classified as HF completions.
As a result of this methodological update, CH4 emissions estimates for HF completions are on average 55 percent
higher across the time series than in the previous Inventory. The 2021 CH4 emissions estimate is 54 percent lower
than in the previous Inventory. The largest increase between the updates and the previous Inventory for CH4
emissions is 134 percent in 1998 with an average increase of 92 percent over the 1990 through 2010 time period.
CH4 emissions decreased or were similar for 2011 forward and the largest decrease between the updates and the
previous Inventory is 54 percent in 2021. CH4 emissions increased on average across the time series, but
particularly in earlier years due to gas well HF completions that were non-REC with venting, particularly in the
Appalachian basin (Eastern Overthrust) [basin 160a], The Appalachian basin (Eastern Overthrust) had a large
number of gas well HF completion events that were non-REC with venting and the highest EF of any basin. The
update resulted in CO2 emissions estimates that are on average 54 percent higher across the time series than in
the previous Inventory. The 2021CO2 emissions estimate is 3 percent lower than in the previous Inventory. The
largest increase between the updates and the previous Inventory for CO2 emissions is 141 percent in 1996, and the
largest decrease between the updates and the previous Inventory is 44 percent in 2013. CO2 emissions increased
predominantly due to non-REC with flaring events in the Appalachian basin (basin 160A) and the East Texas basin
(basin 260); these two basins had the highest EFs of any basin.
Table 3-82: HF Completions National CH4 Emissions (Metric Tons CH4)
Source
1990
2005
2018
2019
2020
2021
2022
HF Completions - Non-REC with Venting
187,841
590,423
1,210
678
75
166
83
HF Completions - Non-REC with Flaring
3,112
11,791
652
399
154
31
1,605
HF Completions - REC with Venting
NO
6,710
28,946
18,150
4,594
2,487
3,376
Energy 3-105
-------
HF Completions - REC with Flaring
NO
2,238 I
1,345
1,148
634
127
190
Total Emissions
190,954
611,162
32,154
20,375
5,458
2,811
5,253
Previous Estimate
111,265
345,098
32,147
20,002
5,220
6,111
NA
NO (Not Occurring)
NA (Not Applicable)
Table 3-83: HF Completions National CO2 Emissions (kt CO2)
Source
1990
2005
2018
2019
2020
2021
2022
HF Completions - Non-REC with Venting
15
50
+
+
+
+
+
HF Completions - Non-REC with Flaring
472
1
1
2,023
57
43
10
2
5
HF Completions - REC with Venting
NO
4
3
+
+
+
+
HF Completions - REC with Flaring
NO
I
496
233
199
87
13
31
Total Emissions
487
1
2,573
293
243
97
15
36
Previous Estimate
289
1,418
290
214
96
15
NA
+ Does not exceed 0.5 kt.
NO (Not Occurring)
NA (Not Applicable)
Non-HF Completions (Methodological Update)
EPA updated the activity data sources and calculation methodology for non-HF completions to use basin-level non-
HF completion counts from Enverus and basin-specific activity factors and emission factors, calculated from
Subpart W data for each control category (i.e., vented, flared). Previously, national non-HF completion counts and
national annual average activity and emission factors calculated using historical data analyses, and Subpart W data
were applied to estimate non-HF gas well completion emissions. In this update, EPA developed national emission
estimates by summing calculated basin-level total emission estimates. The Completions and Workovers memo
presents additional information and considerations for this update.
EPA calculated basin-specific activity factors and CH4 and CO2 emission factors for all basins that reported Subpart
W data. The factors were year-specific for reporting year (RY) 2011 through RY2022. For basin-level non-HF gas
well completion event counts, EPA used Enverus data across the time series. For the fraction of completions in
each control category, EPA implemented at the basin level the previous Inventory's approach and the percent of
non-HF gas well completions that are vented in 2011 is applied to all prior years. For 2011 to 2022, EPA
determined the percent contribution of each control category directly from Subpart W data for each basin. EPA
developed year- and basin-specific Subpart W EFs for 2011 forward. Year 2011 emission factors were applied to all
prior years for each basin. For basins without Subpart W data available, EPA applied national average activity and
emission factors (unweighted average of all Subpart W reported data).
As a result of this methodological update, CH4 emissions estimates for non-HF completions are on average 419
percent higher across the time series than in the previous Inventory. The 2021 CH4 estimate is 43 percent higher
than in the previous Inventory. The largest increase between the updates and the previous Inventory for CH4
emissions is 1,533 percent in 1991, and the largest decrease between the updates and the previous Inventory is 63
percent in 2018. Methane emissions increased primarily due to gas well non-HF completions that were vented. The
Appalachian basin (Eastern Overthrust) [basin 160a] and the Appalachian basin (basin 160) had many non-HF
completion events that were vented and average EFs more than 2 times higher than the national average across
the 1990-2022 time series. The update resulted in CO2 emissions estimates that are on average 1,312 percent
higher across the time series than in the previous Inventory. The 2021 CO2 emissions estimate is 7,331 percent
higher than in the previous Inventory. The largest increase between the updates and the previous Inventory for
CO2 emissions is 7,331 percent in 2021, and the largest decrease between the updates and the previous Inventory
is 47 percent in 2018. The increase in CO2 emissions is due to non-HF completions that were flared, primarily in the
Gulf Coast basin (basin 220). The Gulf Coast basin had the highest fraction of non-HF completions that were flared
of any basin.
3-106 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
Table 3-84: Non-HF Completions National CH4 Emissions (Metric Tons CH4)
Source
1990
2005
2018
2019
2020
2021
2022
Non-HF Completions - vented
44,362 1
84,955
79
319
1,548
284
159
Non-HF Completions - flared
547
545
113
+
NO
98
NO
Total Emissions
44,909
85,500
192
320
1,548
381
159
Previous Estimate
5,736
10,363
513
796
2,659
267
NA
+ Does not exceed 0.5 MT.
NO (Not Occurring)
NA (Not Applicable)
Table 3-85: Non-HF Completions National CO2 Emissions (Metric Tons CO2)
Source
1990
2005
2018
2019
2020
2021
2022
Non-HF Completions - vented
1,398
3,531
1
8
197
8
549
Non-HF Completions - flared
127,316
1
126,743
15,699
42
NO
16,449
NO
Total Emissions
128,714
I
130,274
15,700
50
197
16,457
549
Previous Estimate
4,862
8,784
29,834
81
364
221
NA
NO (Not Occurring)
NA (Not Applicable)
Production
HF Workovers (Methodological Update)
EPA updated the activity data source and calculation methodology for HF workovers to use basin-specific activity
factors and emission factors, calculated from Subpart W data for each control category (i.e., non-reduced emission
completion (REC) with venting, non-REC with flaring, REC with venting, REC with flaring). Previously, national HF
workover counts calculated using analyses for NSPS OOOO (i.e., 1 percent of HF gas wells were worked over
annually) and national annual average emission factors calculated using Subpart W data were applied to estimate
HF gas well workover emissions. In this update, EPA developed national emission estimates by summing calculated
basin-level total emission estimates. The Completions and Workovers memo presents additional information and
considerations for this update.
EPA calculated basin-specific activity factors and CH4 and CO2 emission factors for all basins that reported Subpart
W data. For basin-level workover counts, instead of applying a 1 percent workover rate to HF gas wells, EPA
developed year- and basin-specific Subpart W AFs for 2015 forward that represent the number of HF workovers
per gas well. Year 2015 Subpart W AFs were applied to all prior years for each basin. For the fraction of workovers
in each control subcategory, EPA retained the previous Inventory's assumption that all HF gas well workovers were
non-REC for 1990 to 2000. The previous Inventory also assumed that 10 percent of HF workovers were non-REC
with flaring from 1990 to 2010; EPA updated this value using basin-specific data from Subpart W. For 2011
forward, EPA determined the percent contribution of each control category directly from Subpart W data at the
basin-level. EPA used linear interpolation for interpolation between 2000 and 2011 to determine the percent of
gas wells with RECs. EPA developed year-and basin-specific Subpart W EFs for 2011 forward. Year 2011 emission
factors were applied to all prior years for each basin. For basins without Subpart W data available, EPA applied
national average activity and emission factors (unweighted average of all Subpart W reported data).
As a result of this methodological update, CH4 emissions estimates for HF workovers are on average 43 percent
lower across the time series than in the previous Inventory. The 2021CH4 emissions estimate is 94 percent lower
than in the previous Inventory. The largest increase between the updates and the previous Inventory for CH4
emissions is 5 percent in 1990, and the largest decrease between the updates and the previous Inventory is 99
percent in 2019. The update resulted in CO2 emissions estimates that are on average 60 percent lower than in the
previous Inventory. The 2021CO2 emissions estimate is 67 percent lower than in the previous Inventory. The
largest increase between the updates and the previous Inventory for CO2 emissions is 53 percent in 2013, and the
Energy 3-107
-------
largest decrease between the updates and the previous Inventory is 99 percent in 2019. The decrease in emissions
for both Cm and CO2 was primarily due to the change in calculation method for workover counts. HF gas well
workover counts decreased by an average of 73 percent across the 1990 through 2021 time series compared to
the previous Inventory.
Table 3-86: HF Workovers National CH4 Emissions (Metric Tons CH4)
Source
1990
2005
2018
2019
2020
2021
2022
HF Workovers - Non-REC with Venting
22,198 1
37,917 1
114
96
4
17
35
HF Workovers-Non-REC with Flaring
242 1
483 1
29
2
1
+
19
HF Workovers - REC with Venting
NO I
2311
1,667
73
229
457
113
HF Workovers - REC with Flaring
NO 1
+
6
4
1
2
17
Total Emissions
¦
22,440
38,632
1,816
174
234
476
185
Previous Estimate
21,4271
57,972 |
19,594
13,612
6,771
8,144
NA
+ Does not exceed 0.5 mt.
NO (Not Occurring)
NA (Not Applicable)
Table 3-87: HF Workovers National CChEmissions (kt CO2)
Source
1990
2005
2018
2019
2020
2021
2022
HF Workovers - Non-REC with Venting
1
2 1
+
+
+
+
+
HF Workovers-Non-REC with Flaring
32 1
671
2
+
+
+
+
HF Workovers - REC with Venting
NO
+
+
+
+
+
+
HF Workovers - REC with Flaring
NO i
1
+
+
+
3
Total Emissions
33
69
4
1
+
+
3
Previous Estimate
56
165
99
86
8
1
NA
+ Does not exceed 0.5 kt.
NO (Not Occurring)
NA (Not Applicable)
Non-HF Workovers (Methodological Update)
EPA updated the activity data source and calculation methodology for non-HF workovers to use basin-specific
activity factors and emission factors, calculated from Subpart W data for each control category (i.e., vented,
flared). Previously, national annual average activity and emission factors calculated using historical data analyses
and Subpart W data were applied along with national gas well counts to estimate non-HF gas well workover
emissions. In this update, EPA developed national emission estimates by summing calculated basin-level total
emission estimates. The Completions and Workovers memo presents additional information and considerations for
this update.
EPA calculated basin-specific activity factors and CH4 and CO2 emission factors for all basins that reported Subpart
W data. For basin-level workover counts, EPA developed year-specific Subpart W AFs for 2015 forward. Year 2015
Subpart W data was applied to prior years for each basin. For the fraction of workovers in each control
subcategory, EPA applied year- and basin-specific AFs for 2011 forward, retained the previous Inventory's
assumption that all non-HF workovers were vented in 1990 to 1992, and used linear interpolation between the
1992 and 2011 activity factors at the basin-level. EPA developed year- and basin-specific Subpart W EFs for 2011
forward. Year 2011 emission factors were applied to all prior years for each basin. For basins without Subpart W
data available, EPA applied national average activity and emission factors (unweighted average of all Subpart W
reported data).
As a result of this methodological update, CH4 emissions estimates for non-HF workovers are on average 277
percent higher than in the previous Inventory. The 2021 estimate is 180 percent higher than in the previous
Inventory. The largest increase between the estimates and the previous Inventory for CH4 emissions is 698 percent
3-108 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
in 2017, and the smallest increase is 78 percent in 2018. The increase in Cm emissions is due to non-HF workovers
that were vented. In the Chautauqua Platform basin (basin 355), all non-HF workovers were vented, and it had the
highest Cm EF in 2017. The update resulted in CO2 emissions estimates for non-HF workovers that are higher
across the entire time series than the previous Inventory (on average 3,067 percent higher). The 2021 CO2
emissions estimate is 116 percent higher than in the previous Inventory. The largest increase between estimates
and the previous Inventory for CO2 emissions is 7,432 percent in 1994, and the smallest increase is 78 percent in
2018. The increase in CO2 emissions is due to non-HF workovers that were flared, particularly in the Bend Arch
basin (basin 425). The Bend Arch basin has a high non-HF completion per total gas well AF and a high fraction of
non-HF workovers that were flared compared to other basins over the time series.
Table 3-88: Non-HF Workovers National CH4 Emissions (Metric Tons CH4)
Source
1990
2005
2018
2019
2020
2021
2022
Non-HF Workovers - vented
Non-HF Workovers - flared
1,415
NO
2,083
1,077
1,139
1
866
5
959
+
1,068
43
1,594
3
Total Emissions
1,415
3,159
1,140
870
959
1,111
1,597
Previous Estimate
532
752
415
436
259
396
NA
+ Does not exceed 0.5 MT.
NO (Not Occurring)
NA (Not Applicable)
Table 3-89: Non-HF Workovers National CO2 Emissions (Metric Tons CO2)
Source
1990
2005
2018
2019
2020
2021
2022
Non-HF Workovers - vented
87 1
180
61
51
2,163
199
460
Non-HF Workovers - flared
NO
174,881
269
665
34
9,591
424
Total Emissions
87
175,062
329
716
2,197
9,790
885
Previous Estimate
32
3,701
185
294
476
4,539
NA
NO (Not Occurring)
NA (Not Applicable)
Equipment Leaks (Methodological Update)
In the previous Inventory, EPA updated the CH4 emissions calculation methodology for onshore production
equipment leaks to use basin-specific equipment-level activity factors (e.g., separators per well) from GHGRP data.
However, the CO2 emissions calculation methodology was not updated and instead the previous Inventory still
relied on a national-level methodology to estimate CO2 emissions. For this year's Inventory, EPA calculated
equipment leak CO2 emissions in the same manner as CH4 emissions. EPA calculated CO2 estimates using the basin-
specific equipment-level activity factors for RY2015 through RY2022 from GHGRP, consistent with the
methodology used to calculate the CH4 activity factors, and the CO2 emissions factors for onshore production
segment equipment leaks. Note, this methodological update applies only for activity factors. The previous
Inventory's CO2 emission factors for onshore production segment equipment leaks (by equipment type) were
retained and used to develop CO2 estimates.
The update for CO2 emission estimates for equipment leaks resulted in an average increase of 12 percent across
the time series compared to the previous Inventory. Years 1990 through 2002 were minimally impacted by the
updates, with an increase of 2 percent for CO2 emissions. Years 2020 and 2021 showed larger increases of 69 and
48 percent for 2020 and 2021, respectively, which is mostly due to much higher emissions from meters and piping
in the Powder River Basin.
Methane emissions for equipment leaks were impacted due to recalculations with updated data. CH4 emission
estimates were an average of 5 percent lower across the time series compared to the previous Inventory. The 2021
CH4 estimate is 2 percent lower in 2021 compared to the previous Inventory. These CH4 emissions changes were
due to GHGRP submission revisions.
Energy 3-109
-------
Table 3-90: Production Equipment Leaks National CO2 Emissions (Metric Tons CO2)
Source
1990
2005
2018
2019
2020
2021
2022
Heaters
1,639
3,642
10,485
3,412
4,261
4,014
4,359
Separators
5'372 i
14,404
19,114
19,689
21,306
17,239
15,518
Dehydrators
1,261
1,690
794
677
589
893
683
Meters/Piping
5,400
10,527
11,199
11,775
30,564
24,424
11,022
Compressors
2,673
7,178
7,818
7,236
6,966
9,079
20,310
Total Emissions
16,344
37,441
49,410
42,789
63,686
55,649
51,893
Previous Estimate
18,497
33,300
38,458
37,974
37,608
37,608
NA
NA (Not Applicable)
Table 3-91: Production Equipment Leaks National CH4 Emissions (Metric Tons CH4)
Source
1990
2005
2018
2019
2020
2021
2022
Heaters
12,305
18,436
79,890
16,158
18,568
17,581
18,476
Separators
41,579 ¦¦
80,745 :
121,349
126,037
129,133
109,610
94,591
Dehydrators
12,904
11,381
5,449
3,656
3,070
4,078
3,105
Meters/Piping
43,055 ::::
63,764
79,864
84,730
153,917
130,390
75,719
Compressors
30,307
61,753
71,705
64,771
60,637
73,636
193,389
Total Emissions
140,150
236,079
358,256
295,352
365,325
335,295
385,280
Previous Estimate
137,647
262,188
363,367
298,930
369,466
343,686
NA
NA (Not Applicable)
Chemical Injection Pumps (Methodological Update)
In the previous Inventory, EPA updated the Cm emissions calculation methodology for chemical injection pumps to
use basin-specific equipment-level activity factors (e.g., pumps per well) from GHGRP data. However, the CO2
emissions calculation methodology was not updated and instead the previous Inventory still relied on a national-
level methodology to estimate CO2 emissions. For the current Inventory, EPA calculated chemical injection pump
CO2 emissions in the same manner as CH4 emissions. EPA calculated CO2 estimates using the basin-specific
equipment-level activity factors for RY2015 through RY2022 from GHGRP, consistent with the methodology used
to calculate the CH4 activity factors, and the CO2 emission factor. Note, this methodological update applies only for
activity factors. The previous Inventory's chemical injection pumps CO2 emission factor was retained and used to
develop CO2 estimates.
The update for CO2 emission estimates resulted in an average decrease of 9 percent across the time series
compared to the previous Inventory. There were larger decreases of 31 and 39 percent for 2020 and 2021,
respectively. The recent years of the time series used basin-specific activity factors and certain basins had lower
activity factors compared to the national average factors (e.g., Permian Basin, Denver Basin, San Juan, Paradox, AK
Cook Inlet).
Table 3-92: Chemical Injection Pumps National CO2 Emissions (Metric Tons CO2)
Source
1990
2005
2018
2019
2020
2021
2022
Chemical Injection Pumps
2,153
6,749
10,254
9,392
7,339
6,469
6,428
Previous Estimate
2,275
7,760
11,053
10,899
10,635
10,635
NA
NA (Not Applicable)
Pneumatic Controllers (Methodological Update)
In the previous Inventory, EPA updated the CH4 emissions calculation methodology for pneumatic controllers to
use basin-specific activity factors and emission factors by bleed type (i.e., low, high, intermittent bleed) from
GHGRP data. However, the CO2 emissions calculation methodology was not updated and instead the previous
3-110 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
Inventory still relied on a national-level methodology to estimate CO2 emissions. For the current Inventory, EPA
calculated pneumatic controller CO2 emissions using basin-specific emissions data such that the CO2 emissions
reflect the unique CO2 composition of the gas in a basin.
The update for pneumatic controller CO2 emission estimates resulted in an average decrease of 44 percent across
the time series compared to the previous Inventory. This average decrease is generally consistent for all years.
Methane emissions for pneumatic controllers were impacted due to recalculations with updated data. Methane
emissions estimates are on average 0.3 percent lower across the time-series than in the previous Inventory. The
estimate for 2021 is 2 percent lower than in the previous Inventory. These changes were due to GHGRP submission
revisions.
Table 3-93: Pneumatic Controllers National CO2 Emissions (Metric Tons CO2)
Source
1990
2005
2018
2019
2020
2021
2022
Low Bleed Controllers
High Bleed Controllers
Intermittent Bleed Controllers
NO 1
23,058
15,076
1,318 1
32,794
41,267 1
2,325
4,378
56,974
2,283
2,898
57,147
2,047
2,373
48,052
1,773
2,418
44,964
2,180
1,726
37,897
Total Emissions
38,135
75,378
63,677
62,328
52,472
49,155
41,803
Previous Estimate
70,028
129,648
115,235
115,591
98,144
91,662
NA
NO (Not Occurring)
NA (Not Applicable)
Table 3-94: Pneumatic Controllers National CH41
Emissions (Metric Tons CH4)
Source
1990
2005
2018
2019
2020
2021
2022
Low Bleed Controllers
High Bleed Controllers
Intermittent Bleed Controllers
NO 1
355,671
233,661
22,656 1
480,272
565,070
33,220
86,764
830,140
31,099
52,676
875,168
27,238
42,269
748,219
25,247
41,435
680,708
29,877
30,700
583,144
Total Emissions
589,332
1,067,997
950,124
958,943
817,727
747,391
643,721
Previous Estimate
581,039
1,075,712
956,125
959,080
814,318
760,534
NA
NO (Not Occurring)
NA (Not Applicable)
Storage Tanks (Recalculation with Updated Data)
Methane emissions for production condensate storage tanks are on average lower than the previous Inventory by
less than 0.1 percent across the 1990 to 2021 time series. The 2021 estimate is 6 percent lower than in the
previous Inventory. The production storage tanks CO2 emissions estimates are on average 3 percent higher across
the 1990 to 2021 time series than in the previous Inventory. The 2021 estimate is 25 percent higher than in the
previous Inventory. These changes were due to GHGRP submission revisions.
Table 3-95: Storage Tanks National CH4 Emissions (Metric Tons CH4)
Source
1990
2005
2018
2019
2020
2021
2022
Large Tanks w/Flares
520
337
1,284
819
765
715
467
Large Tanks w/VRU
NO *
27
141
464
743
597
94
Large Tanks w/o Control
16,743
6,828
15,179
3,149
5,323
4,205
3,319
Small Tanks w/Flares
NO
51
235
207
200
161
166
Small Tanks w/o Flares
92,334
31,003
43,050
61,331
47,423
35,460
33,455
Malfunctioning Separator Dump Valves
^ '
4 I
40
79
255
212
67
Total Emissions
109,605
38,250 ...
59,929
66,049
54,708
41,351
37,567
Previous Estimate
106,429
38,461
60,556
67,595
53,613
44,217
NA
NO (Not Occurring)
NA (Not Applicable)
Energy 3-111
-------
Table 3-96: Storage Tanks National CO2 Emissions (kt CO2)
Source
1990
2005
2018
2019
2020
2021
2022
Large Tanks w/Flares
587
4211
1,476
974
1,108
1,049
575
Large Tanks w/VRU
NO
J
+
+
1
1
+
Large Tanks w/o Control
2
36
1
1
1
+
Small Tanks w/Flares
NO
13
86
82
41
27
20
Small Tanks w/o Flares
48
17
26
32
23
17
18
Malfunctioning Separator Dump Valves
+
iiiiii!
+
+
+
1
+
+
Total Emissions
637
455
1,625
1,089
1,175
1,094
613
Previous Estimate
628
456 1
1,507
956
862
873
NA
+ Does not exceed 0.5 kt.
NO (Not Occurring)
NA (Not Applicable)
Liquids Unloading (Recalculation with Updated Data)
Liquids unloading Cm emissions estimates decreased by an average of 9 percent across the 1990 to 2021 time
series compared with the previous Inventory. The 2021 estimate decreased by 17 percent compared with the
previous Inventory. These changes were due to GHGRP submission revisions.
Table 3-97: Liquids Unloading National CH4 Emissions (Metric Tons CH4)
Source
1990
2005
2018
2019
2020
2021
2022
Liquids Unloading With Plunger Lifts
0 1
128,295
82,501
75,081
51,457
33,916
30,384
Liquids Unloading Without Plunger Lifts
77.767
198.728
132,866
104,484
84,251
65,655
54,227
Total Emissions
77,767
327,023
215,367
179,565
135,707
99,572
84,611
Previous Estimate
76,815
358,925
265,173
209,964
158,968
120,145
NA
NA (Not Applicable)
Gas Engines (Recalculation with Updated Data)
Gas engines Cm emissions estimates are on average 2 percent lower across the 1990 to 2021 time series
compared with the previous Inventory. The 2021 estimate is 4 percent lower than in the previous Inventory. These
changes were due to revisions to Enverus data.
Table 3-98: Production Gas Engines National CH4 Emissions (Metric Tons CH4)
Source
1990
2005
1
2018
2019
2020
2021
2022
Gas Engines
116,587 |
194,140
201,660
196,996
192,854
188,374
188,611
Previous Estimate
115,689 I
198,005
207,052
202,060
197,027
197,027
NA
NA (Not Applicable)
Miscellaneous Production Flaring (Recalculation with Updated Data)
Miscellaneous production flaring CO2 emissions estimates are on average 1 percent higher across the 1993 to 2021
time series compared with the previous Inventory, and the 2021 estimate is 12 percent higher compared to the
previous Inventory. These changes were due to GHGRP submission revisions.
Table 3-99: Miscellaneous Production Flaring National Emissions (kt CO2)
Source
1990
2005
2018
2019
2020
2021
2022
Miscellaneous Flaring-Gulf Coast Basin
NO
164
135
395
251
316
206
Miscellaneous Flaring-Williston Basin
NO
+ I
6
3
4
7
+
Miscellaneous Flaring-Permian Basin
NO
263 I
690
926
808
578
231
3-112 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
Miscellaneous Flaring-Other Basins
NO 1
117 I
476
334
237
211
238
Total Emissions
NO
544
1,308
1,659
1,301
1,112
676
Previous Estimate
NO
5431
1,326
1,595
1,298
991
NA
+ Does not exceed 0.5 kt.
NO (Not Occurring)
NA (Not Applicable)
Gathering and Boosting - Station Blowdowns (Recalculation with Updated Data)
Methane emissions estimates for gathering and boosting station blowdowns are on average 0.7 percent lower
across the 1990 to 2021 time series than in the previous Inventory. The 2021 estimate is 17 percent lower than in
the previous Inventory. These changes were due to GHGRP submission revisions.
Table 3-100: Station Blowdowns National Emissions (Metric Tons CH4)
Source
1990
2005
2018
2019
2020
2021
2022
Station Blowdowns
20,218 1
26,155
79,313
39,059
40,519
35,161
32,036
Previous Estimate
20,5171
26,113
78,548
38,412
40,468
42,231
NA
NA (Not Applicable)
Gathering and Boosting - Pneumatic Controllers (Recalculation with Updated Data)
Gathering and boosting pneumatic controllers Cm emissions estimates are on average 0.1 percent higher across
the 1990 to 2021 time series compared with the previous Inventory. The emissions estimate for 2021 is 2 percent
lower than in the previous Inventory, largely because of a decrease in emissions from high-bleed pneumatic
devices. These changes were due to GHGRP submission revisions.
Table 3-101: Pneumatic Controllers National Emissions (Metric Tons CH4)
Source
1990
2005
2018
2019
2020
2021
2022
High-Bleed Pneumatic Devices
Intermittent Bleed Pneumatic Devices
Low-Bleed Pneumatic Devices
17,092 1
78,424
2,713
22,111
101,451
3,509
25,030
173,929
5,799
24,187
184,542
6,996
22,981
171,679
6,965
20,709
156,842
6,564
18,854
145,574
6,572
Total Emissions
98,229
127,072
204,758
215,725
201,625
184,116
171,000
Previous Estimate
99,843
127,073
204,748
215,339
201,415
187,290
NA
NA (Not Applicable)
Gathering and Boosting - Acid Gas Removal Units (Recalculation with Updated Data)
Carbon dioxide emissions estimates for acid gas removal units (AGRU) are on average 0.3 percent lower across the
1990 to 2021 time series compared with the previous Inventory. The emissions estimate for 2021 decreased by 4
percent compared to the previous Inventory. These changes were due to GHGRP submission revisions.
Table 3-102: Acid Gas Removal Units National Emissions (kt CO2)
Source
1990
2005
2018
2019
2020
2021
2022
AGRU
241
3111
707
1,191
1,629
2,222
2,044
Previous Estimate
245
311 1
707
1,288
1,655
2,304
NA
NA (Not Applicable)
Energy 3-113
-------
Gathering and Boosting - Yard Piping (Recalculation with Updated Data)
Methane emissions estimates for yard piping are on average 0.6 percent higher across the 1990 to 2021 time
series compared with the previous Inventory. The emissions estimate for 2021 is 4 percent higher than in the
previous Inventory. These changes were due to GHGRP submission revisions.
Table 3-103: Yard Piping National Emissions (Metric Tons CH4)
Source
1990
2005
2018
2019
2020
2021
2022
Yard Piping
36,319 1
46,984
86,270
94,306
94,463
96,785
101,777
Previous Estimate
36,773
46,802
85,996
94,191
93,253
93,253
NA
NA (Not Applicable)
Processing
Flares (Recalculation with Updated Data)
Processing segment flare Cm emission estimates are on average 0.5 percent lower across the 1993 to 2021 time
series than in the previous Inventory. The Cm estimate for 2021 is 7 percent lower than in the previous Inventory.
The processing segment flare CO2 emission estimates decreased by an average of 0.4 percent over the 1993 to
2021 time series, while the CO2 estimate for 2021 decreased by 6 percent compared to the previous Inventory.
These changes were due to GHGRP submission revisions.
Table 3-104: Processing Segment Flares National CH4 Emissions (Metric Tons CH4)
Source
1990
2005
2018
2019
2020
2021
2022
Flares
NO |
NA |
24,173
43,887
36,985
26,807
33,586
Previous Estimate
NO 1
s
24,148
43,613
36,928
28,784
NA
NA (Not Applicable)
NO (Not Occurring)
Table 3-105: Processing Segment Flares National CO2 Emissions (kt CO2)
Source
1990
2005
2018
2019
2020
2021
2022
Flares
NO
3,517 ¦
5,948
9,776
8,121
6,941
8,533
Previous Estimate
NO =
3,517 i
5,945
9,859
8,120
7,381
NA
NA (Not Applicable)
NO (Not Occurring)
Transmission and Storage
Transmission Compressor Station Leaks and Venting (Methodological Update)
EPA updated the methodology to estimate national level activity data for transmission compressor stations (i.e.,
station and compressor counts). EPA used annual data for 1996 to 2022 from FERC and PHMSA to estimate
national transmission station counts and total transmission compressor counts. FERC requires major interstate
transmission compression facilities to report annual data on station counts, total compressor counts, and total
transmission pipeline miles. EPA compiled annual FERC data and scaled it up to the national level using PHMSA
national transmission pipeline miles for 1996 to 2022. EPA retained existing Inventory activity data for 1990 to
1992 and used linear interpolation to estimate national station and total compressor counts for 1993 to 1995.
Total compressor counts were apportioned to reciprocating and centrifugal compressor types using data from
GHGRP's Subpart W. EPA retained existing Inventory activity data for 1990 to 1992 and used linear interpolation
3-114 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
for intermediate time series years. For more details on this update, refer to the Transmission Station Activity
memo.
This update impacts Cm and CO2 emissions from leaks (including compressor leaks), dehydrator vents, pneumatic
devices, flaring, and venting at transmission compression stations. As a result of this update, CH4 emissions
estimates increased by an average of 2.6 percent across the time series and decreased by 18 percent in 2021,
compared to the previous Inventory. Emissions estimates of CO2 increased by an average of 2.5 percent across the
time series and decreased by 16 percent in 2021, compared to the previous Inventory.
Table 3-106: Transmission Compressor Station National CH4 Emissions (Metric Tons CH4)
Source
1990
2005
2018
2019
2020
2021
2022
Station Total Emissions
1,096,527 1
704,651
581,117
585,617
580,273
585,666
588,497
Station + Compressor
Fugitive Emissions
NE I
NE 1
118,592
120,704
120,256
125,440
128,128
Reciprocating Compressor
NE I
NE
347,685
342,485
333,385
325,130
322,725
Centrifugal Compressor (wet
seals)
NE I
1
NE
NE
50,116
51,544
52,360
53,652
50,524
Centrifugal Compressor (dry
seals)
1
NE
64,724
70,884
74,272
81,444
87,120
Dehydrator vents
1,991
1,931
2,132
2,170
2,162
2,255
2,304
Flaring
305
296
606
432
522
364
492
Pneumatic Devices
213,081 »
87,701 ¦
31,749
31,640
30,120
30,585
31,170
High Bleed
NE
NE
9,781
9,797
9,119
8,951
8,350
Intermittent Bleed
ne :::
ne :
21,132
21,074
20,222
20,882
22,157
Low Bleed
NE
NE
836
769
779
753
663
Station Venting
145,241 ¦
138,843 ;;;;;;
127,165
136,920
136,332
143,132
146,659
Total Emissions
1,457,144
933,422
742,770
756,779
749,409
762,002
769,121
Previous Estimate
1,459,223
871,649
809,418
891,620
911,471
923,868
NA
NE (Not Estimated at individual source level due to lack of data)
NA (Not Applicable)
Table 3-107: Transmission Compressor Station National CO2 Emissions (Metric Tons CO2)
Source
1990
2005
2018
2019
2020
2021
2022
Station Total Emissions
32,285
20,747
17,110
17,242
17,085
17,244
17,327
Station + Compressor Fugitive
Emissions
NE |
¦is
NE 1
3,492
3,554
3,541
3,693
3,773
Reciprocating Compressor
NE
NE
10,237
10,084
9,816
9,573
9,502
Centrifugal Compressor (wet
seals)
iiiiiii
NE Si
NE
1,476
1,518
1,542
1,580
1,488
Centrifugal Compressor (dry
seals)
NE _
NE _
1,906
2,087
2,187
2,398
2,565
Dehydrator vents
59 ii:
63
64
64
66
68
Flaring
78,386
76,030
70,366
78,424
93,510
62,974
82,004
Pneumatic Devices
Transmission
-------
Underground Natural Gas Storage Well Events (Methodological Update)
EPA updated the Inventory with CO2 and Cm estimates resulting from underground natural gas storage well events
that occurred in several years across the inventory time series. Previously, EPA included emissions only from the
Aliso Canyon event (occurring 2015 to 2016). This update incorporates emissions from 9 events identified as
occurring at storage wells. For the update, emissions from individual events were added to the year in which they
occurred.
EPA calculated CH4 emissions using the reported leak size and applying inventory assumptions for CH4 content. The
Cm emissions estimates were then adjusted using a 60 percent combustion efficiency if there was evidence of
combustion or ignition during the event. EPA calculated CO2 emissions only for the events where combustion or
ignition occurred, using the 60 percent combustion efficiency. The Storage Well Events memo presents additional
information and considerations for this update.
One commenter provided feedback and it focused on the 60 percent combustion efficiency. The commenter
suggested using 30 percent as the combustion efficiency instead of 60 percent based on research from Gvakharia
et al., 2017, which lists 30 percent as the lowest flare efficiency.83 The commenter noted the study used to justify
the 60 percent combustion efficiency evaluated engineered flares which may not be representative of combustion
during emergency events.
The newly incorporated events occurred in years 1992,1998, 2001, 2002, 2003, 2004, 2006, 2010, and 2011. They
emitted on average 7,085 metric tons of CH4 and 27,056 metric tons CO2.
Methane emissions estimates for underground natural gas storage wells increased 15 percent across the time
series compared to previous estimates. The largest increase between the updates and the previous estimates
occurs in 2004, which incorporates CH4 emissions from the Moss Bluff event.
Updates to CO2 emissions resulted in an increase of 1,693 percent across the time series compared to previous
estimates. This increase is mostly due to additions for 2001 and 2004 estimates, resulting from Yaggy and Moss
Bluff events, respectively. Combustion occurred at both of these events, resulting in the application of the 60
percent combustion efficiency.
Pipeline Venting (Recalculation with Updated Data)
Transmission pipeline venting CH4 emission estimates are on average 0.1 percent lower across the 1990 to 2021
time series than in the previous Inventory. The CH4 emissions estimate for 2021 is 4 percent lower than in the
previous Inventory. These changes were due to GHGRP submission revisions.
Table 3-108: Pipeline Venting National CH4 Emissions (Metric Tons CH4)
Source
1990
2005
2018
2019
2020
2021
2022
Pipeline Venting
177,951 1
183,159 1
208,438
187,266
220,544
165,703
133,761
Previous Estimate
177 951 ¦
183,159 1
208,438
187,266
220,560
172,287
NA
NA (Not Applicable)
Distribution
There were no methodological updates to the distribution segment and recalculations resulted in an average
decrease in CH4 emissions across the 1990 to 2021 time series of less than 0.1 percent and an average decrease in
calculated CO2 emissions across the 1990 to 2021 time series of less than 0.1 percent, compared to the previous
Inventory.
83 Gvakharia et al, 2017. Methane, Black Carbon, and Ethane Emissions from Natural Gas Flares in the Bakken Shale, North
Dakota. Environmental Science & Technology. 51: 5317-5325. https://doi.org/10.1021/acs.est.6b05183.
3-116 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
Post-Meter
There were no methodological updates to estimate post-meter emissions and recalculations resulted in an average
decrease in Cm emissions across the 1990 to 2021 time series of less than 0.1 percent and an average decrease in
calculated CO2 emissions across the 1990 to 2021 time series of less than 0.1 percent, compared to the previous
Inventory.
Planned Improvements
Planned Improvements for 2025 Inventory
EPA updated the Enverus data and there were notable increases in the number of wells and completions identified
as being hydraulically fractured compared with previous versions of the database. EPA will assess the underlying
Enverus data to determine the cause of these changes.
Upcoming Data, and Additional Data that Could Inform the Inventory
EPA will assess new data received by EPA's Greenhouse Gas Reporting Program and Methane Challenge Program
on an ongoing basis, which may be used to validate or improve existing estimates and assumptions.
EPA continues to track studies that contain data that may be used to update the Inventory. EPA will also continue
to assess studies that include and compare both top-down and bottom-up emission estimates, which could lead to
improved understanding of unassigned high emitters (e.g., identification of emission sources and information on
frequency of high emitters) as recommended in previous stakeholder comments.
3.8 Abandoned Oil and Gas Wells (CRT
Source Categories lB2a and lB2b)
The term "abandoned wells", as used in the Inventory, encompasses various types of oil and gas wells, including
orphaned wells and other non-producing wells:
• Wells with no recent production, and not plugged. Common terms (such as those used in state databases)
might include: inactive, temporarily abandoned, shut-in, dormant, and idle.
• Wells with no recent production and no responsible operator. Common terms might include: orphaned,
deserted, long-term idle, and abandoned.
• Wells that have been plugged to prevent migration of gas or fluids.
The U.S. population of abandoned oil and gas wells (including orphaned wells and other non-producing wells) is
around 3.9 million (with around 3.0 million abandoned oil wells and 0.9 million abandoned gas wells). The methods
to calculate emissions from abandoned wells involve calculating the total populations of plugged and unplugged
abandoned oil and gas wells in the United States and the application of emission factors. An estimate of the
number of orphaned wells within this population is not developed as part of the methodology. Wells that are
plugged have much lower average emissions than wells that are unplugged (less than 1 kg Cm per well per year,
versus over 100 kg Cm per well per year). Around 43 percent of the abandoned well population in the United
States is plugged. This fraction has increased over the Inventory time series (from around 22 percent in 1990) as
more wells fall under regulations and programs requiring or promoting plugging of abandoned wells.
Abandoned oil wells. Abandoned oil wells emitted 235 kt Cm and 5 kt CO2 in 2022. Emissions of both gases
increased by 3 percent from 1990, while the total population of abandoned oil wells increased 40 percent.
Energy 3-117
-------
Abandoned gas wells. Abandoned gas wells emitted 68 kt Cm and 3 kt CO2 in 2022. Emissions of both gases
increased by 33 percent from 1990, while the total population of abandoned gas wells increased 83 percent.
Table 3-109: CH4 Emissions from Abandoned Oil and Gas Wells (MMT CO2 Eq.)
Activity 1990
2005
2018
2019
2020
2021
2022
Abandoned Oil Wells 6.4
6.6
6.6
6.6
6.6
6.6
6.6
Abandoned Gas Wells 1.4
1.6
1.8
1.8
1.9
1.9
1.9
Total 7.8
8.2
8.4
8.5
8.5
8.6
8.5
Note: Totals may not sum due to independent rounding.
Table 3-110: CH4 Emissions from Abandoned Oil and Gas Wells (kt)
Activity 1990
2005
2018
2019
2020
2021
2022
Abandoned Oil Wells 228
236
1 237
237
237
237
235
Abandoned Gas Wells 51
58
1 64
65
66
69
68
Total 279
294
301
302
303
306
303
Note: Totals may not sum due to independent rounding.
Table 3-111: CO2 Emissions from Abandoned Oil and Gas Wells (MMT CO2)
Activity 1990
2005
2018
2019
2020
2021
2022
Abandoned Oil Wells +
~
I
+
+
+
+
Abandoned Gas Wells +
1
+
1
+
+
+
+
Total +
~
~
+
+
+
+
+ Does not exceed 0.05 MMT C02 Eq.
Note: Totals may not sum due to independent rounding.
Table 3-112: CO2 Emissions from Abandoned Oil and Gas Wells (kt)
Activity
1990
2005
2018
2019
2020
2021
2022
Abandoned Oil Wells
5
5 II
5
5
5
5
5
Abandoned Gas Wells
^ I
3 i
3
3
3
3
3
Total
7
7
8
8
8
8
8
Note: Totals may not sum due to independent rounding.
Methodology and Time-Series Consistency
EPA uses a Tier 2 method from IPCC (2019) to quantify emissions from abandoned oil and gas wells. EPA's
approach is based on the number of plugged and unplugged abandoned wells in the Appalachian region and in the
rest of the U.S., and emission factors for plugged and unplugged abandoned wells in Appalachia and the rest of the
U.S. Methods for abandoned wells are unavailable in IPCC (2006). The details of this approach and of the data
sources used are described in the memorandum Inventory of U.S. Greenhouse Gas Emissions and Sinks 1990-2016:
Abandoned Wells in Natural Gas and Petroleum Systems (2018 Abandoned Wells Memo).
EPA developed abandoned well CH4 emission factors using data from Kang et al. (2016) and Townsend-Small et al.
(2016). Plugged and unplugged abandoned well CH4 emission factors were developed at the national-level (using
emission data from Townsend-Small et al.) and for the Appalachia region (using emission data from measurements
in Pennsylvania and Ohio conducted by Kang et al. and Townsend-Small et al., respectively). The Appalachia region
emissions factors were applied to abandoned wells in states in the Appalachian basin region, and the national-level
emission factors were applied to abandoned wells in all other states. EPA developed abandoned well CO2 emission
factors using the CH4 emission factors and an assumed ratio of CCh-to-Cm gas content, similar to the approach
used to calculate CO2 emissions for many sources in Petroleum Systems and Natural Gas Systems. For abandoned
3-118 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
oil wells, EPA used the Petroleum Systems default production segment associated gas ratio of 0.020 MT CO2/MT
Cm, which was derived through API TankCalc modeling runs. For abandoned gas wells, EPA used the Natural Gas
Systems default production segment CH4 and CO2 gas content values (GRI/EPA 1996, GTI 2001) to develop a ratio
of 0.044 MT CO2/MT CH4. The same respective emission factors are applied for each year of the time series.
EPA developed state-level annual counts of abandoned wells for 1990 through 2022 by summing together an
annual estimate of abandoned wells in the Enverus data set (Enverus 2023), and an estimate of total abandoned
wells not included the Enverus dataset (see 2018 Abandoned Wells Memo for additional information on how the
value was calculated) for each state. References reviewed to develop the number of abandoned wells not included
in the Enverus dataset include historical records collected by state agencies and by USGS.
The state-level abandoned well population was then split into plugged and unplugged wells by applying an
assumption that all abandoned wells were unplugged in 1950 and using Enverus data to calculate the fraction of
plugged abandoned wells in 2022. Linear interpolation was applied between the 1950 value and 2022 value to
calculate the plugged fraction for intermediate years. See the memorandum Inventory of U.S. Greenhouse Gas
Emissions and Sinks 1990-2016: Abandoned Wells in Natural Gas and Petroleum Systems (2018 Abandoned Wells
Memo) for details.84
Abandoned Oil Wells
Table 3-113: Abandoned Oil Wells Activity Data, CH4 and CO2 Emissions (kt)
Source
1990
2005
2018
2019
2020
2021
2022
Plugged abandoned oil wells
475,939
810,564
1,156,633
1,192,282
1,227,566
1,263,583
1,281,380
Unplugged abandoned oil wells
1,697,730
(
1
1,787,095 I
1,781,964
1,783,807
1,784,834
1,785,340
1,767,543
Total Abandoned Oil Wells
2,173,669
2,597,659
2,938,597
2,976,089
3,012,400
3,048,923
3,048,923
Abandoned oil wells in
!!S
Appalachia
22%
1
1
20%
19%
19%
18%
18%
18%
Abandoned oil wells outside of
Appalachia
78%
80%
81%
81%
82%
82%
82%
Cl-Mrom plugged abandoned
1
jjj
1
oil wells (kt)
0.17
0.25 III
0.34
0.35
0.36
0.36
0.37
CH4from unplugged abandoned
I
1
oil wells(kt)
227.6
236.1
236.5
236.9
237.0
236.8
235.0
Total CH.i from abandoned oil
wells (kt)
227.7
236.4
236.8
237.2
237.3
237.2
235.4
Total CO' from abandoned oil
wells (kt)
4.6
4.8
4.8
4.8
4.8
4.8
4.8
Abandoned Gas Wells
Table 3-114: Abandoned Gas Wells Activity Data, CH4 and CO2 Emissions (kt)
Source 1990 2005 2018 2019 2020 2021 2022
Plugged abandoned gas wells 110,0891 210,902 1 348,625 359,018 372,605 389,745 395,236
Unplugged abandoned gas
wells 355,6201 404,9601 447,374 448,504 453,988 463,119 457,628
Total Abandoned Gas Wells 465,709j 615,862 1 795,999 807,522 826,593 852,864 852,864
Abandoned gas wells in
Appalachia 28% I 25% I 24% 24% 24% 26% 26%
84 See https://www.epa.Eov/ghgemissions/natural-gas-and-petroleum-svstems.
Energy 3-119
-------
Abandoned gas wells outside
of Appalachia 72% 75% 76% 76% 76% 74% 74%
CH4from plugged abandoned
gas wells (kt) 0.06 j 0.11 i 0.17 0.17 0.19 0.21 0.21
CH4from unplugged
abandoned gas wells (kt) 511 575 641 645 65J) 685 67.8
Total CH.i from abandoned gas
wells (kt) 51.1 57.6 64.3 64.7 66.1 68.7 68.0
Total CO.' from abandoned gas
wells (kt) 2.2 2.5 2.8 2.8 2.9 3.0 3.0
Uncertainty
To characterize uncertainty surrounding estimates of abandoned well emissions, EPA conducted a quantitative
uncertainty analysis using the IPCC Approach 2 methodology (Monte Carlo simulation technique). See the 2018
Abandoned Wells Memo for details of the uncertainty analysis methods. EPA used Microsoft Excel's @RISKadd-in
tool to estimate the 95 percent confidence bound around total methane emissions from abandoned oil and gas
wells in year 2022, then applied the calculated bounds to both Cm and CO2 emissions estimates for each
population. The @RISK add-in provides for the specification of probability density functions (PDFs) for key variables
within a computational structure that mirrors the calculation of the inventory estimate. EPA used measurement
data from the Kang et al. (2016) and Townsend-Small et al. (2016) studies to characterize the CFU emission factor
PDFs. For activity data inputs (e.g., total count of abandoned wells, split between plugged and unplugged), EPA
assigned default uncertainty bounds of ± 10 percent based on expert judgment.
The IPCC guidance notes that in using this method, "some uncertainties that are not addressed by statistical means
may exist, including those arising from omissions or double counting, or other conceptual errors, or from
incomplete understanding of the processes that may lead to inaccuracies in estimates developed from models." As
a result, the understanding of the uncertainty of emission estimates for this category evolves and improves as the
underlying methodologies and datasets improve. The uncertainty bounds reported below only reflect those
uncertainties that EPA has been able to quantify and do not incorporate considerations such as modeling
uncertainty, data representativeness, measurement errors, misreporting or misclassification.
The results presented below in Table 3-115 provide the 95 percent confidence bound within which actual
emissions from abandoned oil and gas wells are likely to fall for the year 2022, using the recommended IPCC
methodology. Abandoned oil well CFU emissions in 2022 were estimated to be between 1.1 and 19.7 MMT CO2 Eq.,
while abandoned gas well CH4 emissions were estimated to be between 0.3 and 5.4 MMT CO2 Eq. at a 95 percent
confidence level. Uncertainty bounds for other years of the time series have not been calculated, but uncertainty is
expected to vary over the time series.
Table 3-115: Approach 2 Quantitative Uncertainty Estimates for CH4 and CO2 Emissions from
Petroleum and Natural Gas Systems (MMT CO2 Eq. and Percent)
Source
Gas
2022 Emission Estimate
(MMTCO. Eq.)1,
Uncertainty Range Relative to Emission Estimate-1
(MMTCO' Eq.) (%)
Lower Upper
Lower
Upper
Bound Bound
Bound
Bound
Abandoned Oil Wells
ch4
6.6
1.1 19.7
-83%
+204%
Abandoned Gas Wells
ch4
1.9
0.3 5.4
-83%
+204%
Abandoned Oil Wells
co2
0.005
0.001 0.014
-83%
+204%
Abandoned Gas Wells
co2
0.003
0.0005 0.008
-83%
+204%
a Range of emission estimates estimated by applying the 95 percent confidence intervals obtained from the Monte Carlo
simulation analysis conducted for total abandoned oil and gas well CH4 emissions in year 2022.
3-120 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
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 the table.
QA/QC and Verification Discussion
The emission estimates in the Inventory are continually reviewed and assessed to determine whether emission
factors and activity factors accurately reflect current industry practices. In order to ensure the quality of emission
estimates for abandoned wells, general (IPCC Tier 1) quality assurance/quality control (QA/QC) procedures were
implemented consistent with the U.S. Inventory QA/QC plan outlined in Annex 8. Additionally, EPA reviewed the
current Enverus dataset and compared it with results from the previous dataset to identify outliers and instances
of significant changes to abandoned oil and gas well counts.
EPA performs a thorough review of information associated with new studies to assess whether the assumptions in
the Inventory are consistent with industry practices and whether new data is available that could be considered for
updates to the estimates. As in previous years, EPA conducted early engagement and communication with
stakeholders on updates prior to public review. EPA held a stakeholder webinar on greenhouse gas data for oil and
gas in October of 2023.
Recalculations Discussion
EPA updated the Inventory with revised abandoned oil and gas well counts developed from Enverus data (Enverus
2023). Compared to the previous Inventory, annual abandoned oil well counts increased by an average of 2
percent across the time series and increased by 4 percent in 2021. Annual abandoned gas well counts increased by
2 percent across the time series and 7 percent in 2021. Similarly, both plugged wells and unplugged wells
increased by 2 percent across the time series. As a result of this update, calculated abandoned oil well Cm and CO2
emissions increased by an average of 2 percent each year across the time series and increased by 3 percent in
2021, compared to the previous Inventory. Abandoned gas well Cm and CO2 emissions increased by an average of
1 percent across the time series and increased by 8 percent in 2021, compared to the previous Inventory.
Planned Improvements
EPA will continue to assess new data and stakeholder feedback on considerations (such as potential use of
emission factor data from regions not included in the measurement studies on which current emission factors are
based) to improve the abandoned well count estimates and emission factors. In future Inventories, EPA will assess
data that become available from Department of Interior and Department of Energy orphan well plugging
programs.
3.9 International Bunker Fuels (CRT Source
Category 1: Memo Items)
Emissions resulting from the combustion of fuels used for international transport activities, termed international
bunker fuels under the Paris Agreement and the UNFCCC, are not included in national emission totals, but are
reported separately based upon location of fuel sales. The decision to report emissions from international bunker
fuels separately, instead of allocating them to a particular country, was made by the Intergovernmental
Energy 3-121
-------
Negotiating Committee in establishing the Framework Convention on Climate Change.85 These decisions are
reflected in the IPCC methodological guidance, including IPCC (2006), in which countries are requested to report
emissions from ships or aircraft that depart from their ports with fuel purchased within national boundaries and
are engaged in international transport separately from national totals (IPCC 2006).86
Two transport modes are addressed under the IPCC definition of international bunker fuels: aviation and marine.87
Greenhouse gases emitted from the combustion of international bunker fuels, like other fossil fuels, include CO2,
Cm and N2O for marine transport modes, and CO2 and N2O for aviation transport modes. Emissions from ground
transport activities—by road vehicles and trains—even when crossing international borders are allocated to the
country where the fuel was loaded into the vehicle and, therefore, are not counted as bunker fuel emissions.
The 2006 IPCC Guidelines distinguish between three different modes of air traffic: civil aviation, military aviation,
and general aviation. Civil aviation comprises aircraft used for the commercial transport of passengers and freight,
military aviation comprises aircraft under the control of national armed forces, and general aviation applies to
recreational and small corporate aircraft. The 2006 IPCC Guidelines further define international bunker fuel use
from civil aviation as the fuel combusted for civil (e.g., commercial) aviation purposes by aircraft arriving or
departing on international flight segments. However, as mentioned above, and in keeping with the 2006 IPCC
Guidelines, only the fuel purchased in the United States and used by aircraft taking-off (i.e., departing) from the
United States are reported here. The standard fuel used for civil and military aviation is kerosene-type jet fuel,
while the typical fuel used for general aviation is aviation gasoline.88
Emissions of CChfrom aircraft are essentially a function of fuel consumption. Nitrous oxide emissions also depend
upon engine characteristics, flight conditions, and flight phase (i.e., take-off, climb, cruise, decent, and landing).
Recent data suggest that little or no CH4 is emitted by modern engines (Anderson et al. 2011), and as a result, CH4
emissions from this category are reported as zero. In jet engines, N2O is primarily produced by the oxidation of
atmospheric nitrogen, and the majority of emissions occur during the cruise phase.
International marine bunkers comprise emissions from fuels burned by ocean-going ships of all flags that are
engaged in international transport. Ocean-going ships are generally classified as cargo and passenger carrying,
military (i.e., U.S. Navy), fishing, and miscellaneous support ships (e.g., tugboats). For the purpose of estimating
greenhouse gas emissions, international bunker fuels are solely related to cargo and passenger carrying vessels,
which is the largest of the four categories, and military vessels. Two main types of fuels are used on sea-going
vessels: distillate diesel fuel and residual fuel oil. Carbon dioxide is the primary greenhouse gas emitted from
marine shipping.
Overall, aggregate greenhouse gas emissions in 2022 from the combustion of international bunker fuels from both
aviation and marine activities were 99.1 MMT CO2 Eq., or 5.2 percent below emissions in 1990 (see Table 3-116
and Table 3-117). Emissions from international flights and international shipping voyages departing from the
United States have increased by 74.4 percent and decreased by 51.7 percent, respectively, since 1990. The
majority of these emissions were in the form of CO2; however, small amounts of CH4 (from marine transport
modes) and N2O were also emitted.
85 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. lc).
86 Note that the definition of international bunker fuels used by the UNFCCC differs from that used by the International Civil
Aviation Organization.
87 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).
88 Naphtha-type jet fuel was used in the past by the military in turbojet and turboprop aircraft engines.
3-122 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
Table 3-116: CO2, CH4, and N2O Emissions from International Bunker Fuels (MMT CO2 Eq.)
Gas/Mode
1990
2005
2018
2019
2020
2021
2022
C02
103.6
113.3
124.3
113.6
69.6
80.2
98.2
Aviation
38.2 ¦
60.2
83.0
78.3
39.8
50.8
66.6
Commercial
30.0
55.6
79.8
75.1
36.7
47.6
63.5
Military
8.2
46
3.2
3.2
3.1
3.2
3.1
Marine
65.4
53.1
41.3
35.4
29.9
29.4
31.6
ch4
0.2
0.1
0.1
0.1
0.1
0.1
0.1
Aviation
NO
NO
NO
NO
NO
NO
NO
Marine
0.2 d
0.1:::!
0.1
0.1
0.1
0.1
0.1
n2o
0.8
0.9
1.0
0.9
0.5
0.6
0.8
Aviation
ro
O
O
Ln
¦nil
0.7
0.7
0.3
0.4
0.6
Marine
0.4
0.4
0.3
0.2
0.2
0.2
0.2
Total
104.6
114.3
125.3
114.6
70.3
80.9
99.1
NO (Not Occurring)
Notes: Totals may not sum due to independent rounding. Includes aircraft cruise altitude emissions.
Table 3-117: CO2, CH4, and N2O Emissions from International Bunker Fuels (kt)
Gas/Mode
1990
2005
2018
2019
2020
2021
2022
C02
103,634
113,328
124,279
113,632
69,638
80,180
98,241
Aviation
38,205 1
60,221
82,953
78,280
39,781
50,812
66,646
Marine
65,429
53,107
41,325
35,351
29,857
29,369
31,595
ch4
7
5
4
4
3
3
3
Aviation
NO
NO
NO
NO
NO
NO
NO
Marine
7
4
4
3
3
3
NzO
3
3
4
3
2
2
3
Aviation
2 if
3
2
1
2
2
Marine
2
1
1
1
1
1
1
NO (Not Occurring)
Notes: Totals may not sum due to independent rounding. Includes aircraft cruise altitude emissions.
Methodology and Time-Series Consistency
Emissions of CO2 were for the most part estimated by applying carbon content and fraction oxidized factors to fuel
consumption activity data. This approach is analogous to that described under Section 3.1. Carbon content and
fraction oxidized factors for jet fuel (except for commercial aviation as per below), distillate fuel oil, and residual
fuel oil are the same as used for CO2 from Fossil Fuel Combustion and are presented in Annex 2.1, Annex 2.2, and
Annex 3.8 of this Inventory. Density conversions were taken from ASTM (1989) and USAF (1998). Heat content for
distillate fuel oil and residual fuel oil were taken from EIA (2024) and USAF (1998), and heat content for jet fuel
was taken from EIA (2024). See below for details on how emission estimates for commercial aviation were
determined.
A complete description of the methodology and a listing of the various factors employed can be found in Annex
2.1. See Annex 3.8 for a specific discussion on the methodology used for estimating emissions from international
bunker fuel use by the U.S. military.
Emission estimates for CH4 and N2O were calculated by multiplying emission factors by measures of fuel
consumption by fuel type and mode. Emission factors used in the calculations of CH4 and N2O emissions were
obtained from the Revised 1996IPCC Guidelines (IPCC/UNEP/OECD/IEA 1997), which is also referenced in the 2006
IPCC Guidelines (IPCC 2006). For aircraft emissions, the following value, in units of grams of pollutant per kilogram
of fuel consumed (g/kg), was employed: 0.1 for N2O (IPCC 2006). For marine vessels consuming either distillate
diesel or residual fuel oil the following values (g/MJ), were employed: 0.315 for CH4 and 0.08 for N2O. Activity data
for aviation included solely jet fuel consumption statistics, while the marine mode included both distillate diesel
and residual fuel oil.
Energy 3-123
-------
Activity data on domestic and international aircraft fuel consumption were developed by the U.S. Federal Aviation
Administration (FAA) using radar-informed data from the FAA Enhanced Traffic Management System (ETMS) for
1990 and 2000 through 2022 as modeled with the Aviation Environmental Design Tool (AEDT). This bottom-up
approach is built from modeling dynamic aircraft performance for each flight occurring within an individual
calendar year. The analysis incorporates data on the aircraft type, date, flight identifier, departure time, arrival
time, departure airport, arrival airport, ground delay at each airport, and real-world flight trajectories. To generate
results for a given flight within AEDT, the radar-informed aircraft data is correlated with engine and aircraft
performance data to calculate fuel burn and exhaust emissions. Information on exhaust emissions for in-
production aircraft engines comes from the International Civil Aviation Organization (ICAO) Aircraft Engine
Emissions Databank (EDB). This bottom-up approach is in accordance with the Tier 3B method from the 2006IPCC
Guidelines (IPCC 2006).
International aviation CO2 estimates for 1990 and 2000 through 2022 were obtained directly from FAA's AEDT
model (FAA 2024). The radar-informed method that was used to estimate CO2 emissions for commercial aircraft
for 1990 and 2000 through 2022 was not possible for 1991 through 1999 because the radar dataset was not
available for years prior to 2000. FAA developed Official Airline Guide (OAG) schedule-informed inventories
modeled with AEDT and great circle trajectories for 1990, 2000, and 2010. Because fuel consumption and CO2
emission estimates for years 1991 through 1999 are unavailable, consumption estimates for these years were
calculated using fuel consumption estimates from the Bureau of Transportation Statistics (DOT 1991 through
2013), adjusted based on 2000 through 2005 data. See Annex 3.3 for more information on the methodology for
estimating emissions from commercial aircraft jet fuel consumption.
Data on U.S. Department of Defense (DoD) aviation bunker fuels and total jet fuel consumed by the U.S. military
was supplied by the Office of the Under Secretary of Defense (Installations and Environment), DoD. Estimates of
the percentage of each Service's total operations that were international operations were developed by DoD.
Military aviation bunkers included international operations, operations conducted from naval vessels at sea, and
operations conducted from U.S. installations principally over international water in direct support of military
operations at sea. Military aviation bunker fuel emissions were estimated using military fuel and operations data
synthesized from unpublished data from DoD's Defense Logistics Agency Energy (DLA Energy 2023). Together, the
data allow the quantity of fuel used in military international operations to be estimated. Densities for each jet fuel
type were obtained from a report from the U.S. Air Force (USAF 1998). Final jet fuel consumption estimates are
presented in Table 3-118. See Annex 3.8 for additional discussion of military data.
Table 3-118: Aviation Jet Fuel Consumption for International Transport (TBtu)
Nationality
1990
2005
2018
2019
2020
2021
2022
U.S. and Foreign Carriers
426
7911
1,104
1,068
521
677
902
U.S. Military
116
64
44
44
43
44
44
Total
542
854
1,148
1,112
564
721
946
Note: Totals may not sum due to independent rounding.
In order to quantify the civilian international component of marine bunker fuels, activity data on distillate diesel
and residual fuel oil consumption by cargo or passenger carrying marine vessels departing from U.S. ports were
collected for individual shipping agents on a monthly basis by the U.S. Customs and Border Protection. This
information was then reported in unpublished data collected by the Foreign Trade Division of the U.S. Department
of Commerce's Bureau of the Census (DOC 1991 through 2022) for 1990 through 2001, 2007 through 2022, and
the Department of Homeland Security's Bunker Report for 2003 through 2006 (DHS 2008). Fuel consumption data
for 2002 was interpolated due to inconsistencies in reported fuel consumption data. Activity data on distillate
diesel consumption by military vessels departing from U.S. ports were provided by DLA Energy (2023). The total
amount of fuel provided to naval vessels was reduced by 21 percent to account for fuel used while the vessels
were not underway (i.e., in port). Data on the percentage of steaming hours underway versus not underway were
provided by the U.S. Navy. These fuel consumption estimates are presented in Table 3-119.
3-124 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
Table 3-119: Marine Fuel Consumption for International Transport (Million Gallons)
Fuel Type
1990
2005
2018
2019
2020
2021
2022
Residual Fuel Oil
4,781
3,8811
2,790
2,246
1,964
1,953
2,172
Distillate Diesel Fuel & Other
617
444 1
684
702
461
437
435
U.S. Military Naval Fuels
522
4711
285
281
296
285
263
Total
5,920
4,796
3,759
3,229
2,721
2,674
2,870
Note: Totals may not sum due to independent rounding.
Methodological recalculations were applied to the entire time series to ensure time-series consistency from 1990
through 2022.
Uncertainty
Emission estimates related to the consumption of international bunker fuels are subject to the same uncertainties
as those from domestic aviation and marine mobile combustion emissions; however, additional uncertainties
result from the difficulty in collecting accurate fuel consumption activity data for international transport activities
separate from domestic transport activities.89 For example, smaller aircraft on shorter routes often carry sufficient
fuel to complete several flight segments without refueling in order to minimize time spent at the airport gate or
take advantage of lower fuel prices at particular airports. This practice, called tankering, when done on
international flights, complicates the use of fuel sales data for estimating bunker fuel emissions. Tankering is less
common with the type of large, long-range aircraft that make many international flights from the United States,
however. Similar practices occur in the marine shipping industry where fuel costs represent a significant portion of
overall operating costs and fuel prices vary from port to port, leading to some tankering from ports with low fuel
costs.
Uncertainties exist with regard to the total fuel used by military aircraft and ships. Total aircraft and ship fuel use
estimates were developed from DoD records, which document fuel sold to the DoD Components (e.g., Army,
Department of Navy and Air Force) from the Defense Logistics Agency Energy. These data may not include fuel
used in aircraft and ships as a result of a Service procuring fuel from, selling fuel to, trading fuel with, or giving fuel
to other ships, aircraft, governments, or other entities.
Additionally, there are uncertainties in historical aircraft operations and training activity data. Estimates for the
quantity of fuel actually used in Navy and Air Force flying activities reported as bunker fuel emissions had to be
estimated based on a combination of available data and expert judgment. Estimates of marine bunker fuel
emissions were based on Navy vessel steaming hour data, which reports fuel used while underway and fuel used
while not underway. This approach does not capture some voyages that would be classified as domestic for a
commercial vessel. Conversely, emissions from fuel used while not underway preceding an international voyage
are reported as domestic rather than international as would be done for a commercial vessel. There is uncertainty
associated with ground fuel estimates for 1997 through 2022, including estimates for the quantity of jet fuel
allocated to ground transportation. Small fuel quantities may have been used in vehicles or equipment other than
that which was assumed for each fuel type.
There are also uncertainties in fuel end-uses by fuel type, emissions factors, fuel densities, diesel fuel sulfur
content, aircraft and vessel engine characteristics and fuel efficiencies, and the methodology used to back-
calculate the data set to 1990 using the original set from 1995. The data were adjusted for trends in fuel use based
on a closely correlating, but not matching, data set. All assumptions used to develop the estimate were based on
process knowledge, DoD data, and expert judgments. The magnitude of the potential errors related to the various
uncertainties has not been calculated but is believed to be small. The uncertainties associated with future military
bunker fuel emission estimates could be reduced through revalidation of assumptions based on data regarding
current equipment and operational tempo, however, it is doubtful data with more fidelity exist at this time.
89 See uncertainty discussions under section 3.1 C02 from Fossil Fuel Combustion.
Energy 3-125
-------
Although aggregate fuel consumption data have been used to estimate emissions from aviation, the recommended
method for estimating emissions of gases other than CO2 in the 2006IPCC Guidelines (IPCC 2006) is to use data by
specific aircraft type, number of individual flights and, ideally, movement data to better differentiate between
domestic and international aviation and to facilitate estimating the effects of changes in technologies. The IPCC
also recommends that cruise altitude emissions be estimated separately using fuel consumption data, while
landing and take-off (LTO) cycle data be used to estimate near-ground level emissions of gases other than CO2.90
There is also concern regarding the reliability of the existing DOC (1991 through 2022) data on marine vessel fuel
consumption reported at U.S. customs stations due to the significant degree of inter-annual variation.
/erification
In order to ensure the quality of the emission estimates from international bunker fuels, General (IPCC Tier 1) and
category-specific (Tier 2) Quality Assurance/Quality Control (QA/QC) procedures were implemented consistent
with the U.S. Inventory QA/QC plan outlined in Annex 8 The Tier 2 procedures that were implemented involved
checks specifically focusing on the activity data and emission factor sources and methodology used for estimating
CO2, CH4, and N2O emissions from international bunker fuels in the United States. Emission totals for the different
sectors and fuels were compared and trends were investigated. No corrective actions were necessary.
Recalculations Discussion
No recalculations were performed for the current Inventory.
Planned Improvements
EPA will evaluate data availability to update the sources for densities, energy contents, and emission factors
applied to estimate emissions from aviation and marine fuels. Many are from sources from the late 1990s, such as
IPCC/UNEP/OECD/IEA (1997). Potential sources with more recent data include the International Maritime
Organization (IMO) greenhouse gas emission inventory, International Air Transport Association (IATA)/ICAO
greenhouse gas reporting system (CORSIA), and the EPA Greenhouse Gas Reporting Program (GHGRP) Technical
Support Document for Petroleum Products. Specifically, EPA will evaluate data availability to support updating the
heat contents and carbon contents of jet fuel with input from EIA.
A longer-term effort is underway to consider the feasibility of including data from a broader range of domestic and
international sources for bunker fuels. Potential sources include the IMO greenhouse gas emission inventory, data
from the U.S. Coast Guard on vehicle operation currently used in criteria pollutant modeling, data from the
International Energy Agency (IEA), relevant updated FAA models to improve aviation bunker fuel estimates, and
researching newly available marine bunker data.
90 U.S. aviation emission estimates for CO, NOx, and NMVOCs are reported by EPA's National Emission Inventory (NEI) Air
Pollutant Emission Trends website, 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.
3-126 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
3.10 Biomass and Biofuels Consumption
(CRT Source Category 1A)
The combustion of biomass—such as wood, charcoal, the biogenic portions of MSW, and wood waste and biofuels
such as ethanol, biogas, and biodiesel—generates CO2 in addition to CH4 and N2O already covered in this chapter.
In line with the reporting requirements for inventories submitted under the Paris Agreement and the UNFCCC, CO2
emissions from biomass and biofuel combustion have been estimated separately from fossil fuel CO2 emissions
and are not directly included in the energy sector contributions to U.S. totals. In accordance with IPCC
methodological guidelines, any such emissions are calculated by accounting for net carbon fluxes from changes in
biogenic carbon reservoirs in wooded or crop lands. For a more complete description of this methodological
approach, see the Land Use, Land-Use Change, and Forestry chapter (Chapter 6), which accounts for the
contribution of any resulting CO2 emissions to U.S. totals within the Land Use, Land-Use Change, and Forestry
sector's approach.
Therefore, CO2 emissions from biomass and biofuel consumption are not included specifically in summing energy
sector totals. However, they are presented here for informational purposes and to provide detail on biomass and
biofuels consumption.
In 2022, total CO2 emissions from the burning of woody biomass in the industrial, residential, commercial, and
electric power sectors were approximately 195.3 MMT CO2 Eq. (195,338 kt) (see Table 3-120 and Table 3-121). As
the largest consumer of woody biomass, the industrial sector was responsible for 62.9 percent of the CO2
emissions from this source. The residential sector was the second largest emitter, constituting 22.3 percent of the
total, while the electric power and commercial sectors accounted for the remainder.
Table 3-120: CO2 Emissions from Wood Consumption by End-Use Sector (MMT CO2 Eq.)
End-Use Sector
1990
2005
2018
2019
2020
2021
2022
Industrial
135.3 1
136.3
134.4
132.1
127.3
128.2
122.8
Residential
59.8
44.3
54.1
56.3
35.6
35.5
43.6
Commercial
6.8
7.2
8.7
8.7
8.6
8.5
8.6
Electric Power
13.3 i
19.1
22.8
20.7
19.1
20.3
20.4
Total
215.2
206.9
220.0
217.7
190.6
192.5
195.3
Note: Totals may not sum due to independent rounding.
Table 3-121: CO2 Emissions from Wood Consumption by End-Use Sector (kt)
End-Use Sector
1990
2005
2018
2019
2020
2021
2022
Industrial
135,348 1
136,269
134,417
132,069
127,301
128,209
122,824
Residential
59,808
44,340
54,122
56,251
35,585
35,484
43,565
Commercial
6,779
7,218
8,669
8,693
8,554
8,528
8,563
Electric Power
13,252
19,074
22,795
20,677
19,115
20,288
20,385
Total
215,186
206,901
220,003
217,690
190,554
192,509
195,338
Note: Totals may not sum due to independent rounding.
Carbon dioxide emissions from combustion of the biogenic components of MSW by the electric power sector were
an estimated 14.9 MMT CO2 (14,864 kt) in 2022. Emissions across the time series are shown in Table 3-122 and
Table 3-123. As discussed in Section 3.3, MSW is combusted to produce electricity and the CO2 emissions from the
fossil portion of the MSW (e.g., plastics, textiles, etc.) are included in the energy sector FFC estimates. The MSW
also includes biogenic components (e.g., food waste, yard trimmings, natural fibers) and the CO2 emissions
associated with that biogenic portion is included here.
Energy 3-127
-------
Table 3-122: CO2 Emissions from Biogenic Components of MSW (MMT CO2 Eq.)
End-Use Sector 1990 2005 2018 2019 2020 2021
2022
Electric Power 18.5 14.7 16.1 15.7 15.6 15.3
14.9
Table 3-123: CO2 Emissions from Biogenic Components of MSW (kt)
End-Use Sector 1990 2005 2018 2019 2020 2021
2022
Electric Power 18,534 14,722 16,115 15,709 15,614 15,329
14,864
The transportation sector is responsible for most of the fuel ethanol consumption in the United States. Ethanol
used for fuel is currently produced primarily from corn grown in the Midwest, but it can be produced from a
variety of biomass feedstocks. Most ethanol for transportation use is blended with gasoline to create a 90 percent
gasoline, 10 percent by volume ethanol blend known as E-10 or gasohol.
In 2022, the United States transportation sector consumed an estimated 1,094.9 trillion Btu of ethanol (94 percent
of total), and as a result, produced approximately 75.0 MMT CO2 Eq. (74,953 kt) (see Table 3-124 and Table 3-125)
of CO2 emissions. Smaller quantities of ethanol were also used in the industrial and commercial sectors. Ethanol
fuel production and consumption has grown significantly since 1990 due to the favorable economics of blending
ethanol into gasoline and federal policies that have encouraged use of renewable fuels.
Table 3-124: CO2 Emissions from Ethanol Consumption (MMT CO2 Eq.)
End-Use Sector
1990
2005
2018
2019
2020
2021
2022
Transportation3
4.11
21.6
78.6
78.7
68.1
75.4
75.0
Industrial
0.1
1.2
1.4
1.6
1.6
1.5
1.9
Commercial
O.l!
0.2
1.9
2.2
2.2
2.1
2.7
Total
4.2
22.9
81.9
82.6
71.8
79.1
79.6
a See Annex 3.2, Table A-74 for additional information on transportation consumption of these fuels.
Note: Totals may not sum due to independent rounding.
Table 3-125: CO2 Emissions from Ethanol Consumption (kt)
End-Use Sector
1990
2005
2018
2019
2020
2021
2022
Transportation3
4,059 _
21,616
78,603
78,739
68,085
75,417
74,953
Industrial
105 1
1,176
1,404
1,610
1,582
1,509
1,919
Commercial
63 1
1511
1,910
2,229
2,182
2,139
2,721
Total
4,227
22,943
81,917
82,578
71,848
79,064
79,593
a See Annex 3.2, Table A-74 for additional information on transportation consumption of these fuels.
Note: Totals may not sum due to independent rounding.
The transportation sector is assumed to be responsible for all of the biodiesel consumption in the United States
(EIA 2024a). Biodiesel is currently produced primarily from soybean oil, but it can be produced from a variety of
biomass feedstocks including waste oils, fats, and greases. Biodiesel for transportation use appears in low-level
blends (less than 5 percent) with diesel fuel, high-level blends (between 6 and 20 percent) with diesel fuel, and 100
percent biodiesel (EIA 2024b).
In 2022, the United States consumed an estimated 211.6 trillion Btu of biodiesel, and as a result, produced
approximately 15.6 MMT CO2 Eq. (15,622 kt) (see Table 3-126 and Table 3-127) of CO2 emissions. Biodiesel
production and consumption has grown significantly since 2001 due to the favorable economics of blending
biodiesel into diesel and federal policies that have encouraged use of renewable fuels (EIA 2024b). There was no
measured biodiesel consumption prior to 2001 EIA (2024a).
3-128 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
Table 3-126: CO2 Emissions from Biodiesel Consumption (MMT CO2 Eq.)
End-Use Sector
1990
2005
2018
2019
2020
2021
2022
Transportation
NO
0.9
17.9
17.1
17.7
16.1
15.6
NO (Not Occurring)
a See Annex 3.2, Table A-74 for additional information on transportation consumption of these fuels.
Table 3-127: CO2 Emissions from Biodiesel Consumption (kt)
End-Use Sector
1990
2005 2018
2019
2020
2021
2022
Transportation3
NO |
856 i 17,936
17,080
17,678
16,112
15,622
NO (Not Occurring)
a See Annex 3.2, Table A-74 for additional information on transportation consumption of these fuels.
Methodology and Time-Series Consistency
Woody biomass emissions were estimated by applying two gross heat contents from EIA (Lindstrom 2006) to U.S.
consumption data (EIA 2024a) (see Table 3-129), provided in energy units for the industrial, residential,
commercial, and electric power sectors. One heat content (16.95 MMBtu/MT wood and wood waste) was applied
to the industrial sector's consumption, while the other heat content (15.43 MMBtu/MT wood and wood waste)
was applied to the consumption data for the other sectors. An EIA emission factor of 0.434 MT C/MT wood
(Lindstrom 2006) was then applied to the resulting quantities of woody biomass to obtain CO2 emission estimates.
The woody biomass is assumed to contain black liquor and other wood wastes, have a moisture content of 12
percent, and undergo complete combustion to be converted into CO2.
Data for total waste incinerated, excluding tires, from 1990 to 2022 was derived following the methodology
described in Section 3.3. Biogenic CO2 emissions associated with MSW combustion were obtained from EPA's
GHGRP FLIGHT data for MSW combustion sources (EPA 2023). Dividing biogenic CO2 emissions from GHGRP
FLIGHT data for MSW combustors by estimated MSW tonnage combusted yielded an annual biogenic CO2 emission
factor. This approach follows the same approach used to develop the fossil CO2 emissions from MSW combustion
as discussed in Section 3.3. As this data was only available following 2011, all years prior use an average of the
emission factors from 2011 through 2015.
Biogenic CO2 emissions from MSW combustion were calculated by multiplying the annual tonnage estimates,
excluding tires, by the calculated emissions factor. Calculated biogenic CO2 emission factors are shown in Table
3-128.
Table 3-128: Calculated Biogenic CO2 Content per Ton Waste (kg CCh/Short Ton Combusted)
End-Use Sector
1990
2005
2018
2019
2020
2021
2022
C02 Emission Factors
556
556 I
553
558
566
550
564
The amount of ethanol allocated across the transportation, industrial, and commercial sectors was based on the
sector allocations of ethanol-blended motor gasoline. The sector allocations of ethanol-blended motor gasoline
were determined using a bottom-up analysis conducted by EPA, as described in the Methodology section of Fossil
Fuel Combustion. Total U.S. ethanol consumption from EIA (2024a) was allocated to individual sectors using the
same sector allocations as ethanol-blended motor gasoline. The emissions from ethanol consumption were
calculated by applying an emission factor of 18.67 MMT C/Qbtu (EPA 2010) to adjusted ethanol consumption
estimates (see Table 3-130). The emissions from biodiesel consumption were calculated by applying an emission
Energy 3-129
-------
factor of 20.1 MMT C/Qbtu (EPA 2010) to U.S. biodiesel consumption estimates that were provided in energy units
(EIA 2024a) (see Table 3-131).91
Table 3-129: Woody Biomass Consumption by Sector (Trillion Btu)
End-Use Sector
1990
2005
2018
2019
2020
2021
2022
Industrial
1,441.9 1
1,451.7
1,432.0
1,407.0
1,356.2
1,365.9
1,308.5
Residential
580.0 S
430.0
524.9
545.5
345.1
344.1
422.5
Commercial
65.7
70.0
84.1
84.3
83.0
82.7
83.0
Electric Power
128.5
185.0 ¦
221.1
200.5
185.4
196.7
197.7
Total
2,216.2
2,136.7
2,262.0
2,237.3
1,969.6
1,989.4
2,011.7
Note: Totals may not sum due to independent rounding.
Table 3-130: Ethanol Consumption by Sector (Trillion Btu)
End-Use Sector
1990
2005
2018
2019
2020
2021
2022
Transportation
59.3
315.8 1
1,148.2
1,150.2
994.6
1,101.7
1,094.9
Industrial
1.5
17.2
20.5
23.5
23.1
22.0
28.0
Commercial
0.9
2.2
27.9
32.6
31.9
31.2
39.7
Total
61.7
335.1
1,196.6
1,206.3
1,049.5
1,155.0
1,162.7
Note: Totals may not sum due to independent rounding.
Table 3-131: Biodiesel Consumption by Sector (Trillion Btu)
End-Use Sector
1990
2005
2018
2019
2020
2021
2022
Transportation
NO
11.6
242.9
231.3
239.4
218.2
211.6
NO (Not Occurring)
Methodological recalculations were applied to the entire time series to ensure time-series consistency from 1990
through 2022.
Uncertainty
It is assumed that the combustion efficiency for biomass is 100 percent, which is believed to be an overestimate of
the efficiency of biomass combustion processes in the United States. Decreasing the combustion efficiency would
decrease emission estimates for CO2. Additionally, the heat content applied to the consumption of woody biomass
in the residential, commercial, and electric power sectors is unlikely to be a completely accurate representation of
the heat content for all the different types of woody biomass consumed within these sectors. Emission estimates
from ethanol and biodiesel production are more certain than estimates from woody biomass consumption due to
better activity data collection methods and uniform combustion techniques.
Recalculations Discussion
EIA (2024a) updated 2020 and 2021 wood energy consumed by the residential sector due to new underlying data
collected by the Residential Energy Consumption Survey (RECS), which collects data about once every 5 years and
uses Annual Energy Outlook growth rates to estimate data for other years. This caused CO2 emissions from
residential wood consumption to decrease by 9.9 MMT CO2 Eq. (4.9 percent) in 2020 and 12.3 MMT CO2 Eq. (6.0
percent) in 2021 compared to estimates in the previous Inventory for these years.
91 C02 emissions from biodiesel do not include emissions associated with the carbon in the fuel that is from the methanol used
in the process. Emissions from methanol use and combustion are assumed to be accounted for under Non-Energy Use of Fuels.
See Annex 2.3 - Methodology for Estimating Carbon Emitted from Non-Energy Uses of Fossil Fuels.
3-130 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
Planned Improvements
Future research will investigate the availability of data on woody biomass heat contents and carbon emission
factors to see if there are newer, improved data sources available for these factors.
Currently, emission estimates from biomass and biomass-based fuels included in this Inventory are limited to
woody biomass, biogenic components of MSW, ethanol, and biodiesel. Additional forms of biomass-based fuel
consumption include biogas, and other renewable diesel fuels. EPA will investigate additional forms of biomass-
based fuel consumption, research the availability of relevant emissions factors, and integrate these into the
Inventory as feasible. EPA will examine EIA data on biogas and other renewable diesel fuels to see if these fuel
types can be included in future Inventories. EIA (2024a) natural gas data already deducts biogas used in the natural
gas supply, so no adjustments are needed to the natural gas fuel consumption data to account for biogas. Distillate
fuel statistics are adjusted in this Inventory to remove other renewable diesel fuels as well as biodiesel.
Additionally, options for including "Other Renewable Fuels," as defined by EIA, will be evaluated.
The availability of facility-level combustion emissions through EPA's GHGRP will be examined to help better
characterize the industrial sector's energy consumption in the United States and further classify woody biomass
consumption by business establishments according to industrial economic activity type. Most methodologies used
in EPA's GHGRP are consistent with IPCC, although for EPA's GHGRP, facilities collect detailed information specific
to their operations according to detailed measurement standards, which may differ with the more aggregated data
collected for the Inventory to estimate total, national U.S. emissions. In addition, and unlike the reporting
requirements for this chapter under Paris Agreement and UNFCCC reporting guidelines, some facility-level fuel
combustion emissions reported under EPA's GHGRP may also include industrial process emissions.92
In line with the Paris Agreement and UNFCCC reporting guidelines, fuel combustion emissions are included in this
chapter, while process emissions are included in the Industrial Processes and Product Use chapter of this report. In
examining data from EPA's GHGRP that would be useful to improve the emission estimates for the CO2 from
biomass combustion category, particular attention will also be made to ensure time-series consistency, as the
facility-level reporting data from EPA's GHGRP are not available for all inventory years as reported in this
Inventory. Additionally, analyses will focus on aligning reported facility-level fuel types and IPCC fuel types per the
national energy statistics, ensuring CO2 emissions from biomass are separated in the facility-level reported data,
and maintaining consistency with national energy statistics provided by EIA. In 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.93
3.11 Energy Sources of Precursor
Greenhouse Gases
In addition to the main greenhouse gases addressed above, energy-related activities are also sources of
greenhouse gas precursors. The reporting requirements of the Paris Agreement and the UNFCCC94 request that
information should be provided on precursor emissions, which include carbon monoxide (CO), nitrogen oxides
(NOx), non-methane volatile organic compounds (NMVOCs), and sulfur dioxide (SO2). These gases are not direct
greenhouse gases, but indirectly impact Earth's radiative balance by altering the concentrations of greenhouse
92 See https://unfccc.int/resource/docs/2013/copl9/ene/10a03.pdf#paee=2.
93 See http://www.ipcc-nggip.iges.or.lp/public/tb/TFI Technical Bulletin l.pdf.
94 See paragraph 51 of Annex to 18/CMA.l available online at:
https://unfccc.int/sites/default/files/resource/CMA2018 03a02E.pdf.
Energy 3-131
-------
gases (e.g., tropospheric ozone) and atmospheric aerosol (e.g., particulate sulfate). Total emissions of NOx, CO,
NMVOCs, and SO2 from energy-related activities from 1990 to 2022 are reported in Table 3-132.
Table 3-132: NOx, CO, NMVOC, and SO2 Emissions from Energy-Related Activities (kt)
Gas/Activity
1990
2005 iiiii
2018
2019
2020
2021
2022
NOx
21,804
18,196
6,812
6,503
5,630
5,570
5,225
Fossil Fuel Combustion
21,678 1
18,188 I
6,804
6,496
5,624
5,563
5,218
Transportationa
12,132
12,628
4,486
4,322
3,618
3,546
3,228
Industrial
2,4751:
' ¦
1,486 E
820
800
751
721
721
Electric Power Sector
6,045
3,440
1,025
898
762
807
781
Commercial
451
288
186
187
193
189
188
Residential
575
346
288
290
300
300
300
Petroleum and Natural Gas Systems
127
8 5
1
1
1
6
6
International Bunker Fuels
1,953
1,699
1,456
1,280
977
1,008
1,132
CO
124,584
63,891
30,237
29,854
27,897
28,283
27,607
Fossil Fuel Combustion
124,353
63,686
30,050
29,660
27,703
28,098
27,426
Transportationa
119,478 ¦
59,540 jj
26,024
25,621
23,546
23,912
23,235
Residential
3,620
2,393
2,751
2,860
2,968
2,950
2,950
Industrial
705
976
620
600
670
658
655
Electric Power Sector
329
582
505
428
362
423
428
Commercial
220
195 1
151
151
157
154
158
Petroleum and Natural Gas Systems
232
205
186
194
194
185
181
International Bunker Fuels
102
131 is
158
150
83
101
128
NMVOCs
12,269
8,081
5,050
4,987
4,822
5,167
5,045
Fossil Fuel Combustion
11,793
6,079 I
2,632
2,593
2,391
2,454
2,329
Transportationa
10,932
5,608
2,127
2,072
1,846
1,912
1,786
Residential
693 |
322
382
397
431
429
429
Commercial
9
18
14
14
14
14
14
Industrial
87
80
81
74
73
74
Electric Power Sector
43
44
30
29
26
27
27
Petroleum and Natural Gas Systems
476 I
2,002
2,418
2,394
2,431
2,713
2,716
International Bunker Fuels
57
54
50
45
32
34
40
S02
22,638
13,331
1,827
1,509
1,288
1,423
1,327
Fossil Fuel Combustion
21,482
13,235
1,770
1,447
1,138
1,272
1,176
Electric Power Sector
14,432 '
9,436 I
1,189
921
758
898
819
Industrial
2,886
1,378
259
234
172
168
159
Transportationa
793 I
724
45
40
23
24
25
Commercial
485
318
18
19
13
14
13
Residential
2,8861
1,378
259
234
172
168
159
Petroleum and Natural Gas Systems
156
96
56
61
150
151
151
International Bunker Fuels
na :
na :
NA
NA
NA
NA
NA
NA (Not Applicable)
a The scope of the NEI for aircraft related precursor emissions included under the transportation is different from the Paris
Agreement and UNFCCC reporting scope. The NEI precursor estimate methodology does not exclude emissions that could be
considered international bunkers given local impacts from these emissions. The precursor estimates are based on modeled
using FAA- and state-supplied landing and take-off data for all aircraft types (including ground support equipment and
auxiliary engines) used for public, private, and military purposes.
Note: Totals may not sum due to independent rounding.
Source: (EPA 2023a). Emission categories from EPA (2023a) are aggregated into sectors and categories reported under the
Paris Agreement and the UNFCCC as shown in Table ES-3.
Methodology and Time-Series Consistency
Emission estimates for 1990 through 2022 were obtained from data published on the National Emissions Inventory
(NEI) Air Pollutant Emissions Trends Data website (EPA 2023a). For Table 3-132, NEI reported emissions of CO, NOx,
NMVOCs, and SO2 were recategorized from NEI Emissions Inventory System (EIS) sectors to source categories more
3-132 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
closely aligned with reporting sectors and categories under the Paris Agreement and the UNFCCC, based on
discussions between the EPA Inventory and NEI staff (see crosswalk documented in Annex 6.3).95 EIS sectors
mapped to the energy sector categories in this report include: fuel combustion for electric utilities, industrial, and
other; petroleum and related industries; highway vehicles; off-highway; and other mobile sources (e.g.,
commercial marine vessels and rail). As described in the NEI Technical Support Documentation (TSD) (EPA 2023b),
NEI emissions are estimated through a combination of emissions data submitted directly to the EPA by state, local,
and tribal air agencies, as well as additional information added by the Agency from EPA emissions programs, such
as the emission trading program, Toxics Release Inventory (TRI), and data collected during rule development or
compliance testing.
Methodological approaches were applied to the entire time series to ensure time-series consistency from 1990
through 2022, which are described in detail in the NEI'sTSD and on EPA's Air Pollutant Emission Trends website
(EPA 2023a; EPA 2023b). No quantitative estimates of uncertainty were calculated for this source category.
95 The NEI estimates and reports emissions from six criteria air pollutants (CAPs) and 187 hazardous air pollutants (HAPs) in
support of National Ambient Air Quality Standards. EPA reported CAP emission trends are grouped into 60 sectors and 15 Tier 1
source categories, which broadly cover similar source categories to those presented in this chapter. For reporting precursor
emissions in the common reporting tables (CRTs), EPA has mapped and regrouped emissions of greenhouse gas precursors (CO,
NOx, S02, and NMVOCs) from NEI's EIS sectors to better align with NIR source categories, and to ensure consistency and
completeness to the extent possible. See Annex 6.3 for more information on this mapping.
Energy 3-133
-------
4. Industrial Processes and Product Use
Industrial Processes and Product Use (IPPU) chapter includes greenhouse gas emissions occurring from industrial
processes and from the use of greenhouse gases in products. The industrial processes and product use categories
included in this chapter are presented in Figure 4-1 and Figure 4-2. Greenhouse gas emissions from industrial
processes can occur in two different ways. First, they may be generated and emitted as the byproducts of various
non-energy-related industrial activities. Second, they may be emitted due to their use in manufacturing processes
or by end-consumers. Combustion-related energy use emissions from industry are reported in Chapter 3, Energy.
In the case of byproduct emissions, the emissions are generated by an industrial process itself and are not directly
a result of energy consumed during the process. For example, raw materials can be chemically or physically
transformed from one state to another. This transformation can result in the release of greenhouse gases such as
carbon dioxide (CO2), methane (CH4), nitrous oxide (N2O), and fluorinated greenhouse gases (e.g., HFC-23). The
greenhouse gas byproduct generating processes included in this chapter include iron and steel production and
metallurgical coke production, cement production, petrochemical production, ammonia production, lime
production, other process uses of carbonates (e.g., flux stone, flue gas desulfurization, ceramics production, non-
metallurgical magnesia production, and soda ash consumption not associated with glass manufacturing), nitric acid
production, adipic acid production, urea consumption for non-agricultural purposes, aluminum production, HCFC-
22 production, other fluorochemical production, glass production, soda ash production, ferroalloy production,
titanium dioxide production, caprolactam production, zinc production, phosphoric acid production, lead
production, and silicon carbide production and consumption.
Greenhouse gases that are used in manufacturing processes or by end-consumers include man-made compounds
such as hydrofluorocarbons (HFCs), perfluorocarbons (PFCs), sulfur hexafluoride (SFs), and nitrogen trifluoride
(NF3). The present contribution of HFCs, PFCs, SF6, and NF3 gases to the radiative forcing effect of all anthropogenic
greenhouse gases is small; however, because of their extremely long lifetimes, many of them will continue to
persist in the atmosphere long after they were first released. In addition, many of these gases have high global
warming potentials; SF6 is the most potent greenhouse gas the Intergovernmental Panel on Climate Change (IPCC)
has evaluated. Use of HFCs continues since they are the primary substitutes for ozone depleting substances (ODS),
which are being phased-out under the Montreal Protocol on Substances that Deplete the Ozone Layer; however,
production and consumption of HFCs are being phased down under the Kigali Amendment to the Montreal
Protocol and in the United States under the American Innovation and Manufacturing Act. Hydrofluorocarbons,
PFCs, SFs, and NF3 are employed and emitted by a number of other industrial sources in the United States, such as
the electronics industry, electric power transmission and distribution, PFCs and SFsfor other product use, and
magnesium metal production and processing. Carbon dioxide is also consumed and emitted through various end-
use applications. In addition, nitrous oxide is used in and emitted by the electronics industry and anesthetic and
aerosol applications.
Industrial Processes and Product Use 4-1
-------
In 2022, IPPU generated emissions of 383.2 million metric tons of CO2 equivalent (MMT CO2 Eq.), or 6.0 percent of
total U.S. greenhouse gas emissions.1 Carbon dioxide emissions from all industrial processes were 168.9 MMT CO2
Eq. (168,937 kt CO2) in 2022, or 3.3 percent of total U.S. CO2 emissions. Methane emissions from industrial
processes resulted in emissions of approximately 0.04 MMT CO2 Eq. (1 kt CH4) in 2022, which was 0.01 percent of
U.S. CH4 emissions. Nitrous oxide emissions from IPPU were 16.1 MMT CO2 Eq. (61 kt N2O) in 2022, or 4.1 percent
of total U.S. N2O emissions. In 2022 combined emissions of HFCs, PFCs, SF6, and NF3 totaled 198.1 MMT CO2 Eq.
Total emissions from IPPU in 2022 were 3.9 percent more than 1990 emissions. Total emissions from IPPU
remained relatively constant between 2021 and 2022, increasing by 0.4 percent due to offsetting trends within the
sector. More information on emissions of greenhouse gas precursors emissions that also result from IPPU are
presented in Section 4.27 of this chapter.
The largest source of IPPU-related emissions is the substitution of ozone depleting substances, which accounted
for 46.5 percent of sector emissions in 2022. These emissions have increased by 79.1 percent since 2005, and 3.2
percent between 2021 and 2022, Cement production was the second largest source of IPPU emissions in 2022,
accounting for 10.9 percent of IPPU emissions in 2022. Iron and steel production and metallurgical coke production
was the third largest source of IPPU emissions, accounting for 10.6 percent of the sector total in 2022.
178
Figure 4-1: Industrial Processes arid Product Use Sector Greenhouse Gas Sources
Substitution of Ozone Depleting Substances
Cement Production
Iron and Steel Production & Metallurgical Coke Production
Petrochemical Production
Ammonia Production
Lime Production
Other Process Uses of Carbonates
Nitric Acid Production
Fluorochemical Production
Urea Consumption for Non-Agricultural Purposes)
Electrical Equipment]
Carbon Dioxide Consumption
Electronics Industry]
N2O from Product Uses,
Aluminum Production
Adipic Acid Production
Glass Production
Soda Ash Production
Titanium Dioxide Production
Ferroalloy Production
Caprolactam, Glyoxal, and Glyoxylic Acid Production
Magnesium Production and Processing
Zinc Production
Phosphoric Acid Production
SFe and PFCs from Other Product Use
Lead Production
Carbide Production and Consumption
0 10 20 30 40 50 60 70
MMT COz Eq.
Industrial Processes and Product
Use as a Portion of All Emissions
I Energy
1 Agriculture
! IPPU
Waste
< 0.5
< 0.5
The increase in overall IPPU emissions since 1990 reflects a range of emission trends among the emission sources,
as shown in Figure 4-2. Emissions resulting from most types of metal production have declined significantly since
1990, largely due to production shifting to other countries, but also due to transitions to less-emissive methods of
1 Emissions reported in the IPPU chapter include those from all 50 states, including Hawaii and Alaska, as well as from U.S.
Territories.
4-2 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
production (in the case of iron and steel) and to improved practices (in the case of PFC emissions from aluminum
production). Carbon dioxide and CFU emissions from some chemical production sources (e.g., petrochemical
production, urea consumption for non-agricultural purposes) have increased since 1990, while emissions from
other chemical production sources (e.g., ammonia production, phosphoric acid production) have decreased.
Emissions from mineral sources have either increased (e.g., cement production) or not changed significantly (e.g.,
lime production) since 1990 and largely follow economic cycles. Hydrofluorocarbon emissions from the
substitution of ODS have increased drastically since 1990 and are the largest source of IPPU emissions (46.5
percent in 2022), while the emissions of HFCs, PFCs, SF6, and NF3 from other sources have generally declined.
Nitrous oxide emissions from the production of nitric acid have decreased. Some emission sources (e.g., adipic
acid) exhibit varied interannual trends. Trends are explained further within each emission source category
throughout the chapter.
Figure 4-2: Trends in Industrial Processes and Product Use Sector Greenhouse Gas Sources
500
450
400
350
S 300
O
u
250
200
150
100
50
0
I Electronics Industry
Other Product Manufacture and Use
I Mineral Industry
I Metal Industry
I Chemical Industry
Substitution of Ozone Depleting Substances
ai rv
^ 2
CJ
U1
m
oj co
00 oo
p-> n
Table 4-1 summarizes emissions for the IPPU chapter in MMT CO2 Eq. using IPCC Fifth Assessment Report (AR5)
GWP values, following the requirements of the current United Nations Framework Convention on Climate Change
(UNFCCC) reporting guidelines for national inventories (IPCC 2007).2 Unweighted gas emissions in kt are also
provided in Table 4-2. The source descriptions that follow in the chapter are presented in the order as reported to
the UNFCCC in the Common Reporting Tables (CRTs), corresponding generally to: mineral industry, chemical
industry, metal industry, and emissions from the uses of HFCs, PFCs, SF6, and NF3.
Each year, some emission and sink estimates in the IPPU sector of the Inventory are recalculated and revised with
improved methods and/or data. In general, recalculations are made to the U.S. greenhouse gas emission estimates
either to incorporate new methodologies or, most commonly, to update recent historical data. These
improvements are implemented consistently across the previous Inventory's time series (i.e., 1990 to 2021) to
2 See http://unfccc.int/resource/docs/2013/copl9/ene/10a03.pdf.
Industrial Processes and Product Use 4-3
-------
ensure that the trend is accurate. Key updates to this year's Inventory include the addition of new categories
previously not estimated: CO2 emissions from ceramics production and non-metallurgical magnesia within other
process use of carbonates category; fluorinated gases from production of fluorochemicals other than HCFC-22
within the fluorochemical production category; and SF6 and PFCs from additional product uses within the other
product manufacture and use category. In addition, there were changes to the petrochemical production
methodology to calculate emissions from methanol production; updates to emission estimates from urea
consumption for non-agricultural purposes driven by revisions to quantities of urea applied, urea imports, and
urea exports; and revisions to the method for estimating SF6 emissions from electrical equipment for estimating
using CARBdata from electrical equipment in California. Together, these methodological and data updates
increased IPPU sector greenhouse gas emissions by an average 20.4 MMT CO2 Eq. (7.2 percent) across the time
series. For more information on specific methodological updates, please see the Recalculations Discussion section
for each category in this chapter.
Table 4-1: Emissions from Industrial Processes and Product Use (MMT CO2 Eq.)
Gas/Source
1990
2005
2018
2019
2020
2021
2022
C02
213.7
195.9
164.4
168.2
160.7
168.8
168.9
Cement Production
33.5
46.2
1
39.0
40.9
40.7
41.3
41.9
Iron and Steel Production &
Metallurgical Coke Production
104.7
70.1
42.9
43.1
37.7
41.9
40.7
Iron and Steel Production
99.1
66.2
I
41.6
40.1
35.4
38.6
37.7
Metallurgical Coke Production
5.6
3.9
1.3
3.0
2.3
3.2
3.0
Petrochemical Production
20.1
1
1
26.9
1
1
27.2
28.5
27.9
30.7
28.8
Ammonia Production
14.4
10.2
12.7
12.4
13.0
12.2
12.6
Lime Production
11.7
1
14.6
J
13.1
12.1
11.3
11.9
12.2
Other Process Uses of Carbonates
7.1
8.5
7.9
9.0
9.0
8.6
10.4
Urea Consumption for Non-
1
1
Agricultural Purposes
3.8
¦
3.7
1
6.1
6.2
5.8
6.6
7.1
Carbon Dioxide Consumption
1.5
1.4
4.1
4.9
5.0
5.0
5.0
Glass Production
2.3
I
I
2.4
1
2.0
1.9
1.9
2.0
2.0
Soda Ash Production
1.4
1.7
1.7
1.8
1.5
1.7
1.7
Titanium Dioxide Production
1.2
1
1
1.8
I
1
1.5
1.3
1.3
1.5
1.5
Aluminum Production
6.8
4.1
1.5
1.9
1.7
1.5
1.4
Ferroalloy Production
2.2
I
I
1.4
I
2.1
1.6
1.4
1.6
1.3
Zinc Production
0.6
1.0
1.0
1.0
1.0
1.0
0.9
Phosphoric Acid Production
1.5
I
1.3
0.9
0.9
0.9
0.9
0.8
Lead Production
0.5
0.6
¦
0.5
0.5
0.5
0.4
0.4
Carbide Production and
1
1
Consumption
0.2
1
1
0.2
0.2
0.2
0.2
0.2
0.2
Substitution of Ozone Depleting
Substances
+
+
+
+
+
+
+
Magnesium Production and
)
1
1
Processing
0.1
1
+
I
+
+
+
+
+
ch4
0.1
+
+
+
+
+
+
Carbide Production and
I
1
1
Consumption
+
I
1
+
1
I
+
+
+
+
+
Ferroalloy Production
+
+
+
+
+
+
+
Iron and Steel Production &
1
1
1
Metallurgical Coke Production
+
mm)
+
+
+
+
+
+
Petrochemical Production
+
+
+
+
+
+
+
n2o
29.6
1
22.2
23.1
18.7
20.8
19.7
16.1
Nitric Acid Production
10.8
10.1
8.5
8.9
8.3
7.9
8.6
N20 from Product Uses
3.8
1
3.8
1
3.8
3.8
3.8
3.8
3.8
4-4 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
Gas/Source
1990
2005
I
mm!
2018
2019
2020
2021
2022
AdipicAcid Production
13.5
6.3
9.3
4.7
7.4
6.6
2.1
Caprolactam, Glyoxal, and Glyoxylic
1
I
Acid Production
1.9
1.3
1.2
1.1
1.2
1.3
Electronics Industry
+
0.1
0.2
0.2
0.3
0.3
0.3
HFCs
47.7
121.7
1
163.9
168.2
170.3
177.0
182.8
Substitution of Ozone Depleting
Substances3
0.3
99.5
157.9
162.1
166.2
172.6
178.1
Fluorochemical Production
47.3 I
22.1
I
111:
5.7
5.7
3.8
4.0
4.3
Electronics Industry
0.2
0.2
0.3
0.3
0.3
0.4
0.3
Magnesium Production and
1
I
Processing
0.0
0.0
I
I
0.1
0.1
0.1
+
+
PFCs
39.5
10.2
7.4
7.3
6.6
6.3
6.7
Fluorochemical Production
17-5 !
4.0
2.9
3.0
2.5
2.6
3.0
Electronics Industry
2.5
3.0
2.9
2.6
2.5
2.6
2.7
Aluminum Production
193 1
3.1
1
1
1.4
1.4
1.4
0.9
0.8
SF6 and PFCs from Other Product
Use
0.1
0.1
0.2
0.2
0.2
0.1
0.2
Substitution of Ozone Depleting
1
I
I
Substances3
NO i
+
mill!
+
+
+
+
+
Electrical Equipment
+
+
0.0
+
+
+
+
sf6
37.9
20.2
1
7.6
8.4
8.1
8.5
7.6
Electrical Equipment
24.7
11.8
5.0
6.1
5.9
6.0
5.1
Magnesium Production and
I
imm
1
Processing
5.6
3.0
1
1.1
0.9
0.9
1.2
1.1
Electronics Industry
0.5
0.8
0.8
0.8
0.8
0.9
0.8
SFe and PFCs from Other Product
!!!!!»
1
1
Use
1-3
1.3
1
0.8
0.6
0.5
0.4
0.6
Fluorochemical Production
5.8
3.3
+
+
+
+
+
nf3
0.3
1.0
I
0.7
1.1
1.3
1.1
1.1
Electronics Industry
+
0.4
0.5
0.5
0.6
0.6
0.6
Fluorochemical Production
0.3
0.6
I
0.1
0.6
0.7
0.5
0.5
Totalb
368.8 si
371.3
367.2
371.9
367.9
381.6
383.2
+ Does not exceed 0.05 MMT C02 Eq.
NO (Not Occurring)
a Small amounts of PFC emissions from this source are included under HFCs due to confidential business information.
bTotal does not include other fluorinated gases, such as HFEs and PFPEs, which are reported separately in Section 4.24.
Note: Totals may not sum due to independent rounding. Emissions of F-HTFs that are not HFCs, PFCs or SF6 are not included
in Inventory totals and are included for informational purposes only in Section 4.24. Emissions presented for informational
purposes include HFEs, PFPMIEs, perfluoroalkylmorpholines, and perfluorotrialkylamines.
Table 4-2: Emissions from Industrial Processes and Product Use (kt)
Gas/Source
1990
2005
2018
2019
2020
2021
2022
co2
213,682
195,933
164,404
168,175
160,701
168,838
168,937
Cement Production
33,484
46,194
38,971
40,896
40,688
41,312
41,884
Iron and Steel Production &
Metallurgical Coke Production
104,740
70,082
42,863
43,095
37,724
41,873
40,672
Iron and Steel Production
99,132
66,161 :
41,581
40,089
35,398
38,648
37,718
Metallurgical Coke Production
5,608
3,921
1,282
3,006
2,325
3,224
2,954
Petrochemical Production
20,075
26,882 3
27,200
28,483
27,926
30,656
28,788
Ammonia Production
14,404
10,234
12,669
12,401
13,006
12,192
12,610
Lime Production
1.1,700
14,552
13,106
12,112
11,299
11,870
12,208
Other Process Uses of Carbonates
7,103 _
8,472
7,938
8,973
9,012
8,583
10,384
Industrial Processes and Product Use 4-5
-------
Gas/Source
1990
2005
2018
2019
2020
2021
2022
Urea Consumption for Non-
¦
Agricultural Purposes
3,784
I
1
3,653
I
6,113
6,150
5,805
6,600
7,053
Carbon Dioxide Consumption
1,472
1,375
4,130
4,870
4,970
4,990
5,000
Glass Production
2,263
I
1
2,402
1
I
1,989
1,940
1,858
1,969
1,956
Soda Ash Production
1,431
1,655
1,714
1,792
1,461
1,714
1,704
Titanium Dioxide Production
1,195
I
mill)
1,755
1
1,541
1,340
1,340
1,474
1,474
Aluminum Production
6,831
4,142
1,455
1,880
1,748
1,541
1,446
Ferroalloy Production
2,152
1
1
1,392
1
I
2,063
1,598
1,377
1,567
1,327
Zinc Production
632
1,030
999
1,026
977
1,007
947
Phosphoric Acid Production
1,529
1,342
1
1
937
909
901
874
840
Lead Production
516
553
527
531
450
439
428
Carbide Production and
1
1
1
Consumption
243
I
213
184
175
154
172
210
Substitution of Ozone Depleting
Substances3
+
I
1
1
3
3
4
4
4
Magnesium Production and
1
|
1
Processing
129
I
1
4
2
2
3
3
3
ch4
3
2
2
1
1
1
1
Carbide Production and
1
1
Consumption
1
1
+
1
1
+
+
+
+
+
Ferroalloy Production
1
+
1
+
+
+
+
Iron and Steel Production &
1
I
1
Metallurgical Coke Production
1
1
1
1
+
+
+
+
+
Petrochemical Production
+
+
+
+
+
+
+
n2o
112
I
84
I
I
87
71
79
74
61
Nitric Acid Production
41
38
32
34
31
30
33
N2O from Product Uses
14
14
I
14
14
14
14
14
AdipicAcid Production
51
24
35
18
28
25
8
Caprolactam, Glyoxal, and Glyoxylic
I
1
Acid Production
6
I
I
7
5
5
4
5
5
Electronics Industry
+
+
1
1
1
1
1
HFCs
M
I
M
1
1
M
M
M
M
M
Substitution of Ozone Depleting
Substances3
M
M
M
M
M
M
M
Fluorochemical Production
M
1
M
1
I
M
M
M
M
M
Electronics Industry
+
+
+
+
+
+
+
Magnesium Production and
1
Processing
0
I
0
1
+
+
+
+
+
PFCs
101
83
179
173
167
139
172
Fluorochemical Production
M
M
1
1
M
M
M
M
M
Electronics Industry
+
+
+
+
+
+
+
Aluminum Production
M
1
M
IB!
i
M
M
M
M
M
SF6 and PFCs from Other Product
Use
101
83
178
173
167
138
172
Substitution of Ozone Depleting
1
1
1
1
Substances3
NO
1
1
+
+
+
+
+
+
Electrical Equipment
+
+
0
+
+
+
+
sf6
2
I
1
1
1
+
+
+
+
+
Electrical Equipment
1
1
+
+
+
+
+
Magnesium Production and
1
1
Processing
+
1
I
+
SI
+
+
+
+
+
Electronics Industry
+
+
1
+
+
+
+
+
4-6 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
Gas/Source
1990
2005
2018
2019
2020
2021
2022
SF6 and PFCs from Other Product
Use
0
o(
+
+
+
+
+
Fluorochemical Production
0
0
+
+
+
+
+
nf3
+
0
+
0
0
0
0
Electronics Industry
+
+
+
+
+
+
+
Fluorochemical Production
+
+
+
+
+
+
+ Does not exceed 0.5 kt.
M (Mixture of gases)
NO (Not Occurring)
a Small amounts of PFC emissions from this source are included under HFCs due to confidential business information.
Note: Totals by gas may not sum due to independent rounding.
This chapter presents emission estimates calculated in accordance with the 2006IPCC Guidelines for National
Greenhouse Gas Inventories (2006 IPCC Guidelines) and its refinements. For additional detail on IPPU sources that
are not included in this Inventory report, please review Annex 5, Assessment of the Sources and Sinks of
Greenhouse Gas Emissions Not Included. These sources are not included due to various national circumstances,
such as emissions from a source may not currently occur in the United States, data are not currently available for
those emission sources (e.g., glyoxal and glyoxylic acid production, Cm from direct reduced iron production),
emissions are included elsewhere within the Inventory report, or data suggest that emissions are not significant
(e.g., other various fluorinated gas emissions from other product uses). In terms of geographic scope, emissions
reported in the IPPU chapter include those from all 50 states, including Hawaii and Alaska, as well as from District
of Columbia and U.S. Territories to the extent to which industries are occurring. While most IPPU sources do not
occur in U.S. Territories (e.g., electronics manufacturing does not occur in U.S. Territories), they are estimated and
accounted for where they are known to occur (e.g., cement production, lime production, electrical equipment).
EPA will review this on an ongoing basis to ensure emission sources are included across all geographic areas if they
occur. Information on planned improvements for specific IPPU source categories can be found in the Planned
Improvements section of the individual source category.
In addition, as mentioned in the Energy chapter of this report (Box 3-5), fossil fuels consumed for non-energy uses
for primary purposes other than combustion for energy (including lubricants, paraffin waxes, bitumen asphalt, and
solvents) are reported in the Energy chapter. According to the 2006 IPCC Guidelines, these non-energy uses of
fossil fuels are to be reported under the IPPU, rather than the Energy sector; however, due to national
circumstances regarding the allocation of energy statistics and carbon balance data, the United States reports
these non-energy uses in the Energy chapter of this Inventory. Although emissions from these non-energy uses are
reported in the Energy chapter, the methodologies used to determine emissions are compatible with the 2006
IPCC Guidelines and are well documented and scientifically based. The methodologies used are described in
Section 3.2, Carbon Emitted from Non-Energy Uses of Fossil Fuels and Annex 2.3, Methodology for Estimating
Carbon Emitted from Non-Energy Uses of Fossil Fuels. The emissions are reported under the Energy chapter to
improve transparency, report a more complete carbon balance, and avoid double counting. For example, only the
emissions from the first use of lubricants and waxes are to be reported under the IPPU sector, and emissions from
use of lubricants in 2-stroke engines and emissions from secondary use of lubricants and waxes in waste
incineration with energy recovery are to be reported under the Energy sector. Reporting non-energy use emissions
from only first use of lubricants and waxes under IPPU would involve making artificial adjustments to the non-
energy use carbon balance and could potentially result in double counting of emissions. These artificial
adjustments would also be required for asphalt and road oil and solvents (which are captured as part of
petrochemical feedstock emissions) and could also potentially result in double counting of emissions. For more
information, see the Methodology discussion in Section 3.1, CO2 from Fossil Fuel Combustion, Section 3.2, Carbon
Emitted from Non-Energy Uses of Fossil Fuels and Annex 2.3, Methodology for Estimating Carbon Emitted from
Non-Energy Uses of Fossil Fuels.
Finally, as stated in the Energy chapter, portions of the fuel consumption data for seven fuel categories—coking
coal, distillate fuel, industrial other coal, petroleum coke, natural gas, residual fuel oil, and other oil—are
Industrial Processes and Product Use 4-7
-------
reallocated to the IPPU chapter, as they are consumed during non-energy related industrial process activity.
Emissions from uses of fossil fuels as feedstocks or reducing agents (e.g., petrochemical production, aluminum
production, titanium dioxide, zinc production) are reported in the IPPU chapter, unless otherwise noted due to
specific national circumstances. This approach is compatible with the 2006IPCC Guidelines and is well documented
and scientifically based. The emissions from these feedstocks and reducing agents are reported under the IPPU
chapter to improve transparency and to avoid double counting of emissions under both the Energy and IPPU
sectors. More information on the methodology to adjust for these emissions within the Energy chapter is
described in the Methodology section of CO2 from Fossil Fuel Combustion (3.1 Fossil Fuel Combustion [CRT Source
Category 1A]) and Annex 2.1 Methodology for Estimating Emissions of CO2 from Fossil Fuel Combustion. Additional
information is listed within each IPPU emission source in which this approach applies.
Box 4-1: Methodological Approach for Estimating and Reporting U.S. Emissions and
Removals
Consistent with Article 13.7(a) of the Paris Agreement and Article 4.1(a) of the UNFCCC as well as relevant
decisions under those agreements, the emissions and removals 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) in the 2006 IPCC Guidelines for National Greenhouse Gas
Inventories (2006 IPCC Guidelines) and its supplements and refinements. Additionally, the calculated emissions
and removals in a given year for the United States are presented in a common format in line with the reporting
guidelines for the reporting of inventories under the Paris Agreement and the UNFCCC. The Parties' use of
consistent methods to calculate emissions and removals for their inventories helps to ensure that these reports
are comparable. The presentation of emissions and removals provided in the IPPU chapter do not preclude
alternative examinations. Rather, this chapter presents emissions and removals in a common format consistent
with how Parties are to report inventories under the Paris Agreement and the UNFCCC. The report itself, and this
chapter, follows this common format, and provides an explanation of the application of methods used to
calculate emissions and removals from industrial processes and from the use of greenhouse gases in products.
QA/QC and Verification Procedures
The quality of IPPU source categories is assured through application of the U.S. Inventory QA/QC plan outlined in
Annex 8. Two types of checks were performed using this plan: (1) general (Tier 1) procedures consistent with
Volume 1, Chapter 6 of the 2006 IPCC Guidelines that focus on annual procedures and checks to be used when
gathering, maintaining, handling, documenting, checking, and archiving the data, supporting documents, and files;
and (2) source category-specific (Tier 2) procedures that focus on checks and comparisons of the emission factors,
activity data, and methodologies used for estimating emissions from the relevant industrial process and product
use sources. Examples of these procedures include: checks to ensure that activity data and emission estimates are
consistent with historical trends; that consistent, complete and data sources are used and documented; that
interpolation or extrapolation techniques are consistent across sources; and that common units, and conversion
factors are used where applicable. Consistent with the 2006 IPCC Guidelines, additional category-specific QC
procedures were performed for more significant emission categories or sources where significant methodological
and data updates have taken place. Any significant findings and errors identified are documented and corrected.
Application of these procedures, specifically category-specific QC procedures and updates/improvements as a
result of QA processes (expert, public, and UNFCCC technical expert reviews), are described further within
respective source categories, in the Recalculations Discussion and Planned Improvement sections.
For sources that use data from EPA's Greenhouse Gas Reporting Program (GHGRP), EPA verifies annual facility-
level reports through a multi-step process (e.g., including a combination of pre-and post-submittal electronic
4-8 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
checks and manual reviews by staff) to identify potential errors and ensure that data submitted to EPA are
accurate, complete, and consistent.3 Based on the results of the verification process, EPA follows up with facilities
to resolve mistakes that may have occurred. The post-submittals checks are consistent with the general and
category-specific QC procedures including: range checks, statistical checks, algorithm checks, and year-to-year
checks of reported data and emissions. See Box 4-2 below for more information on use of GHGRP data in this
chapter.
For most IPPU categories, activity data are obtained via aggregation of facility-level data from EPA's GHGRP (see
Box 4-2 below and Annex 9), national commodity surveys conducted by U.S. Geological Survey (USGS) National
Minerals Information Center, U.S. Department of Energy (DOE), U.S. Census Bureau, and industry associations such
as Air-Conditioning, Heating, and Refrigeration Institute (AHRI), American Chemistry Council (ACC), and American
Iron and Steel Institute (AISI) (specified within each source category). The emission factors used include those
derived from the EPA's GHGRP and application of IPCC default factors.
Box 4-2: Industrial Process and Product Use Data from EPA's Greenhouse Gas Reporting
Program
EPA collects greenhouse gas emissions data from individual facilities and suppliers of certain fossil fuels and
industrial gases through its Greenhouse Gas Reporting Program (GHGRP). The GHGRP 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 sources or suppliers in 41 industrial categories.
Annual 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, but reporting is
required for all facilities in some industries. Calendar year 2010 was the first year for which data were collected
for facilities subject to 40 CFR Part 98, though some source categories first collected data for calendar year 2011.
For more information, see Annex 9, Use of EPA Greenhouse Gas Reporting Program in Inventory.
EPA uses annual GHGRP data in a number of categories to improve the national estimates presented in this
Inventory, consistent with IPCC guidelines (e.g., minerals, chemicals, product uses). Methodologies used in EPA's
GHGRP are consistent with IPCC guidelines, including higher tier methods; however, it should be noted that the
coverage and definitions for source categories (e.g., allocation of energy and IPPU emissions) in EPA's GHGRP
may differ from those used in this Inventory in meeting the Paris Agreement and UNFCCC reporting guidelines
(IPCC 2011) and is an important consideration when incorporating GHGRP data in the Inventory. In line with the
Paris Agreement and UNFCCC reporting guidelines, the Inventory is a comprehensive accounting of all emissions
from source categories identified in the 2006 IPCC Guidelines. EPA has paid particular attention to ensuring both
completeness and time-series consistency for major recalculations that have occurred from the incorporation of
GHGRP data into these categories, consistent with 2006 IPCC Guidelines and the 2019 Refinement, Volume 1,
Chapter 2, Section 2.3, Use of Facility Data in Inventories.4
For certain source categories in this Inventory (e.g., nitric acid production, lime production, cement production,
petrochemical production, carbon dioxide consumption, ammonia production, and urea consumption for non-
agricultural purposes), EPA has integrated data values that have been calculated by aggregating GHGRP data that
are considered confidential business information (CBI) at the facility level. EPA, with industry engagement, has
put forth criteria to confirm that a given data aggregation shields underlying CBI from public disclosure. EPA is
3 See https://www.epa.gov/sites/production/files/2015-07/documents/ghgrp verification factsheet.pdf.
4 See https://www.ipcc-nggip.iges.or.ip/public/2019rf/pdf/l Volumel/19R VI Ch02 DataCollection.pdf.
Industrial Processes and Product Use 4-9
-------
only publishing data values that meet these aggregation criteria.5 Specific uses of aggregated facility-level data
are described in the respective methodological sections (e.g., including other sources using GHGRP data that is
not aggregated CBI, such as aluminum, electronics industry, electrical equipment, HCFC-22 production, and
magnesium production and processing). For other source categories in this chapter, as indicated in the
respective planned improvements sections,6 EPA is continuing to analyze how facility-level GHGRP data may be
used to improve the national estimates presented in this Inventory, giving particular consideration to ensuring
time-series consistency and completeness.
Additionally, EPA's GHGRP has and will continue to enhance QA/QC procedures and assessment of uncertainties
within the IPPU categories (see those categories for specific QA/QC details regarding the use of GHGRP data).
4.1 Cement Production (CRT Source
Category 2A1)
Cement production is an energy- and raw material-intensive process that results in the generation of carbon
dioxide (CO2) both from the energy consumed in making the clinker precursor to cement and from the chemical
process to make the clinker. This reporting category (2A1) includes emissions from production of clinker and use of
cement kiln dust. Per the IPCC methodological guidance, emissions from fuels consumed for energy purposes
during the production of cement are accounted for as part of fossil fuel combustion in the industrial end-use sector
reported under the Energy chapter.
During the clinker production process, the key reaction occurs when calcium carbonate (CaCOs), in the form of
limestone or similar rocks or in the form of cement kiln dust (CKD), is heated in a cement kiln at a temperature
range of about 700 to 1,000 degrees Celsius (1,300 to 1,800 degrees Fahrenheit) to form lime (i.e., calcium oxide,
or CaO) and CO2 in a process known as calcination or calcining. The quantity of CO2 emitted during clinker
production is directly proportional to the lime content of the clinker. During calcination, each mole of CaC03
heated in the clinker kiln forms one mole of CaO and one mole of CO2. The CO2 is vented to the atmosphere as part
of the kiln exhaust:
CaC03 + heat -» CaO + C02
Next, over a temperature range of 1000 to 1450 degrees Celsius, the CaO combines with alumina, iron oxide and
silica that are also present in the clinker raw material mix to form hydraulically reactive compounds within white-
hot semifused (sintered) nodules of clinker. These "sintering" reactions are highly exothermic and produce few CO2
process emissions. The clinker is then rapidly cooled to maintain quality and then very finely ground with a small
amount of gypsum and potentially other materials (e.g., ground granulated blast furnace slag, etc.) to make
Portland and similar cements.
Masonry cement consists of plasticizers (e.g., ground limestone, lime, etc.) and portland cement, and the amount
of portland cement used accounts for approximately 3 percent of total clinker production (USGS 2023b; 2023c). No
5 U.S. EPA Greenhouse Gas Reporting Program. Developments on Publication of Aggregated Greenhouse Gas Data, November
25, 2014. See http://www.epa.gov/ghgreporting/confidential-business-information-ghg-reporting.
6 Ammonia production, glass production, lead production, and other fluorinated gas production.
4-10 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
additional emissions are associated with the production of masonry cement. Carbon dioxide emissions that result
from the production of lime used to produce portland and masonry cement are included in Section 4.2.
Carbon dioxide emitted from the chemical process of cement production is the second largest source of industrial
CO2 emissions in the United States. Cement is produced in 34 states and Puerto Rico. Texas, Missouri, California,
and Florida were the leading cement-producing states in 2022 and accounted for approximately 43 percent of total
U.S. production (USGS 2023b). In 2022, shipments of cement were estimated to have increased by about 3 percent
from 2021, and net imports increased by about 17 percent compared to 2021 (USGS 2023b).
In 2022, U.S. clinker production totaled 80,500 kilotons, which was an increase of 1 percent compared to 2021 and
an increase of 25 percent compared to 1990 (EPA 2023). The resulting CO2 emissions were estimated to be 41.9
MMT CO2 Eq. (41,884 kt) (see Table 4-3 and Table 4-4). The total construction value and cement shipments
increased by 11 percent and 4 percent, respectively, during the first nine months of 2022 compared to the same
time period in 2021. This increase was attributed to continued economic recovery from the COVID-19 pandemic
and the November 2021 passage of the Bipartisan Infrastructure Law. Despite the increases, growth was
constrained by increased costs, labor and production shortages, and ongoing supply chain disruptions (USGS
2023b).
Table 4-3: CO2 Emissions from Cement Production (MMT CO2 Eq.)
Year 1990 2005 2018 2019 2020 2021 2022
Cement Production 33.5 46.2 39.0 40.9 40.7 41.3 41.9
Table 4-4: CO2 Emissions from Cement Production (kt CO2)
Year 1990 2005 2018 2019 2020 2021 2022
Cement Production 33,484 | 46,194 | 38,971 40,896 40,688 41,312 41,884
Greenhouse gas emissions from cement production, which are primarily driven by production levels, increased
every year from 1991 through 2006 but decreased in the following years until 2009. Emissions from cement
production were at their highest levels in 2006 and at their lowest levels in 2009. Emissions in 2009 were
approximately 28 percent lower than 2008 emissions and 12 percent lower than 1990 due to the economic
recession and the associated decrease in demand for construction materials. Since 2009, emissions have increased
by 41 percent due to increasing demand for cement. Cement continues to be a critical component of the
construction industry; therefore, the availability of public and private construction funding, as well as overall
economic conditions, have considerable impact on the level of cement production.
Methodology and Time-Series Consistency
Carbon dioxide emissions from cement production are estimated using the Tier 2 method from the 2006IPCC
Guidelines as this is a key category, in accordance with the IPCC methodological decision tree and available data.
The Tier 2 methodology was used because detailed and complete data (including weights and composition) for
carbonate(s) consumed in clinker production are not available,7 and thus a rigorous Tier 3 approach is impractical.
Tier 2 specifies the use of aggregated plant or national clinker production data and an emission factor, which is the
product of the average lime mass fraction for clinker of 65 percent and a constant reflecting the mass of CO2
released per unit of lime. The U.S. Geological Survey (USGS) mineral commodity expert for cement has confirmed
that this is a reasonable assumption for the United States (Van Oss 2013a). This calculation yields an emission
factor of 0.510 tons of CO2 per ton of clinker produced, which was determined as follows:
7 As discussed further under "Planned Improvements," most cement-producing facilities that report their emissions to the
GHGRP use CEMS to monitor combined process and fuel combustion emissions for kilns, making it difficult to quantify the
process emissions on a facility-specific basis. By the end of 2022, the percentage of facilities not using CEMS was 1 percent.
Industrial Processes and Product Use 4-11
-------
Equation 4-1:2006IPCC Guidelines Tier 1 Emission Factor for Clinker (precursor to Equation
2.4)
Kg \ ( g M tons C02
44.01—V CO, ) H- 56.08—VCaO) = 0.510 ——
mole 2 J \ mole /J ton clinker
During clinker production, some of the raw materials, partially reacted raw materials, and clinker enters the kiln
line's exhaust system as non-calcinated, partially calcinated, or fully calcinated cement kiln dust (CKD). To the
degree that the CKD contains carbonate raw materials which are then calcined, there are associated CChemissions.
At some plants, essentially all CKD is directly returned to the kiln, becoming part of the raw material feed, or is
likewise returned to the kiln after first being removed from the exhaust. In either case, the returned CKD becomes
a raw material, thus forming clinker, and the associated CO2 emissions are a component of those calculated for the
clinker overall. At some plants, however, the CKD cannot be returned to the kiln because it is chemically unsuitable
as a raw material or chemical issues limit the amount of CKD that can be so reused. Any clinker that cannot be
returned to the kiln is either used for other (non-clinker) purposes or is landfilled. The CO2 emissions attributable
to the non-returned calcinated portion of the CKD are not accounted for by the clinker emission factor and thus a
CKD correction factor should be applied to account for those emissions. The USGS reports the amount of CKD used
to produce clinker, but no information is currently available on the total amount of CKD produced annually.8
Because data are not currently available to derive a country-specific CKD correction factor, a default correction
factor of 1.02 (2 percent) was used to account for CKD CO2 emissions, as recommended by the IPCC (IPCC 2006).9
Total cement production emissions were calculated by adding the emissions from clinker production and the
emissions assigned to CKD.
Small amounts of impurities (i.e., not calcium carbonate) may exist in the raw limestone used to produce clinker.
The proportion of these impurities is generally minimal, although a small amount (1 to 2 percent) of magnesium
oxide (MgO) may be desirable as a flux. Per the IPCC Tier 2 methodology, a correction for MgO is not used, since
the amount of MgO from carbonate is likely very small and the assumption of a 100 percent carbonate source of
CaO already yields an overestimation of emissions (IPCC 2006).
The 1990 through 2012 activity data for clinker production were obtained from USGS (Van Oss 2013a; Van Oss
2013b). Clinker production data for 2013 were also obtained from USGS (USGS 2014). USGS compiled the data (to
the nearest ton) through questionnaires sent to domestic clinker and cement manufacturing plants, including
facilities in Puerto Rico. Clinker production values in the current Inventory report utilize GHGRP data for the years
2014 through 2022 (EPA 2023). Clinker production data are summarized in Table 4-5. Details on how this GHGRP
data compares to USGS reported data can be found in the section on QA/QC and Verification.
Table 4-5: Clinker Production (kt)
Year
1990
2005 2018
2019
2020
2021
2022
Clinker Production
64,355 i
38,783 1 74,900
78,600
78,200
79,400
80,500
Note: Clinker production from 1990 through 2022 includes Puerto Rico (relevant U.S.
Territories).
Methodological approaches were applied to the entire time series to ensure time-series consistency from 1990
through 2022. The methodology for cement production spliced activity data from two different sources: USGS for
8 The USGS Minerals Yearbook: Cement notes that CKD values used for clinker production are likely underreported.
9 As stated on p. 2.12 of the 2006 IPCC Guidelines, Vol. 3, Chapter 2: "...As data on the amount of CKD produced may be scarce
(except possibly for plant-level reporting), estimating emissions from lost CKD based on a default value can be considered good
practice. The amount of C02 from lost CKD can vary but range typically from about 1.5 percent (additional C02 relative to that
calculated for clinker) for a modern plant to about 20 percent for a plant losing a lot of highly calcinated CKD (van Oss 2005). In
the absence of data, the default CKD correction factor (CFckd) is 1.02 (i.e., add 2 percent to the CO2 calculated for clinker). If no
calcined CKD is believed to be lost to the system, the CKD correction factor will be 1.00 (van Oss 2005)..."
4-12 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
1990 through 2013 and GHGRP starting in 2014. Consistent with the 2006IPCC Guidelines, the overlap technique
was applied to compare the two data sets for years where there was overlap, with findings that the data sets were
consistent and adjustments were not needed.
Uncertainty
The uncertainties contained in these estimates are primarily due to uncertainties in the lime content of clinker and
in the percentage of CKD recycled inside the cement kiln. Uncertainty is also associated with the assumption that
all calcium-containing raw materials are CaCC>3, when a small percentage likely consists of other carbonate and
non-carbonate raw materials. The lime content of clinker varies from 60 to 67 percent; 65 percent is used as a
representative value (Van Oss 2013a). This contributes to the uncertainty surrounding the emission factor for
clinker which has an uncertainty range of ±3 percent with uniform densities (Van Oss 2013b). The amount of CO2
from CKD loss can range from 1.5 to 8 percent depending upon plant specifications, and uncertainty was estimated
at ±5 percent with uniform densities (Van Oss 2013b). Additionally, some amount of CO2 is reabsorbed when the
cement is used for construction. As cement reacts with water, alkaline substances such as calcium hydroxide are
formed. During this curing process, these compounds may react with CO2 in the atmosphere to create calcium
carbonate. This reaction only occurs in roughly the outer 0.2 inches of the total thickness. Because the amount of
CO2 reabsorbed is thought to be minimal, it was not estimated. EPA assigned uncertainty bounds of ±3 percent and
a normal probability density function for clinker production and uncertainty bounds of ±5 percent and a uniform
probability density function for the emission factor, based on expert judgment (Van Oss 2013b).
The results of the Approach 2 quantitative uncertainty analysis are summarized in Table 4-6. Based on the
uncertainties associated with total U.S. clinker production, the CO2 emission factor for clinker production, and the
emission factor for additional CO2 emissions from CKD, 2022 CO2 emissions from cement production were
estimated to be between 40.1 and 43.8 MMT CO2 Eq. at the 95 percent confidence level. This confidence level
indicates a range of approximately 4 percent below and 5 percent above the emission estimate of 41.9 MMT CO2
Eq.
Table 4-6: Approach 2 Quantitative Uncertainty Estimates for CO2 Emissions from Cement
Production (MMT CO2 Eq. and Percent)
Source
Gas
2022 Emission Estimate
Uncertainty Range Relative to Emission Estimate ¦'
(MMTCO. Eq.)
(MMTCO. Eq.)
(%)
Lower Upper
Lower Upper
Bound Bound
Bound Bound
Cement Production
C02
41.9
40.1 43.8
-4% +5%
a Range of emission estimates predicted by Monte Carlo stochastic simulation for a 95 percent confidence interval.
QA/QC and Verification
General quality assurance/quality control (QA/QC) procedures were applied consistent with the U.S. Inventory
QA/QC plan, which is in accordance with Volume 1, Chapter 6 of 2006 IPCC Guidelines as described in the
introduction of the IPPU chapter (see Annex 8 for more details).
EPA relied upon the latest guidance from the IPCC on the use of facility-level data in national inventories and
applied a category-specific QC process to compare activity data from EPA's GHGRP with existing data from USGS
surveys. This was to ensure time-series consistency of the emission estimates presented in the Inventory. Total U.S.
clinker production is assumed to have low uncertainty because facilities routinely measure this for economic
reasons and because both USGS and GHGRP take multiple steps to ensure that reported totals are accurate. EPA
verifies annual facility-level GHGRP reports through a multi-step process that is tailored to the reporting industry
(e.g., combination of electronic checks including range checks, statistical checks, algorithm checks, year-to-year
Industrial Processes and Product Use 4-13
-------
comparison checks, along with manual reviews involving outside data checks) to identify potential errors and
ensure that data submitted to EPA are accurate, complete, and consistent (EPA 2015). Based on the results of the
verification process, EPA follows up with facilities to resolve mistakes that may have occurred.10 Facilities are also
required to monitor and maintain records of monthly clinker production per section 98.84 of the GHGRP regulation
(40 CFR 98.84).
EPA's GHGRP requires all facilities producing portland cement to report greenhouse gas emissions, including CO2
process emissions from each kiln, CO2 combustion emissions from each kiln, CH4 and N2O combustion emissions
from each kiln, and CO2, Cm, and N2O emissions from each stationary combustion unit other than kilns (40 CFR
Part 98 Subpart H). Source-specific quality control measures for the cement production category are included in
section 98.84, Monitoring and QA/QC Requirements.
As mentioned above, EPA compares GHGRP clinker production data (EPA 2023) to the USGS clinker production
data (USGS 2023a; USGS 2023c). For the year 2014, 2020, and 2022, USGS and GHGRP clinker production data
showed a difference of approximately 1 percent. In 2018, the difference between USGS and GHGRP clinker
production data was approximately 3 percent, which resulted in a difference in emissions of about 1.2 MMT CO2
Eq. In 2015, 2016, 2017, 2019, and 2021, that difference was less than 0.5 percent (less than 0.2 MMT CO2 Eq.)
between the two sets of activity data. The information collected by the USGS National Minerals Information Center
surveys continue to be an important data source.
Recalculations Discussion
No recalculations were performed for the 1990 through 2021 portion of the time series.
Planned Improvements
EPA is continuing to evaluate and analyze data reported under EPA's GHGRP that would be useful to improve the
emission estimates for the Cement Production source category. Most cement production facilities reporting under
EPA's GHGRP use Continuous Emission Monitoring Systems (CEMS) to monitor and report CO2 emissions, thus
reporting combined process and combustion emissions from kilns. In implementing further 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, in addition to category-specific QC methods recommended by the 2006 IPCC
Guidelines.11 EPA's long-term improvement plan includes continued assessment of the feasibility of using
additional GHGRP information beyond aggregation of reported facility-level clinker data, in particular
disaggregating the combined process and combustion emissions reported using CEMS, to separately present
national process and combustion emissions streams consistent with IPCC and UNFCCC guidelines. This long-term
planned analysis is still in development and has not been applied for this current Inventory.
EPA continues to review methods and data used to estimate CO2 emissions from cement production in order to
account for organic material in the raw material and to discuss the carbonation that occurs across the duration of
the cement product. Work includes identifying data and studies on the average carbon content for organic
materials in kiln feed in the United States and on CO2 reabsorption rates via carbonation for various cement
products. This information is not reported by facilities subject to GHGRP reporting. This is a long-term
improvement.
10 See GHGRP Verification Fact Sheet https://www.epa.gov/sites/production/files/2Q15-
07/documents/eherp verification factsheet.pdf.
11 See IPCC Technical Bulletin on Use of Facility-Specific Data in National Greenhouse Gas Inventories http://www.ipcc-
nggip.iges.or.jp/public/tb/TFl Technical Bulletin l.pdf and the 2019 Refinement, Volume 1, Chapter 2, Section 2.3, Use of
Facility Data in Inventories at https://www.ipcc-
nggip.iges.or.jp/public/2019rf/pdf/l Volumel/19R VI Ch02 DataCollection.pdf.
4-14 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
4.2 Lime Production (CRT Source Category
2A2)
Lime is a manufactured product with many industrial, chemical, and environmental applications. This reporting
category (2A2) includes process emissions from the production of lime. Per the IPCC methodological guidance,
emissions from fuels consumed for energy purposes during the production of lime are accounted for as part of
fossil fuel combustion in the industrial end-use sector reported under the Energy chapter.
Lime production involves three main processes: stone preparation, calcination, and hydration. Carbon dioxide
(CO2) is generated during the calcination stage, when limestone—consisting of calcium carbonate (CaCOs) and/or
magnesium carbonate (MgCOs)—is roasted at high temperatures in a kiln to produce calcium oxide (CaO) and CO2.
The CO2 is given off as a gas and is normally emitted to the atmosphere.
CaCO3 —> CaO + C02
Some facilities, however, recover CO2 generated during the production process for use in sugar refining and
precipitated calcium carbonate (PCC) production.12 PCC is used as a filler or coating in the paper, food, and plastic
industries and is derived from reacting hydrated high-calcium quicklime with CO2, a production process that does
not result in net emissions of CO2 to the atmosphere.
For U.S. operations, the term "lime" refers to a variety of chemical compounds. These include CaO, or high-calcium
quicklime; calcium hydroxide (Ca(OH)2), or hydrated lime; dolomitic quicklime ([CaOMgO]); and dolomitic hydrate
([Ca(OH)2*MgO] or [Ca(OH)2»Mg(OH)2]).
The current lime market is approximately distributed across six end-use categories, as follows: metallurgical uses,
35 percent; environmental uses, 29 percent; chemical and industrial uses, 21 percent; construction uses, 10
percent; miscellaneous uses, 3 percent; and refractory dolomite, 1 percent (USGS 2021). The major uses are in
steel making, chemical and industrial applications (such as the manufacture of fertilizer, glass, paper and pulp, and
precipitated calcium carbonate, and in sugar refining), flue gas desulfurization (FGD) systems at coal-fired electric
power plants, construction, and water treatment, as well as uses in mining, pulp and paper and precipitated
calcium carbonate manufacturing (USGS 2023a). Lime is also used as a CO2 scrubber, and there has been
experimentation on the use of lime to capture CO2 from electric power plants. Both lime (CaO) and limestone
(CaCOs) can be used as a sorbent for FGD systems. Emissions from limestone consumption for FGD systems are
reported under Section 4.4 Other Process Uses of Carbonate Production (CRT Source Category 2A4).
Emissions from lime production have fluctuated over the time series depending on lime end-use markets -
primarily the steel making industry and FGD systems for utility and industrial plants - and also energy costs. One
significant change to lime end-use since 1990 has been the increase in demand for lime for FGD at coal-fired
electric power plants, which can be attributed to compliance with sulfur dioxide (SO2) emission regulations of the
Clean Air Act Amendments of 1990. Phase I went into effect on January 1,1995, followed by Phase II on January 1,
2000. To supply lime for the FGD market, the lime industry installed more than 1.8 million tons per year of new
capacity by the end of 1995 (USGS 2023a). The need for air pollution controls continued to drive the FGD lime
market, which had doubled between 1990 and 2019 (2021, 2023a, 2023b).
The U.S. lime industry temporarily shut down some individual gas-fired kilns and, in some case, entire lime plants
during 2000 and 2001, due to significant increases in the price of natural gas. Lime production continued to
12 The amount of C02 captured from lime production for sugar refining and PCC production is reported under CRT Source
Category 2H3 "Other", but within this report, they are included in this chapter.
Industrial Processes and Product Use 4-15
-------
decrease in 2001 and 2002, a result of lower demand from the steel making industry, lime's largest end-use
market, when domestic steel producers were affected by low priced imports and slowing demand (USGS 2023a).
Emissions from lime production peaked in 2006 at approximately 30.3 percent above 1990 levels, due to strong
demand from the steel and construction markets (road and highway construction projects), before dropping to its
second lowest level in 2009 at approximately 2.5 percent below 1990 emissions, driven by the economic recession
and downturn in major markets including construction, mining, and steel (USGS 2023a). In 2010, the lime industry
began to recover as the steel, FGD, and construction markets also recovered (USGS 2023a). Fluctuation in lime
production since 2015 has been driven largely by demand from the steel making industry (USGS 2021). In 2020, a
significant decline in lime production occurred due to plants temporarily closing as a result of the global COVID-19
pandemic (USGS 2023a). This resulted in the lowest level of emissions in 2020 at approximately 3.4 percent below
1990 emissions. Emissions increased annually since then, with 2022 levels similar to emissions in 2019.
Lime production in the United States—including Puerto Rico—was reported to be 16,994 kilotons in 2022, an
increase of about 1.3 percent compared to 2021 levels (USGS 2023a). Compared to 1990, lime production
increased by about 7.3 percent. At year-end 2022, 73 primary lime plants were operating in the United States,
including Puerto Rico (USGS 2023a).13 Principal lime producing states were, in alphabetical order, Kentucky,
Missouri, Ohio, and Texas (USGS 2023a).
U.S. lime production resulted in estimated net CO2 emissions of 12.2 MMT CO2 Eq. (12,208 kt) (see Table 4-7 and
Table 4-8). Carbon dioxide emissions from lime production increased by about 2.8 percent compared to 2021
levels. Compared to 1990, CO2 emissions have increased by about 4.3 percent. The trends in CO2 emissions from
lime production are directly proportional to trends in production, which are described above.
Table 4-7: CO2 Emissions from Lime Production (MMT CO2 Eq.)
Year
1990
2005
2018
2019
2020
2021
2022
Lime Production
11.7
14.6 I
13.1
12.1
11.3
11.9
12.2
Table 4-8: Gross, Recovered, and Net CO2 Emissions from Lime Production (kt CO2)
Year
1990
2005
2018
2019
2020
2021
2022
Gross
11,959 |
15,074
13,609
12,676
11,875
12,586
12,750
Recovered3
259 1
522
503
564
576
716
542
Net Emissions
11,700
14,552
13,106
12,112
11,299
11,870
12,208
Note: Totals may not sum due to independent rounding.
a For sugar refining and PCC production.
Methodology and Time-Series Consistency
To calculate emissions, the amounts of high-calcium and dolomitic lime produced were multiplied by their
respective emission factors, consistent with Tier 2 methodology from the 2006IPCC Guidelines and in accordance
with the IPCC methodological decision tree and available data. The emission factor is the product of the
stoichiometric ratio between CO2 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:
13 In 2022, 68 operating primary lime facilities in the United States reported to the EPA Greenhouse Gas Reporting Program.
4-16 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
Equation 4-2:2006IPCC Guidelines Tier 2 Emission Factor for Lime Production, High-Calcium
Lime (Equation 2.9)
EFnigH-caicjumLime = [(44.01 ^C02) + (56.08^Cao)] x (o.9500^|) = 0.7455
Equation 4-3:2006 IPCC Guidelines Tier 2 Emission Factor for Lime Production, Dolomitic Lime
(Equation 2.9)
EFDComMcu™ = [(88.02^C02)+ (96.39^CaO • Mg0)[ x (0.9500 = 0.8675
Production was adjusted to remove the mass of chemically combined water found in hydrated lime, determined
according to the molecular weight ratios of H2O to (Ca(OH)2 and [Ca(OH)2*Mg(OH)2]) (IPCC 2006). These factors set
the chemically combined water content to 27 percent for high-calcium hydrated lime, and 30 percent for dolomitic
hydrated lime.
The 2006 IPCC Guidelines (Tier 2 method) also recommends accounting for emissions from lime kiln dust (LKD)
through application of a correction factor. LKD is a byproduct of the lime manufacturing process typically not
recycled back to kilns. LKD is a very fine-grained material and is especially useful for applications requiring very
small particle size. Most common LKD applications include soil reclamation and agriculture. Emissions from the
application of lime for agricultural purposes are reported in the Agriculture chapter under 5.5 Liming (CRT Source
Category 3G). Currently, data on annual LKD production is not readily available to develop a country-specific
correction factor. Lime emission estimates were multiplied by a factor of 1.02 to account for emissions from LKD
(IPCC 2006). See the Planned Improvements section associated with efforts to improve uncertainty analysis and
emission estimates associated with LKD.
Lime emission estimates were further adjusted to account for the amount of CO2 captured for use in on-site
processes. All the domestic lime facilities are required to report these data to EPA under its GHGRP. The total
national-level annual amount of CO2 captured for on-site process use was obtained from EPA's GHGRP (EPA 2023)
based on reported facility-level data for years 2010 through 2022. The amount of CO2 captured/recovered for non-
marketed on-site process use is deducted from the total gross emissions (i.e., from lime production and LKD). The
net lime emissions are presented in Table 4-7 and Table 4-8. GHGRP data on CO2 removals (i.e., CO2
captured/recovered) was available only for 2010 through 2022. Since GHGRP data are not available for 1990
through 2009, IPCC "splicing" techniques were used as per the 2006 IPCC Guidelines on time-series consistency
(IPCC 2006, Volume 1, Chapter 5).
Lime production data (i.e., lime sold and non-marketed lime used by the producer) by type (i.e., high-calcium and
dolomitic quicklime, high-calcium and dolomitic hydrated lime, and dead-burned dolomite) for 1990 through 2022
(see Table 4-9) were obtained from U.S. Geological Survey (USGS) Minerals Yearbook (USGS 2023a) and are
compiled by USGS to the nearest ton. Dead-burned dolomite data are additionally rounded by USGS to no more
than one significant digit to avoid disclosing company proprietary data. Production data for the individual
quicklime (i.e., high-calcium and dolomitic) and hydrated lime (i.e., high-calcium and dolomitic) types were not
provided prior to 1997. These were calculated based on total quicklime and hydrated lime production data from
1990 through 1996 and the three-year average ratio of the individual lime types from 1997 to 1999. Natural
hydraulic lime, which is produced from CaO and hydraulic calcium silicates, is not manufactured in the United
States (USGS 2023a). Total lime production was adjusted to account for the water content of hydrated lime by
converting hydrate to oxide equivalent based on recommendations from the IPCC and using the water content
values for high-calcium hydrated lime and dolomitic hydrated lime mentioned above, and is presented in Table
4-10 (IPCC 2006). The CaO and CaOMgO contents of lime, both 95 percent, were obtained from the IPCC (IPCC
2006).
Industrial Processes and Product Use 4-17
-------
Table 4-9: High-Calcium- and Dolomitic-Quicklime, High-Calcium- and Dolomitic-Hydrated,
and Dead-Burned-Dolomite Lime Production (kt)
Year
1990
2005
2018
2019
2020
2021
2022
High-Calcium Quicklime
11,166
14,100
12,400
11,300
10,700
11,200
11,500
Dolomitic Quicklime
2,234
2,990 1
2,810
2,700
2,390
2,700
2,640
High-Calcium Hydrated
1,781
2,220
2,430
2,430
2,320
2,430
2,410
Dolomitic Hydrated
319
474
265
267
252
244
244
Dead-Burned Dolomite
342
200
200
200
200
200
200
Table 4-10: Adjusted
Lime Production (kt)
Year
1990
2005
2018
2019
2020
2021
2022
High-Calcium
12,466
15,721
14,174
13,074
12,394
12,974
13,259
Dolomitic
2,800 1
3,522 1
3,196
3,087
2,766
3,071
3,011
Note: Minus water content of hydrated lime.
Methodological approaches were applied to the entire time series to ensure consistency in emissions from 1990
through 2022.
Uncertainty
The uncertainties contained in these estimates can be attributed to slight differences in the chemical composition
of lime products and CO2 recovery rates for on-site process use over the time series. Although the methodology
accounts for various formulations of lime, it does not account for the trace impurities found in lime, such as iron
oxide, alumina, and silica. Due to differences in the limestone used as a raw material, a rigid specification of lime
material is impossible. As a result, few plants produce lime with exactly the same properties.
In addition, a portion of the CO2 emitted during lime production will actually be reabsorbed when the lime is
consumed, especially at captive lime production facilities. As noted above, lime has many different chemical,
industrial, environmental, and construction applications. In many processes, CO2 reacts with the lime to create
calcium carbonate (e.g., water softening). Carbon dioxide reabsorption rates vary, however, depending on the
application. For example, 100 percent of the lime used to produce precipitated calcium carbonate reacts with CO2,
whereas most of the lime used in steel making reacts with impurities such as silica, sulfur, and aluminum
compounds. Quantifying the amount of CO2 that is reabsorbed would require a detailed accounting of lime use in
the United States and additional information about the associated processes where both the lime and byproduct
CO2 are "reused." Research conducted thus far has not yielded the necessary information to quantify CO2
reabsorption rates.14 Some additional information on the amount of CO2 consumed on site at lime facilities,
however, has been obtained from EPA's GHGRP.
In some cases, lime is generated from calcium carbonate byproducts at pulp mills and water treatment plants.15
The lime generated by these processes is included in the USGS data for commercial lime consumption. In the
pulping industry, mostly using the Kraft (sulfate) pulping process, lime is consumed in order to causticize a process
liquor (green liquor) composed of sodium carbonate and sodium sulfide. The green liquor results from the dilution
of the smelt created by combustion of the black liquor where biogenic carbon (C) is present from the wood. Kraft
14 Representatives of the National Lime Association estimate that C02 reabsorption that occurs from the use of lime may offset
as much as a quarter of the C02 emissions from calcination (Males 2003).
15 Some carbide producers may also regenerate lime from their calcium hydroxide byproducts, which does not result in
emissions of C02. 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 + 2H20 -> 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 + H20],
and no C02 is released.
4-18 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
mills recover the calcium carbonate "mud" after the causticizing operation and calcine it back into lime—thereby
generating CO2—for reuse in the pulping process. Although this re-generation of lime could be considered a lime
manufacturing process, the CO2 emitted during this process is mostly biogenic in origin and therefore is not
included in the industrial processes totals (Miner and Upton 2002). In accordance with IPCC methodological
guidelines, any such emissions are calculated by accounting for net carbon fluxes from changes in biogenic carbon
reservoirs in wooded or crop lands (see the Land Use, Land-Use Change, and Forestry chapter).
In the case of water treatment plants, lime is used in the softening process. Some large water treatment plants
may recover their waste calcium carbonate and calcine it into quicklime for reuse in the softening process. Further
research is necessary to determine the degree to which lime recycling is practiced by water treatment plants in the
United States.
Another uncertainty is the assumption that calcination emissions for LKD are around 2 percent. EPA assigned
uncertainty ranges of ±2 percent and a triangular probability density function for the LKD correction factor based
on expert judgment (RTI 2023). The National Lime Association (NLA) has commented that the estimates of
emissions from LKD in the United States could be closer to 6 percent. They also note that additional emissions
(approximately 2 percent) may also be generated through production of other byproducts/wastes (off-spec lime
that is not recycled, scrubber sludge) at lime plants (Seeger 2013). Publicly available data on LKD generation rates,
total quantities not used in cement production, and types of other byproducts/wastes produced at lime facilities
are limited. NLA compiled and shared historical emissions information and quantities for some waste products
reported by member facilities associated with generation of total calcined byproducts and LKD, as well as
methodology and calculation worksheets that member facilities complete when reporting. There is uncertainty
regarding the availability of data across the time series needed to generate a representative country-specific LKD
factor. Uncertainty of the activity data is also a function of the reliability and completeness of voluntarily reported
plant-level production data. EPA assigned uncertainty ranges of ±1 percent for lime production and a normal
probability density function, based on expert judgment (USGS 2012). Further research, including discussion with
NLA, and data is needed to improve understanding of additional calcination emissions to consider revising the
current assumptions that are based on the 2006 IPCC Guidelines. More information can be found in the Planned
Improvements section below.
The results of the Approach 2 quantitative uncertainty analysis are summarized in Table 4-11. Lime CO2 emissions
for 2022 were estimated to be between 12.1 and 12.3 MMT CO2 Eq. at the 95 percent confidence level. This
confidence level indicates a range of approximately 1 percent below and 1 percent above the emission estimate of
12.2 MMT CO2 Eq.
Table 4-11: Approach 2 Quantitative Uncertainty Estimates for CO2 Emissions from Lime
Production (MMT CO2 Eq. and Percent)
Source
Gas
2022 Emission Estimate
(MMT CO.. Eq.)
Uncertainty Range Relative to Emission Estimate-1
(MMTCO' Eq.) (%)
Lower Upper
Lower
Upper
Bound Bound
Bound
Bound
Lime Production
C02
12.2
12.1 12.3
-1%
+1%
a Range of emission estimates predicted by Monte Carlo stochastic simulation for a 95 percent confidence interval.
QA/QC and Verification
General quality assurance/quality control (QA/QC) procedures were applied consistent with the U.S. Inventory
QA/QC plan, which is in accordance with Volume 1, Chapter 6 of 2006 IPCC Guidelines as noted in the introduction
of the IPPU chapter (see Annex 8 for more details).
More details on the greenhouse gas calculation, monitoring and QA/QC methods associated with reporting on CO2
captured for onsite use applicable to lime manufacturing facilities can be found under Subpart S (lime
Industrial Processes and Product Use 4-19
-------
manufacturing) of the GHGRP regulation (40 CFR Part 98).16EPA verifies annual facility-level GHGRP reports
through a multi-step process (e.g., combination of electronic checks and manual reviews) to identify potential
errors and ensure that data submitted to EPA are accurate, complete, and consistent (EPA 2023).17 Based on the
results of the verification process, EPA follows up with facilities to resolve mistakes that may have occurred. The
post-submittals checks are consistent with a number of general and category-specific QC procedures, including:
range checks, statistical checks, algorithm checks, and year-to-year checks of reported data and emissions.
Recalculations Discussion
No recalculations were performed for the 1990 through 2021 portion of the time series.
Planned Improvements
EPA plans to review GHGRP emissions and activity data reported to EPA under Subpart S of the GHGRP regulation
(40 CFR Part 98), and aggregated activity data on lime production by type in particular. In addition, initial review of
data has identified that several facilities use CEMS to report emissions. Under Subpart S, if a facility is using a
CEMS, they are required to report combined combustion emissions and process emissions. EPA continues to
review how best to incorporate GHGRP and notes that particular attention will be made to also ensuring time-
series consistency of the emissions estimates presented in future Inventory reports, consistent with IPCC and
UNFCCC guidelines. This is required because the facility-level reporting data from EPA's GHGRP, with the program's
initial requirements for reporting of emissions in calendar year 2010, are not available for all inventory years (i.e.,
1990 through 2009) as required for this Inventory. In 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.18
Future improvements involve improving and/or confirming the representativeness of current assumptions
associated with emissions from production of LKD and other byproducts/wastes as discussed in the Uncertainty
section, per comments from the NLA provided during a prior Public Review comment period for a previous (1990
through 2018) Inventory. EPA met with NLA in summer of 2020 for clarification on data needs and available data
and to discuss planned research into GHGRP data. Previously, EPA met with NLA in spring of 2015 to outline
specific information required to apply IPCC methods to develop a country-specific correction factor to more
accurately estimate emissions from production of LKD. In 2016, NLA compiled and shared historical emissions
information reported by member facilities on an annual basis under voluntary reporting initiatives from 2002
through 2011 associated with generation of total calcined byproducts and LKD. Reporting of LKD was only
differentiated for the years 2010 and 2011. This emissions information was reported on a voluntary basis
consistent with NLA's facility-level reporting protocol, which was also provided to EPA. To reflect information
provided by NLA, EPA updated the qualitative description of uncertainty. At the time of this Inventory, this planned
improvement is in process and has not been incorporated into this current Inventory report.
16 See http://www.ecfr.gov/cgi-bin/text-idx7tpN/ecfrbrowse/Title40/40cfr98 main Q2.tpl.
17 See https://www.epa.gov/sites/production/files/2015-07/documents/ghgrp verification factsheet.pdf.
18 See http://www.ipcc-nggip.iges.or.jp/public/tb/TFI Technical Bulletin l.pdf and the 2019 Refinement, Volume 1, Chapter 2,
Section 2.3, Use of Facility Data in Inventories at https://www.ipcc-
nggip.iges.or.jp/public/2019rf/pdf/l Volumel/19R VI Ch02 DataCollection.pdf.
4-20 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
4.3 Glass Production (CRT Source Category
2A3)
Glass production is an energy and raw-material intensive process that results in the generation of carbon dioxide
(CO2) from both the energy consumed in making glass and the glass production process itself. This reporting
category (2A3) includes emissions from the production of glass. Emissions from fuels consumed for energy
purposes during the production of glass are accounted for as part of fossil fuel combustion in the industrial end-
use sector reported under the Energy chapter.
Glass production employs a variety of raw materials in a glass-batch. These include formers, fluxes, stabilizers, and
sometimes colorants. The major raw materials (i.e., fluxes and stabilizers) that emit process-related CO2 emissions
during the glass melting process are limestone, dolomite, and soda ash. The main former in all types of glass is
silica (SiCh). Other major formers in glass include feldspar and boric acid (i.e., borax). Fluxes are added to lower the
temperature at which the batch melts. Most commonly used flux materials are soda ash (sodium carbonate,
Na2CC>3) and potash (potassium carbonate, K2O). Stabilizers make glass more chemically stable and keep the
finished glass from dissolving and/or falling apart. Commonly used stabilizing agents in glass production are
limestone (CaCOs), dolomite (CaCOsMgCOs), alumina (AI2O3), magnesia (MgO), barium carbonate (BaCOs),
strontium carbonate (SrCOs), lithium carbonate (IJ2CO3), and zirconia (ZrCh) (DOE 2002). Glass makers also use a
certain amount of recycled scrap glass (cullet), which comes from in-house return of glassware broken in the
production process or other glass spillage or retention, such as recycling or from cullet broker services.
The raw materials (primarily soda ash, limestone, and dolomite) release CO2 emissions in a complex high-
temperature chemical reaction during the glass melting process. This process is not directly comparable to the
calcination process used in lime manufacturing, cement manufacturing, and process uses of carbonates (i.e.,
limestone/dolomite use) but has the same net effect in terms of generating process CO2 emissions (IPCC 2006).
The U.S. glass industry can be divided into four main categories: containers, flat (window) glass, fiber glass, and
specialty glass. The majority of commercial glass produced is container and flat glass (EPA 2009). The United States
is one of the major global exporters of glass. Domestically, demand comes mainly from the construction, auto,
bottling, and container industries. There are more than 1,700 facilities that manufacture glass in the United States,
with the largest companies being Corning, Guardian Industries, Owens-Illinois, and PPG Industries.19
The glass container sector is one of the leading soda ash consuming sectors in the United States. In 2022, glass
production accounted for 49 percent of total domestic soda ash consumption (USGS 2023). Emissions from soda
ash production are reported in Section 4.12.
In 2022, 2,250 kilotons of soda ash, 1,370 kilotons of limestone, 925 kilotons of dolomite, and 1.9 kilotons of other
carbonates were consumed for glass production (USGS 2023; EPA 2023). Use of soda ash, limestone, dolomite, and
other carbonates in glass production resulted in aggregate CO2 emissions of 2.0 MMT CO2 Eq. (1,956 kt), which are
summarized in Table 4-12 and Table 4-13. Overall, emissions have decreased by 14 percent compared to 1990.
Emissions decreased by 1 percent compared to 2021 levels.
Emissions from glass production have remained relatively consistent over the time series with some fluctuations
since 1990. In general, these fluctuations were related to the behavior of the export market and the U.S. economy.
Specifically, the extended downturn in residential and commercial construction and automotive industries
between 2008 and 2010 resulted in reduced consumption of glass products, causing a drop in global demand for
limestone, dolomite, and soda ash and resulting in lower emissions. Some commercial food and beverage package
19 Excerpt from Glass & Glass Product Manufacturing Industry Profile, First Research. Available online at:
http://www.firstresearch.com/lndustrv-Research/Glass-and-Glass-Product-Manufacturing.html.
Industrial Processes and Product Use 4-21
-------
manufacturers are shifting from glass containers towards lighter and more cost-effective polyethylene
terephthalate (PET) based containers, putting downward pressure on domestic consumption of soda ash (USGS
1995 through 2015b). Glass production in 2022 was steady, changing by no more than 2 percent over the course of
the year (Federal Reserve 2023).
Table 4-12: CO2 Emissions from Glass Production (MMT CO2 Eq.)
Year 1990 2005 2018 2019 2020 2021
2022
Glass Production 2.3 2.4 2.0 1.9 1.9 2.0
2.0
Table 4-13: CO2 Emissions from Glass Production (kt CO2)
Year 1990 2005 2018 2019 2020 2021
2022
Glass Production 2,263 | 2,402 | 1,989 1,940 1,858 1,969
1,956
Methodology and Time-Series Consistency
Carbon dioxide emissions were calculated based on Tier 3 method from the 2006IPCC Guidelines, in accordance
with the IPCC methodological decision tree and available data, by multiplying the quantity of input carbonates (i.e.,
limestone, dolomite, soda ash, and other carbonates) by the carbonate-based emission factor (in metric tons
CCh/metric ton carbonate) and the average carbonate-based mineral mass fraction.
2010 through 2022
The methodology for estimating CO2 emissions from glass production for years 2010 through 2022 used the
quantities of limestone, dolomite, and a group of other carbonates (i.e., barium carbonate, potassium carbonate,
lithium carbonate, and strontium carbonate) used for glass production, obtained from GHGRP (EPA 2023). USGS
data on the quantity of soda ash used for glass production was used because it was obtained directly from the
soda ash producers and includes use by smaller artisanal glass operations, which are excluded in the GHGRP data.
GHGRP collects data from glass production facilities with greenhouse gas emissions greater than 25,000 metric
tons CO2 Eq. The reporting threshold is used to exclude artisanal glass operations that are expected to have much
lower greenhouse gas emissions than the threshold. These smaller facilities have not been accounted for yet for
this portion of the time series for limestone, dolomite, or other carbonates due to limited data. Facilities report the
total quantity of each type of carbonate used in glass production each year to GHGRP, with data collection starting
in 2010 (EPA 2023).
Using the total quantities of each carbonate, EPA calculated the metric tons of emissions resulting from glass
production by multiplying the quantity of input carbonates (i.e., limestone, dolomite, soda ash, and other
carbonates) by carbonate-based emission factors in metric tons CCh/metric ton carbonate (limestone, 0.43971;
dolomite, 0.47732; soda ash, 0.41492; and other carbonates, 0.262), and by the average carbonate-based mineral
mass fraction for each year. IPCC default emission factors were used for limestone, dolomite, and soda ash, and
the emission factor for other carbonates is based on expert judgment (RTI 2022).
1990 through 2009
Data from GHGRP on the quantity of limestone, dolomite, and other carbonates used in glass production are not
available for 1990 through 2009. Additionally, USGS does not collect data on the quantity of other carbonates used
for glass production.
To address time-series consistency, total emissions from 1990 to 2009 were calculated using the Federal Reserve
Industrial Production Index for glass production in the United States as a surrogate for the total quantity of
carbonates used in glass production. The production index measures real output expressed as a percentage of real
output in a base year, which is currently 2017 (Federal Reserve 2023). Since January 1971, the Federal Reserve has
released the monthly glass production index for NAICS code 3272 (Glass and Glass Product Manufacturing) as part
4-22 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
of release G.17, "Industrial Production and Capacity Utilization" (Federal Reserve 2023). The monthly index values
for each year were averaged to calculate an average annual glass production index value. Total annual process
emissions were calculated by taking a ratio of the average annual glass production index for each year to the
average annual glass production index for base year 2017, and multiplying by the calculated 2017 emissions
(process-related) based on GHGRP data.
Emissions from limestone, dolomite, and other carbonate consumption were disaggregated from total annual
emissions, using the average percent contribution of each to annual emissions from these three carbonates for
2010 through 2014 based on GHGRP data: 64.5 percent limestone, 35.5 percent dolomite, and 0.1 percent other
carbonates.
The methodology for estimating CO2 emissions from the use of soda ash for glass production and data sources for
the amount of soda ash used in glass production are consistent with the methodology used for 2010 through 2022.
The average mineral mass fractions for soda ash are only available starting in 2010. The average carbonate-based
mineral mass fractions from the GHGRP, averaged across 2010 through 2014, indicate that soda ash contained
98.7 percent sodium carbonate (Na2COs). This averaged value is used to estimate emissions for 1990 through 2009.
The years 2010 to 2014 were used to determine the average carbonate-based mineral mass fractions because
those years were deemed to better represent historic glass production from 1990 to 2009.
Data on soda ash used for glass production for 1990 through 2022 were obtained from the U.S. Bureau of Mines
(1991 and 1993a), the USGS Minerals Yearbook: Soda Ash (USGS 1995 through 2015b), and USGS Mineral Industry
Surveys for Soda Ash (USGS 2017 through 2023). Data on limestone, dolomite, and other carbonates used for glass
production and on average carbonate-based mineral mass fraction for 2010 through 2022 were obtained from
GHGRP (EPA 2023). The quantities of limestone, dolomite, and other carbonates were calculated for 1990 through
2009 using the Federal Reserve Industrial Production Index (Federal Reserve 2023).
The amount of limestone, dolomite, soda ash, and other carbonates used in glass production each year and the
annual average Federal Reserve production indices for glass production are shown in Table 4-14.
Table 4-14: Limestone, Dolomite, Soda Ash, and Other Carbonates Used in Glass Production
(kt) and Average Annual Production Index for Glass and Glass Product Manufacturing
Activity
1990
2005
2018
2019
2020
2021
2022
Limestone
1,4091
1,690 J
1,442
1,370
1,334
1,397
1,370
Dolomite
714
8571
871
883
824
893
925
Soda Ash
3,177
3,050 J
2,280
2,220
2,130
2,280
2,250
Other Carbonates
2
a ¦
2
2
2
2
1.9
Total
5,302
5,599
4,596
4,475
4,289
4,572
4,547
Production Index3
94.31
H3.11
102.5
99.8
92.4
88.3
86.8
a Average Annual Production Index uses 2017 as the base year.
Note: Totals may not sum due to independent rounding.
As discussed above, methodological approaches were applied to the entire time series to ensure consistency in
emissions from 1990 through 2022. Consistent with the 2006IPCC Guidelines, the overlap technique was applied
to compare USGS and GHGRP data sets for 2010 through 2022. To address the inconsistencies, adjustments were
made as described above.
Uncertainty
The methodology in this Inventory report uses GHGRP data for the average mass fraction of each mineral used in
glass production. These minerals are limestone, dolomite, soda ash, and other carbonates (barium carbonate
(BaCC>3), potassium carbonate (K2CO3), lithium carbonate (U2CO3), and strontium carbonate (SrCOs)). The mass
fractions are reported directly by the glass manufacturers, for each year from 2010 to 2022.
Industrial Processes and Product Use 4-23
-------
The methodology uses the quantities of limestone, dolomite, and other carbonates used in glass manufacturing
which is reported directly by the glass manufacturers for years 2010 through 2022 and the amount of soda ash
used in glass manufacturing which is reported by soda ash producers for the full time series. EPA assigned an
uncertainty range of ±5 percent and a normal probability density function for all carbonate quantities and the
Federal Reserve Industrial Production Index for glass production, and using this suggested uncertainty provided in
Section 2.4.2.2 of the 2006IPCC Guidelines is appropriate based on expert judgment (RTI 2023). EPA assigned an
uncertainty range of ±2 percent for the carbonate emission factors, ±2 percent for the mineral mass fractions, and
±1 percent for the calcination fraction, and using this suggested uncertainty provided in Section 2.4.2.1 of the 2006
IPCC Guidelines is appropriate based on expert judgment (RTI 2023). Per this expert judgment, a triangular
probability density function was assigned for emission factors, mineral mass fractions, and calcination fraction.
The results of the Approach 2 quantitative uncertainty analysis are summarized in Table 4-15. In 2022, glass
production CO2 emissions were estimated to be between 1.9 and 2.0 MMT CO2 Eq. at the 95 percent confidence
level. This indicates a range of approximately 2 percent below and 2 percent above the emission estimate of 2.0
MMTCCh Eq.
Table 4-15: Approach 2 Quantitative Uncertainty Estimates for CO2 Emissions from Glass
Production (MMT CO2 Eq. and Percent)
Source
Gas
2022 Emission Estimate
(MMTCO. Eq.)
Uncertainty Range Relative to Emission Estimate-'
(MMTCO' Eq.) (%)
Lower Upper
Bound Bound
Lower
Bound
Upper
Bound
Glass Production
C02
2.0
1.9 2.0
-2%
+2%
a Range of emission estimates predicted by Monte Carlo stochastic simulation for a 95 percent confidence interval.
QA/QC and Verification
General quality assurance/quality control (QA/QC) procedures were applied consistent with the U.S. Inventory
QA/QC plan, which is in accordance with Volume 1, Chapter 6 of 2006 IPCC Guidelines as described in the
introduction of the IPPU chapter (see Annex 8 for more details). For the GHGRP data, EPA verifies annual facility-
level reports through a multi-step process (e.g., including a combination of pre-and post-submittal electronic
checks and manual reviews by staff) to identify potential errors and ensure that data submitted to EPA are
accurate, complete, and consistent (EPA 2015).20 Based on the results of the verification process, EPA follows up
with facilities to resolve mistakes that may have occurred. The post-submittals checks are consistent with a
number of general and category-specific QC procedures, including: range checks, statistical checks, algorithm
checks, and year-to-year checks of reported data and emissions.
Recalculations Discussion
During annual QC, a transcription error for the 1990 value of CO2 emissions from glass production was identified
and corrected in Table 4-12 and Table 4-13. No recalculations were needed or performed due to this transcription
error, and no other recalculations were performed for the 1990 through 2021 portion of the time series.
20 GHGRP Report Verification Factsheet. See https://www.epa.gov/sites/production/files/2Q15-
07/documents/ehgrp verification factsheet.pdf.
4-24 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
Planned Improvements
EPA plans to evaluate updates to uncertainty levels for the activity data and mineral mass fraction values from
EPA's GHGRP. This is a near-term planned improvement that is anticipated for inclusion in 2025 report.
Some glass producing facilities in the United States do not report to EPA's GHGRP because they fall below the
reporting threshold for this industry. EPA will continue ongoing research on the availability of data to better assess
the completeness of emission estimates from glass production and how to refine the methodology to ensure
complete national coverage of this category. When reporting began in 2010, EPA received data from more facilities
that were above the reporting threshold than expected, and total emissions for these reporting facilities were
higher than expected for all glass production facilities in the United States (EPA 2009). Research will include
reassessing previous assessments of GHGRP industry coverage using the reporting threshold of 25,000 metric tons
CO2 Eq. This is a medium-term planned improvement.
4.4 Other Process Uses of Carbonates (CRT
Source Category 2A4)
Limestone (CaCOs), dolomite (CaCOsMgCOs),21 and other carbonates such as soda ash, magnesite, and siderite are
basic materials used by a wide variety of industries, including construction, agriculture, chemical, metallurgy (i.e.,
iron and steel production, ferroalloy production, and magnesium production), glass production, environmental
pollution control, ceramics production, and non-metallurgical magnesia production. This reporting category (2A4)
includes emissions from other uses of limestone, dolomite, and other carbonates not included in other categories;
the production of ceramics; other uses of soda ash not included elsewhere; and the production of non-metallurgical
magnesia. This section addresses mineral industry use of these carbonates: limestone, dolomite, soda ash, and
magnesite. Emissions from the use of these carbonates are organized into four subcategories: other process uses of
carbonates (i.e., limestone and dolomite consumption), ceramics production, other uses of soda ash, and non-
metallurgical magnesia production.
For industrial applications, carbonates are heated sufficiently enough to calcine the material and generate CO2 as a
byproduct.
CaCO3 —> CaO + C02
MgC03 —> MgO + C02
Examples of such applications include limestone used as a flux or purifier in metallurgical furnaces, as a sorbent in
flue gas desulfurization (FGD) systems for utility and industrial plants, and as a raw material for the production of
glass, lime, and cement.
Emissions from limestone and dolomite used in the production of cement, lime, glass, and iron and steel are
excluded from the other process uses of carbonates category and reported under their respective source categories
(e.g., Section 4.3, Glass Production). Emissions from soda ash production are reported under Section 4.12, Soda Ash
Production (CRT Source Category 2B7). Emissions from soda ash consumption associated with glass manufacturing
are reported under Section 4.3, Glass Production (CRT Source Category 2A3). Emissions from the use of limestone
and dolomite in liming of agricultural soils are included in the Agriculture chapter under Section 5.5, Liming (CRT
Source Category 3G). Emissions from limestone and dolomite used in the production of iron and steel and
magnesium production are reported under Section 4.18, Iron and Steel Production (CRT Source Category 2C1).
21 Limestone and dolomite are collectively referred to as limestone by the industry, and intermediate varieties are seldom
distinguished.
Industrial Processes and Product Use 4-25
-------
Emissions from dolomite used in the production of magnesium are reported under Section 4.21, Magnesium
Production and Processing (CRT Source Category 2C4). As noted in Section 4.19, Ferroalloy Production (CRT Source
Category 2C2), emissions from the production of ferromanganese are not included in this Inventory because of the
small number of manufacturers of these materials in the United States. Government information disclosure rules
prevent the publication of production data for these production facilities. Emissions from fuels consumed for
energy purposes during these processes are accounted for as part of fossil fuel combustion in the industrial end-use
sector reported under the Energy chapter in Section 3.1, Fossil Fuel Combustion (CRT Source Category 1A). Both
lime (CaO) and limestone (CaCOs) can be used as a sorbent for FGD systems. Emissions from lime consumption for
FGD systems and from sugar refining are reported under Section 4.3, Lime Production (CRT Source Category 2A2).
Emissions from the use of dolomite in primary magnesium metal production are reported under Section 4.21,
Magnesium Production and Processing (CRT Source Category 2C4).
Limestone and dolomite are widely distributed throughout the world in deposits of varying sizes and degrees of
purity. Large deposits of limestone occur in nearly every state in the United States, and significant quantities are
extracted for industrial applications. In 2018, the leading limestone producing states were Texas, Florida, Ohio,
Missouri, and Pennsylvania, which contributed 46 percent of the total U.S. output (USGS 2022a). Dolomite deposits
are found in the United States, Canada, Mexico, Europe, Africa, and Brazil. In the United States, the leading dolomite
producing states are Pennsylvania, New York, and Utah which currently contribute more than a third of the total
U.S. output (USGS 2022a).
Ceramics include the production of bricks and roof tiles, vitrified clay pipes, refractory products, expanded clay
products, wall and floor tiles, table and ornamental ware (i.e., household ceramics), sanitary ware, technical
ceramics (e.g., aerospace, automotive, electronic, or biomedical applications), and inorganic bonded abrasives. Most
ceramic products are made from one or more different types of clay (e.g., shales, fire clay, and ball clay) with
varying carbonate contents. The process of manufacturing ceramic products, regardless of the product type or
scale, is essentially the same. This process consists of raw material processing (grinding, calcining, and drying),
forming (wet or dry process), firing (single or multiple stage firing process), and final processing. Process CO2
emissions are produced during the calcination process in the kiln or dryer, where carbonates are heated to high
temperatures which results in metal oxides and CO2. In 2018, the leading clay producing states were Georgia,
Wyoming, Texas, Alabama, and North Carolina, which contributed 60 percent of the total U.S. output (USGS 2022f).
Other uses of soda ash include the consumption of soda ash for non-glass purposes. Excluding glass production,
soda ash consumption by end use in 2022 included chemicals, 54 percent, soap and detergent manufacturing, 9
percent; distributers, 10 percent; flue gas desulfurization, 7 percent; other uses, 17 percent; pulp and paper
production, 1 percent; and water treatment, 2 percent (USGS 2023a). Chemicals produced using soda ash include
sodium-based inorganic chemicals such as sodium bicarbonate, sodium chromates, sodium phosphates, and sodium
silicates. (USGS 2022g). Internationally, two types of soda ash are produced: natural and synthetic. In 2019, 93
percent of the global soda ash production came from China, the United States, Russia, Germany, India, Turkey,
Poland, and France. The United States only produces natural soda ash and only in two states: Wyoming and
California (USGS 2021a).
Non-metallurgical magnesia production comprises of three categories of magnesia products: calcined magnesia,
deadburned magnesia, and fused magnesia. Magnesia is produced by calcining magnesite (MgCOs) which results in
the release of CO2. Non-metallurgical magnesia is used in agricultural, industrial, refractory, and electrical insulating
applications. Specific applications include fertilizers, construction materials, plastics, and flue gas desulphurization.
China, Russia, and Turkey account for 83 percent of global production capacity of magnesia from magnesite (USGS
2022e). In the United States, only one facility located in Nevada produces non-metallurgical magnesia using
magnesite as the raw material.
In 2022,18,671 kilotons (kt) of limestone, 2,052 kt of dolomite, 2,391 kt of soda ash, and 388 kt of magnesite were
consumed for these emissive applications, which excludes consumption for the production of cement, lime, glass,
and iron and steel (Willett 2023; USGS 2022b). Usage of limestone, dolomite, soda ash, and magnesite resulted in
aggregate CO2 emissions of 10.4 MMT CO2 Eq. (10,384 kt) (see Table 4-16 and Table 4-17). The 2022 emissions
increased 21 percent compared to 2021, primarily as a result of increased limestone consumption attributed to
4-26 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
sulfur oxide removal. Growth in the public and private construction markets contributed to an increase in
consumption of crushed stone in 2022. Overall emissions have increased 46 percent from 1990 through 2022.
Table 4-16: CO2 Emissions from Other Process Uses of Carbonates (MMT CO2 Eq.)
Year
1990
2005
2018
2019
2020
2021
2022
Other Uses of Carbonates
4.8
6.2 1
6.3
7.4
7.4
7.0
00
00
Ceramics Production
0.8
0.8 1
0.4
0.4
0.4
0.4
0.4
Other Uses of Soda Asha
1.4
1.3 1
1.1
1.0
1.0
1.0
1.0
Non-Metallurgical
lis
s
Magnesia Production
0.1 1
0.2
0.2
0.2
0.2
0.2
Total
7.1
8.5
7.9
9.0
9.0
8.6
10.4
a Soda ash consumption not associated with glass manufacturing.
Note: Totals may not sum due to independent rounding.
Table 4-17: CO2 Emissions from Other Process Uses of Carbonates (kt CO2)
Year
1990
2005
2018
2019
2020
2021
2022
Other Uses of Carbonates
4,843 1
6'155 S
6,283
7,386
7,441
6,972
8,781
Ceramics Production
757
822
418
399
397
400
407
Other Uses of Soda Asha
1,390 i
1,305 ;
1,069
1,036
958
979
992
Non-Metallurgical
Magnesia Production
113 "
191 -
169
152
216
231
204
Total
7,103
8,472
7,938
8,973
9,012
8,583
10,384
a Soda ash consumption not associated with glass manufacturing.
Note: Totals may not sum due to independent rounding.
Methodology and Time-Series Consistency
Other Uses of Carbonates (Limestone and Dolomite Consumption)
Carbon dioxide emissions from other uses of carbonates, specifically limestone and dolomite consumption, were
calculated using a Tier 2 method from the 2006IPCC Guidelines, in accordance with the IPCC methodological
decision tree and available data, by multiplying the quantity of limestone or dolomite consumed by the emission
factor for limestone or dolomite calcination, respectively: 0.43971 metric ton CCh/metric ton carbonate for
limestone and 0.47732 metric ton CCh/metric ton carbonate for dolomite.22 This methodology was used for
limestone and dolomite used for flux stone, flue gas desulfurization systems, chemical stone, mine dusting or acid
water treatment, and acid neutralization. Flux stone used during the production of iron and steel was deducted
from the other uses of carbonates source category estimate and attributed to the iron and steel production source
category estimate. Similarly, limestone and dolomite consumption for glass manufacturing, cement, and lime
manufacturing are excluded from this category and attributed to their respective categories.
Consumption data for 1990 through 2022 of limestone and dolomite used for flux stone, flue gas desulfurization
systems, chemical stone, mine dusting or acid water treatment, and acid neutralization (see Table 4-18) were
obtained from the U.S. Geological Survey (USGS) Minerals Yearbook: Crushed Stone Annual Report (1995a through
2023b), preliminary data for 2022 from USGS Crushed Stone Commodity Expert (Willett 2023), American Iron and
Steel Institute limestone and dolomite consumption data (AISI 2018 through 2021), and the U.S. Bureau of Mines
(1991 and 1993a), which are reported to the nearest ton. In addition, the estimated values for limestone and
dolomite consumption for flux stone used during the production of iron and steel were adjusted using emissions
data from the EPA's Greenhouse Gas Reporting Program (GHGRP) Subpart Q for the iron and steel sector to
account for the impacts of the COVID-19 pandemic in 2020 and 2021. Iron and steel GHGRP process emissions data
22 2006 IPCC Guidelines, Volume 3: Chapter 2, Table 2.1.
Industrial Processes and Product Use 4-27
-------
decreased by approximately 8 percent from 2021 to 2022 (EPA 2023). This adjustment method is consistent with
the method used in Section 4.18, Iron and Steel Production (CRT Source Category 2C1).
During 1990 and 1992, the USGS did not conduct a detailed survey of limestone and dolomite consumption by
end-use; therefore, data on consumption by end use for 1990 was estimated by applying the 1991 ratios of total
limestone and dolomite consumption by end use to total 1990 limestone and dolomite consumption values.
Similarly, the 1992 consumption figures were approximated by applying an average of the 1991 and 1993 ratios of
total limestone and dolomite use by end uses to the 1992 total values.
In 1991, the U.S. Bureau of Mines, now known as the USGS, began compiling production and end use information
through surveys of crushed stone manufacturers. Manufacturers provided different levels of detail in survey
responses, so information was divided into three categories: (1) production by end-use, as reported by
manufacturers (i.e., "specified" production); (2) production reported by manufacturers without end-uses specified
(i.e., "unspecified-reported" production); and (3) estimated additional production by manufacturers who did not
respond to the survey (i.e., "unspecified-estimated" production). 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; (2) the average percent of total limestone 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.
A large quantity of crushed stone was 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 all other reported end-uses according to each end-use's fraction of
total consumption in that year.23
Table 4-18: Limestone and Dolomite Consumption from Other Uses of Carbonates (kt)
Activity
1990
2005
2018
2019
2020
2021
2022
Limestone
10,0161
10,465
12,816
15,146
13,707
12,788
17,891
Dolomite
3,254
1,356
1,520
2,962
2,826
1,915
Total
10,935
13,719
14,172
16,667
16,669
15,614
19,806
Note: Totals may not sum due to independent rounding.
Ceramics Production
Carbon dioxide emissions from ceramics production were calculated using a Tier 1 method from the 2006IPCC
Guidelines, in accordance with the IPCC methodological decision tree and available data, by multiplying the
quantity of clay consumed for emissive purposes by a carbonate content value of clay of 10 percent, limestone
fraction of 85 percent and dolomite fraction of 15 percent, respectively, and by the emission factor for limestone
or dolomite calcination, respectively: 0.43971 metric ton CCh/metric ton of limestone and 0.47732 metric ton
CCh/metric ton of dolomite.24 To estimate annual process CO2 emissions, EPA evaluated the end-uses of each type
of clay published by USGS to identify the emissive end-uses that fall into the ceramics production subcategory. The
emissive end-uses were organized into three groups: ceramics, glass, and floor & tile; refractories; and heavy clay
products. The total limestone and dolomite consumption from the three emissive groupings for ceramics
production for 1990 through 2022 (see Table 4-19) were obtained from USGS (Simmons 2024).
23 This approach was recommended by USGS, the data collection agency.
24 2006 IPCC Guidelines, Volume 3: Chapter 2, Table 2.1.
4-28 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
Table 4-19: Limestone and Dolomite Consumption from Ceramics Production (kt)
Activity
1990
2005
2018
2019
2020
2021
2022
Limestone
1,4441
1,569
797
762
758
764
776
Dolomite
2551
277
141
135
134
135
137
Total
1,699
1,846
938
897
892
899
913
Note: Totals may not sum due to independent rounding.
Other Uses of Soda Ash
Carbon dioxide emissions from soda ash consumption were calculated using a Tier 1 method from the 2006IPCC
Guidelines, in accordance with the IPCC methodological decision tree and available data. Excluding glass
manufacturing which is reported under Section 4.3 Glass Production (CRT Source Category 2A3), most soda ash is
consumed in chemical production, with smaller amounts used in soap production, pulp and paper, flue gas
desulfurization, and water treatment. In these applications, it is assumed that one mole of carbon is released for
every mole of soda ash used. Thus, approximately 0.113 metric tons of carbon (or 0.415 metric tons of CO2) are
released for every metric ton of soda ash consumed. The activity data for soda ash consumption for 1990 to 2022
(see Table 4-20) were obtained from the U.S. Geological Survey (USGS) Minerals Yearbook for Soda Ash (1994
through 2015b) and USGS Mineral Industry Surveys for Soda Ash (USGS 2017a, 2018, 2019, 2020b, 2021b, 2022a,
2023a). Soda ash consumption data were collected by the USGS from voluntary surveys of the U.S. soda ash
industry.
Table 4-20: Other Uses of Soda Ash Consumption Not Associated with Glass Manufacturing
(kt)
Activity
1990
2005
2018
2019
2020
2021
2022
Soda Asha
3,3511
3,1441
2,576
2,497
2,310
2,360
2,391
a 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).
Non-Metallurgical Magnesia Production
Carbon dioxide emissions from non-metallurgical magnesia production were calculated using a Tier 1 method from
the 2006 IPCC Guidelines, in accordance with the IPCC methodological decision tree and available data, by
multiplying the quantity of magnesium ore extracted from the mine and processed at the facility by the carbonate
content for magnesite or limestone, respectively, and by the emission factor for magnesite or limestone
calcination, respectively: 0.52197 metric ton CCh/metric ton carbonate for magnesite and 0.43971 metric ton
CCh/metric ton carbonate for limestone.25 A USGS report on magnesite deposits at Gabbs, Nevada lists the
carbonate content of magnesite as 98 percent magnesite and 1 percent limestone (USGS 1948). In the absence of
other data, all magnesium ore extracted from the mine is assumed to be used for non-metallurgical magnesium
production. Magnesium ore extracted from the mine and processed at the facility for non-metallurgical magnesia
production for 2002 through 2022 (see Table 4-21) was obtained from the Nevada Department of Environmental
Quality (McNeece 2023). This data was not available for 1990 through 2001. To address this gap in data availability
and time-series consistency, carbonate consumption for 1990 through 2001 were estimated by multiplying the
average ratio of magnesium ore consumption to production capacity for 2002 to 2004 by the production capacity
of the facility in Nevada. Production capacity for 1990 through 2001 was obtained from the USGS Minerals
Yearbook for Magnesium Compounds (USGS 1990 through 2002).
25 2006 IPCC Guidelines, Volume 3: Chapter 2, Table 2.1.
Industrial Processes and Product Use 4-29
-------
Table 4-21: Magnesite and Limestone Consumption from Non-Metallurgical Magnesia
Production (kt)
Activity
1990S
2005
2018
2019
2020
2021
2022
Magnesite
214
3631
321
289
410
439
388
Limestone
2
4
3
3
4
4
4
Total
216
367
325
292
414
443
392
Note: Totals may not sum due to independent rounding.
Methodological approaches were applied to the entire time series to ensure consistency in emissions from 1990
through 2022. Consistent with the 2006IPCC Guidelines, the overlap technique was applied for non-metallurgical
magnesia production to compare the magnesium ore consumption data to production capacity data for years
where there was overlap. To address inconsistencies, adjustments were made, as described above.
Uncertainty
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 and
industry types. USGS estimates the historical response rate for the crushed stone survey to be approximately 70
percent, and 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. EPA contacted the
USGS National Minerals Information Center Crushed Stone commodity expert to assess the current uncertainty
ranges associated with the limestone and dolomite consumption data compiled and published by USGS. During
this discussion, the expert confirmed that EPA's range of uncertainty was still reasonable (Willett 2017). EPA
assigned an uncertainty range of ±10 percent for limestone and dolomite consumption, based on expert
judgement (Willett 2017). EPA assigned an uncertainty range of ±5 percent for soda ash consumption, and using
this suggested uncertainty provided in Volume 3, Chapter 2, Section 2.4.2.2 of the 2006 IPCC Guidelines is
appropriate based on expert judgment (RTI 2023).
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. EPA assigned an uncertainty range of ±3 percent for the
CO2 emission factors for limestone and dolomite consumption, and using this suggested uncertainty provided in
Volume 3, Chapter 2, Section 2.5.2.1 of the 2006 IPCC Guidelines is appropriate based on expert judgment (RTI
2023).
For emissions from ceramics production, data on clay consumption are collected by USGS through voluntary
national surveys. Large fluctuations in reported consumption exist, reflecting year-to-year changes in the number
of survey responders. The accuracy of distribution by end use is also uncertain because this value is reported by
the producer and not the end user. Uncertainty in the estimates also arises in part due to the variations in the
carbonate content of the various clays used for the various types of ceramics. As discussed above, as no
information is available on the carbonate content for each clay, fractions of limestone and dolomite consumed and
a carbonate content for clay from the 2006 IPCC Guidelines are used. EPA assigned an uncertainty range of ±10
percent for the activity data and ±3 percent for the emission factors, consistent with uncertainty ranges for
limestone and dolomite activity data and emission factors for other process uses of carbonates, respectively.
4-30 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
For emissions from soda ash consumption, the primary source of uncertainty results from the fact that these
emissions are dependent upon the type of processing employed by each end-use. Specific emission factors for
each end-use are not available, so a Tier 1 default emission factor is used for all end-uses. Therefore, there is
uncertainty surrounding the emission factors from the consumption of soda ash. Additional uncertainty comes
from the reported consumption and allocation of consumption within sectors that is collected on a quarterly basis
by the USGS. Efforts have been made to categorize company sales within the correct end-use sector. EPA assigned
an uncertainty range of ±2 percent for the CO2 emission factor for soda ash consumption. The uncertainty range is
derived from the default ranges for soda ash consumption for glass production in Volume 3, Chapter 2, Section
2.4.2.1 of the 2006IPCC Guidelines which is representative of soda ash consumption not associated with glass
production, based on expert judgment (RTI 2023).
For non-metallurgical magnesia production, uncertainties arise due to variations in the chemical composition of
the carbonates used in production of caustic-calcined magnesia production. As noted, minor quantities of other
carbonates beyond limestone and magnesite are also used but unknown. These other carbonates are likely small
and have a minimal impact on the derived emission factor. EPA assigned an uncertainty range of ±10 percent for
the activity data and ±3 percent for the emission factors, consistent with uncertainty ranges for limestone and
dolomite activity data and emission factors for other process uses of carbonates, respectively. The results of the
Approach 2 quantitative uncertainty analysis are summarized in Table 4-22.
A normal probability density function was assigned for all activity data, and a triangular probability density
function was assigned for all emission factors (RTI 2023). Carbon dioxide emissions from other process uses of
carbonates in 2022 were estimated to be between 9.2 and 12.0 MMT CO2 Eq. at the 95 percent confidence level.
This indicates a range of approximately 12 percent below and 15 percent above the emission estimate of 10.4
MMTCCh Eq.
Table 4-22: Approach 2 Quantitative Uncertainty Estimates for CO2 Emissions from Other
Process Uses of Carbonates (MMT CO2 Eq. and Percent)
Source
Gas
2022 Emission Estimate
(MMTCO. Eq.)
Uncertainty Range Relative to Emission Estimate-1
(MMTCO' Eq.) (%)
Lower
Upper
Lower Upper
Bound
Bound
Bound Bound
Other Process Uses of
Carbonates
C02
10.4
9.2
12.0
-12% +15%
a Range of emission estimates predicted by Monte Carlo stochastic simulation for a 95 percent confidence interval.
QA/QC and Verification
General quality assurance/quality control (QA/QC) procedures were applied consistent with the U.S. Inventory
QA/QC plan, which is in accordance with Volume 1, Chapter 6 of 2006 IPCC Guidelines as described in the
introduction of the IPPU chapter (see Annex 8 for more details).
Recalculations Discussion
For the current Inventory, emission estimates for new subcategories ceramics production and non-metallurgical
magnesia production were incorporated across the entire time series. No other recalculations were performed for
the 1990 through 2021 portion of the time series.
Planned Improvements
EPA plans to review the uncertainty ranges assigned to activity data. This planned improvement is currently
planned as a medium-term improvement.
Industrial Processes and Product Use 4-31
-------
4.5 Ammonia Production (CRT Source
Category 2B1)
Emissions of carbon dioxide (CO2) occur during the production of synthetic ammonia (NH3), primarily through the
use of natural gas, petroleum coke, or naphtha as a feedstock. The natural gas-, naphtha-, and petroleum coke-
based processes produce CO2 and hydrogen (H2), the latter of which is used in the production of ammonia. The
brine electrolysis process for production of ammonia does not lead to process-based CO2 emissions. This reporting
category (2B1) includes emissions from the production of ammonia. Due to national circumstances, emissions from
fuels consumed for energy purposes during the production of ammonia are accounted for as part of fossil fuel
combustion in the industrial end-use sector reported under the Energy chapter. More information on this
approach can be found in the Methodology section below.
Ammonia production requires a source of nitrogen (N) and hydrogen (H). Nitrogen is obtained from air through
liquid air distillation or an oxidative process where air is burnt and the residual nitrogen is recovered. In the United
States, the majority of ammonia is produced using a natural gas feedstock as the hydrogen source. One synthetic
ammonia production plant located in Kansas is producing ammonia from petroleum coke feedstock. In some U.S.
plants, some of the CO2 produced by the process is captured and used to produce urea rather than being emitted
to the atmosphere. In 2022,16 companies operated 35 ammonia producing facilities in 16 states. Approximately
60 percent of domestic ammonia production capacity is concentrated in Louisiana, Oklahoma, and Texas (USGS
2023).
Synthetic ammonia production from natural gas feedstock consists of five principal process steps. The primary
reforming step converts methane (CH4) to CO2, carbon monoxide (CO), and hydrogen (H2) in the presence of a
catalyst. Only 30 to 40 percent of the CH4 feedstock to the primary reformer is converted to CO and CO2 in this
step of the process. The secondary reforming step converts the remaining CH4 feedstock to CO and CO2. In the shift
conversion step, the CO in the process gas from the secondary reforming step (representing approximately 15
percent of the process gas) is converted to CO2 in the presence of a catalyst, water, and air. Carbon dioxide is
removed from the process gas by the shift conversion process, and the H2 is combined with the nitrogen (N2) gas in
the process gas during the ammonia synthesis step to produce ammonia. The CO2 is included in a waste gas stream
with other process impurities and is absorbed by a scrubber solution. In regenerating the scrubber solution, CO2 is
released from the solution.
The conversion process for conventional steam reforming of CH4, including the primary and secondary reforming
and the shift conversion processes, is approximately as follows:
0.88C7/4 + 1.26Air +1.24H20 0.88C02 + N2 +3H2
N2 + 3H2 -> 2NH3
To produce synthetic ammonia from petroleum coke, the petroleum coke is gasified and converted to CO2 and H2.
These gases are separated, and the H2 is used as a feedstock to the ammonia production process, where it is
reacted with N2 to form ammonia.
Not all of the CO2 produced during the production of ammonia is emitted directly to the atmosphere. Some of the
ammonia and some of the CO2 produced by the synthetic ammonia process are used as raw materials in the
production of urea [CO(NH2>2], which has a variety of agricultural and industrial applications.
The chemical reaction that produces urea is:
2nh3+ C02 -> NH2COONH4 -> CO(NH2)2 +h2o
Only the CO2 emitted directly to the atmosphere from the synthetic ammonia production process is accounted for
in determining emissions from ammonia production. The CO2 that is captured during the ammonia production
process and used to produce urea does not contribute to the CO2 emission estimates for ammonia production
4-32 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
presented in this section. Instead, CO2 emissions resulting from the consumption of urea are attributed to the urea
consumption or urea application source category (under the assumption that the carbon stored in the urea during
its manufacture is released into the environment during its consumption or application). Emissions of CO2 resulting
from agricultural applications of urea are accounted for in Section 5.6. Emissions of CO2 resulting from non-
agricultural applications of urea (e.g., use as a feedstock in chemical production processes) are accounted for in
Section 4.6.
Emissions from fuel used for energy at ammonia plants are accounted for as part of fossil fuel combustion in the
industrial end-use sector reported under the Energy chapter. The consumption of natural gas and petroleum coke
as fossil fuel feedstocks for NH3 production are adjusted for within the Energy chapter as these fuels were
consumed during non-energy related activities. More information on this methodology is described in Annex 2.1,
Methodology for Estimating Emissions of CChfrom Fossil Fuel Combustion.
Total emissions of CO2 from ammonia production in 2022 were 12.6 MMT CO2 Eq. (12,610 kt) and are summarized
in Table 4-23 and Table 4-24. Ammonia production relies on natural gas as both a feedstock and a fuel, and as
such, market fluctuations and volatility in natural gas prices affect the production of ammonia. Since 1990,
emissions from ammonia production have decreased by 12 percent. Emissions in 2022 increased by about 3
percent from the 2021 levels. One facility in Kansas produces ammonia from petroleum coke and began operations
in 2000. All other facilities use natural gas as feedstock.
Emissions from ammonia production increased steadily from 2015 to 2018, due to the addition of new ammonia
production facilities and new production units at existing facilities in 2016, 2017, and 2018. Agriculture continues
to drive demand for nitrogen fertilizers, accounting for approximately 88 percent of domestic ammonia
consumption (USGS 2023).
Table 4-23: CO2 Emissions from Ammonia Production (MMT CO2 Eq.)
Source 1990 2005 2018 2019 2020
2021
2022
Ammonia Production 14.4 10.2 12.7 12.4 13.0
12.2
12.6
Table 4-24: CO2 Emissions from Ammonia Production (kt CO2)
Source 1990 2005 2018 2019 2020
2021
2022
Ammonia Production 14,404 10,234 12,669 12,401 13,006
12,192
12,610
Methodology and Time-Series Consistency
Estimates of CO2 emissions from the production of synthetic ammonia for 2010 through 2022 are estimated using
a country-specific approach consistent with Tier 3 method from the 2006IPCC Guidelines, in accordance with the
IPCC methodological decision tree and available data (IPCC 2006). The methodology for 2010 to 2022 directly uses
the process CO2 emissions reported to subpart G of the U.S. EPA Greenhouse Gas Reporting Program (GHGRP)
(EPA 2018, EPA 2023). Estimates for 1990 to 2009 emissions are based on reported and calculated data on natural
gas and petroleum coke feedstock used for ammonia production, consistent with IPCC Tier 2 methods and in
accordance with the IPCC methodological decision tree and available data.
Emissions from fuel used for energy at ammonia plants are accounted for in the Energy chapter. This approach
differs slightly from the 2006 IPCC Guidelines which indicates that "in the case of ammonia production no
distinction is made between fuel and feedstock emissions with all emissions accounted for in the IPPU Sector."
Disaggregated data on fuel used for ammonia feedstock and fuel used for energy for ammonia production are not
available in the United States. The Energy Information Administration (EIA), where energy use data are obtained
for the Inventory (see the Energy chapter), does not provide data broken out by industrial category. EIA data are
only available at the broad industry sector level. Furthermore, the GHGRP data used to estimate emissions are
based on feedstock use and not fuel use. The method uses the same science informing the 2006 IPCC guidelines
Industrial Processes and Product Use 4-33
-------
and is consistent with avoiding double counting in the reporting of fuel use emissions under Energy and IPPU
reporting. See more information in introduction to this Chapter.
Petroleum Coke Feedstock
Since 2000, one facility in the United States has produced ammonia using petroleum coke as a feedstock. For 2010
to 2022, CO2 emissions from the production of synthetic ammonia from petroleum coke feedstock were estimated
using CO2 emissions reported by the facility to GHGRP (EPA 2018; EPA 2023).
For 2006 to 2009, CO2 emissions from the production of synthetic ammonia from petroleum coke feedstock were
estimated by multiplying the following: quantity of petroleum coke feedstock reported by the facility (CVR 2008
through 2022); the Inventory heating content value for petroleum coke (consistent with values used in the Energy
chapter); the petroleum coke carbon content; and a stoichiometric CO2/C factor of 44/12.
For 2000 to 2005, the quantity of petroleum coke feedstock was not available and was estimated by multiplying
the average ratio of petroleum coke feedstock quantity to ammonia production quantity produced from
petroleum coke from 2006 through 2010 by total ammonia production for 2000 to 2005 (ACC 2023). The years
2006 to 2010 were used to determine the average ratio of petroleum coke feedstock quantity to the ammonia
quantity produced from petroleum coke because that period was deemed to better represent historic ammonia
production from petroleum coke for the period from 2000 to 2005.
For 2000 to 2005, CO2 emissions from the production of synthetic ammonia from petroleum coke feedstock were
estimated by multiplying the following: the average ratio of petroleum coke feedstock quantity to ammonia
production quantity; total ammonia production quantity (ACC 2023); the Inventory heating content value for
petroleum coke (consistent with values used in the Energy chapter); the petroleum coke carbon content; and the
stoichiometric ratio of CO2 to C (44/12).
Natural Gas Feedstock
For 2017 through 2022, facilities directly reported to GHGRP the quantity of natural gas feedstock used for
ammonia production along with the carbon content of the natural gas feedstock (EPA 2018; EPA 2023).
For 2010 through 2016, the quantity of natural gas feedstock was calculated using GHGRP process CO2 emissions
for 2010 through 2016, average molecular weight of the feedstock from 2017 through 2021, and average carbon
content from 2017 through 2021. Data from years 2017 to 2021 were used to determine the average molecular
weight and the average carbon content because that period better represents historic ammonia production from
2010 to 2016. Using all available data from 2017 to 2021 allowed for the maximum number of data points
available at the time of adopting this methodology to ensure that the average was representative. The averages
were not updated using later data to exclude any new facilities that might not be representative of facilities that
were operating during the earlier years of the GHGRP.
For 2010 to 2022, CO2 emissions from the production of synthetic ammonia from natural gas feedstock were
estimated using the CO2 emissions reported to the GHGRP (EPA 2018; EPA 2023) and subtracting the CO2 emissions
from the production of synthetic ammonia from petroleum coke feedstock as determined in the Petroleum Coke
Feedstock section above.
For 1990 to 2009, the quantity of natural gas feedstock was not available and was estimated by multiplying the
average ratio of natural gas feedstock quantity to ammonia production quantity from 2010 through 2014 by total
ammonia production for each year for 1990 to 2009 (ACC 2023). The years 2010 to 2014 were used to determine
the average ratio of natural gas feedstock quantity to ammonia production because that period better represents
4-34 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
historic ammonia production from 1990 to 2009.26 For 1990 to 2009, CO2 emissions from the production of
synthetic ammonia from natural gas feedstock were estimated using the natural gas feedstock quantity as
determined above and the Inventory CO2 emissions factor and heating content value for natural gas (consistent
with values used in the Energy chapter).
Urea Production Adjustments
Emissions of CO2 from ammonia production from both feedstocks and for all years from 1990 to 2022 were
adjusted to account for the use of some CO2 emissions resulting from ammonia production as a raw material in the
production of urea. The CO2 emissions reported for ammonia production are reduced by a factor of 0.733, which
corresponds to a stoichiometric CCh/urea factor of 44/60, assuming complete conversion of ammonia (NH3) and
CO2 to urea (IPCC 2006; EFMA 2000), and multiplied by total annual domestic urea production.
All synthetic ammonia production and subsequent urea production are assumed to be from the same process-
conventional catalytic reforming of natural gas feedstock, with the exception of ammonia production from
petroleum coke feedstock at the one facility located in Kansas.
Table 4-25: Total Ammonia Production, Total Urea Production, and Recovered CO2 Consumed
for Urea Production (kt)
Year
1990
2005
2018
2019
2020
2021
2022
Total Ammonia Production
15/." ¦
10,143 1
16,010
16,410
17,020
15,420
16,800
Total Urea Production
7/1J
5,270
10,700
11,400
11,500
10,521
11,272
Recovered C02 Consumed for
Urea Production
V
3,865
7,847
8,360
8,433
7,715
8,266
Total ammonia production, total urea production, and recovered CO2 consumed for urea production are shown in
Table 4-25. Total ammonia production data for 2011 through 2022 were obtained from American Chemistry
Council (ACC 2023). For years 1990 through 2011, ammonia production data were obtained from the Census
Bureau of the U.S. Department of Commerce (U.S. Census Bureau 1991 through 1994,1998 through 2011) as
reported in Current Industrial Reports Fertilizer Materials and Related Products annual and quarterly reports. Data
on facility-level process emissions for 2010 through 2022 and data on natural gas feedstock used and carbon
content of the natural gas feedstock starting in 2017 were obtained from GHGRP (EPA 2018; EPA 2023). Natural
gas and petroleum coke heating values come from national-level data (EIA 2023), and natural gas and petroleum
coke carbon contents are the same as used in the Energy chapter calculations.
Data on urea production for 2010 through 2022 were obtained from GHGRP (EPA 2018, EPA 2023). Urea
production data for 2009 through 2010 were obtained from the U.S. Census Bureau (U.S. Census Bureau 2010 and
2011). Urea production data for 1990 through 2008 were obtained from the USGS Minerals Yearbook: Nitrogen
(USGS 1994-2009). The U.S. Census Bureau ceased collection of urea production statistics in 2011.
Methodological approaches were applied to the entire time series to ensure consistency in emissions from 1990
through 2022. The methodology for ammonia production spliced activity data from different sources: U. S. Census
Bureau data for 1990 through 2010, ACC data beginning in 2011, and GHGRP data beginning in 2010 and
2017. Consistent with the 2006 IPCC Guidelines, the overlap technique was applied to compare the two data sets
for years where there was overlap, with findings that the data sets were consistent and adjustments were not
needed.
26 The number of facilities reporting to GHGRP has increased since 2010: 22 facilities reported from 2010 to 2012; 23 from
2013 to 2015; 26 in 2016; 28 in 2017 and 29 from 2018 to 2022. Using data from 2010 to 2014 excludes the newer facilities
that might not be representative of facilities in earlier years.
Industrial Processes and Product Use 4-35
-------
Uncertainty
The uncertainties presented in this section are primarily due to how accurately the emission factor used represents
an average across all ammonia plants using natural gas feedstock. Uncertainty in the back calculation of natural gas
feedstock used for 1990 through 2009 also exists. Using the average ratio of natural gas feedstock quantity to
ammonia production, determined using GHGRP data from 2010 to 2014, does not account for efficiency gains in
ammonia production since 1990 (e.g., potential decreases in gas usage per ton of ammonia, manufacturing shift
from steam-driven turbines to electrical-drive turbines). Uncertainties are also associated with ammonia
production estimates and the assumption that all ammonia production and subsequent urea production was from
the same process—conventional catalytic reforming of natural gas feedstock, with the exception of one ammonia
production plant located in Kansas that is manufacturing ammonia from petroleum coke feedstock. Uncertainty is
also associated with the representativeness of the emission factor used for the petroleum coke-based ammonia
process. It is also assumed that ammonia and urea are produced at co-located plants from the same natural gas
raw material. The uncertainty of the total urea production activity data, based on USGS Minerals Yearbook:
Nitrogen data, is a function of the reliability of reported production data and is influenced by the completeness of
the survey responses. EPA assigned an uncertainty range of ±5 percent for ammonia production and a range of ±2
percent for urea production, natural gas feedstock quantity, petroleum coke feedstock quantity, and carbon
content of natural gas feedstock, and using the suggested uncertainty provided in Section 3.2.3.2 of the 2006IPCC
Guidelines is appropriate based on expert judgment (RTI 2023). Per this expert judgement, a normal probability
density function was assigned for all variables.
Recovery of CO2 from ammonia production plants for purposes other than urea production (e.g., commercial sale,
etc.) has not been considered in estimating the CO2 emissions from ammonia production, as data concerning the
disposition of recovered CO2 are not available. Such recovery may or may not affect the overall estimate of CO2
emissions depending upon the end use to which the recovered CO2 is applied. Further research is required to
determine whether byproduct CO2 is being recovered from other ammonia production plants for application to
end uses that are not accounted for elsewhere; however, for reporting purposes, CO2 consumption for urea
production is provided in this chapter.
The results of the Approach 2 quantitative uncertainty analysis are summarized in Table 4-26. Carbon dioxide
emissions from ammonia production in 2022 were estimated to be between 12.2 and 13.1 MMT CO2 Eq. at the 95
percent confidence level. This indicates a range of approximately 4 percent below and 4 percent above the
emission estimate of 12.6 MMT CO2 Eq.
Table 4-26: Approach 2 Quantitative Uncertainty Estimates for CO2 Emissions from Ammonia
Production (MMT CO2 Eq. and Percent)
Source
Gas
2022 Emission Estimate
Uncertainty Range Relative to Emission Estimate"
(MMT CO . Eq.)
(MMTCO. Eq.)
(%)
Lower Upper
Lower Upper
Bound Bound
Bound Bound
Ammonia Production
C02
12.6
12.2 13.1
-4% +4%
a Range of emission estimates predicted by Monte Carlo stochastic simulation for a 95 percent confidence interval.
QA/QC and Verification
General quality assurance/quality control (QA/QC) procedures were applied to ammonia production emission
estimates consistent with the U.S. Inventory QA/QC plan, which is in accordance with Volume 1, Chapter 6 of 2006
IPCC Guidelines as described in the introduction of the IPPU chapter (see Annex 8 for more details). More details
on the greenhouse gas calculation, monitoring and QA/QC methods applicable to ammonia facilities can be found
4-36 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
under Subpart G (Ammonia Production) of the regulation (40 CFR Part 98).27 EPA verifies annual facility-level
GHGRP reports through a multi-step process (e.g., combination of electronic checks and manual reviews) to
identify potential errors and ensure that data submitted to EPA are accurate, complete, and consistent.28 Based on
the results of the verification process, EPA follows up with facilities to resolve mistakes that may have occurred.
The post-submittals checks are consistent with a number of general and category-specific QC procedures, including
range checks, statistical checks, algorithm checks, and year-to-year checks of reported data and emissions.
More details on the greenhouse gas calculation, monitoring, and QA/QC methods applicable to reporting of urea
produced at ammonia production facilities can be found under Section 4.6 Urea Consumption for Non-Agricultural
Purposes.
Recalculations
For 2021, the urea consumption value was changed from a rounded value to a more precise unrounded value. As a
result, recalculations were performed for emissions from ammonia for 2021. Compared to the previous Inventory,
total CO2 emissions from the production of ammonia production (from natural gas and petroleum coke feedstocks)
decreased by less than 1 percent (15 kt) in 2021.
Planned Improvements
Currently the Inventory does not separately track fuel energy use for ammonia production. To be more consistent
with 2006IPCC Guidelines, EPA is considering whether to include natural gas fuel use as part of ammonia
production emissions as a future improvement. The data are still being evaluated as part of EPA's efforts to
disaggregate other industrial sector categories' energy use in the Energy chapter of the Inventory. If possible, this
will be incorporated in future Inventory reports. If incorporated, the fuel energy use and emissions will be removed
from current reporting under Energy to avoid double counting.
4.6 Urea Consumption for Non-Agricultural
Purposes (CRT Source Category 2B10)
Urea is produced using ammonia (NH3) and carbon dioxide (CO2) as raw materials. All urea produced in the United
States is assumed to be produced at ammonia production facilities where both ammonia and CO2 are generated.
There were 35 plants producing ammonia in the United States in 2022, with two additional plants sitting idle for
the entire year (USGS 2023b).
The chemical reaction that produces urea is:
2nh3+ C02 -> NH2COONH4 -> CO(NH2)2 +h2o
This section accounts for CO2 emissions associated with urea consumed exclusively for non-agricultural purposes.
This reporting category (2B10) includes emissions from IPCC assessment reports that do not fall within any other
CRT source category, which includes emissions from urea consumption for non-agricultural purposes. Emissions of
CO2 resulting from agricultural applications of urea are accounted for in Section 5.6 of the Agriculture chapter.
27 See http://www.ecfr.gov/cgi-bin/text-idx7tpk/ecfrbrowse/Title40/40cfr98 main Q2.tpl.
28 See https://www.epa.gov/sites/production/files/2015-07/documents/ghgrp verification factsheet.pdf.
Industrial Processes and Product Use 4-37
-------
The industrial applications of urea include its use in adhesives, binders, sealants, resins, fillers, analytical reagents,
catalysts, intermediates, solvents, dyestuffs, fragrances, deodorizers, flavoring agents, humectants and
dehydrating agents, formulation components, monomers, paint and coating additives, photosensitive agents, and
surface treatments agents. In addition, urea is used for abating nitrogen oxide (NOx) emissions from coal-fired
power plants and diesel transportation motors.
Emissions of CChfrom urea consumed for non-agricultural purposes in 2022 were estimated to be 7.1 MMT CO2
Eq. (7,053 kt) and are summarized in Table 4-27 and Table 4-28. Net CO2 emissions from urea consumption for
non-agricultural purposes have increased by approximately 86 percent from 1990 to 2022 and increased by
approximately 7 percent from 2021 to 2022.
Table 4-27: CO2 Emissions from Urea Consumption for Non-Agricultural Purposes (MMT CO2
Eq.)
Source 1990 2005 2018 2019 2020 2021 2022
Urea Consumption 3.8 3.7 6.1 6.2 5.8 6.6 7.1
Table 4-28: CO2 Emissions from Urea Consumption for Non-Agricultural Purposes (kt CO2)
Source 1990 2005 2018 2019 2020 2021 2022
Urea Consumption 3,784 3,653 6,113 6,150 5,805 6,600 7,053
Methodology and Time-Series Consistency
Emissions of CO2 resulting from urea consumption for non-agricultural purposes are estimated using a country-
specific method consistent with the Tier 1 method used to estimate emissions from ammonia production in the
2006IPCC Guidelines which states that the "CO2 recovered [from ammonia production] for downstream use can be
estimated from the quantity of urea produced where CO2 is estimated by multiplying urea production by 44/60,
the stoichiometric ratio of CO2 to urea" (IPCC 2006). The amount of urea consumed in the United States for non-
agricultural purposes is multiplied by a factor representing the amount of CO2 used as a raw material to produce
the urea. This method is based on the assumption that all of the carbon in urea is released into the environment as
CO2 during use.
The amount of urea consumed for non-agricultural purposes in the United States is estimated by deducting the
quantity of urea fertilizer applied to agricultural lands, which is obtained directly from the Agriculture chapter (see
Table 5-25), from the total domestic supply of urea as reported in Table 4-29. The domestic supply of urea is
estimated based on the amount of urea produced plus urea imports and minus urea exports. A factor of 0.733 tons
of CO2 per ton of urea consumed is then applied to the resulting supply of urea for non-agricultural purposes to
estimate CO2 emissions from the amount of urea consumed for non-agricultural purposes. The 0.733 tons of CO2
per ton of urea emission factor is based on the stoichiometry of carbon in urea. This corresponds to a
stoichiometric ratio of CO2 to urea of 44/60, assuming complete conversion of carbon in urea to CChflPCC 2006;
EFMA 2000).
Urea production data for 1990 through 2008 were obtained from the U.S. Geological Survey (USGS) Minerals
Yearbook: Nitrogen (USGS 1994 through 2009a). Urea production data for 2009 through 2010 were obtained from
the U.S. Census Bureau (2011). The U.S. Census Bureau ceased collection of urea production statistics in 2011.
Urea production data for 2011 through 2022 were obtained from GHGRP (EPA 2018; EPA 2023a; EPA 2023b).
Urea import data for 2022 were not available at the time of publication and were estimated using 2021 values.
Urea import data for 2013 to 2021 were obtained from the USGS Minerals Yearbook: Nitrogen (USGS 2023a). Urea
import data for 2011 and 2012 were taken from U.S. Fertilizer Import/Exports from the United States Department
of Agriculture (USDA) Economic Research Service Data Sets (U.S. Department of Agriculture 2012). USDA
suspended updates to this data after 2012. Urea import data for the previous years were obtained from the U.S.
Census Bureau Current Industrial Reports Fertilizer Materials and Related Products annual and quarterly reports for
4-38 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
1997 through 2010 (U.S. Census Bureau 2001 through 2011), The Fertilizer Institute (TFI 2002) for 1993 through
1996, and the United States International Trade Commission Interactive Tariff and Trade DataWeb (U.S. ITC 2002)
for 1990 through 1992 (see Table 4-29).
Urea export data for 2022 were not available at the time of publication and were estimated using 2021 values.
Urea export data for 2013 to 2021 were obtained from the USGS Minerals Yearbook: Nitrogen (USGS 2023a). Urea
export data for 1990 through 2012 were taken from U.S. Fertilizer Import/Exports from USDA Economic Research
Service Data Sets (U.S. Department of Agriculture 2012). USDA suspended updates to this data after 2012.
Table 4-29: Urea Production, Urea Applied as Fertilizer, Urea Imports, and Urea Exports (kt)
Year
1990
2005
2018
2019
2020
2021
2022
Urea Production
7,450
5,270
10,700
11,400
11,500
10,521
11,272
Urea Applied as Fertilizer
3,296::
4,779 1
6,732
6,865
6,998
7,131
7,265
Urea Imports
1,860
5,026
5,110
4,410
4,190
5,880
5,880
Urea Exports
854
536 I
743
559
777
270
270
Urea Consumed for Non-
Agricultural Purposes
5,160
4,9811
8,335
8,386
7,915
9,000
9,617
Methodological approaches were applied to the entire time series to ensure consistency in emissions from 1990
through 2022. The methodology for urea consumption for non-agricultural purposes spliced activity data from
different sources: USGS data for 1990 through 2008, U. S. Census Bureau data for 2009 and 2010, and GHGRP data
beginning in 2011. Consistent with the 2006IPCC Guidelines, the overlap technique was applied to compare the
data sets for years where there was overlap, with findings that the data sets were consistent and adjustments
were not needed.
Uncertainty
There is limited publicly available data on the quantities of urea produced and consumed for non-agricultural
purposes. Therefore, the amount of urea used for non-agricultural purposes is estimated based on a balance that
relies on estimates of urea production, urea imports, urea exports, and the amount of urea used as fertilizer. EPA
uses an uncertainty range of ±10 percent for urea production and ±5 percent for urea imports and urea exports,
consistent with the ranges for activity data that are not obtained directly from plants, and using this suggested
uncertainty provided in Section 3.2.3.2 of the 2006 IPCC Guidelines is appropriate based on expert judgment (RTI
2023). Per this expert judgment, a normal probability density function was assigned for all activity data. The
primary uncertainties associated with this source category are associated with the accuracy of these estimates as
well as the fact that each estimate is obtained from a different data source. Because urea production estimates are
no longer available from the USGS, there is additional uncertainty associated with urea produced beginning in
2011. There is also uncertainty associated with the assumption that all of the carbon in urea is released into the
environment as CO2 during use.
The results of the Approach 2 quantitative uncertainty analysis are summarized in Table 4-30. Carbon dioxide
emissions associated with urea consumption for non-agricultural purposes during 2022 were estimated to be
between 6.8 and 7.3 MMT CO2 Eq. at the 95 percent confidence level. This indicates a range of approximately 4
percent below and 4 percent above the emission estimate of 7.1 MMT CO2 Eq.
Industrial Processes and Product Use 4-39
-------
Table 4-30: Approach 2 Quantitative Uncertainty Estimates for CO2 Emissions from Urea
Consumption for Non-Agricultural Purposes (MMT CO2 Eq. and Percent)
Source
Gas
2022 Emission Estimate
Uncertainty Range Relative to Emission Estimate"
(MMT CO . Eq.)
(MMT CO
Eq.)
(%)
Lower
Upper
Lower
Upper
Bound
Bound
Bound
Bound
Urea Consumption for
Non-Agricultural
C02
7.1
6.8
7.3
-4%
+4%
Purposes
a Range of emission estimates predicted by Monte Carlo stochastic simulation for a 95 percent confidence interval.
QA/QC and Verification
General quality assurance/quality control (QA/QC) procedures were applied consistent with the U.S. Inventory
QA/QC plan, which is in accordance with Volume 1, Chapter 6 of 2006IPCC Guidelines as described in the
introduction of the IPPU chapter (see Annex 8 for more details).
More details on the greenhouse gas calculation, monitoring and QA/QC methods applicable to reporting of urea
production occurring at ammonia facilities can be found under Subpart G (Ammonia Manufacturing) of the
regulation (40 CFR Part 98).29 EPA verifies annual facility-level GHGRP reports through a multi-step process (e.g.,
combination of electronic checks and manual reviews) to identify potential errors and ensure that data submitted
to EPA are accurate, complete, and consistent.30 Based on the results of the verification process, EPA follows up
with facilities to resolve mistakes that may have occurred. The post-submittals checks are consistent with a
number of general and category-specific QC procedures, including range checks, statistical checks, algorithm
checks, and year-to-year checks of reported data and emissions. EPA also conducts QA checks of GHGRP reported
urea production data against external datasets including the USGS Minerals Yearbook data. The comparison shows
consistent trends in urea production over time.
Recalculations Discussion
Based on updated quantities of urea applied for agricultural uses for 2017 through 2021, updated urea imports
from USGS for 2021, and updated urea exports from USGS for 2021, recalculations were performed for 2017
through 2021. Compared to the previous Inventory, CO2 emissions from urea consumption for non-agricultural
purposes increased by less than 1 percent for 2017 (46 kt CO2) and 2018 (2 kt CO2), decreased by less than 1
percent for 2019 (4 kt CO2) and 2020 (10 kt CO2) and increased by 32 percent for 2021 (1,611 kt CO2).
Planned Improvements
At this time, there are no specific planned improvements for estimating CO2 emissions from urea consumption for
non-agricultural purposes.
29 See http://www.ecfr.gov/cgi-bin/text-idx7tpk/ecfrbrowse/Title40/40cfr98 main Q2.tpl.
30 See https://www.epa.gov/sites/production/files/2015-07/documents/ghgrp verification factsheet.pdf.
4-40 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
4.7 Nitric Acid Production (CRT Source
Category 2B2)
Nitrous oxide (N2O) is emitted during the production of nitric acid (HNO3), an inorganic compound used primarily
to make synthetic commercial fertilizers. Nitric acid is also a major component in the production of adipic acid—a
feedstock for nylon—and explosives. This reporting category (2B2) includes emissions from production of nitric
acid. Emissions from fuels consumed for energy purposes during the production of nitric acid are accounted for as
part of fossil fuel combustion in the industrial end-use sector reported under the Energy chapter.
There are two types of nitric acid: weak nitric acid and high-strength nitric acid. The weak nitric acid production
method utilizes oxidation, condensation, and absorption to produce nitric acid at concentrations between 30 and
70 percent nitric acid. High-strength nitric acid (90 percent or greater nitric acid) can be produced by two methods:
(1) through the dehydration, bleaching, condensing, and absorption of the weak nitric acid or (2) through the
oxidation of ammonia into nitric oxide, which is oxidized and cooled into dinitrogen tetroxide and then pressurized
and oxidized into high-strength nitric acid. Most U.S. plants were built between 1960 and 2000. As of 2022, there
were 31 active nitric acid production plants that produce weak nitric acid in the United States (EPA 2023). One
plant produces both weak and high-strength nitric acid (EPA 2010).
The basic process technology for producing nitric acid has not changed significantly over time. During this process,
N2O is formed as a byproduct and released from reactor vents into the atmosphere.
Nitric acid is made from the reaction of ammonia (NH3) with oxygen (O2) in two stages. The overall reaction is:
4NH3 + 802 -> 4HNO:i + 4H2
Currently, the nitric acid industry in the United States controls emissions of NO and NO2 (i.e., NOx), using a
combination of non-selective catalytic reduction (NSCR) and selective catalytic reduction (SCR) technologies. In the
process of destroying NOx, NSCR systems are also very effective at destroying N2O. Five nitric acid plants had NSCR
systems installed between 1964 and 1977, over half due to the finalization of the Nitric Acid Plant New Source
Performance Standards (NSPS) which went into effect in 1971. Four additional nitric acid plants had NSCR systems
installed between 2016 and 2018, as a result of EPA Consent Decrees to control NOx emissions more effectively.
NSCR systems are used in approximately one-third of the weak acid production plants. For N2O abatement, U.S.
facilities are using both tertiary (i.e., NSCR and SCR) and secondary controls (i.e., catalysts added to the ammonia
reactor to lessen potential N2O production).
Emissions from the production of nitric acid are generally directly proportional to the annual amount of nitric acid
produced because emissions are calculated as the product of the total annual production and plant-specific
emission factors. There are a few instances, however, where that relationship has not been directly proportional.
For example, in 2015 and 2019, nitric acid production decreased and emissions increased compared to the
respective preceding years; in 2016, nitric acid production increased and emissions decreased compared to
2015. N2O emissions for those years are calculated based on data from the GHGRP as discussed in the
Methodology section below. According to data from plants reporting to GHGRP, plant-specific operations can
affect the emission factor used, including: (1) site-specific fluctuations in ambient temperature and humidity, (2)
catalyst age and condition, (3) process changes, such as fluctuations in process pressure or temperature and
replacing the ammonia catalyst, (4) the addition, removal, maintenance, and utilization of abatement technologies,
and (5) the number of nitric acid trains, which are reaction vessels where ammonia is oxidized to form nitric acid.
Changes in those operating conditions for the years in question (2015, 2016, and 2019) caused changes in emission
factors, which resulted in emissions changing disproportionally to production in those years.
Nitrous oxide emissions from this source were estimated to be 8.6 MMT CO2 Eq. (33 kt of N2O) in 2022 and are
summarized in Table 4-31 and Table 4-32. Emissions from nitric acid production have decreased by 20 percent
since 1990, while production has increased by 9 percent over the same time period (see Table 4-31 and Table
Industrial Processes and Product Use 4-41
-------
4-32). Emissions have decreased by 33 percent since 1997, the highest year of production in the time series. From
2021 to 2022, nitric acid production increased by 1 percent, while overall emissions from nitric acid production
increased by 9.4 percent from 2021 to 2022.
Table 4-31: N2O Emissions from Nitric Acid Production (MMT CO2 Eq.)
Year
1990
2005
2018
2019
2020
2021
2022
Nitric Acid Production
10.8 |
10.11
8.5
8.9
8.3
7.9
8.6
Table 4-32: N2O Emissions from Nitric Acid Production (kt N2O)
Year
1990
2005
2018
2019
2020
2021
2022
Nitric Acid Production
41 1
38 i
32
34
31
30
33
Methodology and Time-Series Consistency
Emissions of N2O from nitric acid production are estimated using methods provided by the 2006IPCC Guidelines, in
accordance with the IPCC methodological decision tree and available data. For 2010 through 2022, a Tier 3 method
was used to estimate emissions based on GHGRP data. For 1990 through 2009, a Tier 2 method was used to
estimate emissions from nitric acid production based on U.S. Census Bureau data.
2010 through 2022
Process N2O emissions and nitric acid production data were obtained directly from EPA's GHGRP for 2010 through
2022 by aggregating reported facility-level data (EPA 2018; EPA 2023).31
Since 2010, in the United States, all nitric acid facilities that produce weak nitric acid (30 to 70 percent) have been
required to report annual greenhouse gas emissions data to EPA as per the requirements of the GHGRP (Subpart
V). Beginning with 2018, the rule was changed to include facilities that produce nitric acid of any strength. The only
facility that produces high-strength nitric acid also produces weak nitric acid. All N2O emissions from nitric acid
production originate from the production of weak nitric acid.
Process emissions and nitric acid production reported to the GHGRP provide complete estimates of greenhouse
gas emissions for the United States because there are no reporting thresholds. While facilities are allowed to stop
reporting to the GHGRP if the total reported emissions from nitric acid production are less than 25,000 metric tons
CO2 Eq. per year for five consecutive years or less than 15,000 metric tons CO2 Eq. per year for three consecutive
years, no facilities have stopped reporting as a result of these provisions.32 All nitric acid facilities are required to
either calculate process N2O emissions using a site-specific emission factor that is the average of the emission
factor determined through annual performance tests for each nitric acid train under typical operating conditions or
directly measure process N2O emissions using monitoring equipment.33
Emissions from facilities vary from year to year, depending on the amount of nitric acid produced with and without
abatement technologies and other conditions affecting the site-specific emission factor. To maintain consistency
31 National N20 process emissions, national production, and national share of nitric acid production with abatement and
without abatement technology were aggregated from the GHGRP facility-level data for 2010 to 2022 (i.e., percent production
with and without abatement).
32 See 40 CFR 98.2(i)(l) and 40 CFR 98.2(i)(2) for more information about these provisions.
33 Facilities must use standard methods - either EPA Method 320 or ASTM D6348-03 for annual performance tests—and must
follow associated QA/QC procedures consistent with category-specific QC of direct emission measurements during these
performance tests.
4-42 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
across the time series and with the rounding approaches taken by other data sets, GHGRP nitric acid data are
rounded and are shown in Table 4-33.
1990 through 2009
Using GHGRP data for 2010, country-specific N2O emission factors were calculated for nitric acid production with
abatement and without abatement (i.e., controlled and uncontrolled emission factors). The following 2010
emission factors were derived for production with abatement and without abatement: 3.3 kg INhO/metric ton
HNO3 produced at plants using abatement technologies (e.g., tertiary systems such as NSCR systems) and 5.99 kg
INhO/metric ton HNO3 produced at plants not equipped with abatement technology. Country-specific weighted
emission factors were derived by weighting these emission factors by percent production with abatement and
without abatement over time periods 1990 through 2008 and 2009. These weighted emission factors were used to
estimate N2O emissions from nitric acid production for years prior to the availability of GHGRP data (i.e., 1990
through 2008 and 2009). A separate weighted emission factor is included for 2009 due to data availability for that
year.
EPA verified the installation dates of N2O abatement technologies for all facilities based on GHGRP facility-level
information and confirmed that all abatement technologies were accounted for in the derived emission factors
(Icenhour 2020). Due to the lack of information on abatement equipment utilization, it is assumed that once
abatement technology was installed in facilities, the equipment was consistently operational for the duration of
the time series considered in this report (especially NSCRs).
The country-specific weighted N2O emission factors were used in conjunction with annual production to estimate
N2O emissions for 1990 through 2009, using the following equations:
Equation 4-4:2006IPCC Guidelines Tier 3: N2O Emissions From Nitric Acid Production
(Equation 3.6)
Ei — Pi x EFweighi:eci:i
EFweighted,i = X EFc) + (%PUnc,i X EFunc)\
where,
Ei = Annual N2O Emissions for year i (kg/yr)
Pi = Annual nitric acid production for year i (metric tons HNO3)
EFweighted.i = Weighted N2O emission factor for year i (kg INhO/metric ton HNO3)
%Pc,i = Percent national production of HNO3 with N2O abatement technology (%)
EFc = N2O emission factor, with abatement technology (kg INhO/metric ton HNO3)
%Punc,i = Percent national production of HNO3 without N2O abatement technology (%)
EFunc = N2O emission factor, without abatement technology (kg INhO/metric ton HNO3)
i = year from 1990 through 2009
• For 2009: Weighted N2O emission factor = 5.46 kg INhO/metric ton HNO3.
• For 1990 through 2008: Weighted N2O emission factor = 5.66 kg INhO/metric ton HNO3.
Nitric acid production data for the United States for 1990 through 2009 were obtained from the U.S. Census
Bureau (U.S. Census Bureau 2008, 2009, 2010a, 2010b) (see Table 4-33). EPA used GHGRP facility-level information
to verify that all reported N2O abatement equipment were incorporated into the estimation of N2O emissions from
nitric acid production over the full time series (EPA 2021).
Industrial Processes and Product Use 4-43
-------
Table 4-33: Nitric Acid Production (kt)
Year
1990
2005
2018
2019
2020
2021
2022
Production (kt)
7,200
6,710 |
8,210
8,080
7,970
7,800
7,860
Methodological approaches were applied to the entire time series to ensure consistency in emissions from 1990
through 2022. The methodology for nitric acid production spliced activity data from two different sources: U.S.
Census Bureau production data for 1990 through 2009 and GHGRP production data starting in 2010. Consistent
with the 2006IPCC Guidelines, the overlap technique was applied to compare the two data sets for years where
there was overlap, with findings that the data sets were consistent and adjustments were not needed.
Uncertainty
Uncertainty associated with the parameters used to estimate N2O emissions includes the share of U.S. nitric acid
production attributable to each emission abatement technology (i.e., utilization) over the time series (especially
prior to 2010), and the associated emission factors applied to each abatement technology type. While some
information has been obtained through outreach with industry associations, limited information is available over
the time series (especially prior to 2010) for a variety of facility level variables, including plant-specific production
levels, plant production technology (e.g., low or high pressure, etc.), and abatement technology destruction and
removal efficiency rates. Production data prior to 2010 were obtained from National Census Bureau, which does
not provide uncertainty estimates with their data. Facilities reporting to EPA's GHGRP must measure production
using equipment and practices used for accounting purposes. While emissions are often directly proportional to
production, the emission factor for individual facilities can vary significantly from year to year due to site-specific
fluctuations in ambient temperature and humidity, catalyst age and condition, nitric acid production process
changes, the addition or removal of abatement technologies, and the number of nitric acid trains at the facility. At
this time, EPA does not estimate uncertainty of the aggregated facility-level information. As noted in the QA/QC
and verification section below, EPA verifies annual facility-level reports through a multi-step process (e.g.,
combination of electronic checks and manual reviews by staff) to identify potential errors and ensure that data
submitted to EPA are accurate, complete, and consistent. The annual production reported by each nitric acid
facility under EPA's GHGRP and then aggregated to estimate national N2O emissions is assumed to have low
uncertainty. EPA assigned an uncertainty range of ±5 percent for facility-reported N2O emissions, and using this
suggested uncertainty provided in section 3.4.3.2 of the 2006 IPCC Guidelines is appropriate based on expert
judgment (RTI 2023). EPA assigned an uncertainty range of ±2 percent for nitric acid production, and using this
suggested uncertainty provided in section 3.3.3.2 of the 2006 IPCC Guidelines is appropriate based on expert
judgment (RTI 2023). Per this expert judgment, a normal probability density function was assigned for facility-
reported N2O emissions and nitric acid production.
The results of this Approach 2 quantitative uncertainty analysis are summarized in Table 4-34. Nitrous oxide
emissions from nitric acid production were estimated to be between 8.2 and 9.0 MMT CO2 Eq. at the 95 percent
confidence level. This indicates a range of approximately 5 percent below to 5 percent above the 2022 emissions
estimate of 8.6 MMT CO2 Eq.
Table 4-34: Approach 2 Quantitative Uncertainty Estimates for N2O Emissions from Nitric
Acid Production (MMT CO2 Eq. and Percent)
Source
Gas
2022 Emission Estimate
(MMT CO . Eq.)
Uncertainty Range Relative to Emission Estimate-1
(MMTCO' Eq.) (%)
Lower Upper
Bound Bound
Lower
Bound
Upper
Bound
Nitric Acid Production
N20
8.6
8.2 9.0
-5%
+5%
a Range of emission estimates predicted by Monte Carlo stochastic simulation for a 95 percent confidence interval.
4-44 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
QA/QC and Verification
General quality assurance/quality control (QA/QC) procedures were applied consistent with the U.S. Inventory
QA/QC plan, which is in accordance with Volume 1, Chapter 6 of the 2006IPCC Guidelines as described in the
introduction of the IPPU chapter (see Annex 8 for more details). More details on the greenhouse gas calculation,
monitoring and QA/QC methods applicable to nitric acid facilities can be found under Subpart V: Nitric Acid
Production of the GHGRP regulation (40 CFR Part 98).34
The main QA/QC activities are related to annual performance testing, which must follow either EPA Method 320 or
ASTM D6348-03. EPA verifies annual facility-level GHGRP reports through a multi-step process that is tailored to
the Subpart (e.g., combination of electronic checks including range checks, statistical checks, algorithm checks,
year-to-year comparison checks, along with manual reviews) to identify potential errors and ensure that data
submitted to EPA are accurate, complete, and consistent. Based on the results of the verification process, EPA
follows up with facilities to resolve mistakes that may have occurred (EPA 20 15).35 EPA's review of observed
trends noted that while emissions have generally mirrored production, in 2015 and 2019 nitric acid production
decreased compared to the previous year and emissions increased. While review is ongoing, based on feedback
from the verification process to date, these changes are due to facility-specific changes (e.g., in the nitric
production process and management of abatement equipment).
Recalculations Discussion
No recalculations were performed for the 1990 through 2021 portion of the time series.
Planned Improvements
Pending resources, EPA is considering a near-term improvement to both review and refine quantitative uncertainty
estimates and the associated qualitative discussion.
4.8 Adipic Acid Production (CRT Source
Category 2B3)
Adipic acid is a white crystalline solid used in the manufacture of synthetic fibers, plastics, coatings, urethane
foams, elastomers, and synthetic lubricants. This reporting category (2B3) includes emissions from the production
of adipic acid. Emissions from fuels consumed for energy purposes during the production of adipic acid are
accounted for as part of fossil fuel combustion in the industrial end-use sector reported under the Energy chapter.
Adipic acid is produced through a two-stage process during which nitrous oxide (N2O) is generated in the second
stage. The first stage of manufacturing usually involves the oxidation of cyclohexane to form a
cyclohexanone/cyclohexanol mixture. The second stage involves oxidizing this mixture with nitric acid to produce
34 See Subpart V monitoring and reporting regulation http://www.ecfr.gov/cgi-bin/text-
idx?tpl=/ecfrbrowse/Title40/40cfr98 main Q2.tpl.
35 See GHGRP Verification Factsheet https://www.epa.gov/sites/production/files/2015-
07/documents/ghgrp verification factsheet.pdf.
Industrial Processes and Product Use 4-45
-------
adipic acid. Nitrous oxide is generated as a byproduct of the nitric acid oxidation stage and is emitted in the waste
gas stream (Thiemens and Trogler 1991). The second stage is represented by the following chemical reaction:
('CH2)5CO(cyclohexanone) + (CH2)zCHOH (cyclohexanol) + wHN03
-» HOOC(CH2)4COOH(adipic acid) + xN20 + yH20
Process emissions from the production of adipic acid vary with the types of technologies and level of emission
controls employed by a facility. In 1990, two major adipic acid-producing plants had N2O abatement technologies
in place and, as of 1998, three major adipic acid production facilities had control systems in place (Reimer et al.
1999). In 2022, thermal reduction was applied as an N2O abatement measure at one adipic acid facility (EPA 2023).
Worldwide, only a few adipic acid plants exist. The United States, Europe, and China are the major producers, with
the United States accounting for the largest share of global adipic acid production capacity in recent years. In 2022,
the United States had two companies with a total of two adipic acid production facilities (one in Texas and one in
Florida), following the ceased operations of a third major production facility at the end of 2015 (EPA 2023).
Commercially, adipic acid is the most important of the aliphatic dicarboxylic acids, which are used to manufacture
polyesters. Eighty-four percent of all adipic acid produced in the United States is used in the production of nylon
6,6; 9 percent is used in the production of polyester polyols; 4 percent is used in the production of plasticizers; and
the remaining 4 percent is accounted for by other uses, including unsaturated polyester resins and food
applications (ICIS 2007). Food grade adipic acid is used to provide some foods with a "tangy" flavor (Thiemens and
Trogler 1991).
Compared to 1990, national adipic acid production in 2022 has increased by 3 percent to approximately 780,000
metric tons (ACC 2023). Nitrous oxide emissions from adipic acid production were estimated to be 2.1 MMT CO2
Eq. (8 kt N2O) in 2022 and are summarized in Table 4-35 and Table 4-36. Over the period 1990 through 2022,
facilities have reduced emissions by 84.5 percent due to the widespread installation of pollution control measures
in the late 1990s. The main reason for the 68 percent decrease in N2O emissions from adipic acid production
between 2021 and 2022 is increased utilization of N2O abatement equipment at one adipic acid production facility.
EPA reviewed GHGRP facility reported information on the date of abatement technology installation in order to
better reflect trends and changes in emissions abatement within the industry across the time series. The facility
using the facility-specific emission factor developed through annual performance testing has reported no
installation and no utilization of N2O abatement technology. The facility using direct measurement of N2O
emissions has reported the use of thermal reduction as an N2O abatement technology; the first unit began
operation in 1980, and the second unit began operation in 2023 (Ard 2024; Ascend 2023).
Significant changes in the amount of time that the N2O abatement device at one facility was in operation has been
the main cause of fluctuating emissions in recent years. These fluctuations are most evident for years where trends
in emissions and adipic acid production were not directly proportional: (1) between 2016 and 2017, (2) between
2017 and 2018, (3) between 2019 and 2020, (4) between 2020 and 2021, and (5) between 2021 and 2022. As
noted above, changes in control measures and abatement technologies at adipic acid production facilities,
including maintenance of equipment, can result in annual emission fluctuations. Little additional information is
available on drivers of trends, and the amount of adipic acid produced is not reported under EPA's GHGRP.
Table 4-35: N2O Emissions from Adipic Acid Production (MMT CO2 Eq.)
Year 1990 2005 2018 2019 2020 2021 2022
Adipic Acid Production 13.5 6.3 9.3 4.7 7.4 6.6 2.1
Table 4-36: N2O Emissions from Adipic Acid Production (kt N2O)
Year 1990 2005 2018 2019 2020 2021 2022
Adipic Acid Production 51 24 35 18 28 25 8
4-46 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
Methodology and Time-Series Consistency
Emissions of N2O from adipic acid production are estimated using methods provided by the 2006IPCC Guidelines,
in accordance with the IPCC methodological decision tree and available data. For 2010 through 2022, a Tier 3
method was used to estimate emissions. For 1990 through 2009, emissions are estimated using both Tier 2 and
Tier 3 methods. Due to confidential business information (CBI), plant names are not provided in this section;
therefore, the four adipic acid-producing facilities that have operated over the time series will be referred to as
Plants 1 through 4. As noted above, one currently operating facility uses thermal reduction as an N2O abatement
technology.
2010 through 2022
All emission estimates for 2010 through 2022 were obtained through analysis of GHGRP data (EPA 2010 through
2023). Facility-level greenhouse gas emissions data were obtained from EPA's GHGRP for the years 2010 through
2022 (EPA 2010 through 2023) and aggregated to national N2O emissions. Consistent with IPCC Tier 3 methods, all
adipic acid production facilities are required to either calculate N2O emissions using a facility-specific emission
factor developed through annual performance testing under typical operating conditions or directly measure N2O
emissions using monitoring equipment.36
1990 through 2009
For years 1990 through 2009, which were prior to EPA's GHGRP reporting, for both Plants 1 and 2, emission
estimates were obtained directly from the plant engineers and account for reductions due to control systems in
place at these plants during the time series. These prior estimates are considered CBI and hence are not published
(Desai 2010, 2011). These estimates were based on continuous process monitoring equipment installed at the two
facilities.
For Plant 4,1990 through 2009 N2O emissions were estimated using the following Tier 2 equation from the 2006
IPCC Guidelines:
Equation 4-5:2006 IPCC Guidelines Tier 2: N20 Emissions From Adipic Acid Production
(Equation 3.8)
Eaa = Qaa X EFaa X (1 - [DF X UF])
where,
Eaa = N2O emissions from adipic acid production, metric tons
Qaa = Quantity of adipic acid produced, metric tons
EFaa = Emission factor, metric ton INhO/metric ton adipic acid produced
DF = N2O destruction factor
UF = Abatement system utility factor
The adipic acid production is multiplied by an emission factor (i.e., N2O emitted per unit of adipic acid produced),
which has been estimated to be approximately 0.3 metric tons of N2O per metric ton of product (IPCC 2006). The
"N2O destruction factor" in the equation represents the percentage of N2O emissions that are destroyed by the
installed abatement technology. The "abatement system utility factor" represents the percentage of time that the
36 Facilities must use standard methods, either EPA Method 320 or ASTM D6348-03 for annual performance testing, and must
follow associated QA/QC procedures during these performance tests consistent with category-specific QC of direct emission
measurements.
Industrial Processes and Product Use 4-47
-------
abatement equipment operates during the annual production period. Plant-specific production data for Plant 4
were obtained across the time series through personal communications (Desai 2010, 2011). The plant-specific
production data were then used for calculating emissions as described above.
For Plant 3, 2005 through 2009 emissions were obtained directly from the plant (Desai 2010, 2011). For 1990
through 2004, emissions were estimated using plant-specific production data and the IPCC factors as described
above for Plant 4. Plant-level adipic acid production for 1990 through 2003 was estimated by allocating national
adipic acid production data to the plant level using the ratio of known plant capacity to total national capacity for
all U.S. plants (ACC 2023; CMR 2001,1998; CW 1999; C&EN 1992 through 1995). For 2004, actual plant production
data were obtained and used for emission calculations (CW 2005).
Plant capacities for 1990 through 1994 were obtained from Chemical & Engineering News, "Facts and Figures" and
"Production of Top 50 Chemicals" (C&EN 1992 through 1995). Plant capacities for 1995 and 1996 were kept the
same as 1994 data. The 1997 plant capacities were taken from Chemical Market Reporter, "Chemical Profile: Adipic
Acid" (CMR 1998). The 1998 plant capacities for all four plants and 1999 plant capacities for three of the plants
were obtained from Chemical Week, Product Focus: Adipic Acid/Adiponitrile (CW 1999). Plant capacities for the
year 2000 for three of the plants were updated using Chemical Market Reporter, "Chemical Profile: Adipic Acid"
(CMR 2001). For 2001 through 2003, the plant capacities for three plants were held constant at year 2000
capacities. Plant capacity for 1999 to 2003 for the one remaining plant was kept the same as 1998.
National adipic acid production data (see Table 4-37) from 1990 through 2022 were obtained from the American
Chemistry Council (ACC 2023).
Table 4-37: Adipic Acid Production (kt)
Year
1990
2005
2018
2019
2020
2021
2022
Production (kt)
755
865 I
825
810
710
760
780
Methodological approaches were applied to the entire time series to ensure consistency in emissions from 1990
through 2022. The methodology for adipic acid production spliced activity data from multiple sources: plant-
specific emissions data and publicly available plant capacity data for 1990 through 2009 and GHGRP emission data
starting in 2010. Consistent with the 2006 IPCC Guidelines, the overlap technique was applied to compare the two
data sets for years where there was overlap, with findings that the data sets were consistent and adjustments
were not needed.
Uncertainty
Uncertainty associated with N2O emission estimates includes the methods used by companies to monitor and
estimate emissions. While some information has been obtained through outreach with facilities, limited
information is available over the time series on these methods, abatement technology destruction and removal
efficiency rates, and plant-specific production levels. EPA assigned an uncertainty range of ±5 percent and a
normal probability density function for facility-reported N2O emissions, and using this suggested uncertainty
provided in section 3.4.3.2 of the 2006 IPCC Guidelines is appropriate based on expert judgment (RTI 2023).
The results of this Approach 2 quantitative uncertainty analysis are summarized in Table 4-38. Nitrous oxide
emissions from adipic acid production for 2022 were estimated to be between 2.0 and 2.2 MMT CO2 Eq. at the 95
percent confidence level. These values indicate a range of approximately 4 percent below to 4 percent above the
2022 emission estimate of 2.1 MMT CO2 Eq.
4-48 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
Table 4-38: Approach 2 Quantitative Uncertainty Estimates for N2O Emissions from Adipic
Acid Production (MMT CO2 Eq. and Percent)
Source
Gas
2022 Emission Estimate
Uncertainty Range Relative to Emission Estimate"
(MMT CO . Eq.)
(MMTCO. Eq.)
(%)
Lower Upper
Lower Upper
Bound Bound
Bound Bound
Adipic Acid Production
N20
2.1
2.0 2.2
-4% +4%
a Range of emission estimates predicted by Monte Carlo stochastic simulation for a 95 percent confidence interval.
QA/QC and Verification
General quality assurance/quality control (QA/QC) procedures were applied consistent with the U.S. Inventory
QA/QC plan, which is in accordance with Volume 1, Chapter 6 of the 2006IPCC Guidelines as described in the
introduction of the IPPU chapter (see Annex 8 for more details).
More details on the greenhouse gas calculation, monitoring and QA/QC methods applicable to adipic acid facilities
can be found under Subpart E (Adipic Acid Production) of the GHGRP regulation (40 CFR Part 98).37 The main
QA/QC activities are related to annual performance testing, which must follow either EPA Method 320 or ASTM
D6348-03. EPA verifies annual facility-level GHGRP reports through a multi-step process (e.g., combination of
electronic checks and manual reviews) to identify potential errors and ensure that data submitted to EPA are
accurate, complete, and consistent (EPA 20 15).38 Based on the results of the verification process, EPA follows up
with facilities to resolve mistakes that may have occurred. The post-submittals checks are consistent with a
number of general and category-specific QC procedures, including range checks, statistical checks, algorithm
checks, and year-to-year comparisons of reported data.
Recalculations Discussion
No recalculations were performed for the 1990 through 2021 portion of the time series.
Planned Improvements
EPA has no specific planned improvements related to adipic acid.
4.9 Caprolactam, Glyoxal and Glyoxylic
Acid Production (CRT Source Category 2B4)
This reporting category (2B4) includes emissions from the production of caprolactam, glyoxal (ethanedial), and
glyoxylic acid. Emissions from fuels consumed for energy purposes during the production of caprolactam, glyoxal,
37 See http://www.ecfr.gov/cgi-bin/text-idx7tpk/ecfrbrowse/Title40/40cfr98 main Q2.tpl.
38 See https://www.epa.gov/sites/production/files/2015-07/documents/ghgrp verification factsheet.pdf.
Industrial Processes and Product Use 4-49
-------
and glyoxylic acid are accounted for as part of fossil fuel combustion in the industrial end-use sector reported
under the Energy chapter.
Caprolactam
Caprolactam (CsHnNO) is a colorless monomer produced for nylon-6 fibers and plastics. A substantial proportion
of the fiber is used in carpet manufacturing. Most commercial processes used for the manufacture of caprolactam
begin with benzene, but toluene can also be used. The production of caprolactam can give rise to emissions of
nitrous oxide (N2O).
During the production of caprolactam, emissions of N2O can occur from the ammonia oxidation step, emissions of
carbon dioxide (CO2) from the ammonium carbonate step, emissions of sulfur dioxide (SO2) from the ammonium
bisulfite step, and emissions of non-methane volatile organic compounds (NMVOCs). Emissions of CO2, SO2 and
NMVOCs from the conventional process are unlikely to be significant in well-managed plants. Modified
caprolactam production processes are primarily concerned with elimination of the high volumes of ammonium
sulfate that are produced as a byproduct of the conventional process (IPCC 2006).
In the most commonly used process where caprolactam is produced from benzene, benzene is hydrogenated to
cyclohexane which is then oxidized to produce cyclohexanone (CsHioO). The classical route (Raschig process) and
basic reaction equations for production of caprolactam from cyclohexanone are (IPCC 2006):
NO
Oxidation of NH3 to
' 3 N02
I
C02
NH3 reacted with——to yield ammonium carbonate (NH4)2C03
H20
I
NO
(NH4)2C03 reacted with—— (from NH3 oxidation)to yield ammonium nitrite (NH4N02)
N 02
I
S 02
NH3 reacted with——to yield ammonium bisulphite (NH4HS03)
H20
I
NH4N02 and (NH4HS03)reacted to yield hydroxylamine disulphonate (N0H(S03NH4)2)
I
(N0H(S03NH4)2) hydrolised to yield hydroxylamine sulphate ((NH2OH)2. H2S04) and
ammonium sulphate ((NH4)2S04)
I
Cylohexanone reaction-.
1
C6H10O +-(NH20H)2.H2S04(+NH3 and H2S04) -> C6H10NOH + (NH4)2S04 + H20
I
Beckmann rearrangement:
C6H10NOH (+H2S04 and S02) -> C^NO. H2S04 (+4NH3 and H20) -> C^NO + 2(NH4)2S04
In 2004, three facilities produced caprolactam in the United States (ICIS 2004). Another facility, Evergreen
Recycling, was in operation from 2000 to 2001 (ICIS 2004; Textile World 2000) and from 2007 through 2015 (Shaw
4-50 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
2015). Caprolactam production at Fibrant LLC (formerly DSM Chemicals) in Georgia ceased in 2018 (Cline 2019). As
of 2022, two companies in the United States produced caprolactam at two facilities: AdvanSix (formerly
Honeywell) in Virginia (AdvanSix 2023) and BASF in Texas (BASF 2023).
Nitrous oxide emissions from caprolactam production in the United States were estimated to be 1.3 MMT CO2 Eq.
(5 kt N2O) in 2022 and are summarized in Table 4-39 and Table 4-40. National emissions from caprolactam
production decreased by approximately 10.5 percent over the period of 1990 through 2022. Emissions in 2022
increased by approximately 9.8 percent from the 2021 levels. This annual increase returned caprolactam
production to levels consistent with 2017 before the COVID-19 pandemic.
Table 4-39: N2O Emissions from Caprolactam Production (MMT CO2 Eq.)
Year 1990 2005 2018 2019 2020 2021
2022
Caprolactam Production 1.5 1.9 | 1.3 1.2 1.1 1.2
1.3
Table 4-40: N2O Emissions from Caprolactam Production (kt N2O)
Year 1990 2005 2018 2019 2020 2021
2022
Caprolactam Production 6 7 5 5 4 5
5
Glyoxal
Glyoxal is mainly used as a crosslinking agent for vinyl acetate/acrylic resins, disinfectant, gelatin hardening agent,
textile finishing agent (permanent-press cotton, rayon fabrics), and wet-resistance additive (paper coatings) (IPCC
2006). It is also used for enhanced oil-recovery. It is produced from oxidation of acetaldehyde with concentrated
nitric acid, or from the catalytic oxidation of ethylene glycol, and N2O is emitted in the process of oxidation of
acetaldehyde.
Glyoxal (ethanedial) (C2H2O2) is produced from oxidation of acetaldehyde (ethanal) (C2H4O) with concentrated
nitric acid (HNO3). Glyoxal can also be produced from catalytic oxidation of ethylene glycol (ethanediol)
(CH2OHCH2OH).
Glyoxylic Acid
Glyoxylic acid is produced by nitric acid oxidation of glyoxal. Glyoxylic acid is used for the production of synthetic
aromas, agrochemicals, and pharmaceutical intermediates (IPCC 2006).
Preliminary data suggests that glyoxal and glyoxylic acid may be produced in small quantities domestically but are
largely imported to the United States. EPA does not currently estimate the emissions associated with the
production of glyoxal and glyoxylic acid because activity data are not available. See Annex 5 for more information.
Methodology and Time-Series Consistency
Emissions of N2O from the production of caprolactam are calculated using the Tier 1 methodology from the 2006
IPCC Guidelines, in accordance with the IPCC methodological decision tree and available data. The Tier 1 equation
is as follows:
Equation 4-6:2006 IPCC Guidelines Tier 1: N20 Emissions From Caprolactam Production
(Equation 3.9)
£jv2o = * CP
where,
Enzo = Annual N2O Emissions (kg)
Industrial Processes and Product Use 4-51
-------
EF = N2O emission factor (default) (kg N20/metric ton caprolactam produced)
CP = Caprolactam production (metric tons)
During the caprolactam production process, N2O is generated as a byproduct of the high temperature catalytic
oxidation of ammonia (NH3), which is the first reaction in the series of reactions to produce caprolactam. The
amount of N2O emissions can be estimated based on the chemical reaction shown above. Based on this formula,
which is consistent with an IPCCTier 1 approach, approximately 111.1 metric tons of caprolactam are required to
generate one metric ton of N2O, resulting in an emission factor of 9.0 kg N2O per metric ton of caprolactam (IPCC
2006). When applying the Tier 1 method, the 2006 IPCC Guidelines state that it is good practice to assume that
there is no abatement of N2O emissions and to use the highest default emission factor available in the guidelines.
In addition, EPA did not find support for the use of secondary catalysts to reduce N2O emissions, such as those
employed at nitric acid plants.
The activity data for caprolactam production (see Table 4-41) from 1990 to 2022 were obtained from the American
Chemistry Council's Guide to the Business of Chemistry (ACC 2023). EPA will continue to analyze and assess
alternative sources of production data as a quality control measure.
Table 4-41: Caprolactam Production (kt)
Year
1990 2005
2018
2019
2020
2021
2022
Production (kt)
626 a 795
530
515
480
510
560
Carbon dioxide and methane (CH4) emissions may also occur from the production of caprolactam, but currently the
IPCC does not have methodologies for calculating these emissions associated with caprolactam production.
Methodological approaches, consistent with the 2006 IPCC Guidelines, have been applied to the entire time series
to ensure consistency in emissions from 1990 through 2022.
Uncertainty
Estimation of emissions of N2O from caprolactam production can be treated as analogous to estimation of
emissions of N2O from nitric acid production. Both production processes involve an initial step of NH3 oxidation,
which is the source of N2O formation and emissions (IPCC 2006). Therefore, uncertainties for the default emission
factor values in the 2006 IPCC Guidelines are an estimate based on default values for nitric acid plants. In general,
default emission factors for gaseous substances have higher uncertainties because mass values for gaseous
substances are influenced by temperature and pressure variations and gases are more easily lost through process
leaks. The default values for caprolactam production have a relatively high level of uncertainty due to the limited
information available (IPCC 2006). EPA assigned uncertainty bounds of ±5 percent for caprolactam production,
based on expert judgment. EPA assigned an uncertainty range of ±40 percent for the N2O emission factor, and
using this suggested uncertainty provided in Section 3.5.2.1 of the 2006 IPCC Guidelines is appropriate based on
expert judgment (RTI 2023). Per this expert judgment, a normal probability density function was assigned for
activity data, and a triangular probably density function was assigned for the emission factor.
The results of this Approach 2 quantitative uncertainty analysis are summarized in Table 4-42. Nitrous oxide
emissions from caprolactam, glyoxal and glyoxylic acid production for 2022 were estimated to be between 0.9 and
1.8 MMT CO2 Eq. at the 95 percent confidence level. These values indicate a range of approximately 31 percent
below to 31 percent above the 2022 emission estimate of 1.3 MMT CO2 Eq.
4-52 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
Table 4-42: Approach 2 Quantitative Uncertainty Estimates for N2O Emissions from
Caprolactam, Glyoxal and Glyoxylic Acid Production (MMT CO2 Eq. and Percent)
Source
Gas
2022 Emission Estimate
Uncertainty Range Relative to Emission Estimate-1
(MMT CO . Eq.)
(MMTCO. Eq.)
<%)
Lower Upper
Lower Upper
Bound Bound
Bound Bound
Caprolactam Production
N20
1.3
CO
T—1
cn
0
-31% +31%
a Range of emission estimates predicted by Monte Carlo stochastic simulation for a 95 percent confidence interval.
QA/QC and Verification
General quality assurance/quality control (QA/QC) procedures were applied consistent with the U.S. Inventory
QA/QC plan, which is in accordance with Volume 1, Chapter 6 of the 2006IPCC Guidelines as described in the
introduction of the IPPU chapter (see Annex 8 for more details).
Recalculations Discussion
Recalculations were performed for 2020 and 2021 to reflect updated caprolactam production data from the
American Chemistry Council's Guide to the Business of Chemistry (ACC 2023). Compared to the previous Inventory,
annual N2O emissions decreased by 2 percent in 2020 and 2021, with a decrease of 0.02 MMT CO2 Eq. in 2020 and
2021.
Planned Improvements
Pending resources, EPA will research other available datasets for caprolactam production and industry trends,
including facility-level data. EPA continues to research available activity data and emissions associated with the
production of glyoxal and glyoxylic acid. Preliminary data suggests that glyoxal and glyoxylic acid may be produced
in small quantities domestically but are largely imported to the United States. See Annex 5 for more information.
This planned improvement is subject to data availability and will be implemented in the medium- to long-term.
4.10 Carbide Production and Consumption
(CRT Source Category 2B5 & 2B10)
Carbon dioxide (CO2) and methane (CH4) are emitted from the production of silicon carbide (SiC), a material used
for industrial abrasive, metallurgical, and other non-abrasive applications in the United States, and CO2 is emitted
from the consumption of SiC. Per the IPCC methodological guidance, emissions from fuels consumed for energy
purposes during the production of silicon carbide are accounted for in the industrial end-use sector reported under
the Energy chapter. Additionally, some metallurgical and non-abrasive applications of SiC are emissive at high
temperatures due to the SiC oxidation temperature (Biscay 2021). While emissions should be accounted for where
they occur based on 2006 IPCC Guidelines, emissions from SiC consumption are accounted for here until additional
data on SiC consumption by end-use are available. The reporting category (2B5) includes emissions from the
production of SiC, and the reporting category (2B10) includes emissions from the consumption of SiC.
To produce SiC, silica sand or quartz (Si02> is reacted with carbon (C) in the form of petroleum coke. A portion
(about 35 percent) of the carbon contained in the petroleum coke is retained in the SiC. The remaining carbon is
Industrial Processes and Product Use 4-53
-------
emitted as CO2, Cm, or carbon monoxide (CO). The overall reaction is shown below, but in practice, it does not
proceed according to stoichiometry:
Si02 + 3C —> SiC + 2CO (+ 02 —> 2COz)
Carbon dioxide and CH4 are also emitted during the production of calcium carbide, a chemical used to produce
acetylene. Carbon dioxide is implicitly accounted for in the storage factor calculation for the non-energy use of
petroleum coke in the Energy chapter.
Markets for manufactured abrasives, including SiC, are heavily influenced by activity in the U.S. manufacturing
sector, especially in the aerospace, automotive, furniture, housing, and steel manufacturing sectors. Specific
applications of abrasive-grade SiC in 2018 included antislip abrasives, blasting abrasives, bonded abrasives, coated
abrasives, polishing and buffing compounds, tumbling media, and wire-sawing abrasives (USGS 2021).
Approximately 50 percent of SiC is used in metallurgical applications, which include primarily iron and steel
production, and other non-abrasive applications, which include use in advanced or technical ceramics and
refractories (USGS 2023a; Washington Mills 2023).
As a result of the economic downturn in 2008 and 2009, demand for SiC decreased in those years. Low-cost
imports, particularly from China, combined with high relative operating costs for domestic producers, continue to
put downward pressure on the production of SiC in the United States. Consumption of SiC in the United States has
recovered somewhat from its low in 2009 to 2020; 2021 consumption data was withheld to avoid disclosing
company proprietary data (USGS 1991b through 2021), and 2022 USGS data has not yet been released.
Silicon carbide was manufactured by two facilities in the United States, one of which produced primarily non-
abrasive SiC (USGS 2021). USGS production values for the United States consists of SiC used for abrasives and for
metallurgical and other non-abrasive applications (USGS 2021). During the COVID-19 pandemic in 2020, the U.S.
Department of Homeland Security considered abrasives manufacturing part of the critical manufacturing sector,
and as a result, pandemic "stay-at-home" orders issued in March 2020 did not affect the abrasives manufacturing
industry. These plants remained at full operation (USGS 2021a). In 2022, imports and exports continued to recover
from the negative effects of the COVID-19 pandemic (USGS 2023b). Consumption of SiC increased by
approximately 27 percent from 2021 to 2022, rising above pre-pandemic levels (U.S. Census Bureau 2005 through
2022).
Carbon dioxide emissions from SiC production and consumption in 2022 were 0.2 MMT CO2 Eq. (210 kt CO2), which
are about 14 percent lower than emissions in 1990 (see Table 4-43 and Table 4-44). Approximately 50 percent of
these emissions resulted from SiC production, while the remainder resulted from SiC consumption. Methane
emissions from SiC production in 2022 were 0.01 MMT CO2 Eq. (0.5 kt CH4) (see Table 4-43 and Table 4-44). These
tables indicate minor changes in emissions in recent years.
Table 4-43: CO2 and CH4 Emissions from Silicon Carbide Production and Consumption (MMT
C02 Eq.)
Year
1990
2005
2018
2019
2020
2021
2022
SiC Production
C02
0.2
0.1 1
0.1
0.1
0.1
0.1
0.1
ch4
+
+
+
+
+
+
+
SiC Consumption
C02
0.1
0.1
0.1
0.1
0.1
0.1
0.1
Total
0.2
0.2
0.2
0.2
0.2
0.2
0.2
+ Does not exceed 0.05 MMT C02 Eq.
Note: Totals may not sum due to independent rounding.
4-54 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
Table 4-44: CO2 and CH4 Emissions from Silicon Carbide Production and Consumption (kt)
Year
1990
2005
2018
2019
2020
2021
2022
SiC Production
C02
170
92 |
92
92
92
92
105
ch4
1
+
+
+
+
+
+
SiC Consumption
C02
73
121
93
84
62
80
105
+ Does not exceed 0.5 kt.
Methodology and Time-Series Consistency
Emissions of CO2 and CH4 from the production of SiC are calculated using the Tier 1 method from the 2006IPCC
Guidelines, in accordance with the IPCC methodological decision tree and available data. Emissions of CO2 from the
consumption of SiC are a country-specific source calculated using a country-specific methodology based on
available data. The 2006 IPCC Guidelines do not provide guidance for estimating emissions from use of SiC or SiC
consumption, but the country-specific methodology used is based on the stoichiometry of SiC consumption and is
compatible with the 2006 IPCC Guidelines and consistent with a Tier 1 approach.
Annual estimates of SiC production were multiplied by the default emission factors, as shown below:
Equation 4-78:2006 IPCC Guidelines Tier 1: Emissions from Carbide Production (Equation
3.11)
/I metric ton\
Esc,CH4 = EFSCiCH4 x Qsc x ^ 1000fc J
ESc,C02 — EFSC,C02 X Qs
(1 metric ton\
1000 kg
where,
Esc,co2 = CO2 emissions from production of SiC, metric tons
EFsc,co2 = Emission factor for production of SiC, metric ton CCh/metric ton SiC
Qsc = Quantity of SiC produced, metric tons
Esc,ch4 = Cm emissions from production of SiC, metric tons
EFsc,ch4 = Emission factor for production of SiC, kilogram Cl-U/metric ton SiC
Emission factors were taken from the 2006 IPCC Guidelines:
• 2.62 metric tons C02/metric ton SiC
• 11.6 kg Cl-U/metric ton SiC
Production data includes silicon carbide manufactured for abrasive applications as well as for metallurgical and
other non-abrasive applications (USGS 2021).
Silicon carbide industrial abrasives production data for 1990 through 2022 were obtained from the U.S. Geological
Survey (USGS) Minerals Yearbook: Manufactured Abrasives (USGS 1991a through 2021; USGS 2023a). Silicon
carbide production data published by USGS have been rounded to the nearest 5,000 metric tons to avoid
disclosing company proprietary data. For the period 1990 through 2001, reported USGS production data include
production from two facilities located in Canada that ceased operations in 1995 and 2001. Using SiC production
data from Canada (ECCC 2022), U.S. SiC production for 1990 through 2001 was adjusted to reflect only U.S.
production.
Emissions from SiC consumption are calculated by multiplying the annual SiC consumption for metallurgical and
other non-abrasive uses by the carbon content of SiC (about 30.0 percent), which is based on the molecular weight
Industrial Processes and Product Use 4-55
-------
of SiC, and converted to CO2. This conversion calculation equates to 1.10 and is consistent with the IPCC default
emission factor to calculate CO2 emissions from the consumption of acetylene, a calcium carbide product, and
demonstrates a methodology consistent with the 2006 IPCC Guidelines. The amount of SiC used by other non-
abrasive applications is determined by multiplying the annual SiC consumption by 50 percent (the percentage that
the USGS allocates as usage by metallurgical and other non-abrasive applications) and then subtracting the
amount of SiC used for metallurgical applications (USGS 1991a through 2021; USGS 2023a).
Emissions from SiC consumption are estimated for the entire time series using USGS consumption data (USGS
1991b through 2021) and data from the U.S. International Trade Commission (USITC) database on net imports and
exports of SiC (U.S. Census Bureau 2005 through 2022) (Table 4-45). Total annual SiC consumption (utilization) was
estimated by subtracting annual exports of SiC from the total of annual national SiC production and annual
imports. Data on the annual consumption of SiC for metallurgical uses were obtained from USGS Minerals
Yearbook: Silicon (USGS 1991b-2021; USGS 2023c). USGS withheld consumption data for metallurgical uses from
publication for 2017, 2018, and 2021, and 2022 due to concerns of disclosing company-specific sensitive
information, SiC consumption for 2017 and 2018 were estimated using 2016 values, and SiC consumption for 2021
and 2022 were estimated using the 2020 value (USGS 2023c). Additionally, as the USGS has not yet released the
2022 data, SiC consumption for 2022 was estimated using the 2020 value.
The petroleum coke portion of the total CO2 process emissions from silicon carbide production is adjusted for
within the Energy chapter, as these fuels were consumed during non-energy related activities. Additional
information on the adjustments made within the Energy sector for non-energy use of fuels is described in both the
Methodology section of CO2 from Fossil Fuel Combustion (Section 3.1) and Annex 2.1, Methodology for Estimating
Emissions of CO2 from Fossil Fuel Combustion.
Table 4-45: Production and Consumption of Silicon Carbide (Metric Tons)
Year
1990
2005
2018
2019
2020
2021
2022
SiC Production
SiC Consumption
65,000 I
132,465 1
35,000
220,149
35,000
168,526
35,000
152,412
35,000
113,756
35,000
146,312
40,000
191,133
Methodological approaches were applied to the entire time series to ensure consistency in emissions from 1990
through 2022.
Uncertainty
Silicon carbide production data published by the USGS is rounded to the nearest 5,000 tons and has been
consistently reported at 35,000 tons since 2003 to avoid disclosure of company proprietary data. This translates to
an uncertainty range of ±7 percent and a normal probability density function for SiC production (USGS 2021).
There is uncertainty associated with the emission factors used because they are based on stoichiometry as
opposed to monitoring of actual SiC production plants. An alternative is to calculate emissions based on the
quantity of petroleum coke used during the production process rather than on the amount of silicon carbide
produced; however, these data were not available. For CH4, there is also uncertainty associated with the hydrogen-
containing volatile compounds in the petroleum coke (IPCC 2006). EPA assigned an uncertainty of ±10 percent for
the Tier 1 CO2 and CFU emission factors for the SiC production processes, and using this suggested uncertainty
provided in Section 3.6.3.1 of the 2006 IPCC Guidelines is appropriate based on expert judgment (RTI 2023). Per
this expert judgment, a triangular probability density function was assigned for emission factors. There is also
uncertainty associated with the use or destruction of CH4 generated from the process, in addition to uncertainty
associated with levels of production, net imports, consumption levels, and the percent of total consumption that is
attributed to metallurgical and other non-abrasive uses. EPA assigned an uncertainty range of ±5 percent for the
primary data inputs for consumption (i.e., crude imports, ground and refined imports, crude exports, ground and
refined exports, utilization [metallurgical applications]) to calculate overall uncertainty from SiC production, and
using this suggested uncertainty provided in Section 3.6.3.2 of the 2006 IPCC Guidelines is appropriate based on
expert judgment (RTI 2023).
4-56 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
The results of the Approach 2 quantitative uncertainty analysis are summarized in Table 4-46. Silicon carbide
production and consumption CO2 emissions from 2022 were estimated to be between 10 percent below and 10
percent above the emission estimate of 0.2 MMT CO2 Eq. at the 95 percent confidence level. Silicon carbide
production Cm emissions were estimated to be between 10 percent below and 11 percent above the emission
estimate of 0.01 MMT CO2 Eq. at the 95 percent confidence level.
Table 4-46: Approach 2 Quantitative Uncertainty Estimates for CH4 and CO2 Emissions from
Silicon Carbide Production and Consumption (MMT CO2 Eq. and Percent)
Source
Gas
2022 Emission Estimate
(MMTCO. Eq.)
Uncertainty Range Relative to Emission Estimate-1
(MMTCO' Eq.) (%)
Lower
Upper
Lower
Upper
Bound
Bound
Bound
Bound
Silicon Carbide Production
and Consumption
C02
0.2
0.2
0.2
-10%
+10%
Silicon Carbide Production
ch4
+
+
+
-10%
+11%
+ Does not exceed 0.05 MMT C02 Eq.
a Range of emission estimates predicted by Monte Carlo stochastic simulation for a 95 percent confidence interval.
QA/QC and Verification
General quality assurance/quality control (QA/QC) procedures were applied consistent with the U.S. Inventory
QA/QC plan, which is in accordance with Volume 1, Chapter 6 of 2006IPCC Guidelines as described in the
introduction of the IPPU chapter (see Annex 8 for more details).
Recalculations Discussion
No recalculations were performed for the 1990 through 2021 portion of the time series.
Planned Improvements
EPA is initiating research for data on SiC consumption by end-use for consideration in updating emissions
estimates from SiC consumption and to account for emissions where they occur. This planned improvement is
subject to data availability and will be implemented in the medium- to long-term given significance of emissions.
EPA has not integrated aggregated facility-level GHGRP information to inform estimates of CO2 and Cl-Ufrom SiC
production and consumption. The aggregated information (e.g., activity data and emissions) associated with silicon
carbide did not meet criteria to shield underlying confidential business information (CBI) from public disclosure.
EPA plans to examine the use of GHGRP silicon carbide emissions data for possible use in emission estimates
consistent with both Volume 1, Chapter 6 of the 2006 IPCC Guidelines and the latest IPCC guidance on the use of
facility-level data in national inventories. This planned improvement is ongoing and has not been incorporated into
this Inventory report. This is a long-term planned improvement.
4.11 Titanium Dioxide Production (CRT
Source Category 2B6)
Titanium dioxide (Ti02> is manufactured using one of two processes: the chloride process and the sulfate process.
The chloride process uses petroleum coke and chlorine as raw materials and emits process-related carbon dioxide
Industrial Processes and Product Use 4-57
-------
(CO2). The sulfate process does not use petroleum coke or other forms of carbon as a raw material and does not
emit CO2. The reporting category (2B6) includes emissions from production of TiC>2. In accordance with the IPCC
methodological guidance, emissions from fuels consumed for energy purposes during the production of titanium
dioxide are accounted for as part of fossil fuel combustion in the industrial end-use sector reported under the
Energy chapter. The chloride process is based on the following chemical reactions and does emit CO2:
2FeTi03 + 7Cl2 + 3C —> 2TiCl^ + 2FeCl^ + 3C02
2TiCl4 + 202 ~~* 2Ti02 ~l~ ^Cl2
The carbon in the first chemical reaction is provided by petroleum coke, which is oxidized in the presence of the
chlorine and FeTiCh (rutile ore) to form CO2. Since 2004, all TiC>2 produced in the United States has been produced
using the chloride process, and a special grade of "calcined" petroleum coke is manufactured specifically for this
purpose.
The principal use of TiC>2 is as a white pigment in paint, lacquers, and varnishes. It is also used as a pigment in the
manufacture of plastics, paper, and other products. In 2022, U.S. TiC>2 production totaled 1,100,000 metric tons
(USGS 2023b). Five plants produced TiC>2 in the United States in 2022.
Emissions of CChfrom titanium dioxide production in 2022 were estimated to be 1.5 MMT CO2 Eq. (1,474 kt CO2),
which represents an increase of 23 percent since 1990 (see Table 4-47 and Table 4-48). Compared to 2021,
emissions from titanium dioxide production remained the same because production was consistent from 2021 to
2022. Annual production dipped in 2019 and 2020 and increased in 2021 and 2022.
Table 4-47: CO2 Emissions from Titanium Dioxide (MMT CO2 Eq.)
Year 1990 2005 2018 2019 2020 2021
2022
Titanium Dioxide 1.2 1 1.8 1 1.5 1.3 1.3 1.5
1.5
Table 4-48: CO2 Emissions from Titanium Dioxide (kt CO2)
Year 1990 2005 2018 2019 2020 2021
2022
Titanium Dioxide 1,195 | 1,755 | 1,541 1,340 1,340 1,474
1,474
Methodology and Time-Series Consistency
Emissions of CO2 from TiC>2 production are calculated using a Tier 1 method from the 2006 IPCC Guidelines, in
accordance with the IPCC methodological decision tree and available data. Annual national TiC>2 production is
multiplied by chloride process-specific emission factors provided by IPCC (IPCC 2006). The Tier 1 equation is as
follows:
Equation 4-9:2006 IPCC Guidelines Tier 1: CO2 Emissions from Titanium Production (Equation
3.12)
Etd = EFtd X Qtd
where,
Etd = CO2 emissions from Ti02 production, metric tons
EFtd = Emission factor (chloride process), metric ton CCh/metric ton TiC>2
Qtd = Quantity of Ti02 produced, metric tons
The petroleum coke portion of the total CO2 process emissions from Ti02 production is adjusted for within the
Energy chapter as these fuels were consumed during non-energy related activities. Additional information on the
adjustments made within the Energy sector for Non-Energy Use of Fuels is described in both the Methodology
4-58 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
section of CO2 from Fossil Fuel Combustion (Section 3.1 Fossil Fuel Combustion) and Annex 2.1, Methodology for
Estimating Emissions of CO2 from Fossil Fuel Combustion.
Data were obtained for the total amount of TiC>2 produced each year. For years prior to 2004, it was assumed that
TiC>2 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 production since 2004 used the chloride process (USGS 2005). An emission factor
of 1.34 metric tons CCh/metric ton TiC>2 was applied to the estimated chloride-process production (IPCC 2006). It
was assumed that all TiC>2 produced using the chloride process was produced using petroleum coke, although
some TiC>2 may have been produced with graphite or other carbon inputs.
The emission factor for the TiC>2 chloride process was taken from the 2006 IPCC Guidelines. Titanium dioxide
production data and the percentage of total TiC>2 production capacity that used the chloride process for 1990
through 2018 (see Table 4-49) were obtained through the U.S. Geological Survey (USGS) Minerals Yearbook:
Titanium (USGS 1991 through 2022). Production data for 2019 and 2020 were obtained from the USGS Minerals
Yearbook: Titanium, advanced data release of the 2020 tables (USGS 2023a). Production data for 2020 and 2021
were obtained from the Minerals Commodity Summaries: Titanium and Titanium Dioxide (USGS 2023b).39 Data on
the percentage of total TiC>2 production capacity that used the 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 (Gambogi 2002).
By 2002, only one sulfate process plant remained online in the United States, and this plant closed in 2004 (USGS
2005).
Table 4-49: Titanium Dioxide Production (kt)
Year
1990
2005
2018
2019
2020
2021
2022
Production
979
1,310
1,150
1,000
1,000
1,100
1,100
Methodological approaches were applied to the entire time series to ensure consistency in emissions from 1990
through 2022.
Uncertainty
Each year, the USGS collects titanium industry data for titanium mineral and pigment production operations. If
TiC>2 pigment plants do not respond, production from the operations is estimated based on prior year production
levels and industry trends. Variability in response rates fluctuates from 67 to 100 percent of TiC>2 pigment plants
over the time series. EPA currently uses an uncertainty range of ±5 percent and a normal probability density
function for the primary data inputs (i.e., TiC>2 production and chloride process capacity values) to calculate overall
uncertainty from TiC>2 production, and using this suggested uncertainty provided in Section 3.7.3.2 of the 2006
IPCC Guidelines is appropriate based on expert judgment (RTI 2023). Additionally, the EPA uses an uncertainty
range of ±15 percent and a triangular probability density function for the CO2 chloride process carbon
consumption rate, and using this uncertainty provided in Section 3.7.2.2 of the 2006 IPCC Guidelines is
representative of operations in the United States, based on expert judgment (RTI 2023).
Although some TiC>2 may be produced using graphite or other carbon inputs, information and data regarding these
practices were not available. Titanium dioxide produced using graphite inputs, for example, may generate differing
amounts of CChper unit of TiC>2 produced as compared to that generated using petroleum coke in production.
While the most accurate method to estimate emissions would be to base calculations on the amount of reducing
39 EPA has not integrated aggregated facility-level GHGRP information for titanium dioxide production facilities (40 CFR Part 98
Subpart EE). The relevant aggregated information (activity data, emission factor) from these facilities did not meet criteria to
shield underlying CBI from public disclosure.
Industrial Processes and Product Use 4-59
-------
agent used in each process rather than on the amount of TiC>2 produced, sufficient data were not available to do
so.
As of 2004, the last remaining sulfate-process plant in the United States closed. Since annual TiCh production was
not reported by USGS by the type of production process used (chloride or sulfate) prior to 2004 and only the
percentage of total production capacity by process was reported, the percent of total TiC>2 production capacity that
was attributed to the chloride process was multiplied by total TiC>2 production to estimate the amount of TiC>2
produced using the chloride process. Finally, the emission factor was applied uniformly to all chloride-process
production, and no data were available to account for differences in production efficiency among chloride-process
plants. In calculating the amount of petroleum coke consumed in chloride-process TiC>2 production, literature data
were used for petroleum coke composition. Certain grades of petroleum coke are manufactured specifically for
use in the TiC>2 chloride process; however, this composition information was not available. EPA assigned an
uncertainty range of ±15 percent and a triangular probability density function for the Tier 1CO2 emission factor for
the titanium dioxide (chloride route) production process, and using this uncertainty provided in Table 3.9 of the
2006IPCC Guidelines is representative of operations in the United States based on expert judgment (RTI 2023).
The results of the Approach 2 quantitative uncertainty analysis are summarized in Table 4-49. Titanium dioxide
consumption CO2 emissions from 2022 were estimated to be between 1.3 and 1.7 MMT CO2 Eq. at the 95 percent
confidence level. This indicates a range of approximately 12 percent below and 13 percent above the emission
estimate of 1.5 MMT CO2 Eq.
Table 4-50: Approach 2 Quantitative Uncertainty Estimates for CO2 Emissions from Titanium
Dioxide Production (MMT CO2 Eq. and Percent)
Source
Gas
2022 Emission Estimate
Uncertainty Range Relative to Emission Estimate-'
(MMT CO . Eq.)
(MMTCO. Eq.)
(%)
Lower Upper
Lower Upper
Bound Bound
Bound Bound
Titanium Dioxide Production
C02
1.5
1.3 1.7
-12% +13%
a Range of emission estimates predicted by Monte Carlo stochastic simulation for a 95 percent confidence interval.
QA/QC and Verification
General quality assurance/quality control (QA/QC) procedures were applied consistent with the U.S. Inventory
QA/QC plan, which is in accordance with Volume 1, Chapter 6 of the 2006 IPCC Guidelines as described in the
introduction of the IPPU chapter (see Annex 8 for more details).
Recalculations Discussion
Updated USGS data on Ti02 production was available for 2019 and 2020, resulting in updated emissions estimates
for those years. Compared to the previous Inventory, emissions for 2019 decreased by 9 percent (134 kt CO2), and
emissions for 2020 increased by 12 percent (147 kt CO2).
Planned Improvements
EPA plans to examine the use of GHGRP titanium dioxide emissions and other data for possible use in emission
estimates consistent with both Volume 1, Chapter 6 of the 2006 IPCC Guidelines and the latest IPCC guidance on
the use of facility-level data in national inventories.40 This planned improvement is ongoing and has not been
40 See http://www.ipcc-nggip.iges.or.jp/public/tb/TFI Technical Bulletin l.pdf.
4-60 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
incorporated into this inventory report. This is a long-term planned improvement given the significance of these
emissions.
4.12 Soda Ash Production (CRT Source
Category 2B7)
Carbon dioxide (CO2) is generated as a byproduct of calcining trona ore to produce soda ash (sodium carbonate,
Na2CC>3) and is eventually emitted into the atmosphere. In addition, CO2 may also be released when soda ash is
consumed. This reporting category (2B7) includes emissions from the production of soda ash by any of four
processes, of which calcining trona ore is the only emissive process used in the United States. Emissions from soda
ash consumption associated with glass production are reported under Section 4.3, glass production. Emissions
from soda ash consumption not associated with glass production are reported under Section 4.4, other process
uses of carbonates. Emissions from fuels consumed for energy purposes during the production and consumption
of soda ash are accounted for as part of fossil fuel combustion in the industrial end-use sector reported under the
Energy chapter.
Calcining involves placing crushed trona ore into a kiln to convert sodium bicarbonate into crude sodium carbonate
that will later be filtered into pure soda ash. The emission of CO2 during trona-based production is based on the
following reaction:
2Na2C03 ¦ NaHC03 ¦ 2H20(Trona) -» 3Na2C03(Soda Ash) + 5H20 +C02
Soda ash is a white crystalline solid that is readily soluble in water and strongly alkaline. Commercial soda ash is
used as a raw material in a variety of industrial processes and in many familiar consumer products such as glass,
soap and detergents, paper, textiles, and food. The largest use of soda ash is for glass manufacturing. Emissions
from soda ash used in glass production are reported under Section 4.3. In addition, soda ash is used primarily to
manufacture many sodium-based inorganic chemicals, including sodium bicarbonate, sodium chromates, sodium
phosphates, and sodium silicates (USGS 2018b). Internationally, two types of soda ash are produced: natural and
synthetic. The United States produces only natural soda ash and is second only to China in total soda ash
production. Trona is the principal ore from which natural soda ash is made.
The United States represents about one-fifth of total world soda ash output (USGS 2023a). Only two states
produce natural soda ash: Wyoming and California. Of these two states, net emissions of CO2 from soda ash
production were only calculated for Wyoming where trona ore is used.41 Soda ash end uses in 2022 (excluding
glass production) consisted of chemical production, 54 percent; other uses, 17 percent; wholesale distributors
(e.g., for use in agriculture, water treatment, and grocery wholesale), 10 percent; soap and detergent
manufacturing, 9 percent; flue gas desulfurization, 7 percent; water treatment, 2 percent; and pulp and paper
production, 1 percent (USGS 2023b).42
41 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 C02 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 C02 is generated as a byproduct, the C02 is recovered and recycled for use in the carbonation stage
and is not emitted. A facility in a third state, Colorado, produced soda ash until the plant was idled in 2004. The lone producer
of sodium bicarbonate no longer mines trona ore 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 ore 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.
42 Percentages may not add up to 100 percent due to independent rounding.
Industrial Processes and Product Use 4-61
-------
U.S. natural soda ash is competitive in world markets because it is generally considered a better-quality raw
material than synthetically produced soda ash, and most of the world's soda ash is synthetic. Although the United
States continues to be a major supplier of soda ash, China surpassed the United States in soda ash production in
2003, becoming the world's leading producer.
In 2022, CO2 emissions from the production of soda ash from trona ore were 1.7 MMT CO2 Eq. (1,704 kt CO2) (see
Table 4-51 and Table 4-52). Total emissions from soda ash production in 2022 decreased by approximately 1
percent compared to emissions in 2021, as soda ash production returned to 2018 levels observed before the
COVID-19 pandemic. Emissions have increased by approximately 19 percent from 1990 levels.
Trends in emissions have remained relatively constant over the time series with some fluctuations since 1990. In
general, these fluctuations were related to the behavior of the export market and the U.S. economy. The U.S. soda
ash industry saw a decline in domestic and export sales caused by adverse global economic conditions in 2009,
followed by a steady increase in production through 2019 before a significant decrease in 2020 due to the COVID-
19 pandemic.
Table 4-51: CO2 Emissions from Soda Ash Production (MMT CO2 Eq.)
Year 1990 2005 2018 2019 2020 2021
2022
Soda Ash Production 1.4 1.7 | 1.7 1.8 1.5 1.7
1.7
Table 4-52: CO2 Emissions from Soda Ash Production (kt CO2)
Year 1990 2005 2018 2019 2020 2021
2022
Soda Ash Production 1,431 | 1,655 | 1,714 1,792 1,461 1,714
1,704
Methodology and Time-Series Consistency
Carbon dioxide emissions from soda ash production are calculated using a Tier 1 method from the 2006IPCC
Guidelines, in accordance with the IPCC methodological decision tree and available data. During the soda ash
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 ore can be estimated based on the chemical reaction
shown above. Based on this formula and the IPCC default emission factor of 0.0974 metric tons CO2 per metric ton
of trona ore, both of which are consistent with an IPCC Tier 1 approach, one metric ton of CO2 is emitted when
approximately 10.27 metric tons of trona ore are processed (IPCC 2006).
Data is not currently available for the quantity of trona used in soda ash production. Because trona ore is used
primarily for soda ash production, EPA assumes that all trona ore production was used in soda ash production. The
activity data for trona ore production (see Table 4-53) for 1990 through 2022 were obtained from the U.S.
Geological Survey (USGS) Minerals Yearbook for Soda Ash (1994 through 2015b) and USGS Mineral Industry
Surveys for Soda Ash (USGS 2016 through 2017, 2018a, 2019, 2020, 2021, 2022b, 2023b). Soda ash production43
data were collected by the USGS from voluntary surveys of the U.S. soda ash industry. EPA will continue to analyze
and assess opportunities to use facility-level data from EPA's GHGRP to improve the emission estimates for the
soda ash production source category consistent with IPCC44 and UNFCCC guidelines.
43 EPA has assessed the feasibility of using emissions information (including activity data) from EPA's GHGRP program. At this
time, the aggregated information associated with production of soda ash did not meet criteria to shield underlying confidential
business information (CBI) from public disclosure.
44 See http://www.ipcc-nggip.iges.or.ip/public/tb/TFI Technical Bulletin l.pdf.
4-62 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
Table 4-53: Trona Ore Used in Soda Ash Production (kt)
Year
1990
2005
2018
2019
2020
2021
2022
Trona Ore Usea
14,700 |
17,000
17,600
18,400
15,000
17,600
17,500
a Trona ore use is assumed to be equal to trona ore production.
Methodological approaches were applied to the entire time series to ensure consistency in emissions estimates
from 1990 through 2022.
Uncertainty
Emission estimates from soda ash production have relatively low associated uncertainty levels because reliable
and accurate data sources are available for the emission factor and activity data for trona-based soda ash
production. One source of uncertainty is the purity of the trona ore used for manufacturing soda ash. The emission
factor used for this estimate assumes the ore is 100 percent pure and likely overestimates the emissions from soda
ash manufacture. The average water-soluble sodium carbonate-bicarbonate content for ore mined in Wyoming
ranges from 85.5 to 93.8 percent (USGS 1995c).
EPA is aware of one facility producing soda ash from a liquid alkaline feedstock process, based on EPA's GHGRP.
Soda ash production data was collected by the USGS from voluntary surveys. A survey request was sent to each of
the five soda ash producers, all of which responded, representing 100 percent of the total production data (USGS
2023b). EPA assigned an uncertainty range of ±5 percent for trona production, and using the suggested uncertainty
provided in Section 3.8.2.2 of the 2006IPCC Guidelines is appropriate based on expert judgment (RTI 2023). EPA
assigned an uncertainty range of -15 percent to 0 percent range for the trona emission factor, based on expert
judgment on the purity of mined trona (USGS 1995c). Per this expert judgment, a normal probability density
function was assigned for activity data, and a triangular probability density function was assigned for the emission
factor.
The results of the Approach 2 quantitative uncertainty analysis are summarized in Table 4-54. Soda ash production
CO2 emissions for 2022 were estimated to be between 1.5 and 1.7 MMT CO2 Eq. at the 95 percent confidence
level. This indicates a range of approximately 9 percent below and 8 percent above the emission estimate of 1.7
MMTCCh Eq.
Table 4-54: Approach 2 Quantitative Uncertainty Estimates for CO2 Emissions from Soda Ash
Production (MMT CO2 Eq. and Percent)
Source
Gas
2022 Emission Estimate
(MMT COEq.)
Uncertainty Range Relative to Emission Estimate-'
(MMTCO' Eq.) (%)
Lower Upper
Lower Upper
Bound Bound
Bound Bound
Soda Ash Production
C02
1.7
1.5 1.7
-9% +8%
a Range of emission estimates predicted by Monte Carlo stochastic simulation for a 95 percent confidence interval.
QA/QC and Verification
General quality assurance/quality control (QA/QC) procedures were applied consistent with the U.S. Inventory
QA/QC plan, which is in accordance with Volume 1, Chapter 6 of 2006 IPCC Guidelines as described in the
introduction of the IPPU chapter (see Annex 8 for more details).
Recalculations Discussion
No recalculations were performed for the 1990 through 2021 portion of the time series.
Industrial Processes and Product Use 4-63
-------
Planned Improvements
EPA is assessing planned improvements for future reports, but at this time has no specific planned improvements
for estimating CO2 emissions from soda ash production.
4.13 Petrochemical Production (CRT Source
Category 2B8)
The production of some petrochemicals results in carbon dioxide (CO2) and methane (CH4) emissions.
Petrochemicals are chemicals isolated or derived from petroleum or natural gas. This reporting category (2B8)
includes CO2 emissions from the production of acrylonitrile, carbon black, ethylene, ethylene dichloride, ethylene
oxide, and methanol, and Cm emissions from the production of acrylonitrile. The petrochemical industry uses
primary fossil fuels (i.e., natural gas, coal, petroleum, etc.) for non-fuel purposes in the production of carbon black
and other petrochemicals. Per the IPCC methodological guidance, emissions from fuels and feedstocks transferred
out of the system for use in energy purposes (e.g., indirect or direct process heat or steam production) are
currently accounted for as part of fossil fuel combustion in the industrial end-use sector reported under the Energy
chapter.
Worldwide, more than 90 percent of acrylonitrile (vinyl cyanide, C3H3N) is made by way of direct ammoxidation of
propylene with ammonia (NH3) and oxygen over a catalyst. This process is referred to as the SOHIO process,
named after the Standard Oil Company of Ohio (SOHIO) (IPCC 2006). The primary use of acrylonitrile is as the raw
material for the manufacture of acrylic and modacrylic fibers. Other major uses include the production of plastics
(acrylonitrile-butadiene-styrene [ABS] and styrene-acrylonitrile [SAN]), nitrile rubbers, nitrile barrier resins,
adiponitrile, and acrylamide. All U.S. acrylonitrile facilities use the SOHIO process (AN 2014). The SOHIO process
involves a fluidized bed reaction of chemical-grade propylene, ammonia, and oxygen over a catalyst. The process
produces acrylonitrile as its primary product, and the process yield depends on the type of catalyst used and the
process configuration. The ammoxidation process produces byproduct CO2, carbon monoxide (CO), and water
from the direct oxidation of the propylene feedstock and produces other hydrocarbons from side reactions.
Carbon black is a black powder generated by the incomplete combustion of an aromatic petroleum- or coal-based
feedstock at a high temperature. Most carbon black produced in the United States is added to rubber to impart
strength and abrasion resistance, and the tire industry is by far the largest consumer. The other major use of
carbon black is as a pigment. The predominant process used in the United States to produce carbon black is the
furnace black (or oil furnace) process. In the furnace black process, carbon black oil (a heavy aromatic liquid) is
continuously injected into the combustion zone of a natural gas-fired furnace. Furnace heat is provided by the
natural gas and a portion of the carbon black feedstock; the remaining portion of the carbon black feedstock is
pyrolyzed to carbon black. The resultant CO2 and uncombusted CH4 are released from thermal incinerators used as
control devices, process dryers, and equipment leaks. Three facilities in the United States use other types of
carbon black processes. Specifically, one facility produces carbon black by the thermal cracking of acetylene-
containing feedstocks (i.e., acetylene black process), a second facility produces carbon black by the thermal
cracking of other hydrocarbons (i.e., thermal black process), and a third facility produces carbon black by the open
burning of carbon black feedstock (i.e., lamp black process) (EPA 2000).
Ethylene (C2H4) is consumed in the production processes of the plastics industry including polymers such as high,
low, and linear low density polyethylene (HDPE, LDPE, LLDPE); polyvinyl chloride (PVC); ethylene dichloride;
ethylene oxide; and ethylbenzene. Virtually all ethylene is produced from steam cracking of ethane, propane,
butane, naphtha, gas oil, and other feedstocks. The representative chemical equation for steam cracking of ethane
to ethylene is shown below:
Hf, C2//4 + H2
4-64 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
Small amounts of Cm are also generated from the steam cracking process. In addition, CO2 and Cm emissions
result from combustion units.
Ethylene dichloride (C2H4CI2) is used to produce vinyl chloride monomer, which is the precursor to polyvinyl
chloride (PVC). Ethylene dichloride was also used as a fuel additive until 1996 when leaded gasoline was phased
out. Ethylene dichloride is produced from ethylene by either direct chlorination, oxychlorination, or a combination
of the two processes (i.e., the "balanced process"); most U.S. facilities use the balanced process. The direct
chlorination and oxychlorination reactions are shown below:
C2H4 + Cl2 -» C2H4Cl2 (direct chlorination)
C2H4 + |02 + 2HCI -» C2H4Cl2 + 2H20 (oxychlorination)
C2H4 + 302 -» 2C02 + 2H20 (direct oxidation of ethylene during oxychlorination)
In addition to the byproduct CO2 produced from the direct oxidation of the ethylene feedstock, CO2 and Cm
emissions are also generated from combustion units.
Ethylene oxide (C2H4O) is used in the manufacture of glycols, glycol ethers, alcohols, and amines. Approximately 70
percent of ethylene oxide produced worldwide is used in the manufacture of glycols, including monoethylene
glycol. Ethylene oxide is produced by reacting ethylene with oxygen over a catalyst. The oxygen may be supplied to
the process through either an air (air process) or a pure oxygen stream (oxygen process). The byproduct CO2 from
the direct oxidation of the ethylene feedstock is removed from the process vent stream using a recycled carbonate
solution, and the recovered CO2 may be vented to the atmosphere or recovered for further utilization in other
sectors, such as food production (IPCC 2006). The combined ethylene oxide reaction and byproduct CO2 reaction is
exothermic and generates heat, which is recovered to produce steam for the process. The ethylene oxide process
also produces other liquid and off-gas byproducts (e.g., ethane that may be burned for energy recovery within the
process. Almost all facilities, except one in Texas, use the oxygen process to manufacture ethylene oxide (EPA
2008).
Methanol (CH3OH) is a chemical feedstock most often converted into formaldehyde, acetic acid and olefins. It is
also an alternative transportation fuel, as well as an additive used by municipal wastewater treatment facilities in
the denitrification of wastewater. Methanol is most commonly synthesized from a synthesis gas (i.e., "syngas" - a
mixture containing H2, CO, and CO2) using a heterogeneous catalyst. There are a number of process techniques
that can be used to produce syngas. Worldwide, steam reforming of natural gas is the most common method;
most methanol producers in the United States also use steam reforming of natural gas to produce syngas. Other
syngas production processes in the United States include partial oxidation of natural gas and coal gasification.
Emissions of CO2 and CH4 from petrochemical production in 2022 were 28.8 MMT CO2 Eq. (28,788 kt CO2) and
0.005 MMT CO2 Eq. (0.17 kt CH4), respectively (see Table 4-55 and Table 4-56). Carbon dioxide emissions from
petrochemical production are driven primarily from ethylene production, while CH4 emissions are only from
acrylonitrile production. Since 1990, total CO2emissions from petrochemical production increased by 43 percent,
and Cm emissions declined by 22 percent. Emissions of CO2 were 6 percent lower in 2022 than in 2021, and
emissions of CH4 were 12 percent higher in 2022 than in 2021. The increase in CO2 emissions since 1990 is due
primarily to increased ethylene and methanol production, which have been driven by the increased natural gas
production in the United States. The reduction in CO2 emissions since 2021 is due to a reduction in emissions from
ethylene production, despite an increase in ethylene production. Since CH4 emissions from acrylonitrile are
calculated using a Tier 1 approach based on production as the activity data, the decrease in CH4 emissions since
1990 and the increase since 2021 correspond with changes in the production levels for acrylonitrile.
Table 4-55: CO2 and CH4 Emissions from Petrochemical Production (MMT CO2 Eq.)
Year
1990
2005
2018
2019
2020
2021
2022
C02
20.1
26.9
27.2
28.5
27.9
30.7
28.8
Carbon Black
3.4 1
4.3 ill:
3.4
3.3
2.6
3.0
3.1
Ethylene
13.1
19.0
19.4
20.7
20.7
22.8
20.7
Industrial Processes and Product Use 4-65
-------
Ethylene Dichloride
mm
CO
O
0.5 i
0.4
0.5
0.5
0.4
0.4
Ethylene Oxide
1.1
1.5
1.3
1.4
1.7
1.9
1.7
Methanol
1.0 ¦
0.3
1.4
1.6
1.6
1.7
2.0
Acrylonitrile
1.2
1.3
1.3
1.0
0.9
0.9
1.0
ch4
+
+ s
+
+
+
+
+
Acrylonitrile
+
+
+
+
+
+
+
Total
20.1 ¦
26.9
27.2
28.5
27.9
30.7
28.8
+ Does not exceed 0.05 MMT C02 Eq.
Note: Totals may not sum due to independent rounding.
Table 4-56: CO2 and CH4 Emissions from Petrochemical Production (kt)
Year
1990
2005
2018
2019
2020
2021
2022
C02
20,075
26,882
27,200
28,483
27,926
30,656
28,788
Carbon Black
3/381
4,269 i
3,440
3,300
2,610
3,000
3,060
Ethylene
13,126
19,024
19,400
20,700
20,700
22,800
20,700
Ethylene Dichloride
254 1
455
440
503
456
376
428
Ethylene Oxide
1,123
1,489
1,300
1,370
1,680
1,930
1,650
Methanol
977 1
319 1
1,370
1,620
1,630
1,700
2,000
Acrylonitrile
1,214
1,325
1,250
990
850
850
950
ch4
¦
+ !!!!!!!
+
+
+
+
+
+
Acrylonitrile
+
+
+
+
+
+
+
+ Does not exceed 0.5 kt CH4.
Note: Totals by gas may not sum due to independent rounding.
Methodology and Time-Series Consistency
Emissions of CO2 and CH4 were calculated using the estimation methods provided by the 2006IPCC Guidelines, in
accordance with the IPCC methodological decision tree and available data, and country-specific methods from
EPA's GHGRP. The 2006 IPCC Guidelines Tier 1 method was used to estimate CO2 and Cm emissions from
production of acrylonitrile,45 and a country-specific approach similar to the IPCC Tier 2 method was used to
estimate CO2 emissions from production of carbon black, ethylene oxide, ethylene, ethylene dichloride, and
methanol, as CO2 emissions from petrochemical production is a key category. The Tier 2 method for
petrochemicals is a total feedstock carbon mass balance method used to estimate total CO2 emissions, but it is not
applicable for estimating CH4 emissions.
As noted in the 2006 IPCC Guidelines, the Tier 2 total feedstock carbon mass balance method is based on the
assumption that all of the carbon input to the process is converted either into primary and secondary products or
into CO2. Further, the guideline states that while the total carbon mass balance method estimates total carbon
emissions from the process, it does not directly provide an estimate of the amount of the total carbon emissions
emitted as CO2, Cm, or non-Cm volatile organic compounds (NMVOCs). This method accounts for all the carbon as
CO2, including Cm.
A methodology refinement for emissions from methanol production was implemented in this Inventory to
transition from a Tier 1 method to a country-specific approach similar to a Tier 2 method, using the process C02
emissions reported to Subpart X of the GHGRP. As part of this refinement, CH4 emissions from methanol
production for every year in the time series are now included in the CO2 emissions estimates to avoid double
counting because the GHGRP reporting method is a mass balance method under which all carbon input to the
process is assumed to be converted either into primary and secondary products or into CO2.
45 EPA has not integrated aggregated facility-level GHGRP information for acrylonitrile production. The aggregated information
associated with production of these petrochemicals did not meet criteria to shield underlying CBI from public disclosure.
4-66 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
Note, a subset of facilities reporting under EPA's GHGRP use Continuous Emission Monitoring Systems (CEMS) to
monitor CO2 emissions from process vents and/or stacks from stationary combustion units or use the optional
combustion methodology for ethylene production facilities. These facilities are required to also report CO2, Cm
and N2O emissions from combustion of process off-gas in flares. The CO2 emissions from flares are included in
aggregated CO2 results. Analysis of aggregated annual reports from those facilities shows that flared Cm and N2O
emissions are less than 300 kt CO2 Eq./year. Since data is only available from a subset of facilities and not
consistently reported over time and since CH4 and N2O emissions are shown to be insignificant, they are excluded
from this analysis. See the planned improvements section below and Annex 5.
Carbon Black, Ethylene, Ethylene Dichloride, and Ethylene Oxide
2010 through 2022
Carbon dioxide emissions and national production for carbon black, ethylene, ethylene dichloride, and ethylene
oxide were aggregated directly from EPA's GHGRP dataset for 2010 through 2022 (EPA 2023).
These emissions reflect application of a country-specific approach similar to the IPCC Tier 2 method and were used
to estimate CO2 emissions from the production of carbon black, ethylene, ethylene dichloride, ethylene oxide. In
2022, data reported to the GHGRP included 3,060,000 metric tons of CO2 emissions from carbon black production;
20,700,000 metric tons of CChfrom ethylene production; 428,000 metric tons of CChfrom ethylene dichloride
production; and 1,650,000 metric tons of CO2 from ethylene oxide production.
Since 2010, EPA's GHGRP requires all domestic producers of petrochemicals to report annual emissions and
supplemental emissions information (e.g., production data, etc.) under Subpart X to facilitate verification of
reported emissions. Most petrochemical production facilities are required to use either a mass balance approach
or CEMS to measure and report emissions for each petrochemical process unit to estimate facility-level process
CO2 emissions; ethylene production facilities also have a third option. The mass balance method is used by most
facilities46 and assumes that all the carbon input is converted into primary and secondary products or is emitted to
the atmosphere as CO2. To apply the mass balance, facilities must measure the volume or mass of each gaseous
and liquid feedstock and product, mass rate of each solid feedstock and product, and carbon content of each
feedstock and product for each process unit and sum for their facility. To apply the optional combustion
methodology, ethylene production facilities must measure the quantity, carbon content, and molecular weight of
the fuel to a stationary combustion unit when that fuel includes any ethylene process off-gas. These data are used
to calculate the total CO2 emissions from the combustion unit. The facility must also estimate the fraction of the
emissions that is attributable to burning the ethylene process off-gas portion of the fuel. This fraction is multiplied
by the total emissions to estimate the emissions from ethylene production. The QA/QC and Verification section
below has a discussion of non-CC>2 emissions from ethylene production facilities.
All non-energy uses of residual fuel and some non-energy uses of "other oil" are assumed to be used in the
production of carbon black; therefore, consumption of these fuels is adjusted for within the Energy chapter to
avoid double-counting of emissions from fuel used in the carbon black production presented here within IPPU
sector. Additional information on the adjustments made within the Energy sector for non-energy use of fuels is
described in both the Methodology section of CChfrom Section 3.1 and Annex 2.1.
1990 through 2009
Prior to 2010, for carbon black, ethylene, ethylene dichloride, and ethylene oxide processes, an average national
CO2 emission factor was calculated based on the GHGRP data and applied to production for earlier years in the
time series (i.e., 1990 through 2009) to estimate CO2 emissions. For these 4 types of petrochemical processes, CO2
emission factors were derived from EPA's GHGRP data by dividing annual CO2 emissions for petrochemical type "\"
46 A few facilities producing ethylene dichloride, ethylene, and methanol used C02 CEMS; those C02 emissions have been
included in the aggregated GHGRP emissions presented here.
Industrial Processes and Product Use 4-67
-------
with annual production for petrochemical type "\" and then averaging the derived emission factors obtained for
each calendar year 2010 through 2013 (EPA 2023). The years 2010 through 2013 were used in the development of
carbon dioxide emission factors as these years are more representative of operations in 1990 through 2009 for
these facilities. The average emission factors for each petrochemical type were applied across all prior years
because petrochemical production processes in the United States have not changed significantly since 1990,
though some operational efficiencies have been implemented at facilities over the time series.
The average country-specific CO2 emission factors that were calculated from the GHGRP data are as follows:
• 2.59 metric tons CCh/metric ton carbon black produced
• 0.79 metric tons CCh/metric ton ethylene produced
• 0.040 metric tons CCh/metric ton ethylene dichloride produced
• 0.46 metric tons CCh/metric ton ethylene oxide produced
Annual production data for carbon black for 1990 through 2009 were obtained from the International Carbon
Black Association (Johnson 2003 and 2005 through 2010). Annual production data for ethylene, ethylene
dichloride, and ethylene oxide for 1990 through 2009 were obtained from the American Chemistry Council's (ACC)
Business of Chemistry (ACC 2023).
Methanol
2015 through 2022
Carbon dioxide emissions and national production for methanol were aggregated directly from EPA's GHGRP data
for 2015 through 2022 (EPA 2023). These emissions reflect application of a country-specific approach similar to the
IPCC Tier 2 method and were used to estimate CO2 emissions from the production of methanol. In 2022, data
reported to the GHGRP included 2,000,000 metric tons of CO2 emissions from methanol production.
As noted above, since 2010, EPA's GHGRP requires all domestic producers of petrochemicals to report annual
emissions and supplemental emissions information (e.g., production data, etc.) under Subpart X to facilitate
verification of reported emissions. Methanol production facilities are required to use either a mass balance
approach or CEMS to measure and report emissions for each methanol process unit to estimate facility-level
process CO2 emissions. Most methanol production facilities use the mass balance method. As noted above, when
using the mass balance method, facilities must measure the volume or mass of each gaseous and liquid feedstock
and product, mass rate of each solid feedstock and product, and carbon content of each feedstock and product for
each process unit and sum for their facility. For 2010 to 2014, the methanol data reported to GHGRP is considered
CBI; therefore, the direct use of the GHGRP data starts with the 2015 reported information.
1990 through 2014
In this Inventory, similar to the methodology for other petrochemicals that utilize GHGRP data, an average national
CO2 emission factor for years prior to 2015 was calculated for methanol production based on the GHGRP data and
applied to production for earlier years in the time series (i.e., 1990 through 2014) to estimate CO2 emissions.
Methanol CO2 emission factors were derived from EPA's GHGRP data by dividing annual CO2 emissions for
methanol with annual production for methanol and then averaging the derived emission factors obtained for each
year 2015 through 2022. The average country-specific CO2 emission factor from the GHGRP data for these years
was determined to be 0.26 metric tons CCh/metric ton methanol produced. Annual methanol production data for
1990 through 2014 were obtained from the ACC's Business of Chemistry (ACC 2023). The average country-specific
CO2 emission factor from the GHGRP data is lower than the IPCC Tier 1 emission factor of 0.67 metric tons
CCh/metricton methanol produced value that was used in previous versions of the Inventory. The main difference
between the IPCC Tier 1 emission factor and the GHGRP emission factor is that the IPCC emission factor includes
emissions from combustion of natural gas fuel in the reformer as well as vented CO2 from the process; therefore,
the use of the IPCC Tier 1 emission factor would double count emissions from natural gas combustion in the IPPU
4-68 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
chapter and the Energy chapter. EPA already accounts for emissions from combustion of natural gas fuel in the
reformer as part of fossil fuel combustion in the industrial end-use sector reported under the Energy chapter.
Acrylonitrile
Carbon dioxide and methane emissions from acrylonitrile production were estimated using the Tier 1 method in
the 2006IPCC Guidelines. Acrylonitrile emissions represent about 3 percent of total petrochemical emissions in
2022 so a Tier 1 approach is deemed acceptable, and higher Tier methods could not be used due to data
sensitivities which are described below. Annual acrylonitrile production data were used with IPCC default Tier 1
CO2 and Cm emission factors to estimate emissions for 1990 through 2022. Emission factors used to estimate
acrylonitrile production emissions are as follows:
• 0.18 kg CHVmetric ton acrylonitrile produced
• 1.00 metric tons CCh/metric ton acrylonitrile produced
Annual acrylonitrile production data for 1990 through 2022 were obtained from ACC's Business of Chemistry (ACC
2023). EPA is unable to apply the aggregated facility-level GHGRP information for acrylonitrile production needed
for a Tier 2 approach due to sensitive nature of reported data. The aggregated information associated with
production of these petrochemicals did not meet criteria to shield underlying CBI from public disclosure.
Production of each type of petrochemical are shown in Table 4-57.
Table 4-57: Production of Selected Petrochemicals (kt)
Chemical
1990
2005
2018
2019
2020
2021
2022
Carbon Black
1,307
1,651
1,280
1,210
990
1,140
1,170
Ethylene
16,542 j
23,975
30,500
32,400
33,500
34,700
35,400
Ethylene Dichloride
6,283
11,260
12,500
12,600
11,900
11,500
12,100
Ethylene Oxide
2,429 =
3,220 |
3,310
3,800
4,680
4,860
5,310
Methanol
3,750
1,225
5,830
6,460
6,580
7,110
8,030
Acrylonitrile
1,214 ¦
1,325 "
1,250
990
850
850
950
As noted earlier in the introduction section of the Petrochemical Production section, the allocation and reporting
of emissions from both fuels and feedstocks transferred out of the system for use in energy purposes to the Energy
chapter differs slightly from the 2006 IPCC Guidelines. According to the 2006 IPCC Guidelines, emissions from fuel
combustion from petrochemical production should be allocated to this source category within the IPPU chapter.
Due to national circumstances, EIA data on primary fuel for feedstock use within the energy balance are presented
by commodity only, with no resolution on data by industry sector (i.e., petrochemical production). In addition,
under EPA's GHGRP, reporting facilities began reporting in 2014 on annual feedstock quantities for mass balance
and CEMS methodologies (79 FR 63794), as well as the annual average carbon content of each feedstock (and
molecular weight for gaseous feedstocks) for the mass balance methodology beginning in reporting year 2017 (81
FR 89260).47 The United States is currently unable to report non-energy fuel use from petrochemical production
under the IPPU chapter due to CBI issues. Therefore, consistent with 2006 IPCC Guidelines, fuel consumption data
reported by EIA are adjusted to account for these overlaps to avoid double-counting. More information on the
non-energy use of fossil fuel feedstocks for petrochemical production can be found in Annex 2.3.
Methodological approaches were applied to the entire time series to ensure consistency in emissions from 1990
through 2022. The methodology for ethylene production, ethylene dichloride production, and ethylene oxide
production spliced activity data from two different sources: ACC for 1990 through 2009 and GHGRP for 2010
through 2022. The methodology for methanol production spliced activity data from two different sources: ACC for
1990 through 2014 and GHGRP for 2015 through 2022. Consistent with the 2006 IPCC Guidelines, the overlap
47 See https://www.epa.gov/ehereportine/historical-rulemakines.
Industrial Processes and Product Use 4-69
-------
technique was applied to compare the two data sets for years where there was overlap. For ethylene production,
the data sets were determined to be consistent, and adjustments were not needed. For ethylene dichloride
production, ethylene oxide production, and methanol production, the data sets were determined to be
inconsistent. The GHGRP data includes production of ethylene dichloride and ethylene oxide as intermediates,
while it is unclear if the ACC data does. Methanol production data from GHGRP are significantly higher than the
ACC data for every year since 2015; the reason for the difference is not clear. Therefore, no adjustments were
made to the ethylene dichloride, ethylene oxide, and methanol activity data for 1990 through 2009 because the
2006IPCC Guidelines indicate that it is not good practice to use the overlap technique when the data sets are
inconsistent. The methodology for carbon black production also spliced activity data from two different sources:
ICBA for 1990 through 2009 and GHGRP for 2010 through 2022. The overlap technique was applied to these data
for 2010 and 2011. The data sets were determined to be consistent, and adjustments were not needed.
Uncertainty
The CO2 and CH4 emission factors used for acrylonitrile production are based on a limited number of studies. Using
plant-specific factors instead of default or average factors could increase the accuracy of the emission estimates;
however, such data were not available for the current Inventory report. For acrylonitrile, EPA assigned an
uncertainty range of ±60 percent for the CO2 emission factor, ±10 percent for the CH4 emission factor, and a
normal probability density function for both, and using the suggested uncertainty provided in Table 3.27 of the
2006 IPCC Guidelines is appropriate based on expert judgment, (RTI 2023). The results of the quantitative
uncertainty analysis for the CO2 emissions from carbon black production, ethylene, ethylene dichloride, ethylene
oxide, and methanol are based on reported GHGRP data. Refer to the Methodology section for more details on
how these emissions were calculated and reported to EPA's GHGRP. EPA assigned an uncertainty range of ±5
percent and a normal probability density function for CO2 emissions from carbon black, ethylene, ethylene
dichloride, and ethylene oxide production, and using the suggested uncertainty provided in Table 3.27 of the 2006
IPCC Guidelines is appropriate based on expert judgment (RTI 2023). There is some uncertainty in the applicability
of the average emission factors for each petrochemical type across all prior years. While petrochemical production
processes in the United States have not changed significantly since 1990, some operational efficiencies have been
implemented at facilities over the time series.
The results of the Approach 2 quantitative uncertainty analysis are summarized in Table 4-58. Petrochemical
production CO2 emissions from 2022 were estimated to be between 27.6 and 30.0 MMT CO2 Eq. at the 95 percent
confidence level. This indicates a range of approximately 4 percent below to 4 percent above the emission
estimate of 28.8 MMT CO2 Eq. Petrochemical production CH4 emissions from 2022 were estimated to be between
0.0 and 0.01 MMT CO2 Eq. at the 95 percent confidence level. This indicates a range of approximately 14 percent
below to 14 percent above the emission estimate of 0.005 MMT CO2 Eq.
Table 4-58: Approach 2 Quantitative Uncertainty Estimates for CH4 Emissions from
Petrochemical Production and CO2 Emissions from Petrochemical Production (MMT CO2 Eq.
and Percent)
Source
Gas
2022 Emission Estimate
Uncertainty Range Relative to Emission Estimate-1
(MMTCO. Eq.)
(MMTCO. Eq.)
(%)
Lower Upper
Lower
Upper
Bound Bound
Bound
Bound
Petrochemical Production
C02
28.8
27.6 30.0
-4%
+4%
Petrochemical Production
ch4
+
0.0 0.01
-14%
+14%
+ Does not exceed 0.05 MMT C02 Eq.
a Range of emission estimates predicted by Monte Carlo stochastic simulation for a 95 percent confidence interval.
4-70 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
QA/QC and Verification
For petrochemical production, QA/QC activities were conducted consistent with the U.S. Inventory QA/QC plan, as
described in the QA/QC and Verification Procedures section of the IPPU chapter and Annex 8. Source-specific
quality control measures for this category included the QA/QC requirements and verification procedures of EPA's
GHGRP. More details on the greenhouse gas calculation, monitoring and QA/QC methods applicable to
petrochemical facilities can be found under Subpart X (Petrochemical Production) of the regulation (40 CFR Part
98).48 EPA verifies annual facility-level GHGRP reports through a multi-step process (e.g., combination of electronic
checks and manual reviews) to identify potential errors and ensure that data submitted to EPA are accurate,
complete, and consistent (EPA 20 15).49 Based on the results of the verification process, EPA follows up with
facilities to resolve mistakes that may have occurred. The post-submittals checks are consistent with a number of
general and category-specific QC procedures, including range checks, statistical checks, algorithm checks, and year-
to-year checks of reported data and emissions. EPA also conducts QA checks of GHGRP reported production data
by petrochemical type against external datasets.
For ethylene, ethylene dichloride and ethylene oxide, it is possible to compare CO2 emissions calculated using the
GHGRP data to the CO2 emissions that would have been calculated using the Tier 1 approach if GHGRP data were
not available. For ethylene, the GHGRP emissions were within ±8 percent of the emissions calculated using the Tier
1 approach prior to 2018; for 2018 through 2022, the GHGRP emissions were between 76 percent and 87 percent
of what would be calculated using the Tier 1 approach. For ethylene dichloride, the GHGRP emissions are typically
higher than the Tier 1 emissions by up to 25 percent, but in 2010 and 2021, GHGRP emissions were slightly lower
than the Tier 1 emissions. For ethylene oxide, GHGRP emissions typically vary from the Tier 1 emissions by up to
±20 percent, but in 2021 and 2022, the GHGRP emissions were significantly higher than the Tier 1 emissions. This
was likely due to GHGRP data capturing the production of ethylene oxide as an intermediate in the onsite
production of ethylene glycol.
For methanol, GHGRP production data was consistently higher than ACC production data in all years between 2015
and 2022. Even though the GHGRP production was higher than the ACC production, the GHGRP CO2 emissions
estimated using the methodology refinement in this Inventory are significantly lower than the emissions calculated
using the Tier 1 approach in all years between 2015 and 2022. Additionally, there is a trend towards increasing
differences over these years starting with an 873 kt CO2 difference in 2015 and increasing to a 3,000 kt CO2
difference in 2022. GHGRP emissions were between 43 percent and 61 percent of the Tier 1 emissions in 2015 and
2018, respectively. As discussed in the Methodology and Time-Series Consistency section above, EPA has
determined that using the IPCC Tier 1 emissions factor to calculate methanol emissions results in double counting
of natural gas combustion emissions in both this chapter and in the Energy chapter; therefore, use of the GHGRP
derived emissions is deemed appropriate. For the years 1990 through 2014, the use of the GHGRP derived
emission factor also results in lower emissions than those calculated using the IPCC Tier 1 emission factor. While
this avoids the double counting of emissions with the Energy chapter, as described below in the Planned
Improvements section, EPA intends to examine the emissions from methanol facilities that report to the GHGRP
and may have been operating prior to 2010 to assess whether a more specific process-only emission factor can be
developed from the GHGRP data for use in estimating CO2 emissions from methanol production in 1990 through
2014.
EPA's GHGRP mandates that all petrochemical production facilities report their annual emissions of CO2, CH4, and
N2O from each of their petrochemical production processes. Source-specific quality control measures for the
Petrochemical Production category included the QA/QC requirements and verification procedures of EPA's GHGRP.
The QA/QC requirements differ depending on the calculation methodology used.
48 See http://www.ecfr.gov/cgi-bin/text-idx7tpk/ecfrbrowse/Title40/40cfr98 main Q2.tpl.
49 See https://www.epa.gov/sites/production/files/2015-07/documents/ghgrp verification factsheet.pdf.
Industrial Processes and Product Use 4-71
-------
As part of a planned improvement effort, EPA has assessed the potential of using GHGRP data to estimate Cm
emissions from ethylene production. As discussed in the Methodology section above, CO2 emissions from ethylene
production in this chapter are based on data reported under the GHGRP, and these emissions are calculated using
a Tier 2 approach that assumes all of the carbon in the fuel (i.e., ethylene process off-gas) is converted to CO2.
Ethylene production facilities also calculate and report CH4 emissions under the GHGRP when they use the optional
combustion methodology. The facilities calculate CH4 emissions from each combustion unit that burns off-gas from
an ethylene production process unit using a Tier 1 approach based on the total quantity of fuel burned, a default
or measured higher heating value, and a default emission factor. Because multiple other types of fuel in addition
to the ethylene process unit off-gas may be burned in these combustion units, the facilities also report an estimate
of the fraction of emissions that is due to burning the ethylene process off-gas component of the total fuel.
Multiplying the total emissions by the estimated fraction provides an estimate of the CH4 emissions from the
ethylene production process unit. These ethylene production facilities also calculate CH4 emissions from flares that
burn process vent emissions from ethylene processes. The CO2 emissions are calculated using either a Tier 2
approach based on measured gas volumes and measured carbon content or higher heating value, or a Tier 1
approach based on the measured gas flow and a default emission factor; the CH4 emissions are calculated based
on a Tier 1 approach using the CO2 emissions and default emission factors. Nearly all ethylene production facilities
use the optional combustion methodology under the GHGRP. The CH4 emissions from ethylene production under
the GHGRP have not been included in this chapter because this approach double counts carbon (i.e., all of the
carbon in the CH4 emissions is also included in the CO2 emissions from the ethylene process units). EPA continues
to assess the GHGRP data for ways to better disaggregate the data and incorporate it into the Inventory.
These facilities are also required to report emissions of N2O from combustion of ethylene process off-gas in both
stationary combustion units and flares. Facilities using CEMS (consistent with a Tier 3 approach) are also required
to report emissions of CH4 and N2O from combustion of petrochemical process-off gases in flares. Preliminary
analysis of the aggregated reported CH4 and N2O emissions from facilities using CEMS and N2O emissions from
facilities using the optional combustion methodology suggests that these annual emissions are less than 0.4
percent of total petrochemical emissions, which is not significant enough to prioritize for inclusion in the report at
this time. Pending resources and significance, EPA may include these N2O emissions in future reports to enhance
completeness. Future QC efforts to validate the use of Tier 1 default emission factors and report on the
comparison of Tier 1 emission estimates and GHGRP data are described below in the Planned Improvements
section.
Recalculations Discussion
A methodology refinement for calculating emissions from methanol production was implemented in this Inventory.
As discussed in the Methodology and Time-Series Consistency section, the previously used Tier 1 approach was
replaced with a country-specific approach similar to a Tier 2 method based on emissions aggregated directly from
EPA's GHGRP data for 2015 through 2022 (EPA 2023) and an average country-specific CO2 emission factor from the
GHGRP data applied to production data from ACC's Business of Chemistry for 1990 through 2014 (ACC 2023). For
2015 through 2021, these changes resulted in a reduction in the reported CO2 emissions between 43 percent (873
kt) in 2015 to 61 percent (2,110 kt) in 2018. For 1990 through 2014, the refinement resulted in a reduction of 61
percent each year (287 kt in 2011 to 2,449 kt in 1997).
Additionally, CH4 emissions previously reported from methanol production were reduced to zero for all years of
the time series because, as noted above in the Methodology and Time Series Consistency section, the
methodology refinement is based on the assumption that all carbon input to the process is converted either to
primary and secondary products or to CO2.
Planned Improvements
Improvements include completing category-specific QC of activity data and emission factors, along with further
assessment of CH4 and N2O emissions to enhance completeness in reporting of emissions from U.S. petrochemical
4-72 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
production, pending resources, significance and time-series consistency considerations. For example, EPA is
planning additional assessment of fuel combustion emissions data reported by methanol production facilities for
ways to estimate process-based emissions in the Inventory separately from combustion emissions for 1990
through 2014. If the GHGRP data can be categorized by type of methanol process design, it may be possible to use
GHGRP data for single reformer process units to develop a ratio of process-to-total emissions to adjust the IPCC
emission factor. Potential difficulties with this analysis are that some of the methanol producing facilities also
produce other chemicals and the combustion unit names may not clearly identify the process unit to which they
apply, and some combustion unit data may be aggregated for multiple combustion units. The EPA is also planning
additional assessment of ways to use Cm data from the GHGRP in the Inventory. One possible approach EPA is
assessing would be to adjust the CO2 emissions from the GHGRP downward by subtracting the carbon that is also
included in the reported CH4 emissions, per the discussion in the Petrochemical Production QA/QC and Verification
section, above. As of this current report, timing and resources have not allowed EPA to complete these analyses of
activity data, emissions, and emission factors but they remain priority improvements within the IPPU chapter.
Pending resources, a secondary potential improvement for this source category would focus on continuing to
analyze the fuel and feedstock data from EPA's GHGRP to better disaggregate energy-related emissions and
allocate them more accurately between the Energy and IPPU sectors of the Inventory. EPA will continue to look for
ways to incorporate this data into future Inventories that will allow for easier data integration between the non-
energy uses of fuels category and the petrochemicals category presented in this chapter. This planned
improvement is still under development and has not been completed to report on progress in this current
Inventory.
4.14 HCFC-22 Production (CRT Source
Category 2B9a)
This reporting category (2B9a) includes by-product emissions of HCFC-23 (trifluoromethane or CHF3) from
production of HCFC-22 (chlorodifluoromethane). HFC-23 is generated as a byproduct during the manufacture of
HCFC-22, which is primarily employed in refrigeration and air conditioning systems and as a chemical feedstock for
manufacturing synthetic polymers. Between 1990 and 2000, U.S. production of HCFC-22 increased significantly as
HCFC-22 replaced chlorofluorocarbons (CFCs) in many applications. Between 2000 and 2007, U.S. production
fluctuated but generally remained above 1990 levels. In 2008 and 2009, U.S. production declined markedly and has
remained near 2009 levels since. Because HCFC-22 depletes stratospheric ozone, its production for non-feedstock
uses was phased out in 2020 under the U.S. Clean Air Act.50 Feedstock production, however, is permitted to
continue indefinitely. Per the IPCC methodological guidance, emissions from energy use are currently accounted
for as part of fossil fuel combustion in the industrial end-use sector reported under the Energy chapter.
HCFC-22 is produced by the reaction of chloroform (CHCb) and hydrogen fluoride (HF) in the presence of a catalyst,
SbCls. The reaction of the catalyst and HF produces SbClxFy, (where x + y = 5), which reacts with chlorinated
hydrocarbons to replace chlorine atoms with fluorine. The HF and chloroform are introduced by submerged piping
into a continuous-flow reactor that contains the catalyst in a hydrocarbon mixture of chloroform and partially
fluorinated intermediates. The vapors leaving the reactor contain HCFC-21 (CHCbF), HCFC-22 (CHCIF2), HFC-23
(CHF3), HCI, chloroform, and HF. The under-fluorinated intermediates (HCFC-21) and chloroform are then
condensed and returned to the reactor, along with residual catalyst, to undergo further fluorination. The final
vapors leaving the condenser are primarily HCFC-22, HFC-23, HCI and residual HF. The HCI is recovered as a useful
50 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],
Industrial Processes and Product Use 4-73
-------
byproduct, and the HF is removed. Once separated from HCFC-22, the HFC-23 may be released to the atmosphere,
recaptured for use in a limited number of applications, or destroyed.
Two facilities produced HCFC-22 in the United States in 2022. Emissions of HFC-23 from this activity in 2022 were
estimated to be 1.8 MMT CO2 Eq. (0.1 kt) (see Table 4-59 and Table 4-60). This quantity represents an 18 percent
increase from 2021 emissions and a 95 percent decrease from 1990 emissions. The decrease from 1990 emissions
was caused primarily by changes in the HFC-23 emission rate (kg HFC-23 emitted/kg HCFC-22 produced). The
decrease from 2021 emissions was caused by both a decrease in the HFC-23 emission rate at one plant and a
decrease in the total quantity of HCFC-22 produced. The long-term decrease in the emission rate is primarily
attributable to six factors: (a) five plants that did not capture and destroy the HFC-23 generated have ceased
production of HCFC-22 since 1990; (b) one plant that captures and destroys the HFC-23 generated began to
produce HCFC-22; (c) one plant implemented and documented a process change that reduced the amount of HFC-
23 generated; (d) the same plant began recovering HFC-23, primarily for destruction and secondarily for sale; (e)
another plant began destroying HFC-23; and (f) the same plant, whose emission rate was higher than that of the
other two plants, ceased production of HCFC-22 in 2013.
Emissions from HCFC-22 production are reported under fluorochemical production (CRT category 2B9) in this
Inventory, which also includes the production of fluorochemicals other than HCFC-22 described further in section
4.15 of this chapter.
Table 4-59: HFC-23 Emissions from HCFC-22 Production (MMT CO2 Eq.)
Year 1990 2005 2018 2019 2020 2021 2022
HCFC-22 Production 38.6 16.8 2.7 3.1 1.8 2.2 1.8
Table 4-60: HFC-23 Emissions from HCFC-22 Production (kt HFC-23)
Year 1990 2005 2018 2019 2020 2021 2022
HCFC-22 Production 3 1 + + + + j_
+ Does not exceed 0.5 kt.
Methodology and Time-Series Consistency
To estimate HFC-23 emissions for five of the eight HCFC-22 plants that have operated in the United States since
1990, methods comparable to the Tier 3 methods in the 2006IPCC Guidelines (IPCC 2006) were used throughout
the time series. Emissions for 2010 through 2022 were obtained through reports submitted by U.S. HCFC-22
production facilities to EPA's Greenhouse Gas Reporting Program (GHGRP). EPA's GHGRP mandates that all HCFC-
22 production facilities report their annual emissions of HFC-23 from HCFC-22 production processes and HFC-23
destruction processes. Previously, data were obtained by EPA through collaboration with an industry association
that received voluntarily reported HCFC-22 production and HFC-23 emissions annually from all U.S. HCFC-22
producers from 1990 through 2009. These emissions were aggregated and reported to EPA on an annual basis.
For the other three plants, the last of which closed in 1993, methods comparable to the Tier 1 method in the 2006
IPCC Guidelines were used. Emissions from these three plants have been calculated using the recommended
emission factor for unoptimized plants operating before 1995 (0.04 kg HCFC-23/kg HCFC-22 produced).
The five plants that have operated since 1994 measure (or, for the plants that have since closed, measured)
concentrations of HFC-23 as well as mass flow rates of process streams to estimate their generation of HFC-23.
Plants using thermal oxidation to abate their HFC-23 emissions monitor the performance of their oxidizers to verify
that the HFC-23 is almost completely destroyed. One plant that releases a small fraction of its byproduct HFC-23
periodically measures HFC-23 concentrations at process vents using gas chromatography. This information is
combined with information on quantities of products (e.g., HCFC-22) to estimate HFC-23 emissions.
4-74 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
To estimate 1990 through 2009 emissions, reports from an industry association were used that aggregated HCFC-
22 production and HFC-23 emissions from all U.S. HCFC-22 producers and reported them to EPA (ARAP 1997,1999,
2000, 2001, 2002, 2003, 2004, 2005, 2006, 2007, 2008, 2009, and 2010). To estimate 2010 through 2022
emissions, facility-level data (including both HCFC-22 production and HFC-23 emissions) reported through EPA's
GHGRP were analyzed. In 1997 and 2008, comprehensive reviews of plant-level estimates of HFC-23 emissions and
HCFC-22 production were performed (RTI1997; RTI 2008). The 1997 and 2008 reviews enabled U.S. totals to be
reviewed, updated, and where necessary, corrected. The reviews also allowed plant-level uncertainty analyses
(Monte-Carlo simulations) to be performed for 1990,1995, 2000, 2005, and 2006. Estimates of annual U.S. HCFC-
22 production are presented in Table 4-61.
Table 4-61: HCFC-22 Production (kt)
Year
1990
2005
2012
2018
2019
2020
2021
2022
Production
139
156 I
96 i
C
C
C
C
C
C(CBI)
Note: HCFC-22 production in 2013 through 2022 is considered confidential business information
(CBI) as there were only two producers of HCFC-22 in those years.
Uncertainty
The uncertainty analysis presented in this section was based on a plant-level Monte Carlo stochastic simulation for
2006. The Monte Carlo analysis used estimates of the uncertainties in the individual variables in each plant's
estimating procedure. This analysis was based on the generation of 10,000 random samples of model inputs from
the probability density functions for each input. A normal probability density function was assumed for all
measurements and biases except the equipment leak estimates for one plant; a log-normal probability density
function was used for this plant's equipment leak estimates. The simulation for 2006 yielded a 95-percent
confidence interval for U.S. emissions of 6.8 percent below to 9.6 percent above the reported total.
The relative errors yielded by the Monte Carlo stochastic simulation for 2006 were applied to the U.S. emission
estimate for 2022. The resulting estimates of absolute uncertainty are likely to be reasonably accurate because (1)
the methods used by the two remaining plants to estimate their emissions are not believed to have changed
significantly since 2006, and (2) although the distribution of emissions among the plants has changed between
2006 and 2022 (because one plant has closed), the plant that currently accounts for most emissions had a relative
uncertainty in its 2006 (as well as 2005) emissions estimate that was similar to the relative uncertainty for total
U.S. emissions. Thus, the closure of one plant is not likely to have a large impact on the uncertainty of the national
emission estimate.
The results of the Approach 2 quantitative uncertainty analysis are summarized in Table 4-62. HFC-23 emissions
from HCFC-22 production were estimated to be between 1.7 and 2.0 MMT CO2 Eq. at the 95 percent confidence
level. This indicates a range of approximately 7 percent below and 10 percent above the emission estimate of 1.8
MMTCCh Eq.
Table 4-62: Approach 2 Quantitative Uncertainty Estimates for HFC-23 Emissions from HCFC-
22 Production (MMT CO2 Eq. and Percent)
Source
2022 Emission Estimate
aS (MMTCO' Eq.)
Uncertainty Range Relative to Emission Estimate-1
(MMTCO' Eq.) (%)
Lower Upper
Bound Bound
Lower Upper
Bound Bound
HCFC-22 Production
HFC-23 1.8
1.7 2.0
-7% +10%
a Range of emissions reflects a 95 percent confidence interval.
Industrial Processes and Product Use 4-75
-------
QA/QC and Verification
General quality assurance/quality control (QA/QC) procedures were applied consistent with the U.S. Inventory
QA/QC plan, which is in accordance with Volume 1, Chapter 6 of 2006IPCC Guidelines as described in the
introduction of the IPPU chapter (see Annex 8 for more details). Under the GHGRP, EPA verifies annual facility-level
reports through a multi-step process (e.g., including a combination of pre-and post-submittal electronic checks and
manual reviews by staff) to identify potential errors and ensure that data submitted to EPA are accurate,
complete, and consistent (EPA 2015).51 Based on the results of the verification process, EPA follows up with
facilities to resolve mistakes that may have occurred. The post-submittals checks are consistent with a number of
general and category-specific QC procedures, including: range checks, statistical checks, algorithm checks, and
year-to-year checks of reported data and emissions.
The GHGRP also requires source-specific quality control measures for the HCFC-22 Production category. Under
EPA's GHGRP, HCFC-22 producers are required to (1) measure concentrations of HFC-23 and HCFC-22 in the
product stream at least weekly using equipment and methods (e.g., gas chromatography) with an accuracy and
precision of 5 percent or better at the concentrations of the process samples, (2) measure mass flows of HFC-23
and HCFC-22 at least weekly using measurement devices (e.g., flowmeters) with an accuracy and precision of 1
percent of full scale or better, (3) calibrate mass measurement devices at the frequency recommended by the
manufacturer using traceable standards and suitable methods published by a consensus standards organization,
(4) calibrate gas chromatographs at least monthly through analysis of certified standards, and (5) document these
calibrations.
Recalculations Discussion
The 2019 emissions estimate increased by 0.05 kg of HFC-23 to reflect newly reported emissions from a facility
that destroys HFC-23. This increased the 2019 emissions estimate by two ten thousandths of a percent.
Planned Improvements
At this time, there are no specific planned improvements for estimating HFC-23 emissions from HCFC-22
production.
4.15 Production of Fluorochemicals Other
Than HCFC-22 (CRT Source Category 2B9b)
This reporting category, fluorochemical production (2B9b), facilities in the United States produced or transformed
approximately 200 fluorinated gases other than HCFC-22 in 2022, including saturated and unsaturated
hydrofluorocarbons (HFCs), saturated and unsaturated perfluorocarbons (PFCs), sulfur hexafluoride (SFs), nitrogen
trifluoride (NF3), hydrofluoroethers (HFEs), perfluoroalkylamines, and dozens of others. Emissions from
fluorochemical production may include emissions of the intentionally manufactured chemical as well as reactant
and by-product emissions. The compounds emitted depend upon the production or transformation process, but
may include, e.g., HFCs, PFCs, SF6, nitrous oxide (N2O), NF3, and many others. Potential sources of fluorinated GHG
emissions at fluorochemical production facilities include process vents, equipment leaks, and evacuating returned
51 EPA (2015). Greenhouse Gas Reporting Program Report Verification. Available online at:
https://www.epa.Eov/sites/production/files/2015-07/documents/ehgrp verification factsheet.pdf.
4-76 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
containers52 Production-related emissions of fluorinated GHGs occur from both process vents and equipment
leaks. Process vent emissions occur from manufacturing equipment such as reactors, distillation columns, and
packaging equipment. Equipment leak emissions, or fugitive emissions, occur from valves, flanges, pump seals,
compressor seals, pressure relief valves, connectors, open-ended lines, and sampling connections. In addition,
users of fluorinated GHGs may return empty containers (e.g., cylinders) to the production facility for reuse; prior to
reuse, the residual fluorinated GHGs (often termed "heels") may be evacuated from the container and are a
potential emission source. In many cases, these "heels" are recovered or exhausted to a treatment device for
destruction. In other cases, however, they are released into the atmosphere.53
Emissions of all HFCs, PFCs, NF3, and SF6 from production of fluorochemicals other than hydrochlorofluorocarbon
(HCFC)-22 are presented in Table 4-63 below for the years 1990, 2005, and the period 2018 to 2022. Per the IPCC
methodological guidance, emissions from energy use are currently accounted for as part of fossil fuel combustion
in the industrial end-use sector reported under the Energy chapter.
The fluorinated GHG emissions reported under the Greenhouse Gas Reporting Program (GHGRP) include emissions
of HFCs, PFCs, SFs, NF3, and numerous "other" fluorinated GHGs, such as octafluorotetrahydrofuran (C4F8O),
trifluoromethyl sulfur pentafluoride (SF5CF3), and hexafluoropropylene oxide. Because they are not included
among the seven UNFCCC-reportable gases or gas groups, the "other" fluorinated GHGs are not included in
Inventory totals. However, their emissions are presented below because they often have high GWPs and large
GWP-weighted emissions.
Total emissions of HFCs, PFCs, SF6, and NF3 from fluorochemical production are estimated to have increased from
32 MMT CO2 Eq. (3,400 MT) in 1990 to a peak of 45 MMT C02 Eq. (5,700 MT) in 1999, declining to 3.9 MMT C02
Eq. (860 MT) in 201654 and rising again to 6.0 MMT CO2 Eq. (1,200 MT) in 2022. These trends reflect estimated
changes in fluorinated gas production and increasing use of control devices. Prior to 2000, only 2 facilities are
known to have operated control devices to destroy fluorinated GHG emissions. After 2000, additional production
facilities began to install and use control devices to destroy fluorinated GHG emissions,55 and fluorinated GHG
emissions declined sharply from 45 MMT CO2 Eq. (5,700 MT) in 1999 to 13 MMT CO2 Eq. (2,300 MT) in 2005.
Emissions continued to fall more slowly through 2016, reflecting the installation of controls at an additional 4
facilities in 2011, 2012, 2015, and 2016. Total fluorinated GHG emissions rose from 2017 to 2022 as production
increased at some facilities.
Emissions from the production of fluorochemicals other than HCFC-22 are reported under fluorochemical
production (CRT category 2B9) in conjunction with emissions from HCFC-22 production described in Section 4.14 of
this chapter.
HFC Emissions
Estimated emissions of HFCs increased from 8.7 MMT CO2 Eq. in 1990 to a peak of 14 MMT CO2 Eq. in 1999 (1,200
to 2,600 MT), declining with some fluctuation to 2.5 MMT CO2 Eq. in 2022. Emissions in 1990 were primarily from
facilities producing compounds other than saturated HFCs. The subsequent trends in emissions were driven by the
growth in production of saturated HFCs and the imposition of controls. Production of saturated HFCs is estimated
to have increased from around 0.3 MMT CO2 Eq. (2,000 MT) in 1990 to over 300 MMT CO2 Eq. (100,000 MT) by
52 The totals presented below also include emissions from destruction of previously produced fluorinated GHGs that are
shipped to production facilities for destruction, e.g., because they are found to be irretrievably contaminated.
53 IPCC (2019) 2019 Refinement to the 2006 IPCC Guidelines for National Greenhouse Gas Inventories, Calvo Buendia, E.,
Tanabe, K., Kranjc, A., Baasansuren, J., Fukuda, M., Ngarize S., Osako, A., Pyrozhenko, Y., Shermanau, P. and Federici, S. (eds).
Published: IPCC, Switzerland.
54 Emissions in MMT C02 Eq. were similar in 2017, but the 2017 emissions in MT were considerably higher (4,500 MT) due to
anomalously high emissions of one low-GWP, unsaturated HFC at one facility.
55 One facility is assumed to have installed controls in 2000, another installed controls in 2003, and three facilities are assumed
to have installed controls in 2005.
Industrial Processes and Product Use 4-77
-------
2010 as HFCs replaced ozone-depleting substances, which were being phased out under the Montreal Protocol and
Clean Air Act (EPA 2023a; EPA 2023b). This increase in HFC production drove HFC emissions to their 1999 peak.
However, estimated emissions declined significantly from 1999 to 2005 due to the assumed addition of controls in
2000 and subsequent years. Estimated emissions of HFCs resumed their increase from 2005 to 2010, reaching 7.2
MMT CO2 Eq. (2,300 MT), but again declined sharply in 2011 to 4.6 MMT CO2 Eq. (1,200 MT) based on addition of
controls. Since 2012, HFC emissions have continued to trend downward with some fluctuations, hitting a minimum
of 1.7 MMT CO2 Eq. in 2021. With the phase-out of production of saturated HFCs (Kigali Amendment, and U.S. AIM
program), the downward trend of HFC emissions is expected to continue, but the share of HFC emissions that are
not associated with saturated HFC production (approximately 0.7 MMT CO2 Eq. in 2022) is likely to persist in the
absence of additional controls.
PFC Emissions
Emissions of PFCs increased gradually from 18 MMT CO2 Eq. (2,000 MT) in 1990 to 24 MMT CO2 Eq. (2,800 MT) in
1999 but dropped to 4.0 MMT CO2 Eq. (490 MT) by 2005, reflecting the addition of controls at high-emitting
facilities and apparent changes to the mix of products produced at another facility.56 Overall PFC emissions from
2005 to 2022 have remained relatively steady, oscillating around 2.5 MMT CO2 Eq. The upward trend between
1990 and 1999 was largely driven by the three facilities that reported their historical emissions to the EPA. In the
absence of historical emissions data for other facilities, the quantities of fluorinated GHGs produced or
transformed at other facilities emitting PFCs are estimated to have remained generally steady between 1990 and
2009 and therefore do not contribute to the emissions trend before 2010. For most of the fluorinated GHGs
produced at these facilities, there was no available industry information to inform activity estimates or trends for
1990 to 2009. Therefore, as discussed in the Methodology section below, 2010 production values from EPA's
GHGRP were assumed to have held constant for these compounds from 1990 to 2010.
SFe Emissions
Emissions of SF6 are estimated to have risen gradually from 5.8 MMT CO2 Eq. (250 MT) in 1990 to a peak of 7.5
MMT CO2 Eq. (320 MT) in 1995, to have declined slowly to 7.0 MMT CO2 Eq. in 2000, and then to have declined
more rapidly to a minimum of 0.0004 MMT CO2 Eq. (0.01 MT) in 2017, after which emissions rose and fluctuated
between 0.056 MMT CO2 Eq. (in 2020) and 0.0024 MMT CO2 Eq. (in 2022). The rapid emissions decline after 2000
was driven first by the imposition of controls at one facility and then by the cessation of production in 2010 at a
major U.S. SF6-producing facility.
NF3 Emissions
Since 1990, estimated emissions of NF3 have fluctuated between 0.11 MMT CO2 Eq. and 0.72 MMT CO2 Eq., with
peaks occurring in 2000 (0.71 MMT C02 Eq.), 2010 (0.70 MMT C02 Eq.), and 2020 (0.72 MMT C02 Eq.), and lows
occurring in 1990 (0.29 MMT C02 Eq.), 2003 (0.33 MMT C02 Eq.), and 2018 (0.11 MMT C02 Eq.). NFs may be
emitted both from the production of NF3 and from the production of other fluorochemicals. The dominant source
since 2010 has been production of NF3. Trends after 2010 were driven by changes both in NF3 production and in
56 In a summary of 1990 through 2010 emissions submitted to EPA (described more below), 3M, which owns several facilities
that historically emitted PFCs, noted that the mix of products produced at its various facilities had changed over time, leading
to changes in the magnitude and contents of emissions. This change in magnitude and contents was particularly pronounced at
3M's Decatur facility (referred to elsewhere in this document as "3M Company"), where emissions declined from 15.8 MMT
C02 Eq. in 2000 to 0.53 MMT C02 Eq. in 2002, and where the contents of emissions changed from HFCs, PFCs, SF6 and other
fluorinated GHGs in 2000 to PFCs and other fluorinated GHGs in 2003. (Emissions in 2002 were not differentiated by group).
Emissions were also reduced after the installation of a control device at the Cordova facility. 3M noted that Initial start-up of
the thermal oxidizer occurred in 2003, but that it took time to optimize the operation of the thermal oxidizer and treatment of
the various gas streams, leading to a decrease in emissions over several years.
4-78 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
the emission rate (kg NF3 emitted/kg NF3 produced) for NF3 production, with both contributing to increased
emissions since 2018. For 1990 through 2009, the NF3 that is emitted from the production of NF3 is assumed to be
influenced by the trajectory of NF3 production, which is generally assumed to follow production trends in the
semiconductor industry except where NF3 facility capacity limits production further. Semiconductor production
increased steadily from 1995 to 2007 but is estimated to have declined from 2007 through 2010. As described in
the Methodology section under "Estimated Emissions for 3M facilities," the NF3 that is emitted from production of
other fluorochemicals is assumed to have been emitted as a constant fraction of the "other" fluorinated GHGs
whose 1990 through 2010 emissions were reported by 3M facilities. This fraction was estimated based on the
fraction of "other" fluorinated GHG emissions accounted for by NF3 between 2011 and 2015 and is highly
uncertain. Nevertheless, because the highest-emitting 3M facilities reported decreasing emissions of all other
fluorinated GHG groups between 2000 and 2005 (due to the installation of a control device at one facility and
apparent production changes at another), NF3 emissions also appear likely to have decreased during this period.
Other Fluorinated GHG Emissions
Other fluorinated GHGs, i.e., those not included in the UNFCCC-reportable gases or gas groups, are also emitted in
significant quantities from fluorinated gas production and transformation processes. Estimated emissions of these
other fluorinated GHGs are provided in Table 4-64 for the years 1990, 2005, and the period 2018 to 2022. The
other fluorinated GHGs with the highest estimated emissions in 2022 are presented separately, and the remaining
other fluorinated GHGs are aggregated.
Total emissions of other fluorinated GHGs increased from 4.7 MMT CO2 Eq. (450 MT) in 1990 to a peak of 10.1
MMT CO2 (870 MT) in 2000, declining rapidly to 0.90 MMT CO2 Eq. in 2009 and then declining more slowly to 0.13
MMT CO2 Eq. (40 MT) in 2021 and 2022. Between 1990 and 2009, estimated emissions of other fluorinated GHGs
were primarily driven by the emissions reported by 3M facilities, which showed significant declines between 2000
and 2005, reflecting apparent production changes at one facility and the installation of a control device at another.
The decline in emissions from 2019 to 2020 was due to a decrease in the emission rate at one facility.
Table 4-63: Emissions of HFCs, PFCs, SF6, and NF3 from Production of Fluorochemicals Other
Than HCFC-22 (MMT C02 Eq.)
Gas
1990
2005
2018
2019
2020
2021
2022
HFC-23
6.7
1.7
1.3
1.1
0.9
0.7
1.0
HFC-125
q ^ llllll!
1.9 I
0.4
0.4
0.4
0.4
0.3
HFC-143a
0.1
0.8
0.7
0.6
0.3
0.2
0.3
HFC-134a
+ 5
~ - ¦
0.4 =
0.3
0.3
0.2
0.2
0.3
lH,4H-Perfluorobutane
0.0
0.0
0.0
+
+
+
0.2
lH,6H-Perfluorohexane
0.0!
0.0
0.0
+
+
+
0.2
Other HFCs
1.6
0.5
0.2
0.2
0.2
0.2
0.3
Perfluorocyclobutane
11-2 =
0.7 1
1.3
1.4
1.1
1.3
1.3
PFC-14 (Perfluoromethane)
2.4
1.4
1.0
0.9
0.9
0.9
1.0
Other PFCs
3.9 ¦;
^ 0 "j
0.7
0.7
0.4
0.4
0.6
Nitrogen trifluoride
0.3
0.6
0.1
0.6
0.7
0.5
0.5
Sulfur hexafluoride
5.8
3.3
+
+
+
+
+
Total
32.3
13.2
5.9
6.2
5.2
4.9
5.9
+ Does not exceed 0.05 MMT C02 Eq.
Note: Table does not sum due to independent rounding.
Industrial Processes and Product Use 4-79
-------
Table 4-64: Emissions of HFCs, PFCs, SF6, and NF3 from Production of Fluorochemicals Other
Than HCFC-22 (Metric Tons)
Gas
1990
2005 mi
2018
2019
2020
2021
2022
HFC-23
540
140
100
89
71
56
77
HFC-125
43 "
600
130
130
120
110
105
HFC-143a
30
160
160
120
63
49
57
HFC-134a
37 i
=!
340
200
220
180
180
190
lH,4H-Perfluorobutane
0
0
0
1.2
0.60
1.2
53
lH,6H-Perfluorohexane
0
o ¦
0
0.92
0.47
0.90
41
Other HFCs
500
400
270
260
230
250
270
Perfluorocyclobutane
1,20°
70 1
130
150
120
130
140
PFC-14 (Perfluoromethane)
360
210
150
130
140
140
160
Other PFCs
420
210;
71
77
41
49
59
Nitrogen trifluoride
18
37
6.7
35
45
31
31
Sulfur hexafluoride
250
140
0.15
0.17
0.24
0.21
0.10
Total HFCs, PFCs, SF,„ and NF,
3,400
2,300
1,200
1,200
1,000
1,000
1,200
+ Does not exceed 0.5 MT.
Note: Table does not sum due to independent rounding.
Table 4-65: Emissions of Other Fluorinated GHGs from Production of Fluorochemicals Other
Than HCFC-22 (MMT C02 Eq.)
Gas
1990
2005
2018
2019
2020
2021
2022
Octaf 1 u o rotet ra hyd rof u ra n
2.4
1.9 1
+
0.1
+
+
+
1,1,1,2,2,3,3-Heptafluoro-3-(l,2,2,2-
tetrafluoroethoxy)-propane
+ l
+
+
+
+
+
+
Trifluoromethyl sulphur pentafluoride
1.2
0-9 I
+
0.1
+
+
+
Hexafluoropropylene oxide
0.3
0.3
0.3
0.3
+
+
+
FC-3283/FC-8270 (Perfluorotripropylamine)
+
+
+
+
+
+
Others
0.8
0.5 I
0.1
0.1
0.1
+
+
Total Other Fluorinated GHGs
4.7
3.7
0.6
0.6
0.1
0.1
0.1
+ Does not exceed 0.5 MT.
Note: Table does not sum due to independent rounding.
Table 4-66: Emissions of Other Fluorinated GHGs from Production of Fluorochemicals Other
Than HCFC-22 (Metric Tons)
Gas
1990
2005
2018
2019
2020
2021
2022
Octaf I u o rotet ra hyd rof u ra n
1,1,1,2,2,3,3-Heptafluoro-3-(l, 2,2,2-
tetrafluoroethoxy)-propane
Trifluoromethyl sulphur pentafluoride
Hexafluoropropylene oxide
170
6
66
34
140
4
1
53
34
4
5
3
32
4
6
4
32
4-80 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
FC-3283/FC-8270 (Perfluorotripropylamine)
Others
+
170
120 |
+
76
+
84
1
35
1
33
1
35
Total Other Fluorinated GHGs
450
1 350|
| 120
130
43
43
45
+ Does not exceed 0.5 MT.
Note: Table does not sum due to independent rounding.
Table 4-67: Production and Transformation of Fluorinated GHGs (kt)a
Set of Facilities
1990
2005
2018
2019
2020
2021
2022
Facilities reporting their F-GHG emissions,
production, and transformation to GHGRP
after 2010b
Facilities reporting only their F-GHG
production and transformation to GHGRP
after 2010
86
!
3.3
271
: :
3.3
376
j 11.2
371
9.7
352
8.2
348
7.5
370
11.2
Total Production and Transformation
89
1 274
387
381
360
356
381
a Totals are presented across species to protect confidential business information.
b Includes 1 facility that reported production, but not emissions, of SF6 through 2010.
Note: Tables may not sum due to independent rounding.
Methodology
The 2006IPCC Guidelines as elaborated by the 2019 Refinement include Tier 1, Tier 2, and Tier 3 methods for
estimating fluorinated GHG emissions from production of fluorinated compounds. The Tier 1 method calculates
emissions by multiplying a default emission factor by total production. Specific default emission factors exist for
production of SF6 and NF3; a more general default emission factor covers production of all other fluorinated GHGs.
(The more general default emission factor was developed based on data from U.S. facilities collected under the
GHGRP between 2011 and 2016.) The Tier 2 method calculates emissions using a mass-balance approach. The Tier
3 method is based on the collection of plant-specific data on the types and quantities of fluorinated GHGs emitted
from process vents, leaks, container venting, and other sources, considering any abatement technology. The Tier 3
method is often implemented by developing and applying facility-specific emission factors indexed to production.
Based on available data on emissions and activity, EPA used a form of the IPCC Tier 3 method to estimate
fluorinated GHG emissions from most U.S. production of fluorinated compounds. Emissions from U.S. production
for which there are fewer data are based on the Tier 1 method.
Overview of GHGRP Data for this Source Category
As discussed further below, much of the data used to develop the estimates presented here come from the
GHGRP. The data were collected under two sections of the GHGRP regulation—Subpart L, Fluorinated Gas
Production; and Subpart OO, Suppliers of Industrial Greenhouse Gases. Under Subpart L, certain fluorinated gas
production facilities must report their emissions from a range of processes and sources, detailed further below.
Data collected under Subpart L include emissions data for calendar years 2011 through 2022. Under Subpart 00,
fluorinated GHG suppliers (including fluorinated GHG producers) must report the quantities of each fluorinated
GHG that they produce, transform, destroy, import, or export. Data collected under Subpart 00 include
production and transformation data for calendar years 2010 through 2022. Facilities' production and
transformation data are not shown here because they are considered confidential business information under the
GHGRP.
Industrial Processes and Product Use 4-81
-------
Emissions Reported Under Subpart L of the GHGRP
Under Subpart L, facilities that produce a fluorinated gas must report their greenhouse gas emissions if the facility
emits 25,000 MT CO2 Eq. or more per year in combined emissions from fluorinated gas production, stationary fuel
combustion units, miscellaneous uses of carbonate, and all other applicable source categories listed in the rule.
(For purposes of calculating emissions from fluorinated gas production for inclusion in the total that is compared
to the threshold, emissions are assumed to be uncontrolled.) Facilities must report their fluorinated GHG
emissions from the production and transformation of fluorinated gases, from venting of residual fluorinated GHGs
from containers, and from destruction of previously produced fluorinated GHGs. The emissions reported from
production and transformation include both emissions from process vents and emissions from equipment leaks.
Under the GHGRP, "fluorinated GHGs," whose emissions must be reported, include SF6, NF3, and any fluorocarbon
except for substances with vapor pressures below 1 Torr at 25 degrees C and substances that are regulated as
"controlled substances" under EPA's ozone-protection regulations at 40 CFR Part 82, Subpart A (e.g.,
chlorofluorocarbons [CFCs], hydrochlorofluorocarbons [HCFCs], and halons). In addition to SF6 and NF3, this
definition includes hydrofluorocarbons (HFCs), perfluorocarbons (PFCs), hydrofluoroethers (HFEs), fully fluorinated
tertiary amines, perfluoropolyethers (including PFPMIE), and hydrofluoropolyethers, and others. "Fluorinated
gases," from whose production or transformation emissions must be reported, include the fluorinated GHGs
detailed above as well as CFCs and HCFCs.57
Facilities calculate emissions from process vents using one of two methods. For vents that emit 10,000 MT CO2 Eq.
or more (considering controls) of fluorinated GHGs from continuous processes, facilities must use emissions testing
to establish an emission factor at least every ten years, or sooner if the process changes in a way that will
significantly affect emissions from the vent. For other process vents, facilities may use measurements, engineering
calculations, or engineering assessments to establish the emission factor. Facilities then calculate their annual
emissions based on the measured or calculated emission factor and related activity data, considering the extent to
which the process is controlled and any destruction device or process malfunctions.
To calculate emissions from equipment leaks, facilities that report under Subpart L are required to collect
information on the number and type of pieces of equipment; service of each piece of equipment; concentration of
each fluorinated GHG in the stream; and the time period each piece of equipment was in service. Facilities use one
or more of the following methods to calculate emissions from equipment leaks (EPA 1995):
• Average Emission Factor Approach in EPA Protocol for Equipment Leak Estimates.
• Other Approaches in EPA Protocol for Equipment Leak Estimates in conjunction with EPA Method 21.
• Other Approaches in EPA Protocol for Equipment Leak Estimates in conjunction with site-specific leak
detection methods.
• Site-specific leak detection methods.
Most emissions are reported by chemical; the exceptions are (1) fluorinated GHGs that are emitted in quantities of
1,000 MT CO2 Eq. or less across all production and transformation processes at a facility and (2) fluorinated GHGs
that are emitted from facilities that produce only one fluorinated GHG, where the emitted fluorinated GHG is not
the fluorinated gas produced. In these cases, the emissions are reported in CO2 Eq. by fluorinated GHG group.
There are 12 fluorinated GHG groups, each of which encompasses a set of GHGs with roughly similar atmospheric
behavior, including similar GWPs and atmospheric lifetimes. These include, e.g., fully fluorinated GHGs such as
PFCs and SF6, saturated HFCs with two or fewer hydrogen-carbon bonds, saturated HFCs with more than two
carbon-hydrogen bonds, unsaturated HFCs and PFCs, and others (see Table 4-68 for a full list).
57 HCFC-22 is considered a fluorinated gas under the GHGRP, but emissions from HCFC-22 production are reported separately
from emissions from production of other fluorinated gases.
4-82 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
Table 4-68: Fluorinated GHG Groups Under Which Certain Emissions Are Reported Under
Subpart L of the GHGRP and Associated GWPs
Fluorinated GHG Group
GHGRP Default Global
Warming Potential (100-yr.)
Fully fluorinated GHGs
10,000
Saturated hydrofluorocarbons (HFCs) with 2 or fewer carbon-
hydrogen bonds
3,700
Saturated HFCs with 3 or more carbon-hydrogen bonds
930
Saturated hydrofluoroethers (HFEs) and hydrochlorofluoroethers
(HCFEs) with 1 carbon-hydrogen bond
5,700
Saturated HFEs and HCFEs with 2 carbon-hydrogen bonds
2,600
Saturated HFEs and HCFEs with 3 or more carbon-hydrogen bonds
270
Fluorinated formates
350
Fluorinated acetates, carbonofluoridates, and fluorinated alcohols
other than fluorotelomer alcohols
30
Unsaturated PFCs, unsaturated HFCs, unsaturated HCFCs,
unsaturated halogenated ethers, unsaturated halogenated esters,
fluorinated aldehydes, and fluorinated ketones
1
Fluorotelomer alcohols
1
Fluorinated GHGs with carbon-iodine bond(s)
1
Other fluorinated GHGs
2,000
Two other datasets reported under Subpart L are relevant to estimating uncontrolled emission factors. (As
discussed further below, such uncontrolled emission factors are applied to years before Subpart L reporting began
(for CY 2011) and before emission controls were put into place.) First, in addition to reporting emissions by
chemical at the facility level, facilities report emissions from each production and transformation process at the
facility in tons of CO2 Eq. by fluorinated GHG group. To calculate CO2 Eq. emissions, facilities use a chemical-specific
100-year GWP where one is available for the compound of interest. If no chemical-specific 100-year GWP is
available for the compound of interest, facilities use the GHGRP default GWP for the fluorinated GHG group of
which the compound is a member. These default GWPs are shown in Table 4-63.
Second, for each process, facilities also report the extent to which emissions are abated (the effective destruction
efficiency or EDE) as a range. The EDE is calculated as follows:
CEPV
EDE = 1
UE P
where:
EDE = Effective destruction efficiency of the process
CEpv = Actual GWP-weighted controlled emissions from all vents for the process, MT CO2 Eq.
UE
PV
Hypothetical GWP-weighted uncontrolled emissions from all vents for the process, MT CO2 Eq.
(CEpv will equal UEpv if the process is not controlled, resulting in a calculated EDE of 0).
Note that the EDE is based on the extent to which emissions from process vents are controlled. Emissions from
equipment leaks are not included in the EDE calculation. Table 4-69 provides the EDE ranges available for facilities
to report and the arithmetic means of each range. The use of these datasets to calculate uncontrolled emission
factors is discussed in more detail in the "1990-2010 Emissions Estimates" section below.
Industrial Processes and Product Use 4-83
-------
Verification of GHGRP Reports
Data reported under the GHGRP, including emissions and production, are electronically verified using range
checks, internal consistency checks, and time-series consistency checks. Where the data fail a preliminary check,
EPA contacts the facility to see whether there is an explanation for the issue or whether the data are indeed
erroneous. In the latter case, facilities are required to correct the data. Where one or more of the anomalous data
elements is not explained or corrected, the report for that facility for that year is considered unverified.
1990-2010 Emissions Estimates
For 14 of the 17 fluorinated gas production facilities that have reported their emissions under the GHGRP, 1990
through 2010 emissions are estimated using (1) facility- and chemical-specific emission factors based on the
emissions data discussed under "2011-2022 Emissions" below, (2) reported or estimated production and
transformation of fluorinated GHGs at each facility in each year, i.e., activity data, and (3) reported and estimated
levels of emissions control at each facility in each year. (For the other 3 fluorinated ga production facilities that
have reported their emissions under the GHGRP, 1990 through 2010 emissions were estimated using data
submitted by the company, as explained further below.)
Facility- and Chemical-Specific Emission Factors Reflecting Emissions Controls
Facility- and chemical-specific emission factors were developed based on the 2011 to 2015 emissions reported
under the GHGRP (discussed above) and the 2011 to 2015 production and transformation of fluorinated GHGs
reported under the GHGRP. (Production and transformation of CFCs and HCFCs are not reported under the
GHGRP.) For each emitted fluorinated GHG at each facility, emissions of the fluorinated GHG were summed over
the five-year period. This sum was then divided by the sum of the quantities of all fluorinated GHGs produced or
transformed at the facility over the five-year period.58 As discussed further below in the Uncertainty section,
emissions of any particular fluorinated GHG are likely to occur only from a subset of the production or
transformation processes at each facility, but in the absence of information on chemical-specific emissions at the
process level, it was assumed that all fluorinated GHG production and transformation processes at the facility emit
all fluorinated GHGs at the facility. This yielded the emission factors for each fluorinated GHG at each facility. Both
emissions and activity (production + transformation) totals were summed over the five-year period to account for
the intermittent and variable nature of some emissions and production/transformation processes. Compounds
that were not emitted or produced/transformed between 2011 and 2015 but that were emitted or
produced/transformed later were assumed not to have been emitted or produced/transformed (as applicable)
before 2011.
Facility- and Chemical-Specific Emission Factors Reflecting No Emissions Controls
The 2011 to 2015 emissions reported under the GHGRP reflect emissions controls to the extent those are
implemented at each facility. Because facilities have not always controlled their fluorinated GHG emissions since
1990, uncontrolled emission factors were developed for each facility to apply to years when the facility's emissions
were not believed to be controlled. To estimate uncontrolled emissions, GHGRP data were first used to assess the
2011 to 2015 levels of control for each production or transformation process at each facility.
To calculate uncontrolled emissions from each process and fluorinated GHG group, a point estimate of the
effective destruction efficiency (EDE, described above) was required and was estimated using the arithmetic mean
58 Permit data for two facilities indicated that they began controlling emissions at some point between 2011 and 2015.
However, the actual emissions reported by these facilities did not change substantially after the date when the permit indicated
that controls were imposed. For this reason, the reported 2011 to 2015 emissions and emission factors are believed to be
representative of emissions for these facilities before 2011.
4-84 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
of the lower and upper bounds of the EDE range reported for the process.59 This was consistent with the approach
taken in the 2019 Refinement to develop the Tier 1 factor for fluorinated gas production facilities. The reported
vented emissions for each process and fluorinated GHG group were divided by (1 - arithmetic mean EDE) to obtain
the estimated uncontrolled emissions from process vents for that process and fluorinated GHG group. For each
fluorinated GHG group, the controlled emissions across processes (including emissions from both vents and leaks)
and the uncontrolled emissions across processes (including emissions from both vents and leaks) were then
summed, and the first was divided by the second. This yielded an average level of control for each fluorinated GHG
group at each facility. All fluorinated GHGs within each fluorinated GHG group at each facility were assumed to be
controlled to the same level. To estimate the uncontrolled emissions of each fluorinated GHG within each group at
each facility, the emissions of each fluorinated GHG were divided by the level of control estimated for its
fluorinated GHG group at the facility. The same procedure was used to estimate uncontrolled emission factors as
had been used to estimate controlled emission factors: the estimated uncontrolled 2011 to 2015 emissions of each
fluorinated GHG were summed, and this sum was divided by the sum of the quantities of all fluorinated GHGs
produced or transformed at the facility from 2011 to 2015.
Table 4-69: Destruction Efficiency Range Values Used to Estimate Pre-Abatement Emissions
for Production and Transformation Processes
DE ranges
Lower Bound
Upper Bound
Arithmetic Mean of Bounds
>=0% to <75%
0.0
0.75
0.375
>=75% to <95%
0.75
0.95
0.85
>=95% to <99%
0.95
0.99
0.97
>=99%
0.99
0.9999
0.995
Estimated Levels of Emissions Controls
As discussed above, both uncontrolled emission factors and controlled emission factors were developed for each
facility and fluorinated GHG; these emission factors were developed for estimating emissions from production and
transformation processes for years 1990 to 2010. The following information and assumptions were used to
determine whether and when emissions from facilities were likely to have been controlled from 1990 to 2010. For
the estimated status of emissions controls at each facility reporting under Subpart L, and, where relevant, the
starting year for those controls, see Table 4-70.
• Facilities with publicly available information on the presence and use of control devices were assumed to
control their emissions starting in the year specified in the publicly available information. Publicly
available information included operating permits, news articles on facility modifications, company press
releases, etc. Where the publicly available information documents that a control device was in place
beginning in a certain year, the facility was assumed to control process emissions beginning in that year,
and the controlled emission factor was used in estimating emissions for that year and the following years.
The uncontrolled emission factor was used to estimate emissions in earlier years.
• In the absence of other control information, facilities that never reported DRE ranges other than ">=0% to
<75%" for their production and transformation processes during reporting years 2011 and 2012 were
assumed to have no control devices in place during the time period 1990 to 2012.
• Facilities that reported DRE ranges other than ">=0% to <75%" for at least one production or
transformation process for 2011 or 2012 but for which other control information was not available were
assumed to have begun controlling their emissions in 2005.
59 Note that facilities would report a range of 0% to 75% even if they do not abate emissions at all; thus, the assumption that
emissions are 37.5 percent controlled may overestimate the hypothetical uncontrolled emissions of some facilities, e.g., those
that do not abate any emissions.
Industrial Processes and Product Use 4-85
-------
Table 4-70: Estimated Starting Years for Emission Controls at Each Fluorinated Gas
Production Facility Reporting under Subpart L of the GHGRP
Facility Name
Estimated Start Year
Basis of Estimation
3M COMPANY
No controls
Never reported a DRE range other than ">=0% to <75%"
3MCORDOVA
2003
Climate News Article60
3M Cottage Grove Center -
Site
2016
Reported a DRE range other than ">=0% to <75%" for the first time in
2016
Airgas Therapeutics LLC -
Scott Medical Products
No controls
Never reported a DRE range other than ">=0% to <75%"
ANDERSON DEVELOPMENT
COMPANY
No controls
Never reported a DRE range other than ">=0% to <75%"
ARKEMA, INC.
2005
Reported a DRE range other than ">=0% to <75%" in 2011
Chemours - Corpus Christi
Plant
No controls
Never reported a DRE range other than ">=0% to <75%"
CHEMOURS CHAMBERS
WORKS
2005
Reported a DRE range other than ">=0% to <75%" in 2011
CHEMOURS COMPANY-
2015
Reported a DRE range other than ">=0% to <75%" for the first time in
FAYETTEVILLE WORKS
2015
CHEMOURS EL DORADO
2005
Reported a DRE range other than ">=0% to <75%" in 2011
CHEMOURS LOUISVILLE
WORKS
No controls
Never reported a DRE range other than ">=0% to <75%"
CHEMOURS WASHINGTON
WORKS
2005
Reported a DRE range other than ">=0% to <75%" in 2011
DAIKIN AMERICA INC.
1993
Title V operating permit61
HONEYWELL
INTERNATIONAL INC-
2012
Title V operating permit62
BATON ROUGE PLANT
HONEYWELL
INTERNATIONAL INC -
2011
Title V operating permit63
GEISMAR COMPLEX
Honeywell Metropolis
No controls
Never reported a DRE range other than ">=0% to <75%" (did not
report under Subpart L)
MEXICHEM FLUOR INC.
1993
Title V operating permit64
Versum Materials US, LLC
No controls
Never reported a DRE range other than ">=0% to <75%"
Activity Data
The activity data for production and transformation of fluorinated compounds for 1990 to 2010 are based on
production and transformation data reported to EPA by certain facilities for certain years, on production capacity
data, and on fluorinated GHG production and consumption trends estimated for the various fluorinated GHG-
consuming industries.
Production and Production Capacity Data
Production data are available from reporting to the U.S. GHGRP under Subpart 00, Suppliers of Industrial
Greenhouse Gases, and from an industry survey conducted by U.S. EPA in 2008 and 2009. Production and
transformation data were reported under Subpart 00 for 2010 and later years. The responses to the industry
survey included production data for certain fluorinated gases at certain facilities for the years 2004, 2005, and
60 See https://insideclimatenews.org/news/29122022/3m-cordova-illinois-pfas-cf4-pollution/.
61 Daikin (2013) http://lf.adem.alabama.gov/WebLink/DocView.aspx?id=29951882&dbid=0.
62 Honeywell (2011) https://edms.deq.louisiana.gov/app/doc/view?doc=8579001.
63 Honeywell (2012) https://edms.deq.louisiana.gov/app/doc/view?doc=7812895.
64 See https://edms.deq. louisiana.gov/app/doc/view?doc=1309650.
4-86 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
2006. 2004 to 2006 production data are available for 15 fluorinated compounds. Year 2006 production at an SF6-
producing facility was estimated based on production capacity data as described below (Rand 2007). Production of
certain compounds at one other facility was estimated based on 2003 production capacity estimates from SRI
2004.
Estimated Production
Estimated production for facilities and fluorinated GHGs for which production or production capacity data were
available for some years before 2010.
For facilities and fluorinated GHGs for which production or production capacity data were available for 2006 or
2003, production between 2006 or 2003 (as applicable) and 2010 (or 2011) was estimated by interpolating
between the 2006 production or 2003 production capacity value and the 2010 (or 2011) production value reported
under Subpart OO.
For the years before the earliest year with production or production capacity data (e.g., years 1990 to 2002 or
2003), production was estimated based on growth or consumption trends for the major industries using each
fluorinated GHG.
• For fluorinated compounds that are commonly emitted in the semiconductor industry, estimates of U.S.
layer-weighted semiconductor production (Total Manufactured Layer Area, orTMLA) were used to inform
the fluorinated compound production estimates (EPA 2023c). Fluorinated compound production values
were assumed to vary with TMLA from 1990 to 2002 or 2003. For example, 1998 production of PFC-14 at
a particular facility was estimated by multiplying the 2003 production of PFC-14 at that facility by the ratio
between the TMLA estimated for 1998 and the TMLA estimated for 2003. Fluorinated compounds for
which TMLA was used to estimate production include PFC-14, PFC-116, PFC-218, perfluorocyclobutane (c-
C4F8), and NF3. (Note that the TMLA data were also extrapolated from year 1995 to 1990 based on the
average change per year from 1995 to 2009.)
• SFs is commonly used in electric power systems, magnesium production, and electronics manufacturing.
SFs consumption estimates across these three industries for 1990 to 2003 were used to inform the SF6
production data (EPA 2023d); SF6 production was assumed to vary with consumption totals from 1990 to
2003.
• For HFCs commonly used as replacements for ozone-depleting substances (ODS), such as HFCs used as
substitutes for CFCs and HCFCs in air-conditioning and refrigeration equipment, HFC production data for
certain fluorinated compounds from the Vintaging Model (VM) were used to inform the HFC production
estimates (EPA 2023b). HFC production values were assumed to vary with the VM estimates of
production. The industry trend data were applied to the list of HFCs in Table 4-71.
Industrial Processes and Product Use 4-87
-------
Table 4-71: List of Saturated HFCs, Unsaturated HFCs (Hydrofluoroolefins or HFOs), and
Unsaturated HCFCs (Hydrochlorofluoroolefins or HCFOs) whose 1990-2009 Production Was
Estimated Using Vintaging Model, Virgin Manufacturing by Chemical
Fluorinated Gas
HFC-23
HFC-32
HFC-125
HFC-134a
HFC-143a
HFC-152a
HFC-236fa
HFC-245fa
HFC-365mfc
HCFO-1233zd(E)
HFO-1234yf
HFO-1234ze
HFO-1336mzz(Z)
HFC-4310mee
Estimated production for facilities and fluorinated GHGs for which production data before 2010 were not
available.
In the absence of production data for the period 1990 to 2009, the production data reported to the GHGRP under
Subpart OO were extrapolated backward based on the industry trends discussed above. For compounds for which
industry trend data were unavailable, production was assumed to have remained constant over the time series.
In both cases, 2009 production was estimated by conducting a trend analysis on the Subpart OO production data
for years 2010 to 2015. In instances where there did not appear to be a trend, the average of the production
values for years 2010 to 2015 was used as the estimated production for year 2009. In instances where there was a
trend, the year 2010 (or 2011) production value was used as the estimated production for year 2009.
If the industry trend information discussed above was applicable to a fluorinated compound, it was assumed that
production varied with the industry trend from 1990 to 2009. If no industry trend information was available, it was
assumed that production from 1990 to 2008 remained constant at the 2009 value.
For facilities and fluorinated compounds where information was available on annual production capacity, the
estimated activity data was reviewed and compared to the known production capacity. For instances where the
estimated activity data exceeded known production capacity for a certain year, the production estimate was set
equal to the capacity value. In addition, where information was available on the starting year for production of a
fluorinated GHG at a facility, production was only estimated beginning in the process startup year through 2009.
Estimated Emissions for 3M Facilities
3M provided 1990,1995, 2000, and 2002 through 2010 emissions data for three facilities: 3M Cordova, 3M
Company, and 3M Cottage Grove Center - Site.65 Therefore, speciated 1990-2010 emissions at these facilities were
estimated using a different methodology than that described above.66
65 For 1990,1995, and 2000, 3M provided emissions data for a Pilot Development Center in addition to the other three
facilities. Emissions by group from the Pilot Development Center were added to and are represented by the emissions by group
for 3M Cottage Grove Center - Site.
66 3M's methods for estimating its emissions are described in detail in "3M Global EHS Laboratory Response to EPA Data
Request on Fluorochemical Emissions," February 2024 (3M, 2024). In brief, 3M estimated emissions from its processes using
4-88 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
3M emissions data were provided by facility and by fluorinated GHG group in metric tons of CO2 Eq., weighted by
100-year GWPs from various IPCC Assessment Reports. The fluorinated GHG groups included HFCs, PFCs, SF6, HFEs,
and other fluorinated GHGs. (3M noted that the "other fluorinated GHG" category included NF3.) GWPs from the
IPCC Third Assessment Report (TAR) were used to report totals for 1990,1995, 2000, and 2003 to 2006. GWPs from
the IPCC Fifth Assessment Report (AR5) was used to report totals for 2002, and GWPs from the IPCC Fourth
Assessment Report (AR4) were used to report totals for 2007 to 2010. The emissions were also categorized as
emissions from electrochemical fluorination (ECF) processes and downstream (DS) emissions. The DS emissions for
1990,1995, and 2000 were reported as a total for each facility rather than by fluorinated GHG group.
To present emissions estimates by compound, EPA needed to disaggregate the data provided by 3M. The first step
was to disaggregate the 1990, 1995, and 2000 DS emissions into fluorinated GHG groups. Since the 2003 to 2006
data were calculated using the same set of GWPs as the 1990,1995, and 2000 data, the DS emissions of each
group at each facility for 2003 to 2006 were divided by the total DS 2003 to 2006 emissions at that facility to
obtain a set of fractions. These fractions were then multiplied by the 1990,1995, and 2000 DS totals to sort those
emissions into groups.
For 2002, ECF and DS emissions were only reported under the PFC group. (3M noted that chemical classifications
were not preserved when 2002 emissions were recalculated using GWP values from AR5.) Since data for every
other year showed emissions reported from multiple fluorinated GHG groups, and since 2004 was the closest year
with emissions well sorted into fluorinated GHG groups, the ECF and DS emissions for 2002 were separated into
multiple fluorinated GHG groups using the 2004 ECF and DS groups shares for each facility.
The next step was to disaggregate the emissions of each fluorinated GHG group into emissions of the relevant
compounds in that group. To accomplish this, EPA assumed that emissions of each fluorinated GHG group before
2011 consisted of the same fluorinated GHGs, in the same proportions, as from 2011 through 2015. However, each
compound's share of the GWP-weighted emissions of the group in a given year depends on the GWPs used for that
compound and for the other compounds in the group in that year. To account for this, EPA multiplied the reported
2011 to 2015 emissions of each compound in metric tons by the corresponding GWPs for that compound from the
TAR, AR4, and AR5 to generate three sets of emissions by compound in metric tons of CO2 Eq. For each set, the
sum of emissions across 2011 to 2015 for each compound were divided by the total emissions for the
corresponding fluorinated GHG group for those five years to calculate shares for each group.
The 3M emissions data by group for 1990,1995, 2000, and 2002 to 2010 were then speciated by compound using
the appropriate set of share values for each year. Since 3M Company only reported emissions of one compound in
2011 to 2015 but had emissions from multiple fluorinated GHG groups in 1990,1995, 2000, and 2002 to 2010, the
share values for 3M Cordova were used to speciate the 1990,1995, 2000, and 2002-2010 emissions by group for
3M COMPANY. The speciated emissions in metric tons of CO2 Eq. by compound for each facility were then divided
by the appropriate TAR, AR4, or AR5 GWP for each compound to obtain the estimated emissions in metric tons of
each compound for 1990,1995, 2000, and 2002 to 2010.
Linear interpolation was then used to estimate emissions for 1991 to 1994,1996 to 1999, and 2001 for each
compound for these three facilities.
Estimated Emissions for Facilities that Produce Fluorinated GHGs but Do Not Report Under Subpart L
There is a subset of facilities that report production and transformation of fluorinated gases under Subpart OO and
that also have emission levels less than the threshold value for reporting under Subpart L (i.e., uncontrolled
emissions below the 25,000-MT CO2 Eq. threshold). For these facilities, EPA developed emission estimates based
on aggregated production estimates and the Tier 1 default emission factor in the 2019 Refinement. Because the
emission factors that were developed using methods similar to those used for developing emission factors under the GHGRP.
As under the GHGRP, emission factors were multiplied by different types of activity data (e.g., production) to estimate
emissions for each facility and year. In 2003 and later years, 3M also accounted for emission reductions attributable to
operation of the thermal oxidizer at the Cordova plant.
Industrial Processes and Product Use 4-89
-------
specific fluorinated GHGs emitted are not known, the emissions were assumed to consist of the fluorinated GHGs
shown in Table 3.28b of chapter 3.10.2 of Volume 3 IPPU (IPCC 2019), in the proportions shown in that table.
Emissions are assumed to have been flat at the 2010 value in the years before 2010.
Estimated Emissions for SFe Production Facility
For an SF6 production facility that ceased production in 2010, the year before emissions from fluorinated gas
production were required to be reported under the GHGRP, SF6 emissions were estimated using historical
production capacity, the global growth rate of SF6 sales reported in RAND 2007, and the Tier 1 default emission
factor for production of SF6 in the 2019 Refinement. For this plant, a 1982 SF6 production capacity of 1,200 short
tons (Perkins 1982) was multiplied by the ratio between the RAND survey SF6 sales totals for 2006 and 1982,1.52
(RAND 2007), resulting in estimated production of 1,652 metric tons in 2006. This production was assumed to have
declined linearly to zero in 2011.
2011-2022 Emissions Estimates
For the 17 fluorinated gas production facilities that have reported their emissions under the GHGRP, 2011 to 2022
emissions are estimated using the fluorinated GHG emissions reported under Subpart L of the GHGRP.
As discussed above, most emissions reported under Subpart L are reported by chemical, but some emissions are
reported only by fluorinated GHG group in MT CO2 Eq. Between 2011 and 2022, the share of total CO2 Eq.
emissions reported only by fluorinated GHG group has ranged between 1 and 2 percent. In this analysis, to ensure
that all emissions are reported by species, emissions that are reported only by fluorinated GHG group are assumed
to consist of the fluorinated GHGs in that group that are reported by chemical at the facility. As discussed further
in the Uncertainty section, this is likely to result in incorrect speciation of some emissions, but the impact of this
incorrect speciation is expected to be small.
Estimated Emissions for Facilities that Produce Fluorinated GHGs but Do Not Report Under Subpart L
As discussed above, for facilities that produce fluorinated GHGs but that do not report their emissions under
subpart L, EPA developed emission estimates based on aggregated production estimates and the Tier 1 default
emission factor in the 2019 Refinement. Because the specific fluorinated GHGs emitted are not known, the
emissions were assumed to consist of the fluorinated GHGs shown in Table 3.28b of chapter 3.10.2 of Volume 3
IPPU (IPCC 2019), in the proportions shown in that table.
Uncertainty
The estimates in this memo are subject to a number of uncertainties. These uncertainties are generally greater for
years before 2011, when reporting of fluorinated GHG emissions from fluorinated gas production began under the
GHGRP, than for 2011 and following years. However, the emissions estimated from 2011 to 2022 are also subject
to various uncertainties. The uncertainties for both the 1990 to 2010 and 2011 to 2022 periods are discussed in
more detail below.
1990-2010 Uncertainty
The uncertainty of emissions estimated for 1990 through 2010 is considerably greater than that for emissions for
2011 through 2022 because emissions were not reported under the GHGRP. EPA has estimated emissions using
estimated emission rates, fluorochemical production and transformation activity, and levels of control, and each
set of estimates is subject to uncertainty.
4-90 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
Uncertainty regarding activity data
Identity of emitting processes
In reality, emissions of particular fluorinated GHGs are linked to production and/or transformation of particular
fluorinated gases at facilities. However, GHGRP information/data does not link emissions of specific fluorinated
GHGs to production or transformation of specific fluorinated gases. For the estimates presented here, therefore,
all emissions are indexed to total production across all fluorinated gases. This may not capture trends in emissions
that are driven by trends in production or transformation of subsets of the fluorinated gases produced at a facility.
Produced and emitted gases change over time
The set of gases produced at a facility, and therefore the set of fluorinated GHGs that are emitted by that facility,
may change over time. It is likely that certain production and transformation processes that existed from 2011 to
2015 (the basis of the emission factors used to back-cast emissions in this analysis) did not exist throughout the
entire previous time series (1990 to 2010). In such cases, emissions of the fluorinated GHGs emitted from the new
processes will be overestimated by this analysis for certain years before 2011. On the other hand, it is also likely
that some production and transformation processes, and their associated fluorinated GHG emissions, occurred
only during the 1990 to 2010 period and not later, meaning that their emissions are not represented in the
emission factors developed based on the 2011 to 2015 emissions and production data collected under the GHGRP.
Such emissions will therefore not be captured by this analysis. The most prominent example of the second
situation is probably production of CFCs and HCFCs other than HCFC-22 between 1990 and 2009, which has
declined steadily since 1990 as the production of CFCs and HCFCs for emissive uses has been phased out under the
Montreal Protocol and Clean Air Act. Production of CFCs and HCFCs can sometimes result in emissions of HFCs or
PFCs.
Quantity of produced gases
Where production or production capacity data were available for certain fluorinated gases, facilities, and years
before 2010, those data were incorporated into this analysis. However, even for facilities and compounds for
which data were available in certain years, there were several years for which data were not available. For multiple
produced compounds, data were available only in 2010. To estimate trends in production of compounds for years
before production or production capacity data were available, production of certain compounds was indexed to
known national production or consumption trends for those compounds. This is the case for most HFCs, several
PFCs, SFs, and NF3. National production estimates are available for HFCs, increasing confidence in country-level
production estimates, but the distribution of production among the various HFC-producing facilities is uncertain.
Where estimated production was indexed to consumption (for several PFCs, SF6, and NF3), the uncertainty is larger
than for HFCs because changes in net imports/exports (which are not known) may also affect the production trend.
For certain fluorinated gases, trend information was not available; therefore, production was back-cast by
assuming that it had remained constant at the 2010 level from 1990 through 2009. This is a highly uncertain
assumption.
Some production and transformation activity is not reported under Subpart OO or modeled in back-casting
Under Subpart 00, quantities of fluorinated GHGs that are produced and transformed at the same facility are not
reported to us, although any emissions from such processes are reported under Subpart L. Such unreported
production and transformation are therefore not captured in the 1990 to 2010 activity estimates used to estimate
1990 through 2010 emissions. To the extent that such unreported production and transformation drive emissions
and change over time, the trends will not be captured by this analysis.
Facilities that no longer produce fluorinated gases or that started producing them after 1990
Some facilities may have produced fluorinated gases at some point between 1990 and 2010 that no longer
produced those compounds after 2010. One SF6 producer is known to fall into this category and its 1990 to 2010
emissions were estimated, but there may be other facilities that are not included in this analysis. On the other
hand, some facilities for which 1990 to 2010 emissions were estimated may not have produced them over the
Industrial Processes and Product Use 4-91
-------
entire time series, in which case emissions of the compounds those facilities are assumed to have emitted could be
overestimated.
Uncertainty regarding emission factors
Emission rates change over time
The emission factors used to estimate 1990 to 2010 emissions are based on the emissions and production reported
from 2011 to 2015, reflecting emission rates during that period. For processes that have been used throughout the
timeseries, emission rates may have changed over time as the process was optimized to increase efficiency,
decreasing by-product emissions, or alternatively, as the process was optimized to maximize production, which
sometimes increases by-product emissions. Emission rates also depend on the extent to which emissions are
controlled at the facility, the uncertainties for which are discussed further below.
Emissions from container venting and destruction may not scale with production
In this analysis, emissions from container venting and destruction of previously produced fluorinated GHGs were
included in the emission factors used to estimate 1990 to 2010 emissions. This implicitly assumes that such
emissions scale with production and transformation. While this seems likely to be broadly true, there may be
exceptions. However, since emissions from container venting and destruction are generally a small share of facility
emissions (2 percent, on average), the impact of such exceptions is expected to be small.
Uncertainty regarding levels of control
In this analysis, the arithmetic mean of the DRE range reported by each facility for each process was used to
estimate the DRE for that process and the uncontrolled emissions for that process. Since the emissions implied by
the bounds of each DRE range span at least a factor of four,67 this is an uncertain assumption. The uncertainty is
mitigated somewhat by the fact that there are generally several processes at each facility, meaning that
departures from the assumed mean average out to some extent. There is also uncertainty in the assumptions that
(1) all fluorinated GHGS within a particular fluorinated GHG group are abated to the same extent and (2) facilities
for which control device start dates are unavailable began to control emissions in 2005.
Quantitative uncertainty estimate for uncontrolled emission factors from 2019 Refinement
As noted above, 2011 to 2016 data from the GHGRP was used to develop the Tier 1 default uncontrolled emission
factor for the 2019 Refinement, using methods similar to those described here. A Monte Carlo analysis performed
to assess the uncertainty of the Tier 1 default factor indicated that the uncertainty for each facility's uncontrolled
emission factor was less than 50 percent. This uncertainty estimate considered the uncertainty regarding the levels
of control, but not the uncertainty of applying factors from one time period at the facility to much earlier time
period (although the variability of each facility's emission factor over the 6-year span of the 2019 Refinement
analysis was found to be relatively low).
Uncertainty regarding the identity of3M emitted compounds
For the three 3M facilities that submitted their 1990 through 2010 emissions by fluorinated GHG group, it is
assumed that the emissions of each group consist of the compounds in that group that were reported by species
by the facility from 2011 through 2014. However, 3M indicated that the mix of products made at its facilities had
changed over time, which would have affected the identities of the fluorinated GHGs emitted. For example, at one
facility, only one compound was reported to be emitted for 2011-2015 but 3M's historical emissions data showed
multiple fluorinated GHG groups for that facility for 1990 to 2010. For that facility, it is assumed that emissions
consisted of the compounds in each gas type that were reported by species by the 3M facility that was determined
to be the most similar. Additionally, there were a few years of data for these three facilities where some emissions
were reported as a total rather than separated by group. For these years, the group shares were assumed to be
67 For example, the DRE range 0 to 75% implies emissions of (1-0) x uncontrolled emissions to (1-75%) x uncontrolled
emissions, or, rearranging and calculating, 0.25 x uncontrolled emissions to 1 x uncontrolled emissions, a factor of four.
4-92 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
the same as in nearby years. There is uncertainty involved with each of the assumptions made while speciating
emissions for these three facilities, though the uncertainty is minimal regarding total emissions in metric tons of
CO2 Eq. In at least one case, the speciation method appears to have resulted in an overestimate for an individual
compound. In this analysis, 3M's emissions of C4F8O are estimated to have peaked at around 410 MT in 2000, but
at least one study has inferred considerably lower global emissions (around 120 MT) for that year based on
atmospheric measurements of C4F8O (Vollmer et al. 2019).
2011-2022 Uncertainty
Emissions from 2011 to 2022 reflect reporting by fluorinated gas production facilities under the GHGRP. As
discussed above, emissions reported under the GHGRP are based on facility- and process-specific measurements
or calculations and are therefore expected to be reasonably accurate for the reporting facilities. (Emissions from
the largest sources, process vents emitting 10,000 MT CO2 Eq. or more annually, are estimated using Tier 3
methods.)
Unverified reports
Ninety-five percent (171/180) of the Subpart L reports submitted by fluorinated gas production facilities from 2011
to 2022 are considered to be fully verified; five percent (9/180) of the reports include one or more data elements
that are not verified. One facility accounts for two thirds (6/9) of the unverified reports. Many of the issues in the
unverified reports for this facility relate to time-series inconsistencies that have arisen as the facility updates
reports for recent years, but not previous years, to reflect refinements to estimated emission rates. This facility has
accounted for between 6 percent (in 2011) and 29 percent (in 2022) of the GWP-weighted emissions reported for
this source category. The uncertainties for this facility therefore have an appreciable impact on the uncertainty of
the estimates for the source category as a whole, particularly in years before 2022.
Facilities that do not produce fluorinated gases but may emit fluorinated GHGsfrom other fluorochemical
production processes
Under the GHGRP, EPA collects information from facilities that produce fluorinated gases. While this likely includes
most, and possibly all, U.S. facilities that produce fluorochemicals of any kind, it is possible that some
fluorochemical producers do not report either their production of fluorochemicals or their emissions of fluorinated
GHGs to EPA under the GHGRP. In this case, emissions estimates based only on GHGRP reporting would
underestimate actual emissions.
At fluorinated gas production facilities that currently report their emissions under the GHGRP, it is possible that
some processes that emit fluorinated GHGs neither produce nor transform a fluorinated gas, in which case their
emissions would not be reported under the GHGRP. In that case, emissions estimates based only on GHGRP
reporting would underestimate actual emissions.
Exclusion of nitrous oxide
The GHGRP does not currently require facilities to report emissions of nitrous oxide (N2O) from fluorinated gas
production or transformation, but the IPCC 2019 Refinement includes a default emission factor for N2O from
production of NF3, implying such emissions may occur. The GHGRP data (and this analysis) may therefore
underestimate emissions of N2O from fluorinated gas production. Because the GWP of N2O is considerably lower
than that of saturated HFCs, PFCs, and other fluorinated GHGs, any underestimate is expected to be relatively
small.
Identity of emitted compounds
In this analysis, it is assumed that emissions that are reported only in MT CO2 Eq. by fluorinated GHG group consist
of the compounds in that group that are reported by species by the facility. However, if that were actually the
case, emissions of those compounds would have been included in the speciated emissions rather than reported
separately in MT CO2 Eq. This analysis therefore incorrectly speciates some emissions. As noted in the
Methodology section, the share of total CO2 Eq. emissions reported only by fluorinated GHG group is small, ranging
Industrial Processes and Product Use 4-93
-------
between 1 and 2 percent. Moreover, while the emissions are not assigned to the exact species emitted, they are
assigned to a species that is closely related and likely to have similar atmospheric impacts (e.g., another saturated
HFC with two or fewer carbon-hydrogen bonds). The impact of this uncertainty is therefore limited.
Quantities of Reactants Consumed or Fluorinated Gases Produced
The emissions reported under Subpart L are required to be calculated using process activity data, such as the
quantity of reactants consumed or the quantity of the fluorinated gas product produced. In general, the
uncertainties in process activity levels are expected to be small. The 2019 Refinement places such uncertainties "in
the region of 1 percent."
Because the uncertainties enumerated above are either small or difficult to quantify, EPA did not attempt to
include them in the 2022 quantitative uncertainty estimate for this source category. The 2022 quantitative
uncertainty estimate includes the following uncertainties:
Process Vent Emission Factors
Process vent emission factors that were developed based on stack testing (for continuous process vents emitting
10,000 MT CChEq. or more) were estimated to have an uncertainty (95-percent confidence interval) of ±35
percent based on Subpart L requirements.68 Process vent emission factors that were developed based on
calculations (for batch process vents and for continuous process vents emitting less than 10,000 MT CO2 Eq.) were
estimated to have a larger uncertainty of ±50 percent. Continuous processes were assumed to have two vents per
process; batch processes were assumed to have five vents per process.69
Equipment Leak Estimates
The estimated equipment leaks reported by each facility for each process were estimated to have an uncertainty
of ± 90 percent. The uncertainty of leak estimates depends on the method used to estimate leaks; there are
multiple methods. For simplicity, this analysis uses a conservatively high uncertainty estimate that is appropriate
for the Average EF Approach.
Venting of Residual Gas in Containers
The reported emissions of fluorinated GHGs from venting of residual gas in returned containers were estimated to
have an uncertainty of ± 30 percent for each facility. This estimate is based on the Subpart L requirement to either
measure the contents of each container or to measure the contents of at least 30 representative containers for
each compound and container size and type.
Facilities that produce fluorinated gases but do not report their emissions to the GHGRP
EPA estimated emissions for fluorinated gas production facilities that do not report their emissions under Subpart
L of the GHGRP. The estimates presented here for 2011 to 2022 are based on aggregated production estimates,
the Tier 1 default emission factor in the 2019 Refinement, and the default fluorinated GHG speciation from Table
3.28b (chapter 3.10.2 of 2019 Refinement). There is considerable uncertainty in both the magnitude of the
emissions and the identity of the emitted compounds. The 2019 Refinement estimates the uncertainty of the Tier 1
emission factor as -98 percent to +470 percent. (In the quantitative uncertainty estimate below, which is based on
error propagation, values of ±98 percent were used because error propagation requires the assumption of
symmetric uncertainty bounds.) In 2022, estimated emissions from the six non-reporting facilities accounted for 44
percent of total estimated U.S. emissions from fluorinated gas production and transformation. In contrast,
68 Technical Support Document for Emissions from Production of Fluorinated Gases, Office of Air and Radiation, U.S.
Environmental Protection Agency, November 5, 2010. Available online at: https://www.epa.gov/sites/default/files/2Q15-
02/documents/subpart-l techsuppdoc.pdf.
69 Economic Impact Analysis for the Mandatory Reporting of Greenhouse Gas Emissions F-Gases: Subparts I, L, DD, QQ, SS, U.S.
Environmental Protection Agency, November 2010. Available online at: https://www.regulations.gov/document/EPA-HQ-OAR-
2009-0927-0179.
4-94 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
production and transformation by these facilities accounted for just three percent of total fluorinated GHG
production and transformation across all facilities. Because the emissions estimated for several of the six facilities
exceeded the 25,000 MT CO2 Eq. reporting threshold under Subpart L, but these facilities have not reported their
emissions under Subpart L, it appears likely that emissions from at least some facilities are overestimated.
The four uncertainties listed immediately above were convolved using error propagation to arrive at an overall
uncertainty estimate for 2022. The results of the Approach 1 quantitative uncertainty analysis are summarized in
Table 4-72. Emissions of HFCs, PFCs, SF6, and NF3 from production of fluorochemicals other than HCFC-22 were
estimated to fall between 4.83 and 7.08 MMT CO2 Eq. at the 95 percent confidence level. This indicates a range of
approximately 19 percent below and 19 percent above the emission estimate of 5.95 MMT CO2 Eq.
Table 4-72: Approach 1 Quantitative Uncertainty Estimates for HFC, PFC, SF6, and NF3 from
Production of Fluorochemicals other than HCFC-22 (MMT CO2 Eq. and Percent)
Source
Gas
2022 Emission
Estimate
(MMT CO . Eq.)
Uncertainty Range Relative to Emission Estimate
(MMTCO' Eq.) (%)
Lower
Bound1
Upper
Bound"
Lower Upper
Bound Bound
Production of
Fluorochemicals
HFCs, PFCs, SF6,
and NFs
5.95
4.83
7.08
-19% +19%
other than HCFC-22
a Absolute lower and upper bounds were calculated using the corresponding lower and upper bounds in percentages.
QA/QC and Verification
General quality assurance/quality control (QA/QC) procedures were applied consistent with the U.S. Inventory
QA/QC plan, which is in accordance with Volume 1, Chapter 6 of 2006IPCC Guidelines as described in the
introduction of the IPPU chapter (see Annex 8 for more details). Under the GHGRP, EPA verifies annual facility-level
reports through a multi-step process (e.g., including a combination of pre-and post-submittal electronic checks and
manual reviews by staff) to identify potential errors and ensure that data submitted to EPA are accurate,
complete, and consistent (EPA 20 15).70 Based on the results of the verification process, EPA follows up with
facilities to resolve mistakes that may have occurred. The post-submittals checks are consistent with a number of
general and category-specific QC procedures, including: range checks, statistical checks, algorithm checks, and
year-to-year checks of reported data and emissions.
The GHGRP also requires source-specific quality control measures for the Fluorinated Gas Production category.
Under the GHGRP, fluorinated gas producers are required to (1) develop and periodically update process vent-
specific emission factors using either measurements or engineering calculations, depending on the nature of the
process (continuous vs. batch) and the magnitude of emissions from the vent, (2) take more measurements of vent
emissions where variability is high, (3) use methods for sampling, measuring volumetric flow rates, non-
fluorinated-GHG gas analysis, and measuring stack gas moisture that have been validated using a scientifically
sound validation protocol, (4) use a quality-assured analytical measurement technology capable of detecting the
analyte of interest at the concentration of interest and use a sampling and analytical procedure validated with the
analyte of interest at the concentration of interest, (5) periodically test the performance of destruction devices
used to control emissions, (6) account for any malfunctions in the process or destruction device, (6) account for
emissions from equipment leaks, (7) measure the quantities of residual gas that are vented from returned
containers (or develop an emission factor based on at least 30 measurements per gas and container size and type),
(8) calibrate mass measurement devices at the frequency recommended by the manufacturer using traceable
70 EPA (2015). Greenhouse Gas Reporting Program Report Verification. Available online at:
https://www.epa.Eov/sites/production/files/2015-07/documents/ehgrp verification factsheet.pdf.
Industrial Processes and Product Use 4-95
-------
standards and suitable methods published by a consensus standards organization, (9) calibrate analytical
equipment used to determine the concentration of fluorinated GHGs, and (10) document all measurements and
calibrations.
The 1990,1995, 2000, and 2002 through 2010 emissions data reported by 3M for three facilities was compared to
the 1990 through 2010 emissions previously calculated for those facilities using the same calculation method used
for other facilities that have reported their emissions under the GHGRP since 2011. The overall trajectory of the
3M-reported emissions, as well as the minima and maxima of those emissions, were similar to those previously
calculated, but the increases and decreases in the 3M-reported emissions were more gradual. 3M explained that
the gradual changes were due to changes in the compounds and quantities produced and to the gradual
deployment and optimization of the destruction device at the 3M Cordova facility.
Recalculations
This is a new category included for the current (i.e., 1990 to 2022) Inventory, thus, no recalculations were
performed.
Planned Improvements
EPA is planning to refine its estimates of emissions from non-reporting facilities after confirming with the facilities
that their actual per-facility uncontrolled emissions fall below 25,000 MT CO2 Eq.. EPA is also planning to refine its
estimates of emissions for other facilities between 1990 and 2009, e.g., by comparing these against emissions
inferred from atmospheric measurements. Moreover, EPA is continuing to seek datasets that can be used to
improve and/or QA/QC emissions estimates, particularly for the years 1990 to 2009. These datasets may include,
for example, real-time facility-specific estimates or additional global "top-down," atmosphere-based emissions
estimates that could be used to establish an upper limit on emissions of certain compounds.
4.16 Carbon Dioxide Consumption (CRT
Source Category 2B10)
Carbon dioxide (CO2) is used for a variety of commercial applications, including food processing, chemical
production, carbonated beverage production, and refrigeration, and is also used in petroleum production for
enhanced oil recovery (EOR). CO2 used for EOR is injected underground to enable additional petroleum to be
produced. For the purposes of this analysis, CO2 used in food and beverage applications is assumed to be emitted
to the atmosphere. This reporting category (2B10) includes emissions from IPCC assessment reports that do not
fall within any other CRT source category, which includes emissions from CO2 consumption. A further discussion of
CO2 used in EOR is described in the Energy chapter in Box 3-6 titled "Carbon Dioxide Transport, Injection, and
Geological Storage" and is not included in this section.
Carbon dioxide is produced from naturally-occurring CO2 reservoirs, as a byproduct from the energy and industrial
production processes (e.g., ammonia production, fossil fuel combustion, ethanol production), and as a byproduct
from the production of crude oil and natural gas, which contain naturally occurring CO2 as a component.
In 2022, the amount of CO2 produced and captured for commercial applications and subsequently emitted to the
atmosphere was 5.0 MMT CO2 Eq. (5,000 kt) (see Table 4-73 and Table 4-74). This is less than a 1 percent increase
(10 kt) from 2021 levels and is an increase of approximately 240 percent (3,528 kt) since 1990.
4-96 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
Table 4-73: CO2 Emissions from CO2 Consumption (MMT CO2 Eq.)
Year 1990 2005 2018 2019 2020 2021 2022
C02 Consumption 1.5 | 1.4 4.1 4.9 5.0 5.0 5.0
Table 4-74: CO2 Emissions from CO2 Consumption (kt CO2)
Year 1990 2005 2018 2019 2020 2021 2022
C02 Consumption 1,472 1,375 4,130 4,870 4,970 4,990 5,000
Methodology and Time-Series Consistency
Carbon dioxide emission estimates for 1990 through 2022 utilize a country-specific method and were based on the
quantity of CO2 extracted and transferred for industrial applications (i.e., non-EOR end-uses). Some of the CO2
produced by these facilities is used for EOR, and some is used in other commercial applications (e.g., chemical
manufacturing, food and beverage). The IPCC does not have specific methodological guidelines for CO2
consumption, but the country-specific methodology used is consistent with a Tier 3 approach since it relies on
facility-specific information.
2010 through 2022
For 2010 through 2022, data from EPA's GHGRP (Subpart PP) were aggregated from facility-level reports to
develop a national-level estimate for use in the Inventory (EPA 2023). Facilities report CO2 extracted or produced
from natural reservoirs and industrial sites, and CO2 captured from energy and industrial processes and transferred
to various end-use applications to EPA's GHGRP. This analysis includes only reported CO2 transferred to food and
beverage end-uses. EPA is continuing to analyze and assess integration of CO2 transferred to other end-uses to
enhance the completeness of estimates under this source category. Other end-uses include industrial applications,
such as metal fabrication. EPA is analyzing the information reported to ensure that other end-use data excludes
non-emissive applications and publication will not reveal CBI. Additionally, a small amount of CO2 is used as a
refrigerant; use and emissions from this application are reported under Section 4.25 Substitution of Ozone
Depleting Substances (CRT Source Category 2F). Reporters subject to EPA's GHGRP Subpart PP are also required to
report the quantity of CO2 that is imported and/or exported. Currently, these data are not publicly available
through the GHGRP due to data confidentiality reasons and hence are excluded from this analysis.
Facilities subject to Subpart PP of EPA's GHGRP are required to measure CO2 extracted or produced. More details
on the calculation and monitoring methods applicable to extraction and production facilities can be found under
Subpart PP: Suppliers of Carbon Dioxide of the regulation, Part 9871 The number of facilities that reported data to
EPA's GHGRP Subpart PP (Suppliers of Carbon Dioxide) for 2010 through 2022 is much higher (ranging from 44 to
53) than the number of facilities included in the Inventory for the 1990 to 2009 time period prior to the availability
of GHGRP data (4 facilities). The difference is largely due to the fact the 1990 to 2009 data includes only CO2
transferred to end-use applications from naturally occurring CO2 reservoirs and excludes industrial sites.
1990 through 2009
For 1990 through 2009, data from EPA's GHGRP are not available. For this time period, CO2 production data from
four naturally-occurring CO2 reservoirs were used to estimate annual CO2 emissions. These facilities were Jackson
Dome in Mississippi, Bravo and West Bravo Domes in New Mexico, and McCallum Dome in Colorado. The facilities
in Mississippi and New Mexico produced CO2 for use in both EOR and in other commercial applications (e.g.,
71 See http://www.ecfr.gov/cgi-bin/text-idx7tpN/ecfrbrowse/Title40/40cfr98 main Q2.tpl.
Industrial Processes and Product Use 4-97
-------
chemical manufacturing, food production). The fourth facility in Colorado (McCallum Dome) produced CO2 for
commercial applications only (New Mexico Bureau of Geology and Mineral Resources 2006).
Carbon dioxide production data and the percentage of production that was used for non-EOR applications for the
Jackson Dome, Mississippi facility were obtained from Advanced Resources International (ARI 2006, 2007) for 1990
to 2000, and from the Annual Reports of Denbury Resources (Denbury Resources 2002 through 2010) for 2001 to
2009 (see Table 4-75). Denbury Resources reported the average CO2 production in units of MMCF CO2 per day for
2001 through 2009 and reported the percentage of the total average annual production that was used for EOR.
Production from 1990 to 1999 was set equal to 2000 production, due to lack of publicly available production 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 (ARI 1990 to 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; 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 4-75: CO2 Production (kt CO2) and the Percent Used for Non-EOR Applications
Total CO.
Jackson Dome,
Bravo Dome,
West Bravo
McCallum Dome,
Production
MS
NM
Dome, NM
CO
from Extraction
%
CO ¦ Production
CO ¦ Production
CO' Production
CO ¦ Production
and Capture
Non-
Year
(kt) (% Non-EOR)
(kt) (% Non-EOR)
(kt) (% Non-EOR)
(kt) (% Non-EOR)
Facilities (kt)
EOR1
1990
1,344 (100%)
63 (1%)
+
65 (100%)
NE
NE
2005
1,254 (27%)
58(1%)
+
63 (100%)
NE
NE
2018
IE
IE
IE
IE
58,400b
7%
2019
IE
IE
IE
IE
61,300b
8%
2020
IE
IE
IE
IE
44,700b
11%
2021
IE
IE
IE
IE
43,980b
11%
2022
IE
IE
IE
IE
46,800b
11%
+ Does not exceed 0.5 percent.
NE (Not Estimated)
IE (Included Elsewhere)
a Includes only food and beverage applications.
b For 2010 through 2022, the publicly available GHGRP data were aggregated at the national level based on GHGRP CBI
criteria. The Dome-specific C02 production values are accounted for (i.e., included elsewhere) in the Total C02
Production from Extraction and Capture Facilities values starting in 2010 and are not able to be disaggregated.
Methodological approaches were applied to the entire time series to ensure consistency in emissions from 1990
through 2022. The methodology for CO2 consumption spliced activity data from two different sources: Industry
data for 1990 through 2009 and GHGRP data starting in 2010. Consistent with the 2006IPCC Guidelines, the
overlap technique was applied to compare the two data sets for years where there was overlap (IPCC 2006). The
data sets were determined to be inconsistent; the GHGRP data include CO2 from industrial sources while the
industry data do not. No adjustments were made to the activity data for 1990 through 2009 because the 2006 IPCC
Guidelines indicate that it is not good practice to use the overlap technique when the data sets are inconsistent.
Uncertainty
There is uncertainty associated with the data reported through EPA's GHGRP. Specifically, there is uncertainty
associated with the amount of CO2 consumed for food and beverage applications, given the GHGRP does have
provisions that Subpart PP reporters are not required to report to the GHGRP if their emissions fall below certain
4-98 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
thresholds, in addition to the exclusion of the amount of CO2 transferred to all other end-use categories. This latter
category might include CO2 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. EPA verifies annual
facility-level reports through a multi-step process (e.g., combination of electronic checks and manual reviews by
staff) to identify potential errors and ensure that data submitted to EPA are accurate, complete, and consistent.
Based on the results of the verification process, EPA follows up with facilities to resolve mistakes that may have
occurred.72 Given the lack of specific uncertainty ranges available on the data used, EPA assigned an uncertainty
range of ±5 percent and a normal probability density function for CO2 consumed for food and beverage
applications. The uncertainty range is derived from the default range for solvent use in Section 5.5 of Chapter 3 of
the 2006IPCC Guidelines. These values are representative of CO2 used in food and beverage based on expert
judgment (RTI 2023).
The results of the Approach 2 quantitative uncertainty analysis are summarized in Table 4-76. Carbon dioxide
consumption CO2 emissions for 2022 were estimated to be between 4.8 and 5.2 MMT CO2 Eq. at the 95 percent
confidence level. This indicates a range of approximately 5 percent below to 5 percent above the emission
estimate of 5.0 MMT CO2 Eq.
Table 4-76: Approach 2 Quantitative Uncertainty Estimates for CO2 Emissions from CO2
Consumption (MMT CO2 Eq. and Percent)
Source Gas
2022 Emission Estimate
(MMT CO . Eq.)
Uncertainty Range Relative to Emission Estimate-1
(MMTCO' Eq.) (%)
Lower Upper
Bound Bound
Lower Upper
Bound Bound
C02 Consumption C02
5.0
4.8 5.2
-5% +5%
a Range of emission estimates predicted by Monte Carlo stochastic simulation for a 95 percent confidence interval.
QA/QC and Verification
General quality assurance/quality control (QA/QC) procedures were applied consistent with the U.S. Inventory
QA/QC plan, which is in accordance with Volume 1, Chapter 6 of 2006 IPCC Guidelines as described in the
introduction of the IPPU chapter (see Annex 8 for more details). More details on the greenhouse gas calculation,
monitoring and QA/QC methods applicable to CO2 Consumption can be found under Subpart PP (Suppliers of
Carbon Dioxide) of the regulation (40 CFR Part 98).73 EPA verifies annual facility-level GHGRP reports through a
multi-step process (e.g., combination of electronic checks and manual reviews) to identify potential errors and
ensure that data submitted to EPA are accurate, complete, and consistent (EPA 20 15).74 Based on the results of
the verification process, EPA follows up with facilities to resolve mistakes that may have occurred. The post-
submittals checks are consistent with a number of general and category-specific QC procedures, including range
checks, statistical checks, algorithm checks, and year-to-year checks of reported data and emissions.
Recalculations Discussion
No recalculations were performed for the 1990 through 2021 portion of the time series.
72 See https://www.epa.gov/sites/production/files/2015-07/documents/ghgrp verification factsheet.pdf.
73 See http://www.ecfr.gov/cgi-bin/text-idxPtpk/ecfrbrowse/Title40/40cfr98 main Q2.tpl.
74 See https://www.epa.gov/sites/production/files/2015-07/documents/ghgrp verification factsheet.pdf.
Industrial Processes and Product Use 4-99
-------
Planned Improvements
EPA will continue to evaluate the potential to include additional GHGRP data on other emissive end-uses to
improve the accuracy and completeness of estimates for this source category. Particular attention will be made to
ensuring time-series consistency of the emissions estimates presented in future Inventory reports, consistent with
IPCC and UNFCCC guidelines. This is required as the facility-level reporting data from EPA's GHGRP, with the
program's initial requirements for reporting of emissions in calendar year 2010, are not available for all inventory
years (i.e., 1990 through 2009) as required for this Inventory. In implementing improvements and integration of
data from EPA's GHGRP, EPA will rely on the latest guidance from the IPCC on the use of facility-level data in
national inventories.75
These improvements are still in process and will be incorporated into future Inventory reports. These are near-to
medium-term improvements.
4.17 Phosphoric Acid Production (CRT
Source Category 2B10)
Phosphoric acid (H3PO4) is a basic raw material used in the production of phosphate-based fertilizers. Phosphoric
acid production from natural phosphate rock is a source of carbon dioxide (CO2) emissions, due to the chemical
reaction of the inorganic carbon (calcium carbonate) component of the phosphate rock. This reporting category
(2B10) includes emissions that do not fall within any other CRT source category, which includes production of
phosphoric acid. Emissions from fuels consumed for energy purposes during the production of phosphoric acid are
accounted for as part of fossil fuel combustion in the industrial end-use sector reported under the Energy chapter.
Phosphate rock is mined in Florida and North Carolina, which account for more than 75 percent of total domestic
output, and in Idaho and Utah (USGS 2023). It is used primarily as a raw material for wet-process phosphoric acid
production. The composition of natural phosphate rock varies, depending on the location where it is mined.
Natural phosphate rock mined in the United States generally contains inorganic carbon in the form of calcium
carbonate (limestone) and may also contain organic carbon.
The phosphoric acid production process involves chemical reaction of the calcium phosphate (Ca3(PC>4)2)
component of the phosphate rock with sulfuric acid (H2SO4) and recirculated phosphoric acid (H3PO4) (EFMA 2000).
Phosphate rock also contains naturally occurring limestone (CaCOs), ranging from 0.2 to 4.5 percent (as CO2), with
domestic phosphate rock from Florida containing 3.1 percent limestone (as CO2) (EFMA 2000). The generation of
CChfrom limestone in the phosphate rock is from the associated limestone-sulfuric acid reaction, as shown below:
CaCO3 + //2SO4 + H20 —* CaS04 ¦ 2H20 + C02
Total U.S. phosphate rock production in 2022 was an estimated 21 million metric tons (USGS 2023). Between 1990
and 2022, domestic phosphate rock production decreased by approximately 58 percent. Total imports of
phosphate rock to the United States in 2022 were 2.4 million metric tons (USGS 2023). Between 2018 and 2021,
most of the imported phosphate rock (95 percent) came from Peru, with 5 percent from Morocco (USGS 2023). All
phosphate rock mining companies in the United States are vertically integrated with fertilizer plants that produce
phosphoric acid located near the mines.
Total CO2 emissions from phosphoric acid production were 0.8 MMT CO2 Eq. (840 kt CO2) in 2022 (see Table 4-77
and Table 4-78). Domestic consumption of phosphate rock in 2022 was estimated to have decreased 3.9 percent
75 See http://www.ipcc-nggip.iges.or.jp/public/tb/TFl Technical Bulletin l.pdf.
4-100 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
relative to 2021 levels. The COVID-19 pandemic did not impact the domestic phosphate rock market as both the
fertilizer industry and related agricultural businesses were considered essential industries and were unaffected by
pandemic "stay-at-home" orders issued in March 2020 (USGS 2021a).
Table 4-77: CO2 Emissions from Phosphoric Acid Production (MMT CO2 Eq.)
Year 1990 2005 2018 2019 2020 2021
2022
Phosphoric Acid Production 1.5 | 1.3 | 0.9 0.9 0.9 0.9
0.8
Table 4-78: CO2 Emissions from Phosphoric Acid Production (kt CO2)
Year 1990 2005 2018 2019 2020 2021
2022
Phosphoric Acid Production 1,529 | 1,342 | 937 909 901 874
840
Methodology and Time-Series Consistency
The United States uses a country-specific methodology consistent with and comparable to an IPCC Tier 1 approach
to calculate emissions from production of phosphoric acid from phosphate rock based on the stoichiometry of the
process reaction shown above. The 2006 IPCC Guidelines do not provide a method for estimating process
emissions (CO2) from phosphoric acid production. Carbon dioxide emissions from production of phosphoric acid
from phosphate rock are estimated by multiplying the average amount of inorganic carbon (expressed as CO2)
contained in the natural phosphate rock as calcium carbonate by the amount of phosphate rock that is used
annually to produce phosphoric acid, accounting for domestic production and net imports for consumption. The
estimation methodology is as follows:
Equation 4-10: CO2 Emissions from Phosphoric Acid Production
Epa Cpr * Qpr
where,
Epa = CO2 emissions from phosphoric acid production, metric tons
Cpr = Average amount of carbon (expressed as CO2) in natural phosphate rock, metric ton
CO2/ metric ton phosphate rock
Qpr = Quantity of phosphate rock used to produce phosphoric acid
The CO2 emissions calculation methodology assumes that all of the inorganic carbon (calcium carbonate) content
of the phosphate rock reacts to produce CO2 in the phosphoric acid production process and is emitted with the
stack gas. The methodology also assumes that none of the organic carbon content of the phosphate rock is
converted to CO2 and that all of the organic carbon content remains in the phosphoric acid product.
From 1993 to 2004, the U.S. Geological Survey (USGS) Mineral Yearbook: Phosphate Rock disaggregated phosphate
rock mined annually in Florida and North Carolina from phosphate rock mined annually in Idaho and Utah, and
reported the annual amounts of phosphate rock exported and imported for consumption (see Table 4-79). For the
years 1990 through 1992, and 2005 through 2022, only nationally aggregated mining data was reported by USGS.
For the years 1990,1991, and 1992, the breakdown of phosphate rock mined in Florida and North Carolina and the
amount mined in Idaho and Utah are approximated using data reported by USGS for the average share of U.S.
production in those states from 1993 to 2004. For the years 2005 through 2022, the same approximation method
is used, but the share of U.S. production was assumed to be consistent with the ratio of production capacity in
those states, which were obtained from the USGS commodity specialist for phosphate rock (USGS 2012; USGS
2021b). For 1990 through 2022, data on U.S. domestic consumption of phosphate rock, consisting of domestic
reported sales and use of phosphate rock, exports of phosphate rock (primarily from Florida and North Carolina),
and imports of phosphate rock for consumption, were obtained from USGS Minerals Yearbook: Phosphate Rock
Industrial Processes and Product Use 4-101
-------
(USGS 1994 through 2015b) and from USGS Minerals Commodity Summaries: Phosphate Rock (USGS 2016 through
2021a, 2022). From 2004 through 2022, the USGS reported no exports of phosphate rock from U.S. producers
(USGS 2022).
The carbonate content of phosphate rock varies depending upon where the material is mined. Composition data
for domestically mined and imported phosphate rock were provided by the Florida Institute of Phosphate
Research, now known as the Florida Industrial and Phosphate Research Institute (FIPR 2003a). Phosphate rock
mined in Florida contains approximately 1 percent inorganic C, and phosphate rock imported from Morocco
contains approximately 1.46 percent inorganic C. Calcined phosphate rock mined in North Carolina and Idaho
contains approximately 0.41 percent and 0.27 percent inorganic C, respectively (see Table 4-79). Similar to the
phosphate rock mined in Morocco, phosphate rock mined in Peru contains approximately 5 percent CO2 (Golder
Associates and M3 Engineering 2016).
Carbonate content data for phosphate rock mined in Florida are used to calculate the CO2 emissions from
consumption of phosphate rock mined in Florida and North Carolina (more than 75 percent of domestic
production), and carbonate content data for phosphate rock mined in Morocco and Peru are used to calculate CO2
emissions from consumption of imported phosphate rock. The CO2 emissions calculation assumes that all of the
domestic production of phosphate rock is used in uncalcined form. As of 2006, the USGS noted that one phosphate
rock producer in Idaho produces calcined phosphate rock; however, no production data were available for this
single producer (USGS 2006). The USGS confirmed that no significant quantity of domestic production of
phosphate rock is in the calcined form (USGS 2012).
Table 4-79: Phosphate Rock Domestic Consumption, Exports, and Imports (kt)
Location/Year
1990
2005
2018
2019
2020
2021
2022
U.S. Domestic Consumption3
49,800
35,200
23,300
23,400
22,600
21,900
21,000
FL and NC
42,494
28,160 5
18,170
18,250
17,630
17,080
16,380
ID and UT
7,306
7,040
5,130
5,150
4,970
4,820
4,620
Exports—FL and NC
6,240 is
0 II!
0
0
0
0
0
Imports
451
2,630
2,770
2,140
2,520
2,460
2,400
Total U.S. Consumption
44,011
37,830
26,070
25,540
25,120
24,360
23,400
a U.S. domestic consumption values are based on reported phosphate rock sold or used by producers.
Note: Totals may not sum due to independent rounding.
Table 4-80: Chemical Composition of Phosphate Rock (Percent by Weight)
North
Central
North
Carolina
Idaho
Composition
Florida
Florida
(calcined)
(calcined)
Morocco
Peru
Total Carbon (as C)
1.60
1.76
0.76
0.60
1.56
NA
Inorganic Carbon (as C)
1.00
0.93
0.41
0.27
1.46
NA
Organic Carbon (as C)
0.60
0.83
0.35
0.00
0.10
NA
Inorganic Carbon (as CO2)
3.67
3.43
1.50
1.00
5.00
5.00
NA (Not Available)
Sources: FIPR (2003a), Golder Associates and M3 Engineering (2016)
Methodological approaches were applied to the entire time series to ensure consistency in emissions estimates
from 1990 through 2022.
Uncertainty
Phosphate rock production data used in the emission calculations were developed by the USGS through monthly
and semiannual voluntary surveys of the active phosphate rock mines during 2021. Prior to 2006, USGS provided
the data disaggregated regionally; however, beginning in 2006, only total U.S. phosphate rock production was
4-102 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
reported. Regional production for 2021 was estimated based on regional production data from 2017 to 2020 and
multiplied by regionally-specific emission factors. There is uncertainty associated with the degree to which the
estimated 2021 regional production data represents actual production in those regions. Total U.S. phosphate rock
production data are not considered to be a significant source of uncertainty because all the domestic phosphate
rock producers report their annual production to the USGS. Data for exports of phosphate rock used in the
emission calculations are reported to the USGS by phosphate rock producers and are not considered to be a
significant source of uncertainty. Data for imports for consumption are based on international trade data collected
by the U.S. Census Bureau. These U.S. government economic data are not considered to be a significant source of
uncertainty. Based on expert judgement of the USGS, EPA assigned an uncertainty range of ±5 percent to the
percentage of phosphate rock produced from Florida and North Carolina, and ±5 percent to phosphoric acid
production and imports (USGS 2012). Per this expert judgment, a normal probability density function was assigned
for all activity data.
An additional source of uncertainty in the calculation of CO2 emissions from phosphoric acid production is the
carbonate composition of phosphate rock, as the composition of phosphate rock varies depending upon where the
material is mined and may also vary over time. The Inventory relies on one study (FlPR 2003a) of chemical
composition of the phosphate rock; limited data are available beyond this study. Another source of uncertainty is
the disposition of the organic carbon content of the phosphate rock. A representative of FIPR indicated that in the
phosphoric acid production process, the organic carbon content of the mined phosphate rock generally remains in
the phosphoric acid product, which is what produces the color of the phosphoric acid product (FIPR 2003b).
Organic carbon is therefore not included in the calculation of CO2 emissions from phosphoric acid production.
A third source of uncertainty is the assumption that all domestically-produced phosphate rock is used in
phosphoric acid production and used without first being calcined. Calcination of the phosphate rock would result
in conversion of some of the organic carbon in the phosphate rock into CO2; however, according to air permit
information available to the public, at least one facility has calcining units permitted for operation (NCDENR 2013).
Finally, USGS indicated that in 2021 less than 5 percent of domestically-produced phosphate rock was used to
manufacture elemental phosphorus and other phosphorus-based chemicals, rather than phosphoric acid (USGS
2022). According to USGS, there is only one domestic producer of elemental phosphorus, in Idaho, and no data
were available concerning the annual production of this single producer. Elemental phosphorus is produced by
reducing phosphate rock with coal coke, and it is therefore assumed that 100 percent of the carbonate content of
the phosphate rock will be converted to CO2 in the elemental phosphorus production process. The calculation for
CO2 emissions assumes that phosphate rock consumption, for purposes other than phosphoric acid production,
results in CChemissions from 100 percent of the inorganic carbon content in phosphate rock, but none from the
organic carbon content.
The results of the Approach 2 quantitative uncertainty analysis are summarized in Table 4-81. 2022 phosphoric acid
production CO2 emissions were estimated to be between 0.7 and 1.1 MMT CO2 Eq. at the 95 percent confidence
level. This indicates a range of approximately 18 percent below and 20 percent above the emission estimate of 0.8
MMTCO2 Eq.
Table 4-81: Approach 2 Quantitative Uncertainty Estimates for CO2 Emissions from
Phosphoric Acid Production (MMT CO2 Eq. and Percent)
Gas
2022 Emission Estimate
(MMT CO . Eq.)
Uncertainty Range Relative to Emission Estimate-'
(MMTCO' Eq.) (%)
Source
Lower Upper
Bound Bound
Lower Upper
Bound Bound
Phosphoric Acid Production
C02
0.8
0.7 1.1
-18% +20%
3 Range of emission estimates predicted by Monte Carlo stochastic simulation for a 95 percent confidence interval.
Industrial Processes and Product Use 4-103
-------
QA/QC and Verification
For more information on the general QA/QC process applied to this source category, consistent with the U.S.
Inventory QA/QC plan, which is in accordance with Volume 1, Chapter 6 of 2006IPCC Guidelines as described in the
introduction of the IPPU chapter (see Annex 8 for more details).
Recalculations Discussion
Recalculations were performed for 2021 to reflect updated USGS data on the total U.S. production of phosphate
rock. This update resulted in a decrease of 35 kt CO2 in 2021.
Planned Improvements
EPA continues to evaluate potential improvements to the Inventory estimates for this source category, which
include direct integration of EPA's GHGRP data for 2010 through 2022 along with assessing applicability of
reported GHGRP data to update the inorganic carbon content of phosphate rock for prior years to ensure time-
series consistency. Specifically, EPA would need to assess that averaged inorganic carbon content data (by region
or other approaches) meets GHGRP confidential business information (CBI) screening criteria. EPA would then
need to assess the applicability of GHGRP data for the averaged inorganic carbon content (by region or other
approaches) from 2010 through 2022, along with other information to inform estimates in prior years in the time
series (1990 through 2009) based on the sources of phosphate rock used in production of phosphoric acid over
time. In implementing improvements and integration of data from EPA's GHGRP, EPA will rely upon the latest
guidance from the IPCC on the use of facility-level data in national inventories.76 These long-term planned
improvements are still in development by EPA and have not been implemented into the current Inventory report.
4.18 Iron and Steel Production (CRT Source
Category 2C1) and Metallurgical Coke
Production
Iron and steel production is a multi-step process that generates process-related emissions of carbon dioxide (CO2)
and methane (CH4) as raw materials are refined into iron and then transformed into crude steel. This reporting
category (2C1) includes emissions from the production of iron and steel. Per the IPCC methodological guidance,
emissions from conventional fuels (e.g., natural gas, fuel oil) consumed for energy purposes during the production
of iron and steel are accounted for as part of fossil fuel combustion in the industrial end-use sector reported under
the Energy chapter.
Iron and steel production includes seven distinct production processes: metallurgical coke production, sinter
production, direct reduced iron (DRI) production, pellet production, pig iron.77 production, electric arc furnace
76 See http://www.ipcc-nggiD.iges.or.jp/public/tb/TFl Technical Bulletin l.pdf arid the 2019 Refinement, Volume 1, Chapter 2,
Section 2.3, Use of Facility Data in Inventories at https://www.ipcc-
nggip.iges.or.jp/public/2019rf/pdf/l Volumel/19R VI Ch02 DataCollection.pdf.
77 Pig iron is the common industry term to describe what should technically be called crude iron. Pig iron is a subset of crude
iron that has lost popularity over time as industry trends have shifted. Throughout this report, pig iron will be used
4-104 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
(EAF) steel production, and basic oxygen furnace (BOF) steel production. The number of production processes at a
particular plant is dependent upon the specific plant configuration. Most process CO2 generated from the iron and
steel industry is a result of the production of crude iron.
In addition to the production processes mentioned above, CO2 is also generated at iron and steel mills through the
consumption of process byproducts (e.g., blast furnace gas, coke oven gas) used for various purposes including
heating, annealing, and electricity generation. Process byproducts sold off-site for use as synthetic natural gas are
also accounted for in these calculations. In general, CO2 emissions are generated in these production processes
through the reduction and consumption of various carbon-containing inputs (e.g., ore, scrap, flux, coke
byproducts). Fugitive CH4 emissions can also be generated from these processes, as well as from sinter, direct iron,
and pellet production.
In 2022, approximately eleven integrated iron and steel steelmaking facilities utilized BOFs to refine and produce
steel from iron, and raw steel was produced at 101 facilities across the United States. As of 2020, approximately 29
percent of steel production was attributed to BOFs and 71 percent to EAFs (AISI 2020). The trend in the United
States for integrated facilities has been a shift towards fewer BOFs and more EAFs. EAFs use scrap steel as their
main input and use significantly less energy than BOFs. There are also 14 cokemaking facilities, of which 3 facilities
are co-located with integrated iron and steel facilities (ACCCI 2021). In the United States, seven states account for
roughly 61 percent of total raw steel production: Indiana, Alabama, Tennessee, Kentucky, Mississippi, Arkansas,
and Ohio (AISI 2023).
Total annual production of crude steel in the United States was fairly constant between 2000 and 2008 and ranged
from a low of 99,320,000 tons to a high of 109,880,000 tons (2001 and 2004, respectively). Due to the decrease in
demand caused by the global economic downturn (particularly from the automotive industry), crude steel
production in the United States sharply decreased to 65,459,000 tons in 2009. Crude steel production was fairly
constant from 2011 through 2014, and after a dip in production from 2014 to 2015, crude steel production steadily
increased. Crude steel production dipped again in 2020 due to the COVID-19 pandemic and returned to pre-
pandemic levels in 2021. Production declined by approximately 6 percent in 2022 (AISI 2023). This decline may be
attributable to projections for decreased global end-use consumption due to multiple factors including the conflict
in Ukraine, continuing coronavirus disease 2019 (COVID-19) mitigation measures in China, rising energy costs and
interest rates, and global inflation (USGS 2023a). The United States was the fourth largest producer of raw steel in
the world, behind China, India, and Japan, accounting for approximately 4.3 percent of world production in 2022
(AISI 2004 through 2023).
The majority of CO2 emissions from the iron and steel production process come from the use of metallurgical coke
in the production of pig iron and from the consumption of other process byproducts, with lesser amounts emitted
from the use of carbon-containing flux and from the removal of carbon from pig iron used to produce steel.
According to the 2006IPCC Guidelines, the production of metallurgical coke from coking coal is considered to be an
energy use of fossil fuel, and the use of coke in iron and steel production is considered to be an industrial process
source. The 2006 IPCC Guidelines suggest that emissions from the production of metallurgical coke should be
reported separately in the Energy sector, while emissions from coke consumption in iron and steel production
should be reported in the Industrial Processes and Product Use sector. The approaches and emission estimates for
both metallurgical coke production and iron and steel production, however, are presented here because much of
the relevant activity data is used to estimate emissions from both metallurgical coke production and iron and steel
production. For example, some byproducts (e.g., coke oven gas) of the metallurgical coke production process are
consumed during iron and steel production, and some byproducts of the iron and steel production process (e.g.,
blast furnace gas) are consumed during metallurgical coke production. Emissions associated with the consumption
of these byproducts are attributed at the point of consumption. Emissions associated with the use of conventional
interchangeably with crude iron, but it should be noted that in other data sets or reports pig iron and crude iron may not be
used interchangeably and may provide different values.
Industrial Processes and Product Use 4-105
-------
fuels (e.g., natural gas, fuel oil) for electricity generation, heating and annealing, or other miscellaneous purposes
downstream of the iron and steelmaking furnaces are reported in the Energy chapter.
Metallurgical Coke Production
Emissions of CO2 from metallurgical coke production in 2022 were 3.0 MMT CO2 Eq. (2,954 kt CO2) (see Table 4-82
and Table 4-83). Emissions decreased by 8 percent from 2021 to 2022 and have decreased by 47 percent since
1990. Coke production in 2022 was about 9 percent lower than in 2021 and 59 percent below 1990 (EIA 2023, AISI
2023).
Significant activity data for 2020 through 2022 were not available in time for publication of this report due to
industry consolidation that impacts the publication of data without revealing confidential business information.
Activity data for these years were estimated using 2019 values adjusted based on GHGRP emissions data, as
described in the Methodology and Time-Series Consistency section below.
Table 4-82: CO2 Emissions from Metallurgical Coke Production (MMT CO2 Eq.)
Gas 1990 2005 2018 2019 2020 2021 2022
C02 5.6 3.91 1.3 3.0 2.3 3.2 3.0
Table 4-83: CO2 Emissions from Metallurgical Coke Production (kt CO2)
Gas 1990 2005 2018 2019 2020 2021 2022
C02 5,608 3,921 | 1,282 3,006 2,325 3,224 2,954~
Iron and Steel Production
Emissions of CO2 and CH4 from iron and steel production in 2022 were 37.7 MMT CO2 Eq. (37,718 kt) and 0.0077
MMT CO2 Eq. (0.3 kt CH4), respectively (see Table 4-84 through Table 4-87). Emissions from iron and steel
production decreased by 2 percent from 2021 to 2022 and have decreased by 62 percent since 1990, due to
restructuring of the industry, technological improvements, and increased scrap steel utilization. Carbon dioxide
emission estimates include emissions from the consumption of carbonaceous materials in the blast furnace, EAF,
and BOF, as well as blast furnace gas and coke oven gas consumption for other activities at the steel mill.
Significant activity data for 2020 through 2022 were not available in time for publication of this report due to
industry consolidation that impacts the publication of data without revealing confidential business information.
Activity data for these years were estimated using 2019 values adjusted based on GHGRP emissions data, as
described in the Methodology and Time-Series Consistency section below.
In 2022, domestic production of pig iron decreased by 11 percent from 2021 levels. Overall, domestic pig iron
production has declined since the 1990s; pig iron production in 2022 was 59 percent lower than in 2000 and 60
percent below 1990. Carbon dioxide emissions from iron production have decreased by 81 percent (37.0 MMT CO2
Eq.) since 1990. Carbon dioxide emissions from steel production have decreased by 16 percent (1.3 MMT CO2 Eq.)
since 1990, while overall CO2 emissions from iron and steel production have declined by 62 percent (61.4 MMT
CO2 Eq.) from 1990 to 2022.
Table 4-84: CO2 Emissions from Iron and Steel Production (MMT CO2 Eq.)
Source/Activity Data
1990 2005
2018
2019
2020
2021
2022
Sinter Production
2.4
I 1.7
0.9
0.9
0.7
0.8
0.8
Iron Production
45.71 17.7
I
1
9.6
9.4
8.4
9.0
8.7
Pellet Production
1.81 1.5
0.9
0.9
0.8
0.8
0.8
Steel Production
00
0
lllllll
to
4^
6.0
5.8
5.7
5.8
6.7
Other Activities3
41.2
I 35.9
24.1
23.2
19.8
22.1
20.8
Total
99.1
66.2
41.6
40.1
35.4
38.6
37.7
4-106 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
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 4-85: CO2 Emissions from Iron and Steel Production (kt CO2)
Source/Activity Data
1990
2005
2018
2019
2020
2021
2022
Sinter Production
2,448
1,663
937
876
749
836
787
Iron Production
45,709
17,666
9,589
9,365
8,420
9,038
8,673
Pellet Production
1,817
1,503
924
878
751
838
789
Steel Production
7,964
9,395
5,982
5,812
5,657
5,816
6,655
Other Activities3
41,194 1
35,934
24,149
23,158
19,820
22,119
20,814
Total
99,132
66,161
41,581
40,089
35,398
38,648
37,718
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 4-86: CH4 Emissions from Iron and Steel Production (MMT CO2 Eq.)
Source/Activity Data 1990 2005 2018 2019
2020
2021
2022
Sinter Production + + + +
+
+
+
+ Does not exceed 0.05 MMT C02 Eq.
Table 4-87: CH4 Emissions from Iron and Steel Production (kt CH4)
Source/Activity Data 1990 2005 2018 2019
2020
2021
2022
Sinter Production 0.9 0.6 + +
+
+
+
+ Does not exceed 0.5 kt.
Methodology and Time-Series Consistency
Emission estimates for metallurgical coke, EAF steel production, and BOF steel production presented in this
chapter utilize a country-specific approach based on Tier 2 methodologies provided by the 2006IPCC Guidelines, in
accordance with the IPCC methodological decision tree and available data. These Tier 2 methodologies call for a
mass balance accounting of the carbonaceous inputs and outputs during the iron and steel production process and
the metallurgical coke production process. Estimates for pig iron production apply Tier 2 methods consistent with
the 2006 IPCC Guidelines, in accordance with the IPCC methodological decision tree and available data. Tier 1
methods are used for certain iron and steel production processes (i.e., sinter production, pellet production and DRI
production) for which available data are insufficient to apply a Tier 2 method (e.g., country-specific carbon
contents of inputs and outputs are not known). The majority of emissions are captured with higher tier methods,
as sinter production, pellet production, and DRI production only account for roughly 8 percent of total iron and
steel production emissions.
The Tier 2 methodology equation is as follows:
Equation 4-11: CO2 Emissions from Coke, Pig Iron, EAF Steel, and BOF Steel Production, based
on 2006 IPCC Guidelines Tier 2 Methodologies
Eco2 ~
^(<2a X Ca) - ^(<2fc X Cb)
44
12
where,
Industrial Processes and Product Use 4-107
-------
Eco2 = Emissions from coke, pig iron, EAF steel, or BOF steel production, metric tons
a = Input material a
b = Output material b
Qa = Quantity of input material a, metric tons
Ca = Carbon content of input material a, metric tons C/metric ton material
Qb = Quantity of output material b, metric tons
Cb = Carbon content of output material b, metric tons C/metric ton material
44/12 = Stoichiometric ratio of CO2 to C
The Tier 1 methodology equations are as follows:
Equation 4-12:2006IPCC Guidelines Tier 1: Emissions from Sinter, Direct Reduced Iron, and
Pellet Production (Equations 4.6,4.7, and 4.8)
Es,p = Qs X EFSJ)
Ed,C02 = Qd X EFd,C02
Ep,co2 = Qp x EFp,co2
where,
Es,p
= Emissions from sinter production process for pollutant p (CO2 or CH4), metric ton
a
= Quantity of sinter produced, metric tons
EFs.p
= Emission factor for pollutant p (CO2 or CH4), metric ton p/metric ton sinter
Ed,C02
= Emissions from DRI production process for CO2, metric ton
Qd
= Quantity of DRI produced, metric tons
EFd,C02
= Emission factor for CO2, metric ton C02/metric ton DRI
Ep,C02
= Emissions from pellet production process for CO2, metric ton
Qp
= Quantity of pellets produced, metric tons
EFp,co2
= Emission factor for CO2, metric ton CQ2/metric ton pellets produced
A significant number of activity data that serve as inputs to emissions calculations were unavailable for 2020
through 2022 at the time of publication and were estimated using 2019 values. To estimate annual emissions for
these years, the EPA used process emissions data from the EPA's Greenhouse Gas Reporting Program (GHGRP)
subpart Q for the iron and steel sector to adjust the estimated values for 2020 through 2022. GHGRP process
emissions data decreased by approximately 14 percent from 2019 to 2020, increased by approximately 12 percent
from 2020 to 2021, and decreased by approximately 6 percent from 2021 to 2022 (EPA 2023). These percentage
changes were applied to 2019 activity data values to produce estimates for 2020 through 2022.
Metallurgical Coke Production
Coking coal is used to manufacture metallurgical coke which is used primarily as a reducing agent in the production
of iron and steel but is also used in the production of other metals including zinc and lead (see Zinc Production and
Lead Production sections of this chapter). Emissions associated with producing metallurgical coke from coking coal
are estimated and reported separately from emissions that result from the iron and steel production process. To
estimate emissions from metallurgical coke production, a Tier 2 method provided by the 2006 IPCC Guidelines was
4-108 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
utilized. The amount of carbon contained in materials produced during the metallurgical coke production process
(i.e., coke, coke breeze and coke oven gas) is deducted from the amount of carbon contained in materials
consumed during the metallurgical coke production process (i.e., natural gas, blast furnace gas, and coking coal).
For calculations, activity data for these inputs, including natural gas, blast furnace gas, and coking coke consumed
for metallurgical coke production, are in units consistent with the carbon content values. Light oil, which is
produced during the metallurgical coke production process, is excluded from the deductions due to data
limitations. The amount of carbon contained in these materials is calculated by multiplying the material-specific
carbon content by the amount of material consumed or produced (see Table 4-88). The amount of coal tar
produced was approximated using a production factor of 0.03 tons of coal tar per ton of coking coal consumed.
The amount of coke breeze produced was approximated using a production factor of 0.075 tons of coke breeze per
ton of coking coal consumed (Steiner 2008; DOE 2000). Data on the consumption of carbonaceous materials (other
than coking coal) as well as coke oven gas production were available for integrated steel mills only (i.e., steel mills
with co-located coke plants); therefore, carbonaceous material (other than coking coal) consumption and coke
oven gas production were excluded from emission estimates for merchant coke plants. Carbon contained in coke
oven gas used for coke-oven underfiring was not included in the deductions to avoid double-counting.
Table 4-88: Material Carbon Contents for Metallurgical Coke Production
Material
kg C/kg
Coal Tara
0.62
Cokea
0.83
Coke Breeze3
0.83
Coking Coalb
0.75
Material
kg C/GJ
Coke Oven Gasc
12.1
Blast Furnace Gasc
70.8
a Source: IPCC (2006), Vol. 3 Chapter 4, Table 4.3
b Source: EIA (2017b)
c Source: IPCC (2006), Vol. 2 Chapter 1, Table 1.3
Although the 2006 IPCC Guidelines provide a Tier 1 Cm emission factor for metallurgical coke production (i.e., 0.1 g
Cm per metric ton of coke production), it is not appropriate to use because CO2 emissions were estimated using
the Tier 2 mass balance methodology. The mass balance methodology makes a basic assumption that all carbon
that enters the metallurgical coke production process either exits the process as part of a carbon-containing
output or as CO2 emissions. This is consistent with a preliminary assessment of aggregated facility-level
greenhouse gas CH4 emissions reported by coke production facilities under EPA's GHGRP. The assessment indicates
that CH4 emissions from coke production are insignificant and below 500 kt or 0.05 percent of total national
emissions. Pending resources and significance, EPA continues to assess the possibility of including these emissions
in future Inventories to enhance completeness but has not incorporated these emissions into this report.
Data relating to the mass of coking coal consumed at metallurgical coke plants and the mass of metallurgical coke
produced at coke plants were taken from the Energy Information Administration (EIA) Quarterly Coal Report:
October through December (EIA 1998 through 2019) and EIA Quarterly Coal Report: January through March (EIA
2021 through 2023) (see Table 4-89). Data on the volume of natural gas consumption, blast furnace gas
consumption, and coke oven gas production for metallurgical coke production at integrated steel mills were
obtained from the American Iron and Steel Institute (AISI) Annual Statistical Report (AISI 2004 through 2023) and
through personal communications with AISI (Steiner 2008) (see Table 4-90). These data from the AISI Annual
Statistical Report were withheld for 2020 through 2022, so the 2019 values were used as estimated data for the
missing 2020 through 2022 values and adjusted using GHGRP emissions data, as described earlier in this
Methodology and Time-Series Consistency section.
The factor for the quantity of coal tar produced per ton of coking coal consumed was provided by AISI (Steiner
2008). The factor for the quantity of coke breeze produced per ton of coking coal consumed was obtained through
Table 2-1 of the report Energy and Environmental Profile of the U.S. Iron and Steel Industry (DOE 2000). Data on
Industrial Processes and Product Use 4-109
-------
natural gas consumption and coke oven gas production at merchant coke plants were not available and were
excluded from the emission estimate. Carbon contents for metallurgical coke, coal tar, coke oven gas, and blast
furnace gas were provided by the 2006IPCC Guidelines. The carbon content for coke breeze was assumed to equal
the carbon content of coke. Carbon contents for coking coal was from EIA.
Table 4-89: Production and Consumption Data for the Calculation of CO2 Emissions from
Metallurgical Coke Production (Thousand Metric Tons)
Source/Activity Data
1990
2005
2018
2019
2020
2021
2022
Metallurgical Coke Production
Coking Coal Consumption at Coke Plants
35,269 jjjjjj
21,259;;
16,635
16,261
13,076
15,957
14,523
Coke Production at Coke Plants
25,054
15,167
12,525
11,676
9,392
11,381
10,337
Coke Breeze Production
2,645 «
1,594
1,248
1,220
981
1,197
1,089
Coal Tar Production
1,058
638
499
488
392
479
436
Table 4-90: Production and Consumption Data for the Calculation of CO2 Emissions from
Metallurgical Coke Production (Million ft3)
Source/Activity Data
1990
2005
2018
2019
2020
2021
2022
Metallurgical Coke Production
114,213
Coke Oven Gas Production
250,767 |
80,750
77,692
66,492
74,206
69,829
Natural Gas Consumption
599
2,996
2,275
2,189
1,873
2,091
1,967
Blast Furnace Gas Consumption
24,602 I!
4,460 g
4,022
3,914
3,350
3,738
3,518
Iron and Steel Production
To estimate emissions from pig iron production in the blast furnace, the amount of carbon contained in the
produced pig iron and blast furnace gas were deducted from the amount of carbon contained in inputs (i.e.,
metallurgical coke, sinter, natural ore, pellets, natural gas, fuel oil, coke oven gas, carbonate fluxes or slagging
materials, and direct coal injection). For calculations, activity data for these inputs, including coke consumed for
pig iron production, are in units consistent with the carbon content values. The carbon contained in the pig iron,
blast furnace gas, and blast furnace inputs was estimated by multiplying the material-specific carbon content by
each material type (see Table 4-91). In the absence of a default carbon content value from the 2006 IPCC
Guidelines for pellet, sinter, or natural ore consumed for pig iron production, a country-specific approach based on
Tier 2 methodology is used. Pellet, sinter, and natural ore used as an input for pig iron production is assumed to
have the same carbon content as direct reduced iron (2 percent), based on expert judgment (RTI 2024). Carbon in
blast furnace gas used to pre-heat the blast furnace air is combusted to form CO2 during this process. Carbon
contained in blast furnace gas used as a blast furnace input was not included in the deductions to avoid double-
counting.
Emissions from steel production in EAFs were estimated by deducting the carbon contained in the steel produced
from the carbon contained in the EAF anode, charge carbon, and scrap steel added to the EAF. Small amounts of
carbon from DRI and pig iron to the EAFs were also included in the EAF calculation. For BOFs, estimates of carbon
contained in BOF steel were deducted from carbon contained in inputs such as natural gas, coke oven gas, fluxes
(i.e., limestone and dolomite), and pig iron. In each case, the carbon was calculated by multiplying material-specific
carbon contents by each material type (see Table 4-91). For EAFs, the amount of EAF anode consumed was
approximated by multiplying total EAF steel production by the amount of EAF anode consumed per metric ton of
steel produced (0.002 metric tons EAF anode per metric ton steel produced [Steiner 2008]). The amount of carbon-
containing flux (i.e., limestone and dolomite) used in EAF and BOF steel production was deducted from the "Other
Process Uses of Carbonates" source category (CRT Source Category 2A4) to avoid double-counting.
Carbon dioxide emissions from the consumption of blast furnace gas and coke oven gas for other activities
occurring at the steel mill were estimated by multiplying the amount of these materials consumed for these
purposes by the material-specific carbon content (see Table 4-91).
4-110 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
Table 4-91: Material Carbon Contents for Iron and Steel Production
Material
kg C/kg
Coke
0.83
Direct Reduced Iron
0.02
Dolomite
0.13
EAF Carbon Electrodes
0.82
EAF Charge Carbon
0.83
Limestone
0.12
Pig Iron
0.04
Steel
0.01
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.
Carbon dioxide emissions associated with sinter production, direct reduced iron production, pellet production, pig
iron production, steel production, and other steel mill activities were summed to calculate the total CO2 emissions
from iron and steel production (see Table 4-84 and Table 4-85).
The sinter production process results in fugitive emissions of CH4, which are emitted via leaks in the production
equipment, rather than through the emission stacks or vents of the production plants. The fugitive emissions were
calculated by applying Tier 1 emission factors taken from the 2006 IPCC Guidelines for sinter production (see Table
4-92). Although the 2006 IPCC Guidelines also provide a Tier 1 methodology for CH4 emissions from pig iron
production, it is not appropriate to use because CO2 emissions for pig iron production are estimated using the Tier
2 mass balance methodology. The mass balance methodology makes a basic assumption that all carbon that enters
the pig iron production process either exits the process as part of a carbon-containing output or as CO2 emissions;
the estimation of CH4 emissions is precluded. Annual analysis of facility-level emissions reported during iron
production further supports this assumption and indicates that CH4 emissions are below 500 kt CO2 Eq. and well
below 0.05 percent of total national emissions. The production of direct reduced iron could also result in emissions
of Cm through the consumption of fossil fuels (e.g., natural gas, etc.); however, these emission estimates are
excluded due to data limitations. Pending further analysis and resources, EPA may include these emissions in
future reports to enhance completeness. EPA is still assessing the possibility of including these emissions in future
reports and have not included this data in the current report.
Table 4-92: CH4 Emission Factors for Sinter and Pig Iron Production
Material Produced
Factor
Unit
Sinter
0.07
kg CH4/metric ton
Source: IPCC (2006), Table 4.2.
Emissions of CChfrom sinter production, direct reduced iron production, and pellet production were estimated by
multiplying total national sinter production, total national direct reduced iron production, and total national pellet
production by Tier 1 CO2 emission factors (see Table 4-93). Because estimates of sinter production, direct reduced
iron production, and pellet production were not available, production was assumed to equal consumption.
Industrial Processes and Product Use 4-111
-------
Table 4-93: CO2 Emission Factors for Sinter Production, Direct Reduced Iron Production, and
Pellet Production
Material Produced
Metric Ton CO '/Metric Ton
Sinter
0.2
Direct Reduced Iron
0.7
Pellet Production
0.03
Source: IPCC (2006), Table 4.1.
The consumption of coking coal, natural gas, distillate fuel, and coal used in iron and steel production are adjusted
for within the Energy chapter to avoid double-counting of emissions reported within the IPPU chapter as these
fuels were consumed during non-energy related activities. More information on this methodology and examples of
adjustments made between the IPPU and Energy chapters are described in Annex 2.1, Methodology for Estimating
Emissions of CO2 from Fossil Fuel Combustion.
Sinter consumption and pellet consumption data for 1990 through 2020 were obtained from AISI's Annual
Statistical Report (AISI 2004 through 2022) and through personal communications with AISI (Steiner 2008) (see
Table 4-94). These data from the AISI Annual Statistical Report were withheld for 2020 through 2022, so the 2019
values were used as estimated data for the missing 2020 through 2022 values and adjusted using GHGRP emissions
data, as described earlier in this Methodology and Time-Series Consistency section.
In general, direct reduced iron (DRI) consumption data were obtained from the U.S. Geological Survey (USGS)
Minerals Yearbook - Iron and Steel Scrap (USGS 1991 through 2022; USGS 2023b) and personal communication
with the USGS Iron and Steel Commodity Specialist (Tuck 2023a). Data for DRI consumed in EAFs were not
available for the years 1990 and 1991. EAF DRI consumption in 1990 and 1991 was calculated by multiplying the
total DRI consumption for all furnaces by the EAF share of total DRI consumption in 1992. Data for DRI consumed
in BOFs were not available for the years 1990 through 1993. BOF DRI consumption in 1990 through 1993 was
calculated by multiplying the total DRI consumption for all furnaces (excluding EAFs and cupola) by the BOF share
of total DRI consumption (excluding EAFs and cupola) in 1994.
The Tier 1 CO2 emission factors for sinter production, direct reduced iron production and pellet production were
obtained through the 2006 IPCC Guidelines (IPCC 2006). Time-series data for pig iron production, coke, natural gas,
fuel oil, sinter, and pellets consumed in the blast furnace; pig iron production; and blast furnace gas produced at
the iron and steel mill and used in the metallurgical coke ovens and other steel mill activities were obtained from
AISI's Annual Statistical Report (AISI 2004 through 2021) and through personal communications with AISI (Steiner
2008) (see Table 4-94 and Table 4-95). Data including blast furnace gas, coke oven gas, natural gas, limestone,
sinter, and natural ore consumption for blast furnaces, coke production, and steelmaking furnaces (EAFs and BOFs)
from the AISI Annual Statistical Report were withheld for 2020 through 2022, so the 2019 values were used as
estimated data for the missing 2020 through 2022 values and adjusted using GHGRP emissions data, as described
earlier in this Methodology and Time-Series Consistency section. Similarly, the percent of total steel production for
EAF and BOF steelmaking processes were withheld for 2021 and 2022, so the 2020 values were used as estimated
data for the missing 2021 values and adjusted using GHGRP emissions data, as described earlier in this
Methodology and Time-Series Consistency section.
Data for EAF steel production, carbon-containing flux, EAF charge carbon, and natural gas consumption were
obtained from AISI's Annual Statistical Report (AISI 2004 through 2022) and through personal communications
with AISI (AISI 2006 through 2016, Steiner 2008). The factor for the quantity of EAF anode consumed per ton of
EAF steel produced was provided by AISI (Steiner 2008). Data for BOF steel production, carbon-containing flux,
natural gas, natural ore, pellet, sinter consumption as well as BOF steel production were obtained from AISI's
Annual Statistical Report (AISI 2004 through 2023) and through personal communications with AISI (Steiner 2008).
Data for EAF consumption of natural gas and BOF consumption of coke oven gas, limestone, and natural ore from
the AISI Annual Statistical Report were not available for 2021 and 2022, so 2020 values were used as estimated
data for the missing 2021 and 2022 values and adjusted using GHGRP emissions data, as described earlier in this
4-112 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
Methodology and Time-Series Consistency section. Data for EAF and BOF scrap steel, pig iron, and DRI
consumption were obtained from the USGS Minerals Yearbook - Iron and Steel Scrap (USGS 1991 through 2022;
USGS 2023b) and personal communication with the USGS Iron and Steel Commodity Specialist (Tuck 2023a). Data
on coke oven gas and blast furnace gas consumed at the iron and steel mill (other than in the EAF, BOF, or blast
furnace) were obtained from AISI's Annual Statistical Report (AISI 2004 through 2021) and through personal
communications with AISI (Steiner 2008). These data were not available for 2021 and 2022, so 2020 values were
used as estimated data for the missing 2021 and 2022 values and adjusted using GHGRP emissions data, as
described earlier in this Methodology and Time-Series Consistency section. Some data from the AISI Annual
Statistical Report on natural gas consumption were withheld for 2020 through 2022, so the 2019 values were used
as estimated data for the missing 2020 through 2022 values and adjusted using GHGRP emissions data, as
described earlier in this Methodology and Time-Series Consistency section.
Data on blast furnace gas and coke oven gas sold for use as synthetic natural gas were obtained from ElA's Natural
Gas Annual 2019 (EIA 2020). Carbon contents for direct reduced iron, EAF carbon electrodes, EAF charge carbon,
limestone, dolomite, pig iron, and steel were provided by the 2006IPCC Guidelines. The carbon contents for
natural gas, fuel oil, and direct injection coal were obtained from EIA (EIA 2017b) and EPA (EPA 2010). Heat
contents for fuel oil and direct injection coal were obtained from EIA (EIA 1992, 2011); natural gas heat content
was obtained from Table 37 of AISI's Annual Statistical Report (AISI 2004 through 2021). Heat contents for coke
oven gas and blast furnace gas were provided in Table 37 of AISI's Annual Statistical Report (AISI 2004 through
2021) and confirmed by AISI staff (Carroll 2016).
Table 4-94: Production and Consumption Data for the Calculation of CO2 and CH4 Emissions
from Iron and Steel Production (Thousand Metric Tons)
Source/Activity Data
1990
2005
2018
2019
2020
2021
2022
Sinter Production
12,239
8,315
4,687
4,378
3,747
4,182
3,935
Direct Reduced Iron Production
517 1
1,303 =
C
C
C
C
C
Pellet Production
60,563
50,096
30,793
29,262
25,044
27,949
26,300
Pig Iron Production
I
1
I
1
Coke Consumption
24,946
13,832
7,618
7,291
6,240
6,964
6,553
Pig Iron Production
49,669 I
37,222
24,058
22,302
18,320
22,246
19,791
Direct Injection Coal
I
I
Consumption
1,485
2,573
2,569
2,465
2,110
2,354
2,216
EAF Steel Production
¦
1
¦
1
EAF Anode and Charge Carbon
Consumption
67
1,127
1,133
1,137
1,118
1,129
1,123
Scrap Steel Consumption
42,691
46,600
C
C
C
C
C
Flux Consumption
319
695
998
998
998
998
998
EAF Steel Production
33,511
52,194
58,904
61,172
51,349
57,307
53,926
BOF Steel Production
Pig Iron Consumption
47,307
34,400
C
C
C
C
C
Scrap Steel Consumption
14,713
11,400
C
C
C
C
C
Flux Consumption
576 |
582
408
363
311
347
326
BOF Steel Production
43,973
42,705
27,704
26,591
21,384
23,865
22,457
C (Confidential)
Table 4-95: Production and Consumption Data for the Calculation of CO2 Emissions from Iron
and Steel Production (Million ft3 unless otherwise specified)
Source/Activity Data 1990 2005 1 2018 2019 2020 2021 2022
Pig Iron Production
Natural Gas Consumption 56,273 | 59,844 1 40,204 37,934 32,465 36,232 34,095
Fuel Oil Consumption
(thousand gallons) 163,397 16,170 3,365 2,321 1,986 2,217 2,086
Industrial Processes and Product Use 4-113
-------
Coke Oven Gas Consumption
22,033 :
16,557
1
I
13,337
12,926
11,063
12,346
11,618
Blast Furnace Gas Production
1,439,380
1,299,980
871,860
836,033
715,509
798,522
751,418
EAF Steel Production
I
1
1
Natural Gas Consumption
15,905
19,985
8,556
9,115
7,801
8,706
8,192
BOF Steel Production
!!!!!!:
I
Coke Oven Gas Consumption
3,851
524
¦
405
389
333
372
350
Other Activities
mm;
I
Coke Oven Gas Consumption
224,883
97,132
67,008
64,377
55,096
61,489
57,861
Blast Furnace Gas Consumption
1,414,778 =
1,295,520
867,838
832,119
712,159
794,783
747,900
Methodological approaches were applied to the entire time series to ensure consistency in emissions from 1990
through 2022.
Uncertainty
The estimates of CO2 emissions from metallurgical coke production are based on assessing uncertainties in
material production and consumption data and average carbon contents. Uncertainty is associated with the total
U.S. coking coal consumption, total U.S. coke production, and materials consumed during this process. Data for
coking coal consumption and metallurgical coke production are from different data sources (EIA) than data for
other carbonaceous materials consumed at coke plants (AISI), which does not include data for merchant coke
plants. There is uncertainty associated with the fact that coal tar and coke breeze production were estimated
based on coke production because coal tar and coke breeze production data were not available. Since merchant
coke plant data is not included in the estimate of other carbonaceous materials consumed at coke plants, the mass
balance equation for CO2 from metallurgical coke production cannot be reasonably completed; therefore, for the
purpose of this analysis, uncertainty parameters are applied to primary data inputs to the calculation (i.e., coking
coal consumption and metallurgical coke production) only.
The estimates of CO2 emissions from iron and steel production are based on material production and consumption
data and average carbon contents. There is uncertainty associated with the assumption that pellet production,
direct reduced iron and sinter consumption are equal to production. There is uncertainty with the
representativeness of the associated IPCC default emission factors. There is uncertainty associated with the
assumption that all coal used for purposes other than coking coal is for direct injection coal. There is also
uncertainty associated with the carbon contents for pellets, sinter, and natural ore, which are assumed to equal
the carbon contents of direct reduced iron, when consumed in the blast furnace. There is uncertainty associated
with the consumption of natural ore under current industry practices. For EAF steel production, there is
uncertainty associated with the amount of EAF anode and charge carbon consumed due to inconsistent data
throughout the time series. Also for EAF steel production, there is uncertainty associated with the assumption that
100 percent of the natural gas attributed to "steelmaking furnaces" by AISI is process-related and nothing is
combusted for energy purposes. Uncertainty is also associated with the use of process gases such as blast furnace
gas and coke oven gas. Data are not available to differentiate between the use of these gases for processes at the
steel mill versus for energy generation (i.e., electricity and steam generation); therefore, all consumption is
attributed to iron and steel production. These data and carbon contents produce a relatively accurate estimate of
CO2 emissions; however, there are uncertainties associated with each.
For calculating the emissions estimates from iron and steel and metallurgical coke production, EPA utilizes a
number of data points taken from the AISI Annual Statistical Report (ASR). This report serves as a benchmark for
information on steel companies in United States, regardless if they are a member of AISI, which represents
integrated producers (i.e., blast furnace and EAF). During the compilation of the 1990 through 2016 Inventory
report EPA initiated conversation with AISI to better understand and update the qualitative and quantitative
uncertainty metrics associated with AISI data elements. AISI estimates their data collection response rate to range
from 75 to 90 percent, with certain sectors of the iron and steel industry not being covered by the ASR; therefore,
there is some inherent uncertainty in the values provided in the AISI ASR, including material production and
consumption data. There is also some uncertainty to which materials produced are exported to Canada. As
4-114 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
indicated in the introduction to this section, the trend for integrated facilities has moved to more use of EAFs and
fewer BOFs. This trend may not be completely captured in the current data which also increases uncertainty. EPA
assigned an uncertainty range of ±10 percent for the primary data inputs (i.e., consumption and production values
for each production process, heat and carbon content values), a normal probability density function for
consumption and production values for each production process, and a triangular probability density function for
heat and carbon content values to calculate overall uncertainty from iron and steel production, and using this
suggested uncertainty provided in Table 4.4 of the 2006IPCC Guidelines is appropriate based on expert judgment
(RTI 2023). During EPA's discussion with AISI, AISI noted that an uncertainty range of ±5 percent would be a more
appropriate approximation to reflect their coverage of integrated steel producers in the United States. EPA will
continue to assess the best range of uncertainty for these values. EPA assigned an uncertainty range of ±25
percent and a triangular probability density function for the Tier 1CO2 emission factors for the sinter, direct
reduced iron, and pellet production processes, and using this suggested uncertainty provided in Table 4.4 of the
2006 IPCC Guidelines is appropriate based on expert judgment (RTI 2023).
The results of the Approach 2 quantitative uncertainty analysis are summarized in Table 4-96 for metallurgical coke
production and iron and steel production. Total CO2 emissions from metallurgical coke production and iron and
steel production for 2022 were estimated to be between 34.3 and 47.1 MMT CO2 Eq. at the 95 percent confidence
level. This indicates a range of approximately 16 percent below and 16 percent above the emission estimate of
40.7 MMT CO2 Eq. Total CH4 emissions from metallurgical coke production and iron and steel production for 2022
were estimated to be between 0.007 and 0.008 MMT CO2 Eq. at the 95 percent confidence level. This indicates a
range of approximately 7 percent below and 7 percent above the emission estimate of 0.0077 MMT CO2 Eq.
Table 4-96: Approach 2 Quantitative Uncertainty Estimates for CO2 and CH4 Emissions from
Iron and Steel Production and Metallurgical Coke Production (MMT CO2 Eq. and Percent)
Source
Gas
2022 Emission Estimate
(MMTCO. Eq.)
Uncertainty Range Relative to Emission Estimate-'
(MMTCO' Eq.) (%)
Lower
Upper
Lower
Upper
Bound
Bound
Bound
Bound
Metallurgical Coke & Iron
and Steel Production
C02
40.7
34.3
47.1
-16%
+16%
Metallurgical Coke & Iron
and Steel Production
ch4
+
+
+
-7%
+7%
+ Does not exceed 0.05 MMT C02 Eq.
a Range of emission estimates predicted by Monte Carlo stochastic simulation for a 95 percent confidence interval.
QA/QC and Verification
General quality assurance/quality control (QA/QC) procedures were applied consistent with the U.S. Inventory
QA/QC plan, which is in accordance with Volume 1, Chapter 6 of 2006 IPCC Guidelines as described in the
introduction of the IPPU chapter (see Annex 8 for more details). As part of a multiyear improvement effort, EPA is
reviewing the iron and steel methodology and available data, conducting additional category specific QC checks
and will report on findings when that review is complete (i.e., projected to be complete at earliest for the 2025
report). More information is provided under Planned Improvements below.
Recalculations Discussion
Recalculations were performed for the year 2021 with updated USGS values for DRI, pig iron, and scrap steel
consumption for both BOF and EAF steel production. Additionally, revisions to GHGRP data for 2020 and 2021
resulted in minor changes to activity data that were adjusted using GHGRP data, as described in the Methodology
and Time-Series Consistency section. Compared to the previous Inventory, CO2 emissions from steel production
increased by less than 1 percent (7 kt CO2) in 2020 and by less than 1 percent (211 kt CO2) in 2021.
Industrial Processes and Product Use 4-115
-------
Planned Improvements
Significant activity data for 2020 through 2022 were not available for this report and were estimated using 2019
values and adjusted using GHGRP emissions data. EPA will continue to explore sources of 2020 through 2022 data
and other estimation approaches. EPA will evaluate and analyze data reported under EPA's GHGRP to improve the
emission estimates for Iron and Steel Production process categories. Particular attention will be made to ensure
time-series consistency of the emissions estimates presented in future Inventory reports, consistent with IPCC and
UNFCCC guidelines. This is required as the facility-level reporting data from EPA's GHGRP, with the program's initial
requirements for reporting of emissions in calendar year 2010, are not available for all inventory years (i.e., 1990
through 2009) as required for this Inventory. In implementing improvements and integration of data from EPA's
GHGRP, EPA will rely on the latest guidance from the IPCC on the use of facility-level data in national inventories.78
This is a near to medium-term improvement, and per preliminary work, EPA estimates that the earliest this
improvement could be incorporated is the next (i.e., 2025) Inventory submission.
Additional improvements include accounting for emission estimates for the production of metallurgical coke in the
Energy chapter as well as identifying the amount of carbonaceous materials, other than coking coal, consumed at
merchant coke plants. Other potential improvements include identifying the amount of coal used for direct
injection and the amount of coke breeze, coal tar, and light oil produced during coke production. Efforts will also
be made to identify information to better characterize emissions from the use of process gases and fuels within
the Energy and IPPU chapters. Additional efforts will be made to improve the reporting and transparency in
accounting for fuels between the IPPU and Energy chapters, particularly the inclusion of a quantitative summary of
the carbon balance in the United States. This planned improvement is a long-term improvement and is still in
development. It is not included in this current Inventory report. EPA estimates that the earliest this improvement
could be incorporated is the next (i.e., 2025) Inventory submission.
4.19 Ferroalloy Production (CRT Source
Category 2C2)
Ferroalloys are composites of iron (Fe) and other elements such as silicon (Si), manganese (Mn), and chromium
(Cr). This reporting category (2C2) includes emissions of carbon dioxide (CO2) and methane (CH4) from the
production of several ferroalloys. Per the IPCC methodological guidance, emissions from fuels consumed for
energy purposes during the production of ferroalloys are accounted for as part of fossil fuel combustion in the
industrial end-use sector reported under the Energy chapter. Emissions from the production of two types of
ferrosilicon (25 to 55 percent and 56 to 95 percent silicon), silicon metal (96 to 99 percent silicon), and
miscellaneous alloys (32 to 65 percent silicon) have been calculated.
Emissions from the production of ferrochromium and ferromanganese are not included because of the small
number of manufacturers of these materials in the United States. Government information disclosure rules
prevent the publication of production data for these production facilities. Additionally, production of
ferrochromium in the United States ceased in 2009 (USGS 2013a).
Similar to emissions from the production of iron and steel, CO2 is emitted when metallurgical coke is oxidized
during a high-temperature reaction with iron and the selected alloying element. Due to the strong reducing
78 See http://www.ipcc-nggiD.iges.or.jp/public/tb/TFl Technical Bulletin l.pdf and the 2019 Refinement, Volume 1, Chapter 2,
Section 2.3, Use of Facility Data in Inventories at https://www.ipcc-
nggip.iges.or.jp/public/2019rf/pdf/l Volumel/19R VI Ch02 DataCollection.pdf.
4-116 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
environment, CO is initially produced and eventually oxidized to CO2. A representative reaction equation for the
production of 50 percent ferrosilicon (FeSi) is given below:
While most of the carbon contained in the process materials is released to the atmosphere as CO2, a percentage is
also released as CFU and other volatiles. The amount of CH4 that is released is dependent on furnace efficiency,
operation technique, and control technology.
Ferroalloys are used to alter the material properties of the steel. Ferroalloys are produced in conjunction with the
iron and steel industry, often at co-located facilities, and production trends closely follow that of the iron and steel
industry. As of 2020,11 facilities in the United States produce ferroalloys (USGS 2022b).
Emissions of CO2 from ferroalloy production in 2022 were 1.3 MMT CO2 Eq. (1,327 kt CO2) (see Table 4-97 and
Table 4-98), which is a 15 percent reduction since 2021 and a 38 percent reduction since 1990. Emissions of CFU
from ferroalloy production in 2022 were 0.01 MMT CO2 Eq. (0.4 kt CH4), which is a 15 percent decrease since 2021
and a 45 percent decrease since 1990. Variability in emissions over the past five years is attributable to facility
shutdowns in 2018 and 2020 (USGUS 2020; USGS 2021). The latter facility reopened its ferrosilicon production
facility in 2021, owing to increased demand for ferrosilicon products and improved domestic pricing (USGS 2022c).
Table 4-97: CO2 and CH4 Emissions from Ferroalloy Production (MMT CO2 Eq.)
Gas
1990
2005
2018
2019
2020
2021
2022
C02
1.4 .
2.1
1.6
1.4
1.6
1.3
ch4
+
+
+
+
+
+
+
Total
2.2
1.4
2.1
1.6
1.4
1.6
1.3
+ Does not exceed 0.05 MMT C02 Eq.
Note: Totals may not sum due to independent rounding.
Table 4-98: CO2 and CH4 Emissions from Ferroalloy Production (kt)
Gas
1990
2005
2018
2019
2020
2021
2022
C02
2,152
1,392
2,063
1,598
1,377
1,567
1,327
ch4
1 5*
+ :
1
+
+
+
+
+ Does not exceed 0.5 kt
Emissions of CO2 and CH4 from ferroalloy production are calculated79 using a Tier 1 method from the 2006IPCC
Guidelines, in accordance with the IPCC methodological decision tree and available data. Annual ferroalloy
production is multiplied by material-specific emission factors provided by IPCC (IPCC 2006). The Tier 1 equations
for CO2 and CH4 emissions are as follows:
Equation 4-13:2006 IPCC Guidelines Tier 1: CO2 Emissions for Ferroalloy Production (Equation
4.15)
79 EPA has not integrated aggregated facility-level GHGRP information to inform these estimates. The aggregated information
(e.g., activity data and emissions) associated with production of ferroalloys did not meet criteria to shield underlying
confidential business information (CBI) from public disclosure.
Fe203 + 2Si02 + 7C —> 2FeSi + 7C0
Methodology and Time-Series Consistency
Industrial Processes and Product Use 4-117
-------
where,
Eco2 = CO2 emissions, metric tons
MP, = Production of ferroalloy type /', metric tons
EFi = Generic emission factor for ferroalloy type /', metric tons CCh/metric ton specific ferroalloy
product
Equation 4-14:2006IPCC Guidelines Tier 1: CH4 Emissions for Ferroalloy Production (Equation
4.18)
Ech, = Y^MPi X EFi)
i
where,
Ech4 = Cm emissions, kg
MP, = Production of ferroalloy type /', metric tons
EFi = Generic emission factor for ferroalloy type /', kg Cm/metric ton specific ferroalloy product
Default emission factors were used because country-specific emission factors are not currently available. The
following emission factors were used to develop annual CO2 and CH4 estimates:
• Ferrosilicon, 25 to 55 percent Si and Miscellaneous Alloys, 32 to 65 percent Si: 2.5 metric tons CCh/metric
ton of alloy produced, 1.0 kg Cm/metric ton of alloy produced.
• Ferrosilicon, 56 to 95 percent Si: 4.0 metric tons CCh/metric ton alloy produced, 1.0 kg Cm/metric ton of
alloy produced.
• Silicon Metal: 5.0 metric tons CCh/metric ton metal produced, 1.2 kg Cm/metric ton metal produced.
It was assumed that 100 percent of the ferroalloy production was produced using petroleum coke in an electric arc
furnace process (IPCC 2006), although some ferroalloys may have been produced with coking coal, wood, other
biomass, or graphite carbon inputs. The amount of petroleum coke consumed in ferroalloy production was
calculated assuming that the petroleum coke used is 90 percent carbon (C) and 10 percent inert material (Onder
and Bagdoyan 1993).
The use of petroleum coke for ferroalloy production is adjusted for within the Energy chapter as this fuel was
consumed during non-energy related activities. Additional information on the adjustments made within the Energy
sector for non-energy use of fuels is described in both the Methodology section of CO2 from Fossil Fuel
Combustion (3.1 Fossil Fuel Combustion [CRT Source Category 1A]) and Annex 2.1, Methodology for Estimating
Emissions of CO2 from Fossil Fuel Combustion.
Ferroalloy production data for 1990 through 2022 (see Table 4-99) were obtained from the U.S. Geological Survey
(USGS) through the Minerals Yearbook: Silicon (USGS 1996 through 2022) and the Minerals Industry Survey: Silicon
(USGS 2023a). The following data were available from the USGS publications for the time series:
• Ferrosilicon, 25 to 55 percent Si: Annual production data were available from 1990 through 2010.
• Ferrosilicon, 56 to 95 percent Si: Annual production data were available from 1990 through 2010.
• Silicon Metal: Annual production data were available from 1990 through 2005. Production data for 2005
were used as estimates for 2006 through 2010 because data for these years were not available due to
government information disclosure rules.
• Miscellaneous Alloys, 32 to 65 percent Si: Annual production data were available from 1990 through
1998. Starting 1999, USGS reported miscellaneous alloys and ferrosilicon containing 25 to 55 percent
silicon as a single category.
4-118 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
Starting with the 2011 publication, USGS ceased publication of production quantity by ferroalloy product and
began reporting all the ferroalloy production data as a single category (i.e., Total Silicon Materials Production). This
is due to the small number of ferroalloy manufacturers in the United States and government information
disclosure rules. Ferroalloy product shares developed from the 2010 production data (i.e., ferroalloy product
production divided by total ferroalloy production) were used with the total silicon materials production quantity to
estimate the production quantity by ferroalloy product type for 2011 through 2022 (USGS 2017 through 2022).
Table 4-99: Production of Ferroalloys (Metric Tons)
Year
1990
2005
2018
2019
2020
2021
2022
Ferrosilicon 25%-55%
321,385
123,000
189,846
147,034
126,681
144,227
122,119
Ferrosilicon 56%-95%
109,566 ::
00
cn
8
¦llll!
167,511
129,736
111,778
127,259
107,752
Silicon Metal
145,744
148,000
183,642
142,229
122,541
139,514
118,128
Misc. Alloys 32-65%
72,442 5
NA 1
NA
NA
NA
NA
NA
NA (Not Available) for product type, aggregated with ferrosilicon (25-55% Si)
Methodological approaches were applied to the entire time series to ensure consistency in emissions from 1990
through 2022.
Uncertainty
Annual ferroalloy production was reported by the USGS in three broad categories until the 2010 publication:
ferroalloys containing 25 to 55 percent silicon (including miscellaneous alloys), ferroalloys containing 56 to 95
percent silicon, and silicon metal (through 2005 only, 2005 value used as an estimate for 2006 through 2010).
Starting with the 2011 Minerals Yearbook: Silicon, USGS started reporting all the ferroalloy production under a
single category: total silicon materials production. The total silicon materials quantity was allocated across the
three categories, based on the 2010 production shares for the three categories. Refer to the Methodology section
for further details. Additionally, production data for silvery pig iron (alloys containing less than 25 percent silicon)
are not reported by the USGS to avoid disclosing proprietary company data. Emissions from this production
category, therefore, were not estimated.
Some ferroalloys may be produced using wood or other biomass as a primary or secondary carbon source
(carbonaceous reductants); however, information and data regarding these practices were not available. Emissions
from ferroalloys produced with wood or other biomass would not be counted under this source because wood-
based carbon is of biogenic origin.80 Even though emissions from ferroalloys produced with coking coal or graphite
inputs would be counted in national trends, they may be generated with varying amounts of CO2 per unit of
ferroalloy produced. The most accurate method for these estimates would be to base calculations on the amount
of reducing agent used in the process, rather than the amount of ferroalloys produced. These data, however, were
not available, and are also often considered confidential business information.
Emissions of CH4 from ferroalloy production will vary depending on furnace specifics, such as type, operation
technique, and control technology. Higher heating temperatures and techniques such as sprinkle charging would
reduce Cm emissions; however, specific furnace information was not available or included in the CH4 emission
estimates.
EPA assigned a uncertainty range of ±25 percent for the primary emission factors (i.e., ferrosilicon 25-55% Si,
ferrosilicon 56-95% Si, and silicon metal), and an uncertainty range of ±5 percent for the 2010 production values
for ferrosilicon 25-55% Si, ferrosilicon 56-95% Si, and silicon metal production and the 2021 total silicon materials
production value used to calculate emissions from overall ferroalloy production. Using these suggested
uncertainties provided in in Table 4.9 of Section 4.3.3.2 of the 2006IPCC Guidelines is appropriate based on expert
80 Emissions and sinks of biogenic carbon are accounted for in the Land Use, Land-Use Change, and Forestry chapter.
Industrial Processes and Product Use 4-119
-------
judgment (RTI 2023). Per this expert judgment, a normal probability density function was assumed for all activity
data, and a triangular probability density function was assumed for emission factors.
The results of the Approach 2 quantitative uncertainty analysis are summarized in Table 4-100. Ferroalloy
production CO2 emissions from 2022 were estimated to be between 1.2 and 1.5 MMT CO2 Eq. at the 95 percent
confidence level. This indicates a range of approximately 13 percent below and 13 percent above the emission
estimate of 1.3 MMT CO2 Eq. Ferroalloy production CH4 emissions were estimated to be between a range of
approximately 12 percent below and 13 percent above the emission estimate of 0.01 MMT CO2 Eq.
Table 4-100: Approach 2 Quantitative Uncertainty Estimates for CO2 Emissions from
Ferroalloy Production (MMT CO2 Eq. and Percent)
Source
Gas
2022 Emission Estimate
(MMT CO . Eq.)
Uncertainty Range Relative to Emission Estimate-1
(MMTCO' Eq.) (%)
Lower
Upper
Lower
Upper
Bound
Bound
Bound
Bound
Ferroalloy Production
C02
1.3
1.2
1.5
-13%
+13%
Ferroalloy Production
ch4
+
+
+
-12%
+13%
+ Does not exceed 0.05 MMT C02 Eq.
a Range of emission estimates predicted by Monte Carlo stochastic simulation for a 95 percent confidence interval.
QA/QC and Verification
General quality assurance/quality control (QA/QC) procedures were applied consistent with the U.S. Inventory
QA/QC plan, which is in accordance with Volume 1, Chapter 6 of 2006IPCC Guidelines as described in the
introduction of the IPPU chapter (see Annex 8 for more details).
Recalculations Discussion
No recalculations were performed for the 1990 to 2021 portion of the time series.
Planned Improvements
Pending available resources and prioritization of improvements for more significant sources, EPA will continue to
evaluate and analyze data reported under EPA's GHGRP that would be useful to improve the emission estimates
and category-specific QC procedures for the Ferroalloy Production source category. Given the small number of
facilities and reporting thresholds, particular attention will be made to ensure completeness and time-series
consistency of the emissions estimates presented in future Inventory reports, consistent with IPCC and UNFCCC
guidelines. This is required as the facility-level reporting data from EPA's GHGRP, with the program's initial
requirements for reporting of emissions in calendar year 2010, are not available for all inventory years (i.e., 1990
through 2009) as required for this Inventory. In 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.81 This is a long-term planned improvement, and EPA is still assessing the possibility of incorporating this
improvement into the Inventory. This improvement has not been included in the current Inventory report.
81 See http://www.ipcc-nggip.iges.or.jp/public/tb/TFI Technical Bulletin l.pdf and the 2019 Refinement, Volume 1, Chapter 2,
Section 2.3, Use of Facility Data in Inventories at https://www.ipcc-
nggip.iges.or.jp/public/2019rf/pdf/l Volumel/19R VI Ch02 DataCollection.pdf.
4-120 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
4.20 Aluminum Production (CRT Source
Category 2C3)
Aluminum is a lightweight, malleable, and corrosion-resistant metal that is used in many manufactured products,
including aircraft, automobiles, bicycles, and kitchen utensils. As of recent reporting, the United States was the
ninth82 largest producer of primary aluminum with an estimated aluminum production of 860 thousand metric
tons, with approximately 1.2 percent of the world total production (USGS 2022). The United States was also a
major importer of primary aluminum. This reporting category (2C3) includes emissions from the production of
primary aluminum—in addition to consuming large quantities of electricity—results in process-related emissions of
carbon dioxide (CO2) and two perfluorocarbons (PFCs): perfluoromethane (CF4) and perfluoroethane (C2F6).
Carbon dioxide is emitted during the aluminum smelting process when alumina (aluminum oxide, AI2O3) is reduced
to aluminum using the Hall-Heroult reduction process. The reduction of the alumina occurs through electrolysis in
a molten bath of natural or synthetic cryolite (NasAIFs). The reduction cells contain a carbon (C) lining that serves
as the cathode. Carbon is also contained in the anode, which can be a carbon mass of paste, coke briquettes, or
prebaked carbon blocks from petroleum coke. During reduction, most of this carbon is oxidized and released to the
atmosphere as CO2.
Process emissions of CO2 from aluminum production were estimated to be 1.4 MMT CO2 Eq. (1,446 kt) in 2022 (see
Table 4-101 and Table 4-102). The carbon anodes consumed during aluminum production consist of petroleum
coke and, to a minor extent, coal tar pitch. The petroleum coke portion of the total CO2 process emissions from
aluminum production is considered to be a non-energy use of petroleum coke and is accounted for here and not
under the CO2 from fossil fuel combustion source category of the Energy sector. Similarly, the coal tar pitch portion
of these CO2 process emissions is accounted for here.
Table 4-101: CO2 Emissions from Aluminum Production (MMT CO2 Eq.)
Year 1990 2005 2018 2019 2020 2021 2022
Aluminum Production 6.8 | 4.1 | 1.5 1.9 1.7 1.5 1A~
Table 4-102: CO2 Emissions from Aluminum Production (kt CO2)
Year 1990 I 2005 t 2018 2019 2020 2021 2022
Aluminum Production 6,831 | 4,142 | 1,455 1,880 1,748 1,541 1,446
In addition to CO2 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 High Voltage Anode Effects (HVAEs). HVAEs cause
carbon from the anode and fluorine from the dissociated molten cryolite bath to combine, thereby producing
fugitive emissions of CF4 and C2F6. 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. Another type of anode effect, Low Voltage Anode Effects (LVAEs),
became a concern in the early 2010s as the aluminum industry increasingly began to use cell technologies with
higher amperage and additional anodes (IPCC 2019). LVAEs emit CF4 and are included in PFC emission totals from
2006 forward.
Since 1990, emissions of CF4 and C2F6 have both declined by 96 and 97 percent respectively, to 0.62 MMT CO2 Eq.
82 Based on the U.S. USGS (2022) Aluminum factsheet, assuming all countries grouped under the "other countries" categories
all have lower production than the U.S. Available at: https://pubs.usgs.gov/periodicals/mcs2023/mcs2023-aluminum.pdf.
Industrial Processes and Product Use 4-121
-------
of CF4 (0.1 kt) and 0.08 MMT CO2 Eq. of C2F6 (0.01 kt) in 2022, respectively, as shown in Table 4-103 and Table
4-104. 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 78 percent, while the combined CF4 and C2F6
emission rate (per metric ton of aluminum produced) has been reduced by 78 percent. PFC emissions decreased by
approximately 18 percent between 2021 and 2022. Aluminum production also decreased in 2022, down 3 percent
from 2021.
Table 4-103: PFC Emissions from Aluminum Production (MMT CO2 Eq.)
Gas 1990
2005
2018
2019
2020
2021
2022
CF4 16.1
2.6 1
1.0
1.1
1.2
0.8
0.7
C2F6 3.2 1
0-5 1
0.4
0.3
0.2
0.1
0.1
Total 19.3
3.1
1.4
1.4
1.4
0.9
0.8
Note: Totals may not sum due to independent rounding.
Table 4-104: PFC Emissions from Aluminum Production (kt)
Gas 1990
2005
2018
2019
2020
2021
2022
CF4 2.4 1
0.4 1
0.2
0.2
0.2
0.1
0.1
C2F6 0.29 ¦
0.05 ¦
0.03
0.03
0.02
0.01
0.01
In 2022, U.S. primary aluminum production totaled approximately 0.86 million metric tons, a 3 percent decrease
from 2021 production levels (USGS 2023). In 2022, three companies managed production at six operational
primary aluminum smelters in five states. Two smelters operated at full capacity during 2022.The other four
smelters operated at reduced capacity and one of these four smelters began a temporary shutdown in June (USGS
2023). Domestic smelters were operating at about 52 percent of capacity of 1.64 million tons per year at year end
2022 (USGS 2023).
Methodology and Time-Series Consistency
Process CO2 and PFC (i.e., CF4 and C2F6) emission estimates from primary aluminum production for 2010 through
2022 are available from EPA's GHGRP Subpart F (Aluminum Production) (EPA 2023). Under EPA's GHGRP, facilities
began reporting primary aluminum production process emissions (for 2010) in 2011; as a result, GHGRP data (for
2010 through 2022) are available to be incorporated into the Inventory. EPA's GHGRP mandates that all facilities
that contain an aluminum production process must report: CF4 and C2F6 emissions from anode effects in all
prebake and S0derberg electrolysis cells, CO2 emissions from anode consumption during electrolysis in all prebake
and S0derberg cells, and all CO2 emissions from onsite anode baking. To estimate the process emissions, EPA's
GHGRP uses the process-specific equations detailed in Subpart F (aluminum production).83 These equations are
based on the Tier 2/Tier 3 IPCC (2006) methods for primary aluminum production, and Tier 1 methods when
estimating missing data elements. It should be noted that the same methods (i.e., 2006 IPCC Guidelines) were used
for estimating the emissions prior to the availability of the reported GHGRP data in the Inventory. Prior to 2010,
aluminum production data were provided through EPA's Voluntary Aluminum Industrial Partnership (VAIP).
As previously noted, the use of petroleum coke for aluminum production is adjusted for within the Energy chapter
to avoid double counting emissions as this fuel was consumed during non-energy related activities. Additional
information on the adjustments made within the Energy sector for non-energy use of fuels is described in both the
83 Code of Federal Regulations, Title 40: Protection of Environment, Part 98: Mandatory Greenhouse Gas Reporting, Subpart
F—Aluminum Production. See https://www.ecfr.gov/cgi-bin/text-
idx?SID=24a41781dfe4218b339e914de03e8727&mc=true&node=pt40.23.98&rgn=div5#sp40.23.98.f.
4-122 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
Methodology section of CO2 from Fossil Fuel Combustion (3.2 Carbon Emitted from Non-Energy Uses of Fossil Fuels
[CRT Source Category 1A]) and Annex 2.3, Methodology for Estimating Carbon Emitted from Non-Energy Uses of
Fossil Fuels.
Process CO2 Emissions from Anode Consumption and Anode Baking
Carbon dioxide emission estimates for the years prior to the introduction of EPA's GHGRP in 2010 were estimated
using 2006IPCC Guidelines methods, but individual facility reported data were combined with process-specific
emissions modeling. These estimates were based on information previously gathered from EPA's Voluntary
Aluminum Industrial Partnership (VAIP) program, U.S. Geological Survey (USGS) Mineral Commodity reviews, and
The Aluminum Association (USAA) statistics, among other sources. Since pre- and post-GHGRP estimates use the
same methodology, emission estimates are comparable across the time series.
Most of the CO2 emissions released during aluminum production occur during the electrolysis reaction of the
carbon anode, as described by the following reaction:
2A1203 + 3C -> 4A1 + 3C02
For prebake smelter technologies, CO2 is also emitted during the anode baking process. These emissions can
account for approximately 10 percent of total process CO2 emissions from prebake smelters.
Depending on the availability of smelter-specific data, the CO2 emitted from electrolysis at each smelter was
estimated from: (1) the smelter's annual anode consumption, (2) the smelter's annual aluminum production and
rate of anode consumption (per ton of aluminum produced) for previous and/or following years, or (3) the
smelter's annual aluminum production and IPCC default CO2 emission factors. The first approach tracks the
consumption and carbon content of the anode, assuming that all carbon in the anode is converted to CO2. Sulfur,
ash, and other impurities in the anode are subtracted from the anode consumption to arrive at a carbon
consumption figure. This approach corresponds to either the IPCC Tier 2 or Tier 3 method, depending on whether
smelter-specific data on anode impurities are used. The second approach interpolates smelter-specific anode
consumption rates to estimate emissions during years for which anode consumption data are not available. This
approach avoids substantial errors and discontinuities that could be introduced by reverting to Tier 1 methods for
those years. The last approach corresponds to the IPCC Tier 1 method (IPCC 2006) and is used in the absence of
present or historic anode consumption data.
The equations used to estimate CO2 emissions in the Tier 2 and 3 methods vary depending on smelter type (IPCC
2006). For Prebake cells, the process formula accounts for various parameters, including net anode consumption,
and the sulfur, ash, and impurity content of the baked anode. For anode baking emissions, the formula accounts
for packing coke consumption, the sulfur and ash content of the packing coke, as well as the pitch content and
weight of baked anodes produced. For S0derberg cells, the process formula accounts for the weight of paste
consumed per metric ton of aluminum produced, and pitch properties, including sulfur, hydrogen, and ash
content.
Through the VAIP, anode consumption (and some anode impurity) data have been reported for 1990, 2000, 2003,
2004, 2005, 2006, 2007, 2008, and 2009. Where available, smelter-specific process data reported under the VAIP
were used; however, if the data were incomplete or unavailable, information was supplemented using industry
average values recommended by IPCC (2006). Smelter-specific CO2 process data were provided by 18 of the 23
operating smelters in 1990 and 2000, by 14 out of 16 operating smelters in 2003 and 2004,14 out of 15 operating
smelters in 2005,13 out of 14 operating smelters in 2006, 5 out of 14 operating smelters in 2007 and 2008, and 3
out of 13 operating smelters in 2009. For years where CO2 emissions data or CO2 process data were not reported
by these companies, estimates were developed through linear interpolation, and/or assuming representative (e.g.,
previously reported or industry default) values.
In the absence of any previous historical smelter-specific process data (i.e., 1 out of 13 smelters in 2009; 1 out of
14 smelters in 2006, 2007, and 2008; 1 out of 15 smelters in 2005; and 5 out of 23 smelters between 1990 and
Industrial Processes and Product Use 4-123
-------
2003), C02 emission estimates were estimated using Tier 1 S0derberg and/or Prebake emission factors (metric ton
of CO2 per metric ton of aluminum produced) from IPCC (2006).
Process PFC Emissions from Anode Effects
High Voltage Anode Effects
Smelter-specific PFC emissions from aluminum production for 2010 through 2022 were reported to EPA under its
GHGRP. To estimate their PFC emissions from HVAEs and report them under EPA's GHGRP, smelters use an
approach identical to the Tier 3 approach in the 2006 IPCC Guidelines (IPCC 2006). Specifically, they use a smelter-
specific slope coefficient as well as smelter-specific operating data to estimate an emission factor using the
following equation:
PFC = S xAE
AE = F xD
where,
PFC = CF4 or C2F6, kg/MT aluminum
S = Slope coefficient, PFC/AE
AE = Anode effect, minutes/cell-day
F = Anode effect frequency per cell-day
D = Anode effect duration, minutes
They then multiply this emission factor by aluminum production to estimate PFC emissions from HVAEs. All U.S.
aluminum smelters are required to report their emissions under EPA's GHGRP.
Perfluorocarbon emissions for the years prior to 2010 were estimated using the same equation, but the slope-
factor used for some smelters was technology-specific rather than smelter-specific, making the method a Tier 2
rather than a Tier 3 approach for those smelters. Emissions and background data were reported to EPA under the
VAIP. For 1990 through 2009, smelter-specific slope coefficients were available and were used for smelters
representing between 30 and 94 percent of U.S. primary aluminum production. The percentage changed from year
to year as some smelters closed or changed hands and as the production at remaining smelters fluctuated. For
smelters that did not report smelter-specific slope coefficients, IPCC technology-specific slope coefficients were
applied (IPCC 2006). The slope coefficients were combined with smelter-specific anode effect data collected by
aluminum companies and reported under the VAIP to estimate emission factors over time. For 1990 through 2009,
smelter-specific anode effect data were available for smelters representing between 80 and 100 percent of U.S.
primary aluminum production. Where smelter-specific anode effect data were not available, representative values
(e.g., previously reported or industry averages) were used.
For all smelters, emission factors were multiplied by annual production to estimate annual emissions at the
smelter level. For 1990 through 2009, smelter-specific production data were available for smelters representing
between 30 and 100 percent of U.S. primary aluminum production. (For the years after 2000, this percentage was
near the high end of the range.) Production at non-reporting smelters was estimated by calculating the difference
between the production reported under VAIP and the total U.S. production supplied by USGS, and then allocating
this difference to non-reporting smelters in proportion to their production capacity. Emissions were then
aggregated across smelters to estimate national emissions (see Table 1-5).
Table 4-105: Summary of HVAE Emissions (MMT CO2 Eq.)
Year
1990
2005
2018
2019
2020
2021
2022
HVAE Emissions
19.3
3.1 1
1.4
1.4
1.4
0.9
0.7
4-124 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
Low Voltage Anode Effects
LVAE emissions of CF4 were estimated for 2006 through 2022 (see Table 1-6) based on the Tier 1 (technology-
specific, production-based) method in the 2019 Refinement to the 2006IPCC Guidelines for National Greenhouse
Gas Inventories (IPCC 2019). Prior to 2006, LVAE emissions are believed to have been negligible.84 The Tier 1
method is used in the LVAE emissions calculations from aluminum production in the absence of smelter-specific
data available to quantify the LVAE-specific process emissions. National aluminum production estimates (allocated
to smelters as described below) and the technology used in individual smelters were the best available data to
perform the emissions calculations, as smelter-specific production data is not publicly available.
The following equation was used to estimate LVAE PFC emissions:
Equation 4-15: CF4 Emissions Resulting from Low Voltage Anode Effects
LVAE ECF4 = LVAEEFCF4 X MP
where,
LVAE Ecf4 = LVAE emissions of CF4 from aluminum production, kg CF4
LVAE EFcf4 = LVAE emission factor for CF4 (default by cell technology type)
MP = Metal production by cell technology type, tons Al.
In the LVAE emissions calculations, the Metal Production (MP) factor is calculated differently for the years 2006
through 2009 than for 2010 and beyond. For years prior to GHGRP reporting (2006 through 2009), the MP factor is
calculated by dividing the annual production reported by USGS with the total U.S. capacity reported for this
specific year, based on the USGS yearbook and applying this national utilization factor to each facility's production
capacity to obtain an estimated facility production value. For GHGRP reporting years (2010+), the methodology to
calculate the MP value was changed to allocate the total annual production reported by USAA, based on the
distribution of CO2 emissions amongst the operating smelters in a specific year. The latter improves the accuracy of
the LVAE emissions estimates over assuming capacity utilization is the same at all smelters. The main drawback of
using this methodology to calculate the MP factor is that, in some instances, it led to production estimates that are
slightly larger (less than six percent) than the production capacity reported that year. In practice, this is most likely
explained by the differences in process efficiencies at each facility and to a lesser extent, differences in
measurements and methods used by each facility to obtain their CO2 estimates and the degree of uncertainty in
the USGS annual production reporting.
Once LVAE emissions were estimated, they were then combined with HVAE emissions estimates to calculate total
PFC emissions from aluminum production.
Table 4-106: Summary of LVAE Emissions (MMT CO2 Eq.)
Year
2006
2018
2019
2020
2021
2022
LVAE Emissions
0.13
0.05
0.07
0.06
0.05
0.05
Production Data
Between 1990 and 2009, production data were provided under the VAIP by 21 of the 23 U.S. smelters that
operated during at least part of that period. For the non-reporting smelters, production was estimated based on
84 The 2019 Refinement states, "Since 2006, the global aluminum industry has undergone changes in technology and operating
conditions that make LVAE emissions much more prevalent12; these changes have occurred not only through uptake of newer
technologies (e.g., PFPBL to PFPBM) but also during upgrades within the same technology in order to maximize productivity and
reduce energy use" (IPCC 2019). Footnote #12 uses the example of PFPBL, which is prevalent in the United States, as an older
technology that has been upgraded.
Industrial Processes and Product Use 4-125
-------
the difference between reporting smelters and national aluminum production levels as reported to USGS, with
allocation to specific smelters based on reported production capacities (USGS 1990 through 2009).
National primary aluminum production data for 2010 through 2022 were compiled using USGS Mineral Industry
Surveys, and the USGS Mineral Commodity Summaries (see Table 1-7).
Table 4-107: Production of Primary Aluminum (kt)
Year
1990 2005 2018
2019
2020 2021 2022
Production (kt)
4,048 1 2,481 891
1,093
1,012 889 860
Methodological approaches were applied to the entire time-series to ensure time-series consistency from 1990
through 2022.
Uncertainty
Uncertainty was estimated for the CO2, CF4, and C2F6 emission values reported by each individual facility to EPA's
GHGRP, taking into consideration the uncertainties associated with aluminum production, anode effect minutes,
and slope factors. The uncertainty bounds used for these parameters were established based on information
collected under the VAIP and held constant through 2022. Uncertainty surrounding the reported CO2, CF4, and C2F6
emission values were determined to have a normal distribution with uncertainty ranges of approximately 6
percent below to 6 percent above, 16 percent below to 16 percent above, and 20 percent below to 20 percent
above their 2022 emission estimates, respectively.
For LVAE, since emission values were not reported through EPA's GHGRP but estimated instead through a Tier 1
methodology, the uncertainty analysis examined uncertainty associated with primary capacity data as well as
technology-specific emission factors. Uncertainty for each facility's primary capacity, reported in the USGS
Yearbook, was estimated to have a Pert Beta distribution with an uncertainty range of 7 percent below to 7
percent above the capacity estimates based on the uncertainty of reported capacity data, the number of years
since the facility reported new capacity data, and uncertainty in capacity utilization. Uncertainty was applied to
LVAE emission factors according to technology using the uncertainty ranges provided in the 2019 Refinement to
the 2006IPCC Guidelines. An uncertainty range for Horizontal Stud S0derberg (HSS) technology was not provided
in the 2019 Refinement to the 2006 IPCC Guidelines due to insufficient data, so a normal distribution and
uncertainty range of ±99 percent was applied for that technology based on expert judgment. A Monte Carlo
analysis was applied to estimate the overall uncertainty of the CO2, CF4, and C2F6 emission estimates for the U.S.
aluminum industry as a whole, and the results are provided below.
The results of this Approach 2 quantitative uncertainty analysis are summarized in Table 4-108. Aluminum
production-related CO2 emissions were estimated to be between 1.41 and 1.48 MMT CO2 Eq. at the 95 percent
confidence level. This indicates a range of approximately 3 percent below to 3 percent above the emission
estimate of 1.446 MMT CO2 Eq. Also, production-related CF4 emissions were estimated to be between 0.62 and
0.73 MMT CO2 Eq. at the 95 percent confidence level. This indicates a range of approximately 8 percent below to 8
percent above the emission estimate of 0.676 MMT CO2 Eq. Aluminum production-related C2F6 emissions were
estimated to be between 0.075 and 0.09 MMT CO2 Eq. at the 95 percent confidence level. This indicates a range of
approximately 9 percent below to 9 percent above the emission estimate of 0.083 MMT CO2 Eq. Finally, Aluminum
production-related aggregated PFCs emissions were estimated to be between 0.71 and 0.82 MMT CO2 Eq. at the
95 percent confidence level. This indicates a range of approximately 7 percent below to 7 percent above the
emission estimate of 0.759 MMT CO2 Eq.
4-126 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
Table 4-108: Approach 2 Quantitative Uncertainty Estimates for CO2 and PFC Emissions from
Aluminum Production (MMT CO2 Eq. and Percent)
Source
Gas
2022 Emission Estimate
(MMT CO . Eq.)
Uncertainty Range Relative to Emission Estimate"
(MMTCO' Eq.) (%)
Lower
Upper
Lower
Upper
Bound
Bound
Bound
Bound
Aluminum Production
C02
1.446
1.41
1.48
-3%
+3%
Aluminum Production
cf4
0.676
0.62
0.73
-8%
+8%
Aluminum Production
c2f6
0.083
0.075
0.09
-9%
+9%
Aluminum Production
PFCs
0.759
0.71
0.82
-7%
+7%
a Range of emission estimates predicted by Monte Carlo stochastic simulation for a 95 percent confidence interval.
QA/QC and Verification
General quality assurance/quality control (QA/QC) procedures were applied consistent with the U.S. Inventory
QA/QC plan, which is in accordance with Volume 1, Chapter 6 of 2006IPCC Guidelines as described in the
introduction of the IPPU chapter (see Annex 8 for more details). For the GHGRP data, EPA verifies annual facility-
level reports through a multi-step process (e.g., including a combination of pre-and post-submittal electronic
checks and manual reviews by staff) to identify potential errors and ensure that data submitted to EPA are
accurate, complete, and consistent (EPA 2015).85 Based on the results of the verification process, EPA follows up
with facilities to resolve mistakes that may have occurred. The post-submittals checks are consistent with a
number of general and category-specific QC procedures, including: range checks, statistical checks, algorithm
checks, and year-to-year checks of reported data and emissions.
Recalculations Discussion
No recalculations were performed for the 1990 through 2021 portion of the time series.
Planned Improvements
EPA is assessing planned improvements for future reports, but at this time has no specific planned improvements
for estimating C02and PFC emissions from aluminum production.
4.21 Magnesium Production and Processing
(CRT Source Category 2C4)
The magnesium metal production and casting industry uses sulfur hexafluoride (SFs) as a cover gas to prevent the
rapid oxidation of molten magnesium in the presence of air. This reporting category (2C4) includes emissions from
magnesium metal production and processing. Sulfur hexafluoride has been used in this application around the
world for more than 30 years. A dilute gaseous mixture of SF6 with dry air and/or carbon dioxide (CO2) is blown
over molten magnesium metal to induce and stabilize the formation of a protective crust. A small portion of the
SFs reacts with the magnesium to form a thin molecular film of mostly magnesium oxide and magnesium fluoride.
85 GHGRP Report Verification Factsheet. See https://www.epa.gov/sites/production/files/2Q15-
07/documents/ghgrp verification factsheet.pdf.
Industrial Processes and Product Use 4-127
-------
The amount of SF6 reacting in magnesium production and processing is considered to be negligible and thus all SF6
used is assumed to be emitted into the atmosphere. Alternative cover gases, such as AM-cover™ (containing HFC-
134a), Novec™ 612 (FK-5-1-12) and dilute sulfur dioxide (SO2) systems can and are being used by some facilities in
the United States. However, many facilities in the United States are still using traditional SF6 cover gas systems.
Carbon dioxide is also released during primary magnesium production if carbonate based raw materials, such as
dolomite, are used. During the processing of these raw materials to produce magnesium, calcination occurs which
results in a release of CO2 emissions.
The magnesium industry emitted 1.1 MMT CO2 Eq. (0.05 kt) of SF6, 0.03 MMT CO2 Eq. (0.02 kt) of HFC-134a, and
0.003 MMT CO2 Eq. (2.9 kt) of CO2 in 2022. This represents a decrease of approximately 4 percent from total 2021
emissions (see Table 4-109 and Table 4-110) and a decrease in SF6 emissions by 3 percent. In 2022, total HFC-134a
emissions decreased from 0.040 MMT CO2 Eq. to 0.029 MMT CO2 Eq., or a 28 percent decrease as compared to
2021 emissions. FK 5-1-12 emissions in 2022 were consistent with 2021. The emissions of the carrier gas, CO2,
increased from 2.91 kt in 2021 to 2.94 kt in 2022, or 1 percent.
Table 4-109: SF6, HFC-134a, FK 5-1-12 and CO2 Emissions from Magnesium Production and
Processing (MMT CO2 Eq.)
Year
1990
2005
2018
2019
2020
2021
2022
sf6
5.6
3.0
1.1
0.9
0.9
1.2
1.1
HFC-134a
0.0
0.0
0.1
0.1
0.1
+
+
C02
0.1
+
+
+
+
+
+
FK 5-1-12"
0.0
0.0
+
+
+
+
+
Total
5.7
3.0
1.1
1.0
0.9
1.2
1.2
+ Does not exceed 0.05 MMT C02 Eq.
a Emissions of FK 5-1-12 are not included in totals.
Note: Totals may not sum due to independent rounding.
Table 4-110: SF6, HFC-134a, FK 5-1-12 and CO2 Emissions from Magnesium Production and
Processing (kt)
Year
1990
2005
2018
2019
2020
2021
2022
sf6
0.2
0.1
+
+
+
+
+
HFC-134a
0.0
8—
O
O
0.1
+
+
+
+
C02
129.0
3.6
1.6
2.4
3.0
2.9
2.9
FK 5-1-12°
0-0 1
0.0 |
+
+
+
+
+
+ Does not exceed 0.5 kt
a Emissions of FK 5-1-12 are not included in totals.
Methodology and Time-Series Consistency
Emission estimates for the magnesium industry incorporate information provided by industry participants in EPA's
SFs Emission Reduction Partnership for the Magnesium Industry as well as emissions data reported through
Subpart T (Magnesium Production and Processing) of EPA's GHGRP. The Partnership started in 1999 and, in 2010,
participating companies represented 100 percent of U.S. primary and secondary production and 16 percent of the
casting sector production (i.e., die, sand, permanent mold, wrought, and anode casting). SF6 emissions for 1999
through 2010 from primary production, secondary production (i.e., recycling), and die casting were generally
reported by Partnership participants. Partners reported their SF6 consumption, which is assumed to be equivalent
to emissions. Along with SF6, some Partners reported their HFC-134a and FK 5-1-12 consumed, which is also
assumed to be equal to emissions. The last reporting year under the Partnership was 2010. Emissions data for
2011 through 2020 are obtained through EPA's GHGRP. Under the program, owners or operators of facilities that
have a magnesium production or casting process must report emissions from use of cover or carrier gases, which
include SF6, HFC-134a, FK 5-1-12 and CO2. Consequently, cover and carrier gas emissions from magnesium
4-128 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
production and processing were estimated for three time periods, depending on the source of the emissions data:
1990 through 1998 (pre-EPA Partnership), 1999 through 2010 (EPA Partnership), and 2011 through 2022 (EPA
GHGRP). The methodologies described below also make use of magnesium production data published by the U.S.
Geological Survey (USGS) as available.
1990 through 1998
To estimate emissions for 1990 through 1998, industry SF6 emission factors were multiplied by the corresponding
metal production and consumption (casting) statistics from USGS. For this period, it was assumed that there was
no use of HFC-134a or FK 5-1-12 cover gases, and hence emissions were not estimated for these alternatives.
Sulfur hexafluoride emission factors from 1990 through 1998 were based on a number of sources and
assumptions. Emission factors for primary production were available from U.S. primary producers for 1994 and
1995. The primary production emission factors were 1.2 kg SF6 per metric ton for 1990 through 1993, and 1.1 kg
SFs per metric ton for 1994 through 1997. The emission factor for secondary production from 1990 through 1998
was assumed to be constant at the 1999 average Partner value. An emission factor for die casting of 4.1 kg SF6 per
metric ton, which was available for the mid-1990s from an international survey (Gjestland and Magers 1996), was
used for years 1990 through 1996. For 1996 through 1998, the emission factor for die casting was assumed to
decline linearly to the level estimated based on Partner reports in 1999. This assumption is consistent with the
trend in SF6 sales to the magnesium sector that was reported in the RAND survey of major SF6 manufacturers,
which showed a decline of 70 percent from 1996 to 1999 (RAND 2002). Sand casting emission factors for 1990
through 2001 were assumed to be the same as the 2002 emission factor for all but one facility, which used an
emission factor derived from 2011 GHGRP data and held constant to back cast emissions for 1990-1998. The
emission factors for the other processes (i.e., permanent mold, wrought, and anode casting), about which less is
known, were assumed to remain constant at levels defined in Table 4-110. The emission factors for the other
processes (i.e., permanent mold, wrought, and anode casting) were based on discussions with industry
representatives.
The quantities of CO2 carrier gas used for each production type have been estimated using the 1999 estimated CO2
emissions data and the annual calculated rate of change of SF6 use in the 1990 through 1999 time period. For each
year and production type, the rate of change of SF6 use between the current year and the subsequent year was
first estimated. This rate of change was then applied to the CO2 emissions of the subsequent year to determine the
CO2 emission of the current year.
Carbon dioxide emissions from the calcination of dolomite in the primary production of magnesium were
calculated based on the 2006IPCC Guidelines Tier 2 method by multiplying the estimated primary production of
magnesium by an emissions factor of 3.62 kilogram of CO2 per kilogram of magnesium produced.86 For 1990
through 1998, production was estimated to be equal to the production capacity of the facility.
1999 through 2010
The 1999 through 2010 emissions from primary and secondary production were based on information provided by
EPA's industry Partners. In some instances, there were years of missing Partner data, including SF6 consumption
and metal processed. For these situations, emissions were estimated through interpolation where possible, or by
holding company-reported emissions (as well as production) constant from the previous year. For alternative cover
gases, including HFC-134a and FK 5-1-12, mainly reported data was relied upon. That is, unless a Partner reported
using an alternative cover gas, it was not assumed it was used. Emissions of alternate gases were also estimated
through linear interpolation where possible.
86 See https://www.ipcc-negip.iges.or.ip/public/2006el/pdf/3 Volume3/V3 4 Ch4 Metal lndustrv.pdf.
Industrial Processes and Product Use 4-129
-------
The die casting emission estimates for 1999 through 2010 were also based on information supplied by industry
Partners. When a Partner was determined to be no longer in production, its metal production and usage rates
were set to zero. Missing data on emissions or metal input was either interpolated or held constant at the last
available reported value. In 1999 through 2010, Partners were assumed to account for all die casting tracked by
USGS. For 1999, die casters who were not Partners were assumed to be similar to Partners who cast small parts.
Due to process requirements, these casters consume larger quantities of SF6 per metric ton of processed
magnesium than casters that process large parts. Consequently, emission estimates from this group of die casters
were developed using an average emission factor of 5.2 kg SF6 per metric ton of magnesium. This emission factor
was developed using magnesium production and SF6 usage data for the year 1999. In 2008, the derived emission
factor for die casting began to increase after many years of largely decreasing emission factors. As determined
through an analysis of activity data reported from the USGS, this increase is due to a temporary decrease in
production at many facilities between 2008 and 2010, which reflects the change in production that occurred
during the recession.
The emissions from other casting operations were estimated by multiplying emission factors (kg SF6 per metric ton
of metal produced or processed) by the amount of metal produced or consumed from USGS, with the exception of
some years for which Partner sand casting emissions data are available. The emission factors for sand casting
activities were acquired through the data reported by the Partnership for 2002 to 2006. For 1999 through 2001,
the sandcasting emission factor was held constant at the 2002 Partner-reported level. For 2007 through 2010, the
sandcasting Partner did not report and the reported emission factor from 2005 was applied to the Partner and to a
non GHGRP sand casters. Activity data for 2005 was obtained from USGS (USGS 2005b). One non partner sand
casting facility reported to GHGRP in 2011 and had an emission factor derived for 2011, this factor was used to
back cast emissions for this facility from 1999 to 2010.
The emission factors for primary production, secondary production and sand casting for the 1999 to 2010 are not
published to protect company-specific production information. However, the emission factor for primary
production has not risen above the average 1995 Partner value of 1.1 kg SF6 per metric ton. The emission factors
for the other industry sectors (i.e., permanent mold, wrought, and anode casting) were based on discussions with
industry representatives. The emission factors for casting activities are provided below in Table 4-111.
The emissions of HFC-134a and FK-5-1-12 were included in the estimates for only instances where Partners
reported that information to the Partnership. Emissions of these alternative cover gases were not estimated for
instances where emissions were not reported.
Carbon dioxide carrier gas emissions were estimated using the emission factors developed based on GHGRP-
reported carrier gas and cover gas data, by production type. It was assumed that the use of carrier gas, by
production type, is proportional to the use of cover gases. Therefore, an emission factor, in kg CO2 per kg cover gas
and weighted by the cover gases used, was developed for each of the production types. GHGRP data, on which
these emissions factors are based, was available for primary, secondary, die casting and sand casting. The emission
factors were applied to the quantity of all cover gases used (SF6, HFC-134a, and FK-5-1-12) by production type in
this time period for producers that reported CO2 emissions from 2011-2022 through the GHGP. Carrier gas
emissions for the 1999 through 2010 time period were only estimated for those Partner companies that reported
using CO2 as a carrier gas through the GHGRP. Using this approach helped ensure time-series consistency.
Emissions of carrier gases for permanent mold, wrought, and anode processes were estimated using the ratio of
total CO2 emissions to total cover gas emissions for primary, secondary, die and sand in a given year and the total
SFs emissions from each permanent mold, wrought, and anodes processes respectively in that same year. CO2
emissions from the calcination of dolomite were estimated using the same approach as described above. At the
end of 2001, the sole magnesium production plant operating in the United States that produced magnesium metal
using a dolomitic process that resulted in the release of CO2 emissions ceased its operations (USGS 1995b through
2023).
4-130 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
Table 4-111: SF6 Emission Factors (kg SF6 per metric ton of magnesium)
Year
Die Casting9
Permanent Mold
Wrought
Anodes
1999
1.75b
2
1
1
2000
0.72
2
1
1
2001
0.72
2
1
1
2002
0.71
2
1
1
2003
0.81
2
1
1
2004
0.79
2
1
1
2005
0.77
2
1
1
2006
0.88
2
1
1
2007
0.64
2
1
1
2008
0.97
2
1
1
2009
1.41
2
1
1
2010
1.43
2
1
1
a Weighted average includes all die casters, Partners and non-Partners. For
the majority of the time series (2000 through 2010), Partners made up
100 percent of die casters in the United States.
b Weighted average that includes an estimated emission factor of 5.2 kg
SF6 per metric ton of magnesium for die casters that do not participate in
the Partnership.
2011 through 2022
For 2011 through 2022, for the primary and secondary producers, GHGRP-reported cover and carrier gases
emissions data were used. For sand and die casting, some emissions data was obtained through EPA's GHGRP.
Additionally, in 2018 a new GHGRP reporter began reporting permanent mold emissions. The balance of the
emissions for this industry segment was estimated based on previous Partner reporting (i.e., for Partners that did
not report emissions through EPA's GHGRP) or were estimated by multiplying emission factors by the amount of
metal produced or consumed. Partners who did not report through EPA's GHGRP were assumed to have continued
to emit SFs at the last reported level, which was from 2010 in most cases, unless publicly available sources
indicated that these facilities have closed or otherwise eliminated SF6 emissions from magnesium production (ARB
2015). Many Partners that did report through the GHGRP showed increases in SF6 emissions driven by increased
production related to a continued economic recovery after the 2008 recession. One Partner in particular reported
an anonymously large increase in SF6 emissions from 2010 to 2011, further driving increases in emissions between
the two time periods of inventory estimates. All Partners were assumed to have continued to consume magnesium
at the last reported level. Where the total metal consumption estimated for the Partners fell below the U.S. total
reported by USGS, the difference was multiplied by the emission factors discussed in the section above, i.e., non-
partner emission factors. For the other types of production and processing (i.e., permanent mold, wrought, and
anode casting), emissions were estimated by multiplying the industry emission factors with the metal production
or consumption statistics obtained from USGS (USGS 1995b-2023). USGS data for 2022 were not yet available at
the time of the analysis, so the 2021 values were held constant through 2022 as an estimate.
Emissions of carrier gases for permanent mold, wrought, and anode processes were estimated using an approach
consistent with the 1999 through 2010 time series.
Methodological approaches were applied to the entire time series to ensure time-series consistency from 1990
through 2022. 2006 IPCC Guidance methodologies were used throughout the time series, mainly either a Tier 2 or
Tier 3 approach depending on available data.
Uncertainty
Uncertainty surrounding the total estimated emissions in 2022 is attributed to the uncertainties around SF6, HFC-
134a, and CO2 emission estimates. To estimate the uncertainty surrounding the estimated 2022 SF6 emissions from
Industrial Processes and Product Use 4-131
-------
magnesium production and processing, the uncertainties associated with three variables were estimated: (1)
emissions reported by magnesium producers and processors for 2022 through EPA's GHGRP, (2) emissions
estimated for magnesium producers and processors that reported via the Partnership in prior years but did not
report 2022 emissions through EPA's GHGRP, and (3) emissions estimated for magnesium producers and
processors that did not participate in the Partnership or report through EPA's GHGRP. An uncertainty of 5 percent
was assigned to the emissions (usage) data reported by each GHGRP reporter for all the cover and carrier gases
(per the 2006IPCC Guidelines). If facilities did not report emissions data during the current reporting year through
EPA's GHGRP, SFs emissions data were held constant at the most recent available value reported through the
Partnership. The uncertainty associated with these values was estimated to be 30 percent for each year of
extrapolation (per the 2006 IPCC Guidelines). The uncertainty of the total inventory estimate remained relatively
constant between 2021 and 2022.
Alternate cover gas and carrier gases data was set equal to zero if the facilities did not report via the GHGRP. For
those industry processes that are not represented in the Partnership, such as permanent mold and wrought
casting, SF6 emissions were estimated using production and consumption statistics reported by USGS and
estimated process-specific emission factors (see Table 4-111). The uncertainties associated with the emission
factors and USGS-reported statistics were assumed to be 75 percent and 25 percent, respectively. Emissions
associated with die casting and sand casting activities utilized emission factors based on Partner reported data
with an uncertainty of 75 percent. In general, where precise quantitative information was not available on the
uncertainty of a parameter, a conservative (upper-bound) value was used.
Additional uncertainties exist in these estimates that are not addressed in this methodology, such as the basic
assumption that SF6 neither reacts nor decomposes during use. The melt surface reactions and high temperatures
associated with molten magnesium could potentially cause some gas degradation. Previous measurement studies
have identified SF6 cover gas degradation in die casting applications on the order of 20 percent (Bartos et al. 2007).
Sulfur hexafluoride may also be used as a cover gas for the casting of molten aluminum with high magnesium
content; however, the extent to which this technique is used in the United States is unknown.
The results of this Approach 2 quantitative uncertainty analysis are summarized in Table 4-112. Total emissions
associated with magnesium production and processing were estimated to be between 1.06 and 1.24 MMT CO2 Eq.
at the 95 percent confidence level. This indicates a range of approximately 7.9 percent below to 7.7 percent above
the 2022 emission estimate of 1.15 MMT CO2 Eq. The uncertainty estimates for 2022 are slightly higher to the
uncertainty reported for 2021 in the previous Inventory. This increase in uncertainty is attributed to the increased
number of facilities with interpolated emissions and the increasing number of years for facilities with emissions
held constant.
Table 4-112: Approach 2 Quantitative Uncertainty Estimates for SF6, HFC-134a and CO2
Emissions from Magnesium Production and Processing (MMT CO2 Eq. and Percent)
Source
Gas
2022 Emission Estimate
(MMTCO. Eq.)
Uncertainty Range Relative to Emission Estimate-1
(MMTCO' Eq.) (%)
Lower
Upper
Lower Upper
Bound
Bound
Bound Bound
Magnesium
Production
sf6, hfc-
134a, C02
1.2
1.1
1.2
-7.9% +7.7%
a Range of emission estimates predicted by Monte Carlo stochastic simulation for a 95 percent confidence
interval.
QA/QC and Verification
General quality assurance/quality control (QA/QC) procedures were applied consistent with the U.S. Inventory
QA/QC plan, which is in accordance with Volume 1, Chapter 6 of 2006 IPCC Guidelines as described in the
4-132 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
introduction of the IPPU chapter (see Annex 8 for more details). For the GHGRP data, EPA verifies annual facility-
level reports through a multi-step process (e.g., including a combination of pre-and post-submittal electronic
checks and manual reviews by staff) to identify potential errors and ensure that data submitted to EPA are
accurate, complete, and consistent (EPA 2015).87 Based on the results of the verification process, EPA follows up
with facilities to resolve mistakes that may have occurred. The post-submittals checks are consistent with a
number of general and category-specific QC procedures, including: range checks, statistical checks, algorithm
checks, and year-to-year checks of reported data and emissions.
Recalculations Discussion
One die casting facility which had previously had emissions back cast at a constant level had its back casting
methodology updated using linear growth from 0 to reported emissions levels between 2001 and 2014, resulting in
decreases in SF6 emissions across 2001 to 2013.
Sand Casting Emissions for 2021 were updated based on 2021 specific data available in the 2021 data tables
release from USGS's Mineral Yearbook. 2021 data was previously held constant at 2020 levels due to USGS Mineral
Yearbook data only going through 2020. The updated production of sand cast magnesium was larger than what
was estimated for 2021 in the previous Inventory cycle leading to an increase in SF6 emissions in 2021.
One sand casting facility, which had previously only been estimated from 2011 onward, was confirmed to have
emissions across the time series, an updated emission factor for 2011 was calculated and used to back cast
emissions from 1990 to 2010.
Review of facility responses indicate that changes over time in the emission factors for this industry have occurred
as facilities switch to using systems with cover gases other than SF6 (e.g. SO2) and also during time-periods where
back-up SF6-based systems are used due to the failure of the primary (non-SFs) system have occurred, leading to
the periodic spike in SF6 usage rates.
Planned Improvements
Cover gas research conducted over the last decade has found that SF6 used for magnesium melt protection can
have degradation rates on the order of 20 percent in die casting applications (Bartos et al. 2007). Current emission
estimates assume (per the 2006IPCC Guidelines) that all SF6 utilized is emitted to the atmosphere. Additional
research may lead to a revision of the 2006 IPCC Guidelines to reflect this phenomenon and until such time,
developments in this sector will be monitored for possible application to the Inventory methodology.
Additional emissions are generated as byproducts from the use of alternate cover gases, which are not currently
accounted for. Research on this topic is developing, and as reliable emission factors become available, these
emissions will be incorporated into the Inventory.
4.22 Lead Production (CRT Source Category
2C5)
In 2022, lead was produced in the United States using only secondary production processes. Until 2014, lead
production in the United States involved both primary and secondary processes—both of which emit carbon
87 GHGRP Report Verification Factsheet. See https://www.epa.gov/sites/production/files/2015-
07/documents/eherp verification factsheet.pdf.
Industrial Processes and Product Use 4-133
-------
dioxide (CO2) (Sjardin 2003). This reporting category (2C5) includes emissions from the production of lead. Per the
IPCC methodological guidance, emissions from fuels consumed for energy purposes during the production of lead
are accounted for as part of fossil fuel combustion in the industrial end-use sector reported under the Energy
chapter.
Primary production of lead through the direct smelting of lead concentrate produces CO2 emissions as the lead
concentrates are reduced in a furnace using metallurgical coke (Sjardin 2003). Primary lead production, in the form
of direct smelting, previously occurred at a single smelter in Missouri. This primary lead smelter was closed at the
end of 2013, and a small amount of residual lead was processed during demolition of the facility in 2014 (USGS
2015). Beginning in 2015, primary lead production no longer occurred in the United States.
Similar to primary lead production, CO2 emissions from secondary lead production result when a reducing agent,
usually metallurgical coke, is added to the smelter to aid in the reduction process. Carbon dioxide emissions from
secondary production also occur through the treatment of secondary raw materials (Sjardin 2003). Secondary
production primarily involves the recycling of lead acid batteries and post-consumer scrap at secondary smelters.
Secondary lead production in the United States has fluctuated over the past 20 years, reaching a high of 1,180,000
metric tons in 2007 and again in 2019. In 2022, secondary lead production accounted for 100 percent of total U.S.
lead production. The lead-acid battery industry accounted for about 92 percent of the reported U.S. lead
consumption in 2022 (USGS 2023a).
In 2022, secondary lead production in the United States decreased by approximately 3 percent compared to 2021
(USGS 2023a). Secondary lead production in 2022 is 3 percent higher than in 1990 (USGS 1994 and 2023a). The
United States has become more reliant on imported refined lead, owing to the closure of the last primary lead
smelter in 2013. Exports of spent starting-lighting-ignition (SLI) batteries decreased between 2014 and 2017, and
subsequently recovered beginning in 2018. Exports were 10 percent higher in the first 9 months of 2021 compared
to the same time period in 2014 (USGS 2015 through 2023b). In the first 9 months of 2022, 24.6 million spent SLI
lead-acid batteries were exported, 4 percent less than that in the same time period in 2021 (USGS 2023b).
Emissions of CO2 from lead production in 2022 were 0.4 MMT CO2 Eq. (428 kt), which is a 3 percent decrease
compared to 2021 and a 17 percent decrease compared to 1990 (see Table 4-113 and Table 4-114) (USGS 1994;
USGS 2023a; USGS 2023b).
The United States was the third largest mine producer of lead in the world, behind China and Australia, and
accounted for approximately 6 percent of world production in 2022 (USGS 2023a).
Table 4-113: CO2 Emissions from Lead Production (MMT CO2 Eq.)
Year 1990 2005 2018 2019
2020
2021
2022
Lead Production 0.5 | 0.6 0.5 0.5
0.5
0.4
0.4
Table 4-114: CO2 Emissions from Lead Production (kt CO2)
Year 1990 2005 2018 2019
2020
2021
2022
Lead Production 516 j 553 j 527 531
450
439
428
4-134 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
Methodology and Time-Series Consistency
Carbon dioxide emissions from lead production88 are calculated based on Sjardin's work (Sjardin 2003) for lead
production emissions and use Tier 1 methods from the 2006IPCC Guidelines, in accordance with the IPCC
methodological decision tree and available data. The Tier 1 equation is as follows:
Equation 4-16:2006 IPCC Guidelines Tier 1: CO2 Emissions From Lead Production (Equation
4.32)
C02 Emissions = (DS x EFDS) + (5 x EFS)
where,
DS = Lead produced by direct smelting, metric ton
S = Lead produced from secondary materials
EFds = Emission factor for direct smelting, metric tons CCh/metric ton lead product
EFs = Emission factor for secondary materials, metric tons CCh/metric ton lead product
For primary lead production using direct smelting, Sjardin (2003) and the 2006 IPCC Guidelines provide an emission
factor of 0.25 metric tons CCh/metric ton lead. For secondary lead production, Sjardin (2003) and the 2006 IPCC
Guidelines provide an emission factor of 0.25 metric tons CCh/metric ton lead for direct smelting, as well as an
emission factor of 0.2 metric tons CCh/metric ton lead produced for the treatment of secondary raw materials (i.e.,
pretreatment of lead acid batteries). Since the secondary production of lead involves both the use of the direct
smelting process and the treatment of secondary raw materials, Sjardin recommends an additive emission factor
to be used in conjunction with the secondary lead production quantity. The direct smelting factor (0.25) and the
sum of the direct smelting and pretreatment emission factors (0.45) are multiplied by total U.S. primary and
secondary lead production, respectively, to estimate CO2 emissions.
The production and use of coking coal for lead production is adjusted for within the Energy chapter as this fuel was
consumed during non-energy related activities. Additional information on the adjustments made within the Energy
sector for Non-Energy Use of Fuels is described in both the Methodology section of CO2 from Fossil Fuel
Combustion (Section 3.1 Fossil Fuel Combustion (CRT Source Category 1A)) and Annex 2.1, Methodology for
Estimating Emissions of CO2 from Fossil Fuel Combustion.
The 1990 through 2022 activity data for primary and secondary lead production (see Table 4-115) were obtained
from the U.S. Geological Survey (USGS 1994 through 2023a).
Table 4-115: Lead Production (Metric Tons)
Year
1990
2005
2018
2019
2020
2021
2022
Primary
Secondary
404,000
922,000
143,000
1,150,000 1
0
1,170,000
0
1,180,000
0
1,000,000
0
975,000
0
950,000
Methodological approaches discussed below were applied to applicable years to ensure time-series consistency in
emissions from 1990 through 2022.
88 EPA has not integrated aggregated facility-level Greenhouse Gas Reporting Program (GHGRP) information to inform these
estimates. The aggregated information (e.g., activity data and emissions) associated with Lead Production did not meet criteria
to shield underlying confidential business information (CBI) from public disclosure.
Industrial Processes and Product Use 4-135
-------
Uncertainty
Uncertainty associated with lead production relates to the emission factors and activity data used. The direct
smelting emission factor used in primary production is taken from Sjardin (2003) who averaged the values
provided by three other studies (Dutrizac et al. 2000; Morris et al. 1983; Ullman 1997). For secondary production,
Sjardin (2003) added a CO2 emission factor associated with battery treatment. The applicability of these emission
factors to plants in the United States is uncertain. EPA assigned an uncertainty range of ±20 percent for these
emission factors, and using this suggested uncertainty provided in Table 4.23 of the 2006IPCC Guidelines for a Tier
1 emission factor by process type is appropriate based on expert judgment (RTI 2023). Per this expert judgment, a
triangular probability density function was assumed for emission factors.
There is also a smaller level of uncertainty associated with the accuracy of primary and secondary production data
provided by the USGS which is collected via voluntary surveys; the uncertainty of the activity data is a function of
the reliability of reported plant-level production data and the completeness of the survey response. EPA currently
uses an uncertainty range of ±10 percent for primary and secondary lead production, and using this suggested
uncertainty provided in Table 4.23 of the 2006 IPCC Guidelines for Tier 1 national production data is appropriate
based on expert judgment (RTI 2023). Per this expert judgment, a normal probability density function was
assumed for all activity data.
The results of the Approach 2 quantitative uncertainty analysis are summarized in Table 4-116. Lead production
CO2 emissions in 2022 were estimated to be between 0.4 and 0.5 MMT CO2 Eq. at the 95 percent confidence level.
This indicates a range of approximately 15 percent below and 16 percent above the emission estimate of 0.4 MMT
CO2 Eq.
Table 4-116: Approach 2 Quantitative Uncertainty Estimates for CO2 Emissions from Lead
Production (MMT CO2 Eq. and Percent)
Source
Gas
2022 Emission Estimate
Uncertainty Range Relative to Emission Estimate-1
(MMTCO. Eq.)
(MMTCO. Eq.)
(%)
Lower Upper
Lower Upper
Bound Bound
Bound Bound
Lead Production
C02
0.4
LD
O
O
-15% +16%
a Range of emission estimates predicted by Monte Carlo stochastic simulation for a 95 percent confidence interval.
QA/QC and Verification
General quality assurance/quality control (QA/QC) procedures were applied consistent with the U.S. Inventory
QA/QC plan, which is in accordance with Volume 1, Chapter 6 of 2006 IPCC Guidelines as described in the
introduction of the IPPU chapter (see Annex 8 for more details).
Initial review of activity data show that EPA's GHGRP Subpart R lead production data and resulting emissions are
fairly consistent with those reported by USGS. EPA is still reviewing available GHGRP data, reviewing QC analysis to
understand differences in data reporting (i.e., threshold implications), and assessing the possibility of including this
planned improvement in future Inventory reports (see Planned Improvements section below). Currently, GHGRP
data are used for QA purposes only.
Recalculations Discussion
Recalculations were implemented for 2020 and 2021 based on revised USGS data for secondary lead production.
Compared to the previous Inventory, emissions decreased by 3 percent (14 kt CO2) for 2020 and by 2 percent (7 kt
CO2) for 2021 (USGS 2023b).
4-136 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
Planned Improvements
Pending resources and prioritization of improvements for more significant sources, EPA will continue to evaluate
and analyze data reported under EPA's GHGRP that would be useful to improve the emission estimates and
category-specific QC for the Lead Production source category, in particular considering completeness of reported
lead production given the reporting threshold. Particular attention will be made to ensuring time-series
consistency of the emissions estimates presented in future Inventory reports, consistent with IPCC and UNFCCC
guidelines. This is required as the facility-level reporting data from EPA's GHGRP, with the program's initial
requirements for reporting of emissions in calendar year 2010, are not available for all inventory years (i.e., 1990
through 2009) as required for this Inventory. In 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.89
4.23 Zinc Production (CRT Source Category
2C6)
Zinc production in the United States consists of both primary and secondary processes. Of the primary and
secondary processes currently used in the United States, only the electrothermic and Waelz kiln secondary
processes result in non-energy carbon dioxide (CO2) emissions (Viklund-White 2000). This reporting category (2C6)
includes emissions from the production of zinc. Per the IPCC methodological guidance, emissions from fuels
consumed for energy purposes during the production of zinc are accounted for as part of fossil fuel combustion in
the industrial end-use sector reported under the Energy chapter.
The majority of zinc produced in the United States is used for galvanizing. Galvanizing is a process where zinc
coating is applied to steel in order to prevent corrosion. Zinc is used extensively for galvanizing operations in the
automotive and construction industry. Zinc is also used in the production of zinc alloys and brass and bronze alloys
(e.g., brass mills, copper foundries, and copper ingot manufacturing). Zinc compounds and dust are also used, to a
lesser extent, by the agriculture, chemicals, paint, and rubber industries.
Production of zinc can be conducted with a range of pyrometallurgical (e.g., electrothermic furnace, Waelz kiln,
flame reactor, batch retorts, Pinto process, and PIZO process) and hydrometallurgical (e.g., hydrometallurgical
recovery, solvent recovery, solvent extraction-electrowinning, and electrolytic) processes. Hydrometallurgical
production processes are assumed to be non-emissive since no carbon is used in these processes (Sjardin 2003).
Primary production in the United States is conducted through the non-emissive electrolytic process, while
secondary techniques include the electrothermic and Waelz kiln processes, as well as a range of other processes.
Worldwide primary zinc production also employs a pyrometallurgical process using an Imperial Smelting Furnace;
however, this process is not used in the United States (Sjardin 2003).
In the electrothermic process, roasted zinc concentrate and secondary zinc products enter a sinter feed where
they are burned to remove impurities before entering an electric retort furnace. Metallurgical coke is added to the
electric retort furnace as a carbon-containing reductant. This concentration step, using metallurgical coke and high
temperatures, reduces the zinc oxides and produces vaporized zinc, which is then captured in a vacuum
condenser. This reduction process also generates non-energy CO2 emissions.
ZnO + C -» Zn(gas) + C02 (Reaction 1)
89 See http://www.ipcc-nggiD.iges.or.jp/public/tb/TFl Technical Bulletin l.pdf.
Industrial Processes and Product Use 4-137
-------
ZnO + CO -» Zn(gas) + C02 (Reaction 2)
In the Waelz kiln process, electric arc furnace (EAF) dust, which is captured during the recycling of galvanized steel,
enters a kiln along with a reducing agent (typically carbon-containing metallurgical coke). When kiln temperatures
reach approximately 1,100 to 1,200 degrees Celsius, zinc fumes are produced, which are combusted with air
entering the kiln. This combustion forms zinc oxide, which is collected in a baghouse or electrostatic precipitator,
and is then leached to remove chloride and fluoride. The use of carbon-containing metallurgical coke in a high-
temperature fuming process results in non-energy CO2 emissions. Through this process, approximately 0.33 metric
tons of zinc is produced for every metric ton of EAF dust treated (Viklund-White 2000).
In the flame reactor process, a waste feed stream, which can include EAF dust, is processed in a high-temperature
environment (greater than 2,000 °C) created by the combustion of natural gas or coal and oxygen-enriched air.
Volatile metals, including zinc, are forced into the gas phase and drawn into a combustion chamber, where air is
introduced and oxidation occurs. The metal oxide product is then collected in a dust collection system (EPA 1992).
In 2022, the only companies in the United States that used emissive technology to produce secondary zinc
products were Befesa Holding US Inc (Befesa) and Steel Dust Recycling (SDR). The secondary zinc facilities operated
by Befesa were acquired from American Zinc Recycling (AZR) (formerly "Horsehead Corporation") in 2021. PIZO
Operating Company, LLC (PIZO) operated a secondary zinc production facility that processed EAF dust in
Blytheville, AR from 2009 to 2012.
For Befesa, EAF dust is recycled in Waelz kilns at their Calumet, IL; Palmerton, PA; Rockwood, TN; and Barnwell, SC
facilities. The former AZR facility in Beaumont, TX processed EAF dust via flame reactor from 1993 through 2009
(AZR 2021, Horsehead 2014). These Waelz kiln and flame reactor facilities produce intermediate zinc products
(crude zinc oxide or calcine). Prior to 2014, most of output from these facilities were transported to their Monaca,
PA facility where the products were smelted into refined zinc using electrothermic technology. In April 2014, the
Monaca smelter was permanently closed and replaced by a new facility in Mooresboro, NC in 2014.
The Mooresboro facility uses a hydrometallurgical process (i.e., solvent extraction with electrowinning technology)
to produce zinc products, which is assumed to be non-emissive as described above. Production at the Mooresboro
facility was idled in April 2016 and re-started in March 2020 (Recycling Today 2020). Direct consumption of coal,
coke, and natural gas were replaced with electricity consumption (Horsehead 2012b). The Mooresboro facility uses
leaching and solvent extraction (SX) technology combined with electrowinning, melting, and casting technology. In
this process, Waelz Oxide (WOX) is first washed in water to remove soluble elements such as chlorine, potassium,
and sodium, and then is leached in a sulfuric acid solution to dissolve the contained zinc creating a pregnant liquor
solution (PLS). The PLS is then processed in a solvent extraction step in which zinc is selectively extracted from the
PLS using an organic solvent creating a purified zinc-loaded electrolyte solution. The loaded electrolyte solution is
then fed into the electrowinning process in which electrical energy is applied across a series of anodes and
cathodes submerged in the electrolyte solution causing the zinc to deposit on the surfaces of the cathodes. As the
zinc metal builds up on these surfaces, the cathodes are periodically harvested in order to strip the zinc from their
surfaces (Horsehead 2015).
SDR recycles EAF dust into intermediate zinc products using Waelz kilns and sells the intermediate products to
companies who smelt it into refined products.
Emissions of CO2 from zinc production in 2022 were estimated to be 0.9 MMT CO2 Eq. (947 kt CO2) (see Table
4-117). All 2022 CO2 emissions resulted from secondary zinc production processes. Emissions from zinc production
in the United States have increased overall since 1990 due to a gradual shift from non-emissive primary production
to emissive secondary production. In 2022, emissions were estimated to be 50 percent higher than they were in
1990. Emissions decreased 6 percent from 2021 levels.
Table 4-117: CO2 Emissions from Zinc Production (MMT CO2 Eq.)
Year
1990
2005
2018
2019
2020
2021
2022
Zinc Production
0.6
1.0 1
1.0
1.0
1.0
1.0
0.9
4-138 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
Table 4-118: CO2 Emissions from Zinc Production (kt CO2)
Year
1990
2005 2018
2019
2020
2021
2022
Zinc Production
632
1,030 I 999
1,026
977
1,007
947
U.S. zinc mine production increased by 9 percent in 2022 compared to 2021, due in part to higher mill throughput
and zinc ore grades at the Red Dog Mine in Alaska, the largest zinc mine in the United States. In 2022, United
States primary and secondary refined zinc production were estimated to total 220,000 metric tons (USGS 2023)
(see Table 4-119), remaining at approximately the same production level as in 2021. Secondary zinc production fell
to its lowest point in the time series in 2019, following the closure of the Monaca, PA smelter in 2014 and issues
with the AZR secondary zinc refinery in Mooresboro, NC. Secondary zinc production has increased significantly
since the reopening of the idled Mooresboro facility in March 2020 (USGS 2021; AZP 2021).
Table 4-119: Zinc Production (Metric Tons)
Year
1990
2005
2018
2019
2020
2021
2022
Primary
262,704
191,120
101,000
101,000
101,000
101,000
101,000
Secondary
95,708 i
156,000
15,000
14,000
79,000
119,000
119,000
Total
358,412
347,120
116,000
115,000
180,000
220,000
220,000
Note: Totals may not sum due to independent rounding.
Methodology and Time-Series Consistency
Emissions of CO2 emissions from zinc production90 using the electrothermic primary production and Waelz kiln
secondary production processes are calculated using a Tier 1 method from the 2006IPCC Guidelines, in accordance
with the IPCC methodological decision tree and available data (IPCC 2006). The Tier 1 equation used to estimate
emissions from zinc production is as follows:
Equation 4-17: 2006 IPCC Guidelines Tier 1: CO2 Emissions from Zinc Production (Equation
4.33)
Eco2 ~ Zn x EFdefaldt
where,
Ecoz = CO2 emissions from zinc production, metric tons
Zn = Quantity of zinc produced, metric tons
EFdefauit = Default emission factor, metric tons CCh/metric ton zinc produced
The Tier 1 emission factors provided by IPCC for Waelz kiln-based secondary production were derived from
metallurgical coke consumption factors and other data presented in Vikland-White (2000). These coke
consumption factors as well as other inputs used to develop the Waelz kiln emission factors are shown below. IPCC
does not provide an emission factor for electrothermic processes due to limited information; therefore, the Waelz
kiln-specific emission factors were also applied to zinc produced from electrothermic processes. Starting in 2014,
refined zinc produced in the United States used hydrometallurgical processes and is assumed to be non-emissive.
90 EPA has not integrated aggregated facility-level Greenhouse Gas Reporting Program (GHGRP) information to inform these
estimates. The aggregated information (e.g., activity data and emissions) associated with Zinc Production did not meet criteria
to shield underlying confidential business information (CBI) from public disclosure.
Industrial Processes and Product Use 4-139
-------
For Waelz kiln-based production, IPCC recommends the use of emission factors based on EAF dust consumption, if
possible, rather than the amount of zinc produced since the amount of reduction materials used is more directly
dependent on the amount of EAF dust consumed. Since only a portion of emissive zinc production facilities
consume EAF dust, the emission factor based on zinc production is applied to the non-EAF dust consuming
facilities, while the emission factor based on EAF dust consumption is applied to EAF dust consuming facilities.
The Waelz kiln emission factor based on the amount of zinc produced was developed based on the amount of
metallurgical coke consumed for non-energy purposes per ton of zinc produced (i.e., 1.19 metric tons coke/metric
ton zinc produced) (Viklund-White 2000), and the following equation:
Equation 4-18: Waelz Kiln CO2 Emission Factor for Zinc Produced
1.19 metric tons coke 0.85 metric tons C 3.67 metric tons C07 3.70 metric tons C07
EFU, .... 2
metric tons zinc metric tons coke metric tons C metric tons zinc
Refined zinc production levels for AZR's Monaca, PA facility (utilizing electrothermic technology) were available
from the company for years 2005 through 2013 (Horsehead 2008, 2011, 2012, 2013, and 2014). The Monaca
facility was permanently shut down in April 2014 and replaced by AZR's new facility in Mooresboro, NC. The new
facility uses hydrometallurgical process to produce refined zinc products. Hydrometallurgical production processes
are assumed to be non-emissive since no carbon is used in these processes (Sjardin 2003).
Metallurgical coke consumption for non-EAF dust consuming facilities for 1990 through 2004 were extrapolated
using the percentage change in annual refined zinc production at secondary smelters in the United States, as
provided by the U.S. Geological Survey (USGS) Minerals Yearbook: Zinc (USGS 1994 through 2006). Metallurgical
coke consumption for 2005 through 2013 were based on the secondary zinc production values obtained from the
Horsehead Corporation Annual Report Form 10-K: 2005 through 2008 from the 2008 10-K (Horsehead Corp 2009);
2009 and 2010 from the 2010 10-K (Horsehead Corp. 2011); and 2011 through 2013 from the associated 10-K
(Horsehead Corp. 2012a, 2013, 2014). Metallurgical coke consumption levels for 2014 and later were zero due to
the closure of the AZR (formerly "Horsehead Corporation") electrothermic furnace facility in Monaca, PA. The
secondary zinc produced values for each year were then multiplied by the 3.70 metric tons CCh/metric ton zinc
produced emission factor to develop CO2 emission estimates for the AZR electrothermic furnace facility.
The Waelz kiln emission factor based on the amount of EAF dust consumed was developed based on the amount
of metallurgical coke consumed per ton of EAF dust consumed (i.e., 0.4 metric tons coke/metric ton EAF dust
consumed) (Viklund-White 2000), and the following equation:
Equation 4-19: Waelz Kiln CO2 Emission Factor for EAF Dust Consumed
OA metric tons coke 0.85 metric tons C 3.67 metric tons C02 1.24 metric tons C02
EFuAp Dust — ! : ! ~ X ¦ ; ¦ " X ¦
metric tons EAF Dust metric tons coke metric tons C metric tons EAF Dust
Metallurgical coke consumption for EAF dust consuming facilities for 1990 through 2022 were calculated based on
the values of EAF dust consumed. The total amount of EAF dust consumed by the Waelz kilns currently operated
by Befesa was available from AZR (formerly "Horsehead Corporation") in financial reports for years 2006 through
2015 (Horsehead 2007, 2008, 2010a, 2011, 2012a, 2013, 2014, 2015, and 2016), from correspondence with AZR for
2016 through 2019 (AZR 2020), and from correspondence with Befesa for 2020 through 2022 (Befesa 2022, 2023).
The EAF dust consumption values for each year were then multiplied by the 1.24 metric tons CCh/metric ton EAF
dust consumed emission factor to develop CO2 emission estimates for Befesa's Waelz kiln facilities.
The amount of EAF dust consumed by SDR and their total production capacity were obtained from SDR's facility in
Alabama for the years 2011 through 2022 (SDR 2012, 2014, 2015, 2017, 2018, 2021, 2022, 2023). The SDR facility
has been operational since 2008, underwent expansion in 2011 to include a second unit (operational since early- to
mid-2012), and expanded its capacity again in 2017 (SDR 2018). Annual consumption data for SDR was not publicly
4-140 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
available for the years 2008, 2009, and 2010. These data were estimated using data for AZR's Waelz kilns for 2008
through 2010 (Horsehead 2007, 2008, 2010a, 2010b, 2011). Annual capacity utilization ratios were calculated using
AZR's annual consumption and total capacity for the years 2008 through 2010. AZR's annual capacity utilization
ratios were multiplied with SDR's total capacity to estimate SDR's consumption for each of the years, 2008 through
2010 (SDR 2013). The 1.24 metric tons CCh/metric ton EAF dust consumed emission factor was then applied to
SDR's estimated EAF dust consumption to develop CO2 emission estimates for those Waelz kiln facilities.
PIZO's facility in Arkansas was operational from 2009 to 2012 (PIZO 2021). The amount of EAF dust consumed by
PIZO's facility for 2009 through 2012 was not publicly available. EAF dust consumption for PIZO's facility for 2009
and 2010 were estimated by calculating annual capacity utilization of AZR's Waelz kilns and multiplying this
utilization ratio by PIZO's total capacity (PIZO 2012). EAF dust consumption for PIZO's facility for 2011 through
2012 were estimated by applying the average annual capacity utilization rates for AZR and SDR (Grupo PROMAX)
to PIZO's annual capacity (Horsehead 2012; SDR 2012; PIZO 2012). The 1.24 metric tons C02/metric ton EAF dust
consumed emission factor was then applied to PIZO's estimated EAF dust consumption to develop CO2 emission
estimates for those Waelz kiln facilities.
The production and use of coking coal for zinc production is adjusted for within the Energy chapter as this fuel was
consumed during non-energy related activities. Additional information on the adjustments made within the Energy
sector for non-energy use of fuels is described in both the Methodology section of CO2 from Fossil Fuel
Combustion (3.1 Fossil Fuel Combustion (CRT Source Category 1A)) and Annex 2.1, Methodology for Estimating
Emissions of CO2 from Fossil Fuel Combustion.
Beginning with the 2017 USGS Minerals Commodity Summary: Zinc, United States primary and secondary refined
zinc production were reported as one value, total refined zinc production. Prior to this publication, primary and
secondary refined zinc production statistics were reported separately. For years 2016 through 2022, only one
facility produced primary zinc. Primary zinc produced from this facility was subtracted from the USGS 2016 to 2022
total zinc production statistic to estimate secondary zinc production for these years.
Methodological approaches were applied to the entire time series to ensure consistency in emissions from 1990
through 2022.
Uncertainty
There is uncertainty associated with the amount of EAF dust consumed in the United States to produce secondary
zinc using emission-intensive Waelz kilns. The estimate for the total amount of EAF dust consumed in Waelz kilns is
based on combining the totals for (1) the EAF dust consumption value obtained for the kilns currently operated by
Befesa (and formerly operated by AZR or Horsehead Corporation) and (2) an EAF dust consumption value obtained
from the Waelz kiln facility operated by SDR. For the 1990 through 2015 estimates, EAF dust consumption values
for the kilns currently operated by Befesa were obtained from annual financial reports to the Securities and
Exchange Commission (SEC) by AZR. In 2016, AZR reorganized as a private company and ceased providing annual
reports to the SEC (Recycling Today 2017). EAF dust consumption values for subsequent years from the Befesa
kilns and SDR have been obtained from personal communication with facility representatives. Since actual EAF
dust consumption information is not available for PIZO's facility (2009 through 2010) and SDR's facility (2008
through 2010), the amount is estimated by multiplying the EAF dust recycling capacity of the facility (available
from the company's website) by the capacity utilization factor for AZR (which was available from Horsehead
Corporation financial reports).The EAF dust consumption for PIZO's facility for 2011 through 2012 was estimated
by multiplying the average capacity utilization factor developed from AZR and SDR's annual capacity utilization
rates by PIZO's EAF dust recycling capacity. Therefore, there is uncertainty associated with the assumption used to
estimate PIZO's annual EAF dust consumption values for 2009 through 2012 and SDR's annual EAF dust
consumption values for 2008 through 2010. EPA uses an uncertainty range of ±5 percent for these EAF dust
consumption data inputs, based upon expert elicitation from the USGS commodity specialist. Per this expert
judgment, a normal probability density function was assigned for EAF dust consumption data inputs.
Industrial Processes and Product Use 4-141
-------
There is also uncertainty associated with the emission factors used to estimate CO2 emissions from secondary zinc
production processes. The Waelz kiln emission factors are based on materials balances for metallurgical coke and
EAF dust consumed as provided by Viklund-White (2000). Therefore, the accuracy of these emission factors
depends upon the accuracy of these materials balances. Data limitations prevented the development of emission
factors for the electrothermic process. Therefore, emission factors for the Waelz kiln process were applied to both
electrothermic and Waelz kiln production processes. Consistent with the ranges in Table 4.25 of the 2006IPCC
Guidelines, EPA assigned an uncertainty range of ±20 percent for the Tier 1 Waelz kiln emission factors, which are
provided by Viklund-White in the form of metric tons of coke per metric ton of EAF dust consumed and metric tons
of coke per metric ton of zinc produced. In order to convert coke consumption rates to CO2 emission rates, values
for the heat and carbon content of coke were obtained from Table 4.2 - Tier 2 of the 2006 IPCC Guidelines. An
uncertainty range of ±10 percent was assigned to these coke data elements, and using the suggested uncertainty
provided in Table 4.25, Tier 2 - National Reducing Agent & Process Materials Data of the 2006 IPCC Guidelines is
appropriate based on expert judgment (RTI 2023). Per this expert judgment, a triangular probability density
function was assigned for emission factors and the heat and carbon content of coke.
The results of the Approach 2 quantitative uncertainty analysis are summarized in Table 4-120. Zinc production
CO2 emissions from 2022 were estimated to be between 0.8 and 1.1 MMT CO2 Eq. at the 95 percent confidence
level. This indicates a range of approximately 18 percent below and 20 percent above the emission estimate of 0.9
MMTCO2 Eq.
Table 4-120: Approach 2 Quantitative Uncertainty Estimates for CO2 Emissions from Zinc
Production (MMT CO2 Eq. and Percent)
Source Gas
2022 Emission Estimate
Uncertainty Range Relative to Emission Estimate-1
(MMT CO . Eq.)
(MMT CO . Eq.)
(%)
Lower Upper
Lower Upper
Bound Bound
Bound Bound
Zinc Production C02
0.9
1
1
00
O
-18% +20%
a Range of emission estimates predicted by Monte Carlo stochastic simulation for a 95 percent confidence interval.
QA/QC and Verification
General quality assurance/quality control (QA/QC) procedures were applied consistent with the U.S. Inventory
QA/QC plan, which is in accordance with Volume 1, Chapter 6 of 2006 IPCC Guidelines as described in the
introduction of the IPPU chapter (see Annex 8 for more details).
Recalculations Discussion
Recalculations were performed for the year 2021 based on updated EAF dust consumption data. Compared to the
previous Inventory, 2021 emissions from zinc production increased by 4 percent (38 kt CO2).
Planned Improvements
Pending resources and prioritization of improvements for more significant sources, EPA will continue to evaluate
and analyze data reported under EPA's GHGRP that would be useful to improve the emission estimates and
category-specific QC for zinc production, in particular considering completeness of reported zinc production given
the reporting threshold. Given the small number of facilities in the United States, particular attention will be made
to risks for disclosing CBI and ensuring time-series consistency of the emissions estimates presented in future
Inventory reports, consistent with IPCC and UNFCCC guidelines. This is required as the facility-level reporting data
from EPA's GHGRP, with the program's initial requirements for reporting of emissions in calendar year 2010, are
not available for all inventory years (i.e., 1990 through 2009) as required for this Inventory. In implementing
4-142 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
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.91 This is a long-term planned improvement, and EPA is still
assessing the possibility of including this improvement in future Inventory reports.
4.24 Electronics Industry (CRT Source
Category 2E)
The electronics industry uses multiple greenhouse gases in its manufacturing processes. In semiconductor
manufacturing, these include long-lived fluorinated greenhouse gases used for plasma etching and chamber
cleaning (CRT Source Category 2E1), fluorinated heat transfer fluids used for temperature control and other
applications (CRT Source Category 2E4), and nitrous oxide (N2O) used to produce thin films through chemical vapor
deposition and in other applications (reported under CRT Source Category 2H3). Similar to semiconductor
manufacturing, the manufacturing of micro-electro-mechanical systems (MEMS) devices (reported under CRT
Source Category 2E5 Other) and photovoltaic (PV) cells (CRT Source Category 2E3) requires the use of multiple
long-lived fluorinated greenhouse gases for various processes.
The gases most commonly employed in the electronics industry are trifluoromethane (hydrofluorocarbon (HFC)-23
or CHF3), perfluoromethane (CF4), perfluoroethane (C2F6), nitrogen trifluoride (NF3), and sulfur hexafluoride (SFs),
although other fluorinated compounds such as perfluoropropane (C3F8) and perfluorocyclobutane (c-C4Fs) are also
used. The exact combination of compounds is specific to the process employed.
In addition to emission estimates for these seven commonly used fluorinated gases, this Inventory contains
emissions estimates for N2O and other HFCs and unsaturated, low-GWP PFCs including CsFs, C4F6, HFC-32, HFC-41,
and HFC-134a. These additional HFCs and PFCs are emitted from etching and chamber cleaning processes in much
smaller amounts, accounting for 0.02 percent of emissions (in CO2 Eq.) from these processes.
For semiconductors, a single 300 mm silicon wafer that yields between 400 to 600 semiconductor products
(devices or chips) may require more than 100 distinct fluorinated-gas-using process steps, principally to deposit
and pattern dielectric films. Plasma etching (or patterning) of dielectric films, such as silicon dioxide and silicon
nitride, is performed to provide pathways for conducting material to connect individual circuit components in each
device. The patterning process uses plasma-generated fluorine atoms, which chemically react with exposed
dielectric film to selectively remove the desired portions of the film. The material removed as well as undissociated
fluorinated gases flow into waste streams and, unless emission abatement systems are employed, into the
atmosphere. Plasma enhanced chemical vapor deposition (PECVD) chambers, used for depositing dielectric films,
are cleaned periodically using fluorinated and other gases. During the cleaning cycle the gas is converted to
fluorine atoms in plasma, which etches away residual material from chamber walls, electrodes, and chamber
hardware. Undissociated fluorinated gases and other products pass from the chamber to waste streams and,
unless abatement systems are employed, into the atmosphere.
In addition to emissions of unreacted gases, some fluorinated compounds can also be transformed in the plasma
processes into different fluorinated compounds which are then exhausted, unless abated, into the atmosphere.
For example, when C2F6 is used in cleaning or etching, CF4 is typically generated and emitted as a process
byproduct. In some cases, emissions of the byproduct gas can rival or even exceed emissions of the input gas, as is
the case for NF3 used in remote plasma chamber cleaning, which often generates CF4 as a byproduct.
91 See http://www.ipcc-negip.iges.or.jp/public/tb/TFl Technical Bulletin l.pdf and the 2019 Refinement, Volume 1, Chapter 2,
Section 2.3, Use of Facility Data in Inventories at https://www.ipcc-
nggip.iges.or.jp/public/2019rf/pdf/l Volumel/19R VI Ch02 DataCollection.pdf.
Industrial Processes and Product Use 4-143
-------
Besides dielectric film etching and PECVD chamber cleaning, much smaller quantities of fluorinated gases are used
to etch polysilicon films and refractory metal films like tungsten.
Nitrous oxide is used in manufacturing semiconductor devices to produce thin films by CVD and nitridation
processes as well as for N-doping of compound semiconductors and reaction chamber conditioning (Doering
2000).
Liquid perfluorinated compounds are also used as heat transfer fluids (F-HTFs) for temperature control, device
testing, cleaning substrate surfaces and other parts, and soldering in certain types of semiconductor
manufacturing production processes. Leakage and evaporation of these fluids during use is a source of fluorinated
gas emissions (EPA 2006). Unweighted F-HTF emissions consist primarily of perfluorinated amines,
hydrofluoroethers, perfluoropolyethers (specifically, PFPMIEs), and perfluoroalkylmorpholines. Three percent or
less consist of HFCs, PFCs, and SF6 (where PFCs are defined as compounds including only carbon and fluorine). With
the exceptions of the hydrofluoroethers and most of the HFCs, all of these compounds are very long-lived in the
atmosphere and have global warming potentials (GWPs) near 10,000.92
MEMS and photovoltaic cell manufacturing require thin film deposition and etching of material with a thickness of
one micron or more, so the process is less intricate and complex than semiconductor manufacturing. The
manufacturing process is different than semiconductors, but generally employs similar techniques. Like
semiconductors, MEMS and photovoltaic cell manufacturers use fluorinated compounds for etching, cleaning
reactor chambers, and temperature control. CF4, SF6, and the Bosch process (which consists of alternating steps of
SFs and C4F8) are used to manufacture MEMS (EPA 2010). Photovoltaic cell manufacturing predominately uses CF4,
to etch crystalline silicon wafers, and C2F6 or NF3 during chamber cleaning after deposition of SiNx films (IPCC
2006), although other F-GHGs may be used. Similar to semiconductor manufacturing, both MEMS and photovoltaic
cell manufacturing use N2O in depositing films and other manufacturing processes. MEMS and photovoltaic
manufacturing may also employ HTFs for cooling process equipment (EPA 2010).
Emissions from all fluorinated greenhouse gases (including F-HTFs) and N2O for semiconductors, MEMS and
photovoltaic cells manufacturing are presented in Table 4-121 below for the years 1990, 2005, and the period
2018 to 2022. The rapid growth of the electronics industry and the increasing complexity (growing number of
layers and functions)93 of electronic products led to an increase in emissions of 152 percent between 1990 and
1999, when emissions peaked at 8.4 MMT CO2 Eq. Emissions began to decline after 1999, reaching a low point in
2009 before rebounding to 2006 emission levels and more or less plateauing at the current level, which represents
a 44 percent decline from 1999 to 2022. Together, industrial growth, adoption of emissions reduction technologies
(including but not limited to abatement technologies) and shifts in gas usages resulted in a net increase in
emissions of approximately 41 percent between 1990 and 2022. Total emissions from semiconductor manufacture
in 2022 were lower than 2021 emissions, decreasing by 7.5 percent, largely due to a large decrease in SF6
emissions. The decrease in SF6 are seen in facilities that manufacture 200 mm wafer size that do not have
abatement systems installed as well as 300 mm wafer size that have abatement systems installed.
For U.S. semiconductor manufacturing in 2022, total CC>2-equivalent emissions of all fluorinated greenhouse gases
and N2O from deposition, etching, and chamber cleaning processes were estimated to be 4.6 MMT CO2 Eq. This is a
92 The GWP of PFPMIE, a perfluoropolyether used as an F-HTF, is included in the IPCC Fourth Assessment Report with a value of
10,300. The GWPs of the perfluorinated amines and perfluoroalkylmorpholines that are used as F-HTFs have not been
evaluated in the peer-reviewed literature. However, evaluations by the manufacturer indicate that their GWPs are near 10,000
(78 FR 20632), which is expected given that these compounds are both saturated and fully fluorinated. EPA assigns a default
GWP of 10,000 to compounds that are both saturated and fully fluorinated and that do not have chemical-specific GWPs in
either the Fourth or the Fifth Assessment Reports.
93 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.
4-144 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
decrease in emissions from 1999 of 45 percent, and an increase in emissions from 1990 of 40 percent. These
trends are driven by the above stated reasons.
Photovoltaic cell and MEMS manufacturing emissions of all fluorinated greenhouse gases are in Table 4-121. While
EPA has developed a simple methodology to estimate emissions from non-reporters and to back-cast emissions
from these sources for the entire time series, there is very high uncertainty associated with these emission
estimates.
The emissions reported by facilities manufacturing MEMS included emissions of C2F6, C3F8, C-C4F8, CF4, HFC-23, NF3,
N2O and SFs,94 and were equivalent to only 0.110 percent to 0.264 percent of the total reported emissions from
electronics manufacturing in 2011 to 2022. F-GHG emissions, the primary type of emissions for MEMS, ranged
from 0.0003 to 0.012 MMT CO2 Eq. from 1991 to 2022. Based upon information in the World Fab Forecast (WFF), it
appears that some GHGRP reporters that manufacture both semiconductors and MEMS are reporting their
emissions as only from semiconductor manufacturing (GHGRP reporters must choose a single classification per
fab). Emissions from non-reporters have not been estimated.
Total CC>2-equivalent emissions from manufacturing of photovoltaic cells were estimated to range from 0.0003
MMT CO2 Eq. to 0.0330 MMT CO2 Eq. from 1998 to 2022 and were equivalent to between 0.003 percent to 0.77
percent of the total reported emissions from electronics manufacturing. F-GHG emissions, the primary type of
emissions for photovoltaic cells, ranged from 0.0003 to 0.032 MMT CO2 Eq. from 1998 to 2022. Emissions from
manufacturing of photovoltaic cells were estimated using an emission factor developed from reported data from a
single manufacturer between 2015 and 2016. This emission factor was then applied to production capacity
estimates from non-reporting facilities. Reported emissions from photovoltaic cell manufacturing consisted of CF4,
C2F6, c-C4F8, CHFs, NFb, and N2O.95
Emissions of F-HTFs, grouped by HFCs, PFCs or SF6 are presented in Table 4-121. Emissions of F-HTFs that are not
HFCs, PFCs or SF6 are not included in inventory totals and are included for informational purposes only.
Since reporting of F-HTF emissions began under EPA's GHGRP in 2011, total F-HTF emissions (reported and
estimated non-reported) have fluctuated between 0.4 MMT CO2 Eq. and 0.9 MMT CO2 Eq., with an overall
declining trend between 2011 to 2022. An analysis of the data reported to EPA's GHGRP indicates that F-HTF
emissions account for anywhere between 9 percent and 17 percent of total annual emissions (F-GHG, N2O and F-
HTFs) from semiconductor manufacturing.96 Table 4-123 shows F-HTF emissions in tons by compound group based
on reporting to EPA's GHGRP and the interpolated share of F-HTF emissions to F-GHG emissions for select years
prior to reporting. 97
94 Gases not reported by MEMS manufacturers to the GHGRP are currently listed as "NE" in the CRTs. Since no facilities report
using these gases, emissions of these gases are not estimated for this sub-sector. However, there is insufficient data to
definitively conclude that they are not used by non-reporting facilities.
95 Gases not reported by PV manufacturers to the GHGRP are currently listed as "NE" in the CRTs. Since no facilities report
using these gases, emissions of these gases are not estimated for this sub-sector. However, there is insufficient data to
definitively conclude that they are not used by non-reporting facilities.
96 Emissions data for HTFs (in tons of gas) from the semiconductor industry from 2011 through 2022 were obtained from the
EPA GHGRP annual facility emissions reports.
97 Many fluorinated heat transfer fluids consist of perfluoropolymethylisopropyl ethers (PFPMIEs) of different molecular
weights and boiling points that are distilled from a mixture. "BP 200 °C" (and similar terms below) indicate the boiling point of
the fluid in degrees Celsius. For more information, see https://www.reeulations.gov/document?D=EPA-HQ-QAR-2009-0927-
0276.
Industrial Processes and Product Use 4-145
-------
Table 4-121: PFC, HFC, SF6, NF3, and N2O Emissions from Electronics Industry (MMT CO2 Eq.)
Year
1990
2005
2018
2019
2020
2021
2022
cf4
0.8
1.0
1.6
1.5
1.6
1.7
1.6
c2f6
1-8 I
CO
T—1
1.1
0.9
0.8
0.9
0.9
CsFs
+
0.1
0.1
0.1
0.1
0.1
0.1
C4Fs
O
b
illlH
0.1 ¦¦
0.1
0.1
0.1
0.1
0.1
HFC-23
0.2
0.2
0.3
0.3
0.3
0.4
0.3
sf6
O
Ln
iiimi
00
0
0.8
0.8
0.8
0.9
0.7
nf3
+
0.4
0.6
0.5
0.6
0.7
0.6
C4F6
+ iiii!
+ I!!!!!
+
+
+
+
+
CsFs
+
+
+
+
+
+
+
ch2f2
mm!
+ I
+::
+
+
+
+
+
ch3f
+
+
+
+
+
+
+
CH2FCF3
+ III1
mm!
+
+
+
+
0
Total Semiconductors
3.3
4.3
4.5
4.2
4.3
4.7
4.3
cf4
0.0 1
+ ::
+
+
+
+
+
c2f6
0.0
+
+
+
+
+
+
CsFs
0
0
lllllll
+
0.0
0.0
0.0
0.0
0.0
C4Fs
0.0
+
+
+
+
+
+
HFC-23
0
0
+ i
+
+
+
+
+
sf6
0.0
+
+
+
+
+
+
nf3
0.0 ;;;;;;
¦¦in
0
0
+
+
+
+
+
Total MEMS 0.0 ill +_ + + + + +
cf4
0.0
+ 1
+
+
+
+
+
c2f6
0.0
+
+
+
+
+
+
C4Fs
0
0
lllllll
+
+
+
+
+
HFC-23
0.0
+
+
+
+
+
+
sf6
IBBII—
O
O
0.0
0.0
0.0
0.0
0.0
0.0
nf3
0.0
0.0
0.0
0.0
0.0
0.0
0.0
Total PV
0.0
+
+
+
+
+
+
N20 (Semiconductors)
+
0.1
0.2
0.2
0.3
0.3
0.3
N20 (MEMS)
0
0
+
+
+
+
+
+
N20 (PV)
0.0
+
+
+
+
+
+
Total N20
+
0.1
0.2
0.2
0.3
0.3
0.3
HFC, PFCandSFe F-HTFs
0.0
+
+
+
+
+
+
Total Electronics Industry
3.3
4.5
4.8
4.4
4.5
4.8
4.7
+ Does not exceed 0.05 MMT C02 Eq.
Table 4-122: PFC, HFC, SF6, NF3, and N2O Emissions from Semiconductor Manufacture (Metric
Tons)
Year
1990
2005
2018
2019
2020
2021
2022
cf4
114.8
145.3
237.0
221.1
237.4
252.8
240.9
c2f6
160.0 !
163.4 jji
98.7
83.2
76.0
80.0
80.8
C3Fs
0.4
7.3
12.2
9.9
9.4
11.0
13.4
C4Fs
i»—
O
O
10-9 ¦
6.2
5.7
6.1
6.6
5.9
HFC-23
14.6
14.1
26.4
25.6
26.5
32.6
24.9
sf6
21.7 5
33.4;;
33.8
32.8
33.6
39.8
31.1
nf3
2.8
26.2
34.4
33.5
37.2
40.9
39.1
c4f6
0.7
0.9 |
0.8
0.9
0.8
1.2
0.8
C5Fs
0.5
0.6 ~
0.5
0.4
0.4
0.4
0.4
4-146 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
ch2f2
CHsF
CH2FCF3
N20
+ Does not exceed 0.05 MT.
Table 4-123: F-HTF Emissions from Electronics Manufacture by Compound Group (kt CO2 Eq.)
Year
1990
2005
2018
2019
2020
2021
2022
HFCs
0.0
0.9
2.7
1.1
0.9
1.2
1.5
PFCs
o.o mm
illlM
CO
CO
10.0
8.4
8.0
5.7
7.3
sf6
0.0
5.6
13.2
6.0
13.2
9.4
4.1
HFEs
0.0
39.4
4.6
1.3
5.5
4.0
14.2
PFPMIEs
0.0
110.5
183.3
172.6
154.8
155.4
162.8
Perfluoalkylromorpholines
O
O
(1IHHI!
66.4 =
58.5
56.4
62.8
55.9
19.4
Perfluorotrialkylamines
0.0
209.9
414.7
363.8
391.5
382.2
233.5
Total F-HTFs
0.0
436.5
687.0
608.4
636.6
613.6
443.0
Notes: Emissions of F-HTFs that are not HFCs, PFCs or SF6 are not included in inventory totals and are
included for informational purposes only. Emissions presented for informational purposes include HFEs,
PFPMIEs, perfluoroalkylmorpholines, and perfluorotrialkylamines. Totals may not sum due to
independent rounding.
Methodology and Time-Series Consistency
Emissions are based on data reported through Subpart I, Electronics Manufacture, of EPA's GHGRP, semiconductor
manufacturing Partner-reported emissions data received through EPA's PFC98 Reduction/Climate Partnership,
EPA's PFC Emissions Vintage Model (PEVM)—a model that estimates industry emissions from etching and chamber
cleaning processes in the absence of emission control strategies (Burton and Beizaie 2001)"—and estimates of
industry activity (i.e., total manufactured layer area and manufacturing capacity). The availability and applicability
of reported emissions data from the EPA Partnership and EPA's GHGRP and activity data differ across the 1990
through 2022 time series. Consequently, fluorinated greenhouse gas (F-GHG) emissions from etching and chamber
cleaning processes for semiconductors were estimated using seven distinct methods, one each for the periods
1990 through 1994,1995 through 1999, 2000 through 2006, 2007 through 2010, 2011 and 2012, 2013 and 2014,
and 2015 through 2022. Nitrous oxide emissions were estimated using five distinct methods, one each for the
period 1990 through 1994,1995 through 2010, 2011 and 2012, 2013 and 2014, and 2015 through 2022. The
methodology discussion below for these time periods focuses on semiconductor emissions from etching, chamber
cleaning, and uses of N2O. Other emissions for MEMS, photovoltaic cells, and HTFs were estimated using the
approaches described immediately below.
MEMS
GHGRP-reported emissions (F-GHG and N2O) from the manufacturing of MEMS are available for the years 2011 to
2022. Emissions from manufacturing of MEMS for years prior to 2011 were calculated by linearly interpolating
emissions between 1990 (at zero MMT CO2 Eq.) and 2011, the first year where emissions from manufacturing of
MEMS was reported to the GHGRP. Based upon information in the World Fab Forecast (WFF), it appears that some
GHGRP reporters that manufacture both semiconductors and MEMS are reporting their emissions as only from
0.6 I 0.8 j 0.9 1.0 1.1 1.1 1.0
1.4 1.8 2.6 2.5 3.1 3.2 2.4
+ 5 + g + + + + 0.0
135.9 463.3 881.5 822.2 1,023.0 1,083.6 1,111.2
98 In the context of the EPA Partnership and PEVM, PFC refers to perfluorocompounds, not perfluorocarbons.
99 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.
Industrial Processes and Product Use 4-147
-------
semiconductor manufacturing; however, emissions from MEMS manufacturing are likely being included in
semiconductor totals. Emissions were not estimated for non-reporters.
Photovoltaic Cells
GHGRP-reported emissions (F-GHG and N2O) from the manufacturing of photovoltaic cells are available for 2011,
2012, 2015, and 2016 from two manufacturers. EPA estimates the emissions from manufacturing of PVs from non-
reporting facilities by multiplying the estimated capacity of non-reporters by a calculated F-GHG emission factor
and N2O emission factor based on GHGRP reported emissions from the manufacturer (in MMT CO2 Eq. per
megawatt) that reported emissions in 2015 and 2016. This manufacture's emissions are expected to be more
representative of emissions from the sector, as their emissions were consistent with consuming only CF4for
etching processes and are a large-scale manufacturer, representing 28 percent of the U.S. production capacity in
2016. The second photovoltaic manufacturer only produced a small fraction of U.S. production (<4 percent). They
also reported the use of NF3 in remote plasma cleaning processes, which does not have an emission factor in Part
98 for PV manufacturing, requiring them to report emissions equal to consumption. The total F-GHG emissions
from non-reporters are then disaggregated into individual gases using the gas distribution from the 2015 to 2016
manufacturer. Manufacturing capacities in megawatts were drawn from DisplaySearch, a 2015 Congressional
Research Service Report on U.S. Solar Photovoltaic Manufacturing, and self-reported capacity by GHGRP reporters.
EPA estimated that during the 2015 to 2016 period, 28 percent of manufacturing capacity in the United States was
represented through reported GHGRP emissions. Capacities are estimated for the full time series by linearly scaling
the total U.S. capacity between zero in 1997 to the total capacity reported of crystalline silicon (c-Si) PV
manufacturing in 2000 in DisplaySearch and then linearly scaling between the total capacity of c-Si PV
manufacturing in DisplaySearch in 2009 to the total capacity of c-Si PV manufacturing reported in the
Congressional Research Service report in 2012. Capacities were held constant for non-reporters for 2012 to 2019.
In 2020, non-reporter capacity declined due to the closure of several PV manufacturing plants. This capacity was
held constant for 2021 and 2022. Average emissions per MW from the GHGRP reporter in 2015 and 2016 were
then applied to the total capacity prior to 2015. Emissions for 2014 from the GHGRP reporter that reported in 2015
and 2016 were scaled to the number of months open in 2014. For 1998 through 2022, emissions per MW
(capacity) from the GHGRP reporter were applied to the non-reporters. For 2017 through 2022, there are no
reported PV emissions. Therefore, emissions were estimated using the EPA-derived emission factor and estimated
manufacturing capacity from non-reporters only.
HTFs
Facility emissions of F-HTFs from semiconductor manufacturing are reported to EPA under its GHGRP and are
available for the years 2011 through 2022. EPA estimates the emissions of F-HTFs from non-reporting
semiconductor facilities by calculating the ratio of GHGRP-reported fluorinated HTF emissions to GHGRP reported
F-GHG emissions from etching and chamber cleaning processes, and then multiplying this ratio by the F-GHG
emissions from etching and chamber cleaning processes estimated for non-reporting facilities. Fluorinated HTF use
in semiconductor manufacturing is assumed to have begun in the early 2000s and to have gradually displaced
other HTFs (e.g., de-ionized water and glycol) in semiconductor manufacturing (EPA 2006). For time-series
consistency, EPA interpolated the share of F-HTF emissions to F-GHG emissions between 2000 (at 0 percent) and
2011 (at 17 percent) and applied these shares to the unadjusted F-GHG emissions during those years to estimate
the emissions.
Semiconductors
1990 through 1994
From 1990 through 1994, Partnership data were unavailable, and emissions were modeled using PEVM (Burton
4-148 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
and Beizaie 2001).100 The 1990 to 1994 emissions are assumed to be uncontrolled, since reduction strategies such
as chemical substitution and abatement were yet to be developed.
PEVM is based on the recognition that fluorinated greenhouse gas emissions from semiconductor manufacturing
vary with: (1) the number of layers that comprise different kinds of semiconductor devices, including both silicon
wafer and metal interconnect layers, and (2) silicon consumption (i.e., the area of semiconductors produced) for
each kind of device. The product of these two quantities, Total Manufactured Layer Area (TMLA), constitutes the
activity data for semiconductor manufacturing. PEVM also incorporates an emission factor that expresses
emissions per unit of manufactured layer-area. Emissions are estimated by multiplying TMLA by this emission
factor.
PEVM incorporates information on the two attributes of semiconductor devices that affect the number of layers:
(1) linewidth technology (the smallest manufactured feature size),101 and (2) product type (discrete, memory or
logic).102 For each linewidth technology, a weighted average number of layers is estimated using VLSI product-
specific worldwide silicon demand data in conjunction with complexity factors (i.e., the number of layers per
Integrated Circuit (IC) specific to product type (Burton and Beizaie 2001; ITRS 2007). PEVM derives historical
consumption of silicon (i.e., square inches) by linewidth technology from published data on annual wafer starts
and average wafer size (VLSI Research, Inc. 2012).
The emission factor in PEVM is the average of four historical emission factors, each derived by dividing the total
annual emissions reported by the Partners for each of the four years between 1996 and 1999 by the total TMLA
estimated for the Partners in each of those years. Over this period, the emission factors varied relatively little (i.e.,
the relative standard deviation for the average was 5 percent). Since Partners are believed not to have applied
significant emission reduction measures before 2000, the resulting average emission factor reflects uncontrolled
emissions and hence may be use here to estimate 1990 through 1994 emissions. The emission factor is used to
estimate U.S. uncontrolled emissions using publicly available data on world (including U.S.) silicon consumption.
As it was assumed for this time period that there was no consequential adoption of fluorinated-gas-reducing
measures, a fixed distribution of fluorinated-gas use was assumed to apply to the entire U.S. industry to estimate
gas-specific emissions. This distribution was based upon the average fluorinated-gas purchases made by
semiconductor manufacturers during this period and the application of IPCC default emission factors for each gas
(Burton and Beizaie 2001).
PEVM only addressed the seven main F-GHGs (CF4, C2F6, C3F8, C-C4F8, HFC-23, SF6, and NF3) used in semiconductor
manufacturing. Through reporting under Subpart I of EPA's GHGRP, data on other F-GHGs (C4F6, CsFs, HFC-32, HFC-
41, HFC-134a) used in semiconductor manufacturing became available and EPA was therefore able to extrapolate
this data across the entire 1990 to 2022 timeseries. To estimate emissions for these "other F-GHGs", emissions
data from Subpart I between 2014 to 2016 were used to estimate the average share or percentage contribution of
these gases as compared to total F-GHG emissions. Subpart I emission factors were updated for 2014 by EPA as a
100 Various versions of the PEVM exist to reflect changing industrial practices. From 1990 to 1994 emissions estimates are from
PEVM vl.0, 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.
101 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).
102 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.
Industrial Processes and Product Use 4-149
-------
result of a larger set of emission factor data becoming available, so reported data from 2011 through 2013 was not
utilized for the average. To estimate non-reporter emissions from 2011-2022, the average emissions data from
Subpart I of 2011 to 2022 was used.
To estimate N2O emissions, it was assumed the proportion of N2O emissions estimated for 1995 (discussed below)
remained constant for the period of 1990 through 1994.
1995 through 1999
For 1995 through 1999, total U.S. emissions were extrapolated from the total annual emissions reported by the
Partners (1995 through 1999). Partner-reported emissions are considered more representative (e.g., in terms of
capacity utilization in a given year) than PEVM-estimated emissions and are used to generate total U.S. emissions
when applicable. The emissions reported by the Partners were divided by the ratio of the total capacity of the
plants operated by the Partners and the total capacity of all of the semiconductor plants in the United States; this
ratio represents the share of capacity attributable to the Partnership. This method assumes that Partners and non-
Partners have identical capacity utilizations and distributions of manufacturing technologies. Plant capacity data is
contained in the World Fab Forecast (WFF) database and its predecessors, which is updated quarterly. Gas-specific
emissions were estimated using the same method as for 1990 through 1994.
For this time period emissions of other F-GHGs (C4F6, CsFs, HFC-32, HFC-41, HFC-134a) were estimated using the
method described above for 1990 to 1994.
For this time period, the N2O emissions were estimated using an emission factor that was applied to the annual,
total U.S. TMLA manufactured. The emission factor was developed using a regression-through-the-origin (RTO)
model: GHGRP reported N2O emissions were regressed against the corresponding TMLA of facilities that reported
no use of abatement systems. Details on EPA's GHGRP reported emissions and development of emission factor
using the RTO model are presented in the 2011 through 2012 section. The total U.S. TMLA for 1995 through 1999
was estimated using PEVM.
2000 through 2006
Emissions for the years 2000 through 2006—the period during which Partners began the consequential application
of fluorinated greenhouse gas-reduction measures—were estimated using a combination of Partner-reported
emissions and adjusted PEVM modeled emissions. The emissions reported by Partners for each year were
accepted as the quantity emitted from the share of the industry represented by those Partners. Remaining
emissions, those from non-Partners, were estimated using PEVM, with one change. To ensure time-series
consistency and to reflect the increasing use of remote clean technology (which increases the efficiency of the
production process while lowering emissions of fluorinated greenhouse gases), the average non-Partner emission
factor (PEVM emission factor) was assumed to begin declining gradually during this period. Specifically, the non-
Partner emission factor for each year was determined by linear interpolation, using the end points of 1999 (the
original PEVM emission factor) and 2011 (a new emission factor determined for the non-Partner population based
on GHGRP-reported data, described below).
The portion of the U.S. total emissions attributed to non-Partners is obtained by multiplying PEVM's total U.S.
emissions figure by the non-Partner share of U.S. total silicon capacity for each year as described above.103 Gas-
specific emissions from non-Partners were estimated using linear interpolation between the gas-specific emissions
distributions of 1999 (assumed to be the same as that of the total U.S. Industry in 1994) and 2011 (calculated from
a subset of non-Partners that reported through the GHGRP as a result of emitting more than 25,000 MT CO2 Eq.
per year). Annual updates to PEVM reflect published figures for actual silicon consumption from VLSI Research,
Inc., revisions and additions to the world population of semiconductor manufacturing plants, and changes in IC
103 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.
4-150 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
fabrication practices within the semiconductor industry (see ITRS 2008 and Semiconductor Equipment and
Materials Industry 2011).104,105,106 For this time period emissions of other F-GHGs (C4F6, CsFs, HFC-32, HFC-41,
HFC-134a) were estimated using the method described above for 1990 to 1994.
Nitrous oxide emissions were estimated using the same methodology as the 1995 through 1999 methodology.
2007 through 2010
For the years 2007 through 2010, emissions were also estimated using a combination of Partner reported
emissions and adjusted PEVM modeled emissions to provide estimates for non-Partners; however, two
improvements were made to the estimation method employed for the previous years in the time series. First, the
2007 through 2010 emission estimates account for the fact that Partners and non-Partners employ different
distributions of manufacturing technologies, with the Partners using manufacturing technologies with greater
transistor densities and therefore greater numbers of layers.107 Second, the scope of the 2007 through 2010
estimates was expanded relative to the estimates for the years 2000 through 2006 to include emissions from
research and development (R&D) fabs. This additional enhancement was feasible through the use of more detailed
data published in the WFF. PEVM databases were updated annually as described above. The published world
average capacity utilization for 2007 through 2010 was used for production fabs, while for R&D fabs a 20 percent
figure was assumed (SIA 2009).
In addition, publicly available utilization data was used to account for differences in fab utilization for
manufacturers of discrete and IC products for 2010 emissions for non-Partners. The Semiconductor Capacity
Utilization (SICAS) Reports from SIA provides the global semiconductor industry capacity and utilization,
differentiated by discrete and IC products (SIA 2009 through 2011). PEVM estimates were adjusted using
technology-weighted capacity shares that reflect the relative influence of different utilization. Gas-specific
emissions for non-Partners were estimated using the same method as for 2000 through 2006.
104 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 to 3 year period. To prevent overstating
estimates of partner-capacity shares from plants using 300 mm wafers, design capacities contained in WFF 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 (FQSV) to determine the
average design capacity over the 2006 period.
105 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.
106 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.
107 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.
Industrial Processes and Product Use 4-151
-------
For this time period emissions of other F-GHGs (CsFs, CH2F2, CH3F, CH2FCF3, C2H2F4) were estimated using the
method described above for 1990 to 1994. Nitrous oxide emissions were estimated using the same methodology
as the 1995 through 1999 methodology.
2011 through 2012
The fifth method for estimating emissions from semiconductor manufacturing covers the period 2011 through
2012. This methodology differs from previous years because the EPA's Partnership with the semiconductor
industry ended (in 2010) and reporting under EPA's GHGRP began. Manufacturers whose estimated uncontrolled
emissions equal or exceed 25,000 MT CO2 Eq. per year (based on default F-GHG-specific emission factors and total
capacity in terms of substrate area) are required to report their emissions to EPA. This population of reporters to
EPA's GHGRP included both historical Partners of EPA's PFC Reduction/Climate Partnership as well as non-Partners
some of which use gallium arsenide (GaAs) technology in addition to Si technology.108 Emissions from the
population of manufacturers that were below the reporting threshold were also estimated for this time period
using EPA-developed emission factors and estimates of facility-specific production obtained from WFF. Inventory
totals reflect the emissions from both reporting and non-reporting populations.
Under EPA's GHGRP, semiconductor manufacturing facilities report emissions of F-GHGs (for all types of F-GHGs)
used in etch and clean processes as well as emissions of fluorinated heat transfer fluids. (Fluorinated heat transfer
fluids are used to control process temperatures, thermally test devices, and clean substrate surfaces, among other
applications.) They also report N2O emissions from CVD and other processes. The F-GHGs and N2O were
aggregated, by gas, across all semiconductor manufacturing GHGRP reporters to calculate gas-specific emissions
for the GHGRP-reporting segment of the U.S. industry. At this time, emissions that result from heat transfer fluid
use that are HFC, PFC and SF6 are included in the total emission estimates from semiconductor manufacturing, and
these GHGRP-reported emissions have been compiled and presented in Table 4-121. F-HTF emissions resulting
from other types of gases (e.g., HFEs) are not presented in semiconductor manufacturing totals in Table 4-121 and
Table 4-122 but are shown in Table 4-123 for informational purposes.
Changes to the default emission factors and default destruction or removal efficiencies (DREs) used for GHGRP
reporting affected the emissions trend between 2013 and 2014. These changes did not reflect actual emission rate
changes but data improvements. Therefore, for the current Inventory, EPA adjusted the time series of GHGRP-
reported data for 2011 through 2013 to ensure time-series consistency using a series of calculations that took into
account the characteristics of a facility (e.g., wafer size and abatement use). To adjust emissions for facilities that
did not report abatement in 2011 through 2013, EPA simply applied the revised emission factors to each facility's
estimated gas consumption by gas, process type and wafer size. In 2014, EPA also started collecting information on
fab-wide DREs and the gases abated by process type, which were used in calculations for adjusting emissions from
facilities that abated F-GHGs in 2011 through 2013.
• To adjust emissions for facilities that abated emissions in 2011 through 2013, EPA first calculated the
quantity of gas abated in 2014 using reported F-GHG emissions, the revised default DREs (or the
estimated site-specific DRE,109 if a site-specific DRE was indicated), and the fab-wide DREs reported in
2014.110 To adjust emissions for facilities that abated emissions in 2011 through 2013, EPA first estimated
108 GaAs and Si technologies refer to the wafer on which devices are manufactured, which use the same PFCs but in different
ways.
109 EPA generally assumed site-specific DREs were as follows: CF4, Etch (90 percent); all other gases, Etch (98 percent); NF3,
Clean (95 percent); CF4, Clean (80 percent), and all other gases, Clean (80 percent). There were a few exceptions where a higher
DRE was assumed to ensure the calculations operated correctly when there was 100 percent abatement.
110 If abatement information was not available for 2014 or the reported incorrectly in 2014, data from 2015 or 2016 was
substituted.
4-152 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
the percentage of gas passing through abatement systems for remote plasma clean in 2014 using the ratio
of emissions reported for CF4 and NF3.
• EPA then estimated the quantity of NF3 abated for remote plasma clean in 2014 using the ratio of
emissions reported for CF4 (which is not abated) and NF3. This abated quantity was then subtracted from
the total abated quantity calculated as described in the bullet above.
• To account for the resulting remaining abated quantity, EPA assumed that the percentage of gas passing
through abatement systems was the same across all remaining gas and process type combinations where
abatement was reported for 2014.
• The percentage of gas abated was then assumed to be the same in 2011 through 2013 (if the facility
claimed abatement that year) as in 2014 for each gas abated in 2014.
The revised emission factors and DREs were then applied to the estimated gas consumption for each facility by gas,
process type and wafer size.111
For the segment of the semiconductor industry that is below EPA's GHGRP reporting threshold, and for R&D
facilities, which are not covered by EPA's GHGRP, emission estimates are based on EPA-developed emission factors
for the F-GHGs and N2O and estimates of manufacturing activity. The new emission factors (in units of mass of CO2
Eq./TMLA [million square inches (MSI)]) are based on the emissions reported under EPA's GHGRP by facilities
without abatement and on the TMLA estimates for these facilities based on the WFF (SEMI 2012, 2013).112 In a
refinement of the method used to estimate emissions for the non-Partner population for prior years, different
emission factors were developed for different subpopulations of fabs, disaggregated by wafer size (200 mm and
300 mm). For each of these groups, a subpopulation-specific emission factor was obtained using a regression-
through-the-origin (RTO) model: facility-reported aggregate emissions of seven F-GHGs (CF4, C2F6, C3F8, C-C4F8,
CHF3, SFs and NF3)113 were regressed against the corresponding TMLA to estimate an aggregate F-GHG emissions
factor (CO2 Eq./MSI TMLA), and facility-reported N2O emissions were regressed against the corresponding TMLA to
estimate a N2O emissions factor (CO2 Eq./MSI TMLA). For each subpopulation, the slope of the RTO model is the
emission factor for that subpopulation. Information on the use of point-of-use abatement by non-reporting fabs
was not available; thus, EPA conservatively assumed that non-reporting facilities did not use point-of-use
abatement.
For 2011 and 2012, estimates of TMLA relied on the capacity utilization of the fabs published by the U.S. Census
Bureau's Historical Data Quarterly Survey of Plant Capacity Utilization (USCB 2011, 2012). Similar to the
assumption for 2007 through 2010, facilities with only R&D activities were assumed to utilize only 20 percent of
their manufacturing capacity. All other facilities in the United States are assumed to utilize the average percent of
the manufacturing capacity without distinguishing whether fabs produce discrete products or logic products.
Non-reporting fabs were then broken out into subpopulations by wafer size (200 mm and 300 mm), using
information available through the WFF. The appropriate emission factor was applied to the total TMLA of each
subpopulation of non-reporting facilities to estimate the CC>2-equivalent emissions of that subpopulation.
111 Since facilities did not report by fab before 2014, fab-wide DREs were averaged if a facility had more than one fab. For
facilities that reported more than one wafer size per facility, the percentages of a facility's emissions per wafer size were
estimated in 2014 and applied to earlier years, if possible. If the percentage of emissions per wafer size were unknown, a 50/50
split was used.
112 EPA does not have information on fab-wide DREs for this time period, so it is not possible to estimate uncontrolled
emissions from fabs that reported point-of-use abatement. These fabs were therefore excluded from the regression analysis.
(They are still included in the national totals.)
113 Only seven gases were aggregated because inclusion of F-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 4-153
-------
Gas-specific, CCh-equivalent emissions for each subpopulation of non-reporting facilities were estimated using the
corresponding reported distribution of gas-specific, CC>2-equivalent emissions from which the aggregate emission
factors, based on GHGRP-reported data, were developed. Estimated in this manner, the non-reporting population
accounted for 4.9 and 5.0 percent of U.S. emissions in 2011 and 2012, respectively. The GHGRP-reported emissions
and the calculated non-reporting population emissions are summed to estimate the total emissions from
semiconductor manufacturing.
2013 and 2014
For 2013 and 2014, as for 2011 and 2012, F-GHG and N2O emissions data received through EPA's GHGRP were
aggregated, by gas, across all semiconductor-manufacturing GHGRP reporters to calculate gas-specific emissions
for the GHGRP-reporting segment of the U.S. industry. However, for these years WFF data was not available.
Therefore, an updated methodology that does not depend on the WFF derived activity data was used to estimate
emissions for the segment of the industry that are not covered by EPA's GHGRP. For the facilities that did not
report to the GHGRP (i.e., which are below EPA's GHGRP reporting threshold or are R&D facilities), emissions were
estimated based on the proportion of total U.S. emissions attributed to non-reporters for 2011 and 2012. EPA used
a simple averaging method by first estimating this proportion for both F-GHGs and N2O for 2011, 2012, and 2015
and 2016, resulting in one set of proportions for F-GHGs and one set for N2O, and then applied the average of each
set to the 2013 and 2014 GHGRP reported emissions to estimate the non-reporters' emissions. Fluorinated gas-
specific, CC>2-equivalent emissions for non-reporters were estimated using the corresponding reported distribution
of gas-specific, CC>2-equivalent emissions reported through EPA's GHGRP for 2013 and 2014.
GHGRP-reported emissions in 2013 were adjusted to capture changes to the default emission factors and default
destruction or removal efficiencies used for GHGRP reporting, affecting the emissions trend between 2013 and
2014. EPA used the same method to make these adjustments as described above for 2011 and 2012 GHGRP data.
2015 through 2022
Similar to the methods described above for 2011 and 2012, and 2013 and 2014, EPA relied upon emissions data
reported directly through the GHGRP. For 2015 through 2022, EPA took an approach similar to the one used for
2011 and 2012 to estimate emissions for the segment of the semiconductor industry that is below EPA's GHGRP
reporting threshold, and for R&D facilities, which are not covered by EPA's GHGRP. However, in a change from
previous years, EPA was able to develop new annual emission factors for 2015 through 2022 using TMLA from WFF
and a more comprehensive set of emissions, i.e., fabs with as well as without abatement control, as new
information about the use of abatement in GHGRP fabs and fab-wide were available. Fab-wide DREs represent
total fab CO2 Eq.-weighted controlled F-GHG and N2O emissions (emissions after the use of abatement) divided by
total fab CO2 Eq.-weighted uncontrolled F-GHG and N2O emissions (emission prior to the use of abatement).
Using information about reported emissions and the use of abatement and fab-wide DREs, EPA was able to
calculate uncontrolled emissions (each total F-GHG and N2O) for every GHGRP reporting fab. Using this, coupled
with TMLA estimated using methods described above (see 2011 through 2012), EPA derived emission factors by
year, gas type (F-GHG or N2O), and wafer size (200 mm and less or 300 mm) by dividing the total annual emissions
reported by GHGRP reporters by the total TMLA estimated for those reporters. These emission factors were
multiplied by estimates of non-reporter TMLA to arrive at estimates of total F-GHG and N2O emissions for non-
reporters for each year. For each wafer size, the total F-GHG emissions were disaggregated into individual gases
using the shares of total emissions represented by those gases in the emissions reported to the GHGRP by
unabated fabs producing that wafer size.
Data Sources
GHGRP reporters, which consist of former EPA Partners and non-Partners, estimated their emissions using a
default emission factor method established by EPA. Like the Tier 2c Method in the 2019 Refinement to the 2006
IPCC Guidelines, this method uses different emission and byproduct generation factors for different F-GHGs and
process types and uses factors for different wafer sizes (i.e., 300mm vs. 150 and 200mm) and CVD clean subtypes
4-154 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
(in situ thermal, in situ plasma, and remote plasma). Starting with 2014 reported emissions, EPA's GHGRP required
semiconductor manufacturers to apply updated emission factors to estimate their F-GHG emissions. For the years
2011 through 2013 reported emissions, semiconductor manufacturers used older emission factors to estimate
their F-GHG emissions (Federal Register / Vol. 75, No. 230 /December 1, 2010, 74829). Subpart I emission factors
were updated for 2014 by EPA as a result of a larger set of emission factor data becoming available as part of the
Subpart I petition process, which took place from 2011 through 2013. In addition to semiconductor manufacturing,
GHGRP also includes reported emissions from MEMS and PV producers.
Historically, semiconductor industry partners estimated and reported their emissions using a range of methods
and uneven documentation. It is assumed that most Partners used a method at least as accurate as the IPCC's Tier
2a Methodology, recommended in the 2006IPCC Guidelines. Partners are estimated to have accounted for
between 56 and 79 percent of F-GHG emissions from U.S. semiconductor manufacturing between 1995 and 2010,
with the percentage declining in recent years as Partners increasingly implemented abatement measures.
Estimates of operating plant capacities and characteristics for Partners and non-Partners were derived from the
Semiconductor Equipment and Materials Industry (SEMI) WFF (formerly World Fab Watch) database (1996 through
2012, 2013, 2016, 2018, 2021, and 2023) (e.g., Semiconductor Materials and Equipment Industry 2021). Actual
worldwide capacity utilizations for 2008 through 2010 were obtained from Semiconductor International Capacity
Statistics (SICAS) (SIA 2009 through 2011). Estimates of the number of layers for each linewidth was obtained from
International Technology Roadmap for Semiconductors: 2013 Edition (Burton and Beizaie 2001; ITRS 2007; ITRS
2008; ITRS 2011; ITRS 2013). PEVM utilized the WFF, SICAS, and ITRS, as well as historical silicon consumption
estimates published by VLSI. Actual quarterly U.S. capacity utilizations for 2011, 2012, 2014 to 2022 were obtained
from the U.S. Census Bureau's Historical Data Quarterly Survey of Plant Capacity Utilization (USCB 2011, 2012,
2015, 2016, 2017, 2018, 2019, 2020, 2021, 2022).
Estimates of PV manufacturing capacity, which are used to calculate emissions from non-reporting facilities, are
based on data from two sources. A historical market analysis from DisplaySearch provided estimates of U.S.
manufacturing capacity from 2000 to 2009 (DisplaySearch 2010). Domestic PV cell production for 2012 was
obtained from a Congressional Research Service report titled U.S. Solar Photovoltaic Manufacturing: Industry
Trends, Global Competition, Federal Support (Platzer 2015).
Uncertainty
A quantitative uncertainty analysis of this source category was performed using the IPCC-recommended Approach
2 uncertainty estimation methodology, the Monte Carlo stochastic simulation technique. The Monte Carlo
stochastic simulation was performed on the total emissions estimate from the electronics industry, represented in
equation form as:
Equation 4-20: Total Emissions from Electronics Industry
Total Emissions (Er)
= Semiconductors F-GHG and N20 Emissions (ESemi)
+ MEMS F-GHG and N20 Emissions (EMBMS) + PV F-GHG and N20 Emissions (EPl?)
+ HFC, PFC and SF6 F-HTFs Emissions (EHTF)
The uncertainty in the total emissions for the electronics industry, presented in Table 4-124 below, results from
the convolution of four distributions of emissions, namely from semiconductors manufacturing, MEMS
manufacturing, PV manufacturing and emissions of heat transfer fluids. The approaches for estimating uncertainty
in each of the sources are described below:
Semiconductors Manufacture Emission Uncertainty
The Monte Carlo stochastic simulation was performed on the emissions estimate from semiconductor
manufacturing, represented in equation form as:
Industrial Processes and Product Use 4-155
-------
Equation 4-21: Total Emissions from Semiconductor Manufacturing
Semiconductors F-GHG and N20 Emissions (ESemi)
= GHGRP Reported F-GHG Emissions (ER,F-GHG, Semi)
+ Non-Reporters' Estimated F-GHG Emissions (ENRj_GHGisemi)
+ GHGRP Reported N20 Emissions (ERN2o,serm)
+ Non-Reporters' Estimated N20 Emissions (ENRN2o,semi)
The uncertainty in Esemi results from the convolution of four distributions of emissions, Er,f-ghg,semi ER,N2o,semi Enr,f-
GHG,semi and ENR,N2o,semi. The approaches for estimating each distribution and combining them to arrive at the
reported 95 percent confidence interval (CI) for Esemi are described in the remainder of this section.
The uncertainty estimate of Er, f-ghg,semi, or GHGRP-reported F-GHG emissions, is developed based on gas-specific
uncertainty estimates of emissions for two industry segments, one processing 200 mm or less wafers and one
processing 300 mm wafers. Uncertainties in emissions for each gas and industry segment are based on an
uncertainty analysis conducted during the assessment of emission estimation methods for the Subpart I
rulemaking in 2012 (see Technical Support for Modifications to the Fluorinated Greenhouse Gas Emission
Estimation Method Option for Semiconductor Facilities under Subpart I, docket EPA-HQ-OAR-2011-0028).114 This
assessment relied on facility-specific gas information by gas and wafer size, and incorporated uncertainty
associated with both emission factors and gas consumption quantities. The 2012 analysis did not consider the use
of abatement.
For the industry segment that manufactured 200 mm wafers, estimates of uncertainty at a 95 percent CI ranged
from ±29 percent for C3F8 to ±10 percent for CF4. For the corresponding 300 mm industry segment, estimates of
uncertainty at the 95 percent CI ranged from ±36 percent for C4F8 to ±16 percent for CF4. For gases for which
uncertainty was not analyzed in the 2012 assessment (e.g., CH2F2), EPA applied the 95 percent CI range equivalent
to the range for the gas and industry segment with the highest uncertainty from the 2012 assessment. These gas
and wafer-specific uncertainty estimates were developed to represent uncertainty at a facility-level, but they are
applied to the total emissions across all the facilities that did not abate emissions as reported under EPA's GHGRP
at a national-level. Hence, it is noted that the uncertainty estimates used may be overestimating the uncertainties
at a national-level.
For those facilities reporting abatement of emissions under EPA's GHGRP, estimates of uncertainties for the no
abatement industry segments are modified to reflect the use of full abatement (abatement of all gases from all
cleaning and etching equipment) and partial abatement. These assumptions used to develop uncertainties for the
partial and full abatement facilities are identical for 200 mm and 300 mm wafer processing facilities. For all
facilities reporting gas abatement, a triangular distribution of destruction or removal efficiency is assumed for each
gas. The triangular distributions range from an asymmetric and highly uncertain distribution of zero percent
minimum to 90 percent maximum with 70 percent most likely value for CF4 to a symmetric and less uncertain
distribution of 85 percent minimum to 95 percent maximum with 90 percent most likely value for C4F8, NF3, and
SFs. For facilities reporting partial abatement, the distribution of fraction of the gas fed through the abatement
device, for each gas, is assumed to be triangularly distributed as well. It is assumed that no more than 50 percent
114 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 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.
4-156 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
of the gases are abated (i.e., the maximum value) and that 50 percent is the most likely value, and the minimum is
zero percent. Consideration of abatement then resulted in four additional industry segments, two 200-mm wafer-
processing segments (one fully and one partially abating each gas) and two 300-mm wafer-processing segment
(one fully and the other partially abating each gas). Gas-specific emission uncertainties were estimated by
convolving the distributions of unabated emissions with the appropriate distribution of abatement efficiency for
fully and partially abated facilities using a Monte Carlo simulation.
The uncertainty in ER,F-GHG,semi is obtained by allocating the estimates of uncertainties to the total GHGRP-reported
emissions from each of the six industry segments, and then running a Monte Carlo simulation which results in the
95 percent CI for emissions from GHGRP-reporting facilities (ER,F-GHG,semi).
The uncertainty in ER,N2o,semi is obtained by assuming that the uncertainty in the emissions reported by each of the
GHGRP reporting facilities results from the uncertainty in quantity of N2O consumed and the N2O emission factor
(or utilization). Similar to analyses completed for Subpart I (see Technical Support for Modifications to the
Fluorinated Greenhouse Gas Emission Estimation Method Option for Semiconductor Facilities under Subpart I,
docket EPA-HQ-OAR-2011-0028), the uncertainty of N2O consumed was assumed to be 20 percent. Consumption
of N2O for GHGRP reporting facilities was estimated by back-calculating from emissions reported and assuming no
abatement. The quantity of N2O utilized (the complement of the emission factor) was assumed to have a triangular
distribution with a minimum value of zero percent, mode of 20 percent and maximum value of 84 percent. The
minimum was selected based on physical limitations, the mode was set equivalent to the Subpart I default N2O
utilization rate for chemical vapor deposition, and the maximum was set equal to the maximum utilization rate
found in ISMI Analysis of Nitrous Oxide Survey Data (ISMI 2009). The inputs were used to simulate emissions for
each of the GHGRP reporting, INhO-emitting facilities. The uncertainty for the total reported N2O emissions was
then estimated by combining the uncertainties of each facilities' reported emissions using Monte Carlo simulation.
The estimate of uncertainty in Enr, F-GHG.semi and Enr, N2o,semi entailed developing estimates of uncertainties for the
emissions factors and the corresponding estimates of TMLA.
The uncertainty in TMLA depends on the uncertainty of two variables—an estimate of the uncertainty in the
average annual capacity utilization for each level of production of fabs (e.g., full scale or R&D production) and a
corresponding estimate of the uncertainty in the number of layers manufactured. For both variables, the
distributions of capacity utilizations and number of manufactured layers are assumed triangular for all categories
of non-reporting fabs. The most probable utilization is assumed to be 82 percent, with the highest and lowest
utilization assumed to be 89 percent, and 70 percent, respectively. For the triangular distributions that govern the
number of possible layers manufactured, it is assumed the most probable value is one layer less than reported in
the ITRS; the smallest number varied by technology generation between one and two layers less than given in the
ITRS and largest number of layers corresponded to the figure given in the ITRS.
The uncertainty bounds for the average capacity utilization and the number of layers manufactured are used as
inputs in a separate Monte Carlo simulation to estimate the uncertainty around the TMLA of both individual
facilities as well as the total non-reporting TMLA of each sub-population.
The uncertainty around the emission factors for non-reporting facilities is the total combined uncertainties of
individual gases and the TMLA of each reporting facility in that category. The combined uncertainty of emissions of
individual gases from non-reporters is equal to the uncertainty of total emissions for non-reporting facilities.
The uncertainty around the emission factors for non-reporting facilities is the total combined uncertainties of
individual gases (MT units) and the TMLA of each reporting facility in that category. The combined uncertainty of
emissions of individual gases from non-reporters is equal to the uncertainty of total emissions for non-reporting
facilities. For each wafer size for reporting facilities, emissions of individual gases were regressed on TMLA (with an
intercept forced to zero) for 10,000 emission and 10,000 TMLA values in a Monte Carlo simulation, which results in
10,000 total regression coefficients (emission factors). The 2.5th and the 97.5th percentile of these emission factors
are determined, and the bounds are assigned as the percent difference from the estimated emission factor.
The next step in estimating the uncertainty in emissions of reporting and non-reporting facilities in semiconductor
Industrial Processes and Product Use 4-157
-------
manufacture is convolving the distribution of reported emissions, emission factors, and TMLA using Monte Carlo
simulation. For this Monte Carlo simulation, the distributions of the reported F-GHG gas- and wafer size-specific
emissions are assumed to be normally distributed, and the uncertainty bounds are assigned at 1.96 standard
deviations around the estimated mean. The were some instances, though, where departures from normality were
observed for variables, including for the distributions of the gas- and wafer size-specific N2O emissions, TMLA, and
non-reporter emission factors, both for F-GHGs and N2O. As a result, the distributions for these parameters were
assumed to follow a PERT beta distribution.
MEMS Manufacture Emission Uncertainty
The Monte Carlo stochastic simulation was performed on the emissions estimate from MEMS manufacturing,
represented in equation form as:
Equation 4-22: Total Emissions from MEMS Manufacturing
MEMS F-GHG and N2O Emissions (Emems) = GHGRP Reported F-GHG Emissions (Er,f-ghg,m£ms) + GHGRP
Reported N2O Emissions (Er.nvo, mems)
MEMS F-GHG and N20 Emissions (EMBMS)
= GHGRP Reported F-GHG Emissions (ER F.GHGMBMS)
+ GHGRP Reported N20 Emissions (ERiW2o,mbms)
Emissions from MEMS manufacturing are only quantified for GHGRP reporters. MEMS manufacturers that report
to the GHGRP all report the use of 200 mm wafers. Some MEMS manufacturers report using abatement
equipment. Therefore, the estimates of uncertainty at the 95 percent CI for each gas emitted by MEMS
manufacturers are set equal to the gas-specific uncertainties for manufacture of 200mm semiconductor wafers
with partial abatement. The same assumption is applied for uncertainty levels for GHGRP reported MEMS N2O
emissions (Er,n2o,mems).
PV Manufacture Emission Uncertainty
The Monte Carlo stochastic simulation was performed on the emissions estimate from PV manufacturing,
represented in equation form as:
Equation 4-23: Total Emissions from PV Manufacturing
PV F-GHG and N2O Emissions (Epv) = Non-Reporters' Estimated F-GHG Emissions (Enr,f-ghg,/>k) + Non-
Reporters' Estimated N2O Emissions (Enr,n2
-------
percent were applied to estimate uncertainty associated with the various types of heat transfer fluids, including
PFCs, HFC, and SF6, at the national level.
The results of the Approach 2 quantitative uncertainty analysis for electronics manufacturing are summarized in
Table 4-124. These results were obtained by convolving—using Monte Carlo simulation—the distributions of
emissions for each reporting and non-reporting facility that manufactures semiconductors, MEMS, or PVs and use
heat transfer fluids. The emissions estimate for total U.S. F-GHG, N2O, and HTF emissions from electronics
manufacturing were estimated to be between 4.44 and 5.02 MMT CO2 Eq. at a 95 percent CI level. This range
represents 6 percent below to 6 percent above the 2022 emission estimate of 4.73 MMT CO2 Eq. for all emissions
from electronics manufacture. This range and the associated percentages apply to the estimate of total emissions
rather than those of individual gases. Uncertainties associated with individual gases will be somewhat higher than
the aggregate but were not explicitly modeled.
Table 4-124: Approach 2 Quantitative Uncertainty Estimates for HFC, PFC, SF6, NF3 and N2O
Emissions from Electronics Manufacture (MMT CO2 Eq. and Percent)
Source
Gas
2022 Emission Estimate
(MMTCO. Eq.)
Uncertainty Range Relative to Emission Estimate-1
(MMTCO' Eq.) (%)
Lower
Upper
Lower Upper
Bound1,
Bound1,
Bound Bound
Electronics
Industry
HFC, PFC, SF6, NF3,
and N20
4.7
4.4
5.0
-6% +6%
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.
QA/QC and Verification
For its GHGRP data, EPA verifies annual facility-level reports through a multi-step process (e.g., including a
combination of pre-and post-submittal electronic checks and manual reviews by staff) to identify potential errors
and ensure that data submitted to EPA are accurate, complete, and consistent (EPA 2015).115 Based on the results
of the verification process, EPA follows up with facilities to resolve mistakes that may have occurred. The post-
submittals checks are consistent with a number of general and category-specific QC procedures including range
checks, statistical checks, algorithm checks, and year-to-year checks of reported data and emissions.
For more information on the general QA/QC process applied to this source category, consistent with Volume 1,
Chapter 6 of the 2006IPCC Guidelines, see the QA/QC and Verification Procedures section in the introduction of
the IPPU chapter and Annex 8 for more details.
Recalculations Discussion
Any resubmitted emissions data reported to EPA's GHGRP from all prior years were updated in this Inventory.
Additionally, EPA made the following changes:
• To estimate non-reporter F-GHG and N2O emissions, EPA relies on data reported through Subpart I and
the World Fab Forecast. This process requires EPA to map facilities that report through Subpart I and
which are also represented in the World Fab Forecast. For this Inventory update, EPA identified and made
115 GHGRP Report Verification Factsheet. See https://www.epa.gov/sites/production/files/2Q15-
07/documents/ehgrp verification factsheet.pdf.
Industrial Processes and Product Use 4-159
-------
corrections to a few instances of this mapping based on new information and additional reviews of the
data. This had minimal effects on emission estimates.
• EPA re-ran regression analyses for years 2010 to 2021 to reflect updates to Subpart I and the World Fab
Forecast. These changes had minor effects on the emission factors, standard error, and R2 values for all
years. This resulted in the recalculation of non-reporter's F-GHG and N2O estimates for all years.
• EPA recalculated HTF emissions for years 1990 to 2021 using IPCC Fifth Assessment Report (AR5) GWP
values (IPCC 2013). Emission estimates were incorrectly calculated using the IPCC Fourth Assessment
Report (AR4) GWP values (IPCC 2007) in the previous Inventory (EPA 2023). Overall, the impact of these
recalculations led to an average decrease of 0.009 MMT CO2 Eq. (0.19 percent) across the time series
(1990 through 2021).
• EPA recalculated fluorinated GHG consumption for 2014 to 2021 using the following methodology.
Fluorinated GHG consumption estimates for unabated fabs were calculated using reported GHGRP
emissions data and default emission factors for Subpart I. Because certain fluorinated GHGs are
generated as by-products as well as used as input gases, both input gas emission factors and by-product
gas emission factors were factored into this calculation. For abated fabs, a "consumption factor" was
developed by dividing the reported emissions of each fluorinated GHG from unabated fabs by the
estimated consumption of each fluorinated GHG for each wafter size. Fluorinated GHG consumption for
2021 was estimated based on GDP growth of the 2020 consumption estimate. The consumption estimate
will be updated with reported 2021 GHGRP emissions data.
• EPA refined the non-reporting population for 2015 to 2022 by conducting an analysis into the criteria
being used to determine which fabs should be included and excluded from this population. Overall, the
impact of this refinement led to an average increase in semiconductor emissions by 0.02 MMT CO2 Eq.
(0.45 percent) for the time series 2015 to 2022.
• EPA recalculated non-reporter emissions for 2015 to 2022 by developing emission factors for individual
gases and calculated on an MT basis. Overall, the impact of this refinement led to an average increase in
semiconductor emissions by 0.04 MMT CO2 Eq. (0.84 percent) for the time series 2014 to 2022.
Planned Improvements
The Inventory methodology uses data reported through the EPA Partnership (for earlier years) and EPA's GHGRP
(for later years) to extrapolate the emissions of the non-reporting population. While these techniques are well
developed, the accuracy of the emissions estimates for the non-reporting population could be further increased
through EPA's further investigation of and improvement upon the accuracy of estimated activity in the form of
TMLA.
The Inventory uses utilization from two different sources for various time periods-SEMI to develop PEVM and to
estimate non-Partner emissions for the period 1995 to 2010 and U.S. Census Bureau for 2011 through 2022. SEMI
reported global capacity utilization for manufacturers through 2011. U.S. Census Bureau capacity utilization
include U.S. semiconductor manufacturers as well as assemblers. Further analysis on the impacts of using a new
and different source of utilization data could prove to be useful in better understanding of industry trends and
impacts of utilization data sources on historical emission estimates.
Estimates of semiconductor non-reporter and non-Partner emissions are based on EPA-developed emission factors
for the time periods pre-2010, 2011 through 2012, and 2015 through 2022. Based on the data available for these
time periods, the methods used to develop emission factors for non-reporters and non-Partners are slightly
inconsistent for semiconductors (e.g., how data representing emissions and TMLA from the manufacture of various
wafer sizes are aggregated or disaggregated for purposes of calculating emission factors). Further analyses to
support potentially adjusting the methods for developing these emission factors could be done to better ensure
consistency across the time series.
4-160 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
The methodology for estimating semiconductor emissions from non-reporters uses data from the International
Technology Roadmap for Semiconductors (ITRS) on the number of layers associated with various technology node
sizes. The ITRS has now been replaced by the International Roadmap for Devices and Systems (IRDS), which has
published updated data on the number of layers used in each device type and node size (in nanometers).
Incorporating this updated dataset will improve the accuracy of emissions estimates from non-reporting
semiconductor fabs.
4.25 Substitution of Ozone Depleting
Substances (CRT Source Category 2F)
This reporting category (2F) includes emissions from the substitution of ozone-depleting substance (ODS).
Hydrofluorocarbons (HFCs), perfluorocarbons (PFCs), and carbon dioxide (CO2) are used as alternatives to several
classes of ODS that are being phased out under the terms of the Montreal Protocol and the Clean Air Act
Amendments of 1990.116 Ozone-depleting substances—chlorofluorocarbons (CFCs), halons, carbon tetrachloride,
methyl chloroform, and hydrochlorofluorocarbons (HCFCs)—are used in a variety of industrial applications
including refrigeration and air conditioning equipment, solvent cleaning, foam production, sterilization, fire
extinguishing, and aerosols. Although HFCs and PFCs are not harmful to the stratospheric ozone layer, they are
potent greenhouse gases. On December 27, 2020, the American Innovation and Manufacturing (AIM) Act was
enacted by Congress and directs EPA to address HFCs by phasing down production and consumption (i.e.,
production plus import minus export), maximizing reclamation and minimizing releases from equipment, and
facilitating the transition to next-generation technologies through sector-based restrictions. Emission estimates for
HFCs, PFCs, and CO2 used as substitutes for ODSs are provided in Table 4-125 and Table 4-126.117
Table 4-125: Emissions of HFCs, PFCs, and CO2 from ODS Substitutes (MMT CO2 Eq.)
Gas
1990
2005
2018
2019
2020
2021
2022
HFC-23
0.0
+
+
+
+
+
+
HFC-32
0.0:
0.3
6.1
6.9
7.8
9.4
10.5
HFC-125
+
8.2
48.8
53.1
57.8
66.3
72.0
HFC-134a
+1
72.8
56.4
55.3
54.1
50.0
48.3
HFC-143a
+
10.0
29.7
29.9
29.9
30.0
29.8
HFC-236fa
11111
O
O
i-o 1
0.8
0.7
0.7
0.6
0.6
cf4
0.0
+
+
+
+
+
0.1
C02
+ i
+
+
+
+
+
Other Saturated HFCsa
0.3
6.9
15.9
16.0
15.9
16.3
16.8
Other PFCs and HFOsb
+ it
0.1 ¦
+
+
+
+
+
Total
0.3
99.5
157.9
162.1
166.2
172.7
178.1
+ Does not exceed 0.05 MMT C02 Eq.
a Other Saturated HFCs represents an unspecified mix of saturated HFCs, which includes HFC-152a, HFC-
227ea, HFC-245fa, HFC-365mfc, and HFC-43-10mee.
b Other PFCs and HFOs represents an unspecified mix of PFCs and HFOs, which includes HCFO-1233zd(E),
HFO-1234yf, HFO-1234ze(E), HFO-1336mzz(Z), C4F10, and PFC/PFPEs, the latter being a proxy for a diverse
116 [42 U.S.C § 7671, CAA Title VI],
117 Emissions of ODS are not included here consistent with UNFCCC reporting guidelines for national inventories noted in Box
4-1. See Annex 6.2 for more details on emissions of ODS. Emissions from C02 used in the food and beverage industry are
separately reported in Chapter 4.16 Carbon Dioxide Consumption but does not include C02 in ODS substitute use sectors as a
refrigerant, foam blowing agent, or fire extinguishing agent.
Industrial Processes and Product Use 4-161
-------
collection of PFCs and perfluoropolyethers (PFPEs) employed for solvent applications. For estimating
purposes, the GWP value used for PFC/PFPEs was based upon n-C6Fi4.
Note: Totals may not sum due to independent rounding.
Table 4-126: Emissions of HFCs, PFCs, and CO2 from ODS Substitution (Metric Tons)
Gas
1990
2005
2018
2019
2020
2021
2022
HFC-23
0
1
2
2
2
2
3
HFC-32
0
I
397
1
I
9,008
10,156
11,461
13,958
15,582
HFC-125
+
2,580
15,406
16,761
18,240
20,909
22,704
HFC-134a
+
I
56,029
1
I
43,419
42,558
41,590
38,447
37,167
HFC-143a
+
2,093
6,188
6,230
6,234
6,240
6,203
HFC-236fa
0
127
111!
99
91
84
78
72
cf4
0
3
5
5
4
4
4
C02
14
1,325
|
1
3,093
3,303
3,516
3,734
3,969
Other Saturated HFCsa
M
M
M
M
M
M
M
Other PFCs and HFOsb
M
M
M
M
M
M
M
+ Does not exceed 0.5 MT.
M (Mixture of Gases).
a Other Saturated HFCs represents an unspecified mix of saturated HFCs, which includes HFC-152a, HFC-
227ea, HFC-245fa, HFC-365mfc, and HFC-43-10mee.
b Other PFCs and HFOs represents an unspecified mix of PFCs and HFOs, which includes HCFO-1233zd(E),
HFO-1234yf, HFO-1234ze(E), HFO-1336mzz(Z), C4F10, and PFC/PFPEs, the latter being a proxy for a diverse
collection of PFCs and perfluoropolyethers (PFPEs) employed for solvent applications. For estimating
purposes, the GWP value used for PFC/PFPEs was based upon n-C6Fi4.
In 1990 and 1991, the only significant emissions of HFCs and PFCs as substitutes to ODSs were relatively small
amounts of HFC-152a—used as an aerosol propellant and also a component of the refrigerant blend R-500 used in
chillers. Beginning in 1992, HFC-134a was used in growing amounts as a refrigerant in motor vehicle air-
conditioners and in refrigerant blends such as R-404A.118 In 1993, the use of HFCs in foam production began, and
in 1994 ODS substitutes for halons entered widespread use in the United States as halon production was phased
out. In 1995, these compounds also found applications as solvents. Non-fluorinated ODS substitutes, such as CO2,
have been used in place of ODS in certain foam production and fire extinguishing uses since the 1990s.
The use and subsequent emissions of HFCs, PFCs, and CO2 as ODS substitutes has been increasing from small
amounts in 1990 to 178.1 MMT CO2 Eq. emitted in 2022. This increase was in large part the result of efforts to
phase out CFCs, HCFCs, and other ODSs in the United States. Use and emissions of HFCs are expected to start
decreasing in the next few years and continue downward as production and consumption of HFCs are phased
down to 15 percent of their baseline levels by 2036 through an allowance allocation and trading program
established by EPA. Improvements in recovery practices and the use of alternative gases and technologies, through
voluntary actions and in response to existing and potential future regulations under the AIM Act, will also
contribute to a reduction in HFC use and emissions.
Table 4-127 presents emissions of HFCs, PFCs, and CO2 as ODS substitutes by end-use sector for 1990 through
2022. The refrigeration and air-conditioning sector is further broken down by sub-sector. The end-use sectors that
contributed the most toward emissions of HFCs, PFCs, and CO2 as ODS substitutes in 2022 include refrigeration and
air-conditioning (144.6 MMT CO2 Eq., or approximately 81 percent), aerosols (17.0 MMT CO2 Eq., or approximately
10 percent), and foams (11.7 MMT CO2 Eq., or approximately 7 percent). Within the refrigeration and air-
conditioning end-use sector residential unitary AC, part of the Residential Stationary Air-conditioning subsector
118 R-404A contains HFC-125, HFC-143a, and HFC-134a.
4-162 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
shown below, was the highest emitting end-use (40.8 MMT CO2 Eq.), followed by large retail food, which is part of
the Commercial Refrigeration subsector. Each of the end-use sectors is described in more detail below.
Table 4-127: Emissions of HFCs, PFCs, and CO2 from ODS Substitutes (MMT CO2 Eq.) by Sector
Sector
1990
2005
2018
2019
2020
2021
2022
Refrigeration/Air
Conditioning
+
83.0
122.7
126.5
130.6
139.5
144.6
Commercial
1
1
I
Refrigeration
"1" 111
14.9
1
:::
39.6
40.2
40.6
41.0
41.4
Domestic Refrigeration
+
0.2
1.2
1.2
1.2
1.1
1.0
Industrial Process
l
1
Refrigeration
+ if
1.8
1
13.8
15.0
16.2
17.4
18.6
Transport Refrigeration
+
1.6
6.9
7.4
7.9
8.4
00
00
Mobile Air Conditioning
+ |
61.5
1
1
mm!
28.7
26.6
24.6
22.9
20.8
Residential Stationary
Air Conditioning
+
1.2
26.2
29.4
33.2
41.5
46.4
Commercial Stationary
Air Conditioning
+
1.7
6.2
6.6
6.9
7.3
7.6
Aerosols
0.2
10.2
16.7
17.0
17.3
17.7
17.0
Foams
+
3.5
14.2
14.1
13.7
10.8
11.7
Solvents
+
1.6
2.0
2.0
2.0
2.1
2.1
Fire Protection
+
1.2
2.4
2.5
2.5
2.6
2.6
Total
0.3
99.5
157.9
162.1
166.2
172.7
178.1
+ Does not exceed 0.05 MMT C02 Eq.
Note: Totals may not sum due to independent rounding.
Refrigeration/Air Conditioning
The refrigeration and air-conditioning sector includes a wide variety of equipment types that have historically used
CFCs or HCFCs. End-uses within this sector include motor vehicle air-conditioning, retail food refrigeration,
refrigerated transport (e.g., ship holds, truck trailers, railway freight cars), household refrigeration, residential and
small commercial air-conditioning and heat pumps, chillers (large comfort cooling), cold storage facilities, and
industrial process refrigeration (e.g., systems used in food processing, chemical, petrochemical, pharmaceutical, oil
and gas, metallurgical, and other industries). As the ODS phaseout has taken effect, most equipment has been
retrofitted or replaced to use HFC-based substitutes. Common HFCs in use today in refrigeration/air-conditioning
equipment are HFC-134a, R-410A,119 R-404A, and R-507A.120 Lower-GWP options such as hydrofluoroolefin (HFO)-
1234yf in motor vehicle air-conditioning, R-717 (ammonia) in cold storage and industrial applications, and R-744
(carbon dioxide) and HFC/HFO blends in retail food refrigeration, are also being used. Manufacturers of residential
and commercial air conditioning have announced their plans to use HFC-32 and R-454B121 in the future, and at
least one manufacturer has announced the availability of chillers operating on HFC-32 as of 2023 (Carrier, 2023).
These refrigerants are emitted to the atmosphere during equipment operation (as a result of component failure,
leaks, and purges), as well as at manufacturing (if charged at the factory), installation, servicing, and disposal
events.
119 R-410A contains HFC-32 and HFC-125.
120 R-507A, also called R-507, contains HFC-125 and HFC-143a.
121 R_454B contains HFC-32 and HFO-1234yf.
Industrial Processes and Product Use 4-163
-------
Aerosols
Aerosol propellants are used in metered dose inhalers (MDIs) and a variety of personal care products and
technical/specialty products (e.g., duster sprays and safety horns). Pharmaceutical companies that produce MDIs—
a type of inhaled therapy used to treat asthma and chronic obstructive pulmonary disease—have replaced the use
of CFCs with HFC-propellant alternatives. The earliest ozone-friendly MDIs were produced with HFC-134a, but the
industry is using HFC-227ea as well. Conversely, since the use of CFC propellants in other types of aerosols was
banned in the Unites States in 1978, most non-medical consumer aerosol products have not transitioned to HFCs,
but to "not-in-kind" technologies, such as solid or roll-on deodorants and finger-pump sprays. The transition away
from ODSs in specialty aerosol products has also led to the introduction of non-fluorocarbon alternatives (e.g.,
hydrocarbon propellants) in certain applications, in addition to HFC-134a or HFC-152a. Other low-GWP options
such as HFO-1234ze(E) are being used as well. These propellants are released into the atmosphere as the aerosol
products are used.
Foams
Chlorofluorocarbons and HCFCs have traditionally been used as foam blowing agents to produce polyurethane
(PU), polystyrene, polyolefin, and phenolic foams, which are used in a wide variety of products and applications.
Since the Montreal Protocol, flexible PU foams as well as other types of foam, such as polystyrene sheet,
polyolefin, and phenolic foam, have transitioned almost completely away from fluorocompounds into alternatives
such as CO2 and hydrocarbons. The majority of rigid PU foams have transitioned to HFCs—primarily HFC-134a and
HFC-245fa. Today, these HFCs are used to produce PU appliance, PU commercial refrigeration, PU spray, and PU
panel foams—used in refrigerators, vending machines, roofing, wall insulation, garage doors, and cold storage
applications. In addition, HFC-152a, HFC-134a, and CO2 are used to produce polystyrene sheet/board foam, which
is used in food packaging and building insulation. Low-GWP fluorinated foam blowing agents in use include HFO-
1234ze(E) and HCFO-1233zd(E). Emissions of blowing agents occur when the foam is manufactured as well as
during the foam lifetime and at foam disposal, depending on the particular foam type.
Solvents
Chlorofluorocarbons, methyl chloroform (1,1,1-trichloroethane or TCA), and to a lesser extent carbon tetrachloride
(CCU) were historically used as solvents in a wide range of cleaning applications, including precision, electronics,
and metal cleaning. Since their phaseout, metal cleaning end-use applications have primarily transitioned to non-
fluorocarbon solvents and not-in-kind processes. The precision and electronics cleaning end-uses have transitioned
in part to high-GWP gases, due to their high reliability, excellent compatibility, good stability, low toxicity, and
selective solvency. These applications rely on HFC-43-10mee, HFC-365mfc, HFC-245fa, and to a lesser extent, PFCs.
Electronics cleaning involves removing flux residue that remains after a soldering operation for printed circuit
boards and other contamination-sensitive electronics applications. Precision cleaning may apply to either
electronic components or to metal surfaces, and is characterized by products, such as disk drives, gyroscopes, and
optical components, that require a high level of cleanliness and generally have complex shapes, small clearances,
and other cleaning challenges. The use of these solvents yields fugitive emissions of these HFCs and PFCs.
Fire Protection
Fire protection applications include portable fire extinguishers ("streaming" applications) that originally used halon
1211, and total flooding applications that originally used halon 1301, as well as some halon 2402. Since the
production and import of virgin halons were banned in the United States in 1994, the halon replacement agent of
choice in the streaming sector has been dry chemical, although HFC-236fa is also used to a limited extent. In the
total flooding sector, HFC-227ea has emerged as the primary replacement for halon 1301 in applications that
require clean agents. Other HFCs, such as HFC-23 and HFC-125, are used in smaller amounts. The majority of HFC-
227ea in total flooding systems is used to protect essential electronics, as well as in civil aviation, military mobile
weapons systems, oil/gas/other process industries, and merchant shipping. Fluoroketone FK-5-1-12 is also used as
4-164 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
a low-GWP option and 2-BTP is being use in niche applications. As fire protection equipment is tested or deployed,
emissions of these fire protection agents occur.
Methodology and Time-Series Consistency
Using a Tier 2 method in accordance with the IPCC methodological decision tree, a detailed Vintaging Model of
ODS-containing equipment and products was used to estimate the actual—versus potential—emissions of various
ODS substitutes, including HFCs, PFCs, and CO2. The name of the model refers to the fact that it tracks the use and
emissions of various compounds for the annual "vintages" of new equipment that enter service in each end-use.
The Vintaging Model predicts ODS and ODS substitute use in the United States based on modeled estimates of the
quantity of equipment or products sold each year containing these chemicals and the amount of the chemical
required to manufacture and/or maintain equipment and products over time. Emissions for each end-use were
estimated by applying annual leak rates and release profiles, which account for the lag in emissions from
equipment as they leak over time. By aggregating the data for 80 different end-uses, the model produces
estimates of annual use and emissions of each compound. Further information on the Vintaging Model is
contained in Annex 3.9.
Methodological approaches were applied to the entire time series to ensure time-series consistency from 1990
through 2022.
Uncertainty
Given that emissions of ODS substitutes occur from thousands of different kinds of equipment and from millions of
point and mobile sources throughout the United States, emission estimates must be made using analytical tools
such as the Vintaging Model or the methods outlined in IPCC (2006). Though the model is more comprehensive
than the IPCC default methodology, significant uncertainties still exist with regard to the levels of equipment sales,
equipment characteristics, and end-use emissions profiles that were used to estimate annual emissions for the
various compounds.
The uncertainty analysis quantifies the level of uncertainty associated with the aggregate emissions across the 80
end-uses in the Vintaging Model. In order to calculate uncertainty, functional forms were developed to simplify
some of the complex "vintaging" aspects of some end-use sectors, especially with respect to refrigeration and air-
conditioning, and to a lesser degree, fire extinguishing. These sectors calculate emissions based on the entire
lifetime of equipment, not just equipment put into commission in the current year, thereby necessitating
simplifying equations. The functional forms used variables that included growth rates, emission factors, transition
from ODSs, change in charge size as a result of the transition, disposal quantities, disposal emission rates, and
either stock (e.g., number of air conditioning units in operation) for the current year or ODS consumption before
transition to alternatives began (e.g., in 1985 for most end-uses). Uncertainty was estimated around each variable
within the functional forms based on expert judgment, and a Monte Carlo analysis was performed.
Inputs to the ODS substitutes uncertainty model generally take on a normal distribution with a 90 to 95 percent
confidence interval but do utilize other probability density functions such as a uniform or PERT BETA distribution.
The uncertainty inputs are based on conversations with industry experts and how certain assumptions are
developed in the Vintaging Model. For example, if the Vintaging Model estimates are specifically aligned with
actual reported data, then the uncertainty is decreased. This can be seen with the unitary AC end-use where
annual stock data is aligned with sales data published by the Air-Conditioning, Heating, and Refrigeration Institute
(AHRI). The stock is assumed to be fairly accurate and therefore, uncertainty range for the stock of unitary AC is set
to an upper and lower bound of only 2.5 percent. The most significant sources of uncertainty for the substitution
of ODS source category include the total stock of refrigerant installed in industrial process refrigeration and cold
storage equipment, as well as the charge size for technical aerosols using HFC-134a. For technical aerosols, a
triangular distribution is utilized to apply an asymmetrical range to the inventory value. This is to account for the
Industrial Processes and Product Use 4-165
-------
uncertainty that technical aerosols using HFC-134a might have higher market penetration than what the Vintaging
Model currently estimates.
The results of the Approach 2 quantitative uncertainty analysis are summarized in Table 4-128. Substitution of
ozone depleting substances HFC and PFC emissions were estimated to be between 170.8 and 205.1 MMT CO2 Eq.
at the 95 percent confidence level. This indicates a range of approximately 4.1 percent below to 15.1 percent
above the emission estimate of 178.1 MMT CO2 Eq.
Table 4-128: Approach 2 Quantitative Uncertainty Estimates for HFC and PFC Emissions from
ODS Substitutes (MMT CO2 Eq. and Percent)
Source
Gases
2022 Emission Estimate
(MMTCO. Eq.)
Uncertainty Range Relative to Emission Estimate"
(MMTCO' Eq.) (%)
Lower
Upper
Lower
Upper
Bound
Bound
Bound
Bound
Substitution of Ozone
Depleting Substances
HFCs and
PFCs
178.1
170.8
205.1
-4.1%
+15.1%
a Range of emission estimates predicted by Monte Carlo stochastic simulation for a 95 percent confidence interval.
QA/QC and Verification
For more information on the general QA/QC process applied to this source category, consistent with Volume 1,
Chapter 6 of the 2006IPCC Guidelines, see the QA/QC and Verification Procedures section in the introduction of
the IPPU chapter. Category specific QA/QC findings are described below.
The QA and verification process for individual gases and sources in the Vintaging Model includes review against up-
to-date market information, including equipment stock estimates, leak rates, and sector transitions to new
chemicals and technologies. In addition, comparisons against published emission and consumption sources by gas
and by source are performed when available as described further below. Independent peer reviews of the
Vintaging Model are periodically performed, including one conducted in 2017 (EPA 2018), to confirm Vintaging
Model estimates and identify updates. For the purposes of reporting emissions to protect Confidential Business
Information (CBI), some HFCs and PFCs are grouped into two unspecified mixes of saturated HFCs and other PFCs
and HFOs. The HFCs and PFCs within the unspecified mix of HFCs and PFCs are modelled and verified individually in
the same process as all other gases and sources in the Vintaging Model.
Data from EPA's Greenhouse Gas Reporting Program (GHGRP)122 and emissions of some fluorinated greenhouse
gases estimated for the contiguous United States by scientists at the National Oceanic and Atmospheric
Administration (NOAA) were used to perform additional quality control as specified in 2006 IPCC Guidelines for
National Greenhouse Gas Inventories and the 2019 Refinement to the 2006 IPCC Guidelines for National
Greenhouse Gas Inventories (IPCC 2019). These comparisons are detailed further in Annex 3.9.
Recalculations Discussion
For the current Inventory, updates to the Vintaging Model included updating 2022 growth rates for residential and
commercial unitary air-conditioning to align with annual sales estimates published by AHRI. Projected growth rates
122 For the GHGRP data, EPA verifies annual facility-level and company-level reports through a multi-step process (e.g.,
including a combination of pre-and post-submittal electronic checks and manual reviews by staff) to identify potential errors
and ensure that data submitted to EPA are accurate, complete, and consistent (EPA 2015). Based on the results of the
verification process, EPA follows up with facilities to resolve mistakes that may have occurred. The post-submittals checks are
consistent with a number of general and category-specific QC procedures, including range checks, statistical checks, algorithm
checks, and year-to-year checks of reported data.
4-166 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
were updated for residential unitary air-conditioning to align with projected residential housing available from the
Energy Information Administration (EIA) and commercial unitary air-conditioning growth rates were updated
based on new commercial floorspace growth projections from EIA (EPA 2023a). Growth rates for window units
were updated to align with sales data for Energy Star- and non-Energy Star-certified units (EPA 2023b).
The Vintaging Model was also updated to include the addition of two end-uses representing multi-split air-
conditioning units: small ductless mini-split and multi-split air-conditioning and large ductless mini-split and multi-
split air-conditioning. These end-uses were split from the existing residential unitary air-conditioning end-use.
Stock for ductless mini-split and multi-split air-conditioning systems were estimated based on the EIA Residential
Energy and Consumption Survey (RECS) and growth projected based on annual sales of split systems under 33,000
BTU/h (small mini/multi-splits) and split systems between 33,000 BTU/h and 65,000 BTU/h (large mini/multi-splits)
published by AHRI (EPA 2023c).
In addition, the market size and growth rates of the streaming agents end-use in the Fire Suppression sector was
updated, improving alignment of halon 1211 consumption with the 2022 Fire Suppression Technical Options
Committee (FSTOC) estimates, the lifetime was adjusted to reflect internal inspection timelines, rather than
physical extinguisher lifetimes, and refilling of leaks throughout the lifetime of extinguishers was modeled (EPA
2023d).
Together, these updates increased ODS substitute emissions on average by 0.06 MMT CO2 Eq. (0.6 percent)
between 1990 and 2021, compared to the previous (i.e., 2023) Inventory submission.
Planned Improvements
Future improvements to the Vintaging Model are planned for the Refrigeration and Air-conditioning, Fire
Suppression, and Aerosols sectors. Specifically, refrigerated storage space estimates published biannually from the
United States Department of Agriculture (USDA) are being compared to cold storage warehouse space currently
estimated in the Vintaging Model. Flooding agent fire suppression market transitions are under review to align
more closely with real world activities. In addition, further refinement of HFC consumption in MDIs is expected
from review of data collected on HFC use for MDI production, imports, and exports in response to requests
pursuant to AIM Act regulations for application-specific allowances for MDIs. EPA expects these revisions to be
prepared for the 2025 Inventory submission.
As discussed above, future reporting under the AIM Act may provide useful information for verification purposes
and possible improvements to the Vintaging Model, such as information on HFC stockpiling behaviors. EPA expects
this reporting by late 2023 and incorporation into the 2025 or 2026 report. Should the data suggest structural
changes to the model, such as the handling of stockpiles before use, EPA expects to introduce the revised model
for the 2025 or 2026 Inventory submission.
Several potential improvements to the Inventory were identified in the 2022 Inventory submission based on the
comparisons mentioned above and discussed in Annex 3.9—net supply values from the GHGRP and emission
estimates derived from atmospheric measurements—and remain valid. To estimate HFC emissions for just the
contiguous United States, matching the coverage by the atmospheric measurements, EPA will investigate the
availability of data from Alaska, Hawaii, and U.S. territories. This is planned by the next (i.e., 2025) Inventory
submission. To improve estimates of HFC-125 and HFC-143a, further research into the refrigeration market can be
made. Research in this industry on the shift away from blends such as R-404A or success in lowering emission rates
could be used to improve the Inventory estimate. This is planned for the 2025 Inventory. That said, for the years
where both the atmospheric measurements and the model display a roughly constant emission of HFC-143a at
similar levels, the new results suggest robust estimates for the refrigeration market. Uncertainty estimates by
species would aid in comparisons to atmospheric data. EPA continues to explore the possibility of revising the
Monte Carlo analysis to differentiate between species, starting with the higher-emitted HFCs identified above, in a
future (i.e., 2025) Inventory submission. Reclamation reports and, when available, information gathered under the
AIM Act, could be used to improve the understanding of how chemical moves through the economy and could
resolve some of the temporal effects discussed in Annex 3.9. This would likely require revisions to the basic model
Industrial Processes and Product Use 4-167
-------
structure and could be introduced for the 2026 or 2027 Inventory submission. The additional data from the
atmospheric measurements suggests additional items to investigate. The faster uptick in HFC-32 and HFC-125
emissions suggests additional emissions of R-410A compared to the model's estimation. Further investigation into
the average emission rate, the variability over time of the emission rate, stocks, lifetimes, and other factors will be
investigated for the next (i.e., 2025) Inventory submission.
4.26 Electrical Equipment (CRT Source
Category 2G1)
The largest use of sulfur hexafluoride (SFs), both in the United States and internationally, is as an electrical
insulator and interrupter in equipment that transmits and distributes electricity (RAND 2004). The gas has been
employed by the electric power industry in the United States since the 1950s because of its dielectric strength and
arc-quenching characteristics. It is used in gas-insulated substations, circuit breakers, and other switchgear. SF6 has
replaced flammable insulating oils in many applications and allows for more compact substations in dense urban
areas. Another greenhouse gas emitted in much smaller amounts by the electric power industry is
tetrafluoromethane (CF4), which is mixed with SF6 to avoid liquefaction at low temperatures (Middleton 2000).
While mixed gas circuit breakers are more common in extremely cold climates in geographies outside of the
United States, some U.S. manufacturers of electrical equipment are emitting CF4 during the manufacturing of
equipment designed to hold the SF6/CF4 gas mixture. However, no electrical equipment facilities in the United
States have reported emissions of or equipment using CF4. SF6 emissions exceed PFC emissions from electric power
systems on both a GWP-unweighted and GWP-weighted basis. This reporting category (2G1) includes emissions
from electrical equipment.
Fugitive emissions of SF6 and CF4 can escape from gas-insulated substations and switchgear through seals,
especially from older equipment. The gas can also be released during equipment manufacturing, installation,
servicing, and disposal. Emissions of SF6 and CF4 from equipment manufacturing and from electrical equipment
systems were estimated to be 5.1 MMT CO2 Eq. (0.3 kt) in 2022. This quantity represents a 79 percent decrease
from the estimate for 1990 (see Table 4-129 and Table 4-130). There are a few potential causes for this decrease: a
sharp increase in the price of SF6 during the 1990s and a growing awareness of the environmental impact of SF6
emissions through programs such as EPA's voluntary SF6 Emission Reduction Partnership for Electric Power
Systems (Partnership) and EPA's GHGRP, regulatory drivers at the state and local levels, and research and
development of alternative gases to SF6 that can be used in gas-insulated substations. Utilities participating in the
Partnership have lowered their emission factor from 13 percent in 1999 (kg SF6 emitted per kg of nameplate
capacity) to 0.9 percent in 2022. SF6 emissions reported by electric power systems to EPA's GHGRP have decreased
by 56 percent from 2011 to 2022,123 with much of the reduction seen from utilities that are not participants in the
Partnership. These utilities may be making relatively large reductions in emissions as they take advantage of
relatively large and/or inexpensive emission reduction opportunities (i.e., "low hanging fruit," such as replacing
major leaking circuit breakers) that Partners have already taken advantage of under the voluntary program
123 Analysis of emission trends from facilities reporting to EPA's GHGRP is imperfect due to an inconsistent group of reporters
year to year. A facility that has reported total non-biogenic greenhouse gas emissions below 15,000 metric tons of carbon
dioxide equivalent (MT C02 Eq.) for three consecutive years or below 25,000 MT C02 Eq. for five consecutive years to EPA's
GHGRP can discontinue reporting for all direct emitter subparts. For this sector, most of the variability in the group of reporters
is due to facilities exiting the GHGRP due to being below one of these thresholds; however, facilities must re-enter the program
if their emissions at a later date are above 25,000 MT C02 Eq., which may occur for a variety of reasons, including changes in
facility size and changes in emission rates.
4-168 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
(Ottinger et al. 2014). Total emissions from electrical equipment in 2022 were lower than 2021 emissions,
decreasing by 15.3 percent.
Table 4-129: SF6 and CF4 Emissions from Electric Power Systems and Electrical Equipment
Manufacturers (MMT CO2 Eq.)
1990
2005
2018
2019
2020
2021
2022
Electric Power Systems
Electrical Equipment
Manufacturers
24.3
0.3 1
11.2 1
0-7 1
4.7
0.3
5.7
0.4
5.3
0.5
5.6
0.4
4.8
0.3
Total
24.7
11.9 |
5.0
6.1
5.9
6.0
5.1
Note: Totals may not sum due to independent rounding.
Table 4-130: SF6 and CF4 Emissions from Electric Power Systems and Electrical Equipment
Manufacturers (kt)
1990
2005
2018
2019
2020
2021
2022
SF6 Emissions
1.0 1
0.5
0.2
0.3
0.3
0.3
0.2
CF4 Emissions
+ ¦
+ I
NO
+
+
+
+
+ Does not exceed 0.5 kt.
NO (Not Occurring)
Methodology and Time-Series Consistency
The estimates of emissions from electrical equipment are comprised of emissions from electric power systems and
emissions from the manufacture of electrical equipment. The methodologies for estimating both sets of emissions
are described below.
1990 through 1998 Emissions from Electric Power Systems
Emissions from electric power systems from 1990 through 1998 were estimated based on (1) the emissions
estimated for this source category in 1999, which, as discussed in the next section, were based on the emissions
reported during the first year of EPA's SF6 Emission Reduction Partnership for Electric Power Systems (Partnership),
and (2) the RAND survey of global SF6 emissions. Because most utilities participating in the Partnership reported
emissions only for 1999 through 2011, modeling was used to estimate SF6 emissions from electric power systems
for the years 1990 through 1998. To perform this modeling, U.S. emissions were assumed to follow the same
trajectory as global emissions from this source during the 1990 through 1999 period. To estimate global emissions,
the RAND survey of global SF6 sales was used, together with the following equation for estimating emissions, which
is derived from the mass-balance equation for chemical emissions (Volume 3, Equation 7.3) in the 2006IPCC
Guidelines.124 (Although Equation 7.3 of the 2006 IPCC Guidelines appears in the discussion of substitutes for
ozone-depleting substances, it is applicable to emissions from any long-lived pressurized equipment that is
periodically serviced during its lifetime.)
124 Ideally, sales to utilities in the United States between 1990 and 1999 would be used as a model. However, this information
was not available. There were only two U.S. manufacturers of SF6 during this time period, so it would not have been possible to
conceal sensitive sales information by aggregation.
Industrial Processes and Product Use 4-169
-------
Equation 4-24: Estimation for SF6 Emissions from Electric Power Systems
Emissions (kilograms SFs) = SF6 purchased to refill existing equipment (kilograms) + nameplate capacity of retiring
equipment (kilograms)125
Note that the above equation holds whether the gas from retiring equipment is released or recaptured; if the gas
is recaptured, it is used to refill existing equipment, thereby lowering the amount of SF6 purchased by utilities for
this purpose.
Gas purchases by utilities and equipment manufacturers from 1961 through 2003 are available from the RAND
(2004) survey. To estimate the quantity of SF6 released or recovered from retiring equipment, the nameplate
capacity of retiring equipment in a given year was assumed to equal 81.2 percent of the amount of gas purchased
by electrical equipment manufacturers 40 years previous (e.g., in 2000, the nameplate capacity of retiring
equipment was assumed to equal 81.2 percent of the gas purchased in 1960). The remaining 18.8 percent was
assumed to have been emitted at the time of manufacture. The 18.8 percent emission factor is an average of IPCC
default SFs emission rates for Europe and Japan for 1995 (IPCC 2006). The 40-year lifetime for electrical equipment
is also based on IPCC (2006). The results of the two components of the above equation were then summed to yield
estimates of global SF6 emissions from 1990 through 1999.
U.S. emissions between 1990 and 1999 are assumed to follow the same trajectory as global emissions during this
period. To estimate U.S. emissions, global emissions for each year from 1990 through 1998 were divided by the
estimated global emissions from 1999. The result was a time series of factors that express each year's global
emissions as a multiple of 1999 global emissions. Historical U.S. emissions were estimated by multiplying the factor
for each respective year by the estimated U.S. emissions of SF6 from electric power systems in 1999 (estimated to
be MMTCCh Eq.).
Two factors may affect the relationship between the RAND sales trends and actual global emission trends. One is
utilities' inventories of SF6 in storage containers. When SF6 prices rise, utilities are likely to deplete internal
inventories before purchasing new SF6 at the higher price, in which case SF6 sales will fall more quickly than
emissions. On the other hand, when SF6 prices fall, utilities are likely to purchase more SF6 to rebuild inventories, in
which case sales will rise more quickly than emissions. This effect was accounted for by applying 3-year smoothing
to utility SFs sales data. The other factor that may affect the relationship between the RAND sales trends and
actual global emissions is the level of imports from and exports to Russia and China. SF6 production in these
countries is not included in the RAND survey and is not accounted for in any another manner by RAND. However,
atmospheric studies confirm that the downward trend in estimated global emissions between 1995 and 1998 was
real (see the Uncertainty discussion below).
1999 through 2022 Emissions from Electric Power Systems
Emissions from electric power systems from 1999 to 2022 were estimated based on: (1) reporting from utilities
participating in EPA's SF6 Emission Reduction Partnership for Electric Power Systems (Partners), which began in
1999; (2) reporting from utilities covered by EPA's GHGRP, which began in 2012 for emissions occurring in 2011
(GHGRP-Only Reporters); (3) SF6 emissions from California estimated by the California Air Resources Board (CARB)
and (4) the relationship between utilities' reported emissions and their transmission miles as reported in the 2001,
2004, 2007, 2010, 2013, and 2016 Utility Data Institute (UDI) Directories of Electric Power Producers and
Distributors (UDI 2001, 2004, 2007, 2010, 2013, and 2017), and 2019, 2020, and 2021 Homeland Infrastructure
Foundation-Level Data (HIFLD) (HIFLD 2019, 2020, and 2021), which was applied to the electric power systems that
do not report to EPA (Non-Reporters). Total U.S. transmission mileage was interpolated between 2016 and 2019 to
estimate transmission mileage of electric power systems in 2017 and 2018. (Transmission miles are defined as the
miles of lines carrying voltages above 34.5 kV).
125 Nameplate capacity is defined as the amount of SF6 within fully charged electrical equipment.
4-170 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
Partners
Over the period from 1999 to 2022, Partner utilities, which for inventory purposes are defined as utilities that
either currently are or previously have been part of the Partnership,126 represented 49 percent, on average, of
total U.S. transmission miles. Partner utilities estimated their emissions using a Tier 3 utility-level mass balance
approach (IPCC 2006). If a Partner utility did not provide data for a particular year, emissions were interpolated
between years for which data were available or extrapolated based on Partner-specific transmission mile growth
rates. In 2012, many Partners began reporting their emissions (for 2011 and later years) through EPA's GHGRP
(discussed further below) rather than through the Partnership. In 2022, less than 1 percent of the total emissions
attributed to Partner utilities were reported through Partnership reports. Approximately 99.6 percent of the total
emissions attributed to Partner utilities were reported and verified through EPA's GHGRP.127 Overall, the emission
rates reported by Partners have decreased significantly throughout the time series.
Non-Partners
Non-Partners consist of two groups: Utilities that have reported to the GHGRP beginning in 2012 (reporting 2011
emissions) or later years (GHGRP-only Reporters) and utilities that have never reported to the GHGRP (Non-
Reporters). EPA's GHGRP requires users of SF6 in electric power systems to report emissions if the facility has a
total SFs nameplate capacity that exceeds 17,820 pounds. (This quantity is the nameplate capacity that would
result in annual SF6 emissions equal to 25,000 metric tons of C02 equivalent at the historical emission rate reported
under the Partnership). As under the Partnership, electric power systems that report their SF6 emissions under
EPA's GHGRP are required to use the Tier 3 utility-level mass-balance approach. GHGRP-Only Reporters accounted
for 16 percent of U.S. transmission miles and 14 percent of estimated U.S. emissions from electric power system in
2022.128
From 1999 through 2018, emissions from both GHGRP-only Reporters and Non-Reporters were estimated in the
same way. From 1999 through 2008, emissions were estimated using the results of a regression analysis that
correlated the 1999 emissions from Partner utilities with their 1999 transmission miles.129 The 1999 regression
coefficient (emission factor) was held constant through 2008 and multiplied by the transmission miles estimated
for the non-Partners for each year.
The 1999 regression equation for Non-Partners was developed based on the emissions reported by a subset of
Partner utilities who reported non-zero emissions and non-zero transmission miles (representing approximately 50
percent of total U.S. transmission miles). The regression equation for 1999 is displayed in the equation below.
126 Starting in the 1990 to 2015 Inventory, partners who had reported three years or less of data prior to 2006 were removed.
Most of these Partners had been removed from the list of current Partners but remained in the Inventory due to the
extrapolation methodology for non-reporting partners.
127 Only data reported as of August 21, 2023 are used in the emission estimates for the prior year of reporting. Emissions for
Partners that did not report to the Partnership or GHGRP are extrapolated for three years using a utility-specific transmission
mile growth rate. After four consecutive years of non-reporting they are included in the 'non-reporting Partners' category. It
should be noted that data reported through EPA's GHGRP must go through a verification process. For electric power systems,
verification involved a series of electronic range, completeness, and algorithm checks for each report submitted.
128 GHGRP-reported and Partner transmission miles from a number of facilities were equal to zero with non-zero emissions.
These facilities emissions were added to the emissions totals for their respective parent companies when identifiable and not
included in the regression equation when not identifiable or applicable. Other facilities reported non-zero transmission miles
with zero emissions, or zero transmission miles and zero emissions. These facilities were not included in the development of the
regression equations (discussed further below). These emissions are already implicitly accounted for in the relationship
between transmission miles and emissions.
129 In the United States, SF6 is contained primarily in transmission equipment rated above 34.5 kV.
Industrial Processes and Product Use 4-171
-------
Equation 4-25: Regression Equation for Estimating SF6 Emissions of Non-Reporting Facilities
in 1999
Emissions (kg) = 0.771 x Transmission Miles
The 1999 emission factor (0.77 SF6 emissions/Transmission Miles) for the non-Partners was held constant to
estimate non-Partner emissions from 2000-2008. Non-partner emissions were assumed to decrease beginning in
2009, trending toward the regression coefficient (emission factor) calculated for the GHGRP-only reporters based
on their reported 2011 emissions and transmission miles. Emission factors for 2009 and 2010 were linearly
interpolated between the 1999 and 2011 emission factors. For 2009, the emissions of non-Partners were
estimated by multiplying their transmission miles by the interpolated 2009 emission factor (0.65 kg/transmission
mile).
The 2011 regression equation was developed based on the emissions reported by GHGRP-Only Reporters who
reported non-zero emissions and non-zero transmission miles (representing approximately 23 percent of total U.S.
transmission miles). The regression equation for 2011 is displayed below.
Equation 4-26: Regression Equation for Estimating SF6 Emissions of GHGRP-Only Reporters in
2011
Emissions (kg) = 0.397 x Transmission Miles
For 2011 and later years, the emissions of GHGRP-only reporters were generally equated to their reported
emissions, unless they did not report. The emissions of GHGRP-only reporters that have years of non-reporting
between reporting years are gap filled by interpolating between reported values.
For 2010 and later years, the emissions of non-Reporters were estimated by multiplying their transmission miles by
the estimated 2010 emission factor (0.52 kg/transmission mile), which was held constant from 2010 through 2022.
Off-ramping GHGRP Facilities
The GHGRP program has an "off-ramp" provision (40 CFR Part 98.2(i)) that exempts facilities from reporting under
certain conditions. If reported total greenhouse gas emissions are below 15,000 metric tons of carbon dioxide
equivalent (MT CO2 Eq.) for three consecutive years or below 25,000 MT CO2 Eq. for five consecutive years, the
facility may elect to discontinue reporting. Emissions of GHGRP reporters that have off-ramped are extrapolated
for three years of non-reporting using a utility-specific transmission mile growth rate, unless the utility has
transmission mileage in California. After three consecutive years of non-reporting, emissions for facilities (except
those in California) that off-ramped from GHGRP were estimated using an emissions rate derived from the
reported emissions and transmission miles of GHGRP-only reporters in the respective year. For facilities in
California, a California-specific emissions rate is used as described in the following section.
Table 4-131: GHGRP-only Average Emission Rate (kg per mile)
Year
2011
2018
2019
2020
2021
2022
Average emission rate
0.43 I
0.22
0.29
0.27
0.25
0.22
Table 4-132: Categorization of Utilities and Timeseries for Application of Corresponding
Emission Estimation Methodologies
Categorization of Utilities Timeseries
Partners
1999 - 2021
Non-Partners (GHGRP-Only)
2011-2021
Non-Partners (Remaining Non-
Reporting Utilities)
1999-2021
Off-ramping GHGRP Facilities
2017-2021
4-172 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
California
CARB reports the total SF6 emissions from electrical equipment within the state of California (CARB 2023). Because
California utilities are required to report their SF6 emissions to CARB even when they are not required to report to
the GHGRP, CARB's estimates of California SF6 emissions are expected to be more accurate for the California
utilities that do not report to GHGRP than the methodology described above. As a result, the CARB SF6 emissions
estimates are used as California's contribution to the national total for 2011-2022, except in years where CARB's
estimate is smaller than the California estimates reported to EPA or years for which CARB has not published
estimates. Since CARB's emissions estimates include emissions from facilities that do not report to GHGRP,
emissions for California GHGRP reporters that have off-ramped are not extrapolated. Specifically, CARB estimates
are used for 2011, 2012, 2013, 2014, 2017, 2018, 2019, 2020, and 2021.
For each utility with transmission mileage in California, the GHGRP or voluntarily reported emissions attributed to
California for that utility were determined using the percentage of that utility's transmission mileage within
California based on data from HIFLD. These emissions across all California utilities were summed to find the
California emissions that were reported through GHGRP or voluntarily to the EPA. Then, if CARB's emissions
estimates for the reporting year were larger than the those from GHGRP and voluntary reporting, CARB's
emissions replaced the California emissions from GHGRP and voluntary reporting.
If CARB's emissions estimates were lower than the California emissions from GHGRP and voluntary reporting, it is
assumed there is likely an error, as this would imply negative emissions by GHGRP non-reporters. This was the case
in 2015 and 2016. For these years, the GHGRP and voluntarily reported emissions from California are retained, and
emissions from non-reporting utilities are estimated using a California-specific SF6 emissions rate, which is based
on CARB emission data. The California SF6 emissions rate of 0.41 lbs SF6 per transmission mile is found by taking
the average of CARB emissions divided by the total California transmission mileage in years where CARB estimates
are larger. Emissions from California non-reporting utilities are then found by multiplying the California SF6
emissions rate by the California transmission mileage from non-reporting utilities. This methodology is also used if
CARB has not published emissions estimates for a particular year. CARB has not yet published estimates for 2022.
Table 4-133: California GHGRP and Voluntarily Reported SF6 Emissions Compared to CARB's
SF6 Emissions (MMT CO2 Eq.)
2011
2015
2016
2017
2018
2019
2020
2021
2022
CA GHGRP and Voluntary
0.19
0.16
0.24
0.12
0.11
0.14
0.20
0.14
0.15
CARB (CARB 2023)
0.24
1
0.14
0.10
0.18
0.14
0.17
0.24
0.24
NE
Final CA
0.24
0.21
0.29
0.18
0.14
0.17
0.24
0.24
0.20
NE (Not Estimated)
Total Industry Emissions
Total electric power system emissions from 1999 through 2022 were determined for each year by summing the
Partner reported and estimated emissions (reported data was available through the EPA's SF6 Emission Reduction
Partnership for Electric Power Systems), the GHGRP-only reported emissions, off-ramping GHGRP Facilities (non-
reporters), non-reporters who eventually report to GHGRP, and the non-reporting utilities' emissions (except
California). Then, the California GHGRP and voluntarily reported emissions are subtracted from the total and
replaced with CARB's emissions (or GHGRP and voluntarily reported emissions plus California non-reporting
utilities' emissions).
Non-Partner Transmission Miles
Data on transmission miles for each Non-Reporter for the years 2000, 2003, 2006, and 2009, 2012, and 2016 were
obtained from the 2001, 2004, 2007, 2010, 2013, and 2017 UDI Directories of Electric Power Producers and
Distributors, respectively (UDI 2001, 2004, 2007, 2010, 2013, and 2017). For 2019 to 2022 non-reporter
transmission mileage was derived by subtracting reported transmission mileage data from the total U.S.
Industrial Processes and Product Use 4-173
-------
transmission mileage from 2019 to 2022 HIFLD Data (HIFLD 2019, 2020, 2021, and 2022). The following trends in
transmission miles have been observed over the time series:
• The U.S. transmission system grew by over 22,000 miles between 2000 and 2003 yet declined by almost
4,000 miles between 2003 and 2006. Given these fluctuations, periodic increases are assumed to occur
gradually. Therefore, transmission mileage was assumed to increase at an annual rate of 1.2 percent
between 2000 and 2003 and decrease by 0.20 percent between 2003 and 2006.
• The U.S. transmission system's annual growth rate grew to 1.7 percent from 2006 to 2009 as transmission
miles increased by more than 33,000 miles.
• The annual growth rate for 2009 through 2012 was calculated to be 1.4 percent as transmission miles
grew yet again by over 29,000 miles during this time period.
• The annual transmission mile growth rate for 2012 through 2016 was calculated to be 0.2 percent, as
transmission miles increased by approximately 5,500 miles.
• The annual transmission mile growth rate for 2016 through 2020 was calculated to be 0.9 percent, as
transmission miles increased by approximately 26,000 miles.
• The annual transmission mile growth rate for 2020 through 2021 was calculated to be 2.2 percent, as
transmission miles increased by approximately 16,000 miles.
• The annual transmission mile growth rate for 2021 through 2022 was calculated to be 0.7 percent, as
transmission miles increased by approximately 5,500 miles.
Transmission miles for each year for non-reporters were calculated by interpolating between UDI reported values
obtained from the 2001, 2004, 2007, 2010, 2013 and 2017 UDI directories and HIFLD data for 2019 and
subsequent years. In cases where a non-reporter previously reported the GHGRP or the Partnership, transmission
miles were interpolated between the most recently reported value and the next available UDI value.
1990 through 2022 Emissions from Manufacture of Electrical Equipment
Three different methods were used to estimate 1990 to 2022 emissions from original electrical equipment
manufacturers (OEMs).
• OEM SFs emissions from 1990 through 2000 were derived by assuming that manufacturing emissions
equaled 10 percent of the quantity of SF6 provided with new equipment. The 10 percent emission rate is
the average of the "ideal" and "realistic" manufacturing emission rates (4 percent and 17 percent,
respectively) identified in a paper prepared under the auspices of the International Council on Large
Electric Systems (CIGRE) in February 2002 (O'Connell et al. 2002). The quantity of SF6 provided with new
equipment was estimated based on statistics compiled by the National Electrical Manufacturers
Association (NEMA). These statistics were provided for 1990 to 2000.
• OEM SFs emissions from 2000 through 2010 were estimated by (1) interpolating between the emission
rate estimated for 2000 (10 percent) and an emission rate estimated for 2011 based on reporting by
OEMs through the GHGRP (5.7 percent), and (2) estimating the quantities of SF6 provided with new
equipment for 2001 to 2010. The quantities of SF6 provided with new equipment were estimated using
Partner reported data and the total industry SF6 nameplate capacity estimate (156.5 MMT CO2 Eq. in
2010). Specifically, the ratio of new nameplate capacity to total nameplate capacity of a subset of
Partners for which new nameplate capacity data was available from 1999 to 2010 was calculated. These
ratios were then multiplied by the total industry nameplate capacity estimate for each year to derive the
amount of SF6 provided with new equipment for the entire industry. Additionally, to obtain the 2011
emission rate (necessary for estimating 2001 through 2010 emissions), the estimated 2011 emissions
(estimated using the third methodology listed below) were divided by the estimated total quantity of SF6
4-174 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
provided with new equipment in 2011. The 2011 quantity of SF6 provided with new equipment was
estimated in the same way as the 2001 through 2010 quantities.
• OEM CF4 emissions from 1991 through 2010 were estimated by using an average ratio of reported SF6 and
CF4 emissions from 2011 through 2013. This ratio was applied to the estimated SF6 emissions for 1991
through 2010 to arrive at CF4 emissions. CF4 emissions are estimated starting in 1991 and assumed zero
prior to 1991 based on the entry of the CF4/SF6 gas mixture into the market (Middleton 2000).
• OEM emissions from 2011 through 2022 were estimated using the SF6 and CF4 emissions from OEMs
reporting to the GHGRP, and an assumption that these reported emissions account for a conservatively
low estimate of 50 percent of the total emissions from all U.S. OEMs (those that report and those that do
not).
• OEM SFs emissions from facilities off-ramping from the GHGRP were determined by extrapolation. First,
emission growth rates were calculated for each reporting year for each OEM reporting facility as well as
an average emissions growth rate (2011 through 2022). Averages of reported emissions from last three
consecutive reporting years were multiplied by the average growth rate for each off-ramping OEM to
estimate emissions for the non-reporting year(s).
Methodological approaches were applied to the entire time series to ensure time-series consistency from 1990
through 2022.
Uncertainty
To estimate the uncertainty associated with emissions of SF6 and CF4 from electrical equipment, uncertainties
associated with four quantities were estimated: (1) emissions from Partners, (2) emissions from GHGRP-Only
Reporters, (3) emissions from Non-Reporters, and (4) emissions from manufacturers of electrical equipment. A
Monte Carlo analysis was then applied to estimate the overall uncertainty of the emissions estimate.
Total emissions from the SF6 Emission Reduction Partnership include emissions from both reporting (through the
Partnership or EPA's GHGRP) and non-reporting Partners. For reporting Partners, individual Partner-reported SF6
data was assumed to have an uncertainty of +/-10 percent. Based on a Monte Carlo analysis, the cumulative
uncertainty of all Partner-reported data was estimated to be 4.5 percent. The uncertainty associated with
extrapolated or interpolated emissions from non-reporting Partners was assumed to be 20 percent.
For GHGRP-Only Reporters, reported SF6 data was assumed to have an uncertainty of 10 percent. Based on a
Monte Carlo analysis, the cumulative uncertainty of all GHGRP-Only reported data was estimated to be 7.4
percent.
As discussed below, EPA has substantially revised its method for estimating emissions from non-Reporters,
assuming that the average emission rate of non-Reporters has declined much more slowly than the average
emission rate of reporting facilities rather than declining at the same rate. This assumption brings the U.S. SF6
emissions estimated in this Inventory into better agreement with the U.S. SF6 emissions inferred from atmospheric
observations. However, it must be emphasized that the actual emission rates of non-Reporters remain unknown. It
is possible that they are lower or even higher than estimated here. One possibility is that SF6 sources other than
electric power systems are contributing to the emissions inferred from atmospheric observations, implying that
the emissions from non-Reporters are lower than estimated here. Another is that the emissions inferred from
atmospheric measurements are over- (or under-) estimated, implying that emissions from non-Reporters could be
either lower or higher than estimated here. These uncertainties are difficult to quantify and are not reflected in the
estimated uncertainty below. The estimated uncertainty below accounts only for the two sources of uncertainty
associated with the regression equations used to estimate emissions in 2019 from Non-Reporters: (1) uncertainty
in the coefficients (as defined by the regression standard error estimate), and (2) the uncertainty in total
transmission miles for Non-Reporters. Uncertainties were also estimated regarding (1) estimates of SF6 and CF4
emissions from OEMs reporting to EPA's GHGRP, and (2) the assumption on the percent share of OEM emissions
from OEMs reporting to EPA's GHGRP.
Industrial Processes and Product Use 4-175
-------
The results of the Approach 2 quantitative uncertainty analysis are summarized in Table 1-57. electrical equipment
emissions were estimated to be between 3.8 and 6.4 MMT CO2 Eq. at the 95 percent confidence level, a range of
approximately 25 percent below and 25 percent above the emission estimate of 5.1 MMT CO2 Eq. CF4 emissions
were estimated to be between 0.000006 and 0.000009 MMT CO2 Eq. at the 95 percent confidence level, a range of
approximately 20 percent below and 20 percent above the emission estimate of 0.0000074 MMT CO2 Eq.
Table 4-134: Approach 2 Quantitative Uncertainty Estimates for SF6 and CF4 Emissions from
Electrical Equipment (MMT CO2 Eq. and Percent)
Source
Gas
2022 Emission Estimate
(MMT CO . Eq.)
Uncertainty Range Relative to 2022 Emission Estimate-1
(MMTCO' Eq.) (%)
Lower
Upper
Lower
Upper
Bound
Bound
Bound
Bound
Electrical Equipment
sf6
5.1
3.8
6.4
-25%
+25%
Electrical Equipment
cf4
0.0000074
0.000006
0.000009
-20%
+20%
a Range of emission estimates predicted by Monte Carlo stochastic simulation for a 95 percent confidence interval.
In addition to the uncertainty quantified above for the 2022 estimate, there is uncertainty associated with the
emission rates of GHGRP-only facilities before 2011 and of non-Reporters throughout the time series. As noted
above in the discussion of the uncertainty of non-Reporters for 2022, these uncertainties are difficult to quantify.
There is also uncertainty associated with using global SF6 sales data to estimate U.S. emission trends from 1990
through 1999. However, the trend in global emissions implied by sales of SF6 appears to reflect the trend in global
emissions implied by changing SF6 concentrations in the atmosphere. That is, emissions based on global sales
declined by 29 percent between 1995 and 1998 (RAND 2004), and emissions based on atmospheric measurements
declined by 17 percent over the same period (Levin et al. 2010).
Several pieces of evidence indicate that U.S. SF6 emissions were reduced as global emissions were reduced. First,
the decreases in sales and emissions coincided with a sharp increase in the price of SF6 that occurred in the mid-
1990s and that affected the United States as well as the rest of the world. A representative from DILO, a major
manufacturer of SF6 recycling equipment, stated that most U.S. utilities began recycling rather than venting SF6
within two years of the price rise. Finally, the emissions reported by the one U.S. utility that reported its emissions
for all the years from 1990 through 1999 under the Partnership showed a downward trend beginning in the mid-
1990s.
/erification
For more information on the general QA/QC process applied to this source category, consistent with Volume 1,
Chapter 6 of the 2006IPCC Guidelines, see the QA/QC and Verification Procedures section in the introduction of
the IPPU chapter and Annex 8 for more details. Category specific QC findings are described below.
For the GHGRP data, EPA verifies annual facility-level reports through a multi-step process (e.g., including a
combination of pre-and post-submittal electronic checks and manual reviews by staff) to identify potential errors
and ensure that data submitted to EPA are accurate, complete, and consistent (EPA 2015).130 Based on the results
of the verification process, EPA follows up with facilities to resolve mistakes that may have occurred. The post-
submittals checks are consistent with a number of general and category-specific QC procedures including: range
checks, statistical checks, algorithm checks, and year-to-year checks of reported data and emissions.
130 GHGRP Report Verification Factsheet. See https://www.epa.gov/sites/production/files/2Q15-
07/documents/ehgrp verification factsheet.pdf.
4-176 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
Additionally, EPA provides additional quality control for the SF6 emissions estimates using atmospheric derived
estimates for comparison. The 2019 Refinement to the 2006IPCC Guidelines for National Greenhouse Gas
Inventories (IPCC 2019) Volume 1: General Guidance and Reporting, Chapter 6: Quality Assurance, Quality Control
and Verification notes that atmospheric concentration measurements can provide independent data sets as a basis
for comparison with inventory estimates. Further, it identifies fluorinated gases as particularly suited for such
comparisons. The 2019 Refinement makes this conclusion for fluorinated gases based on their lack of significant
natural sources,131 their generally long atmospheric lifetimes, their well-known loss mechanisms, and the potential
uncertainties in bottom-up inventory methods for some of their sources. Unlike non-fluorinated greenhouse gases
(CO2, Cm, and N2O), SFs has no significant natural sources; therefore, the SF6 estimates derived from atmospheric
measurements are driven overwhelmingly by anthropogenic emissions. The 2019 Refinement provides guidance on
conducting such comparisons (as summarized in Table 6.2 of IPCC (2019) Volume 1, Chapter 6) and provides
guidance on using such comparisons to identify areas of improvement in national inventories (as summarized in
Box 6.5 of IPCC (2019) Volume 1, Chapter 6). Emission estimates derived from atmospheric measurements of SF6
made at NOAA and described in Hu et al. (2022) were used to perform a comparison to the inventory estimates.
This comparison resulted in changes to historical emission estimates, as more thoroughly described in the previous
Inventory cycle (EPA 2022). No further changes were made to the electrical equipment estimates for the current
(i.e., 1990 through 2022) Inventory based on this comparison.
Recalculations Discussion
Several updates to activity data led to recalculations of previous Inventory results. The major updates are as
follows:
• As discussed in the methodology above, CARB estimates of SF6 emissions from electrical equipment in
California were used as California's contribution to the national total starting in 2011, except in cases
where CARB emissions were lower than GHGRP and voluntarily reported emissions from California, or in
years where CARB has not published SF6 emissions estimates.
• Updates were made to reporter emissions where facilities had resubmitted data.
• A correction was made to estimate 2021 nameplate capacities for two off-ramping utilities, which off-
ramped in 2021; these estimations were inadvertently omitted in the previous Inventory.
• SFs emissions from electrical equipment manufacturing was corrected due to an erroneous data pull in
the previous Inventory. This caused emissions to increase in years 2011 through 2019.
• Partner transmission mileage used for calculating average share of Partner utilities across the time series
and for estimating nameplate capacity for non-reporting utilities was corrected. The calculation was
previously referencing partner transmission mileage for the prior year for 2013 through 2021.
Planned Improvements
EPA plans to revisit the methodology for determining emissions from the manufacture of electrical equipment, in
particular, the assumption that emissions reported by OEMs account for a conservatively low estimate of 50
percent of the total emissions from all U.S. OEMs. Additional market research will be required to confirm or modify
the assumptions regarding the portion of industry not reporting to the GHGRP program.
131 See Harnisch and Eisenhauer (1998).
Industrial Processes and Product Use 4-177
-------
4.27 SF6 and PFCs from Other Product Use
(CRT Source Category 2G.2)
There are a variety of other products and processes that use fluorinated greenhouse gases. This section estimates
emissions of sulphur hexafluoride (SFs) and perfluorocarbons (PFCs) from other product use (CRT Source Category
2G.2), including military and scientific applications. Many of these applications utilize SF6 or PFCs to exploit their
unique chemical properties, such as the high dielectric strength of SF6 and the stability of PFCs. Emission profiles
from these processes may vary greatly, ranging from immediate and unavoidable release of all of the chemical to
largely avoidable, delayed release from leak-tight products after decades of use.
Military applications employ SFsand PFCs in many processes, such as the use of SFsin the radar systems of military
reconnaissance planes of the Boeing E-3A type, commonly known as Airborne Warning and Control Systems
(AWACS). These systems use SF6 to prevent electric flashovers in the hollow conductors of the antenna, where
voltages can reach up to 135 kilovolts (kV). During ascent of the planes, SF6 is automatically released from the
AWACS to maintain appropriate pressure difference between the system and the outside air. During descent, the
system is automatically charged with SF6 from an SF6 container on board. Most emissions occur during ascent but
may also occur from system leakage during other phases of flight or during time on the ground. Emissions from
AWACS are largely dependent on the number of active planes and sorties (take-offs) per year.
Other uses of SF6 in military applications include the oxidation of lithium in navel torpedoes and infrared decoys.
SFs has also been documented for use in the quieting of torpedo propellers, as well as a by-product of the
processing of nuclear material for the production of fuel and nuclear warheads.
Military electronics are believed to be a key application for PFC heat transfer fluids, particularly in areas such as
ground and airborne radar avionics, missile guidance systems, and sonar. PFCs may also be used to cool electric
motors, especially for equipment where noise reduction is a priority (e.g., submarines). The specific PFCs used in
military applications are similar to heat transfer fluids identified in the electronics industry (see Section 4.24). PFCs
are typically contained in a closed system, so the emissions are most likely to occur during the manufacture,
maintenance, and disposal of equipment.
SFs and PFCs are also employed in several scientific applications, such as for use in particle accelerators. Particle
accelerators can be found in university and research settings, as well as in industrial and medical applications. SF6 is
typically used as an insulating gas and is operated in a vessel exceeding atmospheric pressure. The amount of SF6
used in particle accelerators is largely dependent on the terminal voltage of the unit. Emissions of SF6 typically
occur when SF6 is transferred to storage tanks while maintenance is occurring, when pressure relief valves are
actuated, and through slow leaks. The emission and charge assumptions for industrial and medical particle
accelerators differ from those of university and research accelerators, as discussed in the methodology below.
PFCs (particularly PFC-14) may also be used in particle accelerators as particle detectors or counters (Workman
2022).
SFs may also be employed in other high-voltage scientific equipment, including lasers, x-rays, and electron
microscopes. SF6 emission estimates for this equipment were not disaggregated from particle accelerators for this
Inventory.
There is a range of unidentified processes that also use SF6 and PFCs, such as R&D activities. PFCs are likely used
primarily as heat transfer fluids (HTFs). Emissions reported for these unknown activities group under "Other
Scientific Applications".
Emissions of SF6 and PFCs from the applications outlined above are presented in Table 4-135.
4-178 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
Table 4-135: SF6 and PFC Emissions from Other Product Use (MMT CO2 Eq.)
Year
1990
2005
2018
2019
2020
2021
2022
sf6
0.6
0.6
0.4
0.4
0.3
+
0.3
Total AWACs
0.6
0.6
0.4
0.4
0.3
+
0.3
sf6
0.3
0.3
0.0
0.0
0.0
0.0
0.0
PFCs
0.1 -
0.1 1
0.2
0.2
0.2
0.1
0.2
Total Other Military Applications
0.4
0.4
0.2
0.2
0.2
0.1
0.2
sf6
0.4 !
°-5 f
0.4
0.2
0.1
0.2
0.2
PFC-14
+
+
+
+
+
+
+
Total Particle Accelerators
0.4
0.5
0.4
0.2
0.1
0.2
0.2
sf6
+
+
+
+
0.1
0.2
0.1
PFCs
+ 1
+ 1
+
+
+
+
+
Total Other Scientific Applications
+
+
+
0.1
0.2
0.1
Total Other Product Use
1.4
1.5
0.9
0.8
0.7
0.5
0.8
+ Does not exceed 0.05 MMT C02 Eq.
Note: PFC subtotals include estimates for HFEs. Totals may not sum due to independent rounding.
Methodology and Time-Series Consistency
Emissions are based primarily on data reported through the Federal Energy Management Program (FEMP).
However, the availability of data from FEMP differs across the 1990 through 2022 time series. Consequently,
additional emission estimates were made through utilizing methodologies from the IPCC. Emissions from military
applications and scientific applications were estimated separately, and the approaches are described immediately
below.
Military Applications
1990 through 2007
FEMP data was not readily available for the 1990 to 2007 period as the first reporting year was in 2008. Thus for
this period, estimated SF6 emissions from AWACS were calculated based using the IPCC Tier 1 methodology (IPCC
2006). IPCC provided a default emissions factor of 740kg of SF6 per plane per year. It was assumed that the U.S.
AWAC fleet between 1990 through 2022 remained constant at 33 planes. The IPCC methodology was utilized for all
years from 1990 to 2007.
Emissions for other military applications were estimated by taking the average of the emissions estimated for
other applications as described in the next section for first five FEMP reporting years (i.e., 2008 through 2012) and
held constant between 1990 through 2007. Emissions from AWACS were not specifically reported by FEMP; the
estimates developed for AWACS using the IPCC methodology above were employed to determine emissions from
remaining unidentified military applications.
2008 through 2021
For the period 2008 through 2021, the Department of Defense (DOD) reported emission data through FEMP which
were used to develop estimates for SF6 and PFCs from other military applications. SF6 emission estimates
developed for AWACS using the IPPC Tier 1 methodology (see 1990 through 2007) were compared against SF6
emissions reported by DOD between 2008 and 2021. In years where SF6 emissions reported by DOD was smaller
than those estimated using the IPCC Tier 1 methodology, DOD-reported emissions were assumed to account for
total AWAC emissions; in years where DOD emissions were greater than the calculated AWAC emissions, the
remainder is assumed to be from other SF6 applications.
Emissions from PFCs, HFEs, and other perfluoro compounds are directly reported by DOD. In years where there are
data gaps from FEMP between two reporting years, expected emissions were interpolated.
Industrial Processes and Product Use 4-179
-------
2022
Estimates for emissions from AWACS and other military uses for 2022 were determined by taking an average of
the previous five reporting years (i.e., 2017 through 2021).
Scientific Applications
1990 through 2007
For the period 1990 through 2007, where no reported data is available from the Department of Energy (DOE),
estimates for emissions of SF6 and PFCs from other product use at Department of Energy Laboratories were
determined by taking an average of the first five reporting years (i.e., 2008 through 2012) and held constant from
1990 to 2007.
SFs emissions from other (non-DOE) research and industrial particle accelerators in the United States was
calculated based on the IPCC Tier 1 methodology for estimating emissions from industrial and university/research
particle accelerators. Default emission factors, charge sizes, and usage rates are provided by size and type of
accelerator in the IPCC methodology. These default assumptions were multiplied against the number of particle
accelerators estimated to be active in the United States by year. This methodology remained the same from 1990
to 2007.
2008 through 2021
For the period 2008 through 2021, SF6 and PFC emissions from government particle accelerators and other
scientific equipment were developed using DOE-reported emissions. SF6 and PFC emissions from particle
accelerators were directly reported by DOE. Other fugitive emissions reported by DOE for SF6 were assumed to
represent emissions from particle accelerators and other scientific equipment, as well as two DOE-managed power
facilities (WAPA and BPA).132 Emissions from these two facilities were subtracted out to present only SF6 emissions
from scientific equipment. Reported fugitive emissions for PFC-14 were assumed to wholly represent particle
accelerator applications. SF6 emissions from non-government particle accelerators were estimated using the IPPC
Tier 1 methodology used for 1990 through 2007.
Process emissions from other applications for SF6 and PFCs were reported by DOE for activities such as R&D, and
these emissions were summed by gas. However, the estimates presented here do not include emissions reported
for semiconductor research and manufacture, or from refrigeration and air conditioning. Emissions from additional
PFCs, HFEs, and other perfluoro compounds are directly reported by DOE and are reported as "Other
Applications." Emissions reported to FEMP were generally calculated based on consumption data. In a number of
years, negative values for emissions were reported due to more gas being returned to supply than purchased in a
given year. When negative values were reported, EPA took the average of that year and the proceeding and
following year and applied that value to all three years. This 3-year average was assumed to be more
representative of actual emissions.
In years where there are data gaps between two reporting years, emissions were interpolated.
2022
For emission estimates developed using DOE-reported emissions, estimates for 2022 were determined by taking
an average of the previous five reporting years (i.e., 2017 through 2021). SF6 emissions from non-government
132 DOE-reported fugitive emissions for SF6 and PFCs includes emissions from high-voltage scientific equipment such as lasers,
x-rays, and electron microscopes. Emissions from this equipment is included in the particle accelerators total.
4-180 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
particle accelerators were estimated using the same IPPC Tier 1 methodology used for 2008 through 2021.
Uncertainty
A quantitative uncertainty analysis of this source category was performed using the IPCC-recommended Approach
2 uncertainty estimation methodology, the Monte Carlo stochastic simulation technique. The Monte Carlo
stochastic simulation was performed on the total emissions estimate from other product use, represented in
equation form as:
Equation 4-27: Total Emissions from Other Product Use
Total Emissions (Er)
= Military Applications SF6 and PFC Emissions (EMilitary)
+ Scientific Applications of SF6 and PFC Emissions (Escientific)
The uncertainty in the total emissions for other product use, presented in Table 4-124 below, results from the
convolution of two distributions of emissions, namely from military applications and scientific applications. The
approaches for estimating uncertainty in each of the sources are described below:
Military Applications Emission Uncertainty
The Monte Carlo stochastic simulation was performed on the emissions estimate from military applications,
represented in equation form as:
Equation 4-28: Total Emissions from Military Applications
Military Applications SF6 and PFC Emissions (EMilitary)
= Military AWACS SF6 Emissions (EAWacs,sf6,Military)
+ Other Military Applications SF6 Emissions (Eother,sf6,Military)
+ Other Military Applications PFC Emissions (Eother,PFC,Military)
The uncertainty in EMmtary results from the convolution of three distributions of emissions, Eawacs,sf6,Military,
Eother,sF6,Military, and Eother,.pfc,Military. The approaches for estimating each distribution and combining them to arrive at
the reported 95 percent confidence interval (CI) for EMmtary are described in the remainder of this section.
The uncertainty estimate of Eawacs,sf6,Military, or SF6 emissions from AWACS, is developed based on the number of
AWACS in commission in the United States and the per-plane emission factor. The estimated number of active
planes installed with AWACS is 33, although estimates range between 31 and 35. The IPCC provides a per-plane
emission factor of 740 kg of SF6 per plane annually and estimates the uncertainty to have bounds of ±14 percent.
The uncertainty in Eother,sfb,Military and Eother,pfc,Military, or SF6 and PFC emissions from other military applications, was
obtained by determining the accuracy of government-reported emissions data and reviewing the methodology the
Department of Defense uses for developing inventory estimates.
The next step in estimating the uncertainty in emissions from military AWACS and other military applications is
convolving the distribution of reported emissions, emission factors, and number of AWACS using Monte Carlo
simulation. For this Monte Carlo simulation, the distributions of the reported emissions and emission factors are
assumed to be normally distributed, and the number of AWACS is assumed to have a uniform distribution since
this is a discrete number of planes. The uncertainty bounds are assigned at 1.96 standard deviations around the
estimated mean.
Scientific Applications Emission Uncertainty
The Monte Carlo stochastic simulation was performed on the emissions estimate from scientific applications,
Industrial Processes and Product Use 4-181
-------
represented in equation form as:
Equation 4-29: Total Emissions from Scientific Applications
Scientific Applications SF6 and PFC Emissions (Escientific)
= Particle Accelerators SF6 Emissions (EAccelerators^,scientific)
+ Particle Accelerators PFC Emissions {^Accelerators,pFc,Military)
+ Other Scientific Applications SF6 Emissions (Eother^,scientific)
+ Other Scientific Applications PFC Emissions (Eother,pfc,scientific)
The uncertainty in Escientific results from the convolution of four distributions of emissions, E Accelerators, SF6, Scientific,
^.Accelerators,pfc,Military, Mother,sf6,scientific, and Mother,pfc,scientific. The approaches for estimating each distribution and
combining them to arrive at the reported 95 percent confidence interval (CI) for Escientific are described in the
remainder of this section.
The uncertainty estimate of ^.Accelerators,sfb,scientific and EAccelerators,PFc,scientific, or SF6 and PFC emissions from particle
accelerators, is developed based on fugitive and process emissions reported by the Department of Energy and
emission estimates from the number active university and industrial particle accelerators in the United States. The
number of active particle accelerators in the United States for the time series 1990 through 2022 was determined
using expert judgment; default emission factors and charge sizes for particle accelerators of various sizes were
provided by IPCC guidelines. Emissions of SF6 from electrical transmission and distribution equipment were
removed from total emissions estimates for this source category, as they are reported elsewhere in the Inventory.
The uncertainty in Mother,sfb,scientific and Eother,pfc,scientific, or SF6 and PFC emissions from other scientific applications,
was obtained by determining the accuracy of government-reported emissions data and reviewing the
methodology the Department of Energy uses for developing inventory estimates.
The next step in estimating the uncertainty in emissions from particle accelerators and other scientific applications
is convolving the distribution of calculated emissions, emission factors, number of accelerators using Monte Carlo
simulation. Similarly, the distributions of the reported emissions and emission factors for this Monte Carlo
simulation are assumed to be normally distributed, and the number of particle accelerators and other scientific
applications is assumed to have a uniform distribution since this is a discrete number of accelerators. The
uncertainty bounds are assigned at 1.96 standard deviations around the estimated mean.
The emissions estimate for total U.S. SF6 and PFC emissions from other product use were estimated to be between
0.5 and 1.1 MMT CO2 Eq. at a 95 percent CI level. This range represents 36 percent below and 38 percent above
the 2022 emission estimate of 0.8 MMT CO2 Eq. for all emissions from others product use. This range and the
associated percentages apply to the estimate of total emissions rather than those of individual gases. Uncertainties
associated with individual gases will be somewhat higher than the aggregate but were not explicitly modeled.
Table 4-136: Approach 2 Quantitative Uncertainty Estimates for SF6 and PFC Emissions from
Other Product Use (MMT CO2 Eq. and Percent)
Source
Gas
2022 Emission Estimate
Uncertainty Range Relative to Emission Estimate-1
(MMTCO. Eq.)
(MMT CO . Eq.)
(%)
Lower Upper
Lower Upper
Bound1, Bound1,
Bound Bound
Other Product Use
SF6 and PFC
0.8
0.5 1.1
-36% +38%
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.
4-182 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
QA/QC and Verification
For more information on the general QA/QC process applied to this source category, consistent with Volume 1,
Chapter 6 of the 2006IPCC Guidelines, see the QA/QC and Verification Procedures section in the introduction of
the IPPU chapter and Annex 8 for more details.
Recalculations Discussion
This is a new category included for the current (i.e., 1990 to 2022) Inventory, thus, no recalculations were
performed.
4.28 Nitrous Oxide from Product Uses (CRT
Source Category 2G3)
Nitrous oxide (N2O) is a clear, colorless, oxidizing liquefied gas with a slightly sweet odor which is used in a wide
variety of specialized product uses and applications. This reporting category (2G3) includes exhalation emissions of
N2O that arise from medical applications and evaporative emissions of N2O from use as a propellant in aerosol
products primarily in food industry. The amount of N2O that is actually emitted depends upon the specific product
use or application.
There are a total of three N2O production facilities currently operating in the United States (Ottinger 2021). Nitrous
oxide is primarily used in carrier gases with oxygen to administer more potent inhalation anesthetics for general
anesthesia, and as an anesthetic in various dental and veterinary applications. The second main use of N2O is as a
propellant in pressure and aerosol products, the largest application being pressure-packaged whipped cream.
Small quantities of N2O also are used in the following applications:
• Oxidizing agent and etchant used in semiconductor manufacturing;
• Oxidizing agent used, with acetylene, in atomic absorption spectrometry;
• Production of sodium azide, which is used to inflate airbags;
• Fuel oxidant in auto racing; and
• Oxidizing agent in blowtorches used by jewelers and others (Heydorn 1997).
Production of N2O in 2022 was approximately 15 kt (see Table 4-137).
Table 4-137: N2O Production (kt)
Year
1990
2005
2018
2019
2020
2021
2022
Production (kt)
16
15
15
15
15
15
15
Nitrous oxide emissions were 3.8 MMT CO2 Eq. (14 kt N2O) in 2022 (see Table 4-138). Production of N2O stabilized
during the 1990s because medical markets had found other substitutes for anesthetics, and more medical
procedures were being performed on an outpatient basis using local anesthetics that do not require N2O. The use
of N2O as a propellant for whipped cream has also stabilized due to the increased popularity of cream products
packaged in reusable plastic tubs (Heydorn 1997).
Table 4-138: N2O Emissions from N2O Product Usage (MMT CO2 Eq.)
Year
1990
2005
2018
2019
2020
2021
2022
N20 Product Usage
00
CO
00
ro
3.8
3.8
3.8
3.8
3.8
Industrial Processes and Product Use 4-183
-------
Table 4-139: N2O Emissions from N2O Product Usage (kt N20)
Year
1990
2005
2018
2019
2020
2021
2022
N20 Product Usage
14
14
14
14
14
14
14
Methodology and Time-Series Consistency
Emissions from N2O product uses are calculated using a country-specific methodology that is consistent with 2006
IPCC Guidelines and based on available data. The 2006IPCC Guidelines do not define tier methodologies for this
source category. Emissions of N2O are estimated using the national N2O production by subcategory use or
application, the share of the subcategory, and the appropriate emission rate for each category. The following
equation is adapted from Equation 8.24 of the 2006 IPCC Guidelines:
Equation 4-30: N2O Emissions from Product Use
Epu = X Sa x ERa)
a
where,
Epu = N2O emissions from product uses, metric tons
P = Total U.S. production of N2O, metric tons
a = specific application
Sa = Share of N2O usage by application a
ERa = Emission rate for application a, percent
The share of total quantity of N2O usage by end-use represents the share of national N2O produced that is used by
the specific subcategory (e.g., anesthesia, food processing). In 2020, the medical/dental industry used an
estimated 89.5 percent of total N2O produced, followed by food processing propellants at 6.5 percent. All other
subcategories, including semiconductor manufacturing, atomic absorption spectrometry, sodium azide production,
auto racing, and blowtorches, used the remainder of the N2O produced. This subcategory breakdown changed
slightly in the mid-1990s. For instance, the small share of N2O usage in the production of sodium azide declined
significantly during the 1990s. Due to the lack of information on the specific time period of the phase-out in this
market subcategory, most of the N2O usage for sodium azide production is assumed to have ceased after 1996,
with the majority of its small share of the market assigned to the larger medical/dental consumption subcategory
(Heydorn 1997). For 1990 through 1996, N2O usage was allocated across the following subcategories: medical
applications, food processing propellant, and sodium azide production. A usage emissions rate was then applied
for each subcategory to estimate the amount of N2O emitted.
Only the medical/dental and food propellant subcategories were assumed to release emissions into the
atmosphere that are not captured under another source category, and therefore these subcategories were the
only usage subcategories with emission rates. Emissions of N2O from semiconductor manufacturing are described
in Section 4.24 and reported under CRT Source Category 2H3. For the medical/dental subcategory, due to the poor
solubility of N2O in blood and other tissues, none of the N2O is assumed to be metabolized during anesthesia and
quickly leaves the body in exhaled breath. Therefore, an emission factor of 100 percent was used for this
subcategory (IPCC 2006). For N2O used as a propellant in pressurized and aerosol food products, none of the N2O is
reacted during the process and all of the N2O is emitted to the atmosphere, resulting in an emission factor of 100
percent for this subcategory (IPCC 2006). For the remaining subcategories, all of the N2O is consumed or reacted
during the process, and therefore the emission rate was considered to be zero percent (Tupman 2002).
The 1990 through 1992 N2O production data were obtained from SRI Consulting's Nitrous Oxide, North America
(Heydorn 1997). Nitrous oxide production data for 1993 through 1995 were not available. Production data for
4-184 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
1996 was specified as a range in two data sources (Heydorn 1997; Tupman 2002). In particular, for 1996, Heydorn
(1997) estimates N2O production to range between 13.6 and 18.1 thousand metric tons. Tupman (2002) provided a
narrower range (15.9 to 18.1 thousand metric tons) for 1996 that falls within the production bounds described by
Heydorn (1997). Tupman (2002) data are considered more industry-specific and current; therefore, the midpoint
of the narrower production range was used to estimate N2O emissions for years 1993 through 2001 (Tupman
2002). The 2002 and 2003 N2O production data were obtained from the Compressed Gas Association Nitrous
Oxide Fact Sheet and Nitrous Oxide Abuse Hotline (CGA 2002, 2003). These data were also provided as a range. For
example, in 2003, CGA (2003) estimates N2O production to range between 13.6 and 15.9 thousand metric tons.
Due to the lack of publicly available data, production estimates for years 2004 through 2022 were held constant at
the 2003 value.
The 1996 share of the total quantity of N2O used by each subcategory was obtained from SRI Consulting's Nitrous
Oxide, North America (Heydorn 1997). The 1990 through 1995 share of total quantity of N2O used by each
subcategory was kept the same as the 1996 number provided by SRI Consulting. The 1997 through 2001 share of
total quantity of N2O usage by sector was obtained from communication with a N2O industry expert (Tupman
2002). The 2002 and 2003 share of total quantity of N2O usage by sector was obtained from CGA (2002, 2003). Due
to the lack of publicly available data, the share of total quantity of N2O usage data for years 2004 through 2021
was assumed to equal the 2003 value. The emission factor for the food processing propellant industry was
obtained from SRI Consulting's Nitrous Oxide, North America (Heydorn 1997) and confirmed by a N2O industry
expert (Tupman 2002). The emission factor for all other subcategories was obtained from communication with a
N2O industry expert (Tupman 2002). The emission factor for the medical/dental subcategory was obtained from
the 2006IPCC Guidelines.
Methodological approaches were applied to the entire time series to ensure consistency in emissions from 1990
through 2022.
Uncertainty
The overall uncertainty associated with the 2022 N2O emission estimate from N2O product usage was calculated
using the 2006 IPCC Guidelines Approach 2 methodology. Uncertainty associated with the parameters used to
estimate N2O emissions include production data, total market share of each end use, and the emission factors
applied to each end use, respectively. The uncertainty associated with N2O production data is ±25 percent, and a
uniform probability density function is assigned, based on expert judgment (RTI 2023). The uncertainty associated
with the market share for the medical/dental subcategory is ±0.56 percent, and uncertainty for the market share
of food propellant subcategory is ±25 percent, both based on expert judgment (RTI 2023). Uncertainty for emission
factors was assumed to be zero, and using this suggested uncertainty provided in the 2006 IPCC Guidelines is
appropriate based on expert judgment (RTI 2023).
The results of this Approach 2 quantitative uncertainty analysis are summarized in Table 4-140. Nitrous oxide
emissions from N2O product usage were estimated to be between 2.9 and 4.6 MMT CO2 Eq. at the 95 percent
confidence level. This indicates a range of approximately 24 percent below to 24 percent above the emission
estimate of 3.8 MMT CO2 Eq.
Table 4-140: Approach 2 Quantitative Uncertainty Estimates for N2O Emissions from N2O
Product Usage (MMT CO2 Eq. and Percent)
Source
Gas
2022 Emission Estimate
Uncertainty Range Relative to Emission Estimate-'
(MMT CO . Eq.)
(MMT CO . Eq.)
(%)
Lower Upper
Lower Upper
Bound Bound
Bound Bound
N20 from Product Uses
N20
3.8
2.9 4.6
-24% +24%
a Range of emission estimates predicted by Monte Carlo stochastic simulation for a 95 percent confidence interval.
Industrial Processes and Product Use 4-185
-------
QA/QC and Verification
General quality assurance/quality control (QA/QC) procedures were applied consistent with the U.S. Inventory
QA/QC plan, which is in accordance with Volume 1, Chapter 6 of 2006IPCC Guidelines as described in the
introduction of the IPPU chapter (see Annex 8 for more details).
Recalculations Discussion
No recalculations were performed for the 1990 to 2021 portion of the time series.
Planned Improvements
EPA recently initiated an evaluation of alternative production statistics for cross-verification and updating time-
series activity data, emission factors, assumptions, etc., and a reassessment of N2O product use subcategories that
accurately represent trends. This evaluation includes conducting a literature review of publications and research
that may provide additional details on the industry. This work remains ongoing, and thus far no additional sources
of data have been found to update this category.
Pending additional resources and planned improvement prioritization, EPA may also evaluate production and use
cycles, and the potential need to incorporate a time lag between production and ultimate product use and
resulting release of N2O. Additionally, planned improvements include considering imports and exports of N2O for
product uses.
Finally, for future Inventories, EPA will examine data from EPA's GHGRP to improve the emission estimates for the
N2O product use subcategory. Particular attention will be made to ensure aggregated information can be published
without disclosing CBI and time-series consistency, as the facility-level reporting data from EPA's GHGRP are not
available for all inventory years as required in this Inventory. This is a lower priority improvement, and EPA is still
assessing the possibility of incorporating aggregated GHGRP CBI data to estimate emissions; therefore, this
planned improvement is still in development and not incorporated in the current Inventory report.
4.29 Industrial Processes and Product Use
Sources of Precursor Gases
In addition to the main greenhouse gases addressed above, many industrial processes can result in emissions of
various greenhouse gas precursors. The reporting requirements of the Paris Agreement and the UNFCCC133
request that information should be provided on precursor emissions, which include carbon monoxide (CO),
nitrogen oxides (NOx), non-methane volatile organic compounds (NMVOCs), and sulfur dioxide (SO2). These gases
are not direct greenhouse gases, but indirectly impact Earth's radiative balance by altering the concentrations of
greenhouse gases (e.g., ozone) and atmospheric aerosol (e.g., particulate sulfate). Combustion byproducts such as
CO and NOx are emitted from industrial applications that employ thermal incineration as a control technology.
NMVOCs, commonly referred to as "hydrocarbons," are the primary gases emitted from most processes employing
organic or petroleum-based products, and can also result from the product storage and handling.
133 See paragraph 51 of Annex to 18/CMA.l available online at:
https://unfccc.int/sites/default/files/resource/CMA2018 03a02E.pdf.
4-186 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
Accidental releases of precursors associated with product use and handling can constitute major emissions in this
category. In the United States, emissions from product use are primarily the result of solvent evaporation,
whereby the lighter hydrocarbon molecules in the solvents escape into the atmosphere. The major categories of
product uses include: degreasing, graphic arts, surface coating, other industrial uses of solvents (e.g., electronics),
dry cleaning, and non-industrial uses (e.g., uses of paint thinner). Product usage in the United States also results in
the emission of hydrofluorocarbons (HFCs) and small amounts of hydrofluoroethers (HFEs), which are included
under Substitution of Ozone Depleting Substances and the Electronics Industry in this chapter.
Total emissions of NOx, CO, NMVOCs, and SO2 from non-energy industrial processes and product use from 1990 to
2022 are reported in Table 4-141.
Table 4-141: NOx, CO, NMVOC, and SO2 Emissions from Industrial Processes and Product Use
(kt)
Gas/Source
1990
2005
2018
2019
2020
2021
2022
NOx
774
672
461
440
393
403
389
Mineral Industry
160 1
200 1
118
114
101
99
95
Other Industrial Processes3
326
355
218
206
187
189
184
Metal Industry
96 =
58 1
63
60
52
60
56
Chemical Industry
192
80
61
59
54
55
53
CO
4,099
1,701
1,022
1,011
855
902
897
Metal Industry
2,261
707
447
448
340
355
335
Other Industrial Processes3
564
662 I
332
331
294
309
329
Mineral Industry
182
120
111
106
96
95
95
Chemical Industry
1,093 1
211
132
126
125
142
138
NMVOCs
6,982
3,668
3,119
2,996
3,366
3,508
3,505
Other Industrial Processes3
6,270
3,396 I!
3,003
2,883
3,261
3,398
3,401
Chemical Industry
601
221
88
86
81
84
79
Mineral Industry
9
101;
7
7
6
6
6
Metal Industry
102
40
21
20
17
19
19
S02
1,488
776
335
309
266
274
261
Other Industrial Processes3
474
256
145
134
120
126
119
Chemical Industry
283 -
242 1
106
97
83
83
75
Mineral Industry
166
138
25
25
26
28
28
Metal Industry
566 1
140 1
58
53
37
38
39
a Other Industrial Processes includes storage and transport, other industrial processes (manufacturing of agriculture, food,
and kindred products; wood, pulp, paper, and publishing products; rubber and miscellaneous plastic products; machinery
products; construction; transportation equipment; and textiles, leather, and apparel products), and miscellaneous sources
(catastrophic/accidental release, other combustion (structural fires), health services, repair shops, 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.
Note: Totals by gas may not sum due to independent rounding.
Source: (EPA 2023a). Emission categories from EPA (2023a) are aggregated into sectors and categories reported under the
Paris Agreement and the UNFCCC as shown in Table ES-3.
Methodology and Time-Series Consistency
Emission estimates for 1990 through 2020 were obtained from data published on the National Emissions Inventory
(NEI) Air Pollutant Emissions Trends Data website (EPA 2023a). For Table 4-141, NEI reported emissions of CO, NOx,
SO2, and NMVOCs were recategorized from NEI Emissions Inventory System (EIS) sectors to source categories more
closely aligned with reporting sectors and categories under the Paris Agreement and the UNFCCC based on
Industrial Processes and Product Use 4-187
-------
discussions between the EPA GHG Inventory and NEI staff (see crosswalk documented in Annex 6.3).134 EIS sectors
mapped to the IPPU sector categories in this report include: chemical and allied product manufacturing, metals
processing, storage and transport, solvent utilization, other industrial processes, and miscellaneous sources. As
described in the NEI Technical Support Documentation (TSD) (EPA 2023c), NEI emissions are estimated through a
combination of emissions data submitted directly to the EPA by state, local, and tribal air agencies, as well as
additional information added by the Agency from EPA emissions programs, such as the emission trading program,
Toxics Release Inventory (TRI), and data collected during rule development or compliance testing.
Methodological approaches were applied to the entire time series to ensure time-series consistency from 1990
through 2021, which are described in detail in the NEI'sTSD and on EPA's Air Pollutant Emission Trends web site
(EPA 2023a; EPA 2023c). A quantitative uncertainty analysis was not performed.
134 The NEI estimates and reports emissions from six criteria air pollutants (CAPs) and 187 hazardous air pollutants (HAPs) in
support of National Ambient Air Quality Standards. EPA reported CAP emission trends are grouped into 60 sectors and 15 Tier 1
source categories, which broadly cover similar source categories to those presented in this chapter. For reporting precursor
emissions in the common reporting tables (CRTs), EPA has mapped and regrouped emissions of greenhouse gas precursors (CO,
NOx, S02, and NMVOCs) from NEI's EIS sectors to better align with NIR source categories, and to ensure consistency and
completeness to the extent possible. See Annex 6.3 for more information on this mapping.
4-188 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
5. Agriculture
Agricultural activities contribute directly to emissions of greenhouse gases through a variety of processes. This
chapter provides an assessment of methane (Cm) from enteric fermentation, livestock manure management, rice
cultivation and field burning of agricultural residues; nitrous oxide (N2O) emissions from agricultural soil
management, livestock manure management, and field burning of agricultural residues; as well as carbon dioxide
(CO2) emissions from liming and urea fertilization (see Figure 5-1). Additional CO2, CH4 and N2O fluxes from
agriculture-related land-use and land-use conversion activities, such as cultivation of cropland, management on
grasslands, grassland fires, aquaculture, and conversion of forest land to cropland, are presented in the Land Use,
Land-Use Change, and Forestry (LULUCF) chapter. Carbon dioxide emissions from stationary and mobile on-farm
energy use and CH4 and N2O emissions from stationary on-farm energy use are reported in the Energy chapter
under the Industrial sector emissions. Methane and N2O emissions from mobile on-farm energy use are reported
in the Energy chapter under mobile fossil fuel combustion emissions.
Figure 5-1: 2022 Agriculture Sector Greenhouse Gas Emission Sources
Agricultural Soil Management
Enteric Fermentation
Manure Management
Rice Cultivation
Urea Fertilization
Liming
Field Burning of Agricultural Residues
Agriculture as a Portion of
All Emissions
9.4%
291
Energy
Agriculture
IPPU
Waste
100 120 140 160 180 200 220
MMT CO2 Eq.
In 2022, the Agriculture sector was responsible for emissions of 593.4 MMT CO2 Eq.,1 or 9.4 percent of total U.S.
greenhouse gas emissions. Emissions of N2O by agricultural soil management through activities such as fertilizer
1 Following the current reporting requirements under the Paris Agreement and the United Nations Framework Convention on
Climate Change (UNFCCC), this Inventory report presents C02 equivalent values based on the IPCC Fifth Assessment Report
(AR5) GWP values. See the Introduction chapter as well as Chapter 9 for more information.
Agriculture 5-1
-------
application and other agricultural practices that increased nitrogen availability in the soil was the largest source of
U.S. N2O emissions, accounting for 74.6 percent, and the largest source of emissions from the Agriculture sector,
accounting for 49.0 percent of total sector emissions. Methane emissions from enteric fermentation and manure
management represented 27.4 percent and 9.2 percent of total CH4 emissions from anthropogenic activities,
respectively, and 32.5 and 10.9 percent of Agriculture sector emissions, respectively. Of all domestic animal types,
beef and dairy cattle were the largest emitters of CH4. Rice cultivation and field burning of agricultural residues
were minor sources of CH4. Manure management and field burning of agricultural residues were also small sources
of N2O emissions. Urea fertilization and liming each accounted for 0.1 percent of total CO2 emissions from
anthropogenic activities.
Table 5-1 and Table 5-2 present emission estimates for the Agriculture sector. Between 1990 and 2022, CChand
Cm emissions from agricultural activities increased by 21 percent and 14.5 percent, respectively, while N2O
emissions from agricultural activities fluctuated from year to year but increased by 1.9 percent overall. Trends in
sources of agricultural emissions over the 1990 to 2022 time series are shown in Figure 5-2.
Figure 5-2: Trends in Agriculture Sector Greenhouse Gas Emission Sources
Each year, some emission estimates in the Agriculture sector of the Inventory are recalculated and revised with
improved methods and/or data. In general, recalculations are made to the U.S. greenhouse gas emission estimates
either to incorporate new methodologies or, most commonly, to update recent historical data. These
improvements are implemented consistently across the previous Inventory's time series (i.e., 1990 through 2021)
to ensure that the trend is accurate. This year's key improvements include: manure management: updates to beef
feedlot and poultry waste management system (WMS) data; field burning of agricultural residues: addition of
residue burning from sugarcane. For more information on specific methodological updates, please see the
Recalculations Discussions within the respective source category sections of this chapter. In total, the
methodological and historic data improvements made to the Agriculture sector in this Inventory increased
greenhouse gas emission estimates by an average of 5.3 MMT CO2 Eq. (0.9 percent) across the time series.
Emissions reported in the Agriculture chapter include those from all states; however, for Hawaii and Alaska some
agricultural practices that can increase nitrogen availability in the soil, and thus cause N2O emissions, are not
5-2 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
included (see chapter sections on Uncertainty and Time-Series Consistency and Planned Improvements for more
details). Emissions from the Agriculture sector occurring in U.S. Territories and the District of Columbia are not
estimated due to incomplete data, with the exception of urea fertilization in Puerto Rico. EPA continues to identify
and review available data on an ongoing basis to include agriculture emissions from U.S. Territories, to the extent
they are occurring, in future Inventories. Other minor outlying U.S. Territories in the Pacific Islands have no
permanent populations (e.g., Baker Island) and therefore EPA assumes no agricultural activities are occurring. See
Annex 5 for more information on EPA's assessment of the sources not included in this Inventory.
Table 5-1: Emissions from Agriculture (MMT CO2 Eq.)
Gas/Source
1990
2005
2018
2019
2020
2021
2022
C02
7.1
7.9
7.2
7.2
8.0
7.6
8.6
Urea Fertilization
2.4
3.5
4.9
5.0
5.1
5.2
5.3
Liming
4.7
4.4
2.2
2.2
2.9
2.4
3.3
ch4
241.7
264.4
285.0
280.2
282.4
281.8
276.8
Enteric Fermentation
183.1
188.2
196.8
197.3
196.3
196.5
192.6
Manure Management
39.1
55.0
67.7
66.7
66.9
66.4
64.7
Rice Cultivation
18.9
20.6
19.9
15.6
18.6
18.3
18.9
Field Burning of Agricultural Residues
0.5
0.6
0.6
0.7
0.6
0.6
0.6
n2o
302.3
309.5
350.2
332.6
309.2
315.3
308.0
Agricultural Soil Management
288.8
294.1
333.4
315.6
292.1
298.0
290.8
Manure Management
13.4
15.2
16.6
16.8
16.9
17.1
17.0
Field Burning of Agricultural Residues
0.2
0.2
0.2
0.2
0.2
0.2
0.2
Total
551.1
581.8
642.4
620.1
599.7
604.8
593.4
Note: Totals may not sum due to independent rounding.
Table 5-2: Emissions from Agriculture (kt)
Gas/Source
1990
2005
2018
2019
2020
2021
2022
C02
7,106
7,856
7,176
7,237
8,019
7,616
8,595
Urea Fertilization
2,417
3,504
4,936
5,034
5,132
5,229
5,327
Liming
4,690
4,351
2,240
2,203
2,887
2,387
3,268
ch4
8,633
9,444
10,179
10,008
10,087
10,066
9,885
Enteric Fermentation
6,539
6,722
7,028
7,045
7,010
7,017
6,878
Manure Management
1,398
1,964
2,418
2,382
2,390
2,373
2,312
Rice Cultivation
677
735
711
558
664
653
674
Field Burning of Agricultural Residues
19
23
22
23
22
22
22
n2o
1,141
1,168
1,322
1,255
1,167
1,190
1,162
Agricultural Soil Management
1,090
1,110
1,258
1,191
1,102
1,124
1,097
Manure Management
50
57
63
63
64
65
64
Field Burning of Agricultural Residues
1
1
1
1
1
1
1
Note: Totals by gas may not sum due to independent rounding.
Box 5-1: Methodological Approach for Estimating and Reporting U.S. Emissions and
Removals
Consistent with Article 13.7(a) of the Paris Agreement and Article 4.1(a) of the UNFCCC as well as relevant
decisions under those agreements, the emissions and removals 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) in the 2006IPCC Guidelines for National Greenhouse Gas
Inventories (2006 IPCC Guidelines). Additionally, the calculated emissions and removals in a given year for the
United States are presented in a common format in line with the reporting guidelines for the reporting of
inventories under the Paris Agreement and the UNFCCC. The Parties' use of consistent methods to calculate
emissions and removals for their inventories helps to ensure that these reports are comparable. The
Agriculture 5-3
-------
presentation of emissions provided in the Agriculture chapter does not preclude alternative examinations (e.g.,
economic sectors). Rather, this chapter presents emissions in a common format consistent with how Parties are
to report their national inventories under the Paris Agreement and the UNFCCC. The report itself, and this
chapter, follow this common format and provide an explanation of the application of methods used to calculate
emissions from agricultural activities.
5.1 Enteric Fermentation (CRT 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 Cm as a byproduct, which can be exhaled or eructated by the animal. The
amount of Cm produced and emitted by an individual animal depends primarily upon the animal's digestive
system, and the amount and type of feed it consumes.2
Ruminant animals (e.g., cattle, buffalo, sheep, goats, and camels) are the major emitters of Cm 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 Cm emissions per unit of body mass among all animal
types.
Non-ruminant animals (e.g., swine, horses, and mules and asses) also produce Cm emissions through enteric
fermentation, although this microbial fermentation occurs in the large intestine. These non-ruminants emit
significantly less Cm on a per-animal-mass basis than ruminants because the capacity of the large intestine to
produce Cm is lower.
In addition to the type of digestive system, an animal's feed quality and feed intake also affect Cm emissions. In
general, lower feed quality and/or higher feed intake leads to higher Cm emissions. Feed intake is positively
correlated to animal size, growth rate, level of activity and production (e.g., milk production, wool growth,
pregnancy, or work). Therefore, feed intake varies among animal types as well as among different management
practices for individual animal types (e.g., animals in feedlots or grazing on pasture).
Methane emission estimates from enteric fermentation are provided in Table 5-3 and Table 5-4. Total livestock Cm
emissions in 2022 were 192.6 MMT CO2 Eq. (6,878 kt). Beef cattle remain the largest contributor of CH4 emissions
from enteric fermentation, accounting for 71 percent in 2022. Emissions from dairy cattle in 2022 accounted for 25
percent, and the remaining methane emissions were from swine, horses, sheep, goats, American bison, mules and
asses.3
2 C02 emissions from livestock are not estimated because annual net C02 emissions are assumed to be zero - the C02
photosynthesized by plants is returned to the atmosphere as respired C02 (IPCC 2006).
3 Enteric fermentation emissions from poultry are not estimated because no IPCC method has been developed for determining
enteric fermentation CH4 emissions from poultry; at this time, developing a country-specific method would require a
disproportionate amount of resources given the small magnitude of this source category. Enteric fermentation emissions from
camels are not estimated because there is no significant population of camels in the United States. Given the insignificance of
estimated camel emissions in terms of the overall level and trend in national emissions, there are no immediate improvement
5-4 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
Table 5-3: CH4 Emissions from Enteric Fermentation (MMT CO2 Eq.)
Livestock Type
1990
2005
2018
2019
2020
2021
2022
Beef Cattle
132.8
139.6
141.2
141.7
140.5
140.3
137.0
Dairy Cattle
43.3 <
41.3
1
i
48.6
48.5
48.8
49.4
48.9
Swine
2.3
2.6
3.1
3.2
3.2
3.1
3.1
Horses
1-1
2.0
1
1.4
1.3
1.2
1.1
1.0
Sheep
2.9
1.5
1.3
1.3
1.3
1.3
1.3
Goats
o.6:
0.7
*
0.7
0.7
0.7
0.7
0.7
American Bison
0.1
0.5
0.4
0.4
0.5
0.5
0.5
Mules and Asses
0.1
I
¦
0.1
0.1
0.1
0.1
0.1
Total
183.1
188.2
I
196.8
197.3
196.3
196.5
192.6
+ Does not exceed 0.05 MMT C02 Eq.
Note: Totals may not sum due to independent rounding.
Table 5-4: CH4 Emissions from Enteric Fermentation (kt CH4)
Livestock Type
1990
2005
2018
2019
2020
2021
2022
Beef Cattle
4,742
4,986
5,042
5,062
5,018
5,010
4,891
Dairy Cattle
1,547
1
1/473 :
1,737
1,732
1,743
1,764
1,748
Swine
81
92
110
115
115
111
110
Horses
40
1
70 I
48
46
43
40
37
Sheep
102
55
47
47
47
47
46
Goats
23
I
26 ::
24
25
25
25
25
American Bison
4
17
15
16
16
17
17
Mules and Asses
1
I
2
3
3
3
3
3
Total
6,539
6,722
7,028
7,045
7,010
7,017
6,878
Note: Totals may not sum due to independent rounding.
From 1990 to 2022, emissions from enteric fermentation have increased by 5.2 percent. From 2021 to 2022,
emissions decreased by 2 percent, largely driven by a decrease in beef cattle populations. While emissions
generally follow trends in cattle populations, there are exceptions across the time series. For example, while dairy
cattle emissions increased 13 percent over the entire time series, the population has declined by 4.5 percent, and
milk production increased 45.9 percent (USDA 2021; USDA 2022). These trends indicate that while emissions per
head are increasing, emissions per unit of product (i.e., meat, milk) are decreasing.
Generally, from 1990 to 1995 emissions from beef cattle increased and then decreased from 1996 to 2004. These
trends were mainly due to fluctuations in beef cattle populations and increased digestibility of feed for feedlot
cattle. Beef cattle emissions generally increased from 2004 to 2007, as beef cattle populations increased, and an
extensive literature review indicated a trend toward a decrease in feed digestibility for those years. Beef cattle
emissions decreased again from 2007 to 2014, as populations again decreased, but increased from 2015 to 2018,
consistent with another increase in population over those same years. Emissions and populations generally
declined from 2018 to 2022, with a slight post-pandemic rebound in 2021.
Emissions from dairy cattle generally trended downward from 1990 to 2004, along with an overall dairy cattle
population decline during the same period. Similar to beef cattle, dairy cattle emissions rose from 2004 to 2007
due to population increases and a decrease in feed digestibility (based on an analysis of more than 350 dairy cow
diets used by producers across the United States). Dairy cattle emissions continued to trend upward from 2007 to
2018, generally in line with dairy cattle population changes.
plans to include this emissions category in the Inventory. See Annex 5 for more information on significance of estimated camel
emissions.
Agriculture 5-5
-------
Regarding trends in other animals, populations of sheep have steadily declined, with an overall decrease of 55
percent since 1990. Horse populations peaked in 2007 and have been declining by an average of 4 percent
annually since 2007, with their current population 6 percent lower than it was in 1990. Goat populations increased
by about 20 percent through 2007 followed by a steady decrease through 2012. Since 2012, goat populations
continue to increase by 1 percent annually. Swine populations have trended upward through most of the time
series, increasing 43 percent from 1990 to 2020. However, swine populations decreased by around 5 percent from
2020 to 2022. The population of American bison more than quadrupled over the 1990 to 2022 time period, while
the population of mules and asses increased by a factor of five.
Methodology and Time-Series Consistency
Livestock enteric fermentation emission estimate methodologies fall into two categories: cattle and other
domesticated animals. Cattle, due to their large population, large size, and particular digestive characteristics,
account for the majority of enteric fermentation CH4 emissions from livestock in the United States. A more detailed
methodology (i.e., IPCC Tier 2) was therefore applied to estimate emissions for all cattle. Emission estimates for
other domesticated animals (horses, sheep, swine, goats, American bison, and mules and asses) were estimated
using the IPCC Tier 1 approach, as suggested by the 2006 IPCC Guidelines (see the Planned Improvements section).
While the large diversity of animal management practices cannot be precisely characterized and evaluated,
significant scientific literature exists that provides the necessary data to estimate cattle emissions using the IPCC
Tier 2 approach. The Cattle Enteric Fermentation Model (CEFM), developed by EPA and used to estimate cattle CH4
emissions from enteric fermentation using IPCC's Tier 2 method, incorporates this information and other analyses
of livestock population, feeding practices, and production characteristics.
Methodological approaches, changes to historic data, and other parameters were applied to the entire time series
to ensure consistency in emissions estimates from 1990 through 2022. See Annex 3.10 for more detailed
information on the methodology and data used to calculate CH4 emissions from enteric fermentation. In addition,
variables and the resulting emissions are also available at the state level in Annex 3.10.
Inventory Methodology for Cattle
National cattle population statistics were disaggregated into the following cattle sub-populations:
• Dairy Cattle
o Calves
o Heifer Replacements
o Cows
• Beef Cattle
o Calves
o Heifer Replacements
o Heifer and Steer Stockers
o Animals in Feedlots (Heifers and Steer)
o Cows
o Bulls
Calf birth rates, end-of-year population statistics, detailed feedlot placement information, and slaughter weight
data were used to create a transition matrix that models cohorts of individual animal types and their specific
emission profiles. The key variables tracked for each of the cattle population categories are described in Annex
3.10. These variables include performance factors such as pregnancy and lactation as well as average weights and
weight gain. Annual cattle population data were obtained from the U.S. Department of Agriculture's (USDA)
National Agricultural Statistics Service (NASS) QuickStats database (USDA 2023).
5-6 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
Diet characteristics were estimated by region for dairy, grazing beef, and feedlot beef cattle. These diet
characteristics were used to calculate digestible energy (DE) values (expressed as the percent of gross energy
intake digested by the animal) and Cm conversion rates (Ym) (expressed as the fraction of gross energy converted
to Cm) for each regional population category. The IPCC recommends Ym ranges of 3.0±1.0 percent for feedlot
cattle and 6.5±1.0 percent for other well-fed cattle consuming temperate-climate feed types (IPCC 2006). Given
the availability of detailed diet information for different regions and animal types in the United States, DE and Ym
values unique to the United States were developed. The diet characterizations and estimation of DE and Ym values
were based on information from state agricultural extension specialists, a review of published forage quality
studies and scientific literature, expert opinion, and modeling of animal physiology.
The diet characteristics for dairy cattle were based on Donovan (1999) and an extensive review of nearly 20 years
of literature from 1990 through 2009. Estimates of DE were national averages based on the feed components of
the diets observed in the literature for the following year groupings: 1990 through 1993,1994 through 1998,1999
through 2003, 2004 through 2006, 2007, and 2008 onward.4 Base year Ym values by region were estimated using
Donovan (1999). As described in ERG (2016), a ruminant digestion model (COWPOLL, as selected in Kebreab et al.
2008) was used to evaluate Ym for each diet evaluated from the literature, and a function was developed to adjust
regional values over time based on the national trend. Dairy replacement heifer diet assumptions were based on
the observed relationship in the literature between dairy cow and dairy heifer diet characteristics.
For feedlot animals, the DE and Ym values used for 1990 were recommended by Johnson (1999). Values for DE and
Ym for 1991 through 1999 were linearly extrapolated based on the 1990 and 2000 data. DE and Ym values for 2000
onwards were based on survey data in Galyean and Gleghorn (2001) and Vasconcelos and Galyean (2007).
For grazing beef cattle, Ym values were based on Johnson (2002), DE values for 1990 through 2006 were based on
specific diet components estimated from Donovan (1999), and DE values from 2007 onwards were developed from
an analysis by Archibeque (2011), based on diet information in Preston (2010) and USDA-APHIS:VS (2010). Weight
and weight gains for cattle were estimated from Holstein (2010), Doren et al. (1989), Enns (2008), Lippke et al.
(2000), Pinchack et al. (2004), Platter et al. (2003), Skogerboe et al. (2000), and expert opinion. See Annex 3.10 for
more details on the method used to characterize cattle diets and weights in the United States.
Calves younger than 4 months are not included in emission estimates because calves consume mainly milk and the
IPCC recommends the use of a Ym of zero for all juveniles consuming only milk. Diets for calves aged 4 to 6 months
are assumed to go through a gradual weaning from milk decreasing to 75 percent at 4 months, 50 percent at age 5
months, and 25 percent at age 6 months. The portion of the diet made up with milk still results in zero emissions.
For the remainder of the diet, beef calf DE and Ym are set equivalent to those of beef replacement heifers, while
dairy calf DE is set equal to that of dairy replacement heifers and dairy calf Ym is provided at 4 and 7 months of age
by Soliva (2006). Estimates of Ym for 5- and 6-month-old dairy calves are linearly interpolated from the values
provided for 4 and 7 months.
To estimate CFU emissions, the population was divided into state, age, sub-type (i.e., dairy cows and replacements,
beef cows and replacements, heifer and steer stockers, heifers and steers in feedlots, bulls, beef calves 4 to 6
months, and dairy calves 4 to 6 months), and production (i.e., pregnant, lactating) groupings to more fully capture
differences in CFU emissions from these animal types. The transition matrix was used to simulate the age and
weight structure of each sub-type on a monthly basis in order to more accurately reflect the fluctuations that
occur throughout the year. Cattle diet characteristics were then used in conjunction with Tier 2 equations from
IPCC (2006) to produce CFU emission factors for the following cattle types: dairy cows, beef cows, dairy
replacements, beef replacements, steer stockers, heifer stockers, steer feedlot animals, heifer feedlot animals,
bulls, and calves. To estimate emissions from cattle, monthly population data from the transition matrix were
multiplied by the calculated emission factor for each cattle type in each state. More details are provided in Annex
3.10.
4 Due to inconsistencies in the 2003 literature values, the 2002 values were used for 2003 as well.
Agriculture 5-7
-------
Non-Cattle Livestock
Emission estimates for other animal types were based on average emission factors (Tier 1 default IPCC emission
factors) representative of entire populations of each animal type. The methodology is in accordance with the
methodological decision tree for methane emissions from enteric fermentation (IPCC 2019). Methane emissions
from these animals accounted for a minor portion of total Cm emissions from livestock in the United States from
1990 through 2022. Additionally, the variability in emission factors for each of these other animal types (e.g.,
variability by age, production system, and feeding practice within each animal type) is less than that for cattle.
Annual livestock population data for 1990 to 2022 for sheep, swine, goats, horses, mules and asses, and American
bison were obtained for available years from USDA-NASS (USDA 2023; USDA 2019). Horse, goat, and mule and ass
population data were available for 1987,1992,1997, 2002, 2007, 2012, and 2017 (USDA 2019); the remaining
years between 1990 and 2022 were interpolated and extrapolated from the available estimates (with the
exception of goat populations being held constant between 1990 and 1992). American bison population estimates
were available from USDA for 2002, 2007, 2012, and 2017 (USDA 2019) and from the National Bison Association
(1999) for 1990 through 1999. Additional years were based on observed trends from the National Bison
Association (1999), interpolation between known data points, and extrapolation beyond 2012, as described in
more detail in Annex 3.10.
Methane emissions from sheep, goats, swine, horses, American bison, and mules and asses were estimated by
using emission factors utilized in Crutzen et al. (1986, cited in IPCC 2006; IPCC 2019). These emission factors are
representative of typical animal sizes, feed intakes, and feed characteristics in developed countries. For American
bison, the emission factor for buffalo was used and adjusted based on the ratio of live weights to the 0.75 power.
The methodology is the same as that recommended by IPCC (2006).
Uncertainty
A quantitative uncertainty analysis for this source category was performed using the IPCC-recommended Approach
2 uncertainty estimation methodology based on a Monte Carlo stochastic simulation technique as described in ICF
(2003). These uncertainty estimates were developed for the 1990 through 2001 Inventory (i.e., 2003 submission to
the UNFCCC). While there are plans to update the uncertainty to reflect recent methodological updates and
forthcoming changes (see Planned Improvements, below), at this time the uncertainty estimates were directly
applied to the 2022 emission estimates in this Inventory.
A total of 185 primary input variables (177 for cattle and 8 for non-cattle) were identified as key input variables for
the uncertainty analysis. A normal distribution was assumed for almost all activity- and emission factor-related
input variables. Triangular distributions were assigned to three input variables (specifically, cow-birth ratios for the
three most recent years included in the 2001 model run) to ensure only positive values would be simulated. For
some key input variables, the uncertainty ranges around their estimates (used for Inventory estimation) were
collected from published documents and other public sources; others were based on expert opinion and best
estimates. In addition, both endogenous and exogenous correlations between selected primary input variables
were modeled. The exogenous correlation coefficients between the probability distributions of selected activity-
related variables were developed through expert judgment.
Among the individual cattle sub-source categories, beef cattle account for the largest amount of Cm emissions, as
well as the largest degree of uncertainty in the emission estimates—due mainly to the difficulty in estimating the
diet characteristics for grazing members of this animal group. Among non-cattle, horses represent the largest
percent of uncertainty in the uncertainty analysis last conducted in 2001 because the Food and Agricultural
Organization (FAO) of the United Nations population estimates used for horses at that time had a higher degree of
uncertainty than for the USDA population estimates used for swine, goats, and sheep. The horse populations are
5-8 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
drawn from the same USDA source as the other animal types5, and therefore the uncertainty range around horses
is likely overestimated. Cattle calves, American bison, mules and asses were excluded from the initial uncertainty
estimate because they were not included in emission estimates at that time.
The uncertainty ranges associated with the activity data-related input variables were ±10 percent or lower.
However, for many emission factor-related input variables, the lower- and/or the upper-bound uncertainty
estimates were over 20 percent. The results of the quantitative uncertainty analysis are summarized in Table 5-5.
Based on this analysis, enteric fermentation Cm emissions in 2022 were estimated to be between 171.4 and 227.2
MMT CO2 Eq. at a 95 percent confidence level, which indicates a range of 11 percent below to 18 percent above
the 2022 emission estimate of 192.6 MMT CO2 Eq.
As a comparison to the Approach 2, a quantitative uncertainty analysis for this source category was performed
using the IPCC (2006) recommended Approach 1 based on simple error propagation. Enteric fermentation CFU
emissions in 2022 were estimated to be between 132.6 and 252.6 MMT CO2 Eq., which indicates a range of ±31
percent above and below the 2022 emission estimate of 192.6 MMT CO2 Eq. A ±10 percent uncertainty factor is
applied to the activity data (e.g., animal populations), and a ±40 percent default uncertainty factor is applied to the
emission factors (IPCC 2019).
Table 5-5: Approach 2 Quantitative Uncertainty Estimates for CH4 Emissions from Enteric
Fermentation (MMT CO2 Eq. and Percent)
2022 Emission
Source
Gas
Estimate
Uncertainty Range Relative to Emission Estimate1- 1
(MMT CO.. Eq.)
(MMT CO . Eq.)
(%)
Lower Upper
Lower Upper
Bound Bound
Bound Bound
Enteric Fermentation
ch4
192.6
171.4 227.2
-11% +18%
a 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 2022 estimates.
c The overall uncertainty calculated in 2003, and applied to the 2022 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.
/erification
In order to ensure the quality of the emission estimates from enteric fermentation, the General (IPCC Tier 1) and
category-specific (Tier 2) Quality Assurance/Quality Control (QA/QC) procedures were implemented consistent
with the U.S. Inventory QA/QC Plan outlined in Annex 8. Category-specific or Tier 2 QA procedures included
independent review of emission estimate methodologies from previous Inventories.
As part of the quality assurance process, average implied emissions factors for U.S. dairy and beef cattle were
developed based on CEFM output and compared to emission factors for other countries provided by IPCC (2006).
This comparison is discussed in further detail in Annex 3.10.
Over the past few years, particular importance has been placed on harmonizing the data exchange between the
enteric fermentation and manure management source categories. The current Inventory utilizes the same
transition matrix from the CEFM for estimating cattle populations and weights for both source categories, and the
CEFM is used to output volatile solids and nitrogen excretion estimates using the diet assumptions in the model in
5 The change from using FAO data to USDA data for horse populations took place during the development of the 1990 through
2011 Inventory, published in 2013.
Agriculture 5-9
-------
conjunction with the energy balance equations from the IPCC (2006). This approach facilitates the QA/QC process
for both of these source categories.
Recalculations Discussion
In the previous Inventory, 1990 to 2020 estimates were retained from the 1990 through 2020 Inventory, and 2021
estimates were based on a simplified approach that used emission factors and extrapolated population estimates
for all animals. For the current Inventory, the CEFM was used for cattle for all years, resulting in different estimates
for 2021 than the prior Inventory.
For cattle, there were also changes to emissions resulting from activity data changes, including:
• The USDA published minor data revisions that EPA incorporated into the CEFM:
o Calf birth data were revised for 2020;
o Dairy cow milk production values were updated for several states for 2020;
o Slaughter data were revised for 2020.
• EPA revised annual milk fat values in the CEFM from 2000 through 2021 with updated annual values from
USDA's Economic Research Services (ERS) dairy data (USDA 2022). In the previous Inventory, EPA derived
annual averages from monthly ERS milk fat values, which is no longer available beyond 2010 (USDA 2021).
• EPA discovered and corrected an error within the CEFM related to the urinary energy input used for
feedlot cattle, which affected VS results for this animal group. The urinary energy default was updated
from 0.04 to 0.02 for feedlot cattle. These updates will affect values in Section 5.2 Manure Management.
Planned Improvements
Regular annual data reviews and updates are necessary to maintain an emissions inventory that reflects the
current base of knowledge. In addition to the documented approaches currently used to address data availability,
EPA conducts the following annual assessments to identify and determine the applicability of newer data when
updating the estimates to extend time series each year:
• Further research to improve the estimation of dry matter intake (as gross energy intake) using data from
appropriate production systems;
• Updating input variables that are from older data sources, such as beef births by month, beef and dairy
annual calving rates, and beef cow lactation rates;
• Investigating the availability of data for dairy births by month, to replace the current assumption that
births are evenly distributed throughout the year;
• Investigating the availability of annual data for the DE, Ym, and crude protein values of specific diet and
feed components for grazing and feedlot animals (including investigating the availability of existing
models to estimate diet characteristics, as well as the use and impact of feed additives on emissions);
• Further investigation on additional sources or methodologies for estimating DE for dairy cattle, given the
many challenges in characterizing dairy cattle diets;
• Further evaluation of the assumptions about weights and weight gains for beef cows, such that trends
beyond 2007 are updated, rather than held constant; and
• Further evaluation of the estimated weight for dairy cows (i.e., 1,500 lbs) that is based solely on Holstein
cows as mature dairy cow weight is likely slightly overestimated, based on knowledge of the breeds of
dairy cows in the United States.
5-10 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
Depending upon the outcome of ongoing investigations, future improvement efforts for enteric fermentation
could include some of the following options which are additional to the regular updates, and may or may not have
implications for regular updates once addressed:
• Potentially updating to a Tier 2 methodology for other animal types (i.e., sheep, swine, goats, horses).
Efforts to move to Tier 2 will consider the emissions significance of livestock types;
• Investigation of methodologies and emission factors for including enteric fermentation emission
estimates from poultry;
• Comparison of the current CEFM with other models that estimate enteric fermentation emissions for
quality assurance and verification;
• Investigation of recent research implications suggesting that certain parameters in enteric models may be
simplified without significantly diminishing model accuracy; and
• Recent changes that have been implemented to the CEFM warrant an assessment of the current
uncertainty analysis; therefore, a revision of the quantitative uncertainty surrounding emission estimates
from this source category will be initiated. EPA plans to perform this uncertainty analysis following the
completed updates to the CEFM.
EPA is continuously investigating these recommendations and potential improvements and working with USDA and
other experts to utilize the best available data and methods for estimating emissions. Many of these
improvements are major updates and may take multiple years to implement in full.
5.2 Manure Management (CRT Source
Category 3B)
The treatment, storage, and transportation of livestock manure can produce anthropogenic CFU and N2O
emissions.6 Methane is produced by the anaerobic decomposition of manure and nitrous oxide is produced from
direct and indirect pathways through the processes of nitrification and denitrification; in addition, there are many
underlying factors that can affect these resulting emissions from manure management, as described below.
When livestock manure is stored or treated in systems that promote anaerobic conditions (e.g., as a liquid/slurry in
lagoons, ponds, tanks, or pits), the decomposition of the volatile solids component in the manure tends to produce
Cm. When manure is handled as a solid (e.g., in stacks or drylots) or deposited on pasture, range, or paddock
lands, it tends to decompose aerobically and produce CO2 and little or no CH4. Ambient temperature, moisture,
and manure storage or residency time affect the amount of CH4 produced because they influence the growth of
the bacteria responsible for CH4 formation. For non-liquid-based manure systems, moist conditions (which are a
function of rainfall and humidity) can promote CH4 production. Manure composition, which varies by animal diet,
growth rate, and animal type (particularly the different animal digestive systems), also affects the amount of CH4
produced. In general, the greater the energy content of the feed, the greater the potential for CH4 emissions.
However, some higher-energy feeds also are more digestible than lower quality forages, which can result in less
overall waste excreted from the animal.
As previously stated, N2O emissions are produced through both direct and indirect pathways. Direct N2O emissions
are produced as part of the nitrogen (N) cycle through the nitrification and denitrification of the N in livestock dung
6 C02 emissions from livestock are not estimated because annual net C02 emissions are assumed to be zero - the C02
photosynthesized by plants is returned to the atmosphere as respired C02 (IPCC 2006).
Agriculture 5-11
-------
and urine.7 There are two pathways for indirect N2O emissions. The first is the result of the volatilization of N in
manure (as NH3 and NOx) and the subsequent deposition of these gases and their products (NH4+ and NO3") onto
soils and the surface of lakes and other waters. The second pathway is the runoff and leaching of N from manure
into the groundwater below, into riparian zones receiving drain or runoff water, or into the ditches, streams,
rivers, and estuaries into which the land drainage water eventually flows.
The production of direct N2O emissions from livestock manure depends on the composition of the manure
(manure includes both feces and urine), the type of bacteria involved in the process, and the amount of oxygen
and liquid in the manure system. For direct N2O emissions to occur, the manure must first be handled aerobically
where organic N is mineralized or decomposed to NH4 which is then nitrified to NO3 (producing some N2O as a
byproduct) (nitrification). Next, the manure must be handled anaerobically where the nitrate is then denitrified to
N2O and N2 (denitrification). NOx can also be produced during denitrification (Groffman et al. 2000; Robertson and
Groffman 2015). These emissions are most likely to occur in dry manure handling systems that have aerobic
conditions, but that also contain pockets of anaerobic conditions due to saturation. Avery small portion of the
total N excreted is expected to convert to N2O in the waste management system (WMS).
Indirect N2O emissions are produced when nitrogen is lost from the system through volatilization (as NH3 or NOx)
or through runoff and leaching. The vast majority of volatilization losses from these operations are NH3. Although
there are also some small losses of NOx, there are no quantified estimates available for use, so losses due to
volatilization are only based on NH3 loss factors. Runoff losses would be expected from operations that house
animals or store manure in a manner that is exposed to weather. Runoff losses are also specific to the type of
animal housed on the operation due to differences in manure characteristics. Little information is known about
leaching from manure management systems as most research focuses on leaching from land application systems.
However, storage systems are often designed to minimize leaching (e.g., clay soil or synthetic liners in lagoons).
Since leaching losses are expected to be minimal, leaching losses are coupled with runoff losses and the
runoff/leaching estimate provided in this chapter does not account for any leaching losses.
Estimates of CFU emissions from manure management in 2022 were 64.7 MMT CO2 Eq. (2,312 kt); in 1990,
emissions were 39.1 MMT CO2 Eq. (1,398 kt). This represents a 65 percent increase in emissions from 1990.
Emissions increased on average by 0.8 MMT CO2 Eq. (2 percent) annually over this period. The majority of this
increase is due to dairy cattle and beef cattle manure, where emissions increased 109 and 146 percent,
respectively. From 2021 to 2022, there was a 3 percent decrease in total CFU emissions from manure
management, mainly due to a decrease in swine, dairy, and beef cattle populations.
Although a large quantity of managed manure in the United States is handled as a solid, producing little CH4, the
general trend in manure management, particularly for dairy cattle and swine (which are both shifting towards
larger facilities), is one of increasing use of liquid systems. Also, new regulations controlling the application of
manure nutrients to land have shifted manure management practices at smaller dairies from daily spread systems
to storage and management of the manure on site. In many cases, manure management systems with the most
substantial methane emissions are those associated with confined animal management operations where manure
is handled in liquid-based systems. Nitrous oxide emissions from manure management vary significantly between
the types of management system used and can also result in indirect emissions due to other forms of nitrogen loss
from the system (IPCC 2006).
While national dairy animal populations have decreased since 1990, some states have seen increases in their dairy
cattle populations as the industry becomes more concentrated in certain areas of the country and the number of
animals contained on each facility increases. These areas of concentration, such as California, New Mexico, and
Idaho, tend to utilize more liquid-based systems to manage (flush or scrape) and store manure. Thus, the shift
7 Direct and indirect N20 emissions from dung and urine spread onto fields either directly as daily spread or after it is removed
from manure management systems (i.e., 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.
5-12 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
toward larger dairy cattle and swine facilities since 1990 has translated into an increasing use of liquid manure
management systems, which have higher potential Cm emissions than dry systems. This significant shift in both
the dairy cattle and swine industries was accounted for by incorporating state and WMS-specific Cm conversion
factor (MCF) values in combination with the 1992,1997, 2002, 2007, 2012, and 2017 farm-size distribution data
reported in the U.S. Department of Agriculture (USDA) Census of Agriculture (USDA 2019d).
In 2022, total N2O emissions from manure management were estimated to be 17.0 MMT CO2 Eq. (64 kt); in 1990,
emissions were 13.4 MMT CO2 Eq. (50 kt). These values include both direct and indirect N2O emissions from
manure management. Nitrous oxide emissions have increased since 1990. Multiple drivers increase N2O emissions,
such as increasing nitrogen excretion rates for some animal types (see Annex, Table A-163) and increasing
numbers of animals on feedlots versus other dry systems (e.g., pasture). Across the entire time series, the overall
net effect is that N2O emissions showed a 27 percent increase from 1990 to 2022, but recent declines in a few
animal populations (e.g., swine and dairy cattle) resulted in a 0.9 percent decrease from 2021 to 2022.
Table 5-6 and Table 5-7 provide estimates of CH4 and N2O emissions from manure management by animal
category.8
Table 5-6: CH4 and N2O Emissions from Manure Management (MMT CO2 Eq.)
Gas/Animal Type
1990
2005
2018
2019
2020
2021
2022
CH4a
39.1
55.0
67.7
66.7
66.9
66.4
64.7
Dairy Cattle
16.0 I
26.4 1
35.7
34.4
34.7
34.3
33.4
Swine
17.4
22.7
24.7
24.9
24.9
24.6
23.8
Poultry
3.8 1
3.4
3.0
3.1
3.0
3.0
3.0
Beef Cattle
1.8
2.2
4.2
4.1
4.2
4.4
4.3
Horses
o-i
0.1
0.1
0.1
0.1
0.1
0.1
Sheep
0.1
0.1
0.1
0.1
0.1
0.1
0.1
Goats
111111!
+
+ iiiiii
+
+
+
+
+
American Bison
+
+
+
+
+
+
+
Mules and Asses
+ 1
+"
+
+
+
+
+
N2Ob
13.4
15.2
16.6
16.8
16.9
17.1
17.0
Beef Cattle
5.2:
6.0 ii
5.9
6.0
6.1
6.4
6.4
Dairy Cattle
5.5
5.5
6.2
6.2
6.2
6.3
6.2
Swine
1-11
1.5
1.8
1.8
1.9
1.8
1.8
Poultry
1.3
1.8
2.3
2.3
2.3
2.3
2.3
Sheep
o-i 1
0.3
0.3
0.3
0.3
0.3
0.3
Horses
0.1
0.1
0.1
0.1
0.1
0.1
0.1
Goats
+ 1
+ «
+
+
+
+
+
Mules and Asses
+
+
+
+
+
+
+
American Bisonc
NA
NA I
NA
NA
NA
NA
NA
Total
52.5
70.2
84.3
83.5
83.8
83.6
81.7
+ Does not exceed 0.05 MMT C02 Eq.
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 N20 emissions.
cThere are no American bison N20 emissions from managed systems; American bison are maintained
entirely on pasture, range, and paddock.
8 Manure management emissions from camels are not estimated because there is no significant population of camels in the
United States. Given the insignificance of estimated camel emissions in terms of the overall level and trend in national
emissions, there are no immediate improvement plans to include this emissions category in the Inventory. See Annex 5 for
more information on significance of estimated camel emissions.
Agriculture 5-13
-------
Notes: N20 emissions from manure deposited on pasture, range and paddock are included in the
agricultural soils management category. Totals may not sum due to independent rounding.
Table 5-7: CH4 and N2O Emissions from Manure Management (kt)
Gas/Animal Type
1990
2005
2018
2019
2020
2021
2022
CH4a
1,398
1,964
2,418
2,382
2,390
2,373
2,312
Dairy Cattle
572
9431
1,274
1,227
1,238
1,226
1,193
Swine
621
812
882
890
888
877
851
Poultry
135 1
123 =!
108
111
109
108
108
Beef Cattle
63
78
149
148
150
157
154
Horses
4
5!
3
3
3
3
2
Sheep
3
2
2
2
2
2
2
Goats
+ 1
+
+
+
+
+
American Bison
+
+
+
+
+
+
+
Mules and Asses
+
iiiii!!
+
+
+
+
+
N2Ob
50
57
63
63
64
65
64
Beef Cattle
NJ
O
llllli!
23 1
22
23
23
24
24
Dairy Cattle
21
21
23
23
24
24
23
Swine
4
6
7
7
7
7
7
Poultry
5
7
9
9
9
9
9
Sheep
;;;;;;;
+ !!!!!!
1 jl
1
1
1
1
1
Horses
+
+
+
+
+
+
+
Goats
¦
+ ¦
!!!!!!!
+ V
+
+
+
+
+
Mules and Asses
+
+
+
+
+
+
+
American Bisonc
NA =
NA 1
NA
NA
NA
NA
NA
+ Does not exceed 0.5 kt.
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 N20 emissions.
cThere are no American bison N20 emissions from managed systems; American bison are maintained
entirely on pasture, range, and paddock.
Notes: N20 emissions from manure deposited on pasture, range and paddock are included in the
agricultural soils management category. Totals by gas may not sum due to independent rounding.
Methodology and Time-Series Consistency
The methodologies presented in IPCC (2006) form the basis of the Cm and N2O emission estimates for each animal
type, including Tier 1, Tier 2, and use of the CEFM previously described for enteric fermentation. These
methodologies use:
• IPCC (2019) Tier 1 default N2O emission factors and MCFs for dry systems
• U.S. specific MCFs for liquid systems (ERG 2001)
• U.S. specific values for volatile solids (VS) production rate and nitrogen excretion rate for some animal
types, including cattle values from the CEFM
This combination of Tier 1 and Tier 2 methods was applied to all livestock animal types and follows guidance for
methodological choice presented in decision trees from the IPCC (2006). This section presents a summary of the
methodologies used to estimate CH4 and N2O emissions from manure management.
See Annex 3.11 for more detailed information on the methodologies (including detailed formulas and emission
factors), data used to calculate CH4 and N2O emissions, and emission results (including input variables and results
at the state-level) from manure management.
5-14 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
Methane Calculation Methods
The following inputs were used in the calculation of manure management Cm emissions for 1990 through 2022:
• Animal population data (by animal type and state);
• Typical animal mass (TAM) data (by animal type);
• Portion of manure managed in each WMS, by state and animal type;
• VS production rate (by animal type and state or United States);
• Methane producing potential (BO) of the volatile solids (by animal type); and
• Methane conversion factors (MCF), the extent to which the Cm producing potential is realized for each
type of WMS (by state and manure management system, including the impacts of any biogas collection
efforts).
Methane emissions were estimated by first determining activity data, including animal population, TAM, WMS
usage, and waste characteristics. The activity data sources are described below:
• Annual animal population data for 1990 through 2022 for all livestock types, except goats, horses, mules
and asses, and American bison were obtained from the USDA-NASS. For cattle, the USDA populations
were utilized in conjunction with birth rates, detailed feedlot placement information, and slaughter
weight data to create the transition matrix in the Cattle Enteric Fermentation Model (CEFM) that models
cohorts of individual animal types and their specific emission profiles. The key variables tracked for each
of the cattle population categories are described in Section 5.1 and in more detail in Annex 3.10. Goat
population data for 1992,1997, 2002, 2007, 2012, and 2017; horse and mule and ass population data for
1987,1992,1997, 2002, 2007, 2012, and 2017; and American bison population for 2002, 2007, 2012, and
2017 were obtained from the Census of Agriculture (USDA 2019d). American bison population data for
1990 through 1999 were obtained from the National Bison Association (1999).
• The TAM is an annual average weight that was obtained for animal types other than cattle from
information in USDA's Agricultural Waste Management Field Handbook (USDA 1996), the American
Society of Agricultural Engineers, Standard D384.1 (ASAE 1998) and others (Meagher 1986; EPA 1992;
Safley 2000; ERG 2003b; IPCC 2006; ERG 2010a). For a description of the TAM data used for cattle, see
Annex 3.10.
• WMS usage was estimated for swine and dairy cattle for different farm size categories using state and
regional data from USDA (USDA APHIS 1996; Bush 1998; Ott 2000; USDA 2016c) and EPA (ERG 2000a; EPA
2002a and 2002b; ERG 2018, ERG 2019). For beef cattle and poultry, manure management system usage
data were not tied to farm size but were based on other data sources (ERG 2000a; USDA APHIS 2000; UEP
1999, ERG 2023). For other animal types, manure management system usage was based on previous
estimates (EPA 1992). American bison WMS usage was assumed to be the same as not on feed (NOF)
cattle, while mules and asses were assumed to be the same as horses.
• VS production rates for all cattle except for calves were calculated by head for each state and animal type
in the CEFM. VS production rates by animal mass for all other animals were determined using data from
USDA's Agricultural Waste Management Field Handbook (USDA 1996 and 2008; ERG 2010b and 2010c)
and data that was not available in the most recent Handbook were obtained from the American Society of
Agricultural Engineers, Standard D384.1 (ASAE 1998) or the 2006 IPCC Guidelines (IPCC 2006). American
bison VS production was assumed to be the same as NOF bulls.
• Bo was determined for each animal type based on literature values (Morris 1976; Bryant et al. 1976;
Hashimoto 1981; Hashimoto 1984; EPA 1992; Hill 1982; Hill 1984).
• MCFs for dry systems were set equal to default IPCC factors based on state climate for each year (IPCC
2019). The IPCC 2019 factors are more representative of U.S. systems and reflect the latest science. MCFs
for liquid/slurry, anaerobic lagoon, and deep pit systems were calculated based on the forecast
performance of biological systems relative to temperature changes as predicted in the van't Hoff-
Arrhenius equation which is consistent with IPCC (2006) Tier 2 methodology.
Agriculture 5-15
-------
• Data from anaerobic digestion systems with CFU capture and combustion were obtained from the EPA
AgSTAR Program, including information available in the AgSTAR project database (EPA 2023). Anaerobic
digester emissions were calculated based on estimated methane production and collection and
destruction efficiency assumptions (ERG 2008).
• For all cattle except for calves, the estimated amount of VS (kg per animal-year) managed in each WMS
for each animal type, state, and year were taken from the CEFM, assuming American bison VS production
to be the same as NOF bulls. For animals other than cattle, the annual amount of VS (kg per year) from
manure excreted in each WMS was calculated for each animal type, state, and year. This calculation
multiplied the animal population (head) by the VS excretion rate (kg VS per 1,000 kg animal mass per
day), the TAM (kg animal mass per head) divided by 1,000, the WMS distribution (percent), and the
number of days per year (365.25).
The estimated amount of VS managed in each WMS was used to estimate the Cm emissions (kg CFU per year) from
each WMS. The amount of VS (kg per year) was multiplied by the Bo (m3 CFU per kg VS), the MCF for that WMS
(percent), and the density of Cm (kg Cm per m3 Cm). The CFU emissions for each WMS, state, and animal type
were summed to determine the total U.S. Cm emissions. See details in Step 5 of Annex 3.11.
Nitrous Oxide Calculation Methods
The following inputs were used in the calculation of direct and indirect manure management N2O emissions for
1990 through 2022:
• Animal population data (by animal type and state);
• TAM data (by animal type);
• Portion of manure managed in each WMS (by state and animal type);
• Total Kjeldahl N excretion rate (Nex);
• Direct N2O emission factor (EFwms);
• Indirect N2O emission factor for volatilization (EFvoiatiiization);
• Indirect N2O emission factor for runoff and leaching (EFrunoff/ieach);
• Fraction of N loss from volatilization of NH3 and NOx (Fracgas); and
• Fraction of N loss from runoff and leaching (Fracmnoff/ieach).
Nitrous oxide emissions were estimated by first determining activity data, including animal population, TAM, WMS
usage, and waste characteristics. The activity data sources (except for population, TAM, and WMS, which were
described above) are described below:
• Nex for all cattle except for calves were calculated by head for each state and animal type in the CEFM.
Nex rates by animal mass for all other animals were determined using data from USDA's Agricultural
Waste Management Field Handbook (USDA 1996 and 2008; ERG 2010b and 2010c) and data from the
American Society of Agricultural Engineers, Standard D384.1 (ASAE 1998) and IPCC (2006). American bison
Nex were assumed to be the same as NOF bulls.9
• All N2O emission factors (direct and indirect) were taken from IPCC (2006).
• Country-specific estimates for the fraction of N loss from volatilization (Fracgas) and runoff and leaching
(FraCrunoff/ieach) were developed. Fracgas values were based on WMS-specific volatilization values as
estimated from EPA's National Emission Inventory - Ammonia Emissions from Animal Agriculture
9 Nex of American bison on grazing lands are accounted for and discussed in the agricultural soil management source category
and included under pasture, range and paddock (PRP) emissions. Because American bison are maintained entirely on
unmanaged WMS and N20 emissions from unmanaged WMS are not included in the manure management source category,
there are no N20 emissions from American bison included in the manure management source category.
5-16 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
Operations (EPA 2005). Fracmnoff/ieaching values were based on regional cattle runoff data from EPA's Office
of Water (EPA 2002b; see Annex 3.11).
To estimate N2O emissions for cattle (except for calves), the estimated amount of N excreted (kg per animal-year)
that is managed in each WMS for each animal type, state, and year were taken from the CEFM. For calves and
other animals, the amount of N excreted (kg per year) in manure in each WMS for each animal type, state, and
year was calculated. The population (head) for each state and animal was multiplied by TAM (kg animal mass per
head) divided by 1,000, the nitrogen excretion rate (Nex, in kg N per 1,000 kg animal mass per day), WMS
distribution (percent), and the number of days per year.
Direct N2O emissions were calculated by multiplying the amount of N excreted (kg per year) in each WMS by the
N2O direct emission factor for that WMS (EFwms, in kg N2O-N per kg N) and the conversion factor of N2O-N to N2O.
These emissions were summed over state, animal, and WMS to determine the total direct N2O emissions (kg of
N2O per year). See details in Step 6 of Annex 3.11.
Indirect N2O emissions from volatilization (kg N2O per year) were then calculated by multiplying the amount of N
excreted (kg per year) in each WMS by the fraction of N lost through volatilization (Fracgas) divided by 100, the
emission factor for volatilization (EFvoiatiiization, in kg N2O per kg N), and the conversion factor of N2O-N to N2O.
Indirect N2O emissions from runoff and leaching (kg N2O per year) were then calculated by multiplying the amount
of N excreted (kg per year) in each WMS by the fraction of N lost through runoff and leaching (Fracmnoff/ieach)
divided by 100, the emission factor for runoff and leaching (EFrunoff/ieach, in kg N2O per kg N), and the conversion
factor of N2O-N to N2O. The indirect N2O emissions from volatilization and runoff and leaching were summed to
determine the total indirect N2O emissions. See details in Step 6 of Annex 3.11.
Following these steps, direct and indirect N2O emissions were summed to determine total N2O emissions (kg N2O
per year) for the years 1990 to 2022.
Methodological approaches, changes to historic data, and other parameters were applied to the entire time series
to ensure consistency in emissions estimates from 1990 through 2022. In some cases, the activity data source
changed over the time series. For example, updated WMS distribution data were applied to 2016 for dairy cows
and 2009 for swine. While previous WMS distribution data were from another data source, EPA integrated the
more recent data source to reflect the best available current WMS distribution data for these animals. EPA
assumed a linear interpolation distribution for years between the two data sources. Refer to Annex 3.11 for more
details on data sources and methodology.
Uncertainty
An analysis (ERG 2003a) was conducted for the manure management emission estimates presented in the 1990
through 2001 Inventory (i.e., 2003 submission to the UNFCCC) to determine the uncertainty associated with
estimating Cm and N2O emissions from livestock manure management. The quantitative uncertainty analysis for
this source category was performed in 2002 through the IPCC-recommended Approach 2 uncertainty estimation
methodology, the Monte Carlo stochastic simulation technique. The uncertainty analysis was developed based on
the methods used to estimate Cm and N2O emissions from manure management systems. The series of equations
used were condensed into a single equation for each animal type and state. The equations for each animal group
contained four to five variables around which the uncertainty analysis was performed for each state. A normal
probability distribution was assumed for all variables in the estimation equations. While there are plans to update
the uncertainty to reflect recent manure management updates and forthcoming changes (see Planned
Improvements, below), at this time the uncertainty estimates were directly applied to the 2022 emission
estimates.
The results of the Approach 2 quantitative uncertainty analysis are summarized in Table 5-8. Manure management
Cm emissions in 2022 were estimated to be between 53.1 and 77.7 MMT CO2 Eq. at a 95 percent confidence level,
which indicates a range of 18 percent below to 20 percent above the actual 2022 emission estimate of 64.7 MMT
CO2 Eq. At the 95 percent confidence level, N2O emissions were estimated to be between 14.3 and 21.1 MMT CO2
Agriculture 5-17
-------
Eq. (or approximately 16 percent below and 24 percent above the actual 2022 emission estimate of 17.0 MMT CO2
Eq.).
A quantitative uncertainty analysis for this source category was also performed using the IPCC (2006)
recommended Approach 1 based on simple error propagation as well. Based on this analysis, manure
management:
• Cm emissions in 2022 were estimated to be between 50.4 and 79.0 MMT CO2 Eq., which indicates a
range of ±21 percent above and below the 2022 emission estimate of 64.7 MMT CO2 Eq. A ±25 percent
uncertainty factor is applied to the activity data (e.g., animal populations), and a ±30 percent default
uncertainty factor for Tier 1 and ±20 percent default uncertainty factor for Tier 2 is applied to the
emission factors (IPCC 2006).
• N2O emissions in 2022 were estimated to be between 11.8 and 22.2 MMT CO2 Eq., which indicates a
range of ±31 percent above and below the 2022 emission estimate of 17.0 MMT CO2 Eq. A ±25 percent
uncertainty factor is applied to the activity data (e.g., animal populations), and a ±50 percent default
uncertainty factor is applied to the emission factors (IPCC 2006).
• CH4 and N2O emissions in 2022 were estimated to be between 66.5 and 96.9 MMT CO2 Eq., which
indicates a range of ±19 percent above and below the 2022 emission estimate of 81.7 MMT CO2 Eq. A ±25
percent uncertainty factor is applied to the activity data (e.g., animal populations), and a ±20-50 percent
default uncertainty factor is applied to the emission factors (IPCC 2006).
Table 5-8: Approach 2 Quantitative Uncertainty Estimates for CH4 and N2O (Direct and
Indirect) Emissions from Manure Management (MMT CO2 Eq. and Percent)
2022 Emission
Source
Gas
Estimate
Uncertainty Range Relative to Emission Estimate-1
(MMTCO. Eq.)
(MMTCO. Eq.)
(%)
Lower Upper
Lower Upper
Bound Bound
Bound Bound
Manure Management
ch4
64.7
53.1 77.7
-18% +20%
Manure Management
n2o
17.0
14.3 21.1
-16% +24%
a Range of emission estimates predicted by Monte Carlo stochastic simulation for a 95 percent confidence interval.
QA/QC and Verification
General (Tier 1) and category-specific (Tier 2) QA/QC activities were conducted consistent with the U.S. Inventory
QA/QC plan outlined in Annex 8. 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.
In addition, 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, both for pasture and managed systems, and the sum
of county estimates for the full time series. This was done to ensure consistency between excreted N within the
manure management sector and those data provided to the managed soils sector. All errors identified were
corrected.
Time-series data, including population, are validated by experts to ensure they are representative of the best
available U.S.-specific data. The U.S.-specific values for TAM, Nex, VS, Bo, 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
5-18 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
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 IPCC (2006) default values.
For additional verification of the 1990 to 2022 estimates, the implied Cm emission factors for manure
management (kg of Cm per head per year) were compared against the default IPCC (2006) values. Table 5-9
presents the implied emission factors of kg of Cm per head per year used for the manure management emission
estimates as well as the IPCC (2006) default emission factors. The U.S. implied emission factors fall within the
range of the IPCC (2006) default values, except in the case of sheep, goats, and some years for horses and dairy
cattle. The U.S. implied emission factors are less than the IPCC (2006) 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 cattle and swine across the time series. This increase reflects the dairy cattle and swine industry trend
towards larger farm sizes; large farms are more likely to manage manure as a liquid and therefore produce more
Cm emissions. See the Recalculations for explanations for changes that affect emissions which impact these
implied emission factors.
Table 5-9: IPCC (2006) Implied Emission Factor Default Values Compared with Calculated
Values for Cmfrom Manure Management (kg/head/year)
IPCC Default
CH.i Emission Factors
Animal Type
(kg/head/year)'1
1990
2005
2018
2019
2020
2021
2022
Dairy Cattle
48-112
29.3
53.0
67.0
65.0
65.9
65.0
64.1
Beef Cattle
1-2
0.8 =
0.9 1
1.8
1.8
1.8
1.9
1.9
Swine
10-45
11.5
13.3
12.0
11.6
11.5
11.8
11.6
Sheep
0.19-0.37
0.3::
0.4
0.4
0.4
0.4
0.4
0.4
Goats
0.13-0.26
0.1
0.1
0.1
0.1
0.1
0.1
0.1
Poultry
0.02-1.4
0.1 =
o-i 1
0.05
0.05
0.05
0.05
0.05
Horses
1.56-3.13
1.9
1.4
1.2
1.2
1.2
1.2
1.2
American Bison
NA
11111
00
O
0.9 I
0.9
0.9
0.9
0.9
0.9
Mules and Asses
0.76-1.14
0.4
0.4
0.4
0.4
0.4
0.4
0.4
NA (Not Applicable)
a Ranges reflect 2006 IPCC Guidelines (Volume 4, Table 10.14) default emission factors for North America across
different climate zones.
In addition, default IPCC (2006) 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.
Recalculations Discussion
In the previous Inventory, 1990 to 2020 estimates were retained from the 1990 through 2020 Inventory, and 2021
estimates were based on a simplified approach that used emission factors and extrapolated population estimates
for all animals. For the current Inventory, the calculations were rerun for all years, resulting in different estimates
for 2021 than the prior Inventory.
There were also changes to emissions resulting from activity data changes, including:
• EPA updated the WMS distributions for broilers, layers, and beef feedlot animal types. For broilers, this
affected 1993 through 2021, for layers 2000 through 2021, and for beef feedlots all years of the time
series (ERG 2023).
Agriculture 5-19
-------
• EPA updated the calf TAM values to coincide with values used within the CEFM. This affected all years of
the time series.
• EPA updated the solid storage direct N2O emission factor to the updated guidance provided in IPCC
(2019).
• EPA updated how poultry digesters were applied, splitting other poultry and caged layers (previously
done for broilers) as well as the year for which select swine anaerobic digesters were shutdown per notes
provided in AgSTAR.
• EPA discovered and corrected an error within the CEFM (see NIR section 5.1 and annex 3.10) related to
the urinary energy input used for feedlot cattle, which affected VS results for this animal group. The
urinary energy default was updated from 0.04 to 0.02 for feedlot cattle.
The cumulative effect of all these recalculations had a minor impact on the overall manure management emissions
estimates:
• Cm emissions increased an average 0.6 percent over the time series, with the largest decrease of 0.2
percent (0.1 MMT CO2 Eq) in 2002 to the largest increase of 1.8 percent (1.2 MMT CO2 Eq.) in 2017.
• N2O emissions increased an average 3.4 percent over the time series, with the largest decrease of 3.6
percent (0.6 MMT CO2 Eq.) in 2020 and the largest increase of 7.8 percent (1.0 MMT CO2 Eq.) in 1990.
• Over the time series the average total emissions increased by 1.2 percent from the previous Inventory.
The changes ranged from the largest decrease 0.5 percent (0.4 MMT CO2 Eq.) in 2020, to the largest
increase 2.1 percent (1.1 MMT CO2 Eq.) in 1990.
Planned Improvements
Regular annual data reviews and updates are necessary to maintain an emissions inventory that reflects the
current base of knowledge. In addition to the documented approaches currently used to address data availability,
EPA conducts data assessments to pursue a number of potential improvements.
Potential improvements (long-term improvements) for future Inventory years include:
• Providing supplemental details on CH4 emissions reductions due to the use of anaerobic digesters (the
Inventory currently estimates only emissions from anaerobic digestion systems).
• Investigating the updated IPCC 2019 Refinement default N2O emissions factor for anaerobic digesters.
Historically, EPA has not estimated N2O emissions from digesters as the default guidance was no
emissions. Incorporating AgSTAR data for N2O emissions, like CH4 emissions, is a longer-term goal for EPA.
• Investigating updates to the current anaerobic digester MCFs based on IPCC (2019).
EPA is aware of the following potential updates or improvements but notes that implementation will be based on
available resources and data availability:
• Updating the Bo data used in the Inventory, as data become available. EPA is conducting outreach with
counterparts from USDA as to available data and research on Bo.
• Comparing CH4 and N2O emission estimates with estimates from other models and more recent studies
and compare the results to the Inventory.
• Comparing manure management emission estimates with on-farm measurement data to identify
opportunities for improved estimates.
• Comparing VS and Nex data to literature data to identify opportunities for improved estimates.
5-20 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
• Determining if there are revisions to the U.S.-specific method for calculating liquid systems for MCFs
based on updated guidance from the IPCC 2019 Refinement. EPA previously began this investigation to
determine the potential differences between the methods.
• Investigating improved emissions estimate methodologies for swine pit systems with less than one month
of storage (the updated swine WMS data included this WMS category).
• Improving the linkages with the Enteric Fermentation source category estimates. For future Inventories, it
may be beneficial to have the CEFM and Manure Management calculations in the same model, as they
rely on much of the same activity data and on each other's outputs to properly calculate emissions. EPA
has begun this investigation and plans to develop a model to calculate emissions for these two categories.
• Continuing to investigate new sources of WMS data. EPA is collaborating with the USDA to collect or use
existing survey data for potential improvements to the Inventory.
• Revising the uncertainty analysis to address changes that have been implemented to the Cm and N2O
estimates. The plan is to align the timing of the updated Manure Management uncertainty analysis with
the uncertainty analysis for Enteric Fermentation.
5.3 Rice Cultivation (CRT Source Category
3C)
Most of the world's rice is grown on flooded fields (Baicich 2013) that create anaerobic conditions leading to CH4
production through a process known as methanogenesis. Approximately 60 to 90 percent of the CH4 produced by
methanogenic bacteria in flooded rice fields is oxidized in the soil and converted to CO2 by methanotrophic
bacteria. The remainder is emitted to the atmosphere (Holzapfel-Pschorn et al. 1985; Sass et al. 1990) or
transported as dissolved CH4 into groundwater and waterways (Neue et al. 1997). Methane is transported to the
atmosphere primarily through the rice plants, but some CH4 also escapes via ebullition (i.e., bubbling through the
water) and to a much lesser extent by diffusion through the water (van Bodegom et al. 2001).
Water management is arguably the most important factor affecting CH4 emissions in rice cultivation, and improved
water management has the largest potential to mitigate emissions (Yan et al. 2009). Upland rice fields are not
flooded, and therefore do not produce CH4, but large amounts of CFUcan be emitted in continuously irrigated
fields, which is the most common practice in the United States (USDA 2012). Single or multiple aeration events
with drainage of a field during the growing season can significantly reduce these emissions (Wassmann et al.
2000a), but drainage may also increase N2O emissions. Deepwater rice fields (i.e., fields with flooding depths
greater than one meter, such as natural wetlands) tend to have fewer living stems reaching the soil, thus reducing
the amount of CH4 transport to the atmosphere through the plant compared to shallow-flooded systems (Sass
2001).
Other management practices also influence CH4 emissions from flooded rice fields including rice residue straw
management and application of organic amendments, in addition to cultivar selection due to differences in the
amount of root exudates10 among rice varieties (Neue et al. 1997). These practices influence the amount of
organic matter available for methanogenesis, and some practices, such as mulching rice straw or composting
organic amendments, can reduce the amount of labile carbon and limit CH4 emissions (Wassmann et al. 2000b).
10 The roots of rice plants add organic material to the soil through a process called "root exudation." Root exudation is thought
to enhance decomposition of the soil organic matter and release nutrients that the plant can absorb for production. The
amount of root exudate produced by a rice plant over a growing season varies among rice varieties.
Agriculture 5-21
-------
Fertilization practices also influence CFU emissions, particularly the use of fertilizers with sulfate, which can reduce
Cm emissions (Wassmann et al. 2000b; Linquist et al. 2012). Other environmental variables also impact the
methanogenesis process such as soil temperature and soil type. Soil temperature regulates the activity of
methanogenic bacteria, which in turn affects the rate of CFU production. Soil texture influences decomposition of
soil organic matter but is also thought to have an impact on oxidation of CFU in the soil (Sass et al. 1994).
Rice is currently cultivated in 12 states, including Arkansas, California, Florida, Illinois, Kentucky, Louisiana,
Minnesota, Mississippi, Missouri, New York, Tennessee, and Texas. Soil types, rice varieties, and cultivation
practices vary across the United States, but most farmers apply fertilizers and do not harvest crop residues. In
addition, a second, ratoon rice crop is sometimes grown in the Southeastern region of the country. Ratoon crops
are produced from regrowth of the stubble remaining after the harvest of the first rice crop. Methane emissions
from ratoon crops are higher than those from the primary crops due to the increased amount of labile organic
matter available for anaerobic decomposition in the form of relatively fresh crop residue straw. Emissions tend to
be higher in rice fields if the residues have been in the field for less than 30 days before planting the next rice crop
(Lindau and Bollich 1993; IPCC 2006; Wang et al. 2013).
A combination of Tier 1 and 3 methods are used to estimate CH4 emissions from rice cultivation across most of the
time series, while a surrogate data method has been applied to estimate national emissions for 2021 to 2022 in
this Inventory due to lack of data in these years of the time series. National emission estimates based on surrogate
data will be recalculated in a future Inventory with the Tier 1 and 3 methods as data becomes available.
Overall, rice cultivation is a minor source of CH4 emissions in the United States relative to other source categories
(see Table 5-10, Table 5-11, and Figure 5-3). Most emissions occur in Arkansas, California, Louisiana, Mississippi,
Missouri, and Texas. In 2022, CFU emissions from rice cultivation were 18.9 MMT CO2 Eq. (674 kt CH4). Annual
emissions fluctuated between 1990 and 2022, which is largely due to differences in the amount of rice harvested
areas over time. There has been a marginal decrease in emissions since 1990. Interestingly, the estimated
emissions in 2022 are roughly the same as emissions in 1990.
Table 5-10: CH4 Emissions from Rice Cultivation (MMT CO2 Eq.)
State
1990
2005
2018
2019
2020
2021
2022
Arkansas
6.3
8.8
8.0
5.6
6.8
NE
NE
California
3-2 1
3-4
3.7
3.5
3.6
NE
NE
Florida
+
+
+
+
+
NE
NE
Illinois
+
+
+
+
NE
NE
Kentucky
+
+
+
+
+
NE
NE
Louisiana
3.5 =
ill!!!
CO
CO
3.7
3.2
3.7
NE
NE
Minnesota
+
+
+
+
+
NE
NE
Mississippi
i-11
CO
1
0.7
0.6
0.6
NE
NE
Missouri
0.5
1.2
1.2
0.9
1.0
NE
NE
New York
+
ii;
+
+
+
+
NE
NE
Tennessee
+
+
+
+
+
NE
NE
Texas
4.3 I
cn
T—1
2.6
1.9
2.9
NE
NE
Total 18.9 20.6 19.9 15.6 18.6 18.3 18.9
+ Does not exceed 0.05 MMT C02 Eq.
NE (Not Estimated). State-level emissions are not estimated for 2021 through 2022 in this Inventory.
A surrogate method is used to estimate emissions for these years only at the national scale.
Note: Totals may not sum due to independent rounding.
5-22 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
Table 5-11: CH4 Emissions from Rice Cultivation (kt CH4)
State
1990
2005
2018
2019
2020
2021
2022
Arkansas
224.2
315.5
287.1
200.5
243.4
NE
NE
California
114.5
122.8
131.5
123.9
129.1
NE
NE
Florida
+
1.2
+
+
+
NE
NE
Illinois
+
0.4
0.1
+
0.1
NE
NE
Kentucky
+
+
+
+
+
NE
NE
Louisiana
124.9
135.0
131.4
113.7
130.4
NE
NE
Minnesota
1.0
1.7
+
0.8
+
NE
NE
Mississippi
39.0
46.5
24.3
20.3
21.5
NE
NE
Missouri
19.5
42.7
43.0
31.8
34.4
NE
NE
New York
0.2
+
+
+
+
NE
NE
Tennessee
+
0.1
+
+
+
NE
NE
Texas
153.5
69.5
93.6
67.0
104.7
NE
NE
Total
677
735
711
558
664
653
674
+ Does not exceed 0,5 kt.
NE (Not Estimated). State-level emissions are not estimated for 2021 through 2022 in this Inventory.
A surrogate method is used to estimate emissions for these years only at the national scale.
Note: Totals may not sum due to independent rounding.
Figure 5-3: Annual CH4 Emissions from Rice Cultivation, 2020, Using the Tier 3 DayCent Model
Note: Only national-scale emissions are estimated for 2021 and 2022 in this Inventory using a surrogate data method described
in the Methodology section; therefore, the fine-scale emission patterns in this map are based on the estimates for 2020.
Agriculture 5-23
-------
Methodology and Time-Series Consistency
The methodology used to estimate Cm emissions from rice cultivation is based on a combination of IPCC Tier 1 and
3 approaches. The Tier 3 method utilizes the DayCent process-based model to estimate Cm emissions from rice
cultivation (Cheng et al. 2013) and has been tested in the United States (see Annex 3.12) and Asia (Cheng et al.
2013, 2014). The model simulates hydrological conditions and thermal regimes, organic matter decomposition,
root exudation, rice plant growth and its influence on oxidation of Cm, as well as CFU transport through the plant
and via ebullition (Cheng et al. 2013). The method captures the influence of organic amendments and rice straw
management on methanogenesis in the flooded soils, and ratooning of rice crops with a second harvest during the
growing season. In addition to CFU emissions, DayCent simulates soil carbon stock changes and N2O emissions
(Parton et al. 1987 and 1998; Del Grosso et al. 2010) and allows for a seamless set of simulations for crop rotations
that include both rice and non-rice crops.
The Tier 1 method is applied to estimate CFU emissions from rice when grown in rotation with crops that are not
simulated by DayCent, such as vegetable crops. The Tier 1 method is also used for areas converted between
agriculture (i.e., cropland and grassland) and other land uses, such as forest land, wetland, and settlements. In
addition, the Tier 1 method is used to estimate CH4 emissions from organic soils (i.e., Histosols) and from areas
with very gravelly, cobbly, or shaley soils (greater than 35 percent by volume). The Tier 3 method using DayCent
has not been fully tested for estimating emissions associated with these conditions.
The Tier 1 method for estimating CH4 emissions from rice production utilizes a default base emission rate and
scaling factors (IPCC 2006). The base emission rate represents emissions for continuously flooded fields with no
organic amendments. Scaling factors are used to adjust the base emission rate for water management and organic
amendments that differ from continuous flooding with no organic amendments. The method accounts for pre-
season and growing season flooding; types and amounts of organic amendments; and the number of rice
production seasons within a single year (i.e., single cropping and double-cropping with ratooning). The Tier 1
analysis is implemented in the Agriculture and Land Use National Greenhouse Gas Inventory (ALU) software (Ogle
et al. 2016).11
Rice cultivation areas are based on crop and land use histories recorded in the USDA National Resources Inventory
(NRI) survey (USDA-NRCS 2020) and extended through 2020 using the USDA-NASS Crop Data Layer Product (USDA-
NASS 2021, Johnson and Mueller 2010). The areas have been modified in the original NRI survey through a process
in which the Forest Inventory and Analysis (FIA) survey data and the National Land Cover Dataset (Yang et al. 2018)
are harmonized with the NRI data (Nelson et al. 2020). This process ensures that the land use areas are consistent
across all land use categories (See Section 6.1, Representation of the U.S. Land Base for more information).
The NRI is a statistically based sample of all non-federal land and includes approximately 604,000 survey locations
in agricultural cropland and grassland for the conterminous United States and Hawaii of which 7,888 include one
or more years of rice cultivation. The Tier 3 method is used to estimate CH4 emissions from 5,998 of the NRI survey
locations, and the remaining 1,890 survey locations are estimated with the Tier 1 method. Each NRI survey location
is associated with a survey weight that allows scaling of CH4 emission to the entire land base with rice cultivation
(i.e., each weight approximates the amount of area with the same land-use/management history as the survey
location). Land-use and some management information in the NRI (e.g., crop type, soil attributes, and irrigation)
were collected on a 5-year cycle beginning in 1982, along with cropping rotation data in four out of five years for
each five-year time period (i.e., 1979 to 1982,1984 to 1987,1989 to 1992, and 1994 to 1997). The NRI program
began collecting annual data in 1998, with data through 2017 (USDA-NRCS 2020). For 2018-2020, the time series is
extended with the crop data provided in USDA-NASS CDL (USDA-NASS 2021). CDL data have a 30 to 58 m spatial
resolution, depending on the year. NRI survey locations are overlaid on the CDL in a geographic information
11 See http ://www. n rel.colostate.edu/projects/ALU softwa re/.
5-24 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
system, and the crop types are extracted to extend the cropping histories. The harvested rice areas in each state
are presented in Table 5-12.
Table 5-12: Rice Area Harvested (1,000 Hectares)
State/Crop
1990
2005
2018
2019
2020
2021
2022
Arkansas
611
782
659
512
663
NE
NE
California
251
237 i
226
218
224
NE
NE
Florida
0
3
0
0
0
NE
NE
Illinois
0
i!
0
0
1
NE
NE
Kentucky
0
0
0
0
0
NE
NE
Louisiana
399
400:
356
313
383
NE
NE
Minnesota
3
6
0
3
0
NE
NE
Mississippi
177
191 ¦
98
96
109
NE
NE
Missouri
48
96
99
74
85
NE
NE
New York
1
0
0
0
0
NE
NE
Tennessee
0
1
0
0
0
NE
NE
Texas
294
104
164
119
167
NE
NE
Total
1,784
1,823
1,603
1,335
1,633
NE
NE
NE (Not Estimated). Area data will be updated in the next Inventory.
Note: Totals may not sum due to independent rounding.
The Southeastern states have sufficient growing periods for a ratoon crop in some years (Table 5-13). For example,
the growing season length is occasionally sufficient for ratoon crops to be grown on about two percent of the rice
fields in Arkansas. No data are available about ratoon crops in Missouri or Mississippi, so the average amount of
ratooning in Arkansas was assigned to these states. Ratoon cropping occurs much more frequently in Louisiana
(LSU 2015 for years 2000 through 2013, 2015) and Texas (TAMU 2015 for years 1993 through 2015), averaging 32
percent and 45 percent of rice acres planted, respectively. Florida also has a large fraction of area with a ratoon
crop (49 percent). Ratoon rice crops are not grown in California. Ratooning practices are assigned to individual NRI
locations using a hot-deck imputation method with six complete imputations for each NRI location to address
uncertainty. The method is based on random assignment of ratooning to approximate the percentages of fields
managed with ratooning provided in Table 5-14.
Table 5-13: Average Ratooned Area as Percent of Primary Growth Area (Percent)
State
1990-2015
Arkansas3
1.9%
California
0%
Florida15
45.2%
Louisiana0
39.5%
Mississippi3
37.8%
Missouri3
2.4%
Texasd
49.5%
3 Arkansas: 1990-2000 (Slaton 1999 through 2001); 2001-2011 (Wilson 2002 through 2007, 2009 through 2012); 2012-2013
(Hardke 2013, 2014). Estimates of ratooning for Missouri and Mississippi are based on the data from Arkansas.
b Florida - Ratoon: 1990-2000 (Schueneman 1997,1999 through 2001); 2001 (Deren 2002); 2002-2003 (Kirstein 2003 through
2004, 2006); 2004 (Cantens 2004 through 2005); 2005-2013 (Gonzalez 2007 through 2014).
c Louisiana: 1990-2013 (Linscombe 1999, 2001 through 2014).
d 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 through 2014).
While rice crop production in the United States includes a minor amount of land with mid-season drainage or
alternate wet-dry periods, the majority of rice growers use continuously flooded water management systems
(Hardke 2015; UCCE 2015; Hollier 1999; Way et al. 2014). Therefore, continuous flooding was assumed in the
DayCent simulations and the Tier 1 analysis. Variation in flooding can be incorporated in future inventories if
updated water management data are available.
Agriculture 5-25
-------
Winter flooding is another key practice associated with water management in rice fields, and the impact of winter
flooding on CFU emissions is addressed in the Tier 3 and Tier 1 analyses. Flooding is used to prepare fields for the
next growing season, and to create waterfowl habitat (Young 2013; Miller et al. 2010; Fleskes et al. 2005).
Fitzgerald et al. (2000) suggests that as much as 50 percent of the annual emissions may occur during winter
flooding. Winter flooding is a common practice with an average of 34 percent of fields managed with winter
flooding in California (Miller et al. 2010; Fleskes et al. 2005), and approximately 21 percent of the fields managed
with winter flooding in Arkansas (Wilson and Branson 2005 and 2006; Wilson and Runsick 2007 and 2008; Wilson
et al. 2009 and 2010; Hardke and Wilson 2013 and 2014; Hardke 2015). No data are available on winter flooding
for Texas, Louisiana, Florida, Missouri, or Mississippi. For these states, the average amount of flooding is assumed
to be similar to Arkansas. In addition, the amount of flooding is assumed to be relatively constant over the
Inventory time series. Similar to ratooning practices, winter flooding is assigned to individual NRI locations using a
hot-deck imputation method with six complete imputations for each NRI location to address uncertainty. The
method is based on random assignment of winter flooding to approximate the percentages of fields managed with
winter flooding as discussed above.
A data splicing method is used to estimate emissions from 2021 to 2022 associated with the rice Cm emissions for
Tier 1 and 3 methods. Specifically, a linear regression model with autoregressive moving average (ARMA) errors
was used to estimate the relationship between the surrogate data and emissions data from 1990 through 2020,
which were derived using the Tier 3 methods (Brockwell and Davis 2016). Surrogate data are based on rice
commodity statistics from USDA-NASS.12 See Box 5-2 for more information about the surrogate data method. For
the Tier 1 method, a linear-time series model is used to estimate emissions for 2021 to 2022 without surrogate
data.
Box 5-2: Surrogate Data Method
An approach to extend the time series is needed to estimate emissions from rice cultivation because there are
gaps in activity data at the end of the time series. This is mainly because the National Resources Inventory (NRI)
does not release data every year, and the NRI is a key data source for estimating greenhouse gas emissions.
A surrogate data method has been selected to impute missing emissions at the end of the time series. A linear
regression model with autoregressive moving average (ARMA) errors (Brockwell and Davis 2016) is used to
estimate the relationship between the surrogate data and the observed 1990 to 2020 emissions data that has
been compiled using the inventory methods described in this section. The model to extend the time series is
given by
Y = Xp + e,
where Y is the response variable (e.g., Cm emissions), xp is the surrogate data that is used to predict the missing
emissions data, and £ is the remaining unexplained error. Models with a variety of surrogate data were tested,
including commodity statistics, weather data, or other relevant information. Parameters are estimated from the
observed data for 1990 to 2020 using standard statistical techniques, and these estimates are used to predict
the missing emissions data for 2021 to 2022.
A critical issue in using splicing methods is to adequately account for the additional uncertainty introduced by
predicting emissions with related information without compiling the full inventory. For example, predicting Cm
emissions will increase the total variation in the emission estimates for these specific years, compared to those
years in which the full inventory is compiled. This added uncertainty is quantified within the model framework
using a Monte Carlo approach. The approach requires estimating parameters for results in each Monte Carlo
simulation for the full inventory (i.e., the surrogate data model is refit with the emissions estimated in each
Monte Carlo iteration from the full inventory analysis with data from 1990 to 2020).
12 See https://quickstats.nass.usda.eov/.
5-26 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
In order to ensure time-series consistency, the same methods are applied from 1990 to 2020, and data splicing
methods are used to approximate emissions for the remainder of the 2021 to 2022 time series based on the
emissions data from 1990 to 2020. The surrogate data method and linear time series approach, used for the Tier 3
and 1 methods, respectively, are consistent with data splicing methods in IPCC (2006).
Uncertainty
Sources of uncertainty in the Tier 3 method include management practices, uncertainties in model structure (i.e.,
algorithms and parameterization), and variance associated with the NRI sample. Sources of uncertainty in the IPCC
(2006) Tier 1 method include the emission factors, management practices, and variance associated with the NRI
sample. The total uncertainty was quantified with two variance components (Ogle et al. 2010) that are combined
using 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. For the first variance component, a Monte Carlo
analysis was used to propagate uncertainties in the Tier 1 and 3 methods for the management data, as well as
emission factors and model structure/parameterization, respectively. The second variance component is
quantifying uncertainty in scaling from the NRI survey to the entire area of rice cultivation, and is computed using a
standard variance estimator for a two-stage sample design (Sarndal et al. 1992). For 2021 to 2022, there is
additional uncertainty propagated through the Monte Carlo analysis associated with the surrogate data method
(See Box 5-2 for information about propagating uncertainty with the surrogate data method). The uncertainties
from the Tier 1 and 3 approaches are combined to produce the final Cm emissions estimate using simple error
propagation (IPCC 2006). Additional details on the uncertainty methods are provided in Annex 3.12.
Rice cultivation CFU emissions in 2022 were estimated to be between 5.1 and 32.6 MMT CO2 Eq. at a 95 percent
confidence level, which indicates a range of 73 percent below to 73 percent above the 2022 emission estimate of
18.9 MMT CO2 Eq. (see Table 5-14).
Table 5-14: Approach 2 Quantitative Uncertainty Estimates for CH4 Emissions from Rice
Cultivation (MMT CO2 Eq. and Percent)
Source
Inventory
Method
Gas
2022 Emission
Estimate
Uncertainty Range Relative to Emission Estimate-'
(MMTCO. Eq.)
(MMT CO.
Eq.)
(%)
Lower
Upper
Lower
Upper
Bound
Bound
Bound
Bound
Rice Cultivation
Tier 3
ch4
15.9
? ?
29.6
-86%
+86%
Rice Cultivation
Tier 1
ch4
3.0
2.0
4.0
-34%
+34%
Rice Cultivation
Total
CH,
18.9
5.1
32.6
-73%
+73%
a Range of emission estimates is the 95 percent confidence interval.
QA/QC and Verification
General (Tier 1) and category-specific (Tier 2) QA/QC activities were conducted consistent with the U.S. Inventory
QA/QC plan outlined in Annex 8. Quality control measures include checking input data, model scripts, and results
to ensure data are properly handled throughout the inventory process. Inventory reporting forms and text are
reviewed and revised as needed to correct transcription errors.
Model results are compared to field measurements to verify that results adequately represent CH4 emissions. The
comparisons included over 17 long-term experiments, representing about 238 combinations of management
treatments across all the sites. A statistical relationship was developed to assess uncertainties in the model
structure and parameterization, adjusting the estimates for model bias and assessing precision in the resulting
estimates (methods are described in Ogle et al. 2007). See Annex 3.12 for more information.
Agriculture 5-27
-------
Recalculations Discussion
Several improvements have been implemented in this Inventory leading to recalculations, including a) updated
time series of land representation data that identifies which points and years were sown with rice (Nelson et al
2020), b) extending the time-series of crop history with CDL data, c) imputing ratooning and winter flooding onto
individual NRI survey points, d) updated fertilizer and organic amendment additions, and e) revisions to the
approach for assigning organic matter amendments and crop residue inputs. As a result of these changes, CO2-
equivalent emissions changed annually with an average annual increase of 0.97 MMT CO2 Eq., or 5.5 percent, over
the time series from 1990 to 2021 compared to the previous Inventory.
Planned Improvements
A key planned improvement for rice cultivation is to refine the model algorithms and re-calibration of the Tier 3
DayCent model using the latest observational data from experiments. Another improvement is collection of more
information about water management and refinement of the application to incorporate mid-season drainage and
alternate wetting and drying systems. Improvements are expected to be completed for the next Inventory (i.e.,
2025 submission to the UNFCCC, 1990 through 2023 Inventory), pending prioritization of resources.
5.4 Agricultural Soil Management (CRT
Source Category 3D)
Nitrous oxide is naturally produced in soils through the microbial processes of nitrification and denitrification that
is driven by the availability of mineral nitrogen (N) (Firestone and Davidson 1989).13 Mineral nitrogen is made
available in soils through decomposition of soil organic matter and plant litter, as well as asymbiotic fixation of
nitrogen from the atmosphere.14 Several agricultural activities increase mineral nitrogen availability in soils that
lead to direct N2O emissions at the site of a management activity (see Figure 5-4) (Mosier et al. 1998). These
activities include synthetic nitrogen fertilization; application of managed livestock manure; application of other
organic materials such as biosolids (i.e., treated sewage sludge); deposition of manure on soils by domesticated
animals in pastures, range, and paddocks (PRP) (i.e., unmanaged manure); retention of crop residues (nitrogen-
fixing legumes and non-legume crops and forages); and drainage of organic soils15 (i.e., Histosols) (IPCC 2006).
Additionally, agricultural soil management activities, including irrigation, drainage, tillage practices, cover crops,
and fallowing of land, can influence nitrogen mineralization from soil organic matter and levels of asymbiotic
nitrogen fixation. Indirect emissions of N2O occur when nitrogen is transported from a site and is subsequently
converted to N2O; there are two pathways for indirect emissions: (1) volatilization and subsequent atmospheric
deposition of applied/mineralized nitrogen, and (2) surface runoff and leaching of applied/mineralized nitrogen
13 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 nitrification and denitrification.
14 Asymbiotic nitrogen fixation is the fixation of atmospheric N2 by bacteria living in soils that do not have a direct relationship
with plants.
15 Drainage of organic soils in former wetlands enhances mineralization of nitrogen-rich organic matter, thereby increasing N20
emissions from these soils.
5-28 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
into groundwater and surface water.16 Direct and indirect emissions from agricultural lands are included in this
section (i.e., cropland and grassland as defined in Section 6.1). Nitrous oxide emissions from forest land and
settlements soils are found in Sections 6.2 and 6.10, respectively.
Figure 5-4: Sources and Pathways of Nitrogen that Result in N2O Emissions from Agricultural
Soil Management
Sources and Pathways ol N that Result in N;0 Emissions trom Agricultural Soil Management
N Rows:
0
N Inputs to
Managed Soils
Direct N2O
Emissions
N Volatilization
and Deposition
Indirect N20
Emissions
Hlstosol
Cultivation
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.
16 These processes entail volatilization of applied or mineralized nitrogen as NH3 and NOx, transformation of these gases in the
atmosphere (or upon deposition), and deposition of the nitrogen primarily in the form of particulate NH4+, nitric acid (HN03),
and NOx. In addition, hydroiogical processes lead to leaching and runoff of N03" that is converted to N20 in aquatic systems,
e.g., wetlands, rivers, streams and lakes. Note: N20 emissions are not estimated for aquatic systems associated with nitrogen
inputs from terrestrial systems in order to avoid double-counting.
Agriculture 5-29
-------
Agricultural soils produce the majority of N2O emissions in the United States. Estimated emissions in 2022 are
290.8 MMT CO2 Eq. (1,097 kt) (see Table 5-15 and Table 5-16). Annual N2O emissions from agricultural soils are 3.2
percent greater in 2022 compared to 1990, but emissions fluctuated between 1990 and 2022 due to inter-annual
variability largely associated with weather patterns, synthetic fertilizer use, and crop production. From 1990 to
2022, cropland accounted for 68 percent of total direct emissions on average from agricultural soil management,
while grassland accounted for 32 percent. On average, 79 percent of indirect emissions are from croplands and 21
percent from grasslands. Estimated direct and indirect N2O emissions by sub-source category are shown in Table
5-17 and Table 5-18.
Table 5-15: N2O Emissions from Agricultural Soils (MMT CO2 Eq.)
Activity
1990
2005
2018
2019
2020
2021
2022
Direct
258.8
265.6
298.3
280.9
262.8
267.7
262.5
Cropland
174.9
180.6
208.9
193.4
182.4
184.3
180.3
Grassland
83.9
85.1
89.4
87.5
80.3
83.4
82.1
Indirect
29.9
28.4
35.1
34.7
29.4
30.3
28.3
Cropland
23.6
22.3
28.1
28.0
23.3
24.1
22.2
Grassland
6.4
6.1 1=
7.0
6.8
6.1
6.2
6.1
Total
288.8
294.1
333.4
315.6
292.1
298.0
290.8
Notes: Estimates for 2021 to 2022 are based on a data splicing method, except for other organic nitrogen amendments that
are based on a data splicing method for 2018 to 2022 (see Methodology section). Totals may not sum due to independent
rounding.
Table 5-16: N2O Emissions from Agricultural Soils (kt N2O)
Activity
1990
2005
2018
2019
2020
2021
2022
Direct
977
1,002
1,126
1,060
992
1,010
990
Cropland
660.0 1
681.4
788.3
729.9
688.5
695.4
680.6
Grassland
316.7
321.1
337.4
330.0
303.1
314.6
309.9
Indirect
113
107
133
131
111
114
107
Cropland
89.0 I
84.2
106.2
105.6
88.0
91.1
83.9
Grassland
24.0 i
23.1 |
26.4
25.5
22.9
23.3
22.9
Total
1,090
1,110
1,258
1,191
1,102
1,124
1,097
Notes: Estimates for 2021 to 2022 are based on a data splicing method, except for other organic nitrogen amendments that
are based on a data splicing method for 2018 to 2022 (see Methodology section). Totals may not sum due to independent
rounding.
Table 5-17: Direct N2O Emissions from Agricultural Soils by Land Use Type and Nitrogen Input
Type (MMT C02 Eq.)
Activity
1990
2005
2018
2019
2020
2021
2022
Cropland
174.9
180.6
208.9
193.4
182.4
184.3
180.3
Mineral Soils
171.5
177.3
205.9
190.5
179.5
181.4
177.4
Synthetic Fertilizer
61.0
64.3
70.3
65.7
63.2
63.4
62.0
Organic Amendment3
n-5
12.7;;
14.7
14.6
14.4
14.7
14.6
Residue Nb
34.1
35.0
39.6
34.5
37.6
33.2
32.4
Mineralization and
!!!!!!!
!!!!!!!
I
Asymbiotic Fixation
64.8 |
65.3
81.3
75.7
64.3
70.1
68.4
Drained Organic Soils
3.4
3.2
3.0
2.9
2.9
2.9
2.9
Grassland
83.9
85.1
89.4
87.5
80.3
83.4
82.1
Mineral Soils
81.6
82.8
87.2
85.2
78.1
81.1
79.8
Synthetic Fertilizer
+
+
+
+
+
+
+
PRP Manure
15.4 1
14.2:
14.0
13.6
13.3
13.9
13.8
Managed Manurec
+
+
+
+
+
+
+
5-30 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
Biosolids (i.e., treated
Sewage Sludge) 0.2 | 0.4 I 0.4 0.4 0.4 0.4 0.4
Residue Nd 27.1 I 28.4 28.0 28.3 28.2 26.3 25.9
Mineralization and
Asymbiotic Fixation 38.9 I 39.8 II 44.8 42.9 36.2 40.5 39.8
Drained Organic Soils 2.3 2.2 22 Z2 22 23 23
Total
258.8
265.6
298.3
280.9
262.8
267.7
262.5
+ Does not exceed 0.05 MMT C02 Eq.
a Organic amendment inputs include managed manure, daily spread manure, and commercial organic fertilizers (i.e., dried
blood, dried manure, tankage, compost, and other).
bCropland residue nitrogen inputs include nitrogen in unharvested cover crops as well as harvested crops.
c Managed manure inputs include managed manure and daily spread manure amendments that are applied to grassland
soils.
d Grassland residue nitrogen inputs include residual biomass, both legumes and grasses, that is ungrazed and becomes dead
organic matter.
Notes: Estimates for 2021 to 2022 are based on a data splicing method, except for other organic nitrogen amendments that
are based on a data splicing method for 2018 to 2022 (see Methodology section). Totals may not sum due to independent
rounding.
Table 5-18: Indirect N2O Emissions from Agricultural Soils (MMT CO2 Eq.)
Activity
1990
2005
2018
2019
2020
2021
2022
Cropland
23.6
22.3
28.1
28.0
23.3
24.1
22.2
Volatilization & Atm.
Deposition
6.6
7.0 1
7.9
7.1
7.5
7.4
7.3
Surface Leaching & Run-Off
17.0
15.3
20.3
20.9
15.8
16.7
14.9
Grassland
6.4
6.1
7.0
6.8
6.1
6.2
6.1
Volatilization & Atm.
I
Deposition
3.4
3.4 I
3.3
3.2
3.0
3.1
3.2
Surface Leaching & Run-Off
2.9
2.7
3.7
3.6
3.1
3.0
2.9
Total
29.9
28.4
35.1
34.7
29.4
30.3
28.3
Notes: Estimates for 2021 to 2022 are based on a data splicing method, except for other organic nitrogen amendments that
are based on a data splicing method for 2018 to 2022 (see Methodology section). Totals may not sum due to independent
rounding.
Figure 5-5 and Figure 5-6 show regional patterns for direct N2O emissions. Figure 5-7 and Figure 5-8 show indirect
N2O emissions from volatilization, and Figure 5-9 and Figure 5-10 show the indirect N2O emissions from leaching
and runoff in croplands and grasslands, respectively.
Direct N2O emissions from croplands occur throughout all of the cropland regions but tend to be high in the
Midwestern Corn Belt Region (particularly, Illinois, Iowa, Kansas, Minnesota, Nebraska), where a large portion of
the land is used for growing highly fertilized corn and nitrogen-fixing soybean crops (see Figure 5-5). There are high
emissions from the Southeastern region, and portions of the Great Plains. Emissions are also high in the Lower
Mississippi River Basin from Missouri to Louisiana, and highly productive irrigated areas, such as Platte River, which
flows from Colorado and Wyoming through Nebraska, Snake River Valley in Idaho, and the Central Valley in
California. Direct emissions from croplands are low in mountainous regions of the Eastern United States because
only a small portion of land is cultivated, and in much of the Western United States where rainfall and access to
irrigation water are limited, in addition to mountainous, which are generally not suitable for crop production.
Direct N2O emissions from grasslands are more evenly distributed throughout the United States compared to
emissions from cropland due to suitable areas for grazing in most regions (see Figure 5-6). Total emissions tend be
highest in the Great Plains and western United States where a large proportion of the land is dominated by
grasslands with cattle and sheep grazing (particularly Kansas, Montana, Nebraska, New Mexico, Oklahoma, South
Dakota, Texas).
Agriculture 5-31
-------
Figure 5-5: Croplands, 2020 Annual Direct N2O Emissions Estimated Using the Tier 3 DayCent
Model
Note: Only riational-scaie emissions are estimated for 2022 using a splicing method, and therefore the fine-scale
emission patterns in this map are based on Inventory data from 2020.
Figure 5-6: Grasslands, 2020 Annual Direct N?.0 Emissions Estimated Using the Tier 3 DayCent
Model
Note: Only national-scale emissions are estimated for 2022 using a splicing method, and therefore the fine-scale
emission patterns in this map are based on Inventory data from 2020.
5-32 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
Indirect N2O emissions from volatilization in croplands have a similar pattern as the direct N?.0 emissions with
higher emissions in the Midwestern Corn Belt, Lower Mississippi River Basin, Southeastern region, and parts of the
Great Plains and irrigated areas of the Western United States. Indirect N2O emissions from volatilization in
grasslands are higher in the Eastern and Central United States, along with relatively small areas scattered around
the Western United States. The higher emissions are partly due to large additions of PRP manure nitrogen, which
in turn, stimulates NH3 volatilization.
Indirect N2O emissions from surface runoff and leaching of applied/mineralized nitrogen in croplands is highest in
the Midwestern Corn Belt. There are also relatively high emissions associated with nitrogen management in the
Lower Mississippi River Basin, Piedmont region of the Southeastern United States and the Mid-Atlantic states. In
addition, areas of high emissions occur in portions of the Great Plains that have irrigated croplands with high
leaching rates of applied/mineralized nitrogen. Indirect N2O emissions from surface runoff and leaching of
applied/mineralized nitrogen in grasslands are higher in the eastern United States and coastal Northwest region.
These regions have greater precipitation and higher levels of leaching and runoff compared to arid to semi-arid
regions in the Western United States.
Figure 5-7: Croplands, 2020 Annual Indirect N2O Emissions from Volatilization Using the Tier
3 DayCent Model
Note: Only national-scale emissions are estimated for 2022 using a splicing method, and therefore the fine-scale
emission patterns in this map are based on Inventory data from 2020.
Agriculture 5-33
-------
Figure 5-8: Grasslands, 2020 Annual Indirect NzO Emissions from Volatilization Using the Tier
3 DayCent Model
Note: Only national-scale emissions are estimated for 2022 using a splicing method, and therefore the fine-scale
emission patterns in this map are based on Inventory data from 2020.
Figure 5-9: Croplands, 2020 Annual Indirect N2O Emissions from Leaching and Runoff Using
the Tier 3 DayCent Model
Note: Only national-scale emissions are estimated for 2022 using a splicing method, and therefore the fine-scale
emission patterns in this map are based on Inventory data from 2020.
5-34 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
Figure 5-10: Grasslands, 2020 Annual Indirect N2O Emissions from Leaching and Runoff Using
the Tier 3 DayCent Model
Note: Only national-scale emissions are estimated for 2022 using a splicing method, and therefore the fine-scale
emission patterns in this map are based on Inventory data from 2020.
Methodology and Time-Series Consistency
The 2006IPCC Guidelines (IPCC 2006) divide emissions from the agricultural soil management source category into
five components, including (1) direct emissions from nitrogen additions to cropland and grassland mineral soils
from synthetic fertilizers, biosolids (i.e., treated sewage sludge), crop residues (legume nitrogen-fixing and non-
legume crops), and organic amendments; (2) direct emissions from soil organic matter mineralization due to land
use and management change; (3) direct emissions from drainage of organic soils in croplands and grasslands; (4)
direct emissions from soils due to manure deposited by livestock on PRP grasslands; and (5) indirect emissions
from soils and water from nitrogen additions and manure deposition to soils that lead to volatilization, leaching, or
runoff of nitrogen and subsequent conversion to N2O.
In this source category, the United States reports on all croplands, as well as all managed grasslands, whereby
anthropogenic greenhouse gas emissions are estimated in a manner consistent with the managed land concept
(IPCC 2006), including direct and indirect N2O emissions from asymbiotic fixation17 and mineralization of nitrogen
associated with decomposition of soil organic matter and residues. One recommendation from IPCC (2006) that
has not been completely adopted is the estimation of emissions from grassland pasture renewal, which involves
occasional plowing to improve forage production in pastures. Currently no data are available to address pasture
renewal.
In addition, estimates of N2O emissions from managed croplands and grasslands are not available for Alaska and
Hawaii except for managed manure and PRP nitrogen, and biosolid additions for Alaska, and managed manure and
17 Nitrogen inputs from asymbiotic nitrogen fixation are not directly addressed in 2006 IPCC Guidelines but are a component of
the nitrogen inputs and total emissions from managed lands and are included in the Tier 3 approach developed for this source.
Agriculture 5-35
-------
PRP nitrogen, biosolids additions, and crop residue for Hawaii. There is a planned improvement to include the
additional sources of emissions in a future Inventory.
Direct N2O Emissions
The methodology used to estimate direct N2O emissions from agricultural soil management in the United States is
based on a combination of IPCC Tier 1 and 3 approaches, along with application of a splicing method for latter
years in the Inventory time series (IPCC 2006; Del Grosso et al. 2010). A Tier 3 process-based model (DayCent) is
used to estimate direct emissions from a variety of crops that are grown on mineral (i.e., non-organic) soils, as well
as the direct emissions from non-federal grasslands except for applications of biosolids (i.e., treated sewage
sludge) (Del Grosso et al. 2010). The Tier 3 approach has been specifically designed and tested to estimate N2O
emissions in the United States, accounting for more of the environmental and management influences on soil N2O
emissions than the IPCC Tier 1 method (see Box 5-3 for further elaboration). Moreover, the Tier 3 approach
addresses direct N2O emissions and soil carbon stock changes from mineral cropland soils in a single analysis.
Carbon and nitrogen dynamics are linked in plant-soil systems through biogeochemical processes of microbial
decomposition and plant production (McGill and Cole 1981). Coupling the two source categories (i.e., agricultural
soil carbon and N2O) in a single inventory analysis ensures that there is consistent activity data and treatment of
the processes, and interactions are considered between carbon and nitrogen cycling in soils.
Crop and land use histories are based on the USDA National Resources Inventory (NRI) (USDA-NRCS 2020) and
extended through 2020 using the USDA-NASS Crop Data Layer Product (USDA-NASS 2021; Johnson and Mueller
2010). The areas have been modified in the original NRI survey through a process in which the Forest Inventory
and Analysis (FIA) survey data and the National Land Cover Dataset (Yang et al. 2018) are harmonized with the NRI
data (Nelson et al. 2020). This process ensures that the land use areas are consistent across all land use categories
(see Section 6.1).
The NRI is a statistically-based sample and includes 364,333 survey locations on agricultural land for the
conterminous United States that are included in the Tier 3 method. The Tier 1 approach is used to estimate the
emissions from an annual average of 239,757 locations in the NRI survey across the time series, which are
designated as cropland or grassland (discussed later in this section). The Tier 1 method is used to estimate
emissions for components that are not simulated by DayCent. DayCent has not been parametrized to simulate
some crop types and soil types, as described below. Each survey location is associated with a survey weight that
allows scaling of N2O emissions from NRI survey locations to the entire country (i.e., each survey weight is an
approximation of the amount of area with the same land-use/management history as the survey location). Each
NRI survey location was sampled on a 5-year cycle from 1982 until 1997. For cropland, data were collected in 4 out
of 5 years in the cycle (i.e., 1979 through 1982,1984 through 1987,1989 through 1992, and 1994 through 1997).
In 1998, the NRI program began collecting annual data, which are currently available through 2017 (USDA-NRCS
2020). For 2018-2020, the time series is extended with the crop data provided in USDA-NASS CDL (USDA-NASS
2021). CDL data have a 30 to 58 m spatial resolution, depending on the year. Specifically, NRI survey locations are
overlaid on the CDL in a geographic information system, and the crop types are extracted to extend the cropping
histories for the inventory analysis.
Box 5-3: Tier 1 vs. Tier 3 Approach for Estimating N2O Emissions
The IPCC (2006) Tier 1 approach is based on multiplying activity data on different nitrogen inputs (i.e., synthetic
fertilizer, manure, nitrogen fixation, etc.) by the appropriate default IPCC emission factors to estimate N2O
emissions on an input-by-input basis. The Tier 1 approach requires a minimal amount of activity data, readily
available in most countries (e.g., total nitrogen applied to crops); calculations are simple; and the methodology
is highly transparent. In contrast, the Tier 3 approach developed for this Inventory is based on application of a
process-based model (i.e., DayCent) that represents the interaction of nitrogen inputs, land use and
management, as well as environmental conditions at specific locations, such as freeze-thaw effects that
generate pulses of N2O emissions (Wagner-Riddle et al. 2017; Del Grosso et al. 2022). Consequently, the Tier 3
approach accounts for land-use and management impacts and their interaction with environmental factors,
5-36 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
such as weather patterns and soil characteristics, in a more comprehensive manner, which will enhance or
dampen anthropogenic influences. However, the Tier 3 approach requires more detailed activity data (e.g.,
crop-specific nitrogen fertilization rates), additional data inputs (e.g., daily weather, soil types), and
considerable computational resources and programming expertise. The Tier 3 methodology is less transparent,
and thus it is critical to evaluate the output of Tier 3 methods against measured data in order to demonstrate
that the method is an improvement over lower tier methods for estimating emissions (IPCC 2006). Another
important difference between the Tier 1 and Tier 3 approaches relates to assumptions regarding nitrogen
cycling. Tier 1 assumes that nitrogen added to a system is subject to N O emissions only during that year and
cannot be stored in soils and contribute to N O emissions in subsequent years. This is a simplifying assumption
that may create bias in estimated N O emissions for a specific year. In contrast, the process-based model in the
Tier 3 approach includes the legacy effect of nitrogen added to soils in previous years that is re-mineralized
from soil organic matter and emitted as N O during subsequent years.
DayCent is used to estimate N2O emissions associated with production of alfalfa hay, barley, corn, cotton, dry
beans, grass hay, grass-clover hay, lentils, oats, onions, peanuts, peas, potatoes, rice, sorghum, soybeans, sugar
beets, sunflowers, sweet potatoes, tobacco, tomatoes, and wheat, but is not applied to estimate N2O emissions
from other crops or rotations with other crops,18 such as sugarcane, some vegetables, and perennial/horticultural
crops. Areas that are converted between agriculture (i.e., cropland and grassland) and other land uses, such as
forest land, wetland and settlements, are not simulated with DayCent. DayCent is also not used to estimate
emissions from land areas with very gravelly, cobbly, or shaley soils in the topsoil (greater than 35 percent by
volume in the top 30 cm of the soil profile), or to estimate emissions from drained organic soils (Histosols). The Tier
3 method has not been fully tested for estimating N2O emissions associated with these crops and rotations, land
uses, as well as organic soils or cobbly, gravelly, and shaley mineral soils. In addition, federal grassland areas are
not simulated with DayCent due to limited activity data on land use histories. For areas that are not included in the
DayCent simulations, Tier 1 methods are used to estimate emissions, including (1) direct emissions from nitrogen
inputs for crops on mineral soils that are not simulated by DayCent; (2) direct emissions from PRP nitrogen
additions on federal grasslands; (3) direct emissions for land application of biosolids (i.e., treated sewage sludge) to
soils; and (4) direct emissions from drained organic soils in croplands and grasslands.
A splicing method is used to estimate soil N2O emissions for 2021 to 2022 at the national scale because new
activity data have not been incorporated into the analysis for those years. Specifically, linear regression models
with autoregressive moving-average (ARMA) errors (Brockwell and Davis 2016) are used to estimate the
relationship between surrogate data and the 1990 to 2020 emissions that are derived using the Tier 3 method.
Surrogate data for these regression models includes corn and soybean yields from USDA-NASS statistics,19 and
weather data from the PRISM Climate Group (PRISM 2022). For the Tier 1 method, a linear-time series model is
used to estimate emissions for 2021-2022 without surrogate data. In addition, the linear time series model is used
to estimate emissions data for 2018 to 2022 for other organic nitrogen amendments (i.e., commercial organic
fertilizer) due to a gap in the activity data during the latter part of the time series (TVA 1991 through 1994;
AAPFCO 1995 through 2022). See Box 5-4 for more information about the splicing method. Emission estimates for
years with imputed data will be recalculated in future Inventory reports when new NRI data and other organic
amendment nitrogen data are available.
18 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.
19 See https://quickstats.nass.usda.gov/.
Agriculture 5-37
-------
Box 5-4: Data Splicing Method
An approach to extend the time series is needed for agricultural soil management because there are typically
activity data gaps at the end of the time series. This is mainly because the NRI survey program, which provides
critical information for estimating greenhouse gas emissions and removals, does not release data every year.
Splicing methods have been used to impute missing data at the end of the emission time series for both the Tier
1 and 3 methods. Specifically, a linear regression model with autoregressive moving-average (ARMA) errors
(Brockwell and Davis 2016) is used to estimate emissions based on the emissions data that has been compiled
using the inventory methods described in this section. The model to extend the time series is given by the
equation:
Y = Xp + e,
where Y is the response variable (e.g., soil nitrous oxide), xp for the Tier 3 method contains specific surrogate
data depending on the response variable, and £ is the remaining unexplained error. Models with a variety of
surrogate data were tested, including commodity statistics, weather data, or other relevant information. The
term xp for the Tier 1 method only contains year as a predictor of emission patterns over the time series
(change in emissions per year), and therefore, is a linear time series model with no surrogate data. Parameters
are estimated using standard statistical techniques, and used in the model described above to predict the
missing emissions data.
A critical issue with splicing methods is to account for the additional uncertainty introduced by predicting
emissions without compiling the full inventory. Specifically, uncertainty will increase for years with imputed
estimates based on the splicing methods, compared to those years in which the full inventory is compiled. This
additional uncertainty is quantified within the model framework using a Monte Carlo approach. Consequently,
the uncertainty from the original inventory data is combined with the uncertainty in the data splicing model.
The approach requires estimating parameters in the data splicing models in each Monte Carlo simulation for the
full inventory (i.e., the surrogate data model is refit with the draws of parameters values that are selected in
each Monte Carlo iteration, and used to produce estimates with inventory data). Therefore, the data splicing
method generates emissions estimates from each surrogate data model in the Monte Carlo analysis, which are
used to derive confidence intervals in the estimates for the missing emissions data. Furthermore, the 95 percent
confidence intervals are estimated using the 3 sigma rules assuming a unimodal density (Pukelsheim 1994).
Tier 3 Approach for Mineral Cropland Soils
The DayCent biogeochemical model (Parton et al. 1998; Del Grosso et al. 2001 and 2011) is used to estimate direct
N2O emissions from mineral cropland soils that are managed for production of a wide variety of crops (see list in
previous section) based on the crop histories in the 2017 NRI (USDA-NRCS 2020) and extended through 2020 using
CDL (USDA-NASS 2021). Crops simulated by DayCent are grown on approximately 85 percent of total cropland area
in the United States. The model simulates net primary productivity (NPP) using the NASA-CASA production
algorithm MODIS Enhanced Vegetation Index (EVI) products, MOD13Q1 and MYD13Q120 (Potter et al. 1993, 2007).
The model simulates soil temperature and water dynamics, using daily weather data from a 4-kilometer gridded
product developed by the PRISM Climate Group (2022), and soil attributes from the Soil Survey Geographic
20 Net Primary Production 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 to 2020. 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.
5-38 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
Database (SSURGO) (Soil Survey Staff 2020). DayCent is used to estimate direct N2O emissions due to mineral
nitrogen available from the following sources: (1) application of synthetic fertilizers; (2) application of livestock
manure; (3) retention of crop residues in the field for nitrogen-fixing legumes and non-legume crops and
subsequent mineralization of nitrogen during microbial decomposition (i.e., leaving residues in the field after
harvest instead of burning or collecting residues); (4) mineralization of nitrogen from decomposition of soil organic
matter; and (5) asymbiotic fixation.
Management activity data from several sources supplement the activity data from the NRI. The USDA-NRCS
Conservation Effects and Assessment Project (CEAP) provides data on a variety of cropland management activities,
and is used to inform the inventory analysis about tillage practices, mineral fertilization, manure amendments,
cover crop management, as well as planting and harvest dates (USDA-NRCS 2022; USDA-NRCS 2018; USDA-NRCS
2012). CEAP data are collected at a subset of NRI survey locations, and currently provide management information
from approximately 2002 to 2006 and 2013 to 2016. These data are combined with other datasets in an
imputation analysis. This imputation analysis is comprised of three steps: a) determine the trends in management
activity across the time series by combining information from several datasets (discussed below); b) use Gradient
Boosting (Friedman 2001) to determine the likely management practice at a given NRI survey location; and c)
assign management practices from the CEAP survey to the specific NRI locations using a predictive mean matching
method for certain variables that are adapted to reflect the trending information (Little 1988, van Buuren 2012).
Gradient boosting is a machine learning technique used in regression and classification tasks, among others. It
combines predictions from multiple weak prediction models and outperforms many complicated machine learning
algorithms. It makes the best predictions at specific NRI survey locations or at state or region level models. The
predictive mean matching method identifies the most similar management activity recorded in the CEAP surveys
that match the prediction from the gradient boosting algorithm. The matching ensures that imputed management
activities are realistic for each NRI survey location, and not odd or physically unrealizable results that could be
generated by the gradient boosting. There are six complete imputations of the management activity data using
these methods.
To determine trends in mineral fertilization and manure amendments, CEAP data are combined with information
on fertilizer use and rates by crop type for different regions of the United States from the USDA Economic
Research Service. The data collection program was known as the Cropping Practices Surveys through 1995 (USDA-
ERS 1997), and is now part of data collection known as the Agricultural Resource Management Surveys (ARMS)
(USDA-ERS 2020). Additional data on fertilization practices are compiled through other sources particularly the
National Agricultural Statistics Service (USDA-NASS 1992,1999, 2004). To determine the trends in tillage
management, CEAP data are combined with Conservation Technology Information Center data between 1989 and
2004 (CTIC 2004) and OpTIS Data Product21 for 2008 to 2020 (Hagen et al. 2020). The CTIC data are adjusted for
long-term adoption of no-till agriculture (Towery 2001). For cover crops, CEAP data are combined with information
from USDA Census of Agriculture (USDA-NASS 2012, 2017) and the OpTIS22 data (Hagen et al. 2020). It is assumed
that cover crop management was minimal prior to 1990 and the rates increased linearly over the decade to the
levels of cover crop management in the CEAP survey.
The IPCC method considers crop residue nitrogen inputs and nitrogen mineralized from soil organic matter as
activity data. However, they are not treated as activity data in DayCent simulations because residue production,
symbiotic nitrogen fixation (e.g., legumes), mineralization of nitrogen from soil organic matter, and asymbiotic
nitrogen fixation are internally generated by the model as part of the simulation. In other words, DayCent accounts
for the influence of symbiotic nitrogen fixation, mineralization of nitrogen from soil organic matter and crop
residue retained in the field, and asymbiotic nitrogen fixation on N2O emissions, but these are not model inputs.
The N2O emissions from crop residues are reduced by approximately 3 percent (the assumed average burned
portion for crop residues in the United States) to avoid double counting associated with non-CC>2 greenhouse gas
21 OpTIS data on tillage practices provided by Regrow Agriculture, Inc.
22 OpTIS data on cover crop management provided by Regrow Agriculture, Inc.
Agriculture 5-39
-------
emissions from agricultural residue burning. Estimated levels of residue burning are based on state inventory data
(ILENR 1993; Oregon Department of Energy 1995; Noller 1996; Wisconsin Department of Natural Resources 1993;
Cibrowski 1996).
Uncertainty in the emission estimates from DayCent is associated with input uncertainty due to missing
management data in the NRI survey that is imputed from other sources; model uncertainty due to incomplete
specification of carbon and nitrogen dynamics in the DayCent model parameters and algorithms; and sampling
uncertainty associated with the statistical design of the NRI survey. Uncertainty is estimated with two variance
components (Ogle et al. 2010). The first variance component quantifies the uncertainty in management activity
data, model structure and parameterization. To assess this uncertainty, carbon and nitrogen dynamics at each NRI
survey location are simulated six times using the imputation product and other model driver data. Uncertainty in
parameterization and model algorithms are determined using a structural uncertainty estimator derived from
fitting a linear mixed-effect model (Ogle et al. 2007; Del Grosso et al. 2010). The data is combined in a Monte Carlo
stochastic simulation with 1,000 iterations for 1990 through 2020. For each iteration, there is a random selection
of management data from the imputation product (select one of the six imputations), and random selection of
parameter values and random effects for the linear mixed-effect model (i.e., structural uncertainty estimator). The
second variance component quantifies uncertainty in scaling from the NRI survey to the entire land base. The
second variance component is computed using the replicate weights provided with the NRI survey data, and a
standard variance estimator for a two-stage sample design (Sarndal et al. 1992). The two variance components are
summed to quantify the total uncertainty and produce confidence intervals associated with the estimated
emissions.
In order to ensure time-series consistency, the DayCent model is applied from 1990 to 2020, and a linear
extrapolation method is used to approximate emissions for 2021 to 2022 based on the pattern in emissions data
from 1990 to 2020 (see Box 5-4). The pattern is determined using a linear regression model with moving-average
(ARMA) errors. Linear extrapolation is a standard data splicing method for approximating missing values at the end
of an inventory time series (IPCC 2006). The time series will be updated with the Tier 3 method in the future as
new activity data are incorporated into the analysis.
Nitrous oxide emissions from managed agricultural lands are the result of interactions among anthropogenic
activities (e.g., nitrogen fertilization, manure application, tillage) and other driving variables, such as weather and
soil characteristics. These factors influence key processes associated with nitrogen dynamics in the soil profile,
including immobilization of nitrogen by soil microbial organisms, decomposition of organic matter, plant uptake,
leaching, runoff, and volatilization, as well as the processes leading to N2O production (nitrification and
denitrification). It is not possible to partition N2O emissions into each anthropogenic activity directly from model
outputs due to the complexity of the interactions (e.g., N2O emissions from synthetic fertilizer applications cannot
be distinguished from those resulting from manure applications). To approximate emissions by activity, the
amount of synthetic nitrogen fertilizer added to the soil, or mineral nitrogen made available through
decomposition of soil organic matter and plant litter, as well as asymbiotic fixation of nitrogen from the
atmosphere, is determined for each nitrogen source and then divided by the total amount of mineral nitrogen in
the soil according to the DayCent model simulation. For 2021 to 2022, the contribution of each nitrogen source is
based on the average of values that are estimated for 2018 to 2020. The percentages are then multiplied by the
total of direct N2O emissions in order to approximate the portion attributed to nitrogen management practices.
This approach is only an approximation because it assumes that all nitrogen made available in soil has an equal
probability of being released as N2O, regardless of its source, which is unlikely to be the case (Delgado et al. 2009).
However, this approach allows for further disaggregation of emissions by source of nitrogen, which is valuable for
reporting purposes and is analogous to the reporting associated with the IPCC (2006) Tier 1 method, in that it
associates portions of the total soil N2O emissions with individual sources of nitrogen.
Tier 1 Approach for Mineral Cropland Soils
The IPCC (2006) Tier 1 methodology is used to estimate direct N2O emissions for mineral cropland soils that are not
simulated by DayCent (e.g., DayCent has not been parametrized to simulate all crop types and some soil types such
5-40 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
as Histosols). For the Tier 1 method, estimates of direct N2O emissions from nitrogen applications are based on
mineral soil N that is made available from the following practices: (1) the application of synthetic commercial
fertilizers; (2) application of managed manure and non-manure commercial organic fertilizers; and (3)
decomposition and mineralization of nitrogen from above- and below-ground crop residues in agricultural fields
(i.e., crop biomass that is not harvested). Non-manure commercial organic amendments are only included in the
Tier 1 analysis because these data are not available at the county-level, which is necessary for the DayCent
simulations. Consequently, all commercial organic fertilizer, as well as manure that is not added to crops in the
DayCent simulations, are included in the Tier 1 analysis. The following sources are used to derive activity data:
• A process-of-elimination approach is used to estimate synthetic nitrogen fertilizer additions for crop areas that
are not simulated by DayCent. The total amount of fertilizer used on farms has been estimated at the county-
level by the USGS using sales records from 1990 to 2012 (Brakebill and Gronberg 2017). For 2013 through
2017, fertilizer sales data from AAPFCO (AAPFCO 2013 through 2022)23 after adjusting for the proportion of
on-farm application to determine the amount applied to crops. The amount of fertilizer applied after 2017 is
estimated using the data splicing method described in Box 5-4 for the linear time series model. Then the
portion of fertilizer applied to crops and grasslands simulated by DayCent is subtracted from the on-farm sales
data (see Tier 3 Approach for mineral cropland soils and direct N2O emissions from grassland soils sections for
information on data sources), and the remainder of the total fertilizer used on farms is assumed to be applied
to crops that are not simulated by DayCent. At a minimum, 3 percent of state-level on-farm fertilizer sales are
assumed to be applied to cropland in the Tier 1 method.
• Similarly, a process-of-elimination approach is used to estimate manure nitrogen additions for crops that are
not simulated by DayCent. The total amount of manure available for land application to soils has been
estimated with methods described in the manure management section (Section 5.2) and annex (Annex 3.11).
The amount of manure nitrogen applied in the Tier 3 approach to crops and grasslands is subtracted from total
annual manure nitrogen available for land application (see Tier 3 Approach for mineral cropland soils and
direct N2O emissions from grassland soils sections for information on data sources). This difference is assumed
to be applied to crops that are not simulated by DayCent.
• Commercial organic fertilizer additions are based on organic fertilizer consumption statistics through 2017,24
which are converted from mass of fertilizer to units of nitrogen using average organic fertilizer nitrogen
content, ranging between 2.3 to 4.2 percent across the time series (TVA 1991 through 1994; AAPFCO 1995
through 2022). Commercial fertilizers include dried manure and biosolids (i.e., treated sewage sludge), but the
amounts are removed from the commercial fertilizer data to avoid double counting25 with the manure
nitrogen dataset described above and the biosolids (i.e., treated sewage sludge) amendment data discussed
later in this section.
• Crop residue nitrogen is derived by combining amounts of above- and below-ground biomass, which are
determined based on NRI crop area data (USDA-NRCS 2020), as extended using the CDL data (USDA-NASS
2021), crop production yield statistics (USDA-NASS 2023), dry matter fractions (IPCC 2006), linear equations to
23 The fertilizer consumption data in AAPFCO are recorded in "fertilizer year" totals, (i.e., July to June), but are converted to
calendar year totals. This is done by assuming that approximately 35 percent of fertilizer usage occurred from July to December
and 65 percent from January to June (TVA 1992b).
24 Soil N20 emissions are imputed using data splicing methods for commercial fertilizers, i.e., other organic fertilizers, after
2017 because the activity data are not available.
25 Commercial organic fertilizers include dried blood, tankage, compost, and other, but the dried manure and biosolids (i.e.,
treated sewage sludge) are also included in other datasets in this Inventory. Consequently, the proportions of dried manure and
biosolids, which are provided in the reports (TVA 1991 through 1994; AAPFCO 1995 through 2022), are used to estimate the
nitrogen amounts in dried manure and biosolids. To avoid double counting, the resulting nitrogen amounts for dried manure
and biosolids are subtracted from the total nitrogen in commercial organic fertilizers before estimating emissions using the Tier
1 method.
Agriculture 5-41
-------
estimate above-ground biomass given dry matter crop yields from harvest (IPCC 2006), ratios of below-to-
above-ground biomass (IPCC 2006), and nitrogen contents of the residues (IPCC 2006). Nitrogen inputs from
residue were reduced by 3 percent to account for average residue burning portions in the United States.
The total amounts of soil mineral nitrogen from applied synthetic and organic fertilizers, manure nitrogen
additions and crop residues are multiplied by the IPCC (2006) default emission factor to derive an estimate of
direct N2O emissions using the Tier 1 method. Further elaboration on the methodology and data used to estimate
N2O emissions from mineral soils are described in Annex 3.12.
In order to ensure time-series consistency, the Tier 1 methods are applied from 1990 to 2020, and a linear
extrapolation method26 is used to approximate emissions for 2021 to 2022 based on the emission patterns
between 1990 and 2020 (see Box 5-4). The exceptions include crop residue nitrogen which is estimating using the
Tier 1 method for 1990 to 2022 with no linear extrapolation, and for other organic nitrogen fertilizers (i.e.,
commercial fertilizers), which are estimated with linear time series model for 2018 to 2022 due to a gap in the
activity data during the latter part of the time series (TVA 1991 through 1994; AAPFCO 1995 through 2022). For the
extrapolation, the emission pattern is determined using a linear regression model with moving-average (ARMA)
errors. Linear extrapolation is a standard data splicing method for approximating missing values at the end of an
inventory time series (IPCC 2006). As with the Tier 3 method, the time series that is based on the splicing methods
will be recalculated in a future Inventory report with updated activity data.
Tier 1 and 3 Approaches from Mineral Grassland Soils
As with N2O emissions from croplands, the Tier 3 process-based approach with application of the DayCent model
and Tier 1 method described in IPCC (2006) are combined to estimate emissions from non-federal grasslands and
PRP manure nitrogen additions for federal grasslands, respectively. Grassland includes pasture and rangeland that
produce grass or mixed grass/legume forage primarily for livestock grazing. Rangelands are extensive areas of
native grassland that are not intensively managed, while pastures are seeded grassland (possibly following tree
removal) that may also have additional management, such as irrigation, fertilization, or inter-seeding legumes.
DayCent is used to simulate N2O emissions from NRI survey locations (USDA-NRCS 2020) on non-federal grasslands
resulting from manure deposited by livestock directly onto pastures and rangelands (i.e., PRP manure), nitrogen
fixation from legume seeding, managed manure amendments (i.e., manure other than PRP manure such as daily
spread or manure collected from other animal waste management systems such as lagoons and digesters), and
synthetic fertilizer application. Other nitrogen inputs are simulated within the DayCent framework, including
nitrogen input from mineralization due to decomposition of soil organic matter and nitrogen inputs from senesced
grass litter, as well as asymbiotic fixation of nitrogen from the atmosphere. The simulations used the same
weather, soil, and synthetic nitrogen fertilizer data as discussed under the Tier 3 Approach in the mineral cropland
soils section. Synthetic nitrogen fertilization rates are based on data from the Carbon Sequestration Rural
Appraisals (CSRA) conducted by the USDA-NRCS (USDA-NRCS, unpublished data). The CSRAwas a solicitation of
expert knowledge from USDA-NRCS staff throughout the United States to support the Inventory. Biological
nitrogen fixation is simulated within DayCent, and therefore is not an input to the model.
Manure nitrogen deposition from grazing animals in PRP systems (i.e., PRP manure nitrogen) is a key input of
nitrogen to grasslands. The amounts of PRP manure nitrogen applied on non-federal grasslands for each NRI
survey location are based on the amount of nitrogen excreted by livestock in PRP systems that is estimated in the
manure management section (see Section 5.2 and Annex 3.11). The total amount of nitrogen excreted in each
county is divided by the grassland area to estimate the nitrogen input rate associated with PRP manure. The
resulting rates are a direct input into the DayCent simulations. The nitrogen input is subdivided between urine and
dung based on a 50:50 split. DayCent simulations of non-federal grasslands accounted for approximately 71
5-42 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
percent of total PRP manure nitrogen in aggregate across the country.27 The remainder of the PRP manure
nitrogen in each state is assumed to be excreted on federal grasslands, and the N2O emissions are estimated using
the IPCC (2006) Tier 1 method.
Biosolids (i.e., treated sewage sludge) are assumed to be applied on grasslands.28 Application of biosolids is
estimated from data compiled by EPA (1993,1999, 2003), McFarland (2001), and NEBRA (2007) (see Section 7.2 for
a detailed discussion of the methodology for estimating treated sewage sludge available for land application
application). Biosolids data are only available at the national scale, and it is not possible to associate application
with specific soil conditions and weather at NRI survey locations. Therefore, DayCent could not be used to simulate
the influence of biosolids on N2O emissions from grassland soils, and consequently, emissions from biosolids are
estimated using the IPCC (2006) Tier 1 method.
Soil N2O emission estimates from DayCent are adjusted using a structural uncertainty estimator accounting for
uncertainty in model algorithms and parameter values (Del Grosso et al. 2010). There is also sampling uncertainty
for the NRI survey that is quantified with replicate sampling weights associated with the survey, as discussed for
Tier 3 method associated with mineral cropland soils. N2O emissions for the PRP manure nitrogen deposited on
federal grasslands and applied biosolids nitrogen are estimated using the Tier 1 method by multiplying the
nitrogen input by the default emission factor. Emissions from manure nitrogen are estimated at the state level and
aggregated to the entire country, but emissions from biosolids nitrogen are calculated exclusively at the national
scale. Further elaboration on the methodology and data used to estimate N2O emissions from mineral soils are
described in Annex 3.12.
Soil N2O emissions and 95 percent confidence intervals are estimated for each year between 1990 and 2020 based
on the Tier 1 and 3 methods, except for biosolids (discussed below). In order to ensure time-series consistency,
emissions from 2021 to 2022 are estimated using a splicing method as described in Box 5-4, with a linear
extrapolation based on the emission patterns in the 1990 to 2020 data. Linear extrapolation is a standard data
splicing method for approximating emissions at the end of a time series (IPCC 2006). As with croplands, estimates
for 2021 to 2022 will be recalculated in a future Inventory when the activity data are updated. Biosolids application
data are compiled through 2022 in this Inventory, and therefore soil N2O emissions and confidence intervals are
estimated using the Tier 1 method for all years without application of the splicing method.
Tier 1 Approach for Drainage of Organic Soils in Croplands and Grasslands
The IPCC (2006) Tier 1 method is used to estimate direct N2O emissions due to drainage of organic soils in
croplands and grasslands at a state scale. State-scale estimates of the total area of drained organic soils are
obtained from the 2017 NRI (USDA-NRCS 2020), and extended through 2022 using CDL (USDA-NASS 2021) and the
Forest Inventory and Analysis (FIA) survey data, which is harmonized with the NRI data (Nelson et al. 2020).
Organic soils are identified using soils data from the Soil Survey Geographic Database (SSURGO) (Soil Survey Staff
2020). The IPCC climate region map is used to subdivide areas into temperate and tropical climates according to
the climate classification from IPCC (2006). To estimate annual emissions, the total temperate area is multiplied by
the IPCC default emission factor for temperate regions, and the total tropical area is multiplied by the IPCC default
emission factor for tropical regions (IPCC 2006). In order to ensure time-series consistency, the Tier 1 methods are
applied from 1990 to 2022.
27 A small amount of PRP nitrogen (less than 1 percent) is deposited in grazed pasture that is in rotation with annual crops and
is reported in the grassland N20 emissions.
28 A portion of biosolids may be applied to croplands, but there is no national dataset to disaggregate the amounts between
cropland and grassland.
Agriculture 5-43
-------
Total Direct N2O Emissions from Cropland and Grassland Soils
Annual direct emissions from the Tier 1 and 3 approaches for mineral and drained organic soils occurring in both
croplands and grasslands are summed to obtain the total direct N2O emissions from agricultural soil management
(see Table 5-15 and Table 5-16). Further elaboration on the methodology and data used to estimate soil N2O
emissions are described in Annex 3.12.
Indirect N2O Emissions Associated with Nitrogen Management in Cropland and
Grasslands
Indirect N2O emissions occur when synthetic nitrogen applied or made available through anthropogenic activity is
transported from the soil either in gaseous or aqueous forms and later converted into N2O. There are two
pathways leading to indirect emissions. The first pathway results from volatilization of nitrogen as NOx (nitrogen
oxides) and NH3 (ammonia) following application of synthetic fertilizer, organic amendments (e.g., manure,
biosolids), and deposition of PRP manure. Nitrogen made available from mineralization of soil organic matter and
residue, including nitrogen incorporated into crops and forage from symbiotic nitrogen fixation, and input of
nitrogen from asymbiotic fixation also contributes to volatilized nitrogen emissions. Volatilized nitrogen can be
returned to soils through atmospheric deposition, and a portion of the deposited nitrogen is emitted to the
atmosphere as N2O. The second pathway occurs via leaching and runoff of soil nitrogen (primarily in the form of
NO3", i.e., nitrate) that is made available through anthropogenic activity on managed lands, including organic and
synthetic fertilization, organic amendments, mineralization of soil organic matter and residue, and inputs of
nitrogen into the soil from asymbiotic fixation. Nitrate is subject to denitrification in water bodies, which leads to
N2O emissions. Regardless of the eventual location of the indirect N2O emissions, the emissions are assigned to the
original source of the nitrogen for reporting purposes, which here includes croplands and grasslands.
Tier 1 and 3 Approaches for Indirect N2O Emissions from Atmospheric Deposition of Volatilized
Nitrogen
The Tier 3 DayCent model and IPCC (2006) Tier 1 methods are combined to estimate the amount of nitrogen that is
volatilized and eventually emitted as N2O. DayCent is used to estimate nitrogen volatilization for land areas whose
direct emissions are simulated with DayCent (i.e., most commodity and some specialty crops and most grasslands).
The nitrogen inputs included are the same as described for direct N2O emissions in the Tier 3 approach for mineral
cropland and grassland soils sections. Nitrogen volatilization from all other areas is estimated using the Tier 1
method with default IPCC fractions for nitrogen subject to volatilization (i.e., synthetic and manure nitrogen on
croplands not simulated by DayCent, other organic nitrogen inputs (i.e., commercial fertilizers), PRP manure
nitrogen excreted on federal grasslands, and biosolids [i.e., treated sewage sludge] application on grasslands).
The IPCC (2006) default emission factor is multiplied by the amount of volatilized nitrogen generated from both
DayCent and Tier 1 methods to estimate indirect N2O emissions occurring with re-deposition of the volatilized
nitrogen from 1990-2020 (see Table 5-18). A linear extrapolation data splicing method, described in Box 5-4, is
applied to estimate emissions from 2021 to 2022 based on the emission patterns from 1990 to 2020. Linear
extrapolation is a standard data splicing method for estimating emissions at the end of a time series (IPCC 2006).
Further elaboration on the methodology and data used to estimate indirect N2O emissions are described in Annex
3.12.
Tier 1 and 3 Approaches for Indirect N2O Emissions from Leaching/Runoff
As with the calculations of indirect emissions from volatilized nitrogen, the Tier 3 DayCent model and IPCC (2006)
Tier 1 method are combined to estimate the amount of nitrogen that is subject to leaching and surface runoff into
water bodies, and eventually emitted as N2O. DayCent is used to simulate the amount of nitrogen transported
from lands in the Tier 3 Approach. Nitrogen transport from all other areas is estimated using the Tier 1 method and
the IPCC (2006) default factor for the proportion of nitrogen subject to leaching and runoff associated with
5-44 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
nitrogen applications on croplands that are not simulated by DayCent, applications of biosolids on grasslands,
other organic N fertilizer applications, crop residue nitrogen inputs, and PRP manure nitrogen excreted on federal
grasslands.
For both the DayCent Tier 3 and IPCC (2006) Tier 1 methods, NO3" leaching is assumed to be an insignificant source
of indirect N2O in cropland and grassland systems in arid regions, as discussed in IPCC (2006). In the United States,
the threshold for significant NO3" leaching is based on the potential evapotranspiration (PET) and rainfall amount,
similar to IPCC (2006), and is assumed to be negligible in regions where the amount of precipitation does not
exceed 80 percent of PET (Note: All irrigated systems are assumed to have significant amounts of leaching of
nitrogen even in drier climates).
For leaching and runoff data estimated by the Tier 3 and Tier 1 approaches, the IPCC (2006) default emission factor
is used to estimate indirect N2O emissions that occur in groundwater and waterways (see Table 5-18). Further
elaboration on the methodology and data used to estimate indirect N2O emissions are described in Annex 3.12.
In order to ensure time-series consistency, indirect soil N2O emissions are estimated using the Tier 1 and 3
approaches from 1990 to 2020 and then a linear extrapolation data splicing method, described in Box 5-4, is
applied to estimate emissions from 2021 to 2022 based on the emission patterns from 1990 to 2020. Linear
extrapolation is a standard data splicing method for estimating emissions at the end of a time series (IPCC 2006).
As with the direct N2O emissions, the time series will be recalculated in a future Inventory when new activity data
are incorporated into the analysis.
Uncertainty
Uncertainty is estimated for each of the following five components of N2O emissions from agricultural soil
management: (1) direct emissions simulated by DayCent; (2) the components of indirect emissions (nitrogen
volatilized and leached or runoff) simulated by DayCent; (3) direct emissions estimated with the IPCC (2006) Tier 1
method; (4) the components of indirect emissions (nitrogen volatilized and leached or runoff) estimated with the
IPCC (2006) Tier 1 method; and (5) indirect emissions estimated with the IPCC (2006) Tier 1 method. Uncertainty in
direct emissions as well as the components of indirect emissions that are estimated from DayCent are derived
from two variance components (Ogle et al. 2010). For the first component, a Monte Carlo Analysis (consistent with
IPCC Approach 2) is used to address uncertainties in management activity data as well as model parameterization
and structure (Del Grosso et al. 2010). The second variance component is quantifying uncertainty in scaling from
the NRI survey to the entire land base, and computed using a standard variance estimator for a two-stage sample
design (Sarndal et al. 1992). The two variance components are combined using 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. For 2021 to 2022 (and 2018 to 2022 for other organic nitrogen fertilizers) there is
additional uncertainty propagated through the Monte Carlo Analysis associated with the splicing method (See Box
5-4) except for the Tier 1 method for biosolids and crop residue nitrogen inputs, which do not use the data splicing
method for 2021 to 2022.
Simple error propagation methods (IPCC 2006) are used to derive confidence intervals for direct emissions
estimated with the IPCC (2006) Tier 1 method, the proportion of volatilization and leaching or runoff estimated
with the IPCC (2006) Tier 1 method, and indirect N2O emissions. Uncertainty in the splicing method is also included
in the error propagation for 2021-2022 (see Box 5-4). Additional details on the uncertainty methods are provided
in Annex 3.12. Table 5-19 shows the combined uncertainty for soil N2O emissions. The estimated direct soil N2O
emissions range from 28 percent below to 28 percent above the 2022 emission estimate of 262.5 MMT CO2 Eq.
The combined uncertainty for indirect soil N2O emissions ranges from 51 percent below to 123 percent above the
2022 estimate of 28.3 MMT CO2 Eq.
Agriculture 5-45
-------
Table 5-19: Quantitative Uncertainty Estimates of N2O Emissions from Agricultural Soil
Management in 2022 (MMT CO2 Eq. and Percent)
2022 Emission
Source
Gas
Estimate
Uncertainty Range Relative to Emission Estimate
(MMTCO. Eq.)
(MMT CO
Eq.)
(%)
Lower
Upper
Lower Upper
Bound
Bound
Bound Bound
Direct Soil N20 Emissions
N20
262.5
189.6
335.3
-28% +28%
Indirect Soil N2O Emissions
n2o
28.3
13.7
63.3
-51% +123%
Note: Due to lack of data, uncertainties in PRP manure nitrogen production, other organic fertilizer amendments, and
biosolids (i.e., treated sewage sludge) amendments to soils are currently treated as certain. These sources of
uncertainty will be included in a future Inventory (IPCC 2006).
Additional uncertainty is associated with an incomplete estimation of N2O emissions from managed croplands and
grasslands in Hawaii and Alaska. The Inventory currently includes the N2O emissions from managed manure and
PRP nitrogen, and biosolid additions for Alaska and managed manure and PRP nitrogen, biosolid additions, and
crop residue for Hawaii. Land areas used for agriculture in Alaska and Hawaii are small relative to major crop
commodity states in the conterminous United States, so the emissions are likely to be minor for the other sources
of nitrogen (e.g., synthetic fertilizer and crop residue inputs). Regardless, there is a planned improvement to
include the additional sources of emissions in a future Inventory.
/erification
General (Tier 1) and category-specific (Tier 2) QA/QC activities were conducted consistent with the U.S. Inventory
QA/QC plan outlined in Annex 8. DayCent results for N2O emissions and NO3" 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. Nitrous oxide measurement data for cropland are available for
64 sites with 769 observations of management practice effects, and measurement data for grassland are available
for 12 sites with 88 observations of management practice effects. Nitrate leaching data are available for 14 sites,
representing 432 observations of management practice effects. In general, DayCent predicted N2O emission and
nitrate leaching for these sites reasonably well. See Annex 3.12 for more detailed information about the
comparisons.
Databases containing input data and probability distribution functions required for DayCent simulations of
croplands and grasslands and unit conversion factors have been checked. In addition, program scripts that are
used to run the Monte Carlo uncertainty analysis have been checked. Errors were found in the synthetic nitrogen
application rates for the Tier 3 method for a subset of years in some states, with overapplication based on
comparisons to the synthetic fertilizer sales data. An error in the uncertainty calculation was found due to
improper formulation of land area variances. A minor error was also identified in manure deposited in pasture,
range, and paddock. Databases containing input data, emission factors, and calculations required for the Tier 1
method have been checked and updated as needed. Quality control identified a problem with error propagation in
the Tier 1 uncertainty analysis associated with the emission factors. There was also an error identified in the
leaching calculation based on irrigation status. All of these errors were corrected. Links between spreadsheets
have also been checked, updated, and corrected as needed.
Recalculations Discussion
Several improvements have been implemented in this Inventory leading to recalculations, including a) updated
time series of land representation data (Nelson et al. 2020), b) re-calibration of the soil carbon module in the
DayCent model (See Annex 3.12); c) a more accurate output variable to estimate asymbiotic nitrogen fixation in
5-46 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
the Tier 3 method, and d) corrections associated with manure deposited on pasture, range and paddock in
addition to estimation of leaching based on irrigation status. The combined impact from these changes resulted in
an average annual increase in emissions of 3.3 MMT CO2 Eq., or 1.1 percent, from 1990 to 2021 relative to the
previous Inventory.
Planned Improvements
Several planned improvements are underway associated with improving the DayCent biogeochemical model.
These improvements include a better representation of plant phenology, particularly senescence events following
grain filling in crops. In addition, crop parameters associated with temperature and water stress effects on plant
production will be further improved in DayCent with additional model calibration. In addition, there is an
improvement underway to calibrate the nitrogen submodule in order to more accurately predict nitrogen-gas
losses and nitrate leaching rates. Experimental study sites will continue to be added for quantifying model
structural uncertainty with priority given to studies that have continuous (daily) measurements of N2O (e.g., Scheer
et al. 2013). In addition, improvements are underway to simulate crop residue burning in the DayCent model
based on the amount of crop residues burned according to the data that is used in the Field Burning of Agricultural
Residues source category (see Section 5.7).
For Tier 1, there is a planned improvement to include all sources of nitrogen for Alaska and Hawaii in the Inventory
for agricultural soil management, which currently only addresses managed manure nitrogen and PRP nitrogen, and
biosolids additions for grasslands in both states, in addition to crop residue nitrogen inputs for Hawaii. There is also
an improvement to incorporate the Tier 1 emission factor for N2O emissions from drained organic soils by using
the revised factors in the 2013 Supplement to the 2006IPCC Guidelines for National Greenhouse Gas Inventories:
Wetlands (IPCC 2014). There is a planned improvement for the Tier 1 method associated with estimating soil N2O
emissions from nitrogen mineralization due to soil organic matter decomposition that is accelerated with land use
conversions to cropland and grassland. Lastly, a review of available data on biosolids (i.e., treated sewage sludge)
application will also be undertaken to improve the distribution of biosolids application on croplands, grasslands
and settlements.
Improvements are expected to be completed for the next Inventory (i.e., 2025 submission to the UNFCCC, 1990
through 2023 Inventory), pending prioritization of resources.
5.5 Liming (CRT Source Category 3G)
Crushed limestone (CaCOs) and dolomite (CaMg(CC>3)2) are added to soils by land managers to increase soil pH
(i.e., to reduce acidification). Carbon dioxide emissions occur as these compounds react with hydrogen ions in
soils. The rate of degradation of applied limestone and dolomite depends on the soil conditions, soil type, climate
regime, and whether limestone or dolomite is applied. Emissions from limestone and dolomite that are used in
industrial processes (e.g., cement production, glass production, etc.) are reported in the IPPU chapter. Emissions
from liming of soils have fluctuated between 1990 and 2022 in the United States, ranging from 2.2 MMT CO2 Eq. to
6.0 MMT CO2 Eq. across the entire time series. In 2022, liming of soils in the United States resulted in emissions of
3.3 MMT CO2 Eq. (0.9 MMT C), representing a 30 percent decrease in emissions since 1990 (see Table 5-20 and
Table 5-21). The trend is driven by variation in the amount of limestone and dolomite applied to soils over the time
period.
Agriculture 5-47
-------
Table 5-20: Emissions from Liming (MMT CO2 Eq.)
Source
1990
2005 1
2018
2019
2020
2021
2022
Limestone
4.1
3.9
2.0
1.9
2.5
2.0
2.9
Dolomite
0.6
0.4
0.2
0.3
0.4
0.4
0.3
Total
4.7
4.4 |
2.2
2.2
2.9
2.4
3.3
Note: Totals may not sum due to independent rounding.
Table 5-21:
Emissions from Liming (MMT C)
Source
1990
2005 1
2018
2019
2020
2021
2022
Limestone
1.1
1.1
0.6
0.5
0.7
0.5
0.8
Dolomite
0.2
0.1
0.1
0.1
0.1
0.1
0.1
Total
1.3
!.2 |
0.6
0.6
0.8
0.7
0.9
+ Does not exceed 0.05 MMT C.
Note: Totals may not sum due to independent rounding.
Methodology and Time-Series Consistency
Carbon dioxide emissions from application of limestone and dolomite to soils were estimated using a Tier 2
methodology consistent with IPCC (2006). The annual amounts of limestone and dolomite, which are applied to
soils (see Table 5-22), were multiplied by CO2 emission factors from West and McBride (2005). These country-
specific emission factors (0.059 metric ton C/metric ton limestone, 0.064 metric ton C/metric ton dolomite) are
lower than the IPCC default emission factors because they account for the portion of carbonates that are
transported from soils through hydrological processes and eventually deposited in ocean basins (West and
McBride 2005). This analysis of lime dissolution is based on studies in the Mississippi River basin, where the vast
majority of lime application occurs in the United States (West 2008). Moreover, much of the remaining lime
application is occurring under similar precipitation regimes, and so the emission factors are considered a
reasonable approximation for all lime application in the United States (West 2008) (see Box 5-5).
The annual application rates of limestone and dolomite were derived from estimates and industry statistics
provided in the U.S. Geological Survey (USGS) Minerals Yearbook (Tepordei 1994 through 2015; Willett 2007a,
2007b, 2009, 2010, 2011a, 2011b, 2013a, 2014, 2015, 2016, 2017, 2020a, 2022a, 2022b, 2022c, 2023a), as well as
preliminary data that will eventually be published in the Minerals Yearbook for the latter part of the time series
(Willett 2023b). Data for the final year of the inventory is based on the Mineral Industry Surveys, as discussed
below (USGS 2023). The U.S. Geological Survey (USGS; U.S. Bureau of Mines prior to 1997) compiled production
and use information through surveys of crushed stone manufacturers. However, manufacturers provided different
levels of detail in survey responses so the estimates of total crushed limestone and dolomite production and use
were divided into three components: (1) production by end-use, as reported by manufacturers (i.e., "specified"
production); (2) production reported by manufacturers without end-uses specified (i.e., "unspecified" production);
and (3) estimated additional production by manufacturers who did not respond to the survey (i.e., "estimated"
production).
Box 5-5: Comparison of the Tier 2 U.S. Inventory Approach and IPCC (2006) Default Approach
Emissions from liming of soils were estimated using a Tier 2 methodology based on emission factors specific to
the United States that are lower than the IPCC (2006) default emission factors. Most lime application in the
United States occurs in the Mississippi River basin, or in areas that have similar soil and rainfall regimes as the
Mississippi River basin. Under these conditions, a significant portion of dissolved agricultural lime leaches
through the soil into groundwater. Groundwater moves into channels and is transported to larger rivers and
eventually the ocean where CaCC>3 precipitates to the ocean floor (West and McBride 2005). The U.S.-specific
emission factors (0.059 metric ton C/metric ton limestone and 0.064 metric ton C/metric ton dolomite) are
about half of the IPCC (2006) emission factors (0.12 metric ton C/metric ton limestone and 0.13 metric ton
5-48 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
C/metric ton dolomite). For comparison, the 2022 U.S. emission estimate from liming of soils is 3.3 MMT CO2
Eq. using the country-specific factors. In contrast, emissions would be estimated at 6.6 MMT CO2 Eq. using the
IPCC (2006) default emission factors.
Data on "specified" limestone and dolomite amounts were used directly in the emission calculation because the
end use is provided by the manufacturers and can be used to directly determine the amount applied to soils.
However, it is not possible to determine directly how much of the limestone and dolomite is applied to soils for
manufacturer surveys in the "unspecified" and "estimated" categories. For these categories, the amounts of
crushed limestone and dolomite applied to soils were determined by multiplying the percentage of total
"specified" limestone and dolomite production that is applied to soils, by the total amounts of "unspecified" and
"estimated" limestone and dolomite production. In other words, the proportion of total "unspecified" and
"estimated" crushed limestone and dolomite that was applied to soils is proportional to the amount of total
"specified" crushed limestone and dolomite that was applied to soils.
In addition, data were not available for 1990,1992, and 2022 on the fractions of total crushed stone production
that were limestone and dolomite, and on the fractions of limestone and dolomite production that were applied to
soils. To estimate the 1990 and 1992 data, a set of average fractions were calculated using the 1991 and 1993
data. These average fractions were applied to the quantity of "total crushed stone produced or used" reported for
1990 and 1992 in the 1994 Minerals Yearbook (Tepordei 1996). To estimate 2022 data, 2021 fractions were applied
to the 2022 estimates of total crushed stone. The basis for these estimates is from the USGS Mineral Industry
Surveys: Crushed Stone and Sand and Gravel in the First Quarter of 2023 (USGS 2023).
The primary source for limestone and dolomite activity data is the Minerals Yearbook, published by the Bureau of
Mines through 1996 and by the USGS from 1997 to the present. In 1994, the "Crushed Stone" chapter in the
Minerals Yearbook began rounding (to the nearest thousand metric tons) quantities for total crushed stone
produced or used. It then reported revised (rounded) quantities for each of the years from 1990 to 1993. In order
to minimize the inconsistencies in the activity data, these revised production numbers have been used in all of the
subsequent calculations.
Table 5-22: Applied Minerals (MMT)
Mineral
1990 2005 I 2018
2019
2020
2021
2022
Limestone
19.0 18.1 9.4
8.9
11.6
9.3
13.5
Dolomite
2.4 1.9 | 0.9
1.2
1.6
1.6
1.5
The same methods are applied throughout the time series. The activity data are extended in the last two years of
the time series based on proportions of specified, unspecified and estimated agricultural limestone and dolomite
so that estimates are consistent with the previous year's data. These years will be recalculated when additional
data are available on the amounts of limestone and dolomite that are used for agricultural purposes.
Uncertainty
Uncertainty regarding the amount of limestone and dolomite applied to soils was estimated at ±15 percent with
normal densities (Tepordei 2003; Willett 2013b). Analysis of the uncertainty associated with the emission factors
included the fraction of lime dissolved by nitric acid versus the fraction that reacts with carbonic acid, and the
portion of bicarbonate that leaches through the soil and is transported to the ocean. Uncertainty regarding the
time associated with leaching and transport was not addressed in this analysis, but is assumed to be a relatively
small contributor to the overall uncertainty (West 2005). The probability distribution functions for the fraction of
lime dissolved by nitric acid and the portion of bicarbonate that leaches through the soil were represented as
triangular distributions between ranges of zero and 100 percent of the estimates. The uncertainty surrounding
these two components largely drives the overall uncertainty. The emission factor distributions were truncated at 0
so that emissions were not less than 0.
Agriculture 5-49
-------
A Monte Carlo (Approach 2) uncertainty analysis was applied to estimate the uncertainty in CO2 emissions from
liming. The results of the Approach 2 quantitative uncertainty analysis are summarized in Table 5-23. Carbon
dioxide emissions from carbonate lime application to soils in 2022 were estimated to be between 0.50 and 6.18
MMT CO2 Eq. at the 95 percent confidence level. This confidence interval represents a range of 85 percent below
to 89 percent above the 2022 emission estimate of 3.3 MMT CO2 Eq. Some carbon in the carbonate lime applied to
agricultural soils is not emitted to the atmosphere due to the dominance of the carbonate lime dissolving in
carbonic acid rather than nitric acid (West and McBride 2005).
Table 5-23: Approach 2 Quantitative Uncertainty Estimates for CO2 Emissions from Liming
(MMT CO2 Eq. and Percent)
2022 Emission Estimate
Uncertainty Range Relative to Emission Estimate-'
Source
Gas
(MMT CO . Eq.)
(MMTCO. Eq.)
(%)
Lower Upper
Bound Bound
Lower Upper
Bound Bound
Liming
C02
3.3
0.50 6.18
-85% +89%
a Range of emission estimates predicted by Monte Carlo stochastic simulation for a 95 percent confidence interval.
QA/QC and Verification
A source-specific QA/QC plan for liming has been developed and implemented, consistent with the U.S. Inventory
QA/QC plan outlined in Annex 8. The quality control effort focused on the Tier 1 procedures for this Inventory, and
no errors were identified in this Inventory.
Recalculations Discussion
Limestone and dolomite application data for 2020 and 2021 were updated with the recent published data from
Willett, J.C. (2023a). With these revisions, the emissions decreased by 1 and 22 percent for 2020 and 2021
(respectively) relative to the previous Inventory.
Planned Improvements
At this time there are no specific planned improvements for estimating emissions from liming.
5.6 Urea Fertilization (CRT Source Category
3H)
The use of urea (CO(NH2)2) as a fertilizer leads to greenhouse gas emissions through the release of C02that was
fixed during the production of urea. In the presence of water and urease enzymes, urea that is applied to soils as
fertilizer is converted into ammonium (NhV), hydroxyl ion (OH), and bicarbonate (HCO3 ). The bicarbonate then
evolves into CO2 and water. Emissions from urea fertilization in the United States were 5.3 MMT CO2 Eq. (1.5 MMT
C) in 2022 (Table 5-24 and Table 5-25). Carbon dioxide emissions have increased by 120 percent between 1990 and
2022 due to an increasing amount of urea that is applied to soils. The variation in emissions across the time series
is driven by differences in the amounts of fertilizer applied to soils each year. Carbon dioxide emissions associated
with urea used for non-agricultural purposes are reported in the IPPU chapter (Section 4.6).
5-50 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
Table 5-24: CO2 Emissions from Urea Fertilization (MMT CO2 Eq.)
Source 1990 2005 2018 2019 2020 2021
2022
Urea Fertilization 2.4 3.5 4.9 5.0 5.1 5.2
5.3
Table 5-25: CO2 Emissions from Urea Fertilization (MMT C)
Source 1990 2005 2018 2019 2020 2021
2022
Urea Fertilization 0.7 1.0 1.3 1.4 1.4 1.4
1.5
Methodology and Time-Series Consistency
Carbon dioxide emissions from the application of urea to agricultural soils were estimated using the IPCC (2006)
Tier 1 methodology following the 2006 IPCC Guidelines Figure 11.5 decision tree for CO2 emissions from urea
fertilization.29 The method assumes that carbon in the urea is released after application to soils and converted to
CO2. The annual amounts of urea applied to croplands (see Table 5-26) were derived from the state-level fertilizer
sales data provided in Commercial Fertilizer reports (TVA 1991,1992,1993,1994; AAPFCO 1995 through 2022).30
These amounts were multiplied by the default IPCC (2006) emission factor (0.20 metric tons of carbon per metric
ton of urea), which is equal to the carbon content of urea on an atomic weight basis. National estimates from urea
fertilization also include emissions from Puerto Rico.
Fertilizer sales data are reported in fertilizer years (July previous year through June current year), so a calculation
was performed to convert the data to calendar years (January through December). According to monthly fertilizer
use data (TVA 1992b), 35 percent of total fertilizer used in any fertilizer year is applied between July and December
of the previous calendar year, and 65 percent is applied between January and June of the current calendar year.
Fertilizer sales data for the 2018 through 2022 fertilizer years were not available for this Inventory. Therefore, urea
application in the 2018 through 2022 fertilizer years were estimated using a linear, least squares trend of
consumption over the data from the previous five years (2013 through 2017) at the state scale. A trend of five
years was chosen as opposed to a longer trend as it best captures the current inter-annual variability in
consumption. State-level estimates of CO2 emissions from the application of urea to agricultural soils were
summed to estimate total emissions for the entire United States. The fertilizer year data is then converted into
calendar year (Table 5-26) data using the method described above.
Table 5-26: Applied Urea (MMT)
1990
2005
2018
2019
2020
2021
2022
Urea Fertilizer3
CO
CO
CO
'sT
6.7
6.9
7.0
7.1
7.3
aThese 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, land converted to 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/conversion category.
The same methods were applied to the entire time series to ensure time-series consistency from 1990 through
2022. In addition, activity data are extended using a data splicing method with a linear extrapolation based on the
last five years of urea fertilization data to ensure consistency in the time series. These years will be recalculated
when additional data are available on urea fertilization.
29 2006 IPCC Guidelines Volume 4, Chapter 11, Figure 11.5 (page 11.33)
30 The amount of urea consumed for non-agricultural purposes in the United States is reported in the Industrial Processes and
Product Use chapter, Section 4.6 Urea Consumption for Non-Agricultural Purposes.
Agriculture 5-51
-------
Uncertainty
An Approach 2 Monte Carlo analysis is conducted as described by the IPCC (2006). The largest source of
uncertainty is the default emission factor, which assumes that 100 percent of the carbon in CO(NH2)2 applied to
soils is emitted as CO2. The uncertainty surrounding this factor incorporates the possibility that some of the carbon
may not be emitted to the atmosphere, and therefore the uncertainty range is set from 50 percent emissions to
the maximum emission value of 100 percent using a triangular distribution. In addition, urea consumption data
have uncertainty that is represented as a normal density. Due to the highly skewed distribution of the resulting
emissions from the Monte Carlo uncertainty analysis, the estimated emissions are based on the analytical solution
to the equation, and the confidence interval is approximated based on the values at 2.5 and 97.5 percentiles.
Carbon dioxide emissions from urea fertilization of agricultural soils in 2022 are estimated to be between 3.05 and
5.49 MMT CO2 Eq. at the 95 percent confidence level. This indicates a range of 43 percent below to 3 percent
above the 2022 emission estimate of 5.33 MMT CO2 Eq. (Table 5-27).
Table 5-27: Quantitative Uncertainty Estimates for CO2 Emissions from Urea Fertilization
(MMT CO2 Eq. and Percent)
Source
Gas
2022 Emission Estimate
(MMT CO . Eq.)
Uncertainty Range Relative to Emission Estimate-1
(MMTCO' Eq.) (%)
Lower Upper
Bound Bound
Lower Upper
Bound Bound
Urea Fertilization
C02
5.33
3.05 5.49
-43% +3%
a Range of emission estimates predicted by Monte Carlo stochastic simulation for a 95 percent confidence interval.
There are additional uncertainties that are not quantified in this analysis. There is uncertainty surrounding the
assumptions underlying conversion of fertilizer years to calendar years. These uncertainties are negligible over
multiple years because an over- or under-estimated value in one calendar year is addressed with a corresponding
increase or decrease in the value for the subsequent year. In addition, there is uncertainty regarding the fate of
carbon in urea that is incorporated into solutions of urea ammonium nitrate (UAN) fertilizer. Emissions of CO2 from
UAN applications to soils are not estimated in the current Inventory (see Planned Improvements).
QA/QC and Verification
A source-specific QA/QC plan for Urea Fertilization has been developed and implemented, consistent with the U.S.
Inventory QA/QC plan. No quality control problems were discovered in this process except a correction to the
emissions factor value in documentation tables.
Recalculations Discussion
Fertilizer consumption data was updated with the latest published estimate. In turn, the fertilizer values were
recalculated using the data splicing method for 2018 to 2021 based on the revised fertilizer amount for 2017. This
update led to an average decrease in emissions for the years 2017 through 2021 of 0.01 MMT CO2 Eq., or 0.1
percent. The remainder of the time series was not affected.
Planned Improvements
A key planned improvement is to incorporate Urea Ammonium Nitrate (UAN) in the estimation of Urea CO2
emissions. Activity data for UAN have been identified, but additional information is needed to fully incorporate this
type of fertilizer into the analysis, which will be completed in a future Inventory.
5-52 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
5.7 Field Burning of Agricultural Residues
(CRT 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. The 2006IPCC Guidelines does not consider field burning of
crop residues to be a net source of CO2 emissions because it is assumed the carbon released to the atmosphere as
CO2 during burning is reabsorbed during the next growing season by the crop (IPCC 2006). However, crop residue
burning is a net source of Cm, N2O, CO, and NOx, which are released during combustion.
In the United States, field burning of agricultural residues 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). In 2022, Cm and N2O emissions from field burning of
agricultural residues were 0.6 MMT CO2 Eq. (22 kt) and 0.2 MMT CO2 Eq. (1 kt), respectively (Table 5-28 and Table
5-29). Annual emissions of Cm and N2O have increased from 1990 to 2022 by 14 percent and 16 percent,
respectively. The increase in emissions over time is partly due to higher yielding crop varieties with larger amounts
of residue production and fuel loads, but also linked with an increase in the area burned for some crop types.
Table 5-28: CH4 and N2O Emissions from Field Burning of Agricultural Residues (MMT CO2 Eq.)
Gas/Crop Type
1990
2005
2018
2019
2020
2021
2022
ch4
0.5
0.6
0.6
0.7
0.6
0.6
0.6
Sugarcane
0.1 III-
0.2 ,
0.1
0.2
0.1
0.1
0.1
Wheat
0.2
0.2
0.1
0.1
0.1
0.1
0.1
Maize
o-i 1
0.11
0.1
0.1
0.1
0.1
0.1
Rice
0.1
0.1
0.1
0.1
0.1
0.1
0.1
Soybeans
+ III
+ 1
+
+
+
+
+
Cotton
+
+
+
+
+
+
+
Sorghum
+ III
+ i
+
+
+
+
+
Other Small Grains
+
+
+
+
+
+
+
Peanuts
+
+
+
+
+
+
Legume Hay
+
+
+
+
+
+
+
Barley
+1
+ II
+
+
+
+
+
Oats
+
+
+
+
+
+
+
Grass Hay
1111111
+ B
+ 5
+
+
+
+
+
Tobacco
+
+
+
+
+
+
+
Vegetables
0.0 !!:
+ 1
+
+
+
+
+
Peas
+
+
+
+
+
+
+
Sunflower
+ =:
+
+
+
+
+
Potatoes
+
+
+
+
+
+
+
Dry Beans
+
+ !
!!!!!!!
+
+
+
+
+
Sugarbeets
+
!!!!!!:
+
+
+
+
+
+
Lentils
0.0
+ 111
+
+
+
+
+
Chickpeas
0.0
0.0
0.0
0.0
0.0
0.0
0.0
n2o
0.2
0.2
0.2
0.2
0.2
0.2
0.2
Wheat
0.1
0.1
+
+
+
+
+
Maize
iiiiiii
+
¦
+ I
+
+
+
+
+
Sugarcane
+
+
+
+
+
+
+
Rice
+ 1
+ E
+
+
+
+
+
Soybeans
+
+
+
+
+
+
+
Cotton
+ i
+1
+
+
+
+
+
Agriculture 5-53
-------
Peanuts
+
+
+
+
+
+
+
Other Small Grains
+ :
+ ;
+
+
+
+
+
Legume Hay
+
+
+
+
+
+
+
Sorghum
+ ¦
+
+
+
+
+
Grass Hay
+
+
+
+
+
+
+
Barley
+ ,
+1
+
+
+
+
+
Oats
+
+
+
+
+
+
+
Potatoes
+
mm:
+
+
+
+
+
Peas
+
+
+
+
+
+
+
Sugarbeets
mm!
+
+ =
+
+
+
+
+
Tobacco
+
+
+
+
+
+
+
Sunflower
+ !!!
+!!!
+
+
+
+
+
Vegetables
0.0
+
+
+
+
+
+
Dry Beans
+
+ i
+
+
+
+
+
Lentils
0.0
+
+
+
+
+
+
Chickpeas
O
O
0.01
0.0
0.0
0.0
0.0
0.0
Total
0.7
0.8
0.8
0.9
0.8
0.8
0.8
+ Does not exceed 0.05 MMT C02 Eq.
Note: Totals may not sum due to independent rounding.
Table 5-29: CH4, N2O, CO, and NOx Emissions from Field Burning of Agricultural Residues (kt)
Gas/Crop Type
1990
2005
2018
2019
2020
2021
2022
ch4
19
23
22
23
22
22
22
Sugarcane
4:::
6 1
5
6
5
5
5
Wheat
6
6
5
5
5
5
5
Maize
2 I
4
5
5
5
5
5
Rice
3
3
2
3
2
3
3
Soybeans
1 I
2 '
2
2
2
2
2
Cotton
1
2
1
1
1
1
1
Sorghum
+
¦
+
¦
+
+
+
+
+
Other Small Grains
+
+
+
+
+
+
+
Peanuts
+
+ i
+
+
+
+
+
Legume Hay
+
+
+
+
+
+
+
Barley
+1
+
+
+
+
+
Oats
+
+
+
+
+
+
+
Grass Hay
+ 1
+
+
+
+
+
Tobacco
+
+
+
+
+
+
+
Vegetables
0::::::
+,
+
+
+
+
+
Peas
+
+
+
+
+
+
+
Sunflower
11111b!
+
+
+
+
+
+
+
Potatoes
+
+
+
+
+
+
+
Dry Beans
+
+¦
+
+
+
+
+
Sugarbeets
+
+
+
+
+
+
+
Lentils
+
+
+
+
+
Chickpeas
0
0
0
0
0
0
0
N20
1
1
1
1
1
1
1
Wheat
+
+
+
+
+
+
+
Maize
+
+=
+
+
+
+
+
Sugarcane
+
+
+
+
+
+
+
Rice
+ :
+!
+
+
+
+
+
Soybeans
+
+
+
+
+
+
+
Cotton
+
+ |
+
+
+
+
+
Peanuts
+
+
+
+
+
+
+
Other Small Grains
+
+
+
+
+
+
Legume Hay
+
+
+
+
+
+
+
Sorghum
+
_l_ ¦
+
+
+
+
+
5-54 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
Grass Hay
+
+
+
+
+
+
+
Barley
+
+ ;
+
+
+
+
+
Oats
+
+
+
+
+
+
+
Potatoes
~:
+
+
+
+
+
Peas
+
+
+
+
+
+
+
Sugarbeets
+1
+
+
+
+
+
Tobacco
+
+
+
+
+
+
+
Sunflower
+ ;;;
¦
+
+
+
+
+
+
Vegetables
0
+
+
+
+
+
+
Dry Beans
+ j
+
+
+
+
+
Lentils
0
+
+
+
+
+
+
Chickpeas
o IE
o I
0
0
0
0
0
CO
407
480
433
468
446
480
501
NOx
16
18
17
18
17
18
19
+ Does not exceed 0.5 kt C02 Eq.
Note: Totals may not sum due to independent rounding.
Methodology and Time-Series Consistency
A country-specific Tier 2 method is used to estimate greenhouse gas emissions from field burning of agricultural
residues from 1990 to 2014 (for more details comparing the country-specific approach to the IPCC (2006) default
approach, see Box 5-6), and a data splicing method with a linear extrapolation is applied to complete the emissions
time series from 2015 to 2022. The exception is sugarcane for which emissions have been estimated from 1990 to
2020, with 2021 to 2022 estimated with the data splicing method. The following equation is used to estimate the
amounts of carbon and nitrogen released (R/, where / is C or N) from burning.
Equation 5-1: Elemental C or N Released through Oxidation of Crop Residues
R, = CP x RCR x DMF x F, x FB x CE
FB =
AB
CAH
where,
Crop Production (CP)
Residue: Crop Ratio (RCR)
Dry Matter Fraction (DMF)
Fraction C or N (Ft)
Fraction Burned (FB)
Combustion Efficiency (CE)
Area Burned (AB)
Crop Area Harvested (CAH)
= Annual production of crop, by state, kt crop production
= Amount of residue produced per unit of crop production, kt residue/kt crop
production
= Amount of dry matter per unit of residue biomass for a crop, kt residue dry
matter/ kt residue biomass
= Fraction of C or N per unit of dry matter for a crop, kt C or N /kt residue dry
matter
= Proportion of residue biomass consumed, unitless
= Proportion of residue actually burned, unitless
= Total area of crop burned, by state, ha
= Total area of crop harvested, by state, ha
Crop production data are available by state and year from USDA-NASS (2019) for 22 crops that are burned in the
conterminous United States, including maize, rice, wheat, barley, oats, other small grains, sorghum, cotton, grass
hay, legume hay, peas, sunflower, tobacco, vegetables, chickpeas, dry beans, lentils, peanuts, soybeans, potatoes,
Agriculture 5-55
-------
sugarbeets, and sugarcane.31 Crop area data are based on the 2015 and 2017 National Resources Inventories (NRI)
(USDA-NRCS 2018; USDA-NRCS 2020). To estimate total crop production, the crop yield data from USDA Quick
Stats (USDA-NASS 2019) are multiplied by the area data for these crops from the NRI survey. The production data
for the crop types are presented in Table 5-30. Alaska and Hawaii are not included in the current analysis, but
there is a planned improvement to estimate residue burning emissions for these two states in a future Inventory.
The amount of elemental carbon or nitrogen released through oxidation of the crop residues is used in the
following equation to estimate the amount of Cm, CO, N2O, and NOx emissions (Eg, where g is the specific gas, i.e.,
Cm, CO, N2O, and NOx) from the field burning of agricultural residues:
Equation 5-2: Emissions from Crop Residue Burning
Eg = KjX EFg X CF
where,
Emission ratio (EFg) = emission ratio by gas, g CH4-C or CO-C/g C released, or g N2O-N or NOx-N/g N
released
Conversion Factor (CF) = conversion by molecular weight ratio of CH4-C to C (16/12), CO-C to C
(28/12), N2O-N to N (44/28), or NOx-N to N (30/14)
Box 5-6: Comparison of Tier 2 U.S. Inventory Approach and IPCC (2006) Default Approach
Emissions from Field Burning of Agricultural Residues are calculated using a Tier 2 methodology that is based on
the method developed by the IPCC/UNEP/OECD/IEA (1997). The rationale for using the IPCC/UNEP/OECD/IEA
(1997) approach rather than the method provided in the 2006 IPCC Guidelines is as follows: (1) the equations
from both guidelines rely on the same underlying variables (though the formats differ); (2) the IPCC (2006)
equation was developed to be broadly applicable to all types of biomass burning, and, thus, is not specific to
agricultural residues; (3) the IPCC (2006) method provides emission factors based on the dry matter content
rather than emission rates related to the amount of carbon and nitrogen in the residues; and (4) the IPCC (2006)
default factors are provided only for four crops (corn, rice, sugarcane, and wheat) while this Inventory includes
emissions from twenty-one crops.
A comparison of the methods in the current Inventory and the default IPCC (2006) approach was undertaken for
2014 to determine the difference in estimates between the two approaches. To estimate greenhouse gas
emissions from field burning of agricultural residues using the IPCC (2006) methodology, the following
equation—cf. IPCC (2006) Equation 2.27—was used with default factors and country-specific values for mass of
fuel.
Equation 5-3: Estimation of Greenhouse Gas Emissions from Fire
Emissions (kt) = AB x MB x Cf x Gef x 10~6
where,
Area Burned (AB) = Total area of crop burned (ha)
Kentucky bluegrass (produced on farms for turf grass installations) may have small areas of burning that are not captured in
the sample of locations that were used in the remote sensing analysis (see Planned Improvements).
5-56 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
Mass of Fuel (M )
= U.S.- Specific Values using NASS Statistics v (metric tons dry matter)
Combustion Factor (C ) = IPCC (2006) default combustion factor with fuel biomass consumption
(metric tons dry matter ha )
Emission Factor (G ) = IPCC (2006) emission factor (g kg dry matter burnt)
The IPCC (2006) Tier 1 method approach resulted in 21 percent lower emissions of CH and 40 percent lower
emissions of N O compared to this Inventory. In summary, the IPCC/UNEP/OECD/IEA (1997) method is
considered more appropriate for U.S. conditions because it is more flexible for incorporating country-specific
data. Emissions are estimated based on specific carbon and nitrogen content of the fuel, which is converted into
CH , CO, N O and NO., compared to IPCC (2006) approach that is based on dry matter rather than elemental
composition.
Emissions from field burning of agricultural residues are calculated using a Tier 2 methodology that is based on
the method developed by the IPCC/UNEP/OECD/IEA (1997). The rationale for using the IPCC/UNEP/OECD/IEA
(1997) approach rather than the method provided in the 2006 IPCC Guidelines is as follows: (1) the equations
from both guidelines rely on the same underlying variables (though the formats differ); (2) the IPCC (2006)
equation was developed to be broadly applicable to all types of biomass burning, and, thus, is not specific to
agricultural residues; (3) the IPCC (2006) method provides emission factors based on the dry matter content
rather than emission rates related to the amount of carbon and nitrogen in the residues; and (4) the IPCC (2006)
default factors are provided only for four crops (corn, rice, sugarcane, and wheat) while this Inventory includes
emissions from 21 crops.
A comparison of the methods in the current Inventory and the default IPCC (2006) approach was undertaken for
2014 to determine the difference in estimates between the two approaches. To estimate greenhouse gas
emissions from field burning of agricultural residues using the IPCC (2006) methodology, the following
equation—cf. IPCC (2006) Equation 2.27—was used with default factors and country-specific values for mass of
fuel.
Equation 5-4: Estimation of Greenhouse Gas Emissions from Fire
Combustion Factor (C ) = IPCC (2006) default combustion factor with fuel biomass consumption
(metric tons dry matter ha )
Emission Factor (G ) = IPCC (2006) emission factor (g kg dry matter burnt)
The IPCC (2006) Tier 1 method approach resulted in 21 percent lower emissions of CH and 40 percent lower
emissions of N O compared to this Inventory. In summary, the IPCC/UNEP/OECD/IEA (1997) method is
considered more appropriate for U.S. conditions because it is more flexible for incorporating country-specific
data. Emissions are estimated based on specific carbon and nitrogen content of the fuel, which is converted into
32 NASS yields are used to derive mass of fuel values because IPCC (2006) only provides default values for 4 of the 21 crops
included in the Inventory.
33 NASS yields are used to derive mass of fuel values because IPCC (2006) only provides default values for 4 of the 21 crops
included in the Inventory.
I-missions (In) = Mi x A/,., x C, x C, x 10 "
where,
Area Burned (AB)
Mass of Fuel (M )
Total area of crop burned (ha)
U.S.- Specific Values using NASS Statistics 11 (metric tons dry matter)
Agriculture 5-57
-------
CH , CO, N O and NO., compared to IPCC (2006) approach that is based on dry matter rather than elemental
composition.
Table 5-30: Agricultural Crop Production (kt of Product)
Crop
1990
2005
2010
2018
2019
2020
Maize
296,065
371,256
398,618
NE
NE
NE
Rice
9,543
1
1
11,751
11,976 :
NE
NE
NE
Wheat
79,805
68,077
68,530
NE
NE
NE
Barley
9,281
I
1
5,161 1
3,942 555
NE
NE
NE
Oats
5,969
2,646
2,364
NE
NE
NE
Other Small Grains
2,651
I
2,051
1'803 :
NE
NE
NE
Sorghum
23,687
14,382
14,052
NE
NE
NE
Cotton
4,605
I
1
6,106
4,638 i
NE
NE
NE
Grass Hay
44,150
49,880
46,761
NE
NE
NE
Legume Hay
90,360
91,819 :
85,813 I
NE
NE
NE
Peas
51
660
839
NE
NE
NE
Sunflower
1,015
1
I
1,448
1,212
NE
NE
NE
Tobacco
1,154
337
470
NE
NE
NE
Vegetables
+
IIP
1,187 iiii:
1,469 1
NE
NE
NE
Chickpeas
+
5 '
+
NE
NE
NE
Dry Beans
467
I
1/143 1
1,461
NE
NE
NE
Lentils
+
101
254
NE
NE
NE
Peanuts
1,856
)
2,176 :
1,925
NE
NE
NE
Soybeans
56,612
86,980
95,198
NE
NE
NE
Potatoes
18,924
1
1
20,026
25,635
19,279 1
NE
NE
NE
Sugarbeets
24,951
I*
33,336
NE
NE
NE
Sugarcane
26,047
Jill
38,928
34,252 I
36,680
37,361
42,400
+ Absolute value does not exceed 0.05 MMT C02 Eq.
NE (Not Estimated)
Note: The amount of crop production has not been compiled for 2015 to 2021 so a data splicing
method is used to estimate emissions for this portion of the time series.
The area burned is determined based on an analysis of remote sensing products (McCarty et al. 2009, 2010, 2011).
The presence of fires has been analyzed at 3,600 survey locations in the NRI from 1990 to 2002 with LANDFIRE
data products developed from 30 m Landsat imagery (LANDFIRE 2008), and from 2003 through 2014 using 1 km
Moderate Resolution Imaging Spectroradiometer imagery (MODIS) Global Fire Location Product (MCD14ML),
combining observations from Terra and Aqua satellites (Giglio et al. 2006). A sample of states are included in the
analysis with high, medium and low burning rates for agricultural residues, including Arkansas, California, Florida,
Indiana, Iowa and Washington. The area burned is determined directly from the analysis for these states for all
crops, with the exception of sugarcane as discussed later in this section.
For other states within the conterminous United States, the area burned for the 1990 through 2014 portion of the
time series is estimated from a logistical regression model that has been developed from the data collected from
the remote sensing products for the six states. The logistical regression model is used to predict occurrence of fire
events. Several variables are tested in the logistical regression including a) the historical level of burning in each
state (high, medium or low levels of burning) based on an analysis by McCarty et al. (2011), b) year that state laws
limit burning of fields, in addition to c) mean annual precipitation and mean annual temperature from a 4-
kilometer gridded product from the PRISM Climate Group (2015). A K-fold model fitting procedure is used due to
low frequency of burning and likelihood that outliers could influence the model fit. Specifically, the model is
trained with a random selection of sample locations and evaluated with the remaining sample. This process is
repeated ten times to select a model that is most common among the set of ten, and avoid models that appear to
be influenced by outliers due to the random draw of survey locations for training the model. In order to address
5-58 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
uncertainty, a Monte Carlo analysis is used to sample the parameter estimates for the logistical regression model
and produce one thousand estimates of burning for each crop in the remaining forty-two states included in this
Inventory. State-level area burned data are divided by state-level crop area data to estimate the percent of crop
area burned by crop type for each state. Table 5-31 shows the resulting percentage of crop residue burned at the
national scale by crop type. State-level estimates are also available upon request.
Table 5-31: U.S. Average Percent Crop Area Burned by Crop (Percent)
Crop
1990
2005
2010
2018
2019
2020
Maize
+
+
+
NE
NE
NE
Rice
12% I
11% :
1.2% S
NE
NE
NE
Wheat
3%
3%
2%
NE
NE
NE
Barley
1% III!!
1% 1
5
NE
NE
NE
Oats
1%
1%
1%
NE
NE
NE
Other Small Grains
5%
4%
4% 1
NE
NE
NE
Sorghum
i%
1%
1%
NE
NE
NE
Cotton
7% :
10% i
9% 1
NE
NE
NE
Grass Hay
+
+
+
NE
NE
NE
Legume Hay
+
+ I
NE
NE
NE
Peas
1%
1%
1%
NE
NE
NE
Sunflower
+;;
+
+ 111
NE
NE
NE
Tobacco
i%
1%
1%
NE
NE
NE
Vegetables
IB!
+ ::
+
!
NE
NE
NE
Chickpeas
+
+
+
NE
NE
NE
Dry Beans
+ I
+ :
+ ¦
NE
NE
NE
Lentils
+
1%
1%
NE
NE
NE
Peanuts
5% 1
5%:
NE
NE
NE
Soybeans
1%
i%
1%
NE
NE
NE
Potatoes
+ ¦
+1
+ -
NE
NE
NE
Sugarbeets
+
+
+
NE
NE
NE
Sugarcane
6%:
5% !
6%
4%
6%
4%
+ Does not exceed 0.5 percent.
NE (Not Estimated)
The method for estimating burned area of sugarcane is similar to the approach for other crops. Areas with
sugarcane production are identified in the 2017 USDA NRI survey (USDA-NRCS 2020) based on Cropland Data Layer
(USDA-NASS 2021).34 We use the MODIS burned area product from 2002 to 2020 to identify NRI survey locations
with sugarcane production that have residue burning, similar to the process for other crops described above
(Giglio et al. 2015). However, area of residue burning for sugarcane was estimated for 1990 to 2001 using a linear
extrapolation of the area burned from 2002 to 2020, instead of analyzing the remote sensing data for this portion
of the time series. This approach is a common data splicing method for filling data gaps in time series (IPCC 2006).
Additional parameters are needed to estimate emissions from the area that has residue burning, including residue:
crop ratios, dry matter fractions, carbon fractions, nitrogen fractions and combustion efficiency. Residue: crop
product mass ratios, residue dry matter fractions, and the residue N contents are obtained from several sources
(IPCC 2006 and sources at bottom of Table 5-32). The residue carbon contents for all crops are based on IPCC
(2006) default value for herbaceous biomass. The combustion efficiency is assumed to be 90 percent for all crop
types (IPCC/UNEP/OECD/IEA 1997). See Table 5-32 for a summary of the crop-specific conversion factors. Emission
ratios and mole ratio conversion factors for all gases are based on the Revised 1996 IPCC Guidelines
(IPCC/UNEP/OECD/IEA 1997) (see Table 5-33).
34 USDA-NRI program aggregates sugarcane with other crops, but areas planted with sugarcane are identified in
the USDA-NASS Crop Data Layer.
Agriculture 5-59
-------
Table 5-32: Parameters for Estimating Emissions from Field Burning of Agricultural Residues
Crop
Residue/Crop
Ratio
Dry Matter
Fraction
Carbon Fraction
Nitrogen Fraction
Combustion
Efficiency
(Fraction)
Maize
0.707
0.56
0.47
0.01
0.90
Rice
1.340
0.89
0.47
0.01
0.90
Wheat
1.725
0.89
0.47
0.01
0.90
Barley
1.181
0.89
0.47
0.01
0.90
Oats
1.374
0.89
0.47
0.01
0.90
Other Small Grains
1.777
0.88
0.47
0.01
0.90
Sorghum
0.780
0.60
0.47
0.01
0.90
Cotton
7.443
0.93
0.47
0.01
0.90
Grass Hay
0.208
0.90
0.47
0.02
0.90
Legume Hay
0.290
0.67
0.47
0.01
0.90
Peas
1.677
0.91
0.47
0.01
0.90
Sunflower
1.765
0.88
0.47
0.01
0.90
Tobacco
0.300
0.87
0.47
0.01
0.90
Vegetables
0.708
0.08
0.47
0.01
0.90
Chickpeas
1.588
0.91
0.47
0.01
0.90
Dry Beans
0.771
0.90
0.47
0.01
0.90
Lentils
1.837
0.91
0.47
0.02
0.90
Peanuts
1.600
0.94
0.47
0.02
0.90
Soybeans
1.500
0.91
0.47
0.01
0.90
Potatoes
0.379
0.25
0.47
0.02
0.90
Sugarbeets
0.196
0.22
0.47
0.02
0.90
Sugarcane
0.410
0.25
0.47
0.02
0.90
NE (Not Estimated)
Notes: Chickpeas: IPCC (2006), Table 11.2; values are for Beans & pulses.
Cotton: Combined sources (Heitholt et al. 1992; Halevy 1976; Wells and Meredith 1984; Sadras and Wilson 1997; Pettigrew
and Meredith 1997; Torbert and Reeves 1994; Gerik et al. 1996; Brouder and Cassmen 1990; Fritschi et al. 2003; Pettigrew et
al. 2005; Bouquet and Breitenbeck 2000; Mahroni and Aharonov 1964; Bange and Milroy 2004; Hollifield et al. 2000;
Mondino et al. 2004; Wallach et al. 1978).
Lentils: IPCC (2006), Table 11.2; Beans & pulses.
Peas: IPCC (2006), Table 11.2; values are for Beans & pulses.
Peanuts: IPCC (2006); Table 11.2; Root ratio and belowground N content values are for Root crops, other.
Sugarbeets: IPCC (2006); Table 11.2; values are forTubers.
Sunflower: IPCC (2006), Table 11.2; values are for Grains.
Sugarcane: combined sources (Wiedenfels 2000, Dua and Sharma 1976; Singels & Bezuidenhout 2002; Stirling et al. 1999;
Sitompul et al. 2000).
Tobacco: combined sources (Beyaert 1996; Moustakas and Ntzanis 2005; Crafts-Brandner et al. 1994; Hopkinson 1967; Crafts-
Brandneretal. 1987).
Vegetables (Combination of carrots, lettuce/cabbage, melons, onions, peppers and tomatoes):
Carrots: McPharlin et al. (1992); Gibberd et al. (2003); Reid and English (2000); Peach et al. (2000); see IPCC Tubers for R:S and
N fraction.
Lettuce, cabbage: combined sources (Huett and Dettman 1991; De Pinheiro Henriques & Marcelis 2000; Huett and Dettman
1989; Peach et al. 2000; Kage et al. 2003; Tan et al. 1999; Kumar et al. 1994; MacLeod et al. 1971; Jacobs et al. 2004; Jacobs
et al. 2001; Jacobs et al. 2002); values from IPCC Grains used for N fraction.
Melons: Valantin et al. (1999); squash for R:S; IPCC Grains for N fraction.
Onion: Peach et al. (2000), Halvorson et al. (2002); IPCC (2006) Tubers for N fraction.
Peppers: combined sources (Costa and Gianquinto 2002; Marcussi et al. 2004; Tadesse et al. 1999; Diaz-Perez et al. 2008);
IPCC Grains for N fraction.
Tomatoes: Scholberg et al. (2000a,b); Akintoye et al. (2005); values for AGR-N and BGR-N are from Grains.
5-60 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
Table 5-33: Greenhouse Gas Emission Ratios and Conversion Factors
Gas
Emission Ratio
Conversion Factor
CH4:C
0.005a
16/12
CO:C
0.060a
28/12
N20:N
0.007b
44/28
NOx:N
0.121b
30/14
a Mass of C compound released (units of C) relative to
mass of total C released from burning (units of C).
b Mass of N compound released (units of N) relative to
mass of total N released from burning (units of N).
To ensure time-series consistency, the same method is applied from 1990 to 2014 for all crops except sugarcane in
which the method was applied for 1990 to 2020. For this Inventory, new activity data on the burned areas have not
been analyzed for 2015 to 2022 for individual crops. The exception is sugarcane in which burned areas have not
been analyzed for 2021 to 2022. To complete the emissions time series, a linear extrapolation of the trend is
applied to estimate the emissions for the latter part of the time series. Specifically, a linear regression model with
autoregressive moving-average (ARMA) errors is used to estimate the trend in emissions over time from 1990
through 2014, and the trend is used to approximate the Cm, N2O, CO and NOx from 2015 to 2022 for all crops
except for sugarcane, which was estimated using this method for 2021 to 2022 (Brockwell and Davis 2016). This
extrapolation method is consistent with data splicing methods in IPCC (2006). The Tier 2 method described
previously will be applied to recalculate the emissions for the latter part of the time series in a future Inventory.
Uncertainty
Emissions are estimated using a linear regression model with autoregressive moving-average (ARMA) errors for
2022. The linear regression ARMA model produced estimates of the upper and lower bounds to quantify
uncertainty, and the results are summarized in Table 5-34. Methane emissions from field burning of agricultural
residues in 2022 are between 0.55 and 0.70 MMT CO2 Eq. at a 95 percent confidence level. This indicates a range
of 11 percent below and 11 percent above the 2022 emission estimate of 0.6 MMT CO2 Eq. Nitrous oxide emissions
are between 0.18 and 0.23 MMT CO2 Eq., or approximately 13 percent below and 13 percent above the 2022
emission estimate of 0.2 MMT CO2 Eq.
Table 5-34: Approach 2 Quantitative Uncertainty Estimates for CH4 and N2O Emissions from
Field Burning of Agricultural Residues (MMT CO2 Eq. and Percent)
Source
Gas
2022 Emission Estimate
(MMTCO. Eq.)
Uncertainty Range Relative to Emission Estimate-1
(MMTCO' Eq.) (%)
Lower
Upper
Lower
Upper
Bound
Bound
Bound
Bound
Field Burning of Agricultural
Residues
ch4
0.6
0.55
0.70
-11%
+11%
Field Burning of Agricultural
Residues
n2o
0.2
0.18
0.23
-13%
+13%
a Range of emission estimates predicted by Monte Carlo stochastic simulation for a 95 percent confidence interval.
Due to data limitations, there are additional uncertainties in agricultural residue burning, particularly the potential
omission of burning associated with Kentucky bluegrass (produced on farms for turf grass installation).
/erification
A source-specific QA/QC plan for field burning of agricultural residues is implemented with Tier 1 analyses,
consistent with the U.S. Inventory QA/QC plan outlined in Annex 8. Quality control measures included checking
input data, model scripts, and results to ensure data are properly handled throughout the inventory process.
Agriculture 5-61
-------
Inventory reporting forms and text are reviewed and revised as needed to correct transcription errors. An error
was identified in the calculation of the emissions using the IPCC (2006) equation after the initial compilation, which
was corrected in Box 5.6. An error was also found with the estimation of non-CC>2 emissions from burning of
sugarcane residue related to the GWP factors. This error was corrected.
Recalculations Discussion
Recalculations have been conducted for this Inventory associated with the addition of residue burning from
sugarcane, which was not included in the previous Inventory. As a result of this change, Cm emissions increased by
an annual average of 0.14 MMT CO2 Eq., or 32 percent, over the time series from 1990 to 2021 compared to the
previous Inventory. In addition, N2O emissions increased by an annual average of 0.03 MMT CO2 Eq., or 21 percent,
over the time series from 1990 to 2021 compared to the previous Inventory.
Planned Improvements
A key planned improvement is to estimate the emissions associated with field burning of agricultural residues in
the states of Alaska and Hawaii. In addition, a method is in development that will directly link agricultural residue
burning with the Tier 3 methods that are used in several other source categories, including agricultural soil
management, cropland remaining cropland, and land converted to cropland chapters of the Inventory. The method
is based on simulating burning events directly within the DayCent process-based model framework using
information derived from remote sensing fire products as described in the Methodology section. This
improvement will lead to greater consistency in the methods across sources, ensuring mass balance of carbon and
nitrogen in the Inventory analysis.
5-62 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
Forestry
This chapter provides an assessment of the greenhouse gas fluxes resulting from land use and land-use change in
the United States.1 The Intergovernmental Panel on Climate Change's 2006IPCC Guidelines for National
Greenhouse Gas Inventories (IPCC 2006) recommends reporting fluxes according to changes within and
conversions between all land use types including: forest land, cropland, grassland, wetlands, and settlements (as
well as other land).
The greenhouse gas flux from forest land remaining forest land is reported for all forest ecosystem carbon (C)
pools (i.e., aboveground biomass, belowground biomass, dead wood, litter, and mineral and organic soils),
harvested wood pools, and non-carbon dioxide (non-CCh) emissions from forest fires, the application of synthetic
nitrogen fertilizers to forest soils, and the draining of organic soils. Fluxes from land converted to forest land are
included for aboveground biomass, belowground biomass, dead wood, litter, and carbon stock changes from
mineral soils, while carbon stock changes from drained organic soils and all non-CC>2 emissions from land
converted to forest land are included in the fluxes from forest land remaining forest land as it is not currently
possible to separate these fluxes by conversion category (e.g., grassland converted to forestland).
Fluxes are reported for four agricultural land use/land-use change categories: cropland remaining cropland, land
converted to cropland, grassland remaining grassland, and land converted to grassland. The reported greenhouse
gas fluxes from these agricultural lands include changes in soil organic carbon stocks in mineral and organic soils
due to land use and management, and for the subcategories of forest land converted to cropland and forest land
converted to grassland, the changes in aboveground biomass, belowground biomass, dead wood, and litter carbon
stocks are also reported. The greenhouse gas flux from grassland remaining grassland also includes estimates of
non-CC>2 emissions from grassland fires occurring on both grassland remaining grassland and land converted to
grassland.
Fluxes from wetlands remaining wetlands include changes in carbon stocks and methane (Cm) and nitrous oxide
(N2O) emissions from managed peatlands, aboveground and belowground biomass, dead organic matter, soil
carbon stock changes and Cl-Uemissions from coastal wetlands, as well as N2O emissions from aquaculture. In
addition, CFU emissions from reservoirs and other constructed waterbodies are included for the subcategory
flooded land remaining flooded land. Estimates for land converted to wetlands include aboveground and
belowground biomass, dead organic matter and soil carbon stock changes, and CFU emissions from land converted
to vegetated coastal wetlands. CO2 and CFU emissions are included for reservoirs and other constructed
waterbodies under the subcategory land converted to flooded land. See Section 6.1 for additional information on
wetlands included in this Inventory.
1 The term "flux" is used to describe the exchange of C02 to and from the atmosphere, with net flux of C02 being either positive
or negative depending on the overall balance. Removal and long-term storage of C02 from the atmosphere is also referred to as
"carbon sequestration."
Land Use, Land-Use Change, and Forestry 6-1
-------
Fluxes from settlements remaining settlements include changes in carbon stocks from organic soils, N2O emissions
from nitrogen fertilizer additions to soils, and CChfluxes from settlement trees and landfilled yard trimmings and
food scraps. The reported greenhouse gas flux from land converted to settlements includes changes in carbon
stocks in mineral and organic soils due to land use and management for all land use conversions to settlements,
and the carbon stock changes in aboveground biomass, belowground biomass, dead wood, and litter are also
included for the subcategory forest land converted to settlements.
In 2022, the Land Use, Land-Use Change, and Forestry (LULUCF) sector resulted in a net increase in carbon stocks
(i.e., net CO2 removals) of 921.8 MMT CO2 Eq. This represents an offset of approximately 14.5 percent of total (i.e.,
gross) greenhouse gas emissions in 2022. Emissions of CFU and N2O from LULUCF activities in 2022 were 58.4 and
9.1 MMT CO2 Eq., respectively, and combined represent 1.1 percent of total greenhouse gas emissions.3 In 2022,
the overall net flux from LULUCF resulted in a removal of 854.2 MMT CO2 Eq. Emissions, removals and net
greenhouse gas flux from LULUCF are summarized in Figure 6-1 and Table 6-1 by land use and category, and Table
6-2 and Table 6-3 by gas in MMT CO2 Eq. and kt, respectively. Trends in LULUCF sources and sinks over the 1990 to
2022 time series are shown in Figure 6-2.
Flooded land remaining flooded land was the largest source of non-CC>2emissions from LULUCF in 2022,
accounting for 65.5 percent of the LULUCF sector non-CC>2 emissions. Non-CC>2emissions from forest fires are the
second largest source of LULUCF sector emissions; these emissions have increased 155.2 percent since 1990 and
accounted for 21.9 percent of LULUCF non-CChemissions in 2022. Coastal wetlands remaining coastal wetlands
and settlements remaining settlements soils accounted for 6.6 and 3.8 percent of non-CChemissions from LULUCF
in 2022, respectively, and the remaining sources account for less than one percent each.
2 LULUCF carbon stock change is the net carbon stock change from the following categories: forest land remaining forest land,
land converted to forest land, cropland remaining cropland, land converted to cropland, grassland remaining grassland, land
converted to grassland, wetlands remaining wetlands, land converted to wetlands, settlements remaining settlements, and
land converted to settlements.
3 LULUCF emissions include the CH4 and N20 emissions reported for peatlands remaining peatlands, forest fires, drained
organic soils, grassland fires, and coastal wetlands remaining coastal wetlands; CH4 emissions from land converted to coastal
wetlands, flooded land remaining flooded land, and land converted to flooded land; and N20 emissions from forest soils and
settlement soils.
6-2 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
Figure 6-1: 2022 LULUCF Chapter Greenhouse Gas Sources and Sinks
Forest Land Remaining Forest Land
Settlements Remaining Settlements
Land Converted to Forest Land
Cropland Remaining Cropland
Wetlands Remaining Wetlands
Non-C02 Emissions from Peatlands Remaining Peatlands
l\lon-C02 Emissions from Drained Organic Soils
CH4 Emissions from Land Converted to Coastal Wetlands
CH4 Emissions from Land Converted to Flooded Land
Land Converted to Wetlands
N2O Emissions from Forest Soils
Non-CC>2 Emissions from Grassland Fires
N2O Emissions from Settlement Soils
Non-C02 Emissions from Coastal Wetlands Remaining Coastal Wetlands
Grassland Remaining Grassland
Non-CC>2 Emissions from Forest Fires
Land Converted to Grassland
Land Converted to Cropland
Non-C02 Emissions from Flooded Land Remaining Flooded Land
Land Converted to Settlements
(787.0)
Carbon Stock Changes
1 [\lon-CO2 Emissions
|< 0.5|
l< 0.5|
l< 0.5|
|< 0.5|
i< 0.5|
l< 0.5|
(250) (200) (150) (100) (50)
MMT CO2 Eq.
50 100
Note: Parentheses in horizontal axis indicate net sequestration.
Figure 6-2: Trends in Emissions and Removals (Net CO2 Flux) from Land Use, Land-Use
Change, and Forestry
600
400
I Forest Land Remaining Forest Land
I Land Converted to Cropland
I Land Converted to Settlements
Land Converted to Wetlands
! Settlements Remaining Settlements
Cropland Remaining Cropland
Wetlands Remaining Wetlands
Grassland Remaining Grassland
Land Converted to Forest Land
Land Converted to Grassland
¦ Net Emissions (Sources and Sinks)
200
r- CO o Ch O o
en. m H- a°.® a s ® a 88 a »
co cr>
$ s a s 5 a
-3-
m a 01
O
u
-200
-400
-600
-800
-1,000
-1,200
owtNmi-iri^Dr^coCTi
OTHrMroTLnv£)r-.cocr>0'-HrMro*rLn<£)rvcocriO'-t(N
OOOOOOOOOOT-HrH^HT-H^-lTHt-HT-H^HT-HrMrMfN
00000000000000000000000
rMNNfMfNNfM(NfNfNINrM(N(NfNfMfN(NfNfNr>l(N(N
Land Use, Land-Use Change, and Forestry 6-3
-------
Table 6-1: Emissions and Removals (Net Flux) from Land Use, Land-Use Change, and Forestry
(MMT C02 Eq.)
Land-Use Category
1990
2005
2018
2019
2020
2021
2022
Forest Land Remaining Forest Land
(968.8)
(860.0)
(863.4)
(807.0)
(846.3)
(823.8)
(771.7)
Changes in Forest Carbon Stocks3
(974.8) I
(876.0)
(873.5)
(813.2)
(862.0)
(844.2)
(787.0)
Non-C02 Emissions from Forest Firesb
5.8
15.5
9.7
5.7
15.3
19.9
14.8
N20 Emissions from Forest Soils0
0.1:
0.4
0.4
0.4
0.4
0.4
0.4
Non-C02 Emissions from Drained Organic
Soilsd
0.1
0.1
0.1
0.1
0.1
0.1
0.1
Land Converted to Forest Land
(100.2)
(100.2)
(100.4)
(100.3)
(100.3)
(100.3)
(100.3)
Changes in Forest Carbon Stockse
(100.2)
(100.2)
(100.4)
(100.3)
(100.3)
(100.3)
(100.3)
Cropland Remaining Cropland
(5.0)
(31.6)
(17.8)
(19.4)
(8.8)
(32.0)
(31.7)
Changes in Mineral and Organic Soil
Carbon Stocks
(5.0)
(31.6)
(17.8)
(19.4)
(8.8)
(32.0)
(31.7)
Land Converted to Cropland
45.4
34.5
31.9
31.4
29.3
34.9
35.1
Changes in all Ecosystem Carbon Stocks'
45.4
34.5
31.9
31.4
29.3
34.9
35.1
Grassland Remaining Grassland
24.6
24.9
29.7
28.9
17.1
11.5
14.0
Changes in Mineral and Organic Soil
Carbon Stocks
24.4
24.1
28.6
28.5
16.1
10.6
13.4
Non-C02 Emissions from Grassland Firese
o.2;
0.8 |
1.1
0.3
1.1
0.9
0.6
Land Converted to Grassland
35.3
21.8
25.2
25.4
28.7
24.5
25.6
Changes in all Ecosystem Carbon Stocks'
35.3 I
21.8 =
25.2
25.4
28.7
24.5
25.6
Wetlands Remaining Wetlands
36.8
39.4
38.2
38.1
38.1
38.1
38.1
Changes in Organic Soil Carbon Stocks in
T—1
T—1
iiiiiii
Peatlands
1.1 I
0.7
0.6
0.6
0.5
0.6
Non-C02 Emissions from Peatlands
Remaining Peatlands
+
+
+
+
+
+
+
Changes in Biomass, DOM, and Soil Carbon
I
1
!!!!!!!
mill!
Stocks in Coastal Wetlands
(10.8)
(10.1)
(11.1)
(11.1)
(11.1)
(11.1)
(11.1)
CH4 Emissions from Coastal Wetlands
Remaining Coastal Wetlands
4.2
4.2
4.3
4.3
4.3
4.3
4.3
N20 Emissions from Coastal Wetlands
IIIIID
E
=
Remaining Coastal Wetlands
0.11
0-2
0.1
0.1
0.1
0.1
0.1
CH4 Emissions from Flooded Land
Remaining Flooded Land
42.3
44.0
44.2
44.2
44.2
44.2
44.2
Land Converted to Wetlands
7.2
1.8
0.7
0.7
0.7
0.7
0.7
Changes in Biomass, DOM, and Soil Carbon
Stocks in Land Converted to Coastal
Wetlands
0.5
0.5
(+)
(+)
(+)
(+)
(+)
CH4 Emissions from Land Converted to
IIIIIII
!!!!!!:
Coastal Wetlands
0.3 !
0.3:
0.2
0.2
0.2
0.2
0.2
Changes in Land Converted to Flooded
Land
3.6
0.6
0.3
0.3
0.3
0.3
0.3
CH4 Emissions from Land Converted to
I
2.9
I
Flooded Land
0.4
0.2
0.2
0.2
0.2
0.2
Settlements Remaining Settlements
(109.1)
(115.2)
(131.0)
(131.5)
(131.8)
(132.3)
(132.3)
Changes in Organic Soil Carbon Stocks
9.9 ""
10.1 I
14.4
14.6
15.1
15.4
15.4
Changes in Settlement Tree Carbon Stocks
(96.6) 1
(117.0)
(134.4)
(135.6)
(136.7)
(137.8)
(138.5)
N20 Emissions from Settlement Soilsh
2-1 a
3-1 '
2.4
2.5
2.5
2.5
2.5
Changes in Yard Trimming and Food Scrap
1
=
Carbon Stocks in Landfills
(24.5)
(11.4)
(13.4)
(13.1)
(12.8)
(12.5)
(11.8)
Land Converted to Settlements
57.2
57.2
77.1
71.4
70.2
68.8
68.2
68.2
Changes in all Ecosystem Carbon Stocks'
77.1
71.4
70.2
68.8
68.2
68.2
LULUCF Emissions'
58.0
68.9
62.8
58.0
68.4
72.9
67.6
ch4
53.1 5
58.5 :
55.5
52.5
59.3
62.1
58.4
n2o
4.8
10.3
7.3
5.5
9.1
10.7
9.1
6-4 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
LULUCF Carbon Stock Change' (1,034.7) (976.6) (978.3) (921.6) (972.8) (983.4) (921.8)
LULUCF Sector Net Total1 (976.7) (907.7) (915.5) (863.6) (904.4) (910.6) (854.2)
+ Absolute value does not exceed 0.05 MMT C02 Eq.
a Includes the net changes to carbon stocks stored in all forest ecosystem pools (estimates include carbon stock changes
from drained organic soils from both forest land remaining forest land and land converted to forest land) and harvested
wood products.
b Estimates include CH4 and N20 emissions from fires on both forest land remaining forest land and land converted to forest
land.
c Estimates include N20 emissions from N fertilizer additions on both forest land remaining forest land and land converted to
forest land.
d Estimates include CH4 and N20 emissions from drained organic soils on both forest land remaining forest land and land
converted to forest land. Carbon stock changes from drained organic soils are included with the forest land remaining
forest land forest ecosystem pools.
e Includes the net changes to carbon stocks stored in all forest ecosystem pools.
f Includes changes in mineral and organic soil carbon stocks for all land-use conversions to cropland, grassland, and
settlements. Also includes aboveground/belowground biomass, dead wood, and litter carbon stock changes for conversion
of forest land to cropland, grassland, and settlements.
s Estimates include CH4 and N20 emissions from fires on both grassland remaining grassland and land converted to
grassland.
h Estimates include N20 emissions from nitrogen fertilizer additions on both settlements remaining settlements and land
converted to settlements because it is not possible to separate the activity data at this time.
' LULUCF emissions include the CH4 and N20 emissions reported for peatlands remaining peatlands, forest fires, drained
organic soils, grassland fires, and coastal wetlands remaining coastal wetlands; CH4 emissions from land converted to
coastal wetlands, flooded land remaining flooded land, and land converted to flooded land; and N20 emissions from forest
soils and settlement soils.
j LULUCF carbon stock change includes any carbon stock gains and losses from all land use and land-use conversion
categories.
k The LULUCF sector net total is the net sum of all LULUCF CH4 and N20 emissions to the atmosphere plus LULUCF net carbon
stock changes in units of MMT C02 Eq.
Notes: Totals may not sum due to independent rounding. Parentheses indicate net sequestration.
The carbon stock changes and emissions of CH4 and N2O from LULUCF are summarized in Table 6-2 (MMT CO2 Eq.)
and Table 6-3 (kt). Total net carbon sequestration in the LULUCF sector decreased by approximately 10.9 percent
between 1990 and 2022. This decrease was primarily due to a decline in the rate of net carbon accumulation in
forest land, as well as an increase in emissions from land converted to settlements.4 Specifically, there was a net
carbon accumulation in settlements remaining settlements, which increased from 1990 to 2022, while the net
carbon accumulation in forest land remaining forest land and land converted to wetlands slowed over this period.
Net carbon accumulation remained steady from 1990 to 2022 in land converted to forest land, cropland remaining
cropland, land converted to cropland, and wetlands remaining wetlands, while net carbon accumulation fluctuated
in grassland remaining grassland.
Flooded land remaining flooded land was the largest source of CH4 emissions from LULUCF in 2022, totaling 44.2
MMT CO2 Eq. (1,579 kt of CH4). Forest fires resulted in CH4 emissions of 9.1 MMT CO2 Eq. (325 kt of CH4).
For N2O emissions, forest fires were the largest source from LULUCF in 2022, totaling 5.7 MMT CO2 Eq. (22 kt of
N2O). Nitrous oxide emissions from fertilizer application to settlement soils in 2022 totaled to 2.5 MMT CO2 Eq. (10
kt of N2O). This represents an increase of 22.8 percent since 1990. Additionally, the application of synthetic
fertilizers to forest soils in 2022 resulted in N2O emissions of 0.4 MMT CO2 Eq. (2 kt of N2O). Nitrous oxide
emissions from fertilizer application to forest soils have increased by 455.1 percent since 1990, but still account for
a relatively small portion of overall emissions.
4 Carbon sequestration estimates are net figures. The carbon stock in a given pool fluctuates due to both gains and losses.
When losses exceed gains, the carbon stock decreases, and the pool acts as a source. When gains exceed losses, the carbon
stock increases, and the pool acts as a sink; also referred to as net carbon sequestration or removal.
Land Use, Land-Use Change, and Forestry 6-5
-------
Table 6-2: Emissions and Removals from Land Use, Land-Use Change, and Forestry by Gas
(MMT C02 Eq.)
Gas/Land-Use Category
1990
2005
2018
2019
2020
2021
2022
Carbon Stock Change (C02)a
(1,034.7)
(976.6)
(978.3)
(921.6)
(972.8)
(983.4)
(921.8)
Forest Land Remaining Forest Land
(974.8)
1
(876.0)
I
1
(873.5)
(813.2)
(862.0)
(844.2)
(787.0)
Land Converted to Forest Land
(100.2)
i
(100.2)
(100.4)
(100.3)
(100.3)
(100.3)
(100.3)
Cropland Remaining Cropland
(5.0)
(31.6)
1
I
(17.8)
(19.4)
(8.8)
(32.0)
(31.7)
Land Converted to Cropland
45.4
34.5
31.9
31.4
29.3
34.9
35.1
Grassland Remaining Grassland
24.4
1
24.1
1
28.6
28.5
16.1
10.6
13.4
Land Converted to Grassland
35.3
21.8
25.2
25.4
28.7
24.5
25.6
Wetlands Remaining Wetlands
(9.8)
1
I
(9.0)
1
1
(10.5)
(10.5)
(10.5)
(10.6)
(10.6)
Land Converted to Wetlands
4.1
1.1
0.3
0.3
0.3
0.3
0.3
Settlements Remaining Settlements
(111.2)
1
I
(118.3)
1
(133.5)
(134.0)
(134.3)
(134.8)
(134.8)
Land Converted to Settlements
57.2
77.1
71.4
70.2
68.8
68.2
68.2
ch4
53.1
1
58.6
1
55.6
52.5
59.3
62.2
58.4
Forest Land Remaining Forest Land:
Forest Firesb
3.4
9.2
6.0
3.4
9.8
12.7
9.1
Forest Land Remaining Forest Land:
1
1
1
Drained Organic Soilsc
+
I
+
I
1
+
+
+
+
+
Grassland Remaining Grassland:
Grassland Firesd
0.1
0.4
0.6
0.2
0.6
0.5
0.3
Wetlands Remaining Wetlands:
1
1
Flooded Land Remaining Flooded
1
1
Land
42.3
44.0
1
1
44.2
44.2
44.2
44.2
44.2
Wetlands Remaining Wetlands:
1
Coastal Wetlands Remaining Coastal
Wetlands
4.2
4.2
4.3
4.3
4.3
4.3
4.3
Wetlands Remaining Wetlands:
¦
Peatlands Remaining Peatlands
+
I
+
1
1
+
+
+
+
+
Land Converted to Wetlands: Land
Converted to Flooded Lands
2.9
0.4
0.2
0.2
0.2
0.2
0.2
Land Converted to Wetlands: Land
1
I
Converted to Coastal Wetlands
0.3
1
0.3
1
1
0.2
0.2
0.2
0.2
0.2
n2o
4.8
10.4
7.2
5.5
9.1
10.8
9.1
Forest Land Remaining Forest Land:
1
1
1
I
1
1
Forest Firesb
2.4
6.3
1
3.7
2.3
5.5
7.2
5.7
Forest Land Remaining Forest Land:
Forest Soilse
0.1
1
0.4
1
0.4
0.4
0.4
0.4
0.4
Forest Land Remaining Forest Land:
1
1
Drained Organic Soilsc
0.1
I
0.1
|
0.1
0.1
0.1
0.1
0.1
Grassland Remaining Grassland:
Grassland Firesd
0.1
0.4
0.5
0.1
0.5
0.4
0.3
Wetlands Remaining Wetlands:
Coastal Wetlands Remaining Coastal
1
1
1
Wetlands
0.1
1
I
0.2
I
I
0.1
0.1
0.1
0.1
0.1
Wetlands Remaining Wetlands:
Peatlands Remaining Peatlands
+
+
+
+
+
+
+
Settlements Remaining Settlements:
1
Settlement Soils'
2.1
i
3.1
2.4
2.5
2.5
2.5
2.5
LULUCF Carbon Stock Change-1
(1,034.7)
(976.6)
(978.3)
(921.6)
(972.8)
(983.4)
(921.8)
LULUCF Emissions'-
57.9
68.9
62.8
58.0
68.4
72.9
67.5
LULUCF Sector Net Total1'
(976.7)
(907.6)
(915.5)
(863.6)
(904.4)
(910.5)
(854.3)
+ Absolute value does not exceed 0.05 MMT C02 Eq.
a LULUCF carbon stock change is the net carbon stock change from the following categories: forest land remaining forest
land, land converted to forest land, cropland remaining cropland, land converted to cropland, grassland remaining
grassland, land converted to grassland, wetlands remaining wetlands, land converted to wetlands, settlements remaining
6-6 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
settlements, and land converted to settlements.
b Estimates include CH4 and N20 emissions from fires on both forest land remaining forest land and land converted to forest
land.
c Estimates include CH4 and N20 emissions from drained organic soils on both forest land remaining forest land and land
converted to forest land.
d Estimates include CH4 and N20 emissions from fires on both grassland remaining grassland and land converted to
grassland.
e Estimates include N20 emissions from nitrogen fertilizer additions on both forest land remaining forest land and land
converted to forest land.
f Estimates include N20 emissions from nitrogen fertilizer additions on both settlements remaining settlements and land
converted to settlements.
g LULUCF emissions include the CH4 and N20 emissions reported for peatlands remaining peatlands, forest fires, drained
organic soils, grassland fires, and coastal wetlands remaining coastal wetlands; CH4 emissions from flooded land remaining
flooded land, land converted to flooded land, and land converted to coastal wetlands; and N20 emissions from forest soils
and settlement soils.
h The LULUCF sector net total is the net sum of all LULUCF CH4 and N20 emissions to the atmosphere plus LULUCF net carbon
stock changes in units of MMT C02 Eq.
Notes: Totals may not sum due to independent rounding. Parentheses indicate net sequestration.
Table 6-3: Emissions and Removals from Land Use, Land-Use Change, and Forestry by Gas (kt)
Gas/Land-Use Category
1990
2005
2018
2019
2020
2021
2022
Carbon Stock Change (C02)a
(1,034,678)
(976,578)
(978,287)
(921,607)
(972,765)
(983,418)
(921,792)
Forest Land Remaining Forest Land
(974,778)
(876,001) i(873,508)
(813,183)
(862,033)
(844,194)
(787,006)
Land Converted to Forest Land
(100,216)
(100,151)
(100,409)
(100,284)
(100,297)
(100,305)
(100,294)
Cropland Remaining Cropland
(5,042)
(31,622)
(17,786)
(19,418)
(8,819)
(31,970)
(31,710)
Land Converted to Cropland
45,403
34,501
31,936
31,376
29,297
34,893
35,110
Grassland Remaining Grassland
24,366
24,071
28,557
28,536
16,086
10,566
13,352
Land Converted to Grassland
35,255
21,792
25,178
25,404
28,696
24,542
25,621
Wetlands Remaining Wetlands
(9,770) !
(8,984) ;
(10,469)
(10,509)
(10,535)
(10,582)
(10,559)
Land Converted to Wetlands
4063
1069
304
309
305
309
313
Settlements Remaining Settlements
(111,203)
(118,335) -
(133,464)
(134,000)
(134,301)
(134,842)
(134,812)
Land Converted to Settlements
57,242
77,081
71,373
70,161
68,836
68,165
68,195
ch4
1,898
2,091
1,983
1,875
2,118
2,218
2,087
Forest Land Remaining Forest Land:
Forest Firesb
122
328
213
120
349
452
327
Forest Land Remaining Forest Land:
I
111!
Drained Organic Soilsc
1
1
1
1
1
1
1
Grassland Remaining Grassland:
Grassland Firesd
4
15
22
6
20
18
12
Wetlands Remaining Wetlands:
1
lllllll
lllllli
Flooded Land Remaining Flooded
1
1<569
Land
1/509
1,578
1,579
1,579
1,579
1,579
Wetlands Remaining Wetlands:
Coastal Wetlands Remaining
I
I
Coastal Wetlands
149
151
153
153
154
154
154
Wetlands Remaining Wetlands:
!!!!!!!
mm!
mm;
sis
Peatlands Remaining Peatlands
+
+:
+
+
+
+
+
Land Converted to Wetlands: Land
Converted to Flooded Lands
102
16 _
8
8
8
8
8
Land Converted to Wetlands: Land
I
iiiiii;
Converted to Coastal Wetlands
10
io I
7
7
7
6
6
n2o
18
39
27
21
34
41
34
Forest Land Remaining Forest Land:
mm;
mm:
Forest Firesb
9
24 !
14
9
21
27
21
Forest Land Remaining Forest Land:
I
1
Forest Soilse
+
2 1
2
2
2
2
2
Land Use, Land-Use Change, and Forestry 6-7
-------
Forest Land Remaining Forest Land:
Drained Organic Soilsc
Grassland Remaining Grassland:
Grassland Firesd
Wetlands Remaining Wetlands:
Coastal Wetlands Remaining
Coastal Wetlands
Wetlands Remaining Wetlands:
Peatlands Remaining Peatlands
Settlements Remaining Settlements:
Settlement Soils'
+:
8
i
I
1
+
i
12 1
10
10
+ Absolute value does not exceed 0.5 kt.
a LULUCF carbon stock change is the net carbon stock change from the following categories: forest land remaining forest
land, land converted to forest land, cropland remaining cropland, land converted to cropland, grassland remaining grassland,
land converted to grassland, wetlands remaining wetlands, land converted to wetlands, settlements remaining settlements,
and land converted to settlements.
b Estimates include CH4 and N20 emissions from fires on both forest land remaining forest land and land converted to forest
land.
c Estimates include CH4 and N20 emissions from drained organic soils on both forest land remaining forest land and land
converted to forest land.
d Estimates include CH4 and N20 emissions from fires on both grassland remaining grassland and land converted to
grassland.
e Estimates include N20 emissions from nitrogen fertilizer additions on both forest land remaining forest land and land
converted to forest land.
f Estimates include N20 emissions from nitrogen fertilizer additions on both settlements remaining settlements and land
converted to settlements.
Notes: Parentheses indicate net sequestration. Totals may not sum due to independent rounding.
Each year, some emission and sink estimates in the LULUCF sector of the Inventory are recalculated and revised
with improved methods and/or data. In general, recalculations are made to the U.S. greenhouse gas emissions and
removals estimates either to incorporate new methodologies or, most commonly, to update recent historical data.
These improvements are implemented consistently across the previous Inventory's time series (i.e., 1990 to 2021)
to ensure that the trend is accurate. Of the updates implemented for this Inventory, the most significant include
(1) managed forest land in Hawaii and several U.S. Territories5 were included for the first time in the current
Inventory which resulted in an increase in managed forest land area of approximate 1.3 M ha and associated
increases in carbon stocks of 286 MMT C for the year 2023 in this Inventory; (2) updated methodological
framework and accounting of carbon in structural components of trees across the United States for total tree
cubic-foot volume, biomass, and carbon which led to an increase in estimated forest carbon stocks; and (3)
incorporating new U.S. Department of Agriculture (USDA) National Resources Inventory (NRI) data through 2017,
incorporating USDA-Natural Resources Conservation Service (NRCS) Conservation Effects Assessment Program
(CEAP) survey data for 2013 to 2016, incorporating cover crop and tillage management information from the OpTIS
remote-sensing data product from 2008 to 2020, in addition to other methodological updates for the estimation of
croplands and grasslands described further in those respective category sections. Together, these and other
updates increased total carbon sequestration estimates by an average of 133.6 MMT CO2 Eq. (15.5 percent) and
decreased total non-CC>2 emissions by 2.2 MMT CO2 Eq. (2.6 percent) across the time series, compared to the
previous Inventory (i.e., 1990 to 2021). For more information on specific methodological updates, please see the
Recalculations Discussion within the respective category section of this chapter.
Emissions and removals reported in the LULUCF chapter include those from all states; however, for Hawaii and
Alaska some emissions and removals from land use and land-use change are not included in most cases (see
chapter sections on Uncertainty and Planned Improvements for more details). In addition, U.S. Territories are not
included for most categories. EPA continues to review available data on an ongoing basis to include emissions and
5 American Samoa, Guam, Norther Mariana Islands, U.S. Virgin Islands, and Puerto Rico
6-8 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
removals from U.S. Territories in future Inventories to the extent they are occurring (e.g., see Box 6-2). See Annex 5
for more information on EPA's assessment of the emissions and removals not included in this Inventory.
Box 6-1: Methodological Approach for Estimating and Reporting U.S. Emissions and
Removals
Consistent with Article 13.7(a) of the Paris Agreement and Article 4.1(a) of the UNFCCC as well as relevant
decisions under those agreements, the gross emissions total presented in this report for the United States
excludes emissions and removals from LULUCF. The LULUCF Sector Net Total presented in this report for the
United States includes emissions and removals from LULUCF. All emissions and removals estimates are
calculated using internationally accepted methods in the 2006IPCC Guidelines for National Greenhouse Gas
Inventories (2006 IPCC Guidelines), 2013 Supplement to the 2006 IPCC Guidelines for National GHG Inventories:
Wetlands, and the 2019 Refinement to the 2006 IPCC Guidelines for National GHG Inventories. Additionally, the
calculated emissions and removals in a given year for the United States are presented in a common manner in
line with the reporting guidelines for the reporting of inventories under the Paris Agreement and the UNFCCC.6
The Parties' use of consistent methods to calculate emissions and removals for their inventories helps to ensure
that these reports are comparable. The presentation of emissions and removals provided in the Land Use Land-
Use Change and Forestry chapter does not preclude alternative examinations. Rather, this chapter presents
emissions and removals in a common format consistent with how Parties are to report their national inventories
under the Paris Agreement and the UNFCCC. The report itself, and this chapter, follow this common format, and
provides an explanation of the application of methods used to calculate emissions and removals.
6.1 Representation of the U.S. Land Base
A national land use representation system that is consistent and complete, both temporally and spatially, is
needed in order to assess land use and land-use change status and the associated greenhouse gas fluxes over the
Inventory time series. This system should be consistent with IPCC (2006), such that all countries reporting on
national greenhouse gas fluxes to the UNFCCC should: (1) describe the methods and definitions used to determine
areas of managed and unmanaged lands in the country (Table 6-4), (2) describe and apply a consistent set of
definitions for land-use categories over the entire national land base and time series (i.e., such that increases in
the land areas within particular land-use categories are balanced by decreases in the land areas of other categories
unless the national land base is changing) (Table 6-5), and (3) account for greenhouse gas fluxes on all managed
lands. The IPCC (2006, Vol. IV, Chapter 1) considers all anthropogenic greenhouse gas emissions and removals
associated with land use and management to occur on managed land, and all emissions and removals on managed
land should be reported based on this guidance (see IPCC (2010), Ogle et al. (2018) for further discussion).
Consequently, managed land serves as a proxy for anthropogenic emissions and removals. This proxy is intended
to provide a practical framework for conducting an inventory, even though some of the greenhouse gas emissions
and removals on managed land are influenced by natural processes that may or may not be interacting with the
anthropogenic drivers. This section of the Inventory has been developed in order to comply with this guidance.
While the 2019 Refinement to the 2006 IPCC Guidelines for National Greenhouse Gas Inventories (IPCC 2019)
provide guidance for factoring out natural emissions and removals, the United States does not apply this guidance
and estimates all emissions/removals on managed land regardless of whether the driver was natural.
6 See http://unfccc.int/resource/docs/2013/copl9/ene/10a03.pdf.
Land Use, Land-Use Change, and Forestry 6-9
-------
Three databases are used to track land management in the United States and are used as the basis to classify
United States land area into the thirty-six IPCC land use and land-use change categories (Table 6-5) (IPCC 2006).
The three primary databases are the U.S. Department of Agriculture (USDA) National Resources Inventory (NRI),7
the USDA Forest Service (USFS) Forest Inventory and Analysis (FIA)8 Database, and the Multi-Resolution Land
Characteristics Consortium (MRLC) National Land Cover Dataset (NLCD).9 See Table 6-6 for an overview of the land
area databases used to characterize land use in federal and non-federal lands in the conterminous United States,
Alaska, and Hawaii.
The total land area included in the United States Inventory is 936 million hectares across the 50 states.10
Approximately 886 million hectares of this land base is considered managed and 50 million hectares is
unmanaged, a distribution that has remained stable over the time series of the Inventory (Table 6-5). In 2022, the
United States had a total of 281 million hectares of managed forest land (0.47 percent decrease compared to
1990). There are 160 million hectares of cropland (8.3 percent decrease compared to 1990), 339 million hectares
of managed grassland (0.35 percent increase compared to 1990), 39 million hectares of managed wetlands (3
percent increase compared to 1990), 47 million hectares of settlements (41 percent increase compared to 1990),
and 21 million hectares of managed other land (1.2 percent decrease compared to 1990) (Table 6-5).
Wetlands are not differentiated between managed and unmanaged with the exception of remote areas in Alaska,
and so are classified and reported mostly as managed land within the coterminous United States.11 In addition,
carbon stock changes are not currently estimated for the entire managed 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 within interior Alaska).1213 Planned improvements are under development to
estimate carbon stock changes and greenhouse gas emissions on all managed land and to 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 and economic patterns (Figure 6-3). Forest land tends to be more common in the
eastern United 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
7 NRI data are available at https://www.nrcs.usda.Eov/wps/portal/nrcs/main/national/techriical/nra/nri/.
8 FIA data are available at https://www.fia.fs.usda.gov/tools-data/index.php.
9 NLCD data are available at http://www.mrlc.gov/ and MRLC is a consortium of several U.S. government agencies.
10 The current land representation does not include areas from U.S. Territories, but there are planned improvements to include
these regions in future Inventories. U.S. Territories represent approximately 0.1 percent of the total land base for the United
States. See Box 6-2.
11 According to the IPCC (2006), wetlands are 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 conterminous United States and Alaska is difficult due to limited data availability. Wetlands are not
characterized within the NRI with information regarding water table management. As a result, all wetlands in the conterminous
United States and Hawaii are reported as managed in the land representation, but emission/removal estimates only developed
for those wetlands that are included under the flooded lands, coastal wetlands or peat extraction categories. Efforts are
underway to better reflect wetland estimates in the future Inventories. See the Planned Improvements section of the Inventory
for future refinements to the wetland area estimates.
12 Other discrepancies occur because the coastal wetlands analysis is based on another land use product (NOAA C-CAP) that is
not currently incorporated into the land representation analysis for this section, which relies on the NRI and NLCD for wetland
areas. EPA anticipates addressing these discrepancies in future Inventories.
13 These "managed area" discrepancies also occur in the Common Reporting Tables (CRTs) submitted to the UNFCCC.
6-10 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
Midwest and eastern portions of the country, as well as coastal regions. 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
1990
2005
2018
2019
2020
2021
2022
Managed Lands
886,533
886,530
886,531
886,531
886,531
886,531
886,531
Forest
282,375 ...
281,806 |
280,971
280,440
281,067
281,071
281,041
Croplands
174,498
165,632
161,394
160,693
160,112
160,079
160,033
Grasslands
337,867 5
340,022
338,927
339,801
339,562
339,260
339,048
Settlements
33,427
40,172
45,971
46,312
46,641
46,960
47,185
Wetlands
37,456
38,310
38,495
38,551
38,430
38,478
38,566
Other
20,911
20,588
20,773
20,734
20,718
20,682
20,657
Unmanaged Lands
49,708
49,711
49,710
49,710
49,710
49,710
49,710
Forest
9,766
9,782
9,814
9,815
9,817
9,818
9,819
Croplands
0
0
0
0
0
0
0
Grasslands
25,090
25,154
25,268
25,266
25,265
25,264
25,262
Settlements
o
0
0
0
0
0
0
Wetlands
4,118
4,057
3,936
3,935
3,935
3,935
3,936
Other
10,734 ¦
10,718 ;=
10,693
10,693
10,693
10,693
10,693
Total Land Areas
936,241
936,241
936,241
936,241
936,241
936,241
936,241
Forest
292,140 :
291,588
290,784
290,255
290,883
290,889
290,861
Croplands
174,498
165,632
161,394
160,693
160,112
160,079
160,033
Grasslands
362,957
365,176 :
364,195
365,068
364,827
364,524
364,310
Settlements
33,427
40,172
45,971
46,312
46,641
46,960
47,185
Wetlands
41,574 '
42,366
42,430
42,486
42,365
42,413
42,502
Other
31,645
31,306
31,466
31,428
31,411
31,375
31,350
Table 6-5: Land Use and Land-Use Change for the U.S. Managed Land Base for All 50 States
(Thousands of Hectares)
Land Use & Land-Use
Change Categories''
1990
2005
2018
2019
2020
2021
2022
Total Forest Land
282,375
281,806
280,971
280,440
281,067
281,071
281,041
FF
281,290
280,587 i
279,683
279,167
279,818
279,829
279,802
CF
208
137
101
88
77
77
76
GF
775 ,
968 I
1,038
1,048
1,036
1,037
1,040
WF
15
23
21
18
16
15
15
SF
11 1
is
20
21
20
19
20
OF
77
73
108
98
99
94
89
Total Cropland
174,498
165,632
161,394
160,693
160,112
160,079
160,033
CC
162,273
150,417
149,721
149,503
149,823
150,591
151,276
FC
173 1
77 '
63
64
60
63
65
GC
11,673
14,623
11,231
10,758
9,914
9,132
8,418
WC
119 ,
178 i
99
98
86
81
75
SC
75
102
107
105
101
97
94
OC
186
235
173
166
129
115
107
Total Grassland
337,867
340,022
338,927
339,801
339,562
339,260
339,048
GG
328,566
315,931
318,960
320,255
320,856
321,910
322,779
FG
572
1,663
4,184
4,202
4,177
4,162
3,894
CG
8,177
I7,746
13,594
13,491
13,205
12,200
11,444
WG
168
466
181
172
159
143
134
SG
43 ,
525
230
190
139
100
93
OG
341
3,692
1,778
1,491
1,026
746
705
Total Wetlands
37,456
__ nnn:
38,310
38,495
38,551
38,430
38,478
38,566
Land Use, Land-Use Change, and Forestry 6-11
-------
WW
36,900
36,288
37,236
37,425
37,448
37,626
37,783
FW
37 1
71
96
85
83
78
76
CW
145
637
362
310
261
221
187
GW
326 :
I,169
564
501
415
342
314
SW
0
38
17
14
10
2
2
OW
47
107 I
220
216
212
210
204
Total Settlements
33,427
40,172
45,971
46,312
46,641
46,960
47,185
SS
30,562 [
31,445 '
40,769
41,615
42,466
43,189
43,748
FS
301
466
468
455
448
446
440
CS
1,231
3,604
1,917
1,726
1,528
1,366
1,228
GS
1,276
4,371
2,630
2,349
2,062
1,830
1,648
WS
4:
59
mm!
30
25
18
14
14
OS
54
229
157
141
120
115
108
Total Other Land
20,911
20,588
20,773
20,734
20,718
20,682
20,657
00
20,177
17,022
18,050
18,293
18,553
18,805
18,874
FO
51 •
77
98
101
101
108
111
CO
287
603
629
582
540
489
444
GO
371 1
2,764
1,772
1,541
1,309
1,068
1,018
WO
22
100
206
206
205
204
200
SO
2 =
21
17
11
10
10
10
Grand Total
886,533
886,530
886,531
886,531
886,531
886,531
886,531
a The abbreviations are "F" for forest land, "C" for cropland, "G" for grassland, "W" for wetlands, "S" for settlements,
and "0" 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).
Notes: 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 U.S. Territories in future Inventories. In addition, carbon 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 (see land use chapters e.g., Forest Land Remaining Forest Land for more
information). Totals may not sum due to independent rounding.
6-12 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
Figure 6-3: Percent of Total Land Area for Each State in the General Land Use Categories for
2022
Croplands
Forest Lands
Grasslands
Other Lands
Settlements
Wetlands
Percent
0 10 30 50 100
Land Use, Land-Use Change, and Forestry 6-13
-------
Methodology and Time-Series Consistency
IPCC (2006) describes three approaches for representing land areas. Approach 1 provides data on the total area for
each individual land use category, but does not provide detailed information on transfer of land area between
categories following land-use change and is not spatially explicit other than at the national or regional level. With
Approach 1, total net conversions between categories can be detected, but not the individual changes (i.e.,
additions and/or losses) between the land-use categories that led to those net changes. Approach 2 introduces
tracking of individual land-use changes between the categories (e.g., forest land converted to cropland, cropland
converted to forest land, and grassland converted to cropland), using survey samples or other forms of data, but
does not provide spatially-explicit location data. Approach 3 extends Approach 2 by providing spatially-explicit
location data, such as surveys with spatially identified sample locations and maps obtained from remote sensing
products. The three approaches are not presented as hierarchical tiers and are not mutually exclusive.
According to IPCC (2006), the approach or mix of approaches selected by an inventory agency should reflect
calculation needs and national circumstances. For this analysis, the NRI, FIA, and the NLCD have been combined to
provide a complete representation of land use for managed lands. These data sources are described in more detail
later in this section. NRI, FIA and NLCD are Approach 3 data sources that provide spatially-explicit representations
of land use and land-use conversions. Lands are treated as remaining in the same category (e.g., cropland
remaining cropland) if a land-use change has not occurred in the last 20 years, consistent with the IPCC guidelines
(2006). Otherwise, the land is classified in a land-use change category based on the current use and most recent
use before conversion to the current use (e.g., cropland converted to forest land).
Definitions of Land Use in the United States
Managed and Unmanaged Land
The United States definition of managed land is similar to the general definition of managed land provided by the
IPCC (2006), but with some additional elaboration to reflect national circumstances. Based on the following
definitions, most lands in the United States are classified as managed:
• Managed Land: Land is considered managed if direct human intervention has influenced its condition.
Direct intervention occurs mostly in areas accessible to human activity and includes altering or
maintaining the condition of the land to produce commercial or non-commercial products or services; to
serve as transportation corridors or locations for buildings, landfills, or other developed areas for
commercial or non-commercial purposes; to extract resources or facilitate acquisition of resources; or to
provide social functions for personal, community, or societal objectives where these areas are readily
accessible to society.14
• Unmanaged Land: All other land is considered unmanaged. Unmanaged land is largely comprised of areas
inaccessible to society due to the remoteness of the locations. Though these lands may be influenced
14 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 within the NRI with information regarding water table management or origin (i.e.,
constructed rather than natural origin). Therefore, unless wetlands are converted into 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, most wetlands
are reported as managed with the exception of wetlands in remote areas of Alaska, but emissions from managed wetlands are
only reported for coastal regions, flooded lands (e.g., reservoirs) and peatlands where peat extraction occurs due to insufficient
activity data to estimate emissions and limited resources to improve the Inventory. See the Planned Improvements section of
the Inventory for future refinements to the wetland area estimates.
6-14 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
indirectly by human actions such as atmospheric deposition of chemical species produced in industry or
CO2 fertilization, they are not influenced by a direct human intervention.15
In addition, land that is previously managed remains in the managed land base for 20 years before re-classifying
the land as unmanaged in order to account for legacy effects of management on carbon stocks.16 Unmanaged land
is also re-classified as managed over time if anthropogenic activity is introduced into the area based on the
definition of managed land.
Land-Use Categories
As with the definition of managed lands, IPCC (2006) provides general non-prescriptive definitions for the six main
land-use categories: forest land, cropland, grassland, wetlands, settlements and other land. In order to reflect
national circumstances, country-specific definitions have been developed, based predominantly on criteria used in
the land-use surveys for the United States. Specifically, the definition of forest land is based on the FIA definition of
forest,17 while definitions of cropland, grassland, and settlements are based on the NRI.18 The definitions for other
land and wetlands are based on the IPCC (2006) definitions for these categories.
• Forest Land: A land-use category that includes areas at least 120 feet (36.6 meters) wide and at least one
acre (0.4 hectare) in size with at least ten percent cover (or equivalent stocking) by live trees including
land that formerly had such tree cover and that will be naturally or artificially regenerated. Trees are
woody plants having a more or less erect perennial stem(s) capable of achieving at least 3 inches (7.6 cm)
in diameter at breast height, or 5 inches (12.7 cm) diameter at root collar, and a height of 16.4 feet (5 m)
at maturity in situ. Forest land includes all areas recently having such conditions and currently
regenerating or capable of attaining such condition in the near future. Forest land also includes transition
zones, such as areas between forest and non-forest lands, that have at least ten percent cover (or
equivalent stocking) with live trees and forest areas adjacent to urban and built-up lands. Unimproved
roads and trails, streams, and clearings in forest areas are classified as forest if they are less than 120 feet
(36.6 m) wide or an acre (0.4 ha) in size. However, land is not classified as forest land if completely
surrounded by urban or developed lands, even if the criteria are consistent with the tree area and cover
requirements for forest land. These areas are classified as settlements. In addition, forest land does not
include land that is predominantly under an agricultural land use (Nelson et al. 2020).
• Cropland: A land-use category that includes areas used for the production of adapted crops for harvest;
this category includes both cultivated and non-cultivated lands. Cultivated crops include row crops or
close-grown crops and also pasture in rotation with cultivated crops. Non-cultivated cropland includes
continuous hay, perennial crops (e.g., orchards) and horticultural cropland. Cropland also includes land
with agroforestry, such as alley cropping and windbreaks,19 if the dominant use is crop production,
assuming the stand or woodlot does not meet the criteria for forest land. Lands in temporary fallow or
15 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.
16 There are examples of managed land transitioning to unmanaged land in the United States. For example, in 2018,100
hectares of managed grassland converted to unmanaged because data indicated that no further grazing occurred. Livestock
data are collected annually by the Department of Agriculture, and no livestock had occurred in the area since the mid-1970s,
and therefore there was no longer active management through livestock grazing. The area is also remote, at least 10 miles from
roads and settlements, and therefore the land was no longer managed based on the implementation criteria.
17 See https://www.fia.fs.usda.Eov/librarv/field-guides-methods-proc/docs/2022/core ver9-2 9 2022 SW HW%20table.pdf.
page 23.
18 See https://www.nrcs.usda.gov/wps/portal/nrcs/main/national/technical/nra/nri/.
19 Currently, there is no data source to account for biomass carbon 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.
Land Use, Land-Use Change, and Forestry 6-15
-------
enrolled in conservation reserve programs (i.e., set-asides20) are also classified as cropland, as long as
these areas do not meet the forest land criteria. Roads through cropland, including interstate highways,
state highways, other paved roads, gravel roads, dirt roads, and railroads are excluded from cropland area
estimates and are, instead, classified as settlements.
• Grassland: A land-use category on which the plant cover is composed principally of grasses, grass-like
plants (i.e., sedges and rushes), forbs, or shrubs suitable for grazing and browsing, and includes both
pastures and native rangelands. This includes areas where practices such as clearing, burning, chaining,
and/or chemicals are applied to maintain the grass vegetation. Land is also categorized as grassland if
there have been three or fewer years of continuous hay production.21 Savannas, deserts, and tundra are
considered grassland. Drained wetlands are considered grassland if the dominant vegetation meets the
plant cover criteria for grassland. Woody plant communities of low forbs, shrubs and woodlands, such as
sagebrush, mesquite, chaparral, mountain shrubland, and pinyon-juniper, are also classified as grassland if
they do not meet the criteria for forest land. Grassland includes land managed with agroforestry
practices, such as silvopasture and windbreaks, if the land is principally grass, grass-like plants, forbs, and
shrubs suitable for grazing and browsing, and assuming the stand or woodlot does not meet the criteria
for forest land. Roads through grassland, including interstate highways, state highways, other paved
roads, gravel roads, dirt roads, and railroads are excluded from grassland and are, instead, classified as
settlements.
• Wetlands: A land-use category that includes land covered or saturated by water for all or part of the year,
in addition to lakes, reservoirs, and rivers. In addition, all coastal wetlands are considered managed
regardless of whether the water level is changed or if they were created by human activity. Certain areas
that fall under the managed wetlands definition are included in other land uses based on the IPCC
guidance and national circumstances, including lands that are flooded for most or just part of the year in
croplands (e.g., rice cultivation and cranberry production), grasslands (e.g., wet meadows dominated by
grass cover) and forest lands (e.g., riparian forests near waterways). See Section 6.8 for more information.
• Settlements: A land-use category representing developed areas consisting of units equal to or greater
than 0.25 acres (0.1 ha) that includes residential, industrial, commercial, and institutional land;
construction sites; public administrative sites; railroad yards; cemeteries; airports; golf courses; sanitary
landfills; sewage treatment plants; water control structures and spillways; parks within urban and built-up
areas; and highways, railroads, and other transportation facilities. Also included are all tracts that may
meet the definition of forest land, and tracts of less than ten acres (4.05 ha) that may meet the definitions
for cropland, grassland, or other land but are completely surrounded by urban or built-up land, and so are
included in the settlements category. Rural transportation corridors located within other land uses (e.g.,
forest land, cropland, and grassland) are also included in settlements.
• Other Land: A land-use category that includes bare soil, rock, ice, and all land areas that do not fall into
any of the other five land-use categories. Following the guidance provided by the IPCC (2006), carbon
stock changes and non-CC>2 emissions are not estimated for other lands because these areas are largely
devoid of biomass, litter and soil carbon pools. However, carbon stock changes and non-CC>2 emissions
should be estimated for land converted to other land during the first 20 years following conversion to
account for legacy effects.
20 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, but is still classified as cropland based on national circumstances.
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. Occasional harvest of hay from grasslands typically
does not involve cultivation or other intensive management practices.
6-16 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
Land Use Data Sources: Description and Application to U.S.
Land Area Classification
U.S. Land Use Data Sources
The three main sources for land use data in the United States are the NRI, FIA, and the NLCD (Table 6-6). These
data sources are combined to account for land use in all 50 states. FIA and NRI data are used when available for an
area because these surveys contain additional information on management, site conditions, crop types, biometric
measurements, and other data that are needed to estimate carbon stock changes, N2O, and Cm emissions on
those lands. If NRI and FIA data are not available for an area, however, then the NLCD product is used to represent
the land use. Sources of land use data included in the land representation in this Inventory are consistent with
those included in the previous Inventory.
Table 6-6: Data Sources Used to Determine Land Use and Land Area for the Conterminous
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
•
National Resources Inventory
For the Inventory, the NRI is the official source of data for land use and land-use change on non-federal lands in the
conterminous United States and Hawaii, and is also used to determine the total land base for the conterminous
United States and Hawaii. The NRI is a statistically-based survey conducted by the USDA Natural Resources
Conservation Service and is designed to assess soil, water, and related environmental resources on non-federal
lands. The NRI has a stratified multi-stage sampling design, where primary sample units are stratified on the basis
of county and township boundaries defined by the United States Public Land Survey (Nusser and Goebel 1997).
Within a primary sample unit (typically a 160 acre [64.75 ha] square quarter-section), three sample points are
selected according to a restricted randomization procedure. Each point in the survey is assigned an area weight
(expansion factor) based on other known areas and land use information (Nusser and Goebel 1997). The NRI
survey utilizes data obtained from remote sensing imagery and site visits in order to provide detailed information
on land use and management, particularly for croplands and grasslands (i.e., agricultural lands), and is used as the
Land Use, Land-Use Change, and Forestry 6-17
-------
basis to account for carbon stock changes in agricultural lands (except federal grasslands). The NRI survey was
conducted every five years between 1982 and 1997, but shifted to annualized data collection in 1998. The land use
between five-year 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 beginning and end of the five-year period (note: most of the data have the same land use at
the beginning and end of the five-year periods). If the land use had changed during a five-year period, then the
change is assigned at random to one of the five years. For crop histories, years with missing data are estimated
based on the sequence of crops grown during years preceding and succeeding a missing year in the NRI history.
This gap-filling approach allows for development of a full time series of land use data for non-federal lands in the
conterminous United States and Hawaii. This Inventory incorporates data through 2017 from the NRI. The land use
patterns are assumed to remain the same from 2018 through 2022 for this Inventory, but the time series will be
updated when new data are integrated into the land representation analysis.
Forest Inventory and Analysis
The FIA program, conducted by the USFS, is the official source of data on forest land area and management data
for the Inventory and is another statistically-based survey for the United States. The Forest Inventory and Analysis
engages in a hierarchical system of sampling, with sampling categorized as Phases 1 through 3, in which sample
points for each consecutive phase are subsets of the previous phase. Phase 1 refers to collection of remotely-
sensed data (either aerial or satellite imagery) primarily to classify land into forest or non-forest and to identify
landscape patterns like fragmentation and urbanization. Phase 2 is the collection of field data on a network of
ground plots that enable classification and summarization of area, tree, and other attributes associated with forest
land uses. Phase 3 plots are a subset of Phase 2 plots where data on indicators of forest health are measured. Data
from all three phases are also used to estimate carbon stock changes for forest land. Historically, FIA inventory
surveys have been conducted periodically, with all plots in a state being measured at a frequency of every five to
ten years. A new national plot design and annual sampling design was introduced by the FIA program in 1998 and
is now used in all states. Annualized sampling means that a portion of plots throughout each state is sampled each
year, with the goal of measuring all plots once every five to seven years in the eastern United States and once
every ten years in the western United States. See Annex 3.13 for the specific survey data available by state. The
most recent year of available data varies state by state (range of most recent data is from 2019 through 2022; see
Table A-202 in Annex 3.13).
National Land Cover Dataset
As noted above, while the NRI survey sample covers the conterminous United States and Hawaii, land use data are
only collected on non-federal lands. Gaps exist in the land representation when the NRI and FIA datasets are
combined, such as federal grasslands operated by Bureau of Land Management (BLM), USDA, and National Park
Service, as well as Alaska.22 The NLCD is used to account for land use on federal lands in the conterminous United
States and Hawaii, in addition to federal and non-federal lands in Alaska with the exception of forest lands in
Alaska.
NLCD products provide land-cover for 1992, 2001, 2004, 2006, 2008, 2011, 2013, 2016, 2019, and 2021 in the
conterminous United States (Yang et al. 2018; Fry et al. 2011; Homer et al. 2007, 2015, Dewitz, 2023), and also for
Alaska in 2001, 2011, and 2016 and Hawaii in 2001. Note that the 2021 NLCD product was not available at the time
the land representation was begun for this Inventory so it was not included. A NLCD change product is not
available for Hawaii because data are only available for one year, i.e., 2001. The NLCD products are based primarily
on Landsat Thematic Mapper imagery at a 30-meter resolution, and the land-cover categories have been
aggregated into the 36 IPCC land-use categories for the conterminous United States and Alaska, and into the six
IPCC land-use categories for Hawaii. The land-use patterns are assumed to remain the same after the last year of
22 The NRI survey program does not include U.S. Territories with the exception of non-federal lands in Puerto Rico. The FIA
program recently began implementing surveys of forest land in U.S. Territories and those data will be used in the years ahead.
Furthermore, NLCD does not include coverage for all U.S. Territories.
6-18 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
data in the time series, which is 2001 for Hawaii, 2019 for the conterminous United States and 2016 for Alaska, but
the time series will be updated when new data are released.
For the conterminous United States, the aggregated maps of IPCC land-use categories obtained from the NLCD
products were used in combination with the NRI database to represent land use and land-use change for federal
lands, with the exception of forest lands, which are based on FIA. Specifically, NRI survey locations designated as
federal lands were assigned a land use/land-use change category based on the NLCD maps that had been
aggregated into the IPCC categories. This analysis addressed shifts in land ownership across years between federal
or non-federal classes as represented in the NRI survey (i.e., the ownership is classified for each survey location in
the NRI). The sources of these additional data are discussed in subsequent sections of the report.
Managed Land Designation
Lands are designated as managed in the United States based on the definition provided earlier in this section. The
following criteria are used in order to apply the definition in an analysis of managed land:
• All croplands and settlements are designated as managed so only grassland, forest land, wetlands or other
lands may be designated as unmanaged land;23
• All forest lands with active fire protection are considered managed;
• All forest lands designated for timber harvests are considered managed;
• All grasslands are considered managed at a county scale if there are grazing livestock in the county;
• Other areas are considered managed if accessible based on the proximity to roads and other
transportation corridors, and/or infrastructure;
• Protected lands maintained for recreational and conservation purposes are considered managed (i.e.,
managed by public and/or private organizations);
• Lands with active and/or past resource extraction are considered managed; and
• Lands that were previously managed but subsequently classified as unmanaged remain in the managed
land base for 20 years following the conversion to account for legacy effects of management on carbon
stocks.
The analysis of managed lands, based on the criteria listed above, is conducted using a geographic information
system (Ogle et al. 2018). Lands that are used for crop production or settlements are determined from the NLCD
(Fry et al. 2011; Homer et al. 2007; Homer et al. 2015). Forest lands with active fire management are determined
from maps of federal and state management plans from the National Atlas (U.S. Department of Interior 2005) and
Alaska Interagency Fire Management Council (1998). It is noteworthy that all forest lands in the conterminous
United States have active fire protection, and are therefore designated as managed regardless of accessibility or
other criteria. In addition, forest lands with timber harvests are designated as managed based on county-level
estimates of timber products in the U.S. Forest Service Timber Products Output Reports (U.S. Department of
Agriculture 2012). Timber harvest data lead to additional designation of managed forest land in Alaska. The
designation of grasslands as managed is based on grazing livestock population data at the county scale from the
USDA National Agricultural Statistics Service (U.S. Department of Agriculture 2015). Accessibility is evaluated based
on a 10-km buffer surrounding road and train transportation networks using the ESRI Data and Maps product (ESRI
2008), and a 10-km buffer surrounding settlements using NLCD.
Lands maintained for recreational purposes are determined from analysis of the Protected Areas Database (U.S.
Geological Survey 2012). The Protected Areas Database includes lands protected from conversion of natural
habitats to anthropogenic uses and describes the protection status of these lands. Lands are considered managed
23 All wetlands are considered managed in this Inventory with the exception of remote areas in Alaska. Distinguishing between
managed and unmanaged wetlands in the conterminous United States and Hawaii is difficult due to limited data availability.
Wetlands are not characterized within the NRI with information regarding water table management. Regardless, a planned
improvement is underway to subdivide managed and unmanaged wetlands.
Land Use, Land-Use Change, and Forestry 6-19
-------
that are protected from development if the regulations allow for extractive or recreational uses or suppression of
natural disturbance (e.g., forest lands with active fire protection). Lands that are protected from development and
not accessible to human intervention, including no suppression of disturbances or extraction of resources, are not
included in the managed land base.
Multiple data sources are used to determine lands with active resource extraction: Alaska Oil and Gas Information
System (Alaska Oil and Gas Conservation Commission 2009), Alaska Resource Data File (U.S. Geological Survey
2012), Active Mines and Mineral Processing Plants (U.S. Geological Survey 2005), and Coal Production and
Preparation Report (U.S. Energy Information Administration 2011). A buffer of 3,300 and 4,000 meters is
established around petroleum extraction and mine locations, respectively, to account for the footprint of
operation and impacts of activities on the surrounding landscape. The buffer size is based on visual analysis of
disturbance to the landscape for approximately 130 petroleum extraction sites and 223 mines. After applying the
criteria identified above, the resulting managed land area is overlaid on the NLCD to estimate the area of managed
land by land use for both federal and non-federal lands in Alaska. The remaining land represents the unmanaged
land base. The resulting spatial product is also used to identify NRI survey locations that are considered managed
and unmanaged for the conterminous United States and Hawaii.24
Approach for Combining Data Sources
The managed land base in the United States has been classified into the 36 IPCC land use/land-use conversion
categories (Table 6-5) using definitions developed to meet national circumstances, while adhering to IPCC
guidelines (2006).25 In practice, the land was initially classified into land-use subcategories within the NRI, FIA, and
NLCD datasets, and then aggregated into the 36 broad land use and land-use change categories identified in IPCC
(2006).
All three datasets provide information on forest land areas in the conterminous United States, but the area data
from FIA serve as the official dataset for forest land. Therefore, another step in the analysis is to address the
inconsistencies in the representation of the forest land among the three databases. NRI and FIA have different
criteria for classifying forest land in addition to different sampling designs, leading to discrepancies in the resulting
estimates of forest land area on non-federal land in the conterminous United States. Similarly, there are
discrepancies between the NLCD and FIA data for defining and classifying forest land on federal lands. Any change
in forest land area in the NRI and NLCD also requires a corresponding change in other land use areas because of
the dependence between the forest land area and the amount of land designated as other land uses, such as the
amount of grassland, cropland, and wetlands (i.e., areas for the individual land uses must sum to the total
managed land area of the country).
FIA is the main database for forest statistics, and consequently, the NRI and NLCD are adjusted to achieve
consistency with FIA estimates of forest land in the conterminous United States. Adjustments are made in the
forest land remaining forest land, land converted to forest land, and forest land converted to other uses (i.e.,
grassland, cropland, settlements, other lands, and wetlands). All adjustments are made at the state scale to
address the discrepancies in areas associated with forest land and conversions to and from forest land. There are
three steps in this process. The first step involves adjustments to land converted to forest land (grassland,
cropland, settlements, other lands, and wetlands), followed by a second step in which there are adjustments in
forest land converted to another land use (i.e., grassland, cropland, settlements, other lands, and wetlands), and
the last step is to adjust forest land remaining forest land.
In the first step, land converted to forest land in the NRI and NLCD are adjusted to match the state-level estimates
in the FIA data for non-federal and federal land converted to forest land, respectively. FIA data have not provided
24 The exception is cropland and settlement areas in the NRI, which are classified as managed, regardless of the managed land
base obtained from the spatial analysis described in this section.
25 Definitions are provided in the previous section.
6-20 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
specific land-use categories that are converted to forest land in the past, but rather a sum of all land converted to
forest land.26 The NRI and NLCD provide information on specific land-use conversions, such as grassland converted
to forest land. Therefore, adjustments at the state level to NRI and NLCD are made proportional to the amount of
specific land-use conversions into forest land for the state, prior to any further adjustments. For example, if 50
percent of the land-use change to forest land is associated with grassland converted to forest land in a state
according to NRI or NLCD, then half of the discrepancy with FIA data in the area of land converted to forest land is
addressed by increasing or decreasing the area in grassland converted to forest land. Moreover, any increase or
decrease in grassland converted to forest land in NRI or NLCD is addressed by a corresponding change in the area
of grassland remaining grassland, so that the total amount of managed area is not changed within an individual
state. Since the sum of all land converted to forest land is used to adjust specific land-use conversions into forest
land for the state-level estimates in the NRI and NLCD, there is the potential for differences in area estimates in
states where specific land-use conversions into forest land do not exist in the FIA data.
In the second step, state-level areas are adjusted in the NRI and NLCD to address discrepancies with FIA data for
forest land converted to other uses. Similar to land converted to forest land, FIA have not provided information on
the specific land-use changes in the past,27 so areas associated with forest land conversion to other land uses in
NRI and NLCD are adjusted proportional to the amount of area in each conversion class in these datasets. Since the
sum of all forest land converted to other uses is used to adjust specific land-used conversions out of forest land for
the state-level estimates in the NRI and NLCD, there is the potential for differences in area estimates in states
where a specific land-use conversion out of forest land does not exist in the FIA data.
In the final step, the area of forest land remaining forest land in each state according to the NRI and NLCD is
adjusted to match the FIA estimates for non-federal and federal land, respectively. It is assumed that the majority
of the discrepancy in forest land remaining forest land is associated with less-precise estimates of grassland
remaining grassland and wetlands remaining wetlands in the NRI and NLCD. This step also assumes that there are
no changes in the land-use conversion categories. Therefore, corresponding adjustments are made in the area
estimates of grassland remaining grassland and wetlands remaining wetlands from the NRI and NLCD. This
adjustment balances the change in forest land remaining forest land area, which ensures no change in the overall
amount of managed land within an individual state. The adjustments are based on the proportion of land within
each of these land-use categories at the state level according to NRI and NLCD (i.e., a higher proportion of
grassland led to a larger adjustment in grassland area).
The modified NRI data are then aggregated to provide the land use and land-use change data for non-federal lands
in the conterminous United States, and the modified NLCD data are aggregated to provide the land use and land-
use change data for federal lands. Data for all land uses in Hawaii are based on NRI for non-federal lands and on
NLCD for federal lands. Land use data in Alaska are based on the NLCD data after adjusting this dataset to be
consistent with forest land areas in the FIA (Table 6-6). The result is land use and land-use change data for the
conterminous United States, Hawaii, and Alaska.
A summary of the details on the approach used to combine data sources for each land use are described below.
• Forest Land: Land representation for both non-federal and federal forest lands in the conterminous
United States and Alaska are based on the FIA. The FIA is used as the basis for both forest land area data
as well as to estimate carbon stocks and fluxes on forest land in the conterminous United States and
Alaska. The FIA does have survey plots in Alaska that are used to determine the carbon stock changes, and
the associated area data for this region are harmonized with NLCD using the methods described above.
NRI is used in the current report to provide forest land areas on non-federal lands in Hawaii, and NLCD is
26 The FIA program has started to collect data on the specific land uses that are converted to forest land, which will be further
investigated and incorporated into a future Inventory.
27 The FIA program has started to collect data on the specific land uses following conversion from forest land, which will be
further investigated and incorporated into a future Inventory.
Land Use, Land-Use Change, and Forestry 6-21
-------
used for federal lands. In Hawaii and the U.S. Territories, FIA data are being collected; these data were
used to compile area estimates and emissions and removals for forest land in this Inventory.
• Cropland: Cropland is classified using the NRI, which covers all non-federal lands within 49 states
(excluding Alaska), including state and local government-owned land as well as tribal lands. The NRI is
used as the basis for both cropland area data as well as to estimate soil carbon stocks and fluxes on
cropland. The NLCD is used to determine cropland area and soil carbon stock changes on federal lands in
the conterminous United States and Hawaii. The NLCD is also used to determine croplands in Alaska, but
carbon stock changes are not estimated for this region in the current Inventory.
• Grassland: Grassland on non-federal lands is classified using the NRI within 49 states (excluding Alaska),
including state and local government-owned land as well as tribal lands. The NRI is used as the basis for
both grassland area data as well as to estimate soil carbon stocks and non-CC>2 greenhouse emissions on
grassland. Grassland area and soil carbon stock changes are determined using the classification provided
in the NLCD for federal land within the conterminous United States. The NLCD is also used to estimate the
areas of federal and non-federal grasslands in Alaska, and the federal grasslands in Hawaii, but the current
Inventory does not include carbon stock changes in these areas.
• Wetlands: The NRI captures wetlands on non-federal lands within 49 states (excluding Alaska), while the
land representation data for federal wetlands and wetlands in Alaska are based on the NLCD.28
• Settlements: The NRI captures non-federal settlement area in 49 states (excluding Alaska). If areas of
forest land or grassland under ten acres (4.05 ha) are contained within settlements or urban areas, they
are classified as settlements (urban) in the NRI database. If these parcels exceed the ten-acre (4.05 ha)
threshold and are grassland, they are classified as grassland by NRI. Regardless of size, a forested area is
classified as non-forest by FIA if it is located within an urban area. Land representation for settlements on
federal lands and Alaska is based on the NLCD.
• Other Land: Any land that is not classified into one of the previous five land-use categories is categorized
as other land using the NRI for non-federal areas in the conterminous United States and Hawaii and using
the NLCD for the federal lands in all regions of the United States and for non-federal lands in Alaska.
Some lands can be classified into one or more categories due to multiple uses that meet the criteria of more than
one definition. However, a ranking has been developed for assignment priority in these cases. The ranking process
is from highest to lowest priority based on the following order:
Settlements > Cropland > Forest Land > Grassland > Wetlands > Other Land
Settlements are given the highest assignment priority because they are extremely heterogeneous with a mosaic of
patches that include buildings, infrastructure, and travel corridors, but also open grass areas, forest patches,
riparian areas, and gardens. The latter examples could be classified as grassland, forest land, wetlands, and
cropland, respectively, but when located in close proximity to settlement areas, they tend to be managed in a
unique manner compared to non-settlement areas. Consequently, these areas are assigned to the settlements
land-use category. Cropland is given the second assignment priority, because cropping practices tend to dominate
management activities on areas used to produce food, forage, or fiber. The consequence of this ranking is that
crops in rotation with pasture are classified as cropland, and land with woody plant cover that is used to produce
crops (e.g., orchards) is classified as cropland, even though these areas may also meet the definitions of grassland
or forest land, respectively. Similarly, wetlands are considered croplands if they are used for crop production, such
as rice or cranberries. Forest land occurs next in the priority assignment because traditional forestry practices tend
to be the focus of the management activity in areas with woody plant cover that are not croplands (e.g., orchards)
28 This analysis does not distinguish between managed and unmanaged wetlands except for remote areas in Alaska, but there
is a planned improvement to subdivide managed and unmanaged wetlands for the entire land base.
6-22 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
or settlements (e.g., housing subdivisions with significant tree cover). Grassland occurs next in the ranking, while
wetlands and then other land complete the list.
The assignment priority does not reflect the level of importance for reporting greenhouse gas emissions and
removals on managed land, but is intended to classify all areas into a discrete land-use category. Currently, the
IPCC does not make provisions in the guidelines for assigning land to multiple uses. For example, a wetland is
classified as forest land if the area has sufficient tree cover to meet the stocking and stand size requirements.
Similarly, wetlands are classified as cropland if they are used for crop production, such as rice, or as grassland if
they are composed principally of grasses, grass-like plants (i.e., sedges and rushes), forbs, or shrubs suitable for
grazing and browsing. Regardless of the classification, emissions and removals from these areas should be included
in the Inventory if the land is considered managed, and therefore impacted by anthropogenic activity in
accordance with the guidance provided by the IPCC (2006).
/erification
The land base obtained from the NRI, FIA, and NLCD was compared to the Topological^ Integrated Geographic
Encoding and Referencing (TIGER) Survey (U.S. Census Bureau 2010). The United States Census Bureau gathers
data on the population and economy and has a database of land areas for the country. The area estimates of land-
use categories, based on NRI, FIA, and NLCD, are obtained from remote sensing data instead of the land survey
approach used by the United States Census Survey. The Census does not provide a time series of land-use change
data or land management information, which is needed for estimating greenhouse gas emissions from land use
and land-use change. Regardless, the Census does provide sufficient information to provide a quality assurance
check on the Inventory data. There are 46 million more hectares of land in the United States according to the
Census, compared to the total area estimate of 936 million hectares obtained from the combined NRI, FIA, and
NLCD data, a 4.8 percent difference. Much of this difference is associated with open water in coastal regions and
the Great Lakes, which is included in the TIGER Survey of the Census, but not included in the land representation
using the NRI, FIA and NLCD. There is only a 0.4 percent difference when open water in coastal regions is removed
from the TIGER data. General QC procedures for data gathering and data documentation also were applied
consistent with the QA/QC and Verification Procedures described in Annex 8.
Recalculations Discussion
The land representation estimates were recalculated from the previous Inventory with the following datasets: a)
updated FIA data from 1990 to 2022 for the conterminous United States and Alaska, b) NRI data from 1990 to 2017
for the conterminous United States and Hawaii, and c) NLCD data for the conterminous United States from 2001
through 2019 and Alaska from 2001 through 2016. There were several changes in methods that resulted in small
changes between this Inventory and the previous Inventory. First, pasture land was previously classified as
cropland in the compilation of forest land conversion estimates using FIA data and is now classified as grassland to
align with methods and definitions used to classify grasslands using NRI data. This led to a decrease in total
managed cropland area and an increase in grassland area. Second, FIA data are now used to classify forest land
and conversions to and from forest land in coastal southeast and southcentral Alaska which resulted in minor
changes, primarily between forest land, wetlands, and grasslands, between this Inventory and the previous
Inventory. Lastly, methods for classifying wetlands using FIA data were refined so that all water bodies are now
classified as wetlands (previously some water bodies were classified as other lands) aligning with methods and
definitions in the NRI. Collectively, these refinements in FIA methods to better align with methods for the other
data sources (i.e., NRI and NLCD) resulted in changes throughout the entire representation of land (see "Approach
for Combining Data Sources"). Specifically, managed wetland area decreased, on average over the time series, by
1.2 percent. Grassland and forest land increased by 0.1 percent and 0.04 percent, respectively. Settlement area
decreased by 0.05 percent and cropland and managed other lands were essentially unchanged in the latest
Inventory.
Land Use, Land-Use Change, and Forestry 6-23
-------
Planned Improvements
Research is underway to harmonize NRI and FIA sampling frames to improve consistency and facilitate estimation
using multi-frame sampling. This includes development of a common land use classification schema between the
two land inventories that can be used in the harmonization process. These steps will allow for population
estimation exclusive of auxiliary information (e.g., NLCD). The multi-frame sample will also serve as reference data
for the development of spatially explicit and spatially continuous map products for each year in the Inventory time
series. Another key planned improvement for the Inventory is to fully incorporate area data by land-use type for
U.S. Territories. Although most of the managed land in the United States is included in the current land use data
for the conterminous United States, Alaska, and Hawaii, a complete reporting of all lands in the United States,
including U.S. Territories, is a key goal for the near future. An initial assessment of data sources for land use area
data for U.S. Territories by land-use category are provided in Box 6-2. In addition, this Inventory includes forest
land areas estimated for American Samoa, Guam, Hawaii, Northern Marianas Islands, U.S. Virgin Islands, and
Puerto Rico using periodic inventories from the FIA program. These estimates are included in the forest land
category, and the methods for compiling these area estimates and the associated carbon stocks and fluxes and
integration of these estimates into the land representation will be refined to compensate for data limitations in
the time series while also taking advantage of new data and data products. See Box 6-2.
Box 6-2: Preliminary Estimates of Land Use in U.S. Territories
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 (NLCD), USFS Pacific
Islands Imagery Project, and the National Oceanic and Atmospheric Administration (NOAA) Coastal Change
Analysis Program (C-CAP). Land-cover data can be used to inform a land use classification if there is a time series
to evaluate the dominant 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 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. 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 (see Table 6-7).
Table 6-7: Total Land Area (Hectares) by Land Use Category for U.S. Territories
Northern
Puerto Rico
U.S. Virgin
Islands
Guam
Marianas
Islands
American
Samoa
Total
Cropland
19,712
138
236
289
389
20,764
Forest Land
404,004
13,107
24,650
25,761
15,440
482,962
Grasslands
299,714
12,148
15,449
13,636
1,830
342,777
Other Land
5,502
1,006
1,141
5,186
298
13,133
Settlements
130,330
7,650
11,146
3,637
1,734
154,496
Wetlands
24,525
4,748
1,633
260
87
31,252
Total
883,788
38,796
54,255
48,769
19,777
1,045,385
Note: Totals may not sum due to independent rounding.
6-24 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
Methods in the 2013 Supplement to the 2006 Guidelines for National Greenhouse Gas Inventories: Wetlands (IPCC
2014) have been applied to estimate emissions and removals from coastal wetlands. Specifically, greenhouse gas
emissions from coastal wetlands have been developed for the Inventory using the NOAA C-CAP land-cover product.
The NOAA C-CAP product is not used directly in the land representation analysis, however, so a planned
improvement for future Inventories is to reconcile the coastal wetlands data from the C-CAP product with the
wetlands area data provided in the NRI, FIA and NLCD. Estimates from flooded lands are also included in this
Inventory, but data are not directly used in the land representation analysis at this time; this is a planned
improvement to include for future inventories. In addition, the current Inventory does not include a classification
of managed and unmanaged wetlands, except for remote areas in Alaska. Consequently, there is a planned
improvement to classify managed and unmanaged wetlands for the conterminous United States and Hawaii, and
more detailed wetlands datasets will be evaluated and integrated into the analysis to meet this objective.
6.2 Forest Land Remaining Forest Land
(CRT Category 4A1)
Changes in Forest Carbon Stocks (CRT Category 4A1)
Delineation of Carbon Pools
For estimating carbon stocks or stock change (flux), carbon in forest ecosystems can be divided into the following
five storage pools (IPCC 2006):
• Aboveground biomass, which includes all living biomass above the soil including stem, stump, branches,
bark, seeds, and foliage. This category includes live understory.
• Belowground biomass, which includes all living biomass of coarse living roots greater than 2 millimeters
(mm) diameter.
• Dead wood, which includes all non-living woody biomass either standing, lying on the ground (but not
including litter), or in the soil.
• Litter, which includes all duff, humus, and fine woody debris above the mineral soil as well as woody
fragments with diameters of up to 7.5 cm.
• Soil organic carbon (SOC), including all organic material in soil to a depth of 1 meter but excluding the
coarse roots of the belowground pools. Organic (e.g., peat and muck) soils have a minimum of 12 to 20
percent organic matter by mass and develop under poorly drained conditions of wetlands. All other soils
are classified as mineral soil types and typically have relatively low amounts of organic matter.
In addition, there are two harvested wood pools included when estimating carbon flux:
• Harvested wood products (HWP) in use.
• HWP in solid waste disposal sites (SWDS).
Forest Carbon Cycle
Carbon is continuously cycled among the previously defined carbon storage pools and the atmosphere as a result
of biogeochemical processes in forests (e.g., photosynthesis, respiration, decomposition, and disturbances such as
fires or pest outbreaks) and anthropogenic activities (e.g., harvesting, thinning, and replanting). As trees
photosynthesize and grow, carbon is removed from the atmosphere and stored in living tree biomass. As trees die
Land Use, Land-Use Change, and Forestry 6-25
-------
and otherwise deposit litter and debris on the forest floor, carbon is released to the atmosphere and is also
transferred to the litter, dead wood, and soil pools by organisms that facilitate decomposition.
The net change in forest carbon is not equivalent to the net flux between forests and the atmosphere because
timber harvests do not cause an immediate flux of all harvested biomass carbon to the atmosphere. Instead,
harvesting transfers a portion of the carbon stored in wood to a "product pool." Once in a product pool, the
carbon is emitted over time as CO2 in the case of decomposition and as CO2, Cm, N2O, CO, and NOxwhen the wood
product combusts. The rate of emission varies considerably among different product pools. For example, if timber
is harvested to produce energy, combustion releases carbon immediately, and these emissions are reported for
information purposes in the Energy sector while the harvest (i.e., the associated reduction in forest carbon stocks)
and subsequent combustion are implicitly estimated in the Land Use, Land-Use Change, and Forestry (LULUCF)
sector (i.e., the portion of harvested timber combusted to produce energy does not enter the HWP pools).
Conversely, if timber is harvested and used as lumber in a house, it may be many decades or even centuries before
the lumber decays and carbon is released to the atmosphere. If wood products are disposed of in SWDS, the
carbon contained in the wood may be released many years or decades later or may be stored almost permanently
in the SWDS. These latter fluxes, with the exception of CH4 from wood in SWDS, which is included in the Waste
sector, are also estimated in the LULUCF sector.
Net Change in Carbon Stocks within Forest Land of the United States
This section describes the general method for quantifying the net changes in carbon stocks in the five carbon
storage pools and two harvested wood pools (a more detailed description of the methods and data is provided in
Annex 3.13). The underlying methodology for determining carbon stock and stock change relies on data from the
national forest inventory (NFI) conducted by the Forest Inventory and Analysis (FIA) program within the USDA
Forest Service. The annual NFI is implemented across all U.S. forest lands within the conterminous 48 states,
Alaska, Puerto Rico, and the U.S. Virgin Islands, and periodic inventories are available for Hawaii and some of the
other U.S. Territories. The methods for estimation and monitoring are continuously improved and these
improvements are reflected in the carbon estimates (Domke et al. 2022; Westfall et al. 2023). First, in the
conterminous 48 states and coastal southeast and southcentral Alaska, the total carbon stocks are estimated for
each carbon storage pool at the individual NFI plot, next the annual net changes in carbon stocks for each pool at
the population level are estimated, and then the changes in stocks are summed for all pools to estimate total net
flux at the population level (e.g., U.S. state). Changes in carbon stocks from disturbances, such natural disturbances
(e.g., wildfires, insects/disease, wind) or harvesting, are included in the net changes (see Box 6-3 for more
information). For instance, an inventory conducted after a fire implicitly includes only the carbon stocks remaining
on the NFI plot. The IPCC (2006) recommends estimating changes in carbon stocks from forest lands according to
several land-use types and conversions, specifically forest land remaining forest land and land converted to forest
land, with the former being lands that have been forest lands for 20 years or longer and the latter being lands (i.e.,
croplands, grassland, wetlands, settlements and other lands) that have been converted to forest lands for less than
20 years.
The methods and data used to delineate forest carbon stock changes by these two categories continue to improve
and in order to facilitate this delineation, a combination of estimation approaches was used to compile estimates
in this Inventory. Methods for compiling carbon stocks and stock changes on forest land in interior Alaska are
different from those used for estimation in the conterminous U.S. and coastal Alaska due to the recency of the
operational FIA inventory in that region and differences in sampling protocols (see Annex 3.13 for more details).
Finally, estimates of carbon stocks and stock changes on forest land in Hawaii and the U.S. Territories of American
Samoa, Guam, Northern Mariana Islands, Puerto Rico, and the U.S. Virgin Islands (hereafter referred to as the U.S.
Territories) are included for the first time in this Inventory. The FIA program has conducted annual inventories in
parts of Puerto Rico (Mainland, Vieques, Culebra) and the U.S. Virgin Islands and periodic inventories in Hawaii,
American Samoa, Guam, Northern Mariana Islands, and Puerto Rico (Mona Island). These inventories in
combination with published estimates of carbon stocks, stock changes, and IPCC (2019) default estimates were
used to compile estimates of carbon stocks and stock changes on forest land for these regions (see Annex 3.13 for
more details).
6-26 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
Forest Area in the United States
Approximately 32 percent of the managed U.S. land area is estimated to be forested based on the U.S. definition of
forest land as provided in Section 6.1. All annual and periodic NFI plots included in the public FIA database as of
September 2023 (which includes data collected through 2022 - note that the COVID 19 pandemic resulted in
delays in data collection in many states) were used in this Inventory. The NFIs from the conterminous United States
(USDA Forest Service 2023a, 2023b), Alaska, Hawaii, and the U.S. Territories comprise an estimated 282 million
hectares of forest land that are considered managed and are included in the current Inventory. Some differences
also exist in forest land area estimates from the latest update to the Resources Planning Act (RPA) Assessment
(Oswalt et al. 2019) and the forest land area estimates included in this report, which are based on the annual and
periodic NFI data through 2022 for all states (USDA Forest Service 2023b; Nelson et al. 2020). The methods for
compiling area estimates for Hawaii and the U.S. Territories in this section are different from those in Section 6.1
because they do not rely on FIA data. Also, it is not possible to separate forest land remaining forest land from land
converted to forest land in Wyoming because of the split annual cycle method used for population estimation (see
Annex 3.13). This prevents harmonization of forest land in Wyoming with the NRI/NLCD method used in Section
6.1. Agroforestry systems that meet the definition of forest land are also not currently included in the current
Inventory since they are not explicitly inventoried (i.e., classified as an agroforestry system) by either the FIA
program or the Natural Resources Inventory (NRI) of the USDA Natural Resources Conservation Service (Perry et al.
2005).
An estimated 67 percent (208 million hectares) of U.S. forests in Alaska, Hawaii and the conterminous United
States are classified as timberland, meaning they meet minimum levels of productivity and have not been removed
from production. Approximately ten percent of Alaska forest land and 73 percent of forest land in the
conterminous United States are classified as timberland. Of the remaining non-timberland in the conterminous
United States, Alaska, and Hawaii, nearly 33 million hectares are reserved forest lands (withdrawn by law from
management for production of wood products) and 102 million hectares are lower productivity forest lands
(Oswalt et al. 2019). Historically, the timberlands in the conterminous United States have been more frequently or
intensively surveyed than the forest lands removed from production because they do not meet the minimum level
of productivity.
Since the late 1980s, gross forest land area in Alaska, Hawaii, and the conterminous United States has increased by
about 13 million hectares (Oswalt et al. 2019). The southern region of the United States contains the most forest
land (Figure 6-4). A substantial portion of this accrued forest land is from the conversion of abandoned croplands
to forest (e.g., Woodall et al. 2015b). Estimated forest land area in the conterminous United States and Alaska
represented in this Inventory is stable, but there are substantial conversions as described in 6.1 and each of the
land conversion sections for each land-use category (e.g., land converted to cropland, land converted to grassland).
The major influences on the net carbon flux from forest land across the 1990 to 2022 time series are management
activities, natural disturbance, particularly wildfire, and the ongoing impacts of current and previous land-use
conversions. These activities affect the net flux of carbon by altering the amount of carbon stored in forest
ecosystems and also the area converted to forest land. For example, intensified management of forests that leads
to an increased rate of growth of aboveground biomass (and possible changes to the other carbon storage pools)
may increase the eventual biomass density of the forest, thereby increasing the uptake and storage of carbon in
the aboveground biomass pool.29 Though harvesting forests removes much of the carbon in aboveground biomass
(and possibly changes carbon density in other pools), on average, the estimated volume of annual net growth in
aboveground tree biomass in the conterminous United States is essentially twice the volume of annual removals
on timberlands (Oswalt et al. 2019). The net effects of forest management and changes in forest land remaining
forest land are captured in the estimates of carbon stocks and fluxes presented in this section.
29 The term "biomass density" refers to the mass of live vegetation per unit area. It is usually measured on a dry-weight basis.
Species-specific carbon fractions are used to convert dry biomass to carbon (Westfall et al. 2023).
Land Use, Land-Use Change, and Forestry 6-27
-------
Figure 6-4: Changes in Forest Area by Region for Forest Land Remaining Forest Land in the
conterminous United States and Alaska (1990-2022)
g 80-
TO
o
0
c
o
ro
CD
aj
to
fl)
60-
40-
20-
> South
•North
Pacific
Coast
Rocky
Mountain
1 | i i i i | i i i i | m I I | I I I I | I I I I | i i i i | i
1990 1995 2000 2005 2010 2015 2020
Year
Forest Carbon Stocks and Stock Change
In the forest land remaining forest land category, forest management practices, the regeneration of forest areas
cleared more than 20 years prior to the reporting year, and timber harvesting have resulted in net removal (i.e.,
net sequestration or accumulation) of carbon each year from 1990 through 2022. The rate of forest clearing in the
17th century following European settlement had slowed by the late 19th century. Through the later part of the 20th
century, many areas of previously forested land in the United States were allowed to revert to forests or were
actively reforested. The impacts of these land-use changes still influence carbon fluxes from these forest lands.
More recently, the 1970s and 1980s saw a resurgence of federally sponsored forest management programs (e.g.,
the Forestry Incentive Program) and soil conservation programs (e.g., the Conservation Reserve Program), which
have focused on tree planting, improving timber management activities, combating soil erosion, and converting
marginal cropland to forests. In addition to forest regeneration and management, forest harvests and natural
disturbance have also affected net carbon fluxes. Because most of the timber harvested from U.S. forest land is
used in wood products, and many discarded wood products are disposed of in SWDS rather than by incineration,
substantial quantities of carbon in harvested wood are transferred to these long-term storage pools rather than
being released rapidly to the atmosphere (Skog 2008). By maintaining current harvesting practices and
regeneration activities on forest lands, along with continued input of harvested wood into the HWP pool, carbon
stocks in the forest land remaining forest land category are likely to continue to increase in the near term, though
6-28 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
possibly at a slower rate. Changes in carbon stocks in the forest ecosystem and harvested wood pools associated
with forest land remaining forest land were estimated to result in net removal of 787.0 MMT CO2 Eq. (214.6 MMT
carbon) in 2022 (Table 6-8, Table 6-9, Table A-203, Table A-204 and state-level estimates in Table A-207). The
estimated net uptake of carbon in the Forest Ecosystem was 694.3 MMT CO2 Eq. (189.3 MMT carbon) in 2022
(Table 6-8 and Table 6-9). The majority of this uptake in 2022, 491.7 MMT CO2 Eq. (134.1 MMT carbon), was from
aboveground biomass. Overall, estimates of average carbon density in forest ecosystems (including all pools)
increased consistently over the time series with an average of approximately 208 MT carbon ha 1 from 1990 to
2022. This was calculated by dividing the forest ecosystem carbon stock estimates by the forest land area
estimates for every year (see Table 6-10 and Table A-206) and then calculating the mean across the entire time
series, i.e., 1990 through 2022. The increasing forest ecosystem carbon density, when combined with relatively
stable forest area, results in net carbon accumulation over time. However, due to an aging forest land base,
increases in the frequency and severity of disturbances in forests in some regions, among other drivers of change,
forest carbon density is increasing at a slower rate resulting in an overall decline in the sink strength of forest land
remaining forest land in the United States. Aboveground live biomass is responsible for the majority of net carbon
uptake among all forest ecosystem pools (Figure 6-5). These increases may be influenced in some regions by
reductions in carbon density or forest land area due to natural disturbances (e.g., wildfire, weather,
insects/disease), particularly in Alaska. The inclusion of all managed forest land in Alaska has increased the
interannual variability in carbon stock change estimates over the time series, and much of this variability can be
attributed to severe fire years (e.g., 2022). The distribution of carbon in forest ecosystems in Alaska is substantially
different from forests in the conterminous United States. In Alaska, more than nine percent of forest ecosystem
carbon is stored in the litter carbon pool whereas in the conterminous United States, less than seven percent of
the total ecosystem carbon stocks are in the litter pool. Much of the litter material in forest ecosystems is
combusted during fire (IPCC 2006) leading to substantial carbon losses in this pool during severe fire years (Figure
6-5, Table A-206).
The estimated net accumulation of carbon in the HWP pool, i.e., the balance of additions from the transfer of
harvested wood from the forest ecosystem and losses from the current decay of wood harvested in the past, was
92.8 MMT CO2 Eq. (25.3 MMT carbon) in 2022 (Table 6-8, Table 6-9, Table A-203, and Table A-204). The majority of
this uptake, 63.9 MMT CO2 Eq. (17.4 MMT carbon), was from solid wood and paper in SWDS. Products in use
accounted for an estimated 28.8 MMT CO2 Eq. (7.9 MMT carbon) in 2022.
Table 6-8: Net CO2 Flux from Forest Ecosystem Pools in Forest Land Remaining Forest Land
and Harvested Wood Pools (MMT CO2 Eq.)
Carbon Pool
1990
2005
2018
2019
2020
2021
2022
Forest Ecosystem
(851.0)
(770.0)
(779.6)
(726.2)
(765.2)
(749.5)
(694.3)
Aboveground Biomass
(600.9) ¦
(550.8)
(536.7)
(516.3)
(522.8)
(513.0)
(491.7)
Belowground Biomass
(116.8)
(107.5)
(105.4)
(102.3)
(102.2)
(100.9)
(96.9)
Dead Wood
<132-°) i
(131.2) =
(138.0)
(133.4)
(136.2)
(135.3)
(131.4)
Litter
(2.4)
20.5
(1.5)
26.5
(3.4)
(0.1)
26.4
Soil (Mineral)
2.0 |
(0.8)
1.3
(1.3)
(1.3)
(0.9)
(1.2)
Soil (Organic)
(1.6)
(1.0)
(0.1)
(0.1)
(0.1)
(0.1)
(0.1)
Drained Organic Soil0
0.8 1
1—¦
00
0
0.8
0.8
0.8
0.8
0.8
Harvested Wood
(123.8)
(106.0)
(93.9)
(86.9)
(96.8)
(94.7)
(92.8)
Products in Use
(54.8) 1
(42.6) j:
(28.8)
(22.6)
(32.3)
(30.4)
(28.8)
SWDS
(69.0)
(63.4)
(65.1)
(64.3)
(64.5)
(64.3)
(63.9)
Total Net Flux
(974.8)
(876.0)
(873.5)
(813.2)
(862.0)
(844.2)
(787.0)
Land Use, Land-Use Change, and Forestry 6-29
-------
a These estimates include carbon stock changes from drained organic soils from both forest land remaining forest
land and land converted to forest land. See the section below on C02, CH4, and N20 emissions from drained organic
soils for the methodology used to estimate the C02 emissions from drained organic soils. Also, Table 6-28 and Table
6-29 for non-C02 emissions from drainage of organic soils from both forest land remaining forest land and land
converted to forest land.
Notes: Managed forest land area for Hawaii and the U.S. Territories was compiled using FIA data in this section which
is different from how area estimates for those lands were compiled in Section 6.1. This results in small differences
(less than 0.5 million hectares) in the forest land area estimates in this section and Section 6.1. Also, it is not
possible to separate forest land remaining forest land from land converted to forest land in Wyoming because of
the split annual cycle method used for population estimation, this prevents harmonization of forest land in
Wyoming with the NRI/NLCD method used in Section 6.1. See Annex 3.13, Table A-206 for annual differences
between the forest area reported in Section 6.1 and Section 6.2. The forest ecosystem carbon stock changes do not
include trees on non-forest land (e.g., agroforestry systems and settlement areas—see Section 6.10 for estimates of
carbon stock change from settlement trees). Forest ecosystem carbon stocks on managed forest land in interior
Alaska, Hawaii, and the U.S. Territories were compiled using the gain-loss method as described in Annex 3.13.
Parentheses indicate net carbon uptake (i.e., a net removal of carbon from the atmosphere). Total net flux is an
estimate of the actual net flux between the total forest carbon pool and the atmosphere. Harvested wood
estimates are based on results from annual surveys (see Annex 3.13, Table A-199) and models. Totals may not sum
due to independent rounding.
Table 6-9: Net Carbon Flux from Forest Ecosystem Pools in Forest Land Remaining Forest
Land and Harvested Wood Pools (MMT C)
Carbon Pool
1990
2005
2018
2019
2020
2021
2022
Forest Ecosystem
(232.1)
(210.0)
(212.6)
(198.1)
(208.7)
(204.4)
(189.3)
Aboveground Biomass
(163.9) 1
(150.2) 1
(146.4)
(140.8)
(142.6)
(139.9)
(134.1)
Belowground Biomass
(31.9)
(29.3)
(28.8)
(27.9)
(27.9)
(27.5)
(26.4)
Dead Wood
(36.0) :
(35.8)
(37.6)
(36.4)
(37.1)
(36.9)
(35.8)
Litter
(0.7)
5.6
(0.4)
7.2
(0.9)
(0.0)
7.2
Soil (Mineral)
0-5 ¦
(0.2)
0.4
(0.4)
(0.3)
(0.2)
(0.3)
Soil (Organic)
(0.4)
(0.3)
(0.0)
(0.0)
(0.0)
(0.0)
(0.0)
Drained Organic Soil3
0-2 1
0.2
0.2
0.2
0.2
0.2
0.2
Harvested Wood
(33.8)
(28.9)
(25.6)
(23.7)
(26.4)
(25.8)
(25.3)
Products in Use
(14.9) |
(11.6)
(7.8)
(6.2)
(8.8)
(8.3)
(7.9)
SWDS
(18.8)
(17.3)
(17.8)
(17.5)
(17.6)
(17.5)
(17.4)
Total Net Flux (265.8) (238.9) (238.2) (221.8) (235.1) (230.2) (214.6)
a These estimates include carbon stock changes from drained organic soils from both forest land remaining forest
land and land converted to forest land. See the section below on C02, CH4, and N20 Emissions from Drained
Organic Soils for the methodology used to estimate the carbon flux from drained organic soils. Also, see Table 6-28
and Table 6-29 for greenhouse gas emissions from non-C02 gases changes from drainage of organic soils from
forest land remaining forest land and land converted to forest land.
Notes: Managed forest land area for Hawaii and the U.S. Territories was compiled using FIA data in this section
which is different from how area estimates for those lands were compiled in Section 6.1 so there are small
differences in the forest land area estimates in this Section and Section 6.1. Also, it is not possible to separate
forest land remaining forest land from land converted to forest land in Wyoming because of the split annual cycle
method used for population estimation, this prevents harmonization of forest land in Wyoming with the NRI/NLCD
method used in Section 6.1). See Annex 3.13, Table A-203 for annual differences between the forest area reported
in Section 6.1 and Section 6.2. The forest ecosystem carbon stock changes do not include trees on non-forest land
(e.g., agroforestry systems and settlement areas—see Section 6.10 for estimates of carbon stock change from
settlement trees). Forest ecosystem carbon stocks on managed forest land in Alaska, Hawaii, and the U.S.
Territories were compiled using the gain-loss method as described in Annex 3.13. Parentheses indicate net carbon
uptake (i.e., a net removal of carbon from the atmosphere). Total net flux is an estimate of the actual net flux
between the total forest carbon pool and the atmosphere. Harvested wood estimates are based on results from
annual surveys and models. Totals may not sum due to independent rounding.
6-30 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
Stock estimates for forest ecosystem and harvested wood carbon storage pools are presented in Table 6-10.
Together, the estimated aboveground biomass and soil carbon pools account for a large proportion of total forest
ecosystem carbon stocks. Forest land area estimates are also provided in Table 6-10, but these do not precisely
match those in Section 6.1 for forest land remaining forest land. This is because the forest land area estimates in
Table 6-10 include estimates of managed forest land in Hawaii and the U.S. Territories compiled using FIA
estimates in this section while the area estimates for managed forest land in Hawaii and the U.S. Territories in
Section 6.1 were compiled using different methods. Differences also exist because forest land area estimates are
based on the latest NFI data through 2022, and woodland areas previously included as forest land have been
separated and included in the grassland categories in this Inventory.30
Table 6-10: Forest Area (1,000 ha) and Carbon Stocks in Forest Land Remaining Forest Land
and Harvested Wood Pools (MMT C)
1990
2005
2019
2020
2021
2022
2023
Forest Area (1,000 ha)
283,500
282,521
281,137
281,779
281,780
281,752
281,725
Carbon Pools (MMT C)
U
iiiiiil
u
Forest Ecosystem
55,142
58,536
61,519
61,717
61,926
62,130
62,320
Aboveground Biomass
12,739 =
15,122 '
17,199
17,340
17,483
17,622
17,757
Belowground Biomass
2,255
2,718
3,124
3,151
3,179
3,207
3,233
Dead Wood
1'977 1
2,521
3,038
3,074
3,111
3,148
3,184
Litter
3,789
3,794
3,775
3,767
3,768
3,768
3,761
Soil (Mineral)
28,407 1
28,401 :
28,400
28,400
28,401
28,401
28,401
Soil (Organic)
5,976
5,981
5,983
5,983
5,983
5,983
5,983
Harvested Wood
1,895
2,353
2,671
2,694
2,721
2,747
2,772
Products in Use
1,249
1,447
1,523
1,530
1,538
1,547
1,555
SWDS
646 jjjjjj:
906
1,147
1,165
1,182
1,200
1,217
Total C Stock
57,037
60,890
64,189
64,411
64,647
64,877
65,092
Notes: Managed forest land area for Hawaii and the U.S. Territories was compiled using FIA data in this section which
is different from how area estimates for those lands were compiled in Section 6.1 so there are small differences in
the forest land area estimates in this section and Section 6.1. Also, it is not possible to separate forest land
remaining forest land from land converted to forest land in Wyoming because of the split annual cycle method used
for population estimation, this prevents harmonization of forest land in Wyoming with the NRI/NLCD method used
in Section 6.1 Representation of the U.S. Land Base (CRT Category 4.1). See Annex 3.13, Table A-213 for annual
differences between the forest area reported in Section 6.1 and Section 6.2. The forest ecosystem carbon stocks do
not include trees on non-forest land (e.g., agroforestry systems and settlement areas—see Section 6.10 for
estimates of carbon stock change from settlement trees). Forest ecosystem carbon stocks on managed forest land in
Alaska, Hawaii, and the U.S. Territories were compiled using the gain-loss method as described in Annex 3.13.
Harvested 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. Population estimates compiled using FIA data 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 2022 requires
estimates of carbon stocks for 2022 and 2023.
30 See Annex 3.13, Table A-203 for annual differences between the forest area reported in Section 6.1 Representation of the
U.S. Land Base and Section 6.2 Forest Land Remaining Forest Land.
Land Use, Land-Use Change, and Forestry 6-31
-------
Figure 6-5: Estimated Net Annual Changes in Carbon Stocks for All Carbon Pools in Forest
Land Remaining Forest Land in the Conterminous United States and Alaska (1990-2022)
0) v
? £
— o
In i—
s? §
£ 2
:l ™
§ "5
c 3
JS 2 greenhouse gas emissions from forest fires that are also
quantified in a separate section below as required by IPCC Guidance and the UNFCCC.
Emissions estimates are developed using IPCC (2019) methodology and based on U.S.-specific data and models
to quantify the primary fire-specific components: area burned; availability and combustibility of fuel; fire
severity (or consumption); and CO2 and non-CC>2 emissions. Estimated CO2 emissions for fires on forest lands in
the United States for 2022 are 129.2 MMT CO2 per year (Table 6-11). This estimate is an embedded component
of the net annual forest carbon stock change estimates provided previously (i.e., Table 6-9), but this separate
approach to estimating CO2 emissions is necessary in order to associate these emissions with fire. See the
discussion in Annex 3.13 for more details on this methodology. Note that in Alaska, a portion of the forest lands
are considered unmanaged, therefore the estimates for Alaska provided in Table 6-11 include only managed
forest land within the state, which is consistent with carbon stock change estimates provided above.
6-32 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
Table 6-11: Estimates of CO2 (MMT per Year) Emissions3 from Forest Fires in the
Conterminous 48 States, Hawaii, Puerto Rico, Guam, and Alaska
1990
2005
2018
2019
2020
2021
2022
C02 emitted from fires on forest land in
the Conterminous 48 States, Hawaii,
Puerto Rico, and Guam (MMT yr1)
11.9
28.6
77.5
19.1
124.0
156.7
71.8
C02 emitted from fires on managed
forest land in Alaska (MM Tyr1)
43.2
113.5
7.1
34
0.4
6.8
57.4
Total C02 emitted (MMTyr1)
55.1
142.2
84.6
53
124.4
163.5
129.2
a These emissions have already been included in the estimates of net annual changes in carbon stocks, which include the
amount sequestered minus any emissions, including the assumption that combusted wood may continue to decay
through time.
Note: Totals may not sum due to independent rounding.
Methodology and Time-Series Consistency
The methodology described herein is consistent with the 2006IPCC Guidelines for National Inventories. Forest
ecosystem carbon stocks and net annual carbon stock change were determined according to the stock-difference
method for the conterminous United States and coastal southeast and southcentral Alaska, which involved
applying carbon estimation factors to annual forest inventories across time to obtain carbon stocks and then
subtracting between the years to obtain the stock change. The gain-loss method was used to estimate carbon
stocks and net annual carbon stock changes in interior Alaska, Hawaii, and the U.S. Territories. The approaches for
estimating carbon stocks and stock changes on forest land remaining forest land are described in Annex 3.13. All
annual and periodic NFI plots available in the public FIA database (USDA Forest Service 2023b) were used in the
current Inventory. Additionally, NFI plots established and measured in 2014 as part of a pilot inventory in interior
Alaska were also included in this Inventory as were plots established and measured since 2015 as part of the
operational NFI in interior Alaska. Some of the data from the pilot and operational NFI in interior Alaska are not yet
available in the public FIA database. Only plots which meet the definition of forest land (see Section 6.1) are
measured in the NFI; as part of the pre-field process in the FIA program, all plots or portions of plots (i.e.,
conditions) are classified into a land-use category. This land use information on each forest and non-forest plot
was used to estimate forest land area and land converted to and from forest land over the time series. The
estimates in this section of the report are based on land use information from the NFI and they may differ from the
other land-use categories where area estimates reported in the Land Representation were not updated (see
Section 6.1). Further, managed forest land area estimates for Hawaii and the U.S. Territories were compiled using
FIA data in this section which is different from how estimates for these lands were compiled in Section 6.1 (see
Annex 3.13 for details on differences).
To implement the stock-difference approach, forest land conditions in the conterminous United States and coastal
Alaska were observed on NFI plots at time to and at a subsequent time ti=to+s, where s is the time step (time
measured in years) and is indexed by discrete (e.g., 5 year) forest age classes. The inventory from to to ti was then
projected to 2022. This projection approach requires simulating changes in the age-class distribution resulting from
forest aging and disturbance events and then applying carbon density estimates for each age class to obtain
population estimates for the nation. In cases where there are ti estimates in the last year (e.g., 2022) of the NFI no
projections are necessary for those plots.
To implement the gain-loss approach in interior Alaska, forest land conditions in Alaska were observed on NFI plots
from 2014 to 2022. Plot-level data from the NFI were harmonized with auxiliary data describing climate, forest
structure, disturbance, and other site-specific conditions to develop non-parametric models to predict carbon
stocks by forest ecosystem carbon pool as well as fluxes over the entire inventory period, 1990 to 2022. First,
carbon stocks for each forest ecosystem carbon pool were predicted for the year 2016 for all NFI plot locations
Land Use, Land-Use Change, and Forestry 6-33
-------
(each plot representing 12,015 ha). Next, the chronosequence of sampled NFI plots and auxiliary information (e.g.,
climate, forest structure, disturbance, and other site-specific data) were used to predict annual gains and losses for
each forest ecosystem carbon pool. The annual gains and losses were then combined with the stock estimates and
disturbance information to compile plot- and population-level carbon stocks and fluxes for each year from 1990 to
2022.
To implement the gain-loss approach in Hawaii and the U.S. Territories, a combination of Tier 1 and Tier 2 methods
were applied. All forest land conditions were observed on annual and periodic NFI plots from 2001 to 2019 (see
Annex 3.13 for specific inventories included for each Island). Plot-level data from the NFI were harmonized with
data describing ecological zone (FAO 2010), soil attributes (Johnson and Kern 2003; Deenik and McClellan, 2007,
IPCC 2019), and dead wood and litter carbon stocks (Oswalt et al. 2008; IPCC 2019). Only estimates of carbon
stocks in live trees were consistently available in the NFI for Hawaii and the U.S. Territories for each inventory.
These estimates were used to obtain average annual carbon stock change estimates for above and belowground
live trees which were applied to each forest plot to capture growth, harvest removals, and mortality. The carbon
stocks and annual stock change estimates were compared with country-specific estimates (Oswalt et al. 2008;
Selmants et al. 2017), and IPCC (2019) default estimates to ensure they were consistent with other sources. There
were limited data available on disturbances and management activities on NFI plots over the times series so Tier 1
methods were applied for dead wood and litter. It was assumed that the average transfer rate into dead wood and
litter pools is equivalent to the average transfer rate out of the dead organic matter pool so there are no net
carbon stock changes included for these pools in the time series (IPCC 2006). Similarly, given data limitations on
forest soils and changes on NFI plots over the time series, a Tier 1 approach was also used for soil carbon with
country-specific estimates (Johnson and Kern 2003) and IPCC (2019) defaults used to estimate soil carbon stocks
with no net carbon stock change reported.
To estimate carbon stock changes in harvested wood, estimates were based on factors such as the allocation of
wood to various primary and end-use products as well as half-life (the time at which half of the amount placed in
use will have been discarded from use) and expected disposition (e.g., product pool, SWDS, combustion). An
overview of the different methodologies and data sources used to estimate the carbon in forest ecosystems within
the conterminous United States and Alaska and harvested wood products for all of the United States is provided
below. See Annex 3.13 for details and additional information related to the methods and data.
Methodological recalculations were applied to the entire time series to ensure time-series consistency from 1990
through 2022. Details on the emission/removal trends and methodologies through time are described in more
detail in the Introduction and Methodology sections.
Forest Ecosystem Carbon from Forest Inventory
The United States applied the compilation approach described in Woodall et al. (2015a) for the current Inventory
which removes the older periodic inventory data, which may be inconsistent with annual inventory data, from the
estimation procedures. This approach enables the attribution of forest carbon accumulation by forest growth,
land-use change, and natural disturbances such as fire. Development will continue on a system that attributes
changes in forest carbon to disturbances and delineates land converted to forest land from forest land remaining
forest land. As part of this development, carbon pool science will continue and will be expanded to improve the
estimates of carbon stock transfers from forest land to other land uses and include techniques to better identify
land-use change (see the Planned Improvements section below).
Unfortunately, the annual FIA inventory system does not extend into the 1970s, necessitating the adoption of a
system to estimate carbon stocks prior to the establishment of the annual forest inventory. The estimation of
carbon stocks prior to the annual national forest inventory consisted of a modeling framework comprised of a
forest dynamics module (age transition matrices) and a land use dynamics module (land area transition matrices).
The forest dynamics module assesses forest uptake, forest aging, and disturbance effects (e.g., disturbances such
as wind, fire, and floods identified by foresters on inventory plots). The land use dynamics module assesses carbon
stock transfers associated with afforestation and deforestation (Woodall et al. 2015b). Both modules are
developed from land use area statistics and carbon stock change or carbon stock transfer by age class. The
6-34 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
required inputs are estimated from more than 625,000 forest and non-forest observations recorded in the FIA
national database (U.S. Forest Service 2023a, b, c). Model predictions prior to the annual inventory period are
constructed from the estimation system using the annual estimates. The estimation system is driven by the annual
forest inventory system conducted by the FIA program (Frayer and Furnival 1999; Bechtold and Patterson 2005;
Westfall et al. 2022; USDA Forest Service 2023d, 2023a). The FIA program relies on a rotating panel statistical
design with a sampling intensity of one 674.5 m2 ground plot per 2,403 ha of land and water area. A five or seven-
panel design, with 20 percent or 14.3 percent of the field plots typically measured each year within a state, is used
in the eastern United States and a ten-panel design, with typically ten percent of the field plots measured each
year within a state, is used in the western United States. The interpenetrating hexagonal design across the U.S.
landscape enables the sampling of plots at various intensities in a spatially and temporally unbiased manner.
Typically, tree and site attributes are measured with higher sample intensity while other ecosystem attributes such
as downed dead wood are sampled during summer months at lower intensities. The first step in incorporating FIA
data into the estimation system is to identify annual and periodic inventory datasets by state and U.S. Territory.
Inventories include data collected on permanent inventory plots on forest lands and were organized as separate
datasets, each representing a complete inventory, or survey, of an individual state at a specified time. Many of the
annual inventories reported for states are represented as "moving window" averages, which mean that a
portion—but not all—of the previous year's inventory is updated each year (USDA Forest Service 2023d). Forest
carbon estimates are organized according to these state surveys, and the frequency of surveys varies by state.
Using this FIA data, separate estimates were prepared for the five carbon storage pools identified by IPCC (2006) as
described above. All estimates for the conterminous United States and Alaska were based on data collected from
the extensive array of permanent, annual forest inventory plots and associated models (e.g., live tree belowground
biomass) in the United States (USDA Forest Service 2023b, 2023c). Carbon conversion factors were applied at the
disaggregated level of each inventory plot and then appropriately expanded to population estimates. Only live
(and in some cases) standing dead wood estimates are available in the annual and periodic FIA inventories in
Hawaii and the U.S. Territories. For this reason, a combination of approaches was used to obtain estimates for all
carbon pools for the time series in these locations.
Carbon in Biomass
Live tree carbon pools include aboveground and belowground (coarse root) biomass of live trees with diameter at
breast height (dbh) of at least 2.54 cm at 1.37 m above the litter. Separate estimates were made for above- and
belowground biomass components. Over the last decade, the USDA Forest Service's FIA program and collaborators
from universities and industry have been developing a new national methodology for the prediction of individual-
tree volume, biomass, and carbon content. The resulting methodology is referred to as the National-Scale Volume
and Biomass (NSVB) framework. The previous methodology used was the Component Ratio Method (CRM)
framework (Woodall et al. 2010). While CRM was nationally consistent, tree biomass was still based on the volume
predicted by regional models and tree carbon was assumed to be 50-percent of biomass, regardless of species.
Hence, the need for NSVB, a nationally consistent methodology for compatible predictions of tree volume,
biomass, and carbon content (Westfall et al. In press).
The NSVB covers timber tree species in the conterminous United States and coastal Alaska. All other trees (i.e.,
trees that are woodland species and trees within Pacific and Caribbean Islands) use regional models for volume
and biomass, with updated carbon fractions (when available). While NSVB did not directly update models for trees
that are considered woodland species or trees within the Pacific (USDA Forest Service 2022a, b) and Caribbean
Islands (collectively referred to hereafter as "non-NSVB trees"), volume, biomass, and carbon estimates for these
trees have also changed. For non-NSVB trees, the standardization of tree defects and how variables are reported
(i.e., whether models for total-stem or merchantable-bole volumes are available) may be reflected as differences
in volume estimates. Additionally, biomass estimates for non-NSVB trees are based on regional biomass models
and no longer are adjusted as they were under the CRM. Finally, updates to carbon fractions (when available) and
calculation of aboveground biomass are reflected in aboveground and belowground biomass carbon estimates
(see Recalculations section and Annex 3.13 for more details).
Land Use, Land-Use Change, and Forestry 6-35
-------
Understory vegetation is a minor component of biomass, which is defined in the FIA program as all biomass of
undergrowth plants in a forest, including woody shrubs and trees less than 2.54 cm dbh. For this Inventory, it was
assumed that ten percent of total understory carbon mass is belowground (Smith et al. 2006). Estimates of carbon
density were based on information in Birdsey (1996) and tree biomass estimates from the FIADB. Understory
biomass represented over one percent of carbon in biomass, but its contribution rarely exceeded two percent of
the total carbon stocks or stock changes across all forest ecosystem carbon pools each year.
Carbon in Dead Organic Matter
Dead organic matter is calculated as three separate pools—standing dead trees, downed dead wood, and litter—
with carbon stocks estimated from sample data or from models as described below. The standing dead tree carbon
pool includes aboveground and belowground (coarse root) biomass for trees of at least 2.54 cm dbh. Calculations
followed the basic methods applied to live trees (Westfall et al. 2023) with additional modifications to account for
decay and structural loss (Harmon et al. 2011). Downed dead wood estimates are based on measurement of a
subset of FIA plots for downed dead wood (Domke et al. 2013; Woodall and Monleon 2008; Woodall et al. 2013).
Downed dead wood is defined as pieces of dead wood greater than 7.5 cm diameter, at transect intersection, that
are not attached to live or standing dead trees. This includes stumps and roots of harvested trees. To facilitate the
downscaling of downed dead wood carbon estimates from the state-wide population estimates to individual plots,
downed dead wood models specific to regions and forest types within each region are used. Litter carbon is the
pool of organic carbon (also known as duff, humus, and fine woody debris) above the mineral soil and includes
woody fragments with diameters of up to 7.5 cm. A subset of FIA plots are measured for litter carbon. A modeling
approach, using litter carbon measurements from FIA plots (Domke et al. 2016), was used to estimate litter carbon
for every FIA plot used in the estimation framework. These estimates are now available in the FIADB (USDA Forest
Service 2023b).
Carbon in Forest Soil
Soil carbon is the largest terrestrial carbon sink with much of that carbon in forest ecosystems. The FIA program
has been consistently measuring soil attributes as part of the annual inventory since 2001 and has amassed an
extensive inventory of soil measurement data on forest land in the conterminous United States and coastal Alaska
(O'Neill et al. 2005). Observations of mineral and organic soil carbon on forest land from the FIA program and the
International Soil Carbon Monitoring Network were used to develop and implement a model framework that
enabled the prediction of mineral and organic (i.e., undrained organic soils) soil carbon to a depth of 100 cm from
empirical measurements collected on sample plots at a depth of 20 cm and included site-, stand-, and climate-
specific variables that yield predictions of soil carbon stocks specific to forest land in the United States (Domke et
al. 2017). These estimates are now available in the FIADB (USDA Forest Service 2023b). This approach allowed for
separation of mineral and organic soils, the latter also referred to as Histosols, in the forest land remaining forest
land category. Note that mineral and organic (i.e., undrained organic soils) soil carbon stock changes are reported
to a depth of 100 cm for forest land remaining forest land to remain consistent with past reporting in this category,
however for consistency across land-use categories, mineral (e.g., cropland, grassland, settlements) soil carbon is
reported to a depth of 30 cm in Section 6.3 Land Converted to Forest Land. Estimates of carbon stock changes
from organic soils shown in Table 6-8 and Table 6-9 include the emissions from drained organic forest soils, and
the methods used to develop these estimates can be found in the Drained Organic Soils section below.
Harvested Wood Carbon
Estimates of the HWP contribution to forest carbon sinks and emissions (hereafter called "HWP contribution")
were based on methods described in Skog (2008) using the WOODCARB II model. These methods are based on
IPCC (2006) guidance for estimating the HWP contribution. IPCC (2006) provides methods that allow for reporting
of the HWP contribution using one of several different methodological approaches: Production, stock change and
atmospheric flow, as well as a default method that assumes there is no change in HWP carbon stocks (see Annex
3.13 for more details about each approach). The United States uses the production approach to report HWP
6-36 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
contribution. Under the production approach, carbon in exported wood was estimated as if it remains in the
United States, and carbon in imported wood was not included in the estimates. Though reported, U.S. HWP
estimates are based on the production approach, estimates resulting from use of the two alternative approaches,
the stock change and atmospheric flow approaches, are also presented for comparison (see Annex 3.13). Annual
estimates of change were calculated by tracking the annual estimated additions to and removals from the pool of
products held in end uses (i.e., products in use such as housing or publications) and the pool of products held in
SWDS. The carbon loss from harvest is reported in the forest ecosystem component of the forest land remaining
forest land and land converted to forest land sections and for informational purposes in the Energy sector, but the
non-CC>2 emissions associated with biomass energy are included in the Energy sector emissions (see Chapter 3).
EPA includes HWP within the forest chapter because forests are the source of wood that goes into the HWP
estimates.
Solidwood products include lumber and panels. End-use categories for solidwood include single and multifamily
housing, alteration and repair of housing, and other end uses. There is one product category and one end-use
category for paper. Additions to and removals from pools were tracked beginning in 1900, with the exception of
additions of softwood lumber to housing, which began in 1800. Solidwood and paper product production and
trade data were taken from USDA Forest Service and USDC Bureau of the Census, among other sources (Hair and
Ulrich 1963; Hair 1958; USDC Bureau of Census 1976; Ulrich 1985,1989; Steer 1948; AF&PA 2006a, 2006b; Howard
2003, 2007; Howard and Jones 2016; Howard and Liang 2019; AF&PA 2021; AF&PA 2023; FAO 2023). Estimates for
disposal of products reflects the change over time in the fraction of products discarded to SWDS (as opposed to
burning or recycling) and the fraction of SWDS that were in sanitary landfills versus dumps.
There are five annual HWP variables that were used in varying combinations to estimate HWP contribution using
any one of the three main approaches listed above. These are:
(IA) annual change of carbon in wood and paper products in use in the United States,
(IB) annual change of carbon in wood and paper products in SWDS in the United States,
(2A) annual change of carbon in wood and paper products in use in the United States and other countries
where the wood came from trees harvested in the United States,
(2B) annual change of carbon in wood and paper products in SWDS in the United States and other countries
where the wood came from trees harvested in the United States,
(3) Carbon in imports of wood, pulp, and paper to the United States,
(4) Carbon in exports of wood, pulp and paper from the United States, and
(5) Carbon in annual harvest of wood from forests in the United States.
The sum of variables 2A and 2B yielded the estimate for HWP contribution under the production estimation
approach. A key assumption for estimating these variables that adds uncertainty in the estimates was that
products exported from the United States and held in pools in other countries have the same half-lives for
products in use, the same percentage of discarded products going to SWDS, and the same decay rates in SWDS as
they would in the United States.
Uncertainty
A quantitative uncertainty analysis placed bounds on the flux estimates for forest ecosystems through a
combination of sample-based and model-based approaches to uncertainty estimation for forest ecosystem CO2
flux using IPCC Approach 1 (Table 6-12 and Table A-214 for state-level uncertainties). A Monte Carlo stochastic
simulation of the methods described above, and probabilistic sampling of carbon conversion factors, were used to
determine the HWP uncertainty using IPCC Approach 2. See Annex 3.13 for additional information. The 2022 net
annual change for forest carbon stocks was estimated to be between -866.5 and -708.3 MMT CO2 Eq. around a
central estimate of-787.0 MMT CO2 Eq. at a 95 percent confidence level. This includes a range of-769.6 to -618.9
MMT CO2 Eq. around a central estimate of-694.3 MMT CO2 Eq. for forest ecosystems and -118.0 to -70.0 MMT
Land Use, Land-Use Change, and Forestry 6-37
-------
C02 Eq. around a central estimate of-92.8 MMT CO2 Eq. for HWP.
Table 6-12: Quantitative Uncertainty Estimates for Net CO2 Flux from Forest Land Remaining
Forest Land: Changes in Forest Carbon Stocks (MMT CO2 Eq. and Percent)
Uncertainty Range Relative to Flux Estimate
(MMT CO2 Eq.) (%)
Lower Upper Lower Upper
Bound Bound Bound Bound
Forest Ecosystem C Pools3 C02 (694.3) (769.6) (618.9) -10.9% +10.9%
Harvested Wood Products" C02 (92.8) (118.0) (70.0) -27.2% +24.6%
Total Forest CO2 (787.0) (866.5) (708.3) -10.1% +10.0%
a Range of flux estimates predicted through a combination of sample-based and model-based uncertainty for a 95 percent
confidence interval, IPCC Approach 1.
b Range of flux estimates predicted by Monte Carlo stochastic simulation for a 95 percent confidence interval, IPCC
Approach 2.
Notes: Parentheses indicate negative values or net uptake. Totals may not sum due to independent rounding.
QA/QC and Verification
The FIA program has conducted consistent forest surveys based on extensive statistically-based sampling of most
of the forest land in the conterminous U.S., dating back to 1952. The FIA program includes numerous quality
assurance and quality control (QA/QC) procedures, including calibration among field crews, duplicate surveys of
some plots, and systematic checking of recorded data. Because of the statistically-based sampling, the large
number of survey plots, and the quality of the data, the survey databases developed by the FIA program form a
strong foundation for carbon stock estimates. Field sampling protocols, summary data, and detailed inventory
databases are archived and are publicly available (USDA Forest Service 2023d).
General quality control procedures were used in performing calculations to estimate carbon stocks based on
survey data. For example, the carbon datasets, which include inventory variables such as areas and volumes, were
compared to standard inventory summaries such as the forest resource statistics of Oswalt et al. (2019) or selected
population estimates generated from the FIA database, which are available at an FIA internet site (USDA Forest
Service 2023b). Agreement between the carbon datasets and the original inventories is important to verify
accuracy of the data used.
Additional verification analyses are currently underway to compare forest carbon stock change estimates
developed using the NSVB model to estimates stemming from other forest biomass models as well as remote
sensing imagery.
Estimates of the HWP variables and the HWP contribution under the production estimation approach use data
from USDC Bureau of the Census and USDA Forest Service surveys of production and trade, among other sources
(Hair and Ulrich 1963; Hair 1958; USDC Bureau of Census 1976; Ulrich 1985,1989; Steer 1948; AF&PA 2006a,
2006b; Howard 2003, 2007; Howard and Jones 2016; Howard and Liang 2019; AF&PA 2021; AF&PA 2023; FAO
2023). Factors to convert wood and paper to units of carbon are based on estimates by industry and U.S. Forest
Service published sources (see Annex 3.13). The WOODCARB II model uses estimation methods suggested by IPCC
(2006). Estimates of annual carbon change in solidwood and paper products in use were calibrated to meet two
independent criteria. The first criterion is that the WOODCARB II model estimate of carbon in houses standing in
2001 needs to match an independent estimate of carbon in housing based on U.S. Census and USDA Forest Service
survey data. Meeting the first criterion resulted in an estimated half-life of about 80 years for single family housing
built in the 1920s, which is confirmed by other U.S. Census data on housing. The second criterion is that the
WOODCARB II model estimate of wood and paper being discarded to SWDS needs to match EPA estimates of
discards used in the Waste sector each year over the period 1990 to 2000 (EPA 2006). These criteria help reduce
uncertainty in estimates of annual change in carbon in products in use in the United States and, to a lesser degree,
2022 Flux Estimate
Source Gas
(MMT CO . Eq.)
6-38 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
reduce uncertainty in estimates of annual change in carbon in products made from wood harvested in the United
States. In addition, WOODCARB II landfill decay rates have been validated by ensuring that estimates of CFU
emissions from landfills based on EPA (2006) data are reasonable in comparison to CFU estimates based on
WOODCARB II landfill decay rates.
Recalculations Discussion
There were several methodological improvements implemented in the current Inventory which have resulted in
substantial changes when compared to the previous (1990 through 2021) Inventory.
First, there were new FIA data included for several states, in some cases, multiple years of new data in this
Inventory resulting from delays that occurred due to the global pandemic. Delays still exist in some states so it is
possible that multiple years of data may be available in the years ahead leading to small changes in forest
ecosystem carbon stocks and stock changes throughout the time series. These changes are most notable in the
conterminous United States (Table 6-14) and coastal southeast and southcentral Alaska (Table 6-15). In coastal
Alaska, remeasurement data from the FIA program facilitated the use of the compilation system used in most of
the conterminous United States, improving consistency and facilitating the disaggregation of forest land
conversions in this region for the first time. This transition in compilation methodology resulted in a 10.6 percent
increase in forest land area (449,287 ha) in coastal Alaska which contributed, in part, to increases in forest
ecosystem carbon stocks across all pools (Table 6-15).
This Inventory also implemented new methods for estimating standing live and dead aboveground biomass carbon
in the FIA program (Westfall et al. 2023). These new methods, leveraging the newly developed national-scale
volume and biomass framework (NSVB), represent nearly a decade of research and development in the FIA
program. The new methods: 1) greatly simplify predictions of aboveground biomass because only five model
specifications are used nationally instead of dozens of species- and species-group specific models used in each
region and/or state, 2) eliminate administrative boundaries (e.g., regions or states) in favor of ecologically-based
regions (i.e., ecodivisons) to capture variation in tree size and volume (or biomass) within species or species
groups, 3) models are based on tree measurements from in-situ data which also facilitates more accurate
quantification of model uncertainty, 4) result in consistent model behavior for all tree species and sizes, and 5) use
species-specific carbon fractions for biomass to carbon conversions compared to the previous method which
assumed a default 50 percent biomass to carbon fraction.
The implementation of the NSVB models resulted in an increase in estimates of aboveground biomass carbon
stocks on forest land in the United States of approximately 11 percent (1,761.9 MMT C) in the current Inventory for
the year 2022 relative to the previous Inventory estimate for the year 2022 (Table 6-13) and accounted for 34
percent of the total increase in estimates of forest carbon stocks in this Inventory relative to the previous Inventory
for the same year (Table 6-13). These increases can largely be attributed to more accurate characterization of top
and limb biomass in the new models (Westfall et al. 2023). This also led to small increases in estimates of
belowground biomass carbon stocks since that model is based on a ratio of aboveground biomass. There were also
increases of more than 11 percent (321 MMT C) in estimates of dead wood carbon stocks due to the
implementation of the new NSVB models in the FIA program for standing dead trees (Table 6-13), which accounted
for 6.2 percent of the increases in estimates of forest ecosystem carbon stocks in this Inventory relative to the
previous Inventory for the same year.
The litter and soil model framework used in this Inventory and implemented across the entire time series was
formally adopted in the FIA program this year and predictions compiled using this framework are now available in
the public FIA database (USDA Forest Service 2023b). As part of the formal adoption of these methods in the FIA
program, all variables and associated datasets used in the models were evaluated. New climate normals for the
time period 1991 to 2020 were included in both the litter and soil models using PRISM data for the conterminous
United States and climate normals for 1981 to 2010 (the only period available) were included for the first time in
coastal Alaska. Collectively, these updates and improvements resulted in carbon stock estimates that were 10.1
percent larger (3,150.7 MMT C) in this Inventory relative to the previous Inventory for mineral and organic soil in
the United States and overall accounted for approximately 48 percent of the total increases in forest carbon stock
Land Use, Land-Use Change, and Forestry 6-39
-------
in the current Inventory for the year 2022 when compared to the previous Inventory for the same year (Table 6-13,
Table 6-14). In coastal Alaska, there were comparable increases in predictions in the Inventory relative to previous
Inventories. These increases can be attributed to increases in estimates of forest land area in coastal Alaska as well
as the incorporation of climate data in the litter and soil models. Estimates of litter carbon stocks also decreased
slightly on forest land remaining forest land relative to the previous Inventory (Table 6-13) due to the
implementation of the NSVB models where the litter model relies on estimates of aboveground biomass as a
model parameter. Finally, new data on wildfire in Interior Alaska in the latest Inventory also contributed to
updated estimates in the time series for this area of forest land where there were decreases in the estimates of all
but the organic soil carbon pools (Table 6-16). Collectively these updates and improvements resulted in a 2.3
percent decrease (119.1 MMT C) in estimates of litter carbon stocks on forest land in the United States
Managed forest land in Hawaii and several U.S. Territories31 were included for the first time in the current
Inventory which resulted in an increase in managed forest land area of approximate 1.3 M ha and associated
increases in carbon stocks of 286 MMT C for the year 2023 in this Inventory. While the inclusion of these forest
land areas represents a relatively small increase in forest area and forest ecosystem carbon stocks overall, their
inclusion represents an important improvement toward completeness in this Inventory.
Table 6-13: Recalculations of Forest Area (1,000 ha) and Carbon Stocks in Forest Land
Remaining Forest Land and Harvested Wood Pools (MMT C)
2022 Estimate,
2022 Estimate,
2023 Estimate,
Previous Inventory
Current Inventory
Current Inventory
Forest Area (1000 ha)
279,800
281,752
281,725
Carbon Pools (MMT C)
Forest
56,951
62,130
62,320
Aboveground Biomass
15,861
17,622
17,757
Belowground Biomass
3,143
3,207
3,233
Dead Wood
2,827
3,148
3,184
Litter
3,888
3,768
3,761
Soil (Mineral)
25,916
28,401
28,401
Soil (Organic)
5,317
5,983
5,983
Harvested Wood
2,749
2,747
2,772
Products in Use
1,549
1,547
1,555
SWDS
1,200
1,200
1,217
Total Stock 59,701 64,877 65,092
Note: Totals may not sum due to independent rounding.
Table 6-14: Recalculations of Forest Area (1,000 ha) and Carbon Stocks in Forest Land
Remaining Forest Land (MMT C) in the Conterminous United States
2022 Estimate,
2022 Estimate,
2023 Estimate,
Previous Inventory
Current Inventory
Current Inventory
Forest Area (1000 ha)
249,821
250,036
249,999
Carbon Pools (MMT C)
Forest
46,575
51,195
51,391
Aboveground Biomass
14,947
16,641
16,773
Belowground Biomass
2,959
3,006
3,032
Dead Wood
2,558
2,883
2,920
Litter
2,474
2,383
2,384
Soil (Mineral)
23,086
25,467
25,467
Soil (Organic)
550
815
815
31 American Samoa, Guam, Norther Mariana Islands, U.S. Virgin Islands, and Puerto Rico
6-40 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
Note: Totals may not sum due to independent rounding.
Table 6-15: Recalculations of Forest Area (1,000 ha) and Carbon Stocks in Forest Land
Remaining Forest Land (MMT C) in Coastal Southeast and Southcentral Alaska
2022 Estimate,
2022 Estimate,
2023 Estimate,
Previous Inventory
Current Inventory
Current Inventory
Forest Area (1000 ha)
4,222
4,671
4,681
Carbon Pools (MMT C)
Forest
1,235
1,458
1,460
Aboveground Biomass
362
421
423
Belowground Biomass
76
85
85
Dead Wood
94
107
107
Litter
129
130
130
Soil (Mineral)
411
470
470
Soil (Organic)
163
245
245
Note: Totals may not sum due to independent rounding.
Table 6-16: Recalculations of Forest Area (1,000 ha) and Carbon Stocks in Forest Land
Remaining Forest Land (MMT C) in Interior Alaska
2022 Estimate,
2022 Estimate,
2023 Estimate,
Previous Inventory
Current Inventory
Current Inventory
Forest Area (1000 ha)
25,758
25,758
25,758
Carbon Pools (MMT C)
Forest
9,142
9,192
9,183
Aboveground Biomass
551
485
484
Belowground Biomass
107
92
92
Dead Wood
176
154
153
Litter
1,285
1,248
1,240
Soil (Mineral)
2,419
2,308
2,308
Soil (Organic)
4,604
4,905
4,905
Note: Totals may not sum due to independent rounding.
Table 6-17: Recalculations of Forest Area (1,000 ha) and Carbon Stocks in Forest Land
Remaining Forest Land (MMT C) in Hawaii and United States Territories
2022 Estimate,
2022 Estimate,
2023 Estimate,
Previous Inventory
Current Inventory
Current Inventory
Forest Area (1000 ha)
NE
1,287
1,287
Carbon Pools (MMT C)
Forest
NE
285
286
Aboveground Biomass
NE
76
77
Belowground Biomass
NE
23
24
Dead Wood
NE
4
4
Litter
NE
7
7
Soil (Mineral)
NE
156
156
Soil (Organic)
NE
19
19
NE (Not Estimated)
Note: Totals may not sum due to independent rounding.
The new FIA data and methodological improvements described throughout this text and specifically in this section
on recalculations of estimates of forest ecosystem carbon stocks extend to forest ecosystem carbon stock changes
in the current Inventory. In total, estimates for the forest land remaining forest land sink increased 21.4 percent
(sink estimates increased by 40.8 MMT C). On average across the time-series, these recalculations resulted in a
Land Use, Land-Use Change, and Forestry 6-41
-------
21.7 percent increase in estimates of the forest sink across the time series (sink estimates increased on average,
43.4 MMT C over the time series) relative to the previous Inventory (Table 6-18).
Changes in estimates of forest ecosystem carbon stock changes accounted for most of the increases between this
Inventory and the previous Inventory (Table 6-18) and of those, new data and improvements in methods in the
conterminous United States accounted for 97 percent (-41.8 MMT C) increases in estimates of forest ecosystem
carbon stock changes in the current Inventory relative to the same year in the previous Inventory (Table 6-19).
Estimates of carbon stock changes in the aboveground biomass pool increased by 25.2 percent in the current
Inventory relative to the same year in the previous Inventory and accounted for 65.2 percent (-27.2 MMT C) of the
increase in estimates from this Inventory (Table 6-19). These changes can be directly attributed to the
implementation of the NSVB models in the FIA program. These increases extend to the belowground biomass pool
where the increases in estimates of aboveground biomass resulted in increases in the estimates of belowground
biomass by 23.8 percent (-5.1 MMT C). These increases accounted for 12.2 percent of the total increase in carbon
stock changes in the forest ecosystem pools in this Inventory. There were also substantial increases in dead wood
carbon stock changes which can also be attributed to the implementation of the NSVB models for standing dead
trees (Westfall et al. In press). There was a 34.4 percent increase (-9.5 MMT C) in estimates of dead wood carbon
stock changes between this Inventory and the same year in the previous Inventory. This increase in the estimates
of dead wood accounts for 22.8 percent of the total increases in estimates of forest ecosystem carbon stock
changes in this Inventory relative to the same year in the previous Inventory (Table 6-19).
There were also small differences in the estimates of carbon stock changes for the litter and the soil carbon pools
in the conterminous United States, coastal and Interior Alaska (Table 6-19, Table 6-20, Table 6-21). These changes
were all relatively small when compared to changes in live and standing dead biomass.
The inclusion of forest land in Hawaii and several U.S. Territories32 also contributed to increases in the estimates of
carbon stock changes in the current Inventory. Collectively, these areas contributed -1.3 MMT C to the forest land
remaining forest land carbon sink in the year 2022 in the current Inventory (Table 6-22).
Finally, new data included in the HWP time series resulted in a minor decrease (< 1 percent) in carbon stocks in the
HWP pools (Table 6-13) and an associated decrease of 7.9 percent (2.2 MMT C) in estimates of carbon stock
changes (Table 6-18). These decreases are the result of decreases in estimates of carbon stock changes for
products in use (19.4 percent) in the current Inventory relative to the same year in the previous Inventory (Table
6-18). With the easing of the global pandemic and the return of consumers to the marketplace, there was a
rebound in the purchase and accumulation of solid wood products. Alternatively, paper products in use have been
declining in recent years, which could be the result of greater digitization across society. These trends are expected
to continue in 2023.
Table 6-18: Recalculations of Net Carbon Flux from Forest Ecosystem Pools in Forest Land
Remaining Forest Land and Harvested Wood Pools (MMT C)
Carbon Pool (MMT C)
2021 Estimate,
Previous Inventory
2021 Estimate,
Current Inventory
2022 Estimate,
Current Inventory
Forest
Aboveground Biomass
Belowground Biomass
Dead Wood
Litter
Soil (Mineral)
Soil (Organic)
Drained organic soil
(161.6)
(111.6)
(204.4)
(139.9)
(27.5)
(36.9)
(0.0)
(0.2)
(0.0)
(189.3)
(134.1)
(26.4)
(35.8)
(22.1)
(27.6)
0.5
(1.1)
0.0
0.2
0.2
(0.3)
(0.0)
0.2
7.2
Harvested Wood
Products in Use
(28.0)
(10.3)
(25.8)
(8.3)
(25.3)
(7.9)
32 American Samoa, Guam, Norther Mariana Islands, U.S. Virgin Islands, and Puerto Rico
6-42 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
SWDS (17.7) (17.5) (17.4)
Total Net Flux (189.6) (230.2) (214.6)
Note: Totals may not sum due to independent rounding. Parentheses indicate net sequestration.
Table 6-19: Recalculations of Net Carbon Flux from Forest Ecosystem Pools in Forest Land
Remaining Forest Land (MMT C) in the Conterminous United States
Carbon Pool (MMT C)
2021 Estimate,
Previous Inventory
2021 Estimate,
Current Inventory
2022 Estimate,
Current Inventory
Forest
(158.5)
(200.3)
(195.9)
Aboveground Biomass
(108.1)
(135.3)
(132.1)
Belowground Biomass
(21.3)
(26.4)
(25.8)
Dead Wood
(27.7)
(37.2)
(36.8)
Litter
(0.5)
(1.0)
(0.8)
Soil (Mineral)
(1.1)
(0.2)
(0.3)
Soil (Organic)
0.0
(0.1)
(0.1)
Note: Totals may not sum due to independent rounding. Parentheses indicate net sequestration.
Table 6-20: Recalculations of Net Carbon Flux from Forest Ecosystem Pools in Forest Land
Remaining Forest Land (MMT C) in Coastal Alaska
Carbon Pool (MMT C)
2021 Estimate,
Previous Inventory
2021 Estimate,
Current Inventory
2022 Estimate,
Current Inventory
Forest
(2.0)
(1.9)
(1.9)
Aboveground Biomass
(1.3)
(1.4)
(1.4)
Belowground Biomass
(0.3)
(0.3)
(0.3)
Dead Wood
(0.3)
(0.1)
(0.1)
Litter
0.0
(0.1)
(0.1)
Soil (Mineral)
0.0
0.0
0.0
Soil (Organic)
0.0
0.0
0.0
Note: Totals may not sum due to independent rounding. Parentheses indicate net sequestration.
Table 6-21: Recalculations of Net Carbon Flux from Forest Ecosystem Pools in Forest Land
Remaining Forest Land (MMT C) in Interior Alaska
Carbon Pool (MMT C)
2021 Estimate,
Previous Inventory
2021 Estimate,
Current Inventory
2022 Estimate,
Current Inventory
Forest
(1.2)
(1.1)
9.6
Aboveground Biomass
(2.1)
(2.1)
0.4
Belowground Biomass
(0.5)
(0.5)
0.0
Dead Wood
0.4
0.4
1.1
Litter
1.0
1.0
8.0
Soil (Mineral)
0.0
0.0
0.0
Soil (Organic)
0.0
0.0
0.0
Note: Totals may not sum due to independent rounding. Parentheses indicate net sequestration.
Table 6-22: Recalculations of Net Carbon Flux from Forest Ecosystem Pools in Forest Land
Remaining Forest Land (MMT C) in Hawaii and United States Territories
2021 Estimate, 2021 Estimate, 2022 Estimate,
Carbon Pool (MMT C) Previous Inventory Current Inventory Current Inventory
Forest NE (1.4) (1.3)
Aboveground Biomass NE (1.0) (1.0)
Belowground Biomass NE (0.3) (0.3)
Dead Wood NE (0.0) 0.0
Land Use, Land-Use Change, and Forestry 6-43
-------
Litter
Soil (Mineral)
Soil (Organic)
NE
NE
NE
0.0
(0.0)
0.0
(0.0)
0.0
0.0
NE (Not Estimated)
Note: Totals may not sum due to independent rounding. Parentheses indicate net sequestration.
Planned Improvements
Reliable estimates of forest carbon stocks and changes across the diverse ecosystems of the United States require
a high level of investment in both annual monitoring and associated analytical techniques. Development of
improved monitoring/reporting techniques is a continuous process that occurs simultaneously with annual
Inventory submissions. Planned improvements can be broadly assigned to the following categories: development
of a robust estimation and reporting system, individual carbon pool estimation, coordination with other land-use
categories, and periodic and annual inventory data incorporation.
While this Inventory submission includes carbon change by forest land remaining forest land and land converted to
forest land and carbon stock changes for all IPCC pools in these two categories, there are many improvements that
are still necessary. The estimation approach used for the conterminous United States in the current Inventory for
the forest land category operates at the state scale, whereas previously the western United States and southeast
and southcentral coastal Alaska operated at a regional scale. While this is an improvement over previous
Inventories and led to improved estimation and separation of land-use categories in the current Inventory,
including coastal Alaska, research is underway to leverage all FIA data (periodic and annual inventories) and
auxiliary information (i.e., remotely sensed information) to operate at finer spatial and temporal scales. As in past
submissions, emissions and removals associated with natural (e.g., wildfire, insects, and disease) and human (e.g.,
harvesting) disturbances are implicitly included in the report given the design of the annual NFI, but not explicitly
estimated. In addition to integrating auxiliary information into the estimation framework and leveraging all NFI
plot measurements, alternative estimators are also being evaluated which will eliminate latency in population
estimates from the NFI, improve annual estimation and characterization of interannual variability, facilitate
attribution of fluxes to particular activities, and allow for streamlined harmonization of NFI data with auxiliary data
products. This will also facilitate separation of prescribed and wildfire emissions in future reports. The
transparency and repeatability of estimation and reporting systems will be improved through the dissemination of
open-source code (e.g., R programming language) in concert with the public availability of the periodic and annual
NFI (USDA Forest Service 2023b). Also, several FIA database processes are being institutionalized to increase
efficiency and QA/QC in reporting and further improve transparency, completeness, consistency, accuracy, and
availability of data used in reporting. Finally, a combination of approaches was used to estimate uncertainty
associated with carbon stock changes in the forest land remaining forest land category in this report. There is
research underway investigating more robust approaches to estimate total uncertainty (Clough et al. 2016), which
will be considered in future Inventory reports.
The modeling framework used to estimate downed dead wood within the dead wood carbon pool (Smith et al.
2022) will be updated similar to the litter (Domke et al. 2016) and soil carbon pools (Domke et al. 2017). With the
implementation of the new models for volume, biomass, and carbon estimation for live and standing dead trees,
the methods for litter and soil carbon estimation used in this Inventory and recent Inventories have been adopted
in the FIA program so there is now alignment and consistency between litter and soil carbon estimates in this
Inventory and the FIA database. Finally, components of other pools, such as carbon in belowground biomass
(Russell et al. 2015) and understory vegetation (Russell et al. 2014; Johnson et al. 2017), are being explored but
may require additional investment in field inventories before improvements can be realized in the Inventory
report.
The foundation of forest carbon estimation and reporting is the annual NFI. The ongoing annual surveys by the FIA
program are expected to improve the accuracy and precision of forest carbon estimates as new state surveys
become available (USDA Forest Service 2023b). With the exception of Wyoming (which will have sufficient
remeasurements in the years ahead), all other states in the conterminous United States and coastal Alaska now
6-44 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
have sufficient annual NFI data to consistently estimate carbon stocks and stock changes for the future using the
state-level compilation system. The FIA program continues to install permanent plots in interior Alaska as part of
the operational NFI, and as more plots are added to the NFI, they will be used to improve estimates for all
managed forest land in Alaska. Estimates of carbon stocks and stock changes for Hawaii and the U.S. Territories
were included in this Inventory using Tier 1 and Tier 2 methods. The methods used to include all managed forest
land in the conterminous United States will be used in future Inventories for Hawaii and U.S. Territories as
additional forest carbon data become available (only a small number of plots from Hawaii are currently available
from the annualized sampling design). To that end, research is underway to incorporate all NFI information (both
annual and periodic data) and the dense time series of remotely sensed data in multiple inferential frameworks for
estimating greenhouse gas emissions and removals as well as change (i.e., disturbance or land-use changes)
detection and attribution across the entire reporting period and all managed forest land in the United States.
Leveraging this auxiliary information will aid the efforts to improve estimates for interior Alaska, Hawaii, and the
U.S. Territories, as well as the entire inventory system. In addition to fully inventorying all managed forest land in
the United States, the more intensive sampling (i.e., more samples) of fine woody debris, litter, and SOC on a
subset of FIA plots continues and will substantially improve spatial and temporal resolution of carbon pools
(Westfall et al. 2013) as this information becomes available. Increased sample intensity of some carbon pools and
using annualized sampling data as it becomes available for those states currently not reporting are planned for
future submissions. The NFI sampling frame extends beyond the forest land-use category (e.g., woodlands, which
fall into the grasslands land-use category, and urban areas, which fall into the settlements land-use category) with
inventory-relevant information for trees outside of forest land. These data will be utilized as they become available
in the NFI.
Non-C02 Emissions from Forest Fires
Emissions of non-CC>2 gases from forest fires were estimated using U.S.-specific data and models for annual area of
forest burned, fuel, consumption, and emission consistent with IPCC (2019). In 2022, emissions from this source
were estimated to be 9.1 MMT CO2 Eq. of CH4 and 5.7 MMT CO2 Eq. of N2O (Table 6-23; kt units provided in Table
6-24). The estimates of non-CC>2 emissions from forest fires include the conterminous 48 states, Hawaii, Puerto
Rico, Guam and managed forest land in Alaska (Ogle et al. 2018) because the fire data in use with the current
methods identifies fires on these areas within the interval 1990 through 2022.
Table 6-23: Non-CCh Emissions from Forest Fires (MMT CO2 Eq.)a
Gas
1990
2005
2018
2019
2020
2021
2022
ch4
3.4
9.2 I
6.0
3.4
9.8
12.7
9.1
n2o
2.4
6.3 J
3.7
2.3
5.5
7.2
5.7
Total
5.8
15.4
9.7
5.7
15.3
19.9
14.8
a These estimates include non-C02 emissions from forest fires on forest land remaining
forest land and land converted to forest land.
Note: Totals may not sum due to independent rounding.
Table 6-24: Non-CCh Emissions from Forest Fires (kt)a
Gas
1990
2005
2018
2019
2020
2021
2022
ch4
122
328
213
120
349
452
327
n2o
9
24
14
9
21
27
21
CO
3179
8447
4648
3054
7266
9598
7593
NOx
49:
124 =;
93
51
123
160
121
a These estimates include non-C02 emissions from forest fires on forest land remaining
forest land and land converted to forest land.
Land Use, Land-Use Change, and Forestry 6-45
-------
Methodology and Time-Series Consistency
Non-CC>2 emissions from forest fires—primarily Cm and N2O emissions—were calculated consistent with IPCC
(2019) methodology, which represent updates of the IPCC (2006) guidance on reporting fire emissions. For the
conterminous states and Alaska, estimates were developed with U.S.-specific data and models on area burned,
fuel, consumption, and emissions as provided through the Wildland Fire Emissions Inventory System calculator
(WFEIS, French et al. 2011, 2014). However, these fire emissions models did not extend to include Hawaii, Puerto
Rico, or Guam, so forest fire estimates for these areas relied on Tier 1 emissions factors (IPCC 2019). Spatial
definitions of wildland burned areas were the starting point for all estimates, from WFEIS or Tier 1. The three
burned area datasets in use are the Monitoring Trends in Burn Severity (MTBS, Eidenshink et al. 2007), MODIS
burned area mapping (MODIS MCD64A1 V6.1, Giglio et al. 2018), and Wildland Fire Interagency Geospatial Service
(WFIGS) fire perimeters (WFIGS 2023). The MTBS data available for this report (MTBS 2023) included fires from
1990 through 2021 for all states and Puerto Rico (the exception was Alaska 2021 where emissions calculations
were not available). The MODIS-based records include 2001 through 2022 for the 48 conterminous states plus
Alaska. The WFIGS-based records for 2020 through 2022 included all states plus Puerto Rico and Guam. Note that
N2O emissions are not included in WFEIS calculations; the emissions provided here are based on the average N2O
to CO2 ratio of 0.000166 (Larkin et al. 2014; IPCC 2019). See the emissions from forest fires section in Annex 3.13
for further details on all fire-related emissions calculations for forests. Consistent use of available data sources,
data processing, and calculation methods were applied to the entire time series to ensure time-series consistency
from 1990 through 2022.
Uncertainty
Uncertainty estimates for non-CC>2 emissions from forest fires are based on a Monte Carlo (IPCC Approach 2)
approach to propagate variability among the alternate WFEIS annual estimates per state. Uncertainty in parts of
the WFEIS system are not currently quantified. Among potential sources for future analysis are burned areas from
MTBS, WFIGS, or MODIS, the fuels models or the Consume model (Prichard et al. 2014). See Annex 3.13 for the
quantities and assumptions employed to define and propagate uncertainty. The results of the Approach 2
quantitative uncertainty analysis are summarized in Table 6-25.
Table 6-25: Quantitative Uncertainty Estimates of Non-CCh Emissions from Forest Fires (MMT
CO2 Eq. and Percent)3
Source
Gas
2022 Emission Estimate
(MMTCO. Eq.)
Uncertainty Range Relative to Emission Estimate1,
(MMTCO' Eq.) (%)
Lower
Upper
Lower
Upper
Bound
Bound
Bound
Bound
Non-C02 Emissions from
Forest Fires
ch4
9.1
6.2
12.1
-32%
+32%
Non-C02 Emissions from
Forest Fires
n2o
5.7
3.6
7.8
-36%
+37%
a These estimates include non-C02 emissions from forest fires on forest land remaining forest land and land converted to
forest land.
b Range of flux estimates predicted by Monte Carlo stochastic simulation for a 95 percent confidence interval.
QA/QC and Verification
Tier 1 and Tier 2 QA/QC activities were conducted consistent with the U.S. QA/QC plan. Source-specific quality
control measures for estimating non-C02 emissions from forest fires included checking input data, documentation,
and calculations to ensure data were properly handled through the inventory process and results were consistent
with values expected from those calculations. The QA/QC procedures did not reveal any inaccuracies or incorrect
input values.
6-46 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
Recalculations Discussion
The methods used in the current (1990 through 2022) Inventory to compile estimates of non-CC>2 emissions from
forest fires represent a slight change relative to the previous (1990 through 2021) Inventory. The basic components
of calculating forest fire emissions (IPCC 2019) remain unchanged, but the WFEIS-based estimates now include
MTBS, WFIGS, and MODIS based burns (depending on year). The MTBS and WFIGS based estimates are now
calculated per burn event (i.e., separately for each forest fire), which improves precision for scaling or allocating
emissions such as to managed versus unmanaged lands in Alaska.
An additional source of change leading to recalculations are recent and ongoing updates to the MTBS fire records
(i.e., including both most-recent as well as possible updates to past years' fires). The WFEIS calculations now use
version 6.1 of the MODIS burned area model and updated versions of the Fuel Characteristic Classification System
(FCCS) fuel layer and the Consume model (see WFEIS 2023) for additional details on updates. The addition of forest
fire emissions for Hawaii, Puerto Rico, and Guam had little effect on the magnitude of overall emissions.
Estimates of non-CC>2 emissions from forest fires (e.g., Table 6-24) are lower for most years over the time series
1990-2021 in comparison with the previous Inventory (EPA 2023), with an average decrease of 12 percent across
all years. For, 2021 the estimate decreased from 24.4 to 19.9 MMT CO2 Eq. (an 18.7 percent decrease). Changes
over the time series are expected because the entire interval is recalculated each year in response to modifications
in the fire datasets that can affect all years. For example, MTBS updates burn perimeters for all years as data
resolution changes. This year, the WFEIS calculator updated a fuel dataset as well as the Consume model (noted
above). The addition of estimates for Hawaii, Puerto Rico, and Guam had negligible effect on the estimates.
Planned Improvements
Continuing improvements are planned for developing better fire and site-specific estimates for forest fires, for
example, improving on the Tier-1 factors currently employed for Hawaii, Puerto Rico, and Guam. Additional focus
will be on addressing three aspects of reporting: best use of WFEIS, better resolution of uncertainty as discussed
above, and identification of burned areas that are not currently captured by the burn records in use.
N20 Emissions from N Additions to Forest Soils
Of the synthetic nitrogen (N) fertilizers applied to soils in the United States, no more than one percent is applied to
forest soils. Application rates are similar to those occurring on cropland soils, but in any given year, only a small
proportion of total forested land receives N fertilizer. This is because forests are typically fertilized only twice
during their approximately 40-year growth cycle (once at planting and once midway through their life cycle). While
the rate of N fertilizer application for the area of forests that receives N fertilizer in any given year is relatively high,
the annual application rate is quite low over the entire area of forest land.
N additions to soils result in direct and indirect N2O emissions. Direct emissions occur on-site due to the N
additions. Indirect emissions result from fertilizer N that is transformed and transported to another location
through volatilization in the form of ammonia (NH3) and nitrogen oxide (NOx), in addition to leaching and runoff of
nitrates (NO3), and later converted into N2O at off-site locations. The indirect emissions are assigned to forest land
because the management activity leading to the emissions occurred in forest land.
Direct soil N2O emissions from forest land remaining forest land and land converted to forest land33 in 2022 were
0.3 MMT CO2 Eq. (1.2 kt), and the indirect emissions were 0.1 MMT CO2 Eq. (0.4 kt). Total emissions for 2022 were
0.4 MMT CO2 Eq. (1.5 kt) and have increased by 455 percent from 1990 to 2022. Total forest soil N2O emissions are
summarized in Table 6-26.
33 The N20 emissions from land converted to forest land are included with forest land remaining forest land because it is not
currently possible to separate the activity data by land use conversion category.
Land Use, Land-Use Change, and Forestry 6-47
-------
Table 6-26: N2O Fluxes from Soils in Forest Land Remaining Forest Land and Land Converted
to Forest Land (MMT CO2 Eq. and kt N2O)
1990
2005
2018
2019
2020
2021
2022
Direct N20 Fluxes from Soils
MMT C02 Eq.
0.1
0.31
0.3
0.3
0.3
0.3
0.3
kt N20
0.2
1.2
1.2
1.2
1.2
1.2
1.2
Indirect N20 Fluxes from Soils
MMTC02 Eq.
+
0.1
0.1
0.1
0.1
0.1
0.1
kt N20
0.1 j
0.4
0.4
0.4
0.4
0.4
0.4
Total (MMT CO . Eq.)
0.1
0.4
0.4
0.4
0.4
0.4
0.4
Total (kt N.'O)
0.3
1.5
1.5
1.5
1.5
1.5
1.5
+ Does not exceed 0.05 MMT C02 Eq. or 0.05 kt.
Note: Totals may not sum due to independent rounding. The N20 emissions from land converted to
forest land are included with forest land remaining forest land because it is not currently possible
to separate the activity data by land use conversion category.
Methodology and Time-Series Consistency
The IPCC Tier 1 approach is used to estimate N2O from soils within forest land remaining forest land and land
converted to forest land. According to U.S. Forest Service statistics for 1996 (USDA Forest Service 2001),
approximately 75 percent of trees planted are for timber, and about 60 percent of national total harvested forest
area is in the southeastern United States. Although southeastern pine plantations represent the majority of
fertilized forests in the United States, this Inventory also incorporated N fertilizer application to commercial
Douglas-fir stands in western Oregon and Washington. For the Southeast, estimates of direct N2O emissions from
fertilizer applications to forests are based on the area of pine plantations receiving fertilizer in the southeastern
United States and estimated application rates (Albaugh et al. 2007; Fox et al. 2007). Fertilizer application is rare for
hardwoods and therefore not included in the inventory (Binkley et al. 1995). For each year, the area of pine
receiving N fertilizer is multiplied by the weighted average of the reported range of N fertilization rates (121 lbs. N
per acre). Area data for pine plantations receiving fertilizer in the Southeast are not available for 2005 through
2022, so data from 2004 are used for these years. For commercial forests in Oregon and Washington, only fertilizer
applied to Douglas-fir is addressed in the inventory because the vast majority (approximately 95 percent) of the
total fertilizer applied to forests in this region is applied to Douglas-fir stands (Briggs 2007). Estimates of total
Douglas-fir area and the portion of fertilized area are multiplied to obtain annual area estimates of fertilized
Douglas-fir stands. Similar to the Southeast, data are not available for 2005 through 2022, so data from 2004 are
used for these years. The annual area estimates are multiplied by the typical rate used in this region (200 lbs. N per
acre) to estimate total N applied (Briggs 2007), and the total N applied to forests is multiplied by the IPCC (2006)
default emission factor of one percent to estimate direct N2O emissions.
For indirect emissions, the volatilization and leaching/runoff N fractions for forest land are calculated using the
IPCC default factors of 10 percent and 30 percent, respectively. The amount of N volatilized is multiplied by the
IPCC default factor of one percent for the portion of volatilized N that is converted to N2O off-site. The amount of
N leached/runoff is multiplied by the IPCC default factor of 0.075 percent for the portion of leached/runoff N that
is converted to N2O off-site. The resulting estimates are summed to obtain total indirect emissions.
The same method is applied in all years of this Inventory to ensure time-series consistency from 1990 through
2022.
Uncertainty
The amount of N2O emitted from forests depends not only on N inputs and fertilized area, but also on a large
number of variables, including organic carbon availability, oxygen gas partial pressure, soil moisture content, pH,
temperature, and tree planting/harvesting cycles. The effect of the combined interaction of these variables on N2O
flux is complex and highly uncertain. IPCC (2006) does not incorporate any of these variables into the default
6-48 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
methodology, except variation in estimated fertilizer application rates and estimated areas of forested land
receiving nitrogen fertilizer. All forest soils are treated equivalently under this methodology. Furthermore, only
applications of synthetic nitrogen fertilizers to forest are captured in this Inventory, so applications of organic
nitrogen fertilizers are not estimated. However, the total quantity of organic nitrogen inputs to soils in the United
States is included in the Inventory within the agricultural soil management source category (Section 5.4) and
settlements remaining settlements (Section 6.10).
Uncertainties exist in the fertilization rates, annual area of forest lands receiving fertilizer, and the emission
factors. Fertilization rates are assigned a default level34 of uncertainty at ±50 percent, and area receiving fertilizer
is assigned a ±20 percent according to expert knowledge (Binkley 2004). IPCC (2006) provided estimates for the
uncertainty associated with direct and indirect N2O emission factor for synthetic N fertilizer application to soils.
Uncertainty is quantified using simple error propagation methods (IPCC 2006). The results of the quantitative
uncertainty analysis are summarized in Table 6-27. Direct N2O fluxes from soils in 2022 are estimated to be
between 0.1 and 1.0 MMT CO2 Eq. at a 95 percent confidence level. This indicates a range of 59 percent below and
211 percent above the emission estimate of 0.3 MMT CO2 Eq. for 2022. Indirect N2O emissions in 2022 are 0.1
MMT CO2 Eq. and have a range are between 0.01 and 0.3 MMT CO2 Eq., which is 86 percent below to 238 percent
above the emission estimate for 2022.
Table 6-27: Quantitative Uncertainty Estimates of N2O Fluxes from Soils in Forest Land
Remaining Forest Land and Land Converted to Forest Land (MMT CO2 Eq. and Percent)
Source
Gas
2022 Emission Estimate
(MMTCO. Eq.)
Uncertainty Range Relative to Emission Estimate
(MMTCO' Eq.) (%)
Forest Land Remaining Forest
Lower
Upper
Lower Upper
Land
Bound
Bound
Bound Bound
Direct N20 Fluxes from Soils
N20
0.3
0.1
1.0
-59% +211%
Indirect N20 Fluxes from Soils
n2o
0.1
+
0.3
-86% +238%
+ Does not exceed 0.05 MMT C02 Eq.
Note: Totals may not sum due to independent rounding.
QA/QC and Verification
The spreadsheet containing fertilizer applied to forests and calculations for N2O and uncertainty ranges are
checked and verified based on the sources of these data consistent with the U.S. Inventory QA/QC plan, which is in
accordance with Volume 1, Chapter 6 of the 2006 IPCC Guidelines (see Annex 8 for more details).
Recalculations Discussion
No recalculations were performed for the current Inventory.
C02, CH4, and N20 Emissions from Drained Organic Soils35
Drained organic soils on forest land are identified separately from other forest soils largely because mineralization
of the exposed or partially dried organic material results in continuous CO2 and N2O emissions (IPCC 2006). In
addition, the 2013 Supplement to the 2006 IPCC Guidelines for National Greenhouse Gas Inventories: Wetlands
34 Uncertainty is unknown for the fertilization rates so a conservative value of ±50 percent is used in the analysis.
35 Estimates of C02 emissions from drained organic soils are described in this section but reported in Table 6-8 and Table 6-9 for
both Forest land remaining forest land and land converted to forest land in order to allow for reporting of all carbon stock changes
on forest lands in a complete and comprehensive manner.
Land Use, Land-Use Change, and Forestry 6-49
-------
(IPCC 2014) calls for estimating Cm emissions from these drained organic soils and the ditch networks used to
drain them.
Organic soils are identified on the basis of thickness of organic horizon and percent organic matter content. All
organic soils are assumed to have originally been wet, and drained organic soils are further characterized by
drainage or the process of artificially lowering the soil water table, which exposes the organic material to drying
and the associated emissions described in this section. The land base considered here is drained inland organic
soils that are coincident with forest area as identified by the NFI of the USDA Forest Service (USDA Forest Service
2022b).
The estimated area of drained organic soils on forest land is 70,849 ha and did not change over the time series
based on the data used to compile the estimates in the current Inventory. These estimates are based on
permanent plot locations of the NFI (USDA Forest Service 2022b) coincident with mapped organic soil locations
(STATSG02 2016), which identifies forest land on organic soils. Forest sites that are drained are not explicitly
identified in the data, but for this estimate, planted forest stands on sites identified as mesic or xeric (which are
identified in USDA Forest Service 2022c, d) are labeled "drained organic soil" sites.
Land use, region, and climate are broad determinants of emissions as are more site-specific factors such as
nutrient status, drainage level, exposure, or disturbance. Current data are limited in spatial precision and thus lack
site specific details. At the same time, corresponding emissions factor data specific to U.S. forests are similarly
lacking. Tier 1 estimates are provided here following IPCC (2014). Total annual non-CC>2 emissions on forest land
with drained organic soils in 2022 are estimated as 0.1 MMT CO2 Eq. per year (Table 6-28; kt units provided in
Table 6-29).
The Tier 1 methodology provides methods to estimate emissions of CO2 from three pathways: direct emissions
primarily from mineralization; indirect, or off-site, emissions associated with dissolved organic carbon releasing
CO2 from drainage waters; and emissions from (peat) fires on organic soils. Data about forest fires specifically
located on drained organic soils are not currently available; as a result, no corresponding estimate is provided
here. Non-CC>2 emissions provided here include CH4 and N2O. Methane emissions generally associated with anoxic
conditions do occur from the drained land surface, but the majority of these emissions originate from ditches
constructed to facilitate drainage at these sites. Emission of N2O can be significant from these drained organic soils
in contrast to the very low emissions from wet organic soils.
Table 6-28: Non-CCh Emissions from Drained Organic Forest Soilsa,b (MMT CO2 Eq.)
Source
1990
2005
2018
2019
2020
2021
2022
ch4
+
+
+
+
+
+
+
n2o
0.1
0-1 1
0.1
0.1
0. 1
0.1
0.1
Total
0.1
0.1
0.1
0.1
0.1
0.1
0.1
+ Does not exceed 0.05 MMT C02 Eq.
a This table includes estimates from forest land remaining forest land and land converted
to forest land.
b Estimates of C02 emissions from drained organic soils are described in this section but
reported in Table 6-8 and Table 6-9 for both forest land remaining forest land and land
converted to forest land in order to allow for reporting of all carbon stock changes on
forest lands in a complete and comprehensive manner.
Note: Totals may not sum due to independent rounding.
Table 6-29: Non-CCh Emissions from Drained Organic Forest Soilsa,b (kt)
Source
1990
2005
2018
2019
2020
2021
2022
ch4
1
1
1
1
1
1
1
n2o
+ I
+ I.
+
+
+
+
+
+ Does not exceed 0.5 kt.
6-50 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
a This table includes estimates from forest land remaining forest land and land converted to
forest land.
b Estimates of C02 emissions from drained organic soils are described in this section but
reported in Table 6-8 and Table 6-9 for both forest land remaining forest land and land
converted to forest land in order to allow for reporting of all carbon stock changes on
forest lands in a complete and comprehensive manner.
Methodology and Time-Series Consistency
The Tier 1 methods for estimating CO2, Cm and N2O emissions from drained inland organic soils on forest lands
follow IPCC (2006), with extensive updates and additional material presented in the 2013 Supplement to the 2006
IPCC Guidelines for National Greenhouse Gas Inventories: Wetlands (IPCC 2014). With the exception of quantifying
area of forest on drained organic soils, which is user-supplied, all quantities necessary for Tier 1 estimates are
provided in Chapter 2, Drained Inland Organic Soils of IPCC (2014).
Estimated area of drained organic soils on forest land is 70,849 ha based on analysis of the permanent NFI of the
USDA Forest Service and did not change over the time series. The most recent plot data per state within the
inventories were used in a spatial overlay with the STATSG02 (2016) soils data, and forest plots coincident with the
soil order histosol were selected as having organic soils. Information specific to identifying "drained organic" are
not in the inventory data so an indirect approach was employed here. Specifically, artificially regenerated forest
stands (inventory field STDORGCD=l) on mesic or xeric sites (inventory field 11
-------
Non-CC>2 emissions, according to the Tier 1 method, include methane (Cm), nitrous oxide (N2O), and carbon
monoxide (CO). Emissions associated with peat fires include factors for CH4 and CO in addition to CO2, but fire
estimates are assumed to be zero for the current Inventory, as discussed above. Methane emissions generally
associated with anoxic conditions do occur from the drained land surface, but the majority of these emissions
originate from ditches constructed to facilitate drainage at these sites. From this, two separate emission factors
are used, one for emissions from the area of drained soils and a second for emissions from drainage ditch
waterways. Calculations are conducted according to Equation 2.6 and Tables 2.3 and 2.4, which includes the
default fraction of the total area of drained organic soil which is occupied by ditches. Emissions of N2O can be
significant from these drained soils in contrast to the very low emissions from wet organic soils. Calculations are
conducted according to Equation 2.7 and Table 2.5, which provide the estimate as kg N per year.
Methodological calculations were applied to the entire set of estimates for 1990 through 2022. Year-specific data
are not available. Estimates are based on a single year and applied as the annual estimates over the interval.
Uncertainty
Uncertainties are based on the sampling error associated with forest area of drained organic soils and the
uncertainties provided in the Chapter 2 (IPCC 2014) emissions factors (Table 6-31). The estimates and resulting
quantities representing uncertainty are based on the IPCC Approach 1-error propagation. However, probabilistic
sampling of the distributions defined for each emission factor produced a histogram result that contained a mean
and 95 percent confidence interval. The primary reason for this approach was to develop a numerical
representation of uncertainty with the potential for combining with other forest components. The methods and
parameters applied here are identical to previous inventories, but input values were resampled for this Inventory,
which results in minor changes in the number of significant digits in the resulting estimates, relative to past values.
The total non-C02 emissions in 2022 from drained organic soils on forest land remaining forest land and land
converted to forest land were estimated to be between zero and 0.150 MMT CO2 Eq. around a central estimate of
0.068 MMT CO2 Eq. at a 95 percent confidence level.
Table 6-31: Quantitative Uncertainty Estimates for Non-CCh Emissions on Drained Organic
Forest Soils (MMT CO2 Eq. and Percent)3
2022 Emission
Source Estimate Uncertainty Range Relative to Emission Estimate
(MMT CO . Eq.) (MMT CQ . Eq.) (%)
Lower
Upper
Lower
Upper
Bound
Bound
Bound
Bound
ch4
+
+
+
-69%
+82%
n2o
0.1
+
0.1
-118%
+132%
Total
0.1
+
0.2
-107%
+121%
+ Does not exceed 0.05 MMT C02 Eq.
a Range of flux estimates predicted through a combination of sample-based and IPCC defaults for a 95
percent confidence interval, IPCC Approach 1.
Note: Totals may not sum due to independent rounding.
QA/QC and Verification
IPCC (2014) guidance cautions of a possibility of double counting some of these emissions. Specifically, the off-site
emissions of dissolved organic carbon from drainage waters may be double counted if soil carbon stock and
change is based on sampling and this carbon is captured in that sampling. Double counting in this case is unlikely
since plots identified as drained were treated separately in this chapter. Additionally, some of the non-C02
emissions may be included in either the wetlands or sections on N2O emissions from managed soils. These paths to
double counting emissions are unlikely here because these issues are taken into consideration when developing
the estimates and this chapter is the only section directly including such emissions on forest land.
6-52 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
Recalculations Discussion
No recalculations were performed for the current Inventory.
Planned Improvements
Additional data will be compiled to update estimates of forest areas on drained organic soils as new reports and
geospatial products become available. For example, current and recent past estimates are based on drained
organic soils identified in a limited number of the conterminous states; if forests on drained organic soils are
identified in additional areas including Alaska, Hawaii, Puerto Rico, or Guam, they will be included in future
Inventories.
6.3 Land Converted to Forest Land (CRT
Category 4A2)
The carbon stock change estimates for land converted to forest land that are provided in this Inventory include all
forest land in an inventory year that had been in another land use(s) during the previous 20 years.36 For example,
cropland or grassland converted to forest land during the past 20 years would be reported in this category.
Converted lands are in this category for 20 years as recommended in the 2006IPCC Guidelines (IPCC 2006), after
which they are classified as forest land remaining forest land. Estimates of carbon stock changes from all pools
(i.e., aboveground and belowground biomass, dead wood, litter and soils), as recommended by IPCC (2006), are
included in the land converted to forest land category of this Inventory.
Area of Land Converted to Forest in the United States37
Land conversion to and from forests has occurred regularly throughout United States history. The 1970s and 1980s
saw a resurgence of federally sponsored forest management programs (e.g., the Forestry Incentive Program) and
soil conservation programs (e.g., the Conservation Reserve Program), which have focused on tree planting,
improving timber management activities, combating soil erosion, and converting marginal cropland to forests.
Recent analyses suggest that net accumulation of forest area continues in areas of the United States, in particular
the northeastern United States (Woodall et al. 2015b). Specifically, the annual conversion of land from other land-
use categories (i.e., cropland, grassland, wetlands, settlements, and other lands) to forest land resulted in a fairly
continuous net annual accretion of forest land area from over the time series at an average rate of 1.0 million ha
year-1.
Over the 20-year conversion period used in the land converted to forest land category, the conversion of grassland
to forest land resulted in the largest source of carbon transfer and uptake, accounting for approximately 38
36 The annual NFI data used to compile estimates of carbon transfer and uptake in this section are based on 5- to 10-yr
remeasurements so the exact conversion period was limited to the remeasured data over the time series.
37 The estimates reported in this section only include the 48 conterminous states in the United States. Land use conversions to
forest land in Alaska are currently included in the Forest Land Remaining Forest Land section because currently there is
insufficient data to separate the changes and estimates for Hawaii were not included because there is insufficient NFI data to
support inclusion at this time. Also, it is not possible to separate forest land remaining forest land from land converted to forest
land in Wyoming because of the split annual cycle method used for population estimation, this prevents harmonization of
forest land in Wyoming with the NRI/NLCD method used in Section 6.1. See Annex 3.13, Table A-203 for annual differences
between the forest area reported in Section 6.1 and Section 6.3.
Land Use, Land-Use Change, and Forestry 6-53
-------
percent of the uptake annually. Estimated carbon uptake has remained relatively stable over the time series across
all conversion categories (see Table 6-24). The net flux of carbon from all forest pool stock changes in 2022 was -
100.3 MMT C02 Eq. (-27.4 MMT C) (see Table 6-24 and Table 6-25).
Mineral soil carbon stocks increased slightly over the time series for land converted to forest land. The small gains
are associated with cropland converted to forest land, settlements converted to forest land, and other land
converted to forest land. Much of this conversion is from soils that are more intensively used under annual crop
production or settlement management, or are conversions from other land, which has little to no soil carbon. In
contrast, grassland converted to forest land leads to a loss of soil carbon across the time series, which negates
some of the gain in soil carbon with the other land-use conversions. Managed pasture to forest land is the most
common conversion. This conversion leads to a loss of soil carbon because pastures are mostly improved in the
United States with fertilization and/or irrigation, which enhances carbon input to soils relative to typical forest
management activities.
Table 6-32: Net CO2 Flux from Forest Carbon Pools in Land Converted to Forest Land by Land
Use Change Category (MMT CO2 Eq.)
Land Use/Carbon Pool
1990
2005
2018
2019
2020
2021
2022
Cropland Converted to Forest Land
(17.6)
(17.5)
(17.3)
(17.3)
(17.2)
(17.2)
(17.2)
Aboveground Biomass
(10.2)
(10.1)
(10.0)
(10.0)
(10.0)
(10.0)
(10.0)
Belowground Biomass
(1.7)
(1.7)
(1.7)
(1.7)
(1.7)
(1.7)
(1.7)
Dead Wood
<2-2> ::
(2.2)
(2.2)
(2.2)
(2.2)
(2.2)
(2.2)
Litter
(3.3)
(3.2)
(3.2)
(3.2)
(3.2)
(3.2)
(3.2)
Mineral Soil
(0.2)
(0.2):::
(0.2)
(0.1)
(0.1)
(0.1)
(0.1)
Grassland Converted to Forest Land
(36.7)
(36.9)
(37.2)
(37.2)
(37.2)
(37.2)
(37.2)
Aboveground Biomass
(22.5)
(22.7)
(22.8)
(22.8)
(22.8)
(22.8)
(22.8)
Belowground Biomass
(2.7)
(2.7)
(2.7)
(2.7)
(2.7)
(2.7)
(2.7)
Dead Wood
(4.0):
(4.1)
(4.1)
(4.1)
(4.1)
(4.1)
(4.1)
Litter
(7.6)
(7.7)
(7.7)
(7.7)
(7.7)
(7.7)
(7.7)
Mineral Soil
0.2
0.2::
0.2
0.2
0.2
0.2
0.2
Other Land Converted to Forest Land
(5.2)
(5.3)
(5.5)
(5.5)
(5.5)
(5.5)
(5.5)
Aboveground Biomass
(2.2)
<2'2> I
(2.3)
(2.3)
(2.3)
(2.3)
(2.3)
Belowground Biomass
(0.3)
(0.4)
(0.4)
(0.4)
(0.4)
(0.4)
(0.4)
Dead Wood
(0.7) :
(0.8)
(0.8)
(0.8)
(0.8)
(0.8)
(0.8)
Litter
(1.2)
(1.2)
(1.2)
(1.2)
(1.2)
(1.2)
(1.2)
Mineral Soil
(0.7)
(0.8)
(1.0)
(0.9)
(0.9)
(0.9)
(0.9)
Settlements Converted to Forest Land
(31.9)
(31.6)
(31.4)
(31.4)
(31.4)
(31.4)
(31.4)
Aboveground Biomass
(19.8)
(19.6)
(19.5)
(19.4)
(19.4)
(19.4)
(19.4)
Belowground Biomass
(3.4)
(3.3)
(3.3)
(3.3)
(3.3)
(3.3)
(3.3)
Dead Wood
(3.8)
(3.8) II
(3.8)
(3.8)
(3.8)
(3.8)
(3.8)
Litter
(4.9)
(4.8)
(4.8)
(4.8)
(4.8)
(4.8)
(4.8)
Mineral Soil
(0.0)
(0.03)
(0.1)
(0.1)
(0.1)
(0.1)
(0.1)
Wetlands Converted to Forest Land
(8.8)
(8.9)
(8.9)
(8.9)
(8.9)
(8.9)
(8.9)
Aboveground Biomass
(4.9) =
(5.0)
(5.0)
(5.0)
(5.0)
(5.0)
(5.0)
Belowground Biomass
(0.9)
(0.9)
(0.9)
(0.9)
(0.9)
(0.9)
(0.9)
Dead Wood
(1-2)
(1.2)
(1.2)
(1.2)
(1.2)
(1.2)
(1.2)
Litter
(1.8)
(1.8)
(1.8)
(1.8)
(1.8)
(1.8)
(1.8)
Mineral Soil
0.0
0.0
0.0
0.0
0.0
0.0
0.0
Total Aboveground Biomass Flux
(59.7)
(59.6)
(59.6)
(59.6)
(59.6)
(59.6)
(59.6)
Total Belowground Biomass Flux
(9.0)
(9.0)
(9.0)
(9.0)
(9.0)
(9.0)
(9.0)
Total Dead Wood Flux
(12.0)
(12.0)
(12.1)
(12.1)
(12.1)
(12.1)
(12.1)
Total Litter Flux
(18.8)
(18.8)
(18.8)
(18.8)
(18.8)
(18.8)
(18.8)
Total Mineral Soil Flux
(0.8)
(0.8)
(1.0)
(0.9)
(0.9)
(0.9)
(0.9)
6-54 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
Total Flux (100.2) (100.2) (100.4) (100.3) (100.3) (100.3) (100.3)
Notes: Totals may not sum due to independent rounding. Parentheses indicate net uptake. Forest ecosystem carbon stock
changes from land conversion in interior Alaska, Hawaii, and the U.S. Territories are currently included in the forest land
remaining forest land section because there is insufficient data to separate the changes at this time. It is not possible to
separate forest land remaining forest land from land converted to forest land in Wyoming because of the split annual cycle
method used for population estimation, this prevents harmonization of forest land in Wyoming with the NRI/NLCD method
used in Section 6.1. See Annex 3.13, Table A-217 for annual differences between the forest area reported in Section 6.1 and
Section 6.3. The forest ecosystem carbon stock changes from land conversion do not include trees on non-forest land (e.g.,
agroforestry systems and settlement areas—see 6.10 for estimates of carbon stock change from settlement trees). It is not
possible to separate emissions from drained organic soils between forest land remaining forest land and land converted to
forest land so estimates for all organic soils are included in Table 6-8 of the Forest Land Remaining Forest Land section of
the Inventory.
Table 6-33: Net Carbon Flux from Forest Carbon Pools in Land Converted to Forest Land by
Land Use Change Category (MMT C)
Land Use/Carbon Pool
1990
2005
mini
2018
2019
2020
2021
2022
Cropland Converted to Forest Land
(4.8)
(4.8)
1
(4.7)
(4.7)
(4.7)
(4.7)
(4.7)
Aboveground Biomass
(2.8)
(2.8)
(2.7)
(2.7)
(2.7)
(2.7)
(2.7)
Belowground Biomass
(0.5)
(0.5)
(0.5)
(0.5)
(0.5)
(0.5)
(0.5)
Dead Wood
(0.6)
(0.6)
1
(0.6)
(0.6)
(0.6)
(0.6)
(0.6)
Litter
(0.9)
(0.9)
(0.9)
(0.9)
(0.9)
(0.9)
(0.9)
Mineral Soil
(0.1) =
(0.1)
I
(0.0)
(0.0)
(0.0)
(0.0)
(0.0)
Grassland Converted to Forest Land
(10.0)
(10.1)
(10.2)
(10.2)
(10.2)
(10.2)
(10.1)
Aboveground Biomass
(6.1) :
(6.2)
I
1
(6.2)
(6.2)
(6.2)
(6.2)
(6.2)
Belowground Biomass
(0.7)
(0.7)
(0.7)
(0.7)
(0.7)
(0.7)
(0.7)
Dead Wood
(1-D
(1.1)
(1.1)
(1.1)
(1.1)
(1.1)
(1.1)
Litter
(2.1)
(2.1)
(2.1)
(2.1)
(2.1)
(2.1)
(2.1)
Mineral Soil
o-i
0.1
I
nine
0.1
0.1
0.1
0.1
0.1
Other Land Converted to Forest Land
(1.4)
(1.4)
(1.5)
(1.5)
(1.5)
(1.5)
(1.5)
Aboveground Biomass
(0.6)
(0.6)
I
1
(0.6)
(0.6)
(0.6)
(0.6)
(0.6)
Belowground Biomass
(0.1)
(0.1)
(0.1)
(0.1)
(0.1)
(0.1)
(0.1)
Dead Wood
(0.2)
(0.2)
¦
(0.2)
(0.2)
(0.2)
(0.2)
(0.2)
Litter
(0.3)
(0.3)
(0.3)
(0.3)
(0.3)
(0.3)
(0.3)
Mineral Soil
(0.2) i
(0.2)
(0.3)
(0.2)
(0.3)
(0.3)
(0.3)
Settlements Converted to Forest Land
(8.7)
(8.6)
(8.6)
(8.6)
(8.6)
(8.6)
(8.6)
Aboveground Biomass
Ln
¦111!!
(5.4)
¦
(5.3)
(5.3)
(5.3)
(5.3)
(5.3)
Belowground Biomass
(0.9)
(0.9)
(0.9)
(0.9)
(0.9)
(0.9)
(0.9)
Dead Wood
(i.o) i
(1.0)
1
1
(1.0)
(1.0)
(1.0)
(1.0)
(1.0)
Litter
(1-3)
(1.3)
(1.3)
(1.3)
(1.3)
(1.3)
(1.3)
Mineral Soil
(0.0) ¦
(0.0)
1
1
(0.0)
(0.0)
(0.0)
(0.0)
(0.0)
Wetlands Converted to Forest Land
(2.4)
(2.4)
(2.4)
(2.4)
(2.4)
(2.4)
(2.4)
Aboveground Biomass
(1.3)
(1.4)
(1.4)
(1.4)
(1.4)
(1.4)
(1.4)
Belowground Biomass
(0.2)
(0.2)
(0.2)
(0.2)
(0.2)
(0.2)
(0.2)
Dead Wood
(0.3) i
(0.3)
1
(0.3)
(0.3)
(0.3)
(0.3)
(0.3)
Litter
(0.5)
(0.5)
(0.5)
(0.5)
(0.5)
(0.5)
(0.5)
Mineral Soil
o.o
0.0
1
0.0
0.0
0.0
0.0
0.0
Total Aboveground Biomass Flux
(16.3)
(16.3)
(16.3)
(16.3)
(16.3)
(16.3)
(16.3)
Total Belowground Biomass Flux
(2.5)
(2.5)
I
I
(2.5)
(2.4)
(2.4)
(2.4)
(2.4)
Total Dead Wood Flux
(3.3)
(3.3)
(3.3)
(3.3)
(3.3)
(3.3)
(3.3)
Total Litter Flux
(5.1)
(5.1)
1
(5.1)
(5.1)
(5.1)
(5.1)
(5.1)
Total Mineral Soil Flux
(0.2)
(0.2)
(0.3)
(0.2)
(0.2)
(0.3)
(0.2)
Total Flux
(27.3)
(27.3)
(27.4)
(27.4)
(27.4)
(27.4)
(27.4)
Land Use, Land-Use Change, and Forestry 6-55
-------
+ Absolute value does not exceed 0.05 MMT C.
Notes: Totals may not sum due to independent rounding. Parentheses indicate net uptake. Forest ecosystem carbon stock
changes from land conversion in interior Alaska, Hawaii, and the U.S. Territories are currently included in the forest land
remaining forest land section because there is not sufficient data to separate the changes at this time. It is not possible to
separate forest land remaining forest land from land converted to forest land in Wyoming because of the split annual cycle
method used for population estimation, this prevents harmonization of forest land in Wyoming with the NRI/NLCD method
used in Section 6.1. See Annex 3.13, Table A-217 for annual differences between the forest area reported in Section 6.1 and
Section 6.3. The forest ecosystem carbon stock changes from land conversion do not include trees on non-forest land (e.g.,
agroforestry systems and settlement areas—see Section 6.10 for estimates of carbon stock change from settlement trees).
It is not possible to separate emissions from drained organic soils between forest land remaining forest land and land
converted to forest land so estimates for organic soils are included in Table 6-9 and Table 6-10 of the Forest Land
Remaining Forest Land section of the Inventory.
Methodology and Time-Series Consistency
The following section includes a description of the methodology used to estimate stock changes in all forest carbon
pools for land converted to forest land. National Forest Inventory data and IPCC (2006) defaults for reference
carbon stocks were used to compile separate estimates for the five carbon storage pools. Estimates for
aboveground and belowground biomass, dead wood and litter were based on data collected from the extensive
array of permanent, annual NFI plots and associated models (e.g., live tree belowground biomass estimates) in the
United States (USDA Forest Service 2023b, 2023c). Carbon conversion factors were applied at the individual plot
and then appropriately expanded to state population estimates, which are summed to provide the national
estimate. To ensure consistency in the land converted to forest land category where carbon stock transfers occur
between land-use categories, all soil estimates are based on methods from Ogle et al. (2003, 2006) and IPCC
(2006).
The methods used for estimating carbon stocks and stock changes in the land converted to forest land are
consistent with those used for forest land remaining forest land. For land-use conversion, IPCC (2006) default
biomass carbon stock values were applied in the year of conversion on individual plots to estimate the C stocks
removed due to land-use conversion from croplands and grasslands. There is no biomass loss data or IPCC (2006)
defaults to include transfers, losses, or gains of carbon in the year of the conversion for other land use (i.e., other
lands, settlements, wetlands) conversions to forest land so these were incorporated for these conversion
categories. All annual NFI plots included in the public FIA database as of September 2023 were used in this
Inventory. Forest land conditions were observed on NFI plots at time to and at a subsequent time ti=to+s, where s
is the time step (time measured in years) and is indexed by discrete (e.g., 5 year) forest age classes. The inventory
from to was then projected from ti to 2022. This projection approach requires simulating changes in the age-class
distribution resulting from forest aging and disturbance events and then applying carbon density estimates for
each age class to obtain population estimates for the nation.
Carbon in Biomass
Live tree carbon pools include aboveground and belowground (coarse root) biomass of live trees with diameter at
breast height (dbh) of at least 2.54 cm at 1.37 m above the forest floor. Separate estimates were made for above
and belowground biomass components. If inventory plots included data on individual trees, aboveground tree
carbon was based on Westfall et al. (2023). The component ratio method (CRM) which is a function of volume,
species, and diameter was used to compile estimates for woodland species where diameter measurements are
taken at root collar and to compile belowground biomass carbon for all tree species (Woodall et al. (2011a). An
additional component of foliage, which was not explicitly included in Woodall et al. (2011a), was added to each
woodland tree following the same CRM method.
Understory vegetation is a minor component of biomass and is defined as all biomass of undergrowth plants in a
forest, including woody shrubs and trees less than 2.54 cm dbh. For the current Inventory, it was assumed that 10
percent of total understory carbon mass is belowground (Smith et al. 2006). Estimates of carbon density were
6-56 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
based on information in Birdsey (1996) and biomass estimates from Jenkins et al. (2003). Understory biomass
represented over one percent of carbon in biomass, but its contribution rarely exceeded 2 percent of the total.
Biomass losses associated with conversion from grassland and cropland to forest land were assumed to occur in
the year of conversion. To account for these losses, IPCC (2006) defaults for aboveground and belowground
biomass on grasslands and aboveground biomass on croplands were subtracted from sequestration in the year of
the conversion. As previously discussed, for all other land use (i.e., other lands, settlements, wetlands) conversions
to forest land no biomass loss data were available, and no IPCC (2006) defaults currently exist to include transfers,
losses, or gains of carbon in the year of the conversion, so none were incorporated for these conversion
categories. As defaults or country-specific data become available for these conversion categories, they will be
incorporated.
Carbon in Dead Organic Matter
Dead organic matter was initially calculated as three separate pools—standing dead trees, downed dead wood,
and litter—with carbon stocks estimated from sample data or from models. The standing dead tree carbon pool
includes aboveground and belowground (coarse root) biomass for trees of at least 2.54 cm dbh. Calculations
followed the basic method applied to live trees (Westfall et al. 2023, Woodall et al. 2011a) with additional
modifications to account for decay and structural loss (Harmon et al. 2011). Downed dead wood estimates are
based on measurement of a subset of FIA plots for downed dead wood (Domke et al. 2013; Woodall and Monleon
2008; Woodall et al. 2013). Downed dead wood is defined as pieces of dead wood greater than 7.5 cm diameter, at
transect intersection, that are not attached to live or standing dead trees. This includes stumps and roots of
harvested trees. To facilitate the downscaling of downed dead wood carbon estimates from the state-wide
population estimates to individual plots, downed dead wood models specific to regions and forest types within
each region are used. Litter carbon is the pool of organic carbon (also known as duff, humus, and fine woody
debris) above the mineral soil and includes woody fragments with diameters of up to 7.5 cm. A subset of FIA plots
are measured for litter carbon. A modeling approach, using litter carbon measurements from FIA plots (Domke et
al. 2016) was used to estimate litter carbon for every FIA plot used in the estimation framework. Dead organic
matter carbon stock estimates are included for all land-use conversions to forest land.
Mineral Soil Carbon Stock Changes
A Tier 2 method is applied to estimate mineral soil C stock changes for land converted to forest land (Ogle et al.
2003, 2006; IPCC 2006). For this method, land is stratified by climate, soil types, land use, and land management
activity, and then assigned reference carbon levels and factors for the forest land and the previous land use. The
difference between the stocks is reported as the stock change under the assumption that the change occurs over
20 years. Reference C stocks have been estimated from data in the National Soil Survey Characterization Database
(USDA-NRCS 1997), and U.S.-specific stock change factors have been derived from published literature (Ogle et al.
2003, 2006). Land use and land-use change patterns are determined from a combination of the Forest Inventory
and Analysis Dataset (FIA) and the 2017 National Resources Inventory (NRI) (USDA-NRCS 2020). The areas have
been modified in the NRI survey through a process in which the Forest Inventory and Analysis (FIA) survey data and
the National Land Cover Dataset (NLCD; Yang et al. 2018) are harmonized with the NRI data. This process ensures
that the land use areas are consistent across all land-use categories (see Section 6.1 for more information). Note
that soil C in this Inventory is reported to a depth of 100 cm in the forest land remaining forest land category
(Domke et al. 2017) while other land-use categories report soil C to a depth of 30 cm. However, to ensure
consistency in the land converted to forest land category where C stock transfers occur between land-use
categories, soil C estimates were based on a 30 cm depth using methods from Ogle et al. (2003, 2006) and IPCC
(2006). See Annex 3.12 for more information about this method.
In order to ensure time-series consistency, the same methods are applied from 1990 to 2020 so that changes
reflect anthropogenic activity and not methodological adjustments. Mineral soil organic C stock changes from 2021
to 2022 are estimated using a linear extrapolation method described in Box 6-4 of the Methodology section in
Cropland Remaining Cropland. The extrapolation is based on a linear regression model with moving-average
Land Use, Land-Use Change, and Forestry 6-57
-------
(ARMA) errors using the 1990 to 2020 emissions data and is a standard data splicing method for estimating
emissions at the end of a time series if activity data are not available (IPCC 2006). The Tier 2 method described
previously will be applied to recalculate the 2021 to 2022 emissions in a future Inventory.
Uncertainty
A quantitative uncertainty analysis placed bounds on the flux estimates for land converted to forest land through a
combination of sample-based and model-based approaches to uncertainty for forest ecosystem CO2 Eq. flux (IPCC
Approach 1). Uncertainty estimates for forest pool carbon stock changes were developed using the same
methodologies as described in the forest land remaining forest land section for aboveground and belowground
biomass, dead wood, and litter. The exception was when IPCC default estimates were used for reference carbon
stocks in certain conversion categories (i.e., cropland converted to forest land and grassland converted to forest
land). In those cases, the uncertainties associated with the IPCC (2006) defaults were included in the uncertainty
calculations. IPCC Approach 2 was used to propagate errors with estimation of mineral soils carbon stock changes
for land-use conversions, and is described in the cropland remaining cropland section.
Uncertainty estimates are presented in Table 6-34 for each land conversion category and carbon pool. Uncertainty
estimates were obtained using a combination of sample-based and model-based approaches for all non-soil
carbon pools (IPCC Approach 1) and a Monte Carlo approach (IPCC Approach 2) was used for mineral soil.
Uncertainty estimates were combined using the error propagation model (IPCC Approach 1). The combined
uncertainty for all carbon stocks in land converted to forest land ranged from 11 percent below to 11 percent
above the 2022 carbon stock change estimate of-100.3 MMT CO2 Eq.
Table 6-34: Quantitative Uncertainty Estimates for Forest Carbon Pool Stock Changes (MMT
CO2 Eq. per Year) in 2022 from Land Converted to Forest Land by Land Use Change
NUX Uncertainty Range Relative to Flux Range"
Land Use/Carbon Pool Estimate
(MMT CO . Eq.) (MMT CQ . Eq.) (%)
Lower
Bound
Upper
Bound
Lower
Bound
Upper
Bound
Cropland Converted to Forest Land
(17.2)
(25.8)
(8.7)
-50%
50%
Aboveground Biomass
(10.0)
(18.4)
(1.7)
-83%
83%
Belowground Biomass
(1.7)
(2.7)
(0.7)
-61%
61%
Dead Wood
(2.2)
(3.4)
(1.0)
-56%
56%
Litter
(3.2)
(4.3)
(2.1)
-34%
34%
Non-federal Mineral Soils
(0.1)
(0.3)
0.0
-142%
142%
Federal Mineral Soils
(0.0)
(0.1)
0.1
-8,796%
8,796%
Grassland Converted to Forest Land
(37.2)
(39.6)
(34.8)
-6%
6%
Aboveground Biomass
(22.8)
(24.2)
(21.5)
-6%
6%
Belowground Biomass
(2.7)
(3.0)
(2.5)
-10%
10%
Dead Wood
(4.1)
(4.3)
(4.0)
-3%
3%
Litter
(7.7)
(8.3)
(7.2)
-7%
7%
Non-federal Mineral Soils
0.2
(0.1)
0.5
-142%
142%
Federal Mineral Soils
0.0
(0.1)
0.1
-1,310%
1,310%
Other Lands Converted to Forest Land
(5.5)
(7.8)
(3.2)
-42%
42%
Aboveground Biomass
(2.3)
(4.4)
(0.2)
-92%
92%
Belowground Biomass
(0.4)
(0.8)
0.1
-121%
121%
Dead Wood
(0.8)
(1.3)
(0.2)
-74%
74%
Litter
(1.2)
(1.8)
(0.6)
-53%
53%
Non-federal Mineral Soils
(0.9)
(1.4)
(0.5)
-49%
49%
Federal Mineral Soils
(0.0)
(0.2)
0.1
-666%
666%
Settlements Converted to Forest Land
(31.4)
(37.9)
(24.9)
-21%
21%
Aboveground Biomass
(19.4)
(25.6)
(13.3)
-32%
32%
Belowground Biomass
(3.3)
(4.6)
(2.0)
-40%
40%
6-58 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
Dead Wood
(3.8)
(4.9)
(2.6)
-31%
31%
Litter
(4.8)
(5.7)
(3.9)
-19%
19%
Non-federal Mineral Soils
(0.1)
(0.1)
(0.0)
-32%
32%
Federal Mineral Soils
(0.0)
(0.0)
0.0
-193%
193%
Wetlands Converted to Forest Land
(8.9)
(9.1)
(8.8)
-2%
2%
Aboveground Biomass
(5.0)
(5.2)
(4.9)
-3%
3%
Belowground Biomass
(0.9)
(0.9)
(0.9)
-3%
3%
Dead Wood
(1.2)
(1.3)
(1.2)
-4%
4%
Litter
(1.8)
(1.9)
(1.7)
-3%
3%
Non-federal Mineral Soils
0.0
0.0
0.0
0%
0%
Federal Mineral Soils
0.0
0.0
0.0
0%
0%
Total: Aboveground Biomass
(59.6)
(70.3)
(48.9)
-18%
18%
Total: Belowground Biomass
(9.0)
(10.7)
(7.2)
-19%
19%
Total: Dead Wood
(12.1)
(13.8)
(10.3)
-15%
15%
Total: Litter
(18.8)
(20.4)
(17.2)
-9%
8%
Total: Mineral Soils
(0.9)
(1.3)
(0.5)
-42%
42%
Total: Lands Converted to Forest Lands
(100.3)
(111.4)
(89.2)
-11%
11%
+ Absolute value does not exceed 0.05 MMT C02 Eq.
NA (Not Applicable)
a Range of flux estimate for 95 percent confidence interval.
Notes: Totals may not sum due to independent rounding. Parentheses indicate net uptake. It is not possible to separate
emissions from drained organic soils between forest land remaining forest land and land converted to forest land so
estimates for organic soils are included in Table 6-8 and Table 6-9 of the Forest Land Remaining Forest Land section of the
Inventory.
/erification
See QA/QC and Verification sections under Forest Land Remaining Forest Land and for mineral soil estimates,
Cropland Remaining Cropland.
Recalculations Discussion
The approach for estimating carbon stock changes in land converted to forest land is consistent with the methods
used for forest land remaining forest Land and is described in Annex 3.13. The land converted to forest land
estimates in this Inventory are based on the land-use change information in the annual NFI. All conversions are
based on empirical estimates compiled using plot remeasurements from the NFI, IPCC (2006) default biomass
carbon stocks removed from croplands and grasslands in the year of conversion on individual plots and the Tier 2
method for estimating mineral soil carbon stock changes (Ogle et al. 2003, 2006; IPCC 2006). All annual NFI plots
included in the public FIA database as of September 2023 were used in this Inventory. This is the fourth year that
remeasurement data from the annual NFI were available throughout the conterminous United States (with the
exception of Wyoming) and coastal southeast and southcentral Alaska to estimate land-use conversion. The
availability of remeasurement data from the annual NFI allowed for consistent plot-level estimation of carbon
stocks and stock changes for forest land remaining forest land and the land converted to forest land categories.
Estimates in the previous Inventory were based on state-level carbon density estimates and a combination of NRI
data and NFI data in the eastern United States. The refined analysis in this Inventory resulted in changes in the land
converted to forest land categories. Overall, the land converted to forest land carbon stock changes increased by
approximately 2 percent in 2022 between the previous Inventory (1990 through 2021) and the current Inventory
(Table 6-35). While the overall change is relatively small, changes by conversion categories were substantial
between this Inventory and the previous Inventory. These changes can be attributed to six methodological
improvements implemented this year. First, managed pastureland was previously classified as cropland and is now
classified as grassland to align with NRI definitions and classifications. This resulted in a substantial structural
decrease in the cropland converted to forest land category area and associated carbon stock changes and a
substantial structural increase in the grassland converted to forest land area and associated carbon stock changes.
Land Use, Land-Use Change, and Forestry 6-59
-------
Second, in this Inventory all NFI plots with evidence of water are classified as wetlands to align with definitions and
classifications in the NRI. In the previous Inventory, some NFI plots with evidence of water were classified as other
land. This methodological improvement resulted in increases in the wetlands converted to forest land category
and decreases in the other land converted to forest land category. Third, estimates of carbon stocks and stock
changes on managed forest land in coastal Alaska are now compiled in the same ways as the conterminous United
States allowing for estimates of land conversions. This led to small increases in the area and associated carbon
stock change estimates in the land converted to forest land category. Fourth, the implementation of new methods
for estimating aboveground biomass carbon in live and standing dead trees resulted in changes across all land use
conversion categories in the aboveground and belowground biomass, dead wood, and litter pools. Fifth, new
climate normals (1990 through 2020) were incorporated in the litter model resulting in additional changes in that
pool. Finally, there were new NFI data incorporated into the latest Inventory which contributed to changes when
compared with the previous Inventory.
Table 6-35: Recalculations of the Net Carbon Flux from Forest Carbon Pools in Land
Converted to Forest Land by Land Use Change Category (MMT C)
Conversion category
2021 Estimate,
2021 Estimate,
2022 Estimate,
and Carbon pool (MMT C)
Previous Inventory
Current Inventory
Current Inventory
Cropland Converted to Forest Land
(10.3)
(4.7)
(4.7)
Aboveground Biomass
(6.0)
(2.7)
(2.7)
Belowground Biomass
(1.2)
(0.5)
(0.5)
Dead Wood
(1.3)
(0.6)
(0.6)
Litter
(1.8)
(0.9)
(0.9)
Mineral Soil
(0.1)
(0.0)
(0.0)
Grassland Converted to Forest Land
(3.4)
(10.2)
(10.1)
Aboveground Biomass
(1.7)
(6.2)
(6.2)
Belowground Biomass
(0.3)
(0.7)
(0.7)
Dead Wood
(0.3)
(1.1)
(1.1)
Litter
(1.1)
(2.1)
(2.1)
Mineral Soil
0.1
0.1
0.1
Other Land Converted to Forest Land
(2.9)
(1.5)
(1.5)
Aboveground Biomass
(1.3)
(0.6)
(0.6)
Belowground Biomass
(0.2)
(0.1)
(0.1)
Dead Wood
(0.4)
(0.2)
(0.2)
Litter
(0.7)
(0.3)
(0.3)
Mineral Soil
(0.3)
(0.3)
(0.3)
Settlements Converted to Forest Land
(9.3)
(8.6)
(8.6)
Aboveground Biomass
(5.7)
(5.3)
(5.3)
Belowground Biomass
(1.1)
(0.9)
(0.9)
Dead Wood
(1.1)
(1.0)
(1.0)
Litter
(1.5)
(1.3)
(1.3)
Mineral soil
(0.0)
(0.0)
(0.0)
Wetlands Converted to Forest Land
(0.9)
(2.4)
(2.4)
Aboveground Biomass
(0.4)
(1.4)
(1.4)
Belowground Biomass
(0.1)
(0.2)
(0.2)
Dead Wood
(0.1)
(0.3)
(0.3)
Litter
(0.4)
(0.5)
(0.5)
Mineral Soil
0.0
0.0
0.0
Total Aboveground Biomass Flux
(15.0)
(16.3)
(16.3)
Total Belowground Biomass Flux
(2.8)
(2.4)
(2.4)
Total Dead Wood Flux
(3.2)
(3.3)
(3.3)
Total Litter Flux
(5.5)
(5.1)
(5.1)
Total SOC (Mineral) Flux
(0.3)
(0.3)
(0.2)
Total Flux
(26.8)
(27.4)
(27.4)
Notes: Totals may not sum due to independent rounding. Parentheses indicate net sequestration.
6-60 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
Planned Improvements
There are many improvements necessary to improve the estimation of carbon stock changes associated with land-
use conversion to forest land over the entire time series. First, soil carbon has historically been reported to a depth
of 100 cm in the forest land remaining forest land category (Domke et al. 2017) while other land-use categories
(e.g., grasslands and croplands) report soil carbon to a depth of 30 cm. To ensure greater consistency in the land
converted to forest land category where carbon stock transfers occur between land-use categories, all mineral soil
estimates in the land converted to forest land category in this Inventory are based on methods from Ogle et al.
(2003, 2006) and IPCC (2006). Methods have recently been developed (Domke et al. 2017) to estimate soil carbon
to depths of 20, 30, and 100 cm in the forest land category using in situ measurements from the FIA program
within the USDA Forest Service and the International Soil Carbon Network. In subsequent Inventories, a common
reporting depth will be defined for all land-use conversion categories and Domke et al. (2017) will be used in the
forest land remaining forest land and land converted to forest land categories to ensure consistent reporting
across all forest land. Second, due to the 5 to 10-year remeasurement periods within the FIA program and limited
land-use change information available over the entire time series, estimates presented in this section may not
reflect the entire 20-year conversion history. Third, since the sum of all land converted to forest land is used to
adjust specific land-use conversions into forest land for the state-level estimates in the NRI and NLCD, there is the
potential for differences in area estimates in states where specific land-use conversions into forest land do not
exist in the FIA data. These difference in area estimates may result in differences between the summed estimates
for mineral soil carbon stock changes across all states and the estimates reported in Table 6-31 through Table 6-34.
Work is underway to integrate the dense time series of remotely sensed data into a new estimation system, which
will facilitate land conversion estimation over the entire time series.
6.4 Cropland Remaining Cropland (CRT
Category 4B1)
Carbon in cropland ecosystems occurs in biomass, dead organic matter, and soils. However, carbon storage in
cropland biomass and dead organic matter is relatively ephemeral and does not need to be reported according to
the IPCC (2006), with the exception of carbon stored in perennial woody crop biomass, such as citrus groves and
apple orchards, in addition to the biomass, downed wood and dead organic matter in agroforestry systems. Within
soils, carbon is found in organic and inorganic forms of carbon, but soil organic carbon is the main source and sink
for atmospheric CO2. IPCC (2006) recommends reporting changes in soil organic carbon stocks due to agricultural
land use and management activities for mineral and organic soils.38
Well-drained mineral soils typically contain from 1 to 6 percent organic carbon by weight, whereas mineral soils
with high water tables for substantial periods of a year may contain significantly more carbon (NRCS 1999).
Conversion of mineral soils from their native state to agricultural land uses can cause up to half of the soil organic
carbon to be lost to the atmosphere due to enhanced microbial decomposition. The rate and ultimate magnitude
of carbon loss depends on subsequent management practices, climate and soil type (Ogle et al. 2005). Agricultural
practices, such as clearing, drainage, tillage, planting, grazing, crop residue management, fertilization, application
of biosolids (i.e., treated sewage sludge) and flooding, can modify both organic matter inputs and decomposition,
and thereby result in a net carbon stock change (Paustian et al. 1997a; Lai 1998; Conant et al. 2001; Ogle et al.
2005; Griscom et al. 2017; Ogle et al. 2019). Eventually, the soil can reach a new equilibrium that reflects a balance
38 Carbon dioxide emissions associated with liming and urea application are also estimated but are included in the Liming and
Urea Fertilization sections of the Agriculture chapter of the Inventory.
Land Use, Land-Use Change, and Forestry 6-61
-------
between carbon inputs (e.g., decayed plant matter, roots, and organic amendments such as manure and crop
residues) and carbon loss through microbial decomposition of organic matter (Paustian et al. 1997b).
Organic soils, also referred to as histosols, include all soils with more than 12 to 20 percent organic carbon by
weight, depending on clay content (NRCS 1999; Brady and Weil 1999). The organic layer of these soils can be very
deep (i.e., several meters), and form under inundated conditions that results in minimal decomposition of plant
residues. When organic soils are prepared for crop production, they are drained and tilled, leading to aeration of
the soil that accelerates both the decomposition rate and CO2 emissions.39 Due to the depth and richness of the
organic layers, carbon loss from drained organic soils can continue over long periods of time, which varies
depending on climate and composition (i.e., decomposability) of the organic matter (Armentano and Menges
1986). Due to deeper drainage and more intensive management practices, the use of organic soils for annual crop
production leads to higher carbon loss rates than drainage of organic soils in grassland or forests (IPCC 2006).
Cropland remaining cropland includes all cropland in an inventory year that has been cropland for a continuous
time period of at least 20 years. This determination is based on the United States Department of Agriculture
(USDA) National Resources Inventory (NRI) for non-federal lands (USDA-NRCS 2020) and the National Land Cover
Dataset for federal lands (Yang et al. 2018; Homer et al. 2007; Fry et al. 2011; Homer et al. 2015). Cropland
includes all land that is used to produce food and fiber, forage that is harvested and used as feed (e.g., hay and
silage), in addition to cropland that has been enrolled in the Conservation Reserve Program (CRP)40 (i.e.,
considered set-aside cropland).
There are two discrepancies between the current land representation (see Section 6.1) and the area data that have
been used in the Inventory for cropland remaining cropland. First, croplands in Alaska are not included in the
Inventory, and second, some miscellaneous croplands that occur throughout the United States are also not
included in the Inventory due to limited understanding of greenhouse gas emissions from these management
systems (e.g., aquaculture). These differences lead to discrepancies between the managed area in cropland
remaining cropland and the cropland area included in the Inventory analysis (Table 6-39). Improvements are
underway to incorporate croplands in Alaska and miscellaneous croplands as part of future Inventories (see
Planned Improvements section).
Land use and land management of mineral soils are the largest contributor to total net carbon stock change,
especially in the early part of the time series (see Table 6-36 and Table 6-37). In 2022, mineral soils are estimated
to sequester 62.0 MMT CO2 Eq. (16.9 MMT C). This level of carbon storage in mineral soils represents a more than
58 percent increase since the initial reporting year of 1990. Carbon dioxide emissions from organic soils are 30.3
MMT CO2 Eq. (8.3 MMT C) in 2022, which is an 11 percent decrease in losses of soil carbon compared to 1990. In
total, United States agricultural soils in cropland remaining cropland sequestered approximately 31.7 MMT CO2 Eq.
(8.6 MMT C) in 2022.
Table 6-36: Net CO2 Flux from Soil Carbon Stock Changes in Cropland Remaining Cropland
(MMT C02 Eq.)
Soil Type
1990
2005
2018
2019
2020
2021
2022
Mineral Soils
(39.2) 1
(61.8) j
(47.1)
(48.5)
(38.2)
(62.2)
(62.0)
Organic Soils
34.2
30.2 J
29.3
29.1
29.4
30.2
30.3
39 N20 emissions from drained organic soils are included in the Agricultural Soil Management section of the Agriculture chapter
of the Inventory.
40 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 to 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.
6-62 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
Total Net Flux (5.0) (31.6) (17.8) (19.4) (8.8) (32.0) (31.7)
Notes: Totals may not sum due to independent rounding. Parentheses indicate net
sequestration.
Table 6-37: Net CO2 Flux from Soil Carbon Stock Changes in Cropland Remaining Cropland
(MMT C)
Soil Type
1990
2005
2018
2019
2020
2021
2022
Mineral Soils
(10.7)
(16.9) |
(12.9)
(13.2)
(10.4)
(17.0)
(16.9)
Organic Soils
51.
8-2 I
8.0
7.9
8.0
8.2
8.3
Total Net Flux
(1-4)
(8.6)
(4.9)
(5.3)
(2.4)
(8.7)
(8.6)
Notes: Totals may not sum due to independent rounding. Parentheses indicate net
sequestration.
Soil organic carbon stocks increase in cropland remaining cropland largely due to conservation tillage (i.e.,
reduced- and no-till practices), land set-aside from production in the Conservation Reserve Program, annual crop
production with hay or pasture in rotations, and manure amendments (Ogle et al. 2023). The mineral soil carbon
stock changes between 1990 and 2022 range from 38.2 to 69.6 MMT CO2 Eq. per year, with a mean of 55.9 MMT
CO2 Eq. Soil organic carbon losses from drainage of organic soils are relatively stable across the time series with a
mean emission of 30.2 MMT CO2 Eq. per year.
The spatial variability in the 2020 annual soil organic carbon stock changes41 are displayed in Figure 6-6 and Figure
6-7 for mineral and organic soils, respectively. Isolated areas with high rates of carbon accumulation occur
throughout the agricultural land base in the United States, but there are more concentrated areas, such as the
Maryland, Delaware, and Virginia where there have been relatively high adoption rates of cover crop
management. The regions with the highest rates of emissions from drainage of organic soils occur in the
Southeastern Coastal Region (particularly Florida and Louisiana), Northeast and upper Midwest surrounding the
Great Lakes, and isolated areas along the Pacific Coast (particularly California), which coincides with the largest
concentrations of organic soils in the United States that are used for agricultural production.
41 Only national-scale emissions are estimated for 2021 to 2022 in this Inventory using the surrogate data method, and
therefore the fine-scale emission patterns in this map are based on Inventory data from 2020.
Land Use, Land-Use Change, and Forestry 6-63
-------
Figure 6-6: Total Net Annual Soil Carbon Stock Changes for Mineral Soils under Agricultural
Management within States, 2020, Cropland Remaining Cropland
1- $
\
V -
&
•i \
S f
J
MT C02 ha1 yr"1
< -4 ~ 1 to 2
-4 to -2 | 2 to 4
-2 to -1 7 > 4
J -1 to 1
v..
:v:
4:.
; <\
\
... i :
W , W" 4 >
/ <«W—
^ ... vVr'^\
Note: Only national-scale soil organic carbon stock changes are estimated for 2021 to 2022 in the current Inventory using a
surrogate data method, and therefore the fine-scale emission patterns in this map are based on Inventory data from 2020.
Negative values represent a net increase in soil organic carbon stocks, and positive values represent a net decrease in soil
organic carbon stocks.
6-64 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
Figure 6-7: Total Net Annual Soil Carbon Stock Changes for Organic Soils under Agricultural
Management within States, 2020, Cropland Remaining Cropland
HIT
¦T*
\ v
\
/ l\
\
...
v. (
~x
MT C02 ha1 yr1
~ < 10
~ 10 to 20
¦ 20 to 30
¦ 30 to 40
¦ > 40
Note: Only national-scale soil organic carbon stock changes are estimated for 2021 to 2022 in the current Inventory using a
surrogate data method, and therefore the fine-scale emission patterns in this map are based on Inventory data from 2020.
Methodology and Time-Series Consistency
The following section includes a description of the methodology used to estimate changes in soil organic carbon
stocks for cropland remaining cropland, including (1) agricultural land use and management activities on mineral
soils; and (2) agricultural land use and management activities on organic soils. Carbon dioxide emissions and
removals42 due to changes in mineral soil organic carbon stocks are estimated using a Tier 3 method for the
majority of annual crops (Ogle et al. 2010, 2023). A Tier 2 IPCC method is used for the remaining crops not included
in the Tier 3 method (see list of crops in the Mineral Soil Carbon Stock Changes section below) (Ogle et al. 2003,
2006). In addition, a Tier 2 method is used for very gravelly, cobbly, or shaley soils (i.e., classified as soils that have
greater than 35 percent of soil volume comprised of gravel, cobbles, or shale, regardless of crop). Emissions from
organic soils are estimated using a Tier 2 IPCC method. While a combination of Tier 2 and 3 methods are used to
estimate carbon stock changes across most of the time series, a surrogate data method has been applied to
42
Removals occur through uptake of C02 into crop and forage biomass that is later incorporated into soil carbon pools.
Land Use, Land-Use Change, and Forestry 6-65
-------
estimate stock changes in the last two years of the Inventory. Stock change estimates based on surrogate data will
be recalculated in a future Inventory report using the Tier 2 and 3 methods when data become available.
Soil organic carbon stock changes on non-federal lands are estimated for cropland remaining cropland (as well as
agricultural land falling into the IPCC categories land converted to cropland, grassland remaining grassland, and
land converted to grassland) according to land-use histories recorded in the USDA NRI survey through 2017 (USDA-
NRCS 2020), and the cropping histories were extended through 2020 using the USDA-NASS Crop Data Layer
Product (CDL) (USDA-NASS 2021). The areas have been modified in the original NRI survey through a process in
which the Forest Inventory and Analysis (FIA) survey data and the National Land Cover Dataset (Yang et al. 2018)
are harmonized with the NRI data. This process ensures that the land-use areas are consistent across all land- use
categories (see Section 6.1 for more information).
The NRI is a statistically-based sample and includes approximately 604,000 survey locations in agricultural land for
the conterminous United States and Hawaii. There are 364,333 survey locations that are included in the Tier 3
method, and another 239,757 locations included in the Tier 2 method. Each survey location is associated with a
weight that allows scaling of carbon stock changes from NRI survey locations to the entire country (i.e., each
weight represents the amount of area that is expected to have the same land use/management history as the
sample point).
Land use and some management information (e.g., crop type, soil attributes, and irrigation) are collected for each
NRI point on a 5-year cycle beginning from 1982 through 1997. For cropland, data has been collected for 4 out of 5
years during each survey cycle (i.e., 1979 through 1982,1984 through 1987,1989 through 1992, and 1994 through
1997). In 1998, the NRI program began collecting annual data, and the annual data are currently available through
2017 (USDA-NRCS 2020). For 2018 to 2020, the time series is extended with the crop data provided in USDA-NASS
CDL (USDA-NASS 2021), by overlaying NRI survey locations on the CDL in a geographic information system and
extracting the crop types to extend the cropping histories. NRI survey locations are classified as cropland remaining
cropland in a given year between 1990 and 2020 if the land use has been cropland for a continuous time period of
at least 20 years. The 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 1998. 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 between 1971 and 1978.
Soil Carbon Stock Changes for Mineral Soils
An IPCC Tier 3 model-based approach (Ogle et al. 2010) is applied to estimate organic carbon stock changes for
mineral soils on the majority of land that is used to produce annual crops and forage crops that are harvested and
used as feed (e.g., hay and silage) in the United States. These crops include alfalfa hay, barley, corn, cotton, dry
beans, grass hay, grass-clover hay, lentils, oats, onions, peanuts, peas, potatoes, rice, sorghum, soybeans, sugar
beets, sunflowers, tobacco, tomatoes, and wheat, but is not applied to estimate organic carbon stock changes
from other crops or rotations with other crops. The model-based approach uses the DayCent ecosystem model
(Parton et al. 1998; Del Grosso et al. 2001, 2011) to estimate soil organic carbon stock changes, soil nitrous oxide
(N2O) emissions from agricultural soil management, and methane (CH4) emissions from rice cultivation. Carbon and
nitrogen dynamics are linked in plant-soil systems through the biogeochemical processes of microbial
decomposition and plant production (McGill and Cole 1981). Coupling the two source categories (i.e., agricultural
soil carbon and N2O) in a single inventory analysis ensures that there is a consistent treatment of the processes
and interactions between carbon and nitrogen cycling in soils.
The remaining crops on mineral soils are estimated using an IPCC Tier 2 method (Ogle et al. 2003), including some
vegetables, perennial/horticultural crops, and crops that are rotated with these crops. The Tier 2 method is also
used for very gravelly, cobbly, or shaley soils (greater than 35 percent by volume), and soil organic carbon stock
changes on federal croplands. Mineral soil organic carbon stocks are estimated using a Tier 2 method for these
areas because the DayCent model, which is used for the Tier 3 method, has not been fully tested for estimating
carbon stock changes associated with these crops and rotations, as well as cobbly, gravelly, or shaley soils. In
addition, there is insufficient information to simulate croplands on federal lands using DayCent.
6-66 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
A surrogate data method is used to estimate soil organic carbon stock changes from 2021 to 2022 at the national
scale for land areas included in the Tier 2 and Tier 3 methods. Specifically, linear regression models with
autoregressive moving-average (ARMA) errors (Brockwell and Davis 2016) are used to estimate the relationship
between surrogate data and the 1990 to 2020 stock change data that are derived using the Tier 2 and 3 methods.
Surrogate data for these regression models include corn and soybean yields from USDA-NASS statistics,43 and
weather data from the PRISM Climate Group (PRISM 2022). See Box 6-4 for more information about the surrogate
data method. Stock change estimates for 2021 to 2022 will be recalculated in future Inventories with an updated
time series of activity data.
Box 6-4: Surrogate Data Method
Time series extension is needed because there are typically gaps at the end of the time series. This is mainly
because the NRI, which provides critical data for estimating greenhouse gas emissions and removals, does not
release new activity data every year.
A surrogate data method has been used to impute missing emissions at the end of the time series for soil
organic carbon stock changes in cropland remaining cropland, land converted to cropland, grassland remaining
grassland, and land converted to grassland. A linear regression model with autoregressive moving-average
(ARMA) errors (Brockwell and Davis 2016) is used to estimate the relationship between the surrogate data and
the modeled 1990 to 2020 emissions data that has been compiled using the inventory methods described in this
section. The model to extend the time series is given by
Y = Xp + e,
where Y is the response variable (e.g., soil organic carbon), xp contains specific surrogate data depending on the
response variable, and £ is the remaining unexplained error. Models with a variety of surrogate data were
tested, including commodity statistics, weather data, or other relevant information. Parameters are estimated
from the emissions data for 1990 to 2020 using standard statistical techniques, and these estimates are used to
predict the missing emissions data for 2021 to 2022.
A critical issue with the application of splicing methods is to adequately account for the additional uncertainty
introduced by predicting emissions rather than compiling the full inventory. Consequently, uncertainty will
increase for years with imputed estimates based on the splicing methods, compared to those years in which the
full inventory is compiled. This added uncertainty is quantified within the model framework using a Monte Carlo
approach. The approach requires estimating parameters for results in each iteration of the Monte Carlo analysis
for the full inventory (i.e., the surrogate data model is refit with the emissions estimated in each Monte Carlo
iteration from the full inventory analysis with data from 1990 to 2020), estimating emissions from each model
and deriving confidence intervals combining uncertainty across all iterations. This approach propagates
uncertainties through the calculations from the original inventory and the surrogate data method. Furthermore,
the 95 percent confidence intervals are estimated using the 3 sigma rules assuming a unimodal density
(Pukelsheim 1994).
Tier 3 Approach. Mineral soil organic carbon stocks and stock changes are estimated to a 30 cm depth using the
DayCent ecosystem model (Parton et al. 1998; Del Grosso et al. 2001, 2011), which simulates cycling of carbon,
nitrogen, and other nutrients in cropland, grassland, forest, and savanna ecosystems. The DayCent model utilizes
the soil carbon modeling framework developed in the Century model (Parton et al. 1987,1988,1994; Metherell et
al. 1993), but has been refined to simulate dynamics at a daily time-step. Input data on land use and management
are specified at a daily resolution and include land-use type, crop/forage type, and management activities (e.g.,
43 See https://quickstats.nass.usda.eov/.
Land Use, Land-Use Change, and Forestry 6-67
-------
planting, harvesting, fertilization, manure amendments, tillage, irrigation, cover crops, and grazing; more
information is provided below). The model simulates net primary productivity (NPP) using the NASA-CASA
production algorithm MODIS Enhanced Vegetation Index (EVI) products, MOD13Q1 and MYD13Q1, for most
croplands44 (Potter et al. 1993, 2007). The model simulates soil temperature and water dynamics, using daily
weather data from a 4-kilometer gridded product developed by the PRISM Climate Group (2022), and soil
attributes from the Soil Survey Geographic Database (SSURGO) (Soil Survey Staff 2020). This method is more
accurate than the Tier 1 and 2 approaches provided by the IPCC (2006) because the simulation model treats
changes as continuous over time as opposed to the simplified discrete changes represented in the default method
(see Box 6-5 for additional information).
Box 6-5: Tier 3 Approach for Soil Carbon Stocks Compared to Tier 1 or 2 Approaches
A Tier 3 model-based approach is used to estimate soil organic carbon stock changes for the majority of
agricultural land with mineral soils. This approach results in a more complete and accurate estimation of soil
organic carbon stock changes and entails several fundamental differences from the IPCC Tier 1 or 2 methods, as
described below.
1) The IPCC Tier 1 and 2 methods are simplified approaches for estimating soil organic carbon stock
changes and classify land areas into discrete categories based on highly aggregated information about
climate (six regions), soil (seven types), and management (eleven management systems) in the United
States. In contrast, the Tier 3 model incorporates the same variables (i.e., climate, soils, and
management systems) with considerably more detail both temporally and spatially, and captures multi-
dimensional interactions through the more complex model structure.
2) The IPCC Tier 1 and 2 methods have a coarser spatial resolution in which data are aggregated to soil
types in climate regions, of which there about 30 combinations in the United States. In contrast, the
Tier 3 model simulates soil carbon dynamics at about 364,000 individual NRI survey locations in crop
fields and grazing lands.
The IPCC Tier 1 and 2 methods use a simplified approach for estimating changes in carbon stocks that assumes a
step-change from one equilibrium level of the carbon stock to another equilibrium level. In contrast, the Tier 3
approach simulates a continuum of carbon stock changes that may reach a new equilibrium over an extended
period of time depending on the environmental conditions (i.e., a new equilibrium often requires hundreds to
thousands of years to reach). More specifically, the DayCent model, which is used in the United States
Inventory, simulates soil carbon dynamics (and CO2 emissions and uptake) on a daily time step based on carbon
emissions and removals from plant production and decomposition processes. These changes in soil organic
carbon stocks are influenced by multiple factors that affect primary production and decomposition, including
changes in land use and management, weather variability and secondary feedbacks between management
activities, climate, and soils.
Historical land-use patterns and irrigation histories are simulated with DayCent based on the 2017 USDA NRI
survey (USDA-NRCS 2020). Additional sources of activity data are used to supplement the activity data from the
NRI. The USDA-NRCS Conservation Effects and Assessment Project (CEAP) provides data on a variety of cropland
management activities, and is used to inform the inventory analysis about tillage practices, mineral fertilization,
manure amendments, cover crop management, as well as planting and harvest dates (USDA-NRCS 2022; USDA-
44 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 to 2020. 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.
6-68 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
NRCS 2018; USDA-NRCS 2012). CEAP data are collected at a subset of NRI survey locations, and currently provide
management information from approximately 2002 to 2006 and 2013 to 2016. These data are combined with
other datasets in an imputation analysis. This imputation analysis is comprised of three steps: a) determine the
trends in management activity across the time series by combining information from several datasets (discussed
below); b) use Gradient Boosting (Friedman 2001) to determine the likely management practice at a given NRI
survey location; and c) assign management practices from the CEAP survey to the specific NRI locations using a
predictive mean matching method for certain variables that are adapted to reflect the trending information (Little
1988; van Buuren 2012). Gradient boosting is a machine learning technique used in regression and classification
tasks, among others. It combines predictions from multiple weak prediction models and outperforms many
complicated machine learning algorithms. It makes the best predictions at specific NRI survey locations or at state
or region level models. The predictive mean matching method identifies the most similar management activity
recorded in the CEAP surveys that match the prediction from the gradient boosting algorithm. The matching
ensures that imputed management activities are realistic for each NRI survey location, and not odd or physically
unrealizable results that could be generated by the gradient boosting. There are six complete imputations of the
management activity data using these methods.
To determine trends in mineral fertilization and manure amendments, CEAP data are combined with information
on fertilizer use and rates by crop type for different regions of the United States from the USDA Economic
Research Service. The data collection program was known as the Cropping Practices Surveys through 1995 (USDA-
ERS 1997), and is now part of data collection known as the Agricultural Resource Management Surveys (ARMS)
(USDA-ERS 2020). Additional data on fertilization practices are compiled through other sources particularly the
National Agricultural Statistics Service (USDA-NASS 1992,1999, 2004). To determine the trends in tillage
management, CEAP data are combined with Conservation Technology Information Center data between 1989 and
2004 (CTIC 2004) and OpTIS Data Product45 for 2008 to 2020 (Hagen et al. 2020). The CTIC data are adjusted for
long-term adoption of no-till agriculture (Towery 2001). For cover crops, CEAP data are combined with information
from USDA Census of Agriculture (USDA-NASS 2012, 2017) and the OpTIS data (Hagen et al. 2020). It is assumed
that cover crop management was minimal prior to 1990 and the rates increased linearly over the decade to the
levels of cover crop management in the CEAP survey.
Uncertainty in the carbon stock estimates from DayCent associated with management activity includes input
uncertainty due to missing management data in the NRI survey, which is imputed from other sources as discussed
above; model uncertainty due to incomplete specification of carbon and nitrogen dynamics in the DayCent model
algorithms and associated parameterization; and sampling uncertainty associated with the statistical design of the
NRI survey. Uncertainty is estimated with two variance components (Ogle et al. 2010). The first variance
component quantifies the uncertainty in management activity data, model structure and parameterization. To
assess this uncertainty, carbon and nitrogen dynamics at each NRI survey location are simulated six times using the
imputation product and other model driver data. Uncertainty in parameterization and model algorithms are
determined using a structural uncertainty estimator derived from fitting a linear mixed-effect model (Ogle et al.
2007, 2010, 2023). The data are combined in a Monte Carlo stochastic simulation with 1,000 iterations for 1990
through 2020. For each iteration, there is a random selection of management data from the imputation product
(select one of the six imputations), and random selection of parameter values and random effects for the linear
mixed-effect model (i.e., structural uncertainty estimator). The second variance component quantifies uncertainty
in scaling from the NRI survey to the entire land base, and is computed with the NRI replicate weights using a
standard variance estimator for a two-stage sample design (Sarndal et al. 1992). The two variance components are
combined using 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. Carbon stocks and 95 percent
confidence intervals are estimated for each year between 1990 and 2020 using the DayCent model. Further
elaboration on the methodology and data used to estimate carbon stock changes from mineral soils are described
in Annex 3.12.
45 OpTIS data on tillage and cover crop practices provided by Regrow Agriculture, Inc.
Land Use, Land-Use Change, and Forestry 6-69
-------
In order to ensure time-series consistency, the Tier 3 method is applied from 1990 to 2020 so that changes reflect
anthropogenic activity and not methodological adjustments. In addition, soil organic carbon stock changes from
2021 to 2022 are approximated with a linear extrapolation of emission patterns from 1990 to 2020. The
extrapolation is based on a linear regression model with moving-average (ARMA) errors (see Box 6-4). Linear
extrapolation is a standard data splicing method for approximating emissions at the end of a time series (IPCC
2006). The time series of activity data will be updated in a future inventory, and emissions from 2021 to 2022 will
be recalculated.
Tier 2 Approach. In the IPCC Tier 2 method, data on climate, soil types, land use, and land management activity are
used to classify land area and apply appropriate factors to estimate soil organic carbon stock changes to a 30 cm
depth (Ogle et al. 2003, 2006). The primary source of activity data for land use, crop and irrigation histories is the
2017 NRI survey (USDA-NRCS 2020). Each NRI survey location is classified by soil type, climate region, and
management condition using data from other sources. Survey locations on federal lands are included in the NRI,
but land use and cropping history are not compiled for these locations in the survey program (i.e., NRI is restricted
to data collection on non-federal lands). Therefore, land-use patterns for the NRI survey locations on federal lands
are based on the National Land Cover Database (NLCD) (Yang et al. 2018; Fry et al. 2011; Homer et al. 2007; Homer
et al. 2015).
Additional management activities needed for the Tier 2 method are based on the imputation product described for
the Tier 3 approach, including tillage practices, mineral fertilization, and manure amendments that are assigned to
NRI survey locations. Activity data used exclusively in the Tier 2 method are wetland restoration for Conservation
Reserve Program land from Euliss and Gleason (2002). Climate zones in the United States are determined from the
IPCC climate map (IPCC 2006), and then assigned to NRI survey locations.
Reference carbon stocks are estimated using the National Soil Survey Characterization Database (NRCS 1997) with
cultivated cropland as the reference condition, rather than native vegetation as used in IPCC (2006). Soil
measurements under agricultural management are much more common and easily identified in the National Soil
Survey Characterization Database (NRCS 1997) than are soils under a native condition, and therefore cultivated
cropland provides a more robust sample for estimating the reference condition. Country-specific carbon stock
change factors are derived from published literature to determine the impact of management practices on soil
organic carbon storage (Ogle et al. 2003, 2006). The factors represent changes in tillage, cropping rotations,
intensification, and land-use change between cultivated and uncultivated conditions. However, country-specific
factors associated with organic matter amendments are not estimated due to an insufficient number of studies in
the United States to analyze the impacts of this practice. Instead, factors from IPCC (2006) are used to estimate the
effect of those activities.
Uncertainty in soil carbon stock changes is estimated with two variance components (Ogle et al., 2010). The first
variance component quantifies the uncertainty in management activity data and carbon stock change factors. To
assess this uncertainty, changes in soil organic carbon stocks for mineral soils are estimated 1,000 times for 1990
through 2020 using a Monte Carlo stochastic simulation approach and probability distribution functions for the
country-specific stock change factors, reference carbon stocks, and land use activity data (Ogle et al. 2003; Ogle et
al. 2006). The second variance component quantifies uncertainty in scaling from the NRI survey to the entire land
base, and is computed with the NRI replicate weights using a standard variance estimator for a two-stage sample
design (Sarndal et al. 1992). The two variance components are combined using 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. Further elaboration on the methodology and data used to estimate stock changes from
mineral soils are described in Annex 3.12.
In order to ensure time-series consistency, the Tier 2 method is applied from 1990 to 2020 so that changes reflect
anthropogenic activity and not methodological adjustments. In addition, soil organic carbon stock changes for the
remainder of the time series are approximated with a linear extrapolation of emission patterns from 1990 to 2020.
The extrapolation is based on a linear regression model with moving-average (ARMA) errors (see Box 6-4). Linear
extrapolation is a standard data splicing method for approximating emissions at the end of a time series (IPCC
6-70 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
2006). As with the Tier 3 method, time series of activity data will be updated in a future inventory, and emissions
from 2021 to 2022 will be recalculated (see Planned Improvements section).
Soil Carbon Stock Changes for Organic Soils
Annual carbon emissions from drained organic soils in cropland remaining cropland are estimated using the Tier 2
method provided in IPCC (2006), with country-specific carbon loss rates (Ogle et al. 2003) rather than default IPCC
rates. As with mineral soils, uncertainty is estimated with two variance components (Ogle et al., 2010). The first
variance component quantifies the uncertainty in management activity data and emission factors. A Monte Carlo
stochastic simulation with 1,000 iterations is used to quantify this uncertainty with probability distribution
functions for the country-specific organic soil emission factors and land use activity data. The second variance
component quantifies uncertainty in scaling from the NRI survey to the entire land base, and is computed with the
NRI replicate weights using a standard variance estimator for a two-stage sample design (Sarndal et al. 1992). The
two variance components are combined using 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. Further
elaboration on the methodology and data used to estimate stock changes from organic soils are described in
Annex 3.12.
In order to ensure time-series consistency, the same Tier 2 method is applied from 1990 to 2020 so that changes
reflect anthropogenic activity and not methodological adjustments. In addition, soil organic carbon stock changes
for the remainder of the time series are approximated with a linear extrapolation of emission patterns from 1990
to 2020. The extrapolation is based on a linear regression model with moving-average (ARMA) errors (see Box 6-4).
Linear extrapolation is a standard data splicing method for approximating emissions at the end of a time series
(IPCC 2006). Estimates for 2021 to 2022 will be recalculated in a future inventory when new activity data are
incorporated into the analysis.
Uncertainty
Uncertainty is quantified for changes in soil organic carbon stocks associated with cropland remaining cropland
(including both mineral and organic soils). Uncertainty estimates are presented in Table 6-38 for each subsource
(mineral and organic soil carbon stocks) and the methods that are used in the Inventory analyses (i.e., Tier 2 and
Tier 3). Uncertainty for the Tier 2 and 3 approaches is derived from two variance components (Ogle et al. 2010).
For the first component, a Monte Carlo approach is used to address uncertainties in management activity data as
well as model parameterization and structure or emissions factors for the Tier 3 and Tier 2 methods, respectively
(Ogle et al. 2010, 2023). The second variance component is quantifying uncertainty in scaling from the NRI survey
to the entire land base, and is computed using a standard variance estimator for a two-stage sample design
(Sarndal et al. 1992). The two variance components are combined using 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 (see Annex 3.12 for further discussion). For 2021 to 2022, additional uncertainty is
propagated through a Monte Carlo analysis that is associated with the surrogate data method (see Box 6-3). Soil
organic carbon stock changes from the Tier 2 and 3 approaches are combined using the simple error propagation
method provided by the IPCC (2006). The combined uncertainty is calculated by taking the square root of the sum
of the squares of the standard deviations of the uncertain quantities.
The combined uncertainty for soil organic carbon stocks in cropland remaining cropland ranges from 212 percent
below to 212 percent above the 2022 stock change estimate of -31.7 MMT CO2 Eq. The large relative uncertainty
around the 2022 stock change estimate is mostly due to variation in soil organic carbon stock changes that is not
explained by the surrogate data method, leading to high prediction error.
Land Use, Land-Use Change, and Forestry 6-71
-------
Table 6-38: Approach 2 Quantitative Uncertainty Estimates for Soil Carbon Stock Changes
occurring within Cropland Remaining Cropland (MMT CO2 Eq. and Percent)
Source
2022 Flux Estimate
Uncertainty Range Relative to Flux Estimate3
(MMT C02 Eq.)
(MMT C02
Eq.)
(%)
Lower
Upper
Lower
Upper
Bound
Bound
Bound
Bound
Mineral Soil C Stocks: Cropland Remaining
Cropland, Tier 3 Inventory Methodology
(58.8)
(123.6)
5.9
-110%
+110%
Mineral Soil C Stocks: Cropland Remaining
Cropland, Tier 2 Inventory Methodology
(3.2)
(8.4)
2.0
-162%
+162%
Organic Soil C Stocks: Cropland Remaining
Cropland, Tier 2 Inventory Methodology
30.3
12.7
47.9
-58%
+58%
Combined Uncertainty for Flux associated with
Agricultural Soil Carbon Stock Change in
(31.7)
(99.0)
35.6
-212%
+212%
Cropland Remaining Cropland
a Range of C stock change estimates is a 95 percent confidence interval.
Notes: Totals may not sum due to independent rounding. Parentheses indicate net sequestration.
Uncertainty is also associated with lack of reporting of agricultural woody biomass and dead organic matter carbon
stock changes. However, woody biomass carbon stock changes are likely minor in perennial crops, with relatively
small amounts of woody crops such as orchards and nut plantations. There will be removal and replanting of tree
crops each year, but the net effect on biomass carbon stock changes is probably minor because the overall area
and tree density is relatively constant across time series. In contrast, agroforestry practices, such as shelterbelts,
riparian forests and intercropping with trees, may have more significant changes over the Inventory time series,
compared to perennial woody crops, at least in some regions of the United States, but there are currently no
datasets to evaluate the trends. Changes in litter carbon stocks are also assumed to be negligible in croplands over
annual time frames, although there are certainly significant changes at sub-annual time scales across seasons. This
trend may change in the future, particularly if crop residue becomes a viable feedstock for bioenergy production.
QA/QC and Verification
Quality control measures included checking input data, model scripts, and results to ensure data are properly
handled throughout the inventory process consistent with the U.S. Inventory QA/QC plan, which is in accordance
with Volume 1, Chapter 6 of the 2006IPCC Guidelines (see Annex 8 for more details). Inventory reporting forms
and text are reviewed and revised as needed to correct transcription errors. In addition, results from the DayCent
model are compared to field measurements and soil monitoring sites associated with the NRI (Spencer et al. 2011),
and a statistical relationship has been developed to assess uncertainties in the predictive capability of the model
(Ogle et al. 2007). The comparisons include 69 long-term experiment sites and 145 NRI soil monitoring network
sites, with 1406 observations across all of the sites (see Annex 3.12 for more information). Quality control
uncovered several errors in the Tier 2 method following the initial analysis, such as no estimation for some NRI
survey locations (i.e., federal lands and Hawaii), double counting some NRI survey locations with aggregation to
the national scale, and errors in the estimation of the two variance components associated with the uncertainty
analysis. The errors have been corrected following the diagnosis of the quality control issues.
Recalculations Discussion
Several improvements have been implemented in this Inventory leading to the need for recalculations. These
improvements included a) incorporating new USDA-NRCS NRI data through 2017; b) extending the time series for
crop histories through 2020 using USDA-NASS CDL data; c) incorporating USDA-NRCS CEAP survey data for 2013 to
2016; d) incorporating cover crop and tillage management information from the OpTIS remote-sensing data
product from 2008 to 2020; e) modifying the statistical imputation method for the management activity data
associated with about tillage practices, mineral fertilization, manure amendments, cover crop management, and
6-72 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
planting and harvest dates using gradient boosting instead of an artificial neural network; f) updating time series of
synthetic nitrogen fertilizer sales data, PRP nitrogen and manure nitrogen available for application to soils; g)
constraining synthetic nitrogen fertilization and manure nitrogen applications in the Tier 3 method at the state
scale rather than the national scale; h) re-calibrating the soil carbon module in the DayCent model using Bayesian
methods; and i) expanding the crops in the Tier 3 method to include dry beans, lentils, onions, peas and tomatoes,
which shifted some NRI survey locations from the Tier 2 to the Tier 3 method. The combined impact from these
improvements resulted in an average annual increase in soil C stocks of 4.2 MMT CO2 Eq., or 26 percent, from 1990
to 2021 relative to the previous Inventory.
Planned Improvements
A key improvement is conducting an analysis of carbon stock changes in Alaska for cropland. The improvement will
be conducted using the Tier 2 method for mineral and organic soils that is described earlier in this section. The
analysis will initially focus on land-use change, which typically has a larger impact on soil organic carbon stock
changes than management practices, but will be further refined over time to incorporate management data. The
improvement will resolve most of the differences between the managed land base for cropland remaining
cropland and amount of area currently included in cropland remaining cropland (see Table 6-39).
Table 6-39: Comparison of Managed Land Area in Cropland Remaining Cropland and Area in
the Current Cropland Remaining Cropland Inventory (Thousand Hectares)
Area (Thousand Hectares)
Year
Managed Land
Inventory
Difference
1990
162,273
162,247
26
1991
161,840
161,814
26
1992
161,343
161,317
26
1993
159,577
159,551
26
1994
157,890
157,864
26
1995
157,277
157,251
26
1996
156,639
156,613
26
1997
156,018
155,992
26
1998
152,335
152,309
26
1999
151,432
151,406
26
2000
151,257
151,231
26
2001
150,734
150,708
26
2002
150,426
150,400
26
2003
151,055
151,029
26
2004
150,787
150,761
26
2005
150,417
150,391
26
2006
149,908
149,882
26
2007
150,117
150,091
26
2008
149,718
149,692
26
2009
149,660
149,634
26
2010
149,222
149,196
26
2011
148,626
148,600
26
2012
148,297
148,271
26
2013
148,660
148,633
26
2014
149,141
149,115
26
2015
148,525
148,499
26
2016
148,436
148,410
26
Land Use, Land-Use Change, and Forestry 6-73
-------
2017
148,331
148,305
26
2018
149,720
149,694
26
2019
149,503
149,477
26
2020
149,822
149,796
26
2021
150,591
*
*
2022
151,276
*
*
Activity data on land use have not been incorporated into the Inventory
after 2020, designated with asterisks (*).
There are several other planned improvements underway related to the plant production module in DayCent. Crop
parameters associated with temperature effects on plant production will be further improved in DayCent with
additional model calibration. Senescence events following grain filling in crops, such as wheat, are being modified
based on recent model algorithm development, and will be incorporated. There will also be further testing and
parameterization of the DayCent model to reduce uncertainty, particularly the submodules that are used to
approximate the cycling of nitrogen through the plant-soil system, which will also have impacts on carbon cycling
in the model simulations.
Improvements are underway to simulate crop residue burning in the DayCent model based on the amount of crop
residues burned according to the data that are used in the field burning of agricultural residues source category
(see Section 5.7). This improvement will more accurately represent the carbon inputs to the soil that are
associated with residue burning. In addition, a review of available data on biosolids (i.e., treated sewage sludge)
application will be undertaken to improve the distribution of biosolids application on croplands, grasslands and
settlements.
Another improvement is to estimate biomass carbon stock changes in agroforestry systems and perennial tree
crops. Methods combining survey data and remote sensing imagery are under development to determine the
extent of land with agroforestry and perennial tree crops. In addition, a meta-analysis is being conducted to derive
country-specific factors for biomass C stock changes in agroforestry systems. Although the influence of perennial
tree crop biomass is expected to be minor, carbon stock changes may be significantly impacted by the effect of
agroforestry practices.
Many of these improvements are expected to be completed for the next (1990 through 2023) Inventory (i.e., 2025
submission), pending prioritization of resources.
6.5 Land Converted to Cropland (CRT
Category 4B2)
Land converted to cropland includes all current cropland in an inventory year that had been in another land use(s)
during the previous 20 years (IPCC 2006), and used to produce food or fiber, or forage that is harvested and used
as feed (e.g., hay and silage). For example, grassland or forest land converted to cropland during the past 20 years
would be reported in this category. Recently converted lands are retained in this category for 20 years as
recommended by IPCC (2006).
Land use change can lead to large losses of carbon to the atmosphere, particularly conversions from forest land
(Houghton et al. 1983; Houghton and Nassikas 2017). Moreover, conversion of forest to another land use (i.e.,
deforestation) is one of the largest anthropogenic sources of emissions to the atmosphere globally, although this
source may be declining (Tubiello et al. 2015).
6-74 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
The 2006IPCC Guidelines recommend reporting changes in biomass, dead organic matter and soil organic carbon
stocks with land use change. All soil organic carbon stock changes are estimated and reported for land converted
to cropland, but reporting of carbon stock changes for aboveground and belowground biomass, dead wood, and
litter pools is limited to forest land converted to cropland and grassland converted to cropland for woodland
conversions (i.e., woodland conversion to cropland).46
Grassland converted to cropland is the largest source of emissions from 1990 to 2000, while forest land converted
to cropland is the largest source of emissions from 2001 to 2022. This shift is largely due to reduced losses of
carbon from mineral soils after 2001. The high losses of carbon from forest land converted to cropland is due to
reductions in biomass and dead organic matter carbon following conversion from forests (Table 6-40 and Table
6-41). The net change in total carbon stocks for 2022 led to CO2 emissions to the atmosphere of 35.1 MMT CO2 Eq.
(9.6 MMT C), including 12.1 MMT CO2 Eq. (3.3 MMT C) from aboveground biomass carbon losses, 2.0 MMT CO2 Eq.
(0.6 MMT C) from belowground biomass carbon losses, 2.3 MMT CO2 Eq. (0.6 MMT C) from dead wood carbon
losses, 3.4 MMT CO2 Eq. (0.9 MMT C) from litter carbon losses, 12.6 MMT CO2 Eq. (3.4 MMT C) from mineral soils
and 2.7 MMT CO2 Eq. (0.7 MMT C) from drainage and cultivation of organic soils. The overall net loss of carbon has
declined by 23 percent from 1990 to 2022.
Table 6-40: Net CO2 Flux from Soil, Dead Organic Matter and Biomass Carbon Stock Changes
in Land Converted to Cropland by Land-Use Change Category (MMT CO2 Eq.)
1990
2005
2018
2019
2020
2021
2022
Grassland Converted to Cropland
27.3
17.2
13.7
13.0
10.6
16.1
16.3
Aboveground Live Biomass
0.1
al 1
0.1
0.1
0.1
0.1
0.1
Belowground Live Biomass
+
+
+
+
+
+
+
Dead Wood
+ 1
+ I
+
+
+
+
+
Litter
+
+
+
+
+
+
+
Mineral Soils
24.6 s
13.7
10.7
10.1
8.0
13.5
13.6
Organic Soils
2.4
3.3
2.7
2.7
2.4
2.4
2.4
Forest Land Converted to Cropland
19.2
19.2
19.7
19.7
19.7
19.6
19.6
Aboveground Live Biomass
11.4
11.6
11.9
11.9
11.9
11.9
11.9
Belowground Live Biomass
1.9
2.0
2.0
2.0
2.0
2.0
2.0
Dead Wood
2.2
2.2
2.2
2.2
2.2
2.2
2.2
Litter
3.2 if
3-3 1
3.3
3.4
3.4
3.4
3.4
Mineral Soils
0.4
0.2
0.1
0.1
0.1
0.1
0.1
Organic Soils
o-i
+
+
+
+
+
+
Other Lands Converted to Cropland
(1-8)
(2.5)
(1.6)
(1.6)
(1.2)
(1.1)
(1.1)
Mineral Soils
(1.9)
(2.6) 1
(1.7)
(1.6)
(1.2)
(1.1)
(1.1)
Organic Soils
0.1
0.1
+
+
+
+
+
Settlements Converted to Cropland
(0.0)
(0.1)
(0.1)
(0.1)
(0.1)
(0.1)
(0.1)
Mineral Soils
(0.1)
(0.1)
(0.2)
(0.2)
(0.2)
(0.2)
(0.2)
Organic Soils
!!!!:
+ is
+::::::
+
+
+
+
+
Wetlands Converted to Cropland
0.7
0.7
0.4
0.4
0.4
0.4
0.4
Mineral Soils
I—
-------
Litter
3.3
3.3
3.4
3.4
3.4
3.4
3.4
Total Mineral Soil Flux
23.2
11.3
9.2
8.6
6.9
12.5
12.6
Total Organic Soil Flux
3.2
3.9
3.0
3.0
2.6
2.6
2.7
Total Net Flux
45.4
34.5
31.9
31.4
29.3
34.9
35.1
+ Does not exceed 0.05 MMT C02 Eq.
Notes: Totals may not sum due to independent rounding. Parentheses indicate negative values or net
sequestration.
Table 6-41: Net CO2 Flux from Soil, Dead Organic Matter and Biomass Carbon Stock Changes
in Land Converted to Cropland (MMT C)
1990
2005
2018
2019
2020
2021
2022
Grassland Converted to Cropland
7.4
4.7
3.7
3.6
2.9
4.4
4.4
Aboveground Live Biomass
+ :
+
+
+
+
+
+
Belowground Live Biomass
+
+
+
+
+
+
+
Dead Wood
+ 1
+
+
+
+
+
+
Litter
+
+
+
+
+
+
+
Mineral Soils
6.7
3.7
2.9
2.8
2.2
3.7
3.7
Organic Soils
0.7
0.9
0.7
0.7
0.6
0.6
0.7
Forest Land Converted to Cropland
5.2
5.2
5.4
5.4
5.4
5.4
5.4
Aboveground Live Biomass
3.1
3.2
3.2
3.3
3.3
3.3
3.3
Belowground Live Biomass
0.5
0.5 -
0.5
0.6
0.6
0.6
0.6
Dead Wood
0.6
0.6
0.6
0.6
0.6
0.6
0.6
Litter
°-9 1
°-9
0.9
0.9
0.9
0.9
0.9
Mineral Soils
0.1
+
+
+
+
+
+
Organic Soils
+ •
+ :*
+
+
+
+
+
Other Lands Converted to Cropland
(0.5)
(0.7)
(0.4)
(0.4)
(0.3)
(0.3)
(0.3)
Mineral Soils
(0.5) it
(o.7);;
(0.5)
(0.4)
(0.3)
(0.3)
(0.3)
Organic Soils
+
+
+
+
+
+
+
Settlements Converted to Cropland
+
+
+
+
+
+
+
Mineral Soils
+
+
+
+
+
+
+
Organic Soils
+ :
+ si
+
+
+
+
+
Wetlands Converted to Cropland
0.2
0.2
0.1
0.1
0.1
0.1
0.1
Mineral Soils
o.i
0.1 1
+
+
+
+
+
Organic Soils
0.1
0.1
0.1
0.1
0.1
0.1
0.1
Aboveground Live Biomass
3.2
3.2
3.3
3.3
3.3
3.3
3.3
Belowground Live Biomass
0.5
0.5
0.6
0.6
0.6
0.6
0.6
Dead Wood
0.6
0.6
0.6
0.6
0.6
0.6
0.6
Litter
0.9
0.9
0.9
0.9
0.9
0.9
0.9
Total Mineral Soil Flux
6.3
3.1
2.5
2.4
1.9
3.4
3.4
Total Organic Soil Flux
0.9
1.1
0.8
0.8
0.7
0.7
0.7
Total Net Flux
12.4
9.4
8.7
8.6
8.0
9.5
9.6
+ Does not exceed 0.05 MMT C.
Notes: Totals may not sum due to independent rounding. Parentheses indicate negative values or net
sequestration.
Methodology and Time-Series Consistency
The following section includes a description of the methodology used to estimate carbon stock changes for land
converted to cropland, including (1) loss of aboveground and belowground biomass, dead wood and litter carbon
with conversion of forest lands to croplands, as well as (2) the impact from all land use conversions to cropland on
mineral and soil organic carbon stocks.
6-76 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
Biomass, Dead Wood and Litter Carbon Stock Changes
A Tier 2 method is applied to estimate biomass, dead wood, and litter carbon stock changes for forest land
converted to cropland. Estimates are calculated in the same way as those in the forest land remaining forest land
category using data from the USDA Forest Service, Forest Inventory and Analysis (FIA) program (USDA Forest
Service 2023). However, there is no country-specific data for cropland biomass, so only a default biomass estimate
(IPCC 2006) for croplands was used to estimate carbon stock changes (litter and dead wood carbon stocks were
assumed to be zero since no reference carbon density estimates exist for croplands). The difference between the
stocks is reported as the stock change under the assumption that the change occurred in the year of the
conversion. Details for each of the carbon attributes described below are available in Domke et al. (2022) and
Westfall et al. (2023). If FIA plots include data on individual trees, aboveground and belowground carbon density
estimates are based on Woodall et al. (2011) and Westfall et al. (2023). Aboveground and belowground biomass
estimates also include live understory which is a minor component of biomass defined as all biomass of
undergrowth plants in a forest, including woody shrubs and trees less than 2.54 cm dbh. For this Inventory, it was
assumed that 10 percent of total understory carbon mass is belowground (Smith et al. 2006). Estimates of carbon
density are based on information in Birdsey (1996) and biomass estimates from Jenkins et al. (2003).
For dead organic matter, if FIA plots include data on standing dead trees, standing dead tree carbon density is
estimated following the basic method applied to live trees (Woodall et al. 2011; Westfall et al. 2023) with
additional modifications for woodland species to account for decay and structural loss (Domke et al. 2011; Harmon
et al. 2011). If FIA plots include data on downed dead wood, downed dead wood carbon density is estimated based
on measurements of a subset of FIA plots for downed dead wood (Domke et al. 2013; Woodall and Monleon
2008). Downed dead wood is defined as pieces of dead wood greater than 7.5 cm diameter, at transect
intersection, that are not attached to live or standing dead trees. This includes stumps and roots of harvested
trees. To facilitate the downscaling of downed dead wood carbon estimates from the state-wide population
estimates to individual plots, downed dead wood models specific to regions and forest types within each region
are used. Litter carbon is the pool of organic carbon (also known as duff, humus, and fine woody debris) above the
mineral soil and includes woody fragments with diameters of up to 7.5 cm. A subset of FIA plots are measured for
litter carbon. If FIA plots include litter material, a modeling approach using litter carbon measurements from FIA
plots is used to estimate litter carbon density (Domke et al. 2016). See Annex 3.13 for more information about
reference carbon density estimates for forest land and the compilation system used to estimate carbon stock
changes from forest land.
Soil Carbon Stock Changes
Soil organic stock changes are estimated for land converted to cropland according to land use histories recorded in
the 2017 USDA NRI survey for non-federal lands (USDA-NRCS 2020). Land use and some management information
(e.g., crop type, soil attributes, and irrigation) had been collected for each NRI point on a five-year cycle beginning
in 1982. In 1998, the NRI program began collecting annual data, which are currently available through 2017 (USDA-
NRCS 2020), and the time series for cropping histories was extended through 2020 using the USDA-NASS Crop Data
Layer Product (CDL) (USDA-NASS 2021) and National Land Cover Dataset (NLCD) (Yang et al. 2018; Fry et al. 2011;
Homer et al. 2007, 2015). The areas have been modified in the original NRI survey through a process in which the
Forest Inventory and Analysis (FIA) survey data and the National Land Cover Dataset (NLCD; Yang et al. 2018) are
harmonized with the NRI data. This process ensures that the land use areas are consistent across all land use
categories (see Section 6.1 for more information).
NRI survey locations are classified as land converted to cropland in a given year between 1990 and 2020 if the land
use is cropland but had been another use during the previous 20 years. 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 1998, 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 from 1971 to 1978. For federal
lands, the land use history is derived from land cover changes in the NLCD (Yang et al. 2018; Homer et al. 2007; Fry
et al. 2011; Homer et al. 2015).
Land Use, Land-Use Change, and Forestry 6-77
-------
Soil Carbon Stock Changes for Mineral Soils
An IPCC Tier 3 model-based approach using the DayCent ecosystem model (Ogle et al. 2010, 2023) is applied to
estimate carbon stock changes from 1990 to 2020 for mineral soils on the majority of land that is used to produce
annual crops and forage crops that are harvested and used as feed (e.g., hay and silage) in the United States. These
crops include alfalfa hay, barley, corn, cotton, dry beans, grass hay, grass-clover hay, lentils, oats, onions, peanuts,
peas, potatoes, rice, sorghum, soybeans, sugar beets, sunflowers, tobacco, tomatoes, and wheat. Soil organic
carbon stock changes on the remaining mineral soils are estimated with the IPCC Tier 2 method (Ogle et al. 2003,
2006), including land used to produce some vegetables and perennial/horticultural crops and crops rotated with
these crops; land on very gravelly, cobbly, or shaley soils (greater than 35 percent by volume); and land converted
from another land use or federal ownership.47
For the years 2021 to 2022, a surrogate data method is used to estimate soil organic carbon stock changes at the
national scale for land areas included in the Tier 2 and Tier 3 methods. Specifically, linear regression models with
autoregressive moving-average (ARMA) errors (Brockwell and Davis 2016) are used to estimate the relationship
between surrogate data and the 1990 to 2020 stock change data from the Tier 2 and 3 methods. Surrogate data
for these regression models include corn and soybean yields from USDA-NASS statistics,48 and weather data from
the PRISM Climate Group (PRISM 2022). See Box 6-4 in the Methodology section of Cropland Remaining Cropland
for more information about the surrogate data method. Stock change estimates for 2021 to 2022 will be
recalculated in future Inventories when the time series of activity data are updated.
Tier 3 Approach. For the Tier 3 method, mineral soil organic carbon stocks and stock changes are estimated using
the DayCent ecosystem model (Parton et al. 1998; Del Grosso et al. 2001, 2011). The DayCent model utilizes the
soil carbon modeling framework developed in the Century model (Parton et al. 1987,1988,1994; Metherell et al.
1993), but has been refined to simulate dynamics at a daily time-step. National estimates are obtained by using
the model to simulate historical land use change patterns as recorded in the USDA NRI survey (USDA-NRCS 2020).
Carbon stocks and 95 percent confidence intervals are estimated for each year between 1990 and 2020. See the
cropland remaining cropland section and Annex 3.12 for additional discussion of the Tier 3 methodology for
mineral soils.
In order to ensure time-series consistency, the Tier 3 method is applied from 1990 to 2020 so that changes reflect
anthropogenic activity and not methodological adjustments. Soil organic carbon stock changes from 2021 to 2022
are approximated using a linear extrapolation of emission patterns from 1990 to 2020. The extrapolation is based
on a linear regression model with moving-average (ARMA) errors (described in Box 6-4 of the Methodology section
in cropland remaining cropland). Linear extrapolation is a standard data splicing method for estimating emissions
at the end of a time series (IPCC 2006). Time series of activity data will be updated in a future Inventory, and
emissions from 2020 to 2022 will be recalculated.
Tier 2 Approach. For the mineral soils not included in the Tier 3 analysis, soil organic carbon stock changes are
estimated using a Tier 2 Approach, as described in the Tier 2 Approach for mineral soils in cropland remaining
cropland. In order to ensure time-series consistency, the Tier 2 method is applied from 1990 to 2020 so that
changes reflect anthropogenic activity and not methodological adjustments. In addition, soil organic carbon stock
changes are approximated for the remainder of the 2021 to 2022 time series with a linear extrapolation of
emission patterns from 1990 to 2020. The extrapolation is based on a linear regression model with moving-
average (ARMA) errors (see Box 6-4 of the Methodology section in Cropland Remaining Cropland). Linear
extrapolation is a standard data splicing method for estimating emissions at the end of a time series (IPCC 2006).
As with the Tier 3 method, time series of activity data will be updated in a future Inventory, and emissions from
2021 to 2022 will be recalculated.
47 Federal 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 2018).
48 See https://auickstats.nass.usda.gov/.
6-78 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
Soil Carbon Stock Changes for Organic Soils
Annual carbon emissions from drained organic soils in land converted to cropland are estimated using the Tier 2
method provided in IPCC (2006), with country-specific carbon loss rates (Ogle et al. 2003) as described in the
cropland remaining cropland section for organic soils. Further elaboration on the methodology is also provided in
Annex 3.12.
In order to ensure time-series consistency, the Tier 2 methods are applied from 1990 to 2020 so that changes
reflect anthropogenic activity and not methodological adjustments. In addition, soil organic carbon stock changes
for the remainder of the time series (i.e., 2021 to 2022) are approximated with a linear extrapolation of emission
patterns from 1990 to 2020. The extrapolation is based on a linear regression model with moving-average (ARMA)
errors (see Box 6-4 of the Methodology section in cropland remaining cropland). Linear extrapolation is a standard
data splicing method for approximating emissions at the end of a time series (IPCC 2006). Estimates for 2021 to
2022 will be recalculated in a future Inventory when new activity data are incorporated into the analysis.
Uncertainty
The uncertainty analyses for biomass, dead wood and litter carbon losses with forest land converted to cropland
and grassland converted to cropland for woodland conversions are conducted in the same way as the uncertainty
assessment for forest ecosystem carbon flux associated with forest land remaining forest land. Sample and model-
based error are combined using simple error propagation methods provided by the IPCC (2006) by taking the
square root of the sum of the squares of the standard deviations of the uncertain quantities. For additional details,
see the Uncertainty Analysis in Annex 3.13.
The uncertainty analyses for soil organic carbon stock changes using the Tier 3 and Tier 2 methodologies are
quantified from two variance components (Ogle et al. 2010), as described in cropland remaining cropland. For
2021 to 2022, there is additional uncertainty propagated through the Monte Carlo analysis associated with the
surrogate data method, which is also described in cropland remaining cropland.
Uncertainty estimates are presented in Table 6-42 for each sub-source (i.e., biomass carbon stocks, dead wood
carbon stocks, litter carbon stocks, soil organic carbon stocks for mineral and organic soils) and the method applied
in the Inventory analysis (i.e., Tier 2 and Tier 3). Uncertainty estimates for the total carbon stock changes for
biomass, dead organic matter and soils are combined using the simple error propagation methods provided by the
IPCC (2006). The combined uncertainty for total carbon stock changes in land converted to cropland ranged from
93 percent below to 93 percent above the 2022 stock change estimate of 35.1 MMT CO2 Eq. The large relative
uncertainty in the 2022 estimate is mostly due to variation in soil organic carbon stock changes that is not
explained by the surrogate data method, leading to high prediction error with this splicing method.
Table 6-42: Approach 2 Quantitative Uncertainty Estimates for Soil, Dead Organic Matter and
Biomass Carbon Stock Changes occurring within Land Converted to Cropland (MMT CO2 Eq.
and Percent)
2022 Flux Estimate Uncertainty Range Relative to Flux Estimate-'
(MMTCO. Eq.) (MMTCO. Eq.) (%)
Lower
Bound
Upper
Bound
Lower
Bound
Upper
Bound
Grassland Converted to Cropland
16.3
(12.1)
44.7
-174%
174%
Aboveground Live Biomass
0.1
+
0.3
-128%
124%
Belowground Live Biomass
+
+
+
-100%
56%
Dead Wood
+
+
0.1
-100%
173%
Litter
+
+
0.1
-100%
147%
Mineral Soil C Stocks: Tier 3
11.0
(17.2)
39.3
-256%
256%
Mineral Soil C Stocks: Tier 2
2.6
0.6
4.5
-77%
77%
Land Use, Land-Use Change, and Forestry 6-79
-------
Organic Soil C Stocks: Tier 2
2.4
0.5
4.4
-79%
79%
Forest Land Converted to Cropland
19.6
3.4
35.8
-82%
82%
Aboveground Live Biomass
11.9
(3.2)
27.1
-127%
127%
Belowground Live Biomass
2.0
(0.6)
4.6
-127%
127%
Dead Wood
2.2
(0.6)
5.1
-128%
127%
Litter
3.4
(0.9)
7.6
-127%
128%
Mineral Soil C Stocks: Tier 2
0.1
+
0.2
-107%
107%
Organic Soil C Stocks: Tier 2
+
+
0.1
-429%
429%
Other Lands Converted to Cropland
(1.1)
(2.1)
+
-99%
99%
Mineral Soil C Stocks: Tier 2
(1.1)
(2.1)
+
-99%
99%
Organic Soil C Stocks: Tier 2
+
+
+
0%
0%
Settlements Converted to Cropland
(0.1)
(0.3)
+
-97%
97%
Mineral Soil C Stocks: Tier 2
(0.2)
(0.3)
(0.1)
-71%
71%
Organic Soil C Stocks: Tier 2
+
+
0.1
-103%
103%
Wetlands Converted to Croplands
0.4
(0.1)
0.8
-115%
115%
Mineral Soil C Stocks: Tier 2
0.2
+
0.3
-124%
124%
Organic Soil C Stocks: Tier 2
0.2
(0.2)
0.6
-173%
173%
Total: Land Converted to Cropland
35.1
2.4
67.8
-93%
93%
Aboveground Live Biomass
12.1
(3.1)
27.2
-126%
126%
Belowground Live Biomass
2.0
(0.5)
4.6
-126%
126%
Dead Wood
2.3
(0.6)
5.1
-126%
125%
Litter
3.4
(0.9)
7.7
-126%
126%
Mineral Soil C Stocks: Tier 3
11.0
(17.2)
39.3
-256%
256%
Mineral Soil C Stocks: Tier 2
1.6
(0.7)
3.8
-144%
144%
Organic Soil C Stocks: Tier 2
2.7
0.7
4.7
-73%
73%
+ Does not exceed 0.05 MMT C02 Eq.
a Range of C stock change estimates is a 95 percent confidence interval.
Notes: Totals may not sum due to independent rounding. Parentheses indicate negative values or net sequestration.
Uncertainty is also associated with lack of reporting of agricultural biomass and dead organic matter carbon stock
changes. Biomass carbon stock changes are likely minor in perennial crops, such as orchards and nut plantations,
given the small amount of change in land that is 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
larger changes in biomass carbon stocks at least in some regions of the United States. However, there are currently
no datasets to evaluate the trends. Changes in dead organic matter carbon stocks are assumed to be negligible
with conversion of land to croplands with the exception of forest lands, which are included in this analysis. This
assumption will be further explored in a future Inventory.
/erification
See the QA/QC and Verification section in Cropland Remaining Cropland for information on QA/QC steps.
Recalculations Discussion
Several improvements have been implemented in this Inventory leading to the need for recalculations. These
improvements included a) incorporating new USDA-NRCS NRI data through 2017; b) extending the time series for
crop histories through 2020 using USDA-NASS CDL data; c) incorporating USDA-NRCS CEAP survey data for 2013 to
2016; d) incorporating cover crop and tillage management information from the OpTIS remote-sensing data
product from 2008 to 2020; e) modifying the statistical imputation method for the management activity data
associated with about tillage practices, mineral fertilization, manure amendments, cover crop management,
planting and harvest dates using gradient boosting instead of an artificial neural network; f) updating time series of
synthetic nitrogen fertilizer sales data, PRP nitrogen and manure nitrogen available for application to soils; g)
6-80 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
constraining synthetic nitrogen fertilization and manure nitrogen applications in the Tier 3 method at the state
scale rather than the national scale; h) re-calibrating the soil carbon module in the DayCent model using Bayesian
methods; i) expanding the crops in the Tier 3 method to include dry beans, lentils, onions, peas and tomatoes,
which shifted some NRI survey locations from the Tier 2 to the Tier 3 method, and j) updated FIA data from 1990
to 2022 on biomass, dead wood and litter carbon stocks associated with forest land converted to cropland. Finally,
see further updates in Section 6.2, describing updates to the estimates for aboveground volume and biomass
which impacted lands converted to cropland estimates. As a result, land converted to cropland has an estimated
smaller carbon loss of 20.7 MMT CO2 Eq. on average over the time series. This represents a 37 percent average
decrease in carbon stock change losses for land converted to cropland compared to the previous Inventory, and is
mainly due to less loss of carbon associated with forest land converted to cropland.
Planned Improvements
A key improvement is to estimate the biomass carbon stock changes for other land use changes beyond only forest
land converted to cropland and grassland converted to cropland for woodland conversion, which is included in the
current Inventory. Additional planned improvements are discussed in the Planned Improvements section of
Cropland Remaining Cropland.
6.6 Grassland Remaining Grassland (CRT
Category 4C1)
Carbon in grassland ecosystems occurs in biomass, dead organic matter, and soils. Soils are the largest pool of
carbon in grasslands, and have the greatest potential for longer-term storage or release of carbon. Biomass and
dead organic matter carbon pools are relatively ephemeral compared to the soil carbon pool, with the exception of
carbon stored in tree and shrub biomass that occurs in grasslands. The 2006IPCC Guidelines recommend reporting
changes in biomass, dead organic matter and soil organic carbon stocks with land use and management. Carbon
stock changes for aboveground and belowground biomass, dead wood and litter pools are reported for woodlands
(i.e., a subcategory of grasslands49), and may be extended to include agroforestry management associated with
grasslands in the future. For soil organic carbon, the 2006 IPCC Guidelines (IPCC 2006) recommend reporting
changes due to (1) agricultural land use and management activities on mineral soils, and (2) agricultural land use
and management activities on organic soils.50
Grassland remaining grassland includes all grassland in an inventory year that had been grassland for a continuous
time period of at least 20 years (USDA-NRCS 2018). Grassland includes pasture and rangeland that are primarily,
but not exclusively used for livestock grazing. Rangelands are typically extensive areas of native grassland that are
not intensively managed, while pastures are typically seeded grassland (possibly following tree removal) that may
also have additional management, such as irrigation or inter-seeding of legumes. Woodlands are also considered
grassland and are areas of continuous tree cover that do not meet the definition of forest land (see Section 6.1 for
more information about the criteria for forest land).
There is a discrepancy between the current land representation (see Section 6.1) and the area data that have been
used in the inventory for grassland remaining grassland. Specifically, grasslands in Alaska are not included in the
Inventory, and this land base is approximately 50 million hectares. This difference leads to a discrepancy between
49 Woodlands are considered grasslands in the U.S. land representation because they do not meet the definition of forest land.
50 C02 emissions associated with liming and urea fertilization are also estimated but included in the Agriculture chapter of the
report.
Land Use, Land-Use Change, and Forestry 6-81
-------
the managed area in grassland remaining grassland in the land representation and the grassland area included in
the emissions and removals estimation for the grassland remaining grassland land-use category (Table 6-46).
Improvements are underway to incorporate grasslands in Alaska as part of future Inventories (see Planned
Improvements section).
For grassland remaining grassland, there has been considerable variation in carbon stocks between 1990 and 2022.
These changes are driven by variability in weather patterns and associated interaction with land management
activity. Moreover, changes are small on a per hectare rate basis across the time series even in the years with a
larger total change in stocks. The net change in total carbon stocks for 2022 led to net CO2 emissions to the
atmosphere of 13.4 MMT CO2 Eq. (3.6 MMT C), including -1.3 MMT CO2 Eq. (-0.4 MMT C) from net gains of
aboveground biomass C, -0.2 MMT CO2 Eq. (-0.1 MMT C) from net gains in belowground biomass carbon, 2.8 MMT
CO2 Eq. (0.8 MMT C) from net losses in dead wood carbon, less than 0.05 MMT CO2 Eq. (less than 0.05 MMT C)
from net gains in litter C, 6.5 MMT CO2 Eq. (1.8 MMT C) from net losses in mineral soil organic carbon, and 5.5
MMT CO2 Eq. (1.5 MMT C) from losses of carbon due to drainage and cultivation of organic soils (Table 6-43 and
Table 6-44). Losses of carbon are 45.2 percent lower in 2022 compared to 1990, but as noted previously, stock
changes are highly variable from 1990 to 2022, with an average annual change of 19.9 MMT CO2 Eq. (5.4 MMT C).
Table 6-43: Net CO2 Flux from Soil, Dead Organic Matter and Biomass Carbon Stock Changes
in Grassland Remaining Grassland (MMT CO2 Eq.)
1990
2005
2018
2019
2020
2021
2022
Aboveground Live Biomass
(2.7) |
(2'1)I
(1.4)
(1.4)
(1.4)
(1.4)
(1.3)
Belowground Live Biomass
(°'4) 1
(°.3) I
(0.3)
(0.2)
(0.2)
(0.2)
(0.2)
Dead Wood
3.2
3.11
2.9
2.9
2.9
2.9
2.8
Litter
(0'4) 1
(0.2) I
+
+
+
+
+
Mineral Soils
18.6
18.6 J
22.0
22.0
9.3
3.8
6.5
Organic Soils
6.1
5-11
5.3
5.3
5.5
5.5
5.5
Total Net Flux
24.4
24.1
28.6
28.5
16.1
10.6
13.4
+ Does not exceed 0.05 MMT C02 Eq.
Notes: Totals may not sum due to independent rounding. Parentheses indicate negative values or net
sequestration.
Table 6-44: Net CO2 Flux from Soil, Dead Organic Matter and Biomass Carbon Stock Changes
in Grassland Remaining Grassland (MMT C)
1990
2005
2018
2019
2020
2021
2022
Aboveground Live Biomass
(0.7)
(0.6) 1
(0.4)
(0.4)
(0.4)
(0.4)
(0.4)
Belowground Live Biomass
(o.i)
(0.1) i
(0.1)
(0.1)
(0.1)
(0.1)
(0.1)
Dead Wood
0.9
0.8
0.8
0.8
0.8
0.8
0.8
Litter
(0.1)
(o.i)
+
+
+
+
+
Mineral Soils
5.1
5.1
6.0
6.0
2.5
1.0
1.8
Organic Soils
—at
r-.
vi
1.4 I
1.4
1.4
1.5
1.5
1.5
Total Net Flux
6.6 ill!
6.6
7.8
7.8
4.4
2.9
3.6
+ Does not exceed 0.05 MMT C.
Notes: Totals may not sum due to independent rounding. Parentheses indicate negative values or net
sequestration.
The spatial variability in soil organic carbon stock changes for 20 2051 is displayed in Figure 6-8 for mineral soils and
in Figure 6-9 for organic soils. Although relatively small on a per-hectare basis, grassland soils gained carbon in
51 Only national-scale emissions are estimated for 2021 to 2022 in the current Inventory using the surrogate data method, and
therefore the fine-scale emission patterns in this map are based on land use data from 2020.
6-82 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
isolated areas that mostly occurred in pastures of the upper Midwest and eastern United States; losses occurred
primarily in the northwestern region. For organic soils, the regions with the highest rates of emissions coincide
with the largest concentrations of organic soils that occur in managed grassland, including the Southeastern
Coastal Region (particularly Florida), areas surrounding the Great Lakes in the upper Midwest and Northeast, and a
few isolated areas along the Pacific Coast.
Figure 6-8: Total Net Annual Soil Carbon Stock Changes for Mineral Soils under Agricultural
Management within States, 2020, Grassland Remaining Grassland
'.?? <
I 4
PL— ,
* v • " * X
I
r-
t
/ V
w
• Av:
• -v
Vv. - *
illy
/.
I
V.*;
> :vt •'
«v—_ ,i-.
'..'r ¦ ' :
pi v h:
¦ Vt . .*
f
: ; :
' 1. ~
~T«- '•
v fir
X' "v
1 -cl. •
/
p -
¦" « ' *'•-
'V. r. 4
* ¦ r —
t-V:
¦ MtdrSHVttar f-i JBL
. *
ggil18^
. \ • % '
\
•/
S_
i ¦
MT C02 ha1 yr1
¦ < -4 ~ 1 to 2
¦ -4 to -2 ~ 2 to 4
~ -2 to -1 ¦ > 4
~ -1 to 1
\ . • •: y
\ • ' "v
^1 ¦ Us- <• ';4. u^afSssr - ¦ r-i
frty-ifwy.i/il: : ' -v",i
M «-W
V
\
-t r.—--- u|v
• 'i\
* i
V
. ..J"
Note: Only national-scale soil organic carbon stock changes are estimated for 2021 to 2022 in the current Inventory
using a surrogate data method, and therefore the fine-scale emission patterns in this map are based on inventory
data from 2020. Negative values represent a net increase in soil organic carbon stocks, and positive values represent a
net decrease in soil organic carbon stocks.
Land Use, Land-Use Change, and Forestry 6-83
-------
Figure 6-9: Total Net Annual Soil Carbon Stock Changes for Organic Soils under Agricultural
Management within States, 2020, Grassland Remaining Grassland
MT C02 ha1 yr1
~ < 10
~ 10 to 20
¦ 20 to 30
¦ 30 to 40
¦ > 40
Note: Only national-scale soil organic carbon stock changes are estimated for 2021 to 2022 in the current Inventory
using a surrogate data method, and therefore the fine-scale emission patterns in this map are based on inventory
data from 2020.
Methodology and Time-Series Consistency
The following section includes a description of the methodology used to estimate carbon stock changes for
grassland remaining grassland, including (1) aboveground and belowground biomass, dead wood and litter carbon
for woodlands, as well as (2) soil organic carbon stocks for mineral and organic soils.
Biomass, Dead Wood and Litter Carbon Stock Changes
Woodlands are lands that do not meet the definition of forest land or agroforestry (see Section 6.1), but include
woody vegetation with carbon storage in aboveground and belowground biomass, dead wood and litter carbon
(IPCC 2006) as described in the Forest Land Remaining Forest Land section. Carbon stocks and net annual carbon
stock change were determined according to the stock-difference method for the conterminous United States,
which involved applying carbon estimation factors to annual forest inventories across time to obtain carbon stocks
and then subtracting the values between years to estimate the stock changes. The methods for estimating carbon
stocks and stock changes for woodlands in grassland remaining grassland are consistent with those in the forest
6-84 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
land remaining forest land section and are described in Annex 3.13. All annual National Forest Inventory (NFI) plots
available in the public FIA database (USDA Forest Service 2023) were used in the current Inventory. While the NFI is
an all-lands inventory, only those plots that meet the definition of forest land are typically measured. However, in
some cases, particularly in the Central Plains and Southwest United States, woodlands have been measured as part
of the survey. This analysis is limited to those plots and is not considered a comprehensive assessment of trees
outside of forest land that meet the definition of grassland. The same methods are applied from 1990 to 2022 in
order to ensure time-series consistency. This methodology is consistent with IPCC guidance (2006).
Soil Carbon Stock Changes
The following section includes a brief description of the methodology used to estimate changes in soil organic
carbon stocks for grassland remaining grassland, including: (1) agricultural land use and management activities on
mineral soils; and (2) agricultural land use and management activities on organic soils. Further elaboration on the
methodologies and data used to estimate stock changes from mineral and organic soils is provided in the Cropland
Remaining Cropland section and Annex 3.12.
Soil organic carbon stock changes are estimated for grassland remaining grassland on non-federal lands according
to land use histories recorded in the USDA National Resources Inventory (NRI) (USDA-NRCS 2020). Land use and
some management information (e.g., grass type, soil attributes, and irrigation) were originally collected for each
NRI survey location on a five-year cycle beginning in 1982. In 1998, the NRI program began collecting annual data,
and the annual data are currently available through 2017 (USDA-NRCS 2020). For 2018-2020, the time series is
extended with the data provided in the National Land Cover Dataset (NLCD) (Yang et al. 2018; Fry et al. 2011;
Homer et al. 2007, 2015). The areas have been modified in the original NRI survey through a process in which the
Forest Inventory and Analysis (FIA) survey data and the NLCD are harmonized with the NRI data. This process
ensures that the land use areas are consistent across all land use categories (see Section 6.1 for more information).
NRI survey locations are classified as grassland remaining grassland in a given year between 1990 and 2020 if the
land use had been grassland for 20 years. 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 1998. 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 between 1971 and 1978. For federal lands, the land use history is derived from
land cover changes in the NLCD (Yang et al. 2018; Homer et al. 2007; Fry et al. 2011; Homer et al. 2015).
Soil Carbon Stock Changes for Mineral Soils
An IPCC Tier 3 model-based approach (Ogle et al. 2010) is applied to estimate carbon stock changes from 1990 to
2020 for most mineral soils in grassland remaining grassland. The carbon stock changes for the remaining soils are
estimated with an IPCC Tier 2 method (Ogle et al. 2003), including gravelly, cobbly, or shaley soils (greater than 35
percent by volume), as well as additional stock changes associated with biosolids (i.e., treated sewage sludge)
amendments and federal land.52
A surrogate data method is used to estimate soil organic carbon stock changes from 2021 to 2022 at the national
scale for land areas included in the Tier 2 and Tier 3 methods. Specifically, linear regression models with
autoregressive moving-average (ARMA) errors (Brockwell and Davis 2016) are used to estimate the relationship
between surrogate data and the 1990 to 2020 emissions data from the Tier 2 and 3 methods. Surrogate data for
these regression models are based on weather data from the PRISM Climate Group (PRISM Climate Group 2022).
See Box 6-4 in the Methodology section of Cropland Remaining Cropland for more information about the surrogate
data method.
52 Federal 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 2020).
Land Use, Land-Use Change, and Forestry 6-85
-------
Tier 3 Approach. Mineral soil organic carbon stocks and stock changes for grassland remaining grassland are
estimated using the DayCent ecosystem model (Parton et al. 1998; Del Grosso et al. 2001, 2011), as described in
Cropland Remaining Cropland. The DayCent model utilizes the soil carbon modeling framework developed in the
Century model (Parton et al. 1987,1988,1994; Metherell et al. 1993), but has been refined to simulate dynamics
at a daily time-step. Historical land-use patterns and irrigation histories are simulated with DayCent based on the
2017 USDA NRI survey (USDA-NRCS 2020). The amount of manure produced by each livestock type is calculated for
managed and unmanaged waste management systems based on methods described in Section 5.2 and Annex 3.11.
Manure nitrogen deposition from grazing animals (i.e., pasture/range/paddock (PRP) manure) is an input to the
DayCent model to estimate the influence of PRP manure on carbon stock changes for lands included in the Tier 3
method. Carbon stocks and 95 percent confidence intervals are estimated for each year between 1990 and 2020
using the NRI survey data. Further elaboration on the Tier 3 methodology and data used to estimate carbon stock
changes from mineral soils are described in Annex 3.12.
In order to ensure time-series consistency, the same methods are applied from 1990 to 2020 so that changes
reflect anthropogenic activity and not methodological adjustments. In addition, soil organic carbon stock changes
from 2021 to 2022 are approximated using a linear extrapolation of emission patterns from 1990 to 2020. The
extrapolation is based on a linear regression model with moving-average (ARMA) errors, described in Box 6-4 of
the Methodology section in Cropland Remaining Cropland. Linear extrapolation is a standard data splicing method
for estimating emissions at the end of a time series (IPCC 2006). Stock change estimates for 2021 to 2022 will be
recalculated in future Inventories with an updated time series of activity data (see the Planned Improvements
section in Cropland Remaining Cropland).
Tier 2 Approach. The Tier 2 approach is based on the same methods described in the Tier 2 portion of the
Cropland Remaining Cropland section for mineral soils, with the exception of the manure nitrogen deposition from
grazing animals (i.e., PRP manure), and the land use and management data that are used in the Inventory for
federal grasslands. First, the PRP nitrogen manure is included in the Tier 2 method that is not deposited on lands
included in the Tier 3 method. Second, the NRI (USDA-NRCS 2020) provides land use and management histories for
all non-federal lands, and is the basis for the Tier 2 analysis for these areas. However, NRI does not provide land
use information on federal lands. The land use data for federal lands is based on the NLCD (Yang et al. 2018; Fry et
al. 2011; Homer et al. 2007; Homer et al. 2015). In addition, the Bureau of Land Management (BLM) manages
some of the federal grasslands, and compiles information on grassland condition through the BLM Rangeland
Inventory (BLM 2014). To estimate soil organic carbon stock changes from federal grasslands, rangeland conditions
in the BLM data are aligned with IPCC grassland management categories of nominal, moderately degraded, and
severely degraded in order to apply the appropriate emission factors. Further elaboration on the Tier 2
methodology and data used to estimate carbon stock changes from mineral soils are described in Annex 3.12.
In order to ensure time-series consistency, the Tier 2 method is applied from 1990 to 2020 so that changes reflect
anthropogenic activity and not methodological adjustments. In addition, soil organic carbon stock changes are
approximated for the remainder of the time series with a linear extrapolation of emission patterns from 1990 to
2020. The extrapolation is based on a linear regression model with moving-average (ARMA) errors (see Box 6-4 of
the Methodology section in Cropland Remaining Cropland). Linear extrapolation is a standard data splicing method
for estimating emissions at the end of a time series (IPCC 2006). As with the Tier 3 method, time series of activity
data will be updated in a future Inventory, and emissions from 2021 to 2022 will be recalculated.
Additional Mineral Carbon Stock Change Calculations
A Tier 2 method is used to adjust annual carbon stock change estimates for mineral soils between 1990 and 2022
to account for additional carbon stock changes associated with biosolids (i.e., treated sewage sludge)
amendments. Estimates of the amounts of biosolids nitrogen applied to agricultural land are derived from national
data on biosolids generation, disposition, and nitrogen content (see Section 7.2 for a detailed discussion of the
methodology for estimating treated sewage sludge available for land application application). Although biosolids
can be added to land managed for other land uses, it is assumed that agricultural amendments only occur in
grassland remaining grassland. Total biosolids generation data for 1988,1996, and 1998, in dry mass units, are
6-86 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
obtained from EPA (1999) and estimates for 2004 are obtained from an independent national biosolids survey
(NEBRA 2007). These values are linearly interpolated to estimate values for the intervening years, and linearly
extrapolated to estimate values for years since 2004. Nitrogen application rates from Kellogg et al. (2000) are used
to determine the amount of area receiving biosolids amendments. The soil organic carbon storage rate is
estimated at 0.38 metric tons carbon per hectare per year for biosolids amendments to grassland as described
above. The stock change rate is based on country-specific factors and the IPCC default method (see Annex 3.12 for
further discussion).
Soil Carbon Stock Changes for Organic Soils
Annual carbon emissions from drained organic soils in grassland remaining grassland are estimated using the Tier 2
method in IPCC (2006), which utilizes country-specific carbon loss rates (Ogle et al. 2003) rather than default IPCC
rates. For more information, see the cropland remaining cropland section for organic soils and Annex 3.12.
In order to ensure time-series consistency, the Tier 2 methods are applied from 1990 to 2020 so that changes
reflect anthropogenic activity and not methodological adjustments. In addition, soil organic carbon stock changes
for the remainder of the time series (i.e., 2021 to 2022) are approximated with a linear extrapolation of emission
patterns from 1990 to 2020. The extrapolation is based on a linear regression model with moving-average (ARMA)
errors (see Box 6-4 of the Methodology section in cropland remaining cropland). Linear extrapolation is a standard
data splicing method for approximating emissions at the end of a time series (IPCC 2006). Estimates for 2021 to
2022 will be recalculated in future Inventories with an updated time series of activity data.
Uncertainty
The uncertainty analysis for biomass, dead wood and litter carbon losses with woodlands is conducted in the same
way as the uncertainty assessment for forest ecosystem carbon flux associated with forest land remaining forest
land. Sample and model-based error are combined using simple error propagation methods provided by the IPCC
(2006) by taking the square root of the sum of the squares of the standard deviations of the uncertain quantities.
For additional details, see the Uncertainty Analysis in Annex 3.13.
Uncertainty analysis for soil organic carbon stock changes using the Tier 3 and Tier 2 methodologies are quantified
from two variance components (Ogle et al. 2010), as described in Cropland Remaining Cropland. For 2021 to 2022,
there is additional uncertainty propagated through the Monte Carlo analysis associated with the surrogate data
method.
Uncertainty estimates are presented in Table 6-45 for each subcategory (i.e., soil organic carbon stocks for mineral
and organic soils) 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 total carbon stock changes in grassland remaining grassland ranges from
more than 926 percent below and above the 2022 stock change estimate of 13.4 MMT CO2 Eq. The large relative
uncertainty in the 2022 estimate is mostly due to variation in soil organic carbon stock changes that is not
explained by the surrogate data method, leading to high prediction error with this data splicing method.
Land Use, Land-Use Change, and Forestry 6-87
-------
Table 6-45: Approach 2 Quantitative Uncertainty Estimates for Carbon Stock Changes
Occurring Within Grassland Remaining Grassland (MMT CO2 Eq. and Percent)
2022 Flux Uncertainty Range Relative to Flux Estimate-1
Source Estimate (MMTCO.Eq.) (%)
(MMT CO . Eq.)
Lower
Bound
Upper
Bound
Lower
Bound
Upper
Bound
Woodland Biomass:
Aboveground live biomass
(1.3)
(1.5)
(1.2)
-10%
12%
Belowground live biomass
(0.2)
(0.3)
(0.2)
-8%
8%
Dead wood
2.8
2.5
3.2
-13%
14%
Litter
+
+
0.1
-22%
22%
Mineral Soil C Stocks Grassland Remaining Grassland,
Tier 3 Methodology
7.4
(116.1)
131.0
-1,663%
1,663%
Mineral Soil C Stocks: Grassland Remaining Grassland,
Tier 2 Methodology
0.1
(0.4)
0.6
-448%
448%
Mineral Soil C Stocks: Grassland Remaining Grassland,
Tier 2 Methodology (Change in Soil C due to Biosolids
[i.e., Treated Sewage Sludge] Amendments)
(1.0)
(1.5)
(0.5)
-50%
50%
Organic Soil C Stocks: Grassland Remaining Grassland,
Tier 2 Methodology
5.5
1.2
9.9
-79%
79%
Combined Uncertainty for Flux Associated with
Carbon Stock Changes Occurring in Grassland
Remaining Grassland
13.4
(110.3)
137.0
-926%
926%
+ Does not exceed 0.05 MMT C02 Eq.
a Range of C stock change estimates is a 95 percent confidence interval.
Notes: Totals may not sum due to independent rounding. Parentheses indicate negative values or net
sequestration.
Uncertainty is also associated with a lack of reporting on biomass, dead wood and litter carbon stock changes for
agroforestry systems. Changes in biomass and dead organic matter carbon stocks are assumed to be negligible in
other grasslands, largely comprised of herbaceous biomass, although there are significant changes at sub-annual
time scales across seasons.
QA/QC and Verification
See the QA/QC and Verification section in Cropland Remaining Cropland.
Recalculations Discussion
Several improvements have been implemented in this Inventory leading to recalculations. These improvements
included a) incorporating new USDA-NRCS NRI data through 2017; b) updated FIA data from 1990 to 2022 on
biomass, dead wood and litter carbon stocks in woodlands for grassland remaining grassland; c) constraining
manure N applications in the Tier 3 method at the state scale rather than the national scale; and d) re-calibrating
the soil carbon module in the DayCent model using Bayesian methods. See the Recalculations Discussion in the
cropland remaining cropland section for other improvements. As a result of these improvements, grassland
remaining grassland has a larger average loss of 10.7 MMT CO2 Eq. across the time series compared to the
previous Inventory, which is an 1,850 percent change on average over the time series. The large average value for
the percentage change is due to an increase from near zero to 7.0 MMT CO2 Eq. for the estimated carbon stock
change in 1994.
6-88 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
Planned Improvements
A key improvement planned for the Inventory includes conducting an analysis of carbon stock changes for
grasslands in Alaska. This improvement will be a significant development that will resolve the majority of the
discrepancy between the managed land base for grassland remaining grassland and amount of area currently
included in grassland remaining grassland emissions and removals calculations (see Table 6-46).
Table 6-46: Comparison of Managed Land Area in Grassland Remaining Grassland and the
Area in the current Grassland Remaining Grassland Inventory (Thousand Hectares)
Area (Thousand Hectares)
Year
Managed Land
Inventory
Difference
1990
328,565
279,705
48,861
1991
328,058
279,205
48,853
1992
327,601
278,755
48,846
1993
325,869
277,030
48,839
1994
324,249
275,418
48,831
1995
323,373
274,549
48,824
1996
322,517
273,701
48,816
1997
321,752
272,944
48,808
1998
319,811
271,010
48,801
1999
318,903
270,110
48,793
2000
317,917
269,131
48,785
2001
317,060
268,282
48,778
2002
316,443
267,883
48,560
2003
316,545
268,206
48,340
2004
316,350
268,232
48,118
2005
315,930
268,034
47,897
2006
315,422
267,748
47,675
2007
315,164
267,712
47,452
2008
315,090
267,861
47,228
2009
315,163
268,159
47,005
2010
314,765
267,984
46,781
2011
314,270
267,712
46,557
2012
313,977
267,586
46,391
2013
314,640
268,416
46,224
2014
315,329
269,271
46,058
2015
315,427
269,535
45,891
2016
315,327
269,602
45,725
2017
316,056
270,339
45,717
2018
318,959
273,168
45,791
2019
320,255
274,471
45,784
2020
320,855
275,079
45,777
2021
321,909
*
*
2022
322,779
*
*
Activity data on land use have not been incorporated into the
Inventory after 2020, designated with asterisks (*).
Additionally, a review of available data on biosolids (i.e., treated sewage sludge) application will be undertaken to
improve the distribution of biosolids application on croplands, grasslands and settlements. For information about
Land Use, Land-Use Change, and Forestry 6-89
-------
other improvements, see the Planned Improvements section in Cropland Remaining Cropland.
Non-C02 Emissions from Grassland Fires (CRT Source Category
4C1)
Fires are common in grasslands and are thought to have been a key feature shaping the evolution of the grassland
vegetation in North America (Daubenmire 1968; Anderson 2004). Fires can occur naturally through lightning strikes
but are also an important management practice to remove standing dead vegetation and improve forage for
grazing livestock. Woody and herbaceous biomass will be oxidized in a fire, although in this section the current
focus is primarily on herbaceous biomass.53 Biomass burning emits a variety of trace gases including non-CC>2
greenhouse gases such as Cm and N2O, as well as CO and NOx that can become greenhouse gases when they react
with other gases in the atmosphere (Andreae and Merlet 2001). IPCC (2006) recommends reporting non-CC>2
greenhouse gas emissions from all wildfires and prescribed burning occurring in managed grasslands.
Biomass burning in grasslands of the United States (including burning emissions in grassland remaining grassland
and land converted to grassland) is a relatively small source of emissions, but it has increased by 184 percent since
1990. In 2022, CH4 and N2O emissions from biomass burning in grasslands were 0.3 MMT CO2 Eq. (12 kt) and 0.3
MMT CO2 Eq. (1 kt), respectively. Annual emissions from 1990 to 2022 have averaged approximately 0.4 MMT CO2
Eq. (14 kt) of CH4 and 0.3 MMT C02 Eq. (1 kt) of l\l20 (see Table 6-47 and Table 6-48).
Table 6-47: CH4 and N2O Emissions from Biomass Burning in Grassland (MMT CO2 Eq.)
1990 1
2005
2018
2019
2020
2021
2022
ch4 0.1
0.4 1
0.6
0.2
0.6
0.5
0.3
T—1
0
0
z
0.4
0.5
0.2
0.5
0.4
0.3
Total Net Flux 0.2
0.8
1.1
0.3
1.1
0.9
0.6
Note: Totals may not sum due to independent rounding.
Table 6-48: CH4, N2O, CO, and NOx Emissions from Biomass Burning ii
1990
2005
2018
2019
2020
2021
2022
CH4 4
15
22
6
20
18
12
n2o +:
1 if
2
1
2
2
1
CO 122
430
610
170
575
509
346
NOx 7
37
10
35
31
21
+ Does not exceed 0.5 kt.
Methodology and Time-Series Consistency
The following section includes a description of the methodology used to estimate non-CC>2 greenhouse gas
emissions from biomass burning in grassland, including (1) determination of the land base that is classified as
managed grassland; (2) assessment of managed grassland area that is burned each year, and (3) estimation of
emissions resulting from the fires. For this Inventory, the IPCC Tier 1 method is applied to estimate non-CC>2
greenhouse gas emissions from biomass burning in grassland from 1990 to 2020 (IPCC 2006). A data splicing
method is used to estimate the emissions from 2021 to 2022, which is discussed later in this section.
The land area designated as managed grassland is based primarily on the USDA National Resources Inventory (NRI)
(Nusser and Goebel 1997; USDA-NRCS 2020). NRI has survey locations across the entire United States, but does not
53 A planned improvement is underway to incorporate woodland tree biomass into the Inventory for non-C02 emissions from
grassland fires.
6-90 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
classify land use on federally-owned areas, and so survey locations on federal lands are designated as grassland
using land cover data from the National Land Cover Dataset (NLCD) (Fry et al. 2011; Homer et al. 2007; Homer et
al. 2015) (see Section 6.1).
The area of biomass burning in grasslands (grassland remaining grassland and land converted to grassland) is
determined using 30-m burned area data from the Monitoring Trends in Burn Severity (MTBS) program for 1990
through 2020 (MTBS 2023; Picotte, et al. 2020).54 NRI survey locations on grasslands are designated as burned in a
year if there is a fire within 500 m of the survey point according to the MTBS fire data. The area of biomass burning
is estimated from the NRI spatial weights and aggregated to the country (Table 6-49).
Table 6-49: Thousands of Grassland Hectares Burned Annually
Year
1990
2005
2018
2019
2020
2021
2022
Thousand Hectares
457
1-612 I
2,290
637
2156
NE
NE
NE (Not Estimated)
Notes: Burned area was not estimated (NE) for 2021 to 2022, but will be updated in a future
Inventory.
For 1990 to 2020, the total area of grassland burned is multiplied by the IPCC default factor for grassland biomass
(4.1 tonnes dry matter per ha) (IPCC 2006) to estimate the amount of combusted biomass. A combustion factor of
1 is assumed in this Inventory, and the resulting biomass estimate is multiplied by the IPCC default grassland
emission factors for CH4 (2.3 g CH4 per kg dry matter), N2O (0.21 g N2O per kg dry matter), CO (65 g CO per kg dry
matter) and NOx (3.9 g NOx per kg dry matter) (IPCC 2006).
A linear extrapolation of the trend in the time series is applied to estimate emissions for 2021 to 2022. Specifically,
a linear regression model with autoregressive moving-average (ARMA) errors (Brockwell and Davis 2016) is used to
derive the trend in emissions over time from 1990 to 2020, and the trend is used to approximate the 2021 to 2022
emissions. The Tier 1 method described previously will be applied to recalculate the 2021 to 2022 emissions in a
future Inventory.
The same methods are applied from 1990 to 2020, and a data splicing method is used to extend the time series
from 2021 to 2022 ensuring a consistent time series of emissions data. The trend extrapolation is a standard data
splicing method for estimating emissions at the end of a time series if activity data are not available (IPCC 2006).
Uncertainty
Emissions are estimated using a linear regression model with ARMA errors for 2021 to 2022. The model produces
estimates for the upper and lower bounds of the emission estimate and the results are summarized in Table 6-50.
Methane emissions from biomass burning in grassland for 2022 are estimated to be between approximately 0.0
and 0.8 MMT CO2 Eq. at a 95 percent confidence level. This indicates a range of 100 percent below and 137
percent above the 2022 emission estimate of 0.3 MMT CO2 Eq. Nitrous oxide emissions are estimated to be
between approximately 0.0 and 0.7 MMT CO2 Eq., or 100 percent below and 137 percent above the 2022 emission
estimate of 0.3 MMT CO2 Eq.
54 See http://www.mtbs.gov.
Land Use, Land-Use Change, and Forestry 6-91
-------
Table 6-50: Uncertainty Estimates for Non-CCh Greenhouse Gas Emissions from Biomass
Burning in Grassland (MMT CO2 Eq. and Percent)
Source
Gas
2022 Emission Estimate
Uncertainty Range Relative to Emission Estimate-'
(MMT CO . Eq.)
(MMT CO.
Eq.)
(%)
Lower
Upper
Lower Upper
Bound
Bound
Bound Bound
Grassland Burning
ch4
0.3
+
0.8
-100% +137%
Grassland Burning
n2o
0.3
+
0.7
-100% +137%
+ Does not exceed 0.05 MMT C02 Eq.
a Range of emission estimates predicted by linear regression time-series model for a 95 percent confidence interval.
Uncertainty is also associated with lack of reporting of emissions from biomass burning in grasslands of Alaska.
Grassland burning emissions could be relatively large in this region of the United States, and therefore extending
this analysis to include Alaska is a planned improvement for the Inventory. There is also uncertainty due to lack of
reporting on the combustion of woody biomass, and this is another planned improvement.
/erification
Quality control measures included checking input data, model scripts, and results to ensure data are properly
handled throughout the inventory process consistent with the U.S. Inventory QA/QC plan, which is in accordance
with Volume 1, Chapter 6 of the 2006IPCC Guidelines (see Annex 8 for more details). Inventory reporting forms
and text are reviewed and revised as needed to correct transcription errors.
Recalculations Discussion
While the methods for calculating non-CC>2 emissions from grassland burning remained the same, the two primary
data sources have been updated from the previous Inventory. We used the current NRI 2017 dataset (USDA-NRCS
2020) and the current release of MTBS burn perimeter data (MTBS 2023). In the original estimation of non-CC>2
emissions, the same set of NRI survey locations was used for the entire time series, but the locations identified
with burning were allowed to vary inter-annually with this revision. These changes resulted in a net increase in
CCh-equivalent emissions by an annual average of 0.1 MMT CO2 Eq., or 19 percent from 1990 to 2021 compared to
the previous Inventory.
Planned Improvements
Two key planned improvements have been identified for this source category, including 1) incorporation of
country-specific grassland biomass factors, and 2) extending the analysis to include Alaska. In the current
Inventory, biomass factors are based on a global default for grasslands that is provided by the IPCC (2006). There is
considerable variation in grassland biomass, however, which would affect the amount of fuel available for
combustion in a fire. Alaska has an extensive area of grassland and includes tundra vegetation, although some of
the areas are not managed. There has been an increase in fire frequency in boreal forest of the region (Chapin et
al. 2008), and this may have led to an increase in burning of neighboring grassland areas. There is also an effort
under development to incorporate grassland fires into DayCent model simulations. Lastly, a future Inventory will
incorporate non-CC>2 greenhouse emissions from burning woodland tree biomass in grasslands. These
improvements are expected to reduce uncertainty and produce more accurate estimates of non-CC>2 greenhouse
gas emissions from grassland burning.
6-92 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
6.7 Land Converted to Grassland (CRT
Category 4C2)
Land converted to grassland includes all current grassland in an inventory year that had been in another land
use(s) during the previous 20 years (IPCC 2006).55 For example, cropland or forest land converted to grassland
during the past 20 years would be reported in this category. Recently converted lands are retained in this category
for 20 years as recommended by IPCC (2006). Grassland includes pasture and rangeland that are used primarily but
not exclusively for livestock grazing. Rangelands are typically extensive areas of native grassland that are not
intensively managed, while pastures are typically seeded grassland (possibly following tree removal) that may also
have additional management, such as irrigation or interseeding of legumes.
Land use change can lead to large losses of carbon to the atmosphere, particularly conversions from forest land
(Houghton et al. 1983, Houghton and Nassikas 2017). Moreover, conversion of forest to another land use (i.e.,
deforestation) is one of the largest anthropogenic sources of emissions to the atmosphere globally, although this
source may be declining according to a recent assessment (Tubiello et al. 2015).
IPCC (2006) recommends reporting changes in biomass, dead organic matter, and soil organic carbon stocks due to
land use change. All soil organic carbon stock changes are estimated and reported for land converted to grassland,
but there is limited reporting of other pools in this Inventory. Losses of aboveground and belowground biomass,
dead wood and litter carbon from forest land converted to grassland are reported, as well as gains and losses
associated with conversions to woodlands56 from other land uses, including croplands converted to grasslands,
settlements converted to grasslands and other lands converted to grasslands. However, the current Inventory does
not include the gains and losses in aboveground and belowground biomass, dead wood and litter carbon for other
land use conversions to grassland that are not woodlands.57
There is a discrepancy between the current land representation (see Section 6.1) and the area data that have been
used in the inventory for land converted to grassland. Specifically, grassland in Alaska is not included in the
Inventory, and this leads to a difference between the managed area in land converted to grassland in the land
representation and the grassland area included in the emissions and removal calculations for land converted to
grassland (Table 6-54). Improvements are underway to incorporate grassland area in Alaska as part of future
Inventories (see Planned Improvements section).
The largest carbon losses with land converted to grassland are associated with aboveground biomass,
belowground biomass, and litter carbon losses from forest land converted to grassland (see Table 6-51 and Table
6-52). These three pools led to net emissions in 2022 of 31.3, 4.3, and 8.0 MMT CO2 Eq. (8.5,1.2, and 2.2 MMT C),
respectively. The losses associated with forest land converted to grassland are partially offset by gains associated
with other land converted to grassland and due to cropland converted to grassland, which leads to less intensive
management of the soil. Drainage of organic soils for grassland management led to CO2 emissions to the
atmosphere of 1.4 MMT CO2 Eq. (0.4 MMT C). The total net carbon stock change in 2022 for land converted to
55 USDA 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 between 1971 and 1978.
56 Woodlands are considered grasslands in the U.S. land representation because they do not meet the definition of forest land.
57 Changes in biomass carbon stocks are not currently reported for other conversions to grassland (other than forest land
conversion to grassland and other land-use conversions to woodlands), but this is a planned improvement for a future
Inventory. Note: changes in dead organic matter are assumed negligible for other land use conversions (i.e., other than forest
land) to grassland based on the Tier 1 method in IPCC (2006).
Land Use, Land-Use Change, and Forestry 6-93
-------
grassland is estimated as a loss of 25.6 MMT CO2 Eq. (7.0 MMT C) or a net source of emissions, which represents a
decrease in carbon stock loss by 27 percent compared to the initial reporting year of 1990.
Table 6-51: Net CO2 Flux from Soil, Dead Organic Matter and Biomass Carbon Stock Changes
for Land Converted to Grassland (MMT CO2 Eq.)
1990
2005
2018
2019
2020
2021
2022
Cropland Converted to Grassland
(10.2)
(16.9)
(10.8)
(10.3)
(9.3)
(13.6)
(12.5)
Aboveground Live Biomass
(0.2) 1
(0.1) 1
(0.1)
(0.1)
(0.1)
(0.1)
(0.1)
Belowground Live Biomass
+
+
+
+
+
+
+
Dead Wood
(0.1) "
(0.1) 1
+
+
+
+
+
Litter
(0.1)
(0.1) 1
+
+
+
+
+
Mineral Soils
(10.4) |
(18-D 1
(11.7)
(11.1)
(10.1)
(14.4)
(13.3)
Organic Soils
0.6
1.4
1.1
1.1
1.0
1.0
1.0
Forest Land Converted to Grassland
50.2
49.0
46.9
46.9
46.8
46.8
46.8
Aboveground Live Biomass
34.5
33.4
31.8
31.8
31.8
31.8
31.8
Belowground Live Biomass
4.8 I
4-6 1
4.4
4.4
4.4
4.4
4.4
Dead Wood
2.4
2.4
2.4
2.4
2.4
2.4
2.4
Litter
8.6
—III
in
CO
8.2
8.2
8.2
8.2
8.2
Mineral Soils
(0.1)
(0.1)
(0.1)
(0.1)
(0.1)
(0.1)
(0.1)
Organic Soils
_|_ mils
O.i I
0.1
0.1
0.1
0.1
0.1
Other Lands Converted to Grassland
(4.0)
(9.6)
(10.2)
(10.5)
(8.2)
(8.0)
(8.0)
Aboveground Live Biomass
(0.1) :
(0.1) 1
(0.1)
(0.1)
(0.1)
(0.1)
(0.1)
Belowground Live Biomass
+
+
+
+
+
+
+
Dead Wood
+:
+ I
+
+
+
+
+
Litter
(0.1)
(0.1)
(0.1)
(0.1)
(0.1)
(0.1)
(0.1)
Mineral Soils
(3-8) j
(9.4) |
(10.1)
(10.4)
(8.1)
(7.9)
(7.9)
Organic Soils
+
+
0.1
0.1
0.1
0.1
0.1
Settlements Converted to Grassland
(0.6)
(0.7)
(0.8)
(0.8)
(0.8)
(0.8)
(0.8)
Aboveground Live Biomass
(0.2)
(0.2)
(0.2)
(0.2)
(0.2)
(0.2)
(0.2)
Belowground Live Biomass
+ E
+ "
+
+
+
+
+
Dead Wood
(0.1)
(o.i)
+
+
+
+
+
Litter
(0.1)
(0.1) 1
(0.1)
(0.1)
(0.1)
(0.1)
(0.1)
Mineral Soils
(0.1)
(0.3)
(0.5)
(0.5)
(0.4)
(0.4)
(0.4)
Organic Soils
+ ;
+
+
+
+
+
+
Wetlands Converted to Grassland
(0.1)
0.1
0.1
0.1
0.1
0.1
Aboveground Live Biomass
(o.i) :
-------
Table 6-52: Net CO2 Flux from Soil, Dead Organic Matter and Biomass Carbon Stock Changes
for Land Converted to Grassland (MMT C)
1990
2005
111!
Illl
2018
2019
2020
2021
2022
Cropland Converted to Grassland
(2.8)
(4.6)
(3.0)
(2.8)
(2.5)
(3.7)
(3.4)
Aboveground Live Biomass
+
+
I
+
+
+
+
+
Belowground Live Biomass
+
+
+
+
+
+
+
Dead Wood
+
1
+
+
+
+
+
Litter
+
+
+
+
+
+
+
Mineral Soils
(2.8);;;
(4.9)
I
(3.2)
(3.0)
(2.8)
(3.9)
(3.6)
Organic Soils
0.2
0.4
0.3
0.3
0.3
0.3
0.3
Forest Land Converted to Grassland
13.7
13.3
1
I
12.8
12.8
12.8
12.8
12.8
Aboveground Live Biomass
9.4
9.1
8.7
8.7
8.7
8.7
8.7
Belowground Live Biomass
i-3 ::
1.3
I
1.2
1.2
1.2
1.2
1.2
Dead Wood
0.7
0.7
0.7
0.7
0.7
0.7
0.7
Litter
2.4
2.3
I
2.2
2.2
2.2
2.2
2.2
Mineral Soils
+
+
+
+
+
+
+
Organic Soils
+
+
Jj
+
+
+
+
+
Other Lands Converted to Grassland
(1.1)
(2.6)
(2.8)
(2.9)
(2.2)
(2.2)
(2.2)
Aboveground Live Biomass
+ i
+
I
+
+
+
+
+
Belowground Live Biomass
+
+
+
+
+
+
+
Dead Wood
+ 1
+
¦
+
+
+
+
+
Litter
+
+
+
+
+
+
+
Mineral Soils
(1.0)
(2.6)
I
(2.8)
(2.8)
(2.2)
(2.2)
(2.2)
Organic Soils
+
+
+
+
+
+
+
Settlements Converted to Grassland
(0.2)
(0.2)
1
(0.2)
(0.2)
(0.2)
(0.2)
(0.2)
Aboveground Live Biomass
(0.1)
(0.1)
(0.1)
(0.1)
(0.1)
(0.1)
(0.1)
Belowground Live Biomass
+ i
+
I
+
+
+
+
+
Dead Wood
+
+
+
+
+
+
+
Litter
+ !
+
1
I
+
+
+
+
+
Mineral Soils
+
(0.1)
(0.1)
(0.1)
(0.1)
(0.1)
(0.1)
Organic Soils
+
+
1
+
+
+
+
+
Wetlands Converted to Grassland
+
+
+
+
+
+
+
Aboveground Live Biomass
+
1
+
+
+
+
+
Belowground Live Biomass
+
+
+
+
+
+
+
Dead Wood
+
+
mini
1
+
+
+
+
+
Litter
+
+
+
+
+
+
+
Mineral Soils
+ -
+
1
+
+
+
+
+
Organic Soils
+
0.1
0.1
0.1
0.1
0.1
0.1
Aboveground Live Biomass
9.2
9.0
nun:
8.6
8.6
8.5
8.5
8.5
Belowground Live Biomass
1.3
1.2
1.2
1.2
1.2
1.2
1.2
Dead Wood
0.6
0.6
0.6
0.6
0.6
0.6
0.6
Litter
2.3
2.2
2.2
2.2
2.2
2.2
2.2
Total Mineral Soil Flux
(4.0)
(7.6)
(6.1)
(6.0)
(5.1)
(6.2)
(5.9)
Total Organic Soil Flux
0.2
0.5
0.4
0.4
0.4
0.4
0.4
Total Net Flux
9.6
5.9
6.9
6.9
7.8
6.7
7.0
+ Does not exceed 0.05 MMT C.
Notes: Totals may not sum due to independent rounding. Parentheses indicate negative values or net
sequestration.
Methodology and Time-Series Consistency
The following section includes a description of the methodology used to estimate carbon stock changes for land
converted to grassland, including (1) loss of aboveground and belowground biomass, dead wood and litter carbon
with forest land converted to grassland and other land use conversions to woodlands, as well as (2) the impact
from all land use conversions to grassland on mineral and organic soil organic carbon stocks.
Land Use, Land-Use Change, and Forestry 6-95
-------
Biomass, Dead Wood, and Litter Carbon Stock Changes
A Tier 3 method is applied to estimate biomass, dead wood and litter carbon stock changes for forest land
converted to grassland. Estimates are calculated in the same way as those in the forest land remaining forest land
category using data from the USDA Forest Service, Forest Inventory and Analysis (FIA) program (USDA Forest
Service 2023) and in the Eastern US, IPCC (2006) defaults for biomass in grasslands. There is limited data on
grassland carbon stocks so only default biomass estimates (IPCC 2006) for grasslands were used to estimate carbon
stock changes (litter and dead wood carbon stocks were assumed to be zero since no reference carbon density
estimates exist for croplands) in the eastern United States. The difference between the stocks is reported as the
stock change under the assumption that the change occurred in the year of the conversion.
The amount of biomass carbon that is lost abruptly with forest land converted to grasslands is estimated based on
the amount of carbon before conversion and the amount of carbon following conversion according to
remeasurements in the FIA program. This approach is consistent with IPCC (2006) that assumes there is an abrupt
change during the first year, but does not necessarily capture the slower change over the years following
conversion until a new steady state is reached. It was determined that using an IPCC Tier 1 approach that assumes
all carbon is lost in the year of conversion for forest land converted to grasslands in the West and Great Plains
states does not accurately characterize the transfer of carbon in woody biomass during abrupt or gradual land use
change. To estimate this transfer of carbon in woody biomass, state-specific carbon densities for woody biomass
remaining on these former forest lands following conversion to grasslands were developed and included in the
estimation of carbon stock changes from forest land converted to grasslands in the West and Great Plains states. A
review of the literature in grassland and rangeland ecosystems (Asner et al. 2003; Huang et al. 2009; Tarhouni et
al. 2016), as well as an analysis of FIA data, suggests that a conservative estimate of 50 percent of the woody
biomass carbon density was lost during conversion from forest land to grasslands. This estimate was used to
develop state-specific carbon density estimates for biomass, dead wood, and litter for grasslands in the West and
Great Plains states, and these state-specific carbon densities were applied in the compilation system to estimate
the carbon losses associated with conversion from forest land to grassland in the West and Great Plains states.
Further, losses from forest land to what are often characterized as woodlands are included in this category using
FIA plot remeasurements and the methods and models briefly described below and in detail in Domke et al. (2022)
and Westfall et al. (2023).
If FIA plots include data on individual trees, aboveground and belowground carbon density estimates are based on
Woodall et al. (2011) and Westfall et al. (2023). Aboveground and belowground biomass estimates also include live
understory which is a minor component of biomass defined as all biomass of undergrowth plants in a forest,
including woody shrubs and trees less than 2.54 cm dbh. For this Inventory, it was assumed that 10 percent of total
understory carbon mass is belowground (Smith et al. 2006). Estimates of carbon density are based on information
in Birdsey (1996) and biomass estimates from Jenkins et al. (2003). If FIA plots include data on standing dead trees,
standing dead tree carbon density is estimated following the basic method applied to live trees (Woodall et al.
2011, Westfall et al. 2023) with additional modifications to woodland species to account for decay and structural
loss (Domke et al. 2011; Harmon et al. 2011).
If FIA plots include data on downed dead wood, downed dead wood carbon density is estimated based on
measurements of a subset of FIA plots for downed dead wood (Domke et al. 2013; Woodall and Monleon 2008).
Downed dead wood is defined as pieces of dead wood greater than 7.5 cm diameter, at transect intersection, that
are not attached to live or standing dead trees. This includes stumps and roots of harvested trees. To facilitate the
downscaling of downed dead wood carbon estimates from the state-wide population estimates to individual plots,
downed dead wood models specific to regions and forest types within each region are used. Litter carbon is the
pool of organic carbon (also known as duff, humus, and fine woody debris) above the mineral soil and includes
woody fragments with diameters of up to 7.5 cm. A subset of FIA plots are measured for litter carbon. If FIA plots
include litter material, a modeling approach using litter carbon measurements from FIA plots is used to estimate
litter carbon density (Domke et al. 2016). See Annex 3.13 for more information about reference carbon density
estimates for forest land.
6-96 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
Soil Carbon Stock Changes
Soil organic carbon stock changes are estimated for land converted to grassland according to land use histories
recorded in the 2017 USDA NRI survey for non-federal lands (USDA-NRCS 2020). Land use and some management
information (e.g., crop type, soil attributes, and irrigation) were originally collected for each NRI survey location on
a five-year cycle beginning in 1982. In 1998, the NRI Program began collecting annual data, and the annual data are
currently available through 2017 (USDA-NRCS 2020). For 2018 through 2020, the time series is extended with the
crop data provided in USDA-NASS CDL (USDA-NASS 2021), while survey locations identified as grasslands are
assumed to not change over this time period. However, the areas have been modified in the original NRI survey
through a process in which the Forest Inventory and Analysis (FIA) survey data and the National Land Cover
Dataset (NLCD; Yang et al. 2018) are harmonized with the NRI data. This process ensures that the land use areas
are consistent across all land use categories (see Section 6.1 for more information).
NRI survey locations are classified as land converted to grassland in a given year between 1990 and 2020 if the
land use is grassland but had been classified as another use during the previous 20 years. 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 1998. 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 between 1971 and 1978. For
federal lands, the land use history is derived from land cover changes in the NLCD (Yang et al. 2018; Homer et al.
2007; Fry et al. 2011; Homer et al. 2015).
Soil Carbon Stock Changes for Mineral Soils
An IPCC Tier 3 model-based approach (Ogle et al. 2010) is applied to estimate carbon stock changes in mineral soils
for most of the area in land converted to grassland. Carbon stock changes on the remaining area are estimated
with an IPCC Tier 2 approach (Ogle et al. 2003), including prior cropland used to produce vegetables, tobacco, and
perennial/horticultural crops; land areas with very gravelly, cobbly, or shaley soils (greater than 35 percent by
volume); and land converted to grassland from another land use other than cropland.
A surrogate data method is used to estimate soil organic carbon stock changes from 2021 to 2022 at the national
scale for land areas included in the Tier 2 and Tier 3 methods. Specifically, linear regression models with
autoregressive moving-average (ARMA) errors (Brockwell and Davis 2016) are used to estimate the relationship
between surrogate data and the 1990 to 2020 emissions data that are derived using the Tier 2 and 3 methods.
Surrogate data for these regression models includes weather data from the PRISM Climate Group (PRISM Climate
Group 2022). See Box 6-4 in the Methodology section of cropland remaining cropland for more information about
the surrogate data method.
Tier 3 Approach. Mineral soil organic carbon stocks and stock changes are estimated using the DayCent ecosystem
model (Parton et al. 1998; Del Grosso et al. 2001, 2011). The DayCent model utilizes the soil carbon modeling
framework developed in the Century model (Parton et al. 1987,1988,1994; Metherell et al. 1993), but has been
refined to simulate dynamics at a daily time-step. Historical land use patterns and irrigation histories are simulated
with DayCent based on the 2017 USDA NRI survey (USDA-NRCS 2018). Carbon stocks and 95 percent confidence
intervals are estimated for each year between 1990 and 2020. See the cropland remaining cropland section and
Annex 3.12 for additional discussion of the Tier 3 methodology for mineral soils.
In order to ensure time-series consistency, the same methods are applied from 1990 to 2020 so that changes
reflect anthropogenic activity and not methodological adjustments. In addition, soil organic carbon stock changes
from 2021 to 2022 are approximated using a linear extrapolation of emission patterns from 1990 to 2020. The
extrapolation is based on a linear regression model with moving-average (ARMA) errors, described in Box 6-4 of
the Methodology section in cropland remaining cropland. Linear extrapolation is a standard data splicing method
for estimating emissions at the end of a time series (IPCC 2006). Stock change estimates for 2021 to 2022 will be
recalculated in future Inventories with an updated time series of activity data (see the Planned Improvements
section in cropland remaining cropland).
Land Use, Land-Use Change, and Forestry 6-97
-------
Tier 2 Approach. For the mineral soils not included in the Tier 3 analysis, soil organic carbon stock changes are
estimated using a Tier 2 Approach, as described in the Tier 2 Approach for mineral soils in grassland remaining
grassland and Annex 3.12. In order to ensure time-series consistency, the Tier 2 method is applied from 1990 to
2020 so that changes reflect anthropogenic activity and not methodological adjustments. In addition, soil organic
carbon stock changes are approximated for the remainder of the time series with a linear extrapolation of
emission patterns from 1990 to 2020. The extrapolation is based on a linear regression model with moving-
average (ARMA) (see Box 6-4 of the Methodology section in Cropland Remaining Cropland). Linear extrapolation is
a standard data splicing method for estimating emissions at the end of a time series (IPCC 2006). As with the Tier 3
method, stock change estimates for 2021 to 2022 will be recalculated in future Inventories with an updated time
series of activity data.
Soil Carbon Stock Changes for Organic Soils
Annual carbon emissions from drained organic soils in land converted to grassland are estimated using the Tier 2
method provided in IPCC (2006), with country-specific carbon loss rates (Ogle et al. 2003) as described in the
cropland remaining cropland section. Further elaboration on the methodology is also provided in Annex 3.12 for
organic soils.
In order to ensure time-series consistency, the Tier 2 method is applied from 1990 to 2020 so that changes reflect
anthropogenic activity and not methodological adjustments. In addition, soil organic carbon stock changes are
approximated for the remainder of the time series with a linear extrapolation of emission patterns from 1990 to
2020. The extrapolation is based on a linear regression model with moving-average (ARMA) errors (see Box 6-4 of
the Methodology section in Cropland Remaining Cropland). Linear extrapolation is a standard data splicing method
for estimating emissions at the end of a time series (IPCC 2006). Annual carbon emissions from drained organic
soils from 2021 to 2022 will be recalculated in future Inventories with an updated time series of activity data.
Uncertainty
The uncertainty analyses for biomass, dead wood and litter carbon losses with forest land converted to grassland
and other land use conversions to woodlands are conducted in the same way as the uncertainty assessment for
forest ecosystem carbon flux in the forest land remaining forest land category. Sample and model-based error are
combined using simple error propagation methods provided by the IPCC (2006), by taking the square root of the
sum of the squares of the standard deviations of the uncertain quantities. For additional details see the
Uncertainty Analysis in Annex 3.13.
The uncertainty analyses for soil organic carbon stock changes using the Tier 3 and Tier 2 methodologies are
quantified from two variance components (Ogle et al. 2010), as described in cropland remaining cropland. For
2021 to 2022, there is additional uncertainty propagated through the Monte Carlo analysis associated with a
surrogate data method, which is also described in the Cropland Remaining Cropland section.
Uncertainty estimates are presented in Table 6-53 for each subsource (i.e., biomass carbon stocks, mineral and
organic carbon stocks in soils) 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), as discussed in the previous paragraph. The combined uncertainty for total carbon stocks in land
converted to grassland ranges from 156 percent below to 156 percent above the 2022 stock change estimate of
25.6 MMT CO2 Eq. The large relative uncertainty around the 2022 stock change estimate is partly due to large
uncertainties in biomass and dead organic matter carbon losses with forest land conversion to grassland, in
addition to variation in soil organic carbon stock changes that is not explained by the surrogate data method.
6-98 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
Table 6-53: Approach 2 Quantitative Uncertainty Estimates for Soil, Dead Organic Matter and
Biomass Carbon Stock Changes occurring within Land Converted to Grassland (MMT CO2 Eq.
and Percent)
Source
2022 Flux Estimate-1
Uncertainty Range Relative to Flux Estimate-'
(MMT CO . Eq.)
(MMT CO.
Eq.)
(%)
Lower
Upper
Lower
Upper
Bound
Bound
Bound
Bound
Cropland Converted to Grassland
(12.5)
(32.1)
7.0
-156%
156%
Aboveground Live Biomass
(0.1)
(0.3)
+
-136%
134%
Belowground Live Biomass
+
+
+
-78%
100%
Dead Wood
+
(0.1)
+
-128%
100%
Litter
+
(0.1)
+
-170%
100%
Mineral Soil C Stocks: Tier 3
(11.6)
(31.0)
7.8
-167%
167%
Mineral Soil C Stocks: Tier 2
(1.7)
(3.7)
0.2
-114%
114%
Organic Soil C Stocks: Tier 2
1.0
+
2.0
-96%
96%
Forest Land Converted to Grassland
46.8
12.2
81.4
-74%
74%
Aboveground Live Biomass
31.8
(1.4)
64.9
-104%
104%
Belowground Live Biomass
4.4
(0.2)
9.0
-104%
105%
Dead Wood
2.4
(0.1)
5.0
-105%
104%
Litter
8.2
(0.4)
16.8
-104%
104%
Mineral Soil C Stocks: Tier 2
(0.1)
(0.2)
+
-140%
140%
Organic Soil C Stocks: Tier 2
0.1
(0.0)
0.2
-130%
130%
Other Lands Converted to Grassland
(8.0)
(12.9)
(3.1)
-61%
61%
Aboveground Live Biomass
(0.1)
(0.1)
+
-69%
44%
Belowground Live Biomass
+
+
+
-100%
100%
Dead Wood
+
(0.1)
+
-85%
100%
Litter
(0.1)
(0.1)
+
-60%
47%
Mineral Soil C Stocks: Tier 2
(7.9)
(12.8)
(3.0)
-62%
62%
Organic Soil C Stocks: Tier 2
0.1
+
0.1
-111%
111%
Settlements Converted to Grassland
(0.8)
(1.0)
(0.5)
-35%
35%
Aboveground Live Biomass
(0.2)
(0.3)
(0.1)
-56%
61%
Belowground Live Biomass
+
+
+
-42%
100%
Dead Wood
+
(0.1)
+
-67%
100%
Litter
(0.1)
(0.1)
+
-66%
59%
Mineral Soil C Stocks: Tier 2
(0.4)
(0.6)
(0.2)
-56%
56%
Organic Soil C Stocks: Tier 2
+
+
+
-432%
432%
Wetlands Converted to Grasslands
0.1
(0.2)
0.3
-289%
283%
Aboveground Live Biomass
(0.1)
(0.1)
+
-85%
38%
Belowground Live Biomass
+
+
+
-100%
100%
Dead Wood
+
+
+
-95%
100%
Litter
+
+
+
-112%
100%
Mineral Soil C Stocks: Tier 2
+
+
+
-173%
173%
Organic Soil C Stocks: Tier 2
0.2
+
0.4
-112%
112%
Total: Land Converted to Grassland
25.6
(14.4)
65.7
-156%
156%
Aboveground Live Biomass
31.3
(1.8)
64.5
-106%
106%
Belowground Live Biomass
4.3
(0.2)
9.0
-106%
106%
Dead Wood
2.3
(0.3)
4.8
-111%
111%
Litter
8.0
(0.6)
16.6
-107%
107%
Mineral Soil C Stocks: Tier 3
(11.6)
(31.0)
7.8
-167%
167%
Mineral Soil C Stocks: Tier 2
(10.1)
(15.4)
(4.9)
-52%
52%
Land Use, Land-Use Change, and Forestry 6-99
-------
Organic Soil C Stocks: Tier 2 1A 04 2A -74% 74%
+ Absolute value does not exceed 0.05 MMT C02 Eq.
a Range of C stock change estimates is a 95 percent confidence interval.
Notes: Totals may not sum due to independent rounding. Parentheses indicate negative values or net sequestration.
Uncertainty is also associated with a lack of reporting on biomass, dead wood and litter carbon stock changes for
conversions to agroforestry systems and herbaceous grasslands. The influence of agroforestry is difficult to address
because there are currently no datasets to evaluate the trends in the area and associated carbon stocks in
agroforestry systems. The influence of land use change to herbaceous grasslands and agroforestry will be further
explored in a future Inventory.
/erification
See the QA/QC and Verification section in Cropland Remaining Cropland for information on QA/QC steps.
Recalculations Discussion
Several improvements have been implemented in this Inventory leading to recalculations. These improvements
included a) incorporating new USDA-NRCS NRI data through 2017; b) updated FIA data from 1990 to 2022 on
biomass, dead wood and litter carbon stocks associated with forest land converted to grassland (see Recalculations
Discussion of Chapter 6.2 Forest Land Remaining Forest Land for more details); c) constraining manure nitrogen
applications in the Tier 3 method at the state scale rather than the national scale; and d) re-calibrating the soil
carbon module in the DayCent model using Bayesian methods. See the Recalculations Discussion in the cropland
remaining cropland section for other improvements. Finally, see further updates in Section 6.2, describing updates
to the estimates for aboveground volume and biomass which impacted lands converted to grassland estimates. As
a result, land converted to grassland has an estimated increase in losses of carbon stock changes, leading to a net
change of 53 MMT CO2 Eq. on average over the time series, representing a 237 percent change on average
compared to the previous Inventory. Land converted to grassland is a net source of emissions across the time
series based on the recalculations in this Inventory. This change from a net sink to a net source is mostly due to
larger estimated losses of biomass and dead organic matter with forest land converted to grassland, and smaller
estimated gains in mineral soil carbon stocks for cropland and other lands converted to grasslands.
Planned Improvements
The key improvement planned for the inventory is conducting an analysis of carbon stock changes for grassland in
Alaska. This will resolve the majority of the discrepancy between the managed land base for land converted to
grassland and amount of area currently included in land converted to grassland emissions and removals
calculations (see Table 6-54).
Table 6-54: Comparison of Managed Land Area in Land Converted to Grassland and Area in
the current Land Converted to Grassland Inventory (Thousand Hectares)
Area (Thousand Hectares)
Year
Managed Land
Inventory
Difference
1990
9,301
9,297
4
1991
9,492
9,488
4
1992
9,710
9,706
4
1993
11,619
11,615
4
1994
13,372
13,368
4
1995
14,039
14,035
4
1996
14,727
14,723
4
6-100 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
1997
15,411
15,408
4
1998
19,289
19,285
4
1999
20,143
20,139
4
2000
21,257
21,253
4
2001
22,349
22,345
4
2002
23,087
22,817
270
2003
22,986
22,445
541
2004
23,920
23,108
811
2005
24,091
23,009
1,082
2006
24,693
23,341
1,352
2007
24,694
23,072
1,622
2008
25,266
23,373
1,893
2009
25,424
23,260
2,163
2010
25,769
23,336
2,434
2011
26,176
23,471
2,704
2012
26,164
23,292
2,871
2013
25,154
22,116
3,038
2014
23,981
20,776
3,205
2015
24,101
20,730
3,372
2016
23,531
19,993
3,538
2017
22,808
19,270
3,538
2018
19,968
16,429
3,538
2019
19,546
16,008
3,538
2020
18,706
15,168
3,538
2021
17,351
*
*
2022
16,269
*
*
Activity data on land use have not been incorporated into the Inventory
after 2020, designated with asterisks (*).
In addition, the amount of biomass carbon that is lost abruptly or the slower changes that continue to occur over a
decade or longer with forest land converted to grasslands will be further refined in a future Inventory. The current
values are estimated based on the amount of carbon before conversion and an estimated level of carbon left after
conversion based on limited plot data from the FIA and published literature for the Western United States and
Great Plains Regions. The amount of carbon left after conversion will be further investigated with additional data
collection, particularly in the Western United States and Great Plains, including tree biomass, understory biomass,
dead wood and litter carbon pools. In addition, biomass carbon stock changes will be estimated for conversions
from other land uses to herbaceous grasslands. For information about other improvements, see the Planned
Improvements section in Cropland Remaining Cropland.
6.8 Wetlands Remaining Wetlands (CRT
Category 4D1)
Wetlands remaining wetlands includes all wetlands in an inventory year that have been classified as a wetland for
the previous 20 years, and in this Inventory, the flux estimates include peatlands, coastal wetlands, and flooded
land.
Land Use, Land-Use Change, and Forestry 6-101
-------
Peatlands Remaining Peatlands
Emissions from Managed Peatlands
Managed peatlands are peatlands that have been cleared and drained for the production of peat. The production
cycle of a managed peatland has three phases: land conversion in preparation for peat extraction (e.g., clearing
surface biomass, draining), extraction (which results in the emissions reported under peatlands remaining
peatlands), and abandonment, restoration, rewetting, or conversion of the land to another use.
Carbon dioxide emissions from the removal of biomass and the decay of drained peat constitute the major
greenhouse gas flux from managed peatlands. Managed peatlands may also emit Cm and N2O. The natural
production of Cm is largely reduced but not entirely eliminated when peatlands are drained in preparation for
peat extraction (Strack et al. 2004 as cited in the 2006IPCC Guidelines). Drained land surface and ditch networks
contribute to the Cm flux in peatlands managed for peat extraction. Methane emissions were considered
insignificant under the IPCC Tier 1 methodology (IPCC 2006), but are included in the emissions estimates for
peatlands remaining peatlands consistent with the 2013 Supplement to the 2006 IPCC Guidelines for National
Greenhouse Gas Inventories: Wetlands (IPCC 2013). Nitrous oxide emissions from managed peatlands depend on
site fertility. In addition, abandoned and restored peatlands continue to release greenhouse gas emissions.
Although methodologies are provided to estimate emissions and removals from rewetted organic soils (which
includes rewetted/restored peatlands) in IPCC (2013) guidelines, information on the areal extent of
rewetted/restored peatlands in the United States is currently unavailable. The current Inventory estimates CO2,
Cm and N2O emissions from peatlands managed for peat extraction in accordance with IPCC (2006 and 2013)
guidelines.
CO2, N2O, and CH4 Emissions from Peatlands Remaining Peatlands
IPCC (2013) recommends reporting CO2, N2O, and CH4 emissions from lands undergoing active peat extraction (i.e.,
peatlands remaining peatlands) as part of the estimate for emissions from managed wetlands. Peatlands occur
where plant biomass has sunk to the bottom of water bodies and water-logged areas and exhausted the oxygen
supply below the water surface during the course of decay. Due to these anaerobic conditions, much of the plant
matter does not decompose but instead forms layers of peat over decades and centuries. In the United States,
peat is extracted for horticulture and landscaping growing media, and for a wide variety of industrial, personal
care, and other products. It has not been used for fuel in the United States for many decades. Peat is harvested
from two types of peat deposits in the United States: Sphagnum bogs in northern states (e.g., Minnesota) and
wetlands in states further south (e.g., Florida). The peat from Sphagnum bogs in northern states, which is nutrient-
poor, is generally corrected for acidity and mixed with fertilizer. Production from more southerly states is relatively
coarse (i.e., fibrous) but nutrient-rich.
IPCC (2006 and 2013) recommend considering both on-site and off-site emissions when estimating CO2 emissions
from peatlands remaining peatlands using the Tier 1 approach. Current IPCC methodologies estimate only on-site
N2O and CH4 emissions. This is because off-site N2O estimates are complicated by the risk of double-counting
emissions from nitrogen fertilizers added to horticultural peat where subsequent runoff or leaching into
waterbodies can result in indirect N2O emissions that are already included within the agricultural soil management
category.
On-site emissions from managed peatlands occur as the land is drained and cleared of vegetation, and the
underlying peat is exposed to sun, weather and oxygen. As this occurs, some of the peat deposit is lost and CO2 is
emitted from the oxidation of the peat. Since N2O emissions from saturated ecosystems tend to be low unless
there is an exogenous source of nitrogen, N2O emissions from drained peatlands are dependent on nitrogen
mineralization and therefore on soil fertility. Peatlands occurring on highly fertile/nutrient-rich soils, mostly
located in the southern peatlands in Florida, contain significant amounts of organic nitrogen in inert/microbially
inaccessible forms. Draining land in preparation for peat extraction allows bacteria to convert the organic nitrogen
6-102 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
into nitrates through nitrogen mineralization which leach to the surface where they are reduced to N2O during
nitrification. Nitrate availability also contributes to the activity of methanogens and methanotrophs that result in
Cm emissions (Blodau 2002; Treat et al. 2007 as cited in IPCC 2013). Drainage ditches, which are constructed to
drain the land in preparation for peat extraction, also contribute to the flux of CH4 through in situ production and
lateral transfer of CH4 from the organic soil matrix (IPCC 2013).
Off-site CO2 emissions from managed peatlands occur from waterborne dissolved organic carbon losses and the
horticultural and landscaping use of peat. Dissolved organic carbon from water drained off peatlands reacts within
aquatic ecosystems and is converted to CO2, which is then emitted to the atmosphere (Billet et al. 2004 as cited in
IPCC 2013). During the horticultural and landscaping use of peat, nutrient-poor (but fertilizer-enriched) peat tends
to be used in bedding plants and in greenhouse and plant nursery production, whereas nutrient-rich (but relatively
coarse) peat is used directly in landscaping, athletic fields, golf courses, and plant nurseries. Most (nearly 94
percent) of the CO2 emissions from peat occur off-site, as the peat is processed and sold to firms which, in the
United States, use it predominantly for the aforementioned horticultural and landscaping purposes.
Total emissions from peatlands remaining peatlands are estimated to be 0.6 MMT CO2 Eq. in 2022 (see Table 6-55
and Table 6-56) comprising 0.6 MMT C02 Eq. (572 kt) of C02, 0.004 MMT C02 Eq. (0.13 kt) of CH4 and 0.0004 MMT
CO2 Eq. (0.002 kt) of N2O. Total emissions in 2022 are 4.7 percent greater than total emissions in 2021.
Total emissions from peatlands remaining peatlands have fluctuated between 0.6 and 1.3 MMT CO2 Eq. across the
time series with a decreasing trend from 1990 until 1993, followed by an increasing trend until reaching peak
emissions in 2000. After 2000, emissions generally decreased until 2006 and then increased until 2009. The trend
reversed in 2009 and total emissions have generally decreased between 2009 and 2021, however, total emissions
from peatlands increased slightly in 2022 compared to 2021. Carbon dioxide emissions from peatlands remaining
peatlands have fluctuated between 0.6 and 1.3 MMT CO2 across the time series, and these emissions drive the
trends in total emissions. Methane and N2O emissions remained close to zero across the time series.
Table 6-55: Emissions from Peatlands Remaining Peatlands (MMT CO2 Eq.)
Gas
1990
2005
2018
2019
2020
2021
2022
C02
1.1
1.1
0.7
0.6
0.6
0.5
0.6
Off-site
1.0
1-0 1
0.6
0.6
0.5
0.5
0.5
On-site
0.1
0.1
+
+
+
+
+
CH4 (On-site)
+
+
+
+
+
+
+
N20 (On-site)
+
+
+
+
+
+
+
Total
1.1
1.1
0.7
0.6
0.6
0.6
0.6
+ Does not exceed 0.05 MMT C02 Eq.
Note: Totals may not sum due to independent rounding.
Table 6-56: Emissions from Peatlands Remaining Peatlands (kt)
Gas
1990
2005
2018
2019
2020
2021
2022
C02
1,055
1,101
650
613
590
547
572
Off-site
985
1,030 !
608
572
550
509
533
On-site
70
71
42
41
41
38
39
CH4 (On-site)
+
+
+
+
+
+
+
N20 (On-site)
+
+
+
+
+
+
+
+ Does not exceed 0.5 kt.
Note: Totals may not sum due to independent rounding.
Land Use, Land-Use Change, and Forestry 6-103
-------
Methodology and Time-Series Consistency
Off-Site CO2 Emissions
Carbon dioxide emissions from domestic peat production were estimated using a Tier 1 methodology consistent
with IPCC (2006). Off-site CO2 emissions from peatlands remaining peatlands were calculated by apportioning the
annual weight of peat produced in the United States (Table 6-57) into peat extracted from nutrient-rich deposits
and peat extracted from nutrient-poor deposits using annual percentage-by-weight figures. These nutrient-rich
and nutrient-poor production values were then multiplied by the appropriate default C fraction conversion factor
taken from IPCC (2006) in order to obtain off-site emission estimates. For the conterminous 48 states, both annual
percentages of peat type by weight and domestic peat production data were sourced from estimates and industry
statistics provided in the Minerals Yearbook and Mineral Commodity Summaries from the U.S. Geological Survey
(USGS; USGS 1995 through 2018; USGS 2023a; USGS 2023b; USGS 2023c). Hawaii is assumed to have no peat
production due to its absence from these sources. To develop these data, the USGS (U.S. Bureau of Mines prior to
1997) obtained production and use information by surveying domestic peat producers. On average, about 75
percent of the peat operations respond to the survey; USGS estimates data for non-respondents on the basis of
prior-year production levels (Apodaca 2011).
The estimates for Alaska rely on reported peat production from the annual Alaska's Mineral Industry reports
(DGGS 1993 through 2015). Similar to the U.S. Geological Survey, the Alaska Department of Natural Resources,
Division of Geological & Geophysical Surveys (DGGS) solicits voluntary reporting of peat production from producers
for the Alaska's Mineral Industry report. However, the report does not estimate production for the non-reporting
producers, resulting in larger inter-annual variation in reported peat production from Alaska depending on the
number of producers who report in a given year (Szumigala 2011). In addition, in both the conterminous United
States and Alaska, large variations in peat production can also result from variation in precipitation and the
subsequent changes in moisture conditions, since unusually wet years can hamper peat production. The
methodology estimates emissions from Alaska separately from the conterminous United States because Alaska
previously conducted its own mineral surveys and reported peat production by volume, rather than by weight
(Table 6-58). However, volume production data were used to calculate off-site CO2 emissions from Alaska applying
the same methodology but with volume-specific C fraction conversion factors from IPCC (2006).58 Peat production
was not reported for 2015 in Alaska's Mineral Industry 2014 report (DGGS 2015), and reliable data are not
available beyond 2012, so Alaska's peat production in 2013 through 2021 (reported in cubic yards) was assumed to
be equal to the 2012 value.
Consistent with IPCC (2013) guidelines, off-site CO2 emissions from dissolved organic carbon were estimated based
on the total area of peatlands managed for peat extraction, which is calculated from production data using the
methodology described in the On-Site CO2 Emissions section below. Carbon dioxide emissions from dissolved
organic carbon were estimated by multiplying the area of managed peatlands by the default emission factor for
dissolved organic C provided in IPCC (2013).
The United States has largely imported peat from Canada for horticultural purposes; in 2022, imports of Sphagnum
moss (nutrient-poor) peat from Canada represented 96 percent of total U.S. peat imports and 80 percent of U.S.
domestic consumption (USGS 2023c). Most peat produced in the United States is reed-sedge peat, generally from
southern states, which is classified as nutrient-rich by IPCC (2006). To be consistent with the Tier 1 method, only
domestic peat production is accounted for when estimating off-site emissions. Higher-tier calculations of CO2
emissions from apparent consumption would involve consideration of the percentages of peat types stockpiled
(nutrient-rich versus nutrient-poor) as well as the percentages of peat types imported and exported.
58 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).
6-104 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
Table 6-57: Peat Production of Conterminous 48 States (kt)
Type of Deposit
1990
2005
2018
2019
2020
2021
2022
Nutrient-Rich
595.1
657.61
338.4
329.4
343.4
291.6
306.0
Nutrient-Poor
55.4
27.4|
50.6
36.6
10.6
32.4
34.0
Total Production
692.0
685.0
389.0
366.0
354.0
324.0
340.0
Note: Totals may not sum due to independent rounding.
Sources: United States Geological Survey (USGS) (1991-2017) Minerals Yearbook: Peat (1994-
2016); United States Geological Survey (USGS) (2018) Minerals Yearbook: Peat - Tables-only
release (2018); United States Geological Survey (USGS) (2023) Mineral Commodity Summaries:
Peat (2023).
Table 6-58: Peat Production of Alaska (Thousand Cubic Meters)
1990
2005
2018
2019
2020
2021
2022
Total Production
49.7|
47.81
93.1
93.1
93.1
93.1
93.1
Sources: Division of Geological & Geophysical Surveys (DGGS), Alaska Department of Natural
Resources (1997-2015) Alaska's Mineral Industry Report (1997-2014).
On-site CO2 Emissions
IPCC (2006) suggests basing the calculation of on-site emission estimates on the area of peatlands managed for
peat extraction differentiated by the nutrient type of the deposit (rich versus poor). Information on the area of
land managed for peat extraction is currently not available for the United States, but consistent with IPCC (2006),
an average production rate for the industry was applied to derive a land area estimate. In a mature industrialized
peat industry, such as exists in the United States and Canada, the vacuum method can extract up to 100 metric
tons per hectare per year (Cleary et al. 2005 as cited in IPCC 2006).59 The area of land managed for peat extraction
in the conterminous United States was estimated using both nutrient-rich and nutrient-poor production data and
the assumption that 100 metric tons of peat are extracted from a single hectare in a single year, see Table 6-59.
The annual land area estimates were then multiplied by the IPCC (2013) default emission factor in order to
calculate on-site CO2 emission estimates.
Production data are not available by weight for Alaska. In order to calculate on-site emissions resulting from
peatlands remaining peatlands in Alaska, the production data by volume were converted to weight using annual
average bulk peat density values, and then converted to land area estimates using the assumption that a single
hectare yields 100 metric tons, see Table 6-60. The IPCC (2006) on-site emissions equation also includes a term
that accounts for emissions resulting from the change in carbon stocks that occurs during the clearing of
vegetation prior to peat extraction. Area data on land undergoing conversion to peatlands for peat extraction is
also unavailable for the United States. However, USGS records show that the number of active operations in the
United States has been declining since 1990; therefore, it seems reasonable to assume that no new areas are being
cleared of vegetation for managed peat extraction. Other changes in carbon stocks in living biomass on managed
peatlands are also assumed to be zero under the Tier 1 methodology (IPCC 2006 and 2013).
Table 6-59: Peat Production Area of Conterminous 48 States (Hectares)
1990'
2005
2018
2019
2020
2021
2022
Nutrient-Rich
5,951 1
6,576 1
3,384
3,294
3,434
2,916
3,060
Nutrient-Poor
554 1
274 1
506
366
106
324
340
Total Production
6,920
6,850
3,890
3,660
3,540
3,240
3,400
59 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 6-105
-------
a A portion of the production in 1990 is of unknown nutrient type, resulting in a total production
value greater than the sum of nutrient-rich and nutrient-poor.
Note: Totals may not sum due to independent rounding.
Table 6-60: Peat Production Area of Alaska (Hectares)
1990
2005
2018
2019
2020
2021
2022
Nutrient-Rich
0
0
0
0
0
0
0
Nutrient-Poor
286 1
104 I
212
329
428
428
428
Total Production
286
104
212
329
428
428
428
Note: Totals may not sum due to independent rounding.
On-site N2O Emissions
IPCC (2006) indicates the calculation of on-site N2O emission estimates using Tier 1 methodology only considers
nutrient-rich peatlands managed for peat extraction. These area data are not available directly for the United
States, but the on-site CO2 emissions methodology above details the calculation of nutrient-rich area data from
production data. In order to estimate N2O emissions, the land area estimate of nutrient-rich peatlands remaining
peatlands was multiplied by the appropriate default emission factor taken from IPCC (2013). See the Planned
Improvements section for additional information on identified research activities to improve peatland land area
estimates.
On-site CH4 Emissions
IPCC (2013) also suggests basing the calculation of on-site Cm emission estimates on the total area of peatlands
managed for peat extraction. Area data is derived using the calculation from production data described in the On-
site CO2 Emissions section above. In order to estimate CH4 emissions from drained land surface, the land area
estimate of peatlands remaining peatlands was multiplied by the emission factor for direct Cm emissions taken
from IPCC (2013). In order to estimate CH4 emissions from drainage ditches, the total area of peatland was
multiplied by the default fraction of peatland area that contains drainage ditches, and the appropriate emission
factor taken from IPCC (2013). See Table 6-61 for the calculated area of ditches and drained land.
Table 6-61: Peat Production (Hectares)
1990
2005
2018
2019
2020
2021
2022
Conterminous 48 States
Area of Drained Land
6,574 I
6,508 1
3,696
3,477
3,363
3,078
3,230
Area of Ditches
346 1
343 |
195
183
177
162
170
Total Production
6,920
6,850
3,890
3,660
3,540
3,240
3,400
Alaska
Area of Drained Land
272 I
99 1
202
312
407
407
407
Area of Ditches
14 1
5
11
16
21
21
21
Total Production
286
104
212
329
428
428
428
Note: Totals may not sum due to independent rounding.
Methodological recalculations were applied to the entire time series to ensure time-series consistency from 1990
through 2022. The same data sources were used throughout the time series, when available. When data were
unavailable or the available data were outliers, missing values were estimated based on the past available data.
Uncertainty
A Monte Carlo (Approach 2) uncertainty analysis was applied to estimate the uncertainty of CO2, Cm, and N2O
emissions from peatlands remaining peatlands for 2022, using the following assumptions:
6-106 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
• The uncertainty associated with peat production data was estimated to be ± 25 percent (Apodaca 2008)
and assumed to be normally distributed.
• The uncertainty associated with peat production data stems from the fact that the USGS receives data
from smaller peat producers but estimates production from some larger peat distributors. The peat type
production percentages were assumed to have the same uncertainty values and distribution as the peat
production data (i.e., ± 25 percent with a normal distribution).
• The uncertainty associated with the reported production data for Alaska was assumed to be the same as
for the conterminous United States, or ± 25 percent with a normal distribution. It should be noted that
the DGGS estimates that around half of producers do not respond to their survey with peat production
data; therefore, the production numbers reported are likely to underestimate Alaska peat production
(Szumigala 2008).
• The uncertainty associated with the average bulk density values was estimated to be ± 25 percent with a
normal distribution (Apodaca 2008).
• IPCC (2006 and 2013) gives uncertainty values for the emissions factors for the area of peat deposits
managed for peat extraction based on the range of underlying data used to determine the emission
factors. The uncertainty associated with the emission factors was assumed to be triangularly distributed.
• The uncertainty values surrounding the C fractions were based on IPCC (2006) and the uncertainty was
assumed to be uniformly distributed.
• The uncertainty values associated with the fraction of peatland covered by ditches was assumed to be ±
100 percent with a normal distribution based on the assumption that greater than 10 percent coverage,
the upper uncertainty bound, is not typical of drained organic soils outside of The Netherlands (IPCC
2013).
The results of the Approach 2 quantitative uncertainty analysis are summarized in Table 6-62. Carbon dioxide
emissions from peatlands remaining peatlands in 2022 were estimated to be between 0.5 and 0.7 MMT CO2 Eq. at
the 95 percent confidence level. This indicates a range of 16 percent below to 16 percent above the 2022 emission
estimate of 0.6 MMT CO2 Eq. Methane emissions from peatlands remaining peatlands in 2022 were estimated to
be between 0.001 and 0.006 MMT CO2 Eq. This indicates a range of 59 percent below to 79 percent above the
2022 emission estimate of 0.004 MMT CO2 Eq. Nitrous oxide emissions from peatlands remaining peatlands in
2022 were estimated to be between 0.0002 and 0.0006 MMT CO2 Eq. at the 95 percent confidence level. This
indicates a range of 52 percent below to 53 percent above the 2022 emission estimate of 0.0004 MMT CO2 Eq.
Table 6-62: Approach 2 Quantitative Uncertainty Estimates for CO2, CH4, and N2O Emissions
from Peatlands Remaining Peatlands (MMT CO2 Eq. and Percent)
Source
Gas
2022 Emission Estimate
Uncertainty Range Relative to Emission Estimate"
(MMT CO . Eq.)
(MMT CO.
Eq.)
(%)
Lower
Upper
Lower
Upper
Bound
Bound
Bound
Bound
Peatlands Remaining Peatlands
C02
0.6
0.5
0.7
-16%
+16%
Peatlands Remaining Peatlands
ch4
+
+
+
-59%
+79%
Peatlands Remaining Peatlands
n2o
+
+
+
-52%
+53%
+ Does not exceed 0.05 MMT C02 Eq.
a Range of emission estimates predicted by Monte Carlo stochastic simulation for a 95 percent confidence interval.
QA/QC and Verification
A QA/QC analysis was performed to review input data and calculations, and no issues were identified. In addition,
the emission trends were analyzed to ensure they reflected activity data trends.
Land Use, Land-Use Change, and Forestry 6-107
-------
Recalculations Discussion
The conterminous United States peat production estimates for peatlands remaining peatlands were updated using
the Peat section of the Mineral Commodity Summaries 2023. The 2023 edition updated 2018, 2019, 2020, and
2021 peat production data and provided peat type production estimates for 2022. Updated data decreased
previously estimated emissions for 2018 by 18 percent, 2019 by 19 percent, 2020 by 19 percent, and 2021 by 22
percent versus estimated emissions for 2018, 2019, 2020, and 2021 in the previous (i.e., 1990 through 2021)
Inventory for peatlands remaining peatlands. According to USGS, peat production estimations for 2018 through
2022 were revised in the Mineral Commodity Summaries 2023 due to a company having shut down sometime in
2017 (USGS 2023d). Previously, USGS was estimating production for this company due to lack of peat production
survey responses.
Although Alaska peat production data for 2015 through 2022 were unavailable, 2014 data are available in the
Alaska's Mineral Industry 2014 report. However, the reported values represented an apparent 98 percent
decrease in production since 2012. Due to the uncertainty of the most recent data, 2013 through 2022 value were
assumed to be equal to the 2012 value, seen in the Alaska's Mineral Industry 2013 report. If updated Alaska data
are available for the next Inventory cycle, this will result in a recalculation in the next (i.e., 1990 through 2023)
Inventory report.
Planned Improvements
Edits to the trends and methodology sections are planned based on expert review comments.
EPA notes the following improvements may be implemented or investigated within the next two or three Inventory
cycles pending time and resource:
• The implied emission factors will be calculated and included in this chapter for future Inventories.
Currently, the N2O emissions calculation uses different land areas than the CO2 and Cm emission
calculations (see Methodology and Time Series Consistency in this chapter), so estimating the implied
emission factor per total land area is not appropriate. The inclusion of implied emission factors in this
chapter will provide another method of QA/QC and verification for Inventory data.
EPA notes the following improvements will continue to be investigated as time and resources allow, but there are
no immediate plans to implement until data are available or identified:
• In order to further improve estimates of CO2, N2O, and Cm emissions from peatlands remaining
peatlands, future efforts will investigate if improved data sources exist for determining the quantity of
peat harvested per hectare and the total area of land undergoing peat extraction.
• EPA plans to identify a new source for Alaska peat production. The current source has not been reliably
updated since 2012 and Alaska Department of Natural Resources indicated future publication of data has
been discontinued.
Coastal Wetlands Remaining Coastal Wetlands
Consistent with ecological definitions of wetlands,60 the United States has historically included under the category
of wetlands those coastal shallow water areas of estuaries and bays that lie within the extent of the Land
Representation. Guidance on quantifying greenhouse gas emissions and removals on coastal wetlands is provided
in the 2013 Supplement to the 2006IPCC Guidelines for National Greenhouse Gas Inventories: Wetlands (Wetlands
Supplement), which recognizes the particular importance of vascular plants in sequestering CO2 from the
atmosphere within biomass, dead organic material (DOM; including litter and dead wood stocks) and soils. Thus,
the Wetlands Supplement provides specific guidance on quantifying emissions and removals on organic and
60 See https://water.usgs.Eov/nwsum/WSP2425/definitioris.htm]; accessed August 2023.
6-108 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
mineral soils that are covered or saturated for part of the year by tidal fresh, brackish or saline water and are
vegetated by vascular plants and may extend seaward to the maximum depth of vascular plant vegetation. The
United States calculates emissions and removals based upon the stock change method for soil carbon (C) and the
gain-loss method for biomass and DOM. Presently, this Inventory does not calculate the lateral flux of carbon to or
from any land use. Lateral transfer of organic carbon to coastal wetlands and to marine sediments within U.S.
waters is the subject of ongoing scientific investigation; there is currently no IPCC methodological guidance for
lateral fluxes of carbon.
The United States recognizes both vegetated wetlands and unvegetated open water as coastal wetlands. Per
guidance provided by the Wetlands Supplement, sequestration of carbon into biomass, DOM and soil carbon pools
is recognized only in vegetated coastal wetlands and does not occur in unvegetated open water coastal wetlands.
The United States takes the additional step of recognizing that carbon stock losses occur when vegetated coastal
wetlands are converted to Unvegetated open water coastal wetlands.
This Inventory includes all privately- and publicly-owned coastal wetlands (i.e., mangroves and tidal marsh) along
the oceanic shores of the conterminous United States, including the District of Columbia., but does not include
coastal wetlands remaining coastal wetlands in Alaska, Hawaii, or any of the United States Territories. Seagrasses
are not currently included within the Inventory due to insufficient data on distribution, change through time and
carbon stocks or carbon stock changes as a result of anthropogenic influence (see Planned Improvements).
Under the coastal wetlands remaining coastal wetlands category, the following emissions and removals are
quantified in this chapter:
1) Carbon stock changes and Cm emissions on vegetated coastal wetlands remaining vegetated coastal
wetlands,
2) Carbon stock changes on vegetated coastal wetlands converted to unvegetated open water coastal
wetlands,
3) Carbon stock changes on unvegetated open water coastal wetlands converted to vegetated coastal
wetlands, and
4) Nitrous oxide emissions from aquaculture in coastal wetlands.
Vegetated coastal wetlands hold carbon in all five carbon pools (i.e., aboveground biomass, belowground biomass,
dead organic matter [DOM; dead wood and litter], and soil), though typically soil carbon and, to a lesser extent,
aboveground and belowground biomass are the dominant pools, depending on wetland type (i.e., forested vs.
marsh), vegetated coastal wetlands are net accumulators of carbon over centuries to millennia as soils accumulate
carbon under anaerobic soil conditions and carbon accumulates in plant biomass. Large emissions from soil carbon
and biomass stocks occur when vegetated coastal wetlands are converted to unvegetated open water coastal
wetlands (e.g., when vegetated coastal wetlands are lost due to subsidence, channel cutting through vegetated
coastal wetlands), but are still recognized as coastal wetlands in this Inventory. These carbon stock losses resulting
from conversion to unvegetated open water coastal wetlands can cause the release of decades to centuries of
accumulated soil carbon, as well as the standing stock of biomass carbon. Conversion of unvegetated open water
coastal wetlands to vegetated coastal wetlands, either through restoration efforts or naturally, initiates the
building of carbon stocks within soils and biomass. In applying the Wetlands Supplement methodologies for
estimating Cm emissions, coastal wetlands in salinity conditions greater than 18 parts per thousand have little to
no Cm emissions compared to those experiencing lower salinity brackish and freshwater conditions. Therefore,
conversion of vegetated coastal wetlands to or from unvegetated open water coastal wetlands are conservatively
assumed to not result in a change in salinity condition and are assumed to have no impact on CH4 emissions. The
Wetlands Supplement provides methodologies to estimate N2O emissions from coastal wetlands that occur due to
aquaculture. The N2O emissions from aquaculture result from the nitrogen derived from consumption of the
applied food stock that is then excreted as nitrogen load available for conversion to N2O. While N2O emissions can
also occur due to anthropogenic nitrogen loading from the watershed and atmospheric deposition, these
Land Use, Land-Use Change, and Forestry 6-109
-------
emissions are not reported here to avoid double-counting of indirect N2O emissions with the agricultural soils
management, forest land and settlements categories.
The Wetlands Supplement provides methodologies for estimating carbon stock changes and CH4 emissions from
mangroves, tidal marshes and seagrasses. Depending upon their height and area, carbon stock changes from
mangroves may be reported under the forest land category or under coastal wetlands. If mangrove stature is 5 m
or greater or if there is evidence that trees can obtain that height, mangroves are reported under the forest land
category because they meet the definition of forest land. Mangrove forests that are less than 5 m are reported
under coastal wetlands because they meet the definition of wetlands. All other non-drained, intact coastal
marshes are reported under coastal wetlands.
Because of human activities and level of regulatory oversight, all coastal wetlands within the conterminous United
States are included within the managed land area described in Section 6.1, and as such, estimates of carbon stock
changes, emissions of CH4, and emissions of N2O from aquaculture from all coastal wetlands are included in this
Inventory. At the present stage of inventory development, coastal wetlands are not explicitly shown in the land
representation analysis while work continues to harmonize data from NOAA's Coastal Change Analysis Program (C-
CAP)61 with NRI, FIA and NLDC data used to compile the land representation (see Section 6.1). However, a check
was undertaken to confirm that coastal wetlands recognized by C-CAP represented a subset of wetlands
recognized by the NRI for marine coastal states.
The greenhouse gas fluxes for all four wetland categories described above are summarized in Table 6-63. Coastal
wetlands remaining coastal wetlands are generally a net carbon sink, with the fluxes ranging from -5.6 to -6.7 MMT
CO2 Eq. across the majority of the time series; however, between 2006 and 2010, they were a net source of
emissions (ranging from 3.2 to 53.5 MMT CO2 Eq.), resulting from a large loss of vegetated coastal wetlands to
open water due to hurricanes (Table 6-63). Recognizing removals of CChto soil of 12.5 MMT CO2 Eq. and Cm
emissions of 4.3 MMT CO2 Eq. in 2022, vegetated coastal wetlands remaining vegetated coastal wetlands are a net
sink of 8.2MMT CO2 Eq. Loss of coastal wetlands, primarily in the Mississippi Delta as a result of hurricane impacts
and sediment diversion and other human impacts, recognized as vegetated coastal wetlands converted to
unvegetated coastal wetlands, drive an emission of 1.5 MMT CO2 Eq. since 2011, primarily from soils. Building of
new wetlands from open water, recognized as unvegetated coastal wetlands converted to vegetated coastal,
results each year in removal of 0.1 MMT CO2 Eq. Aquaculture is a minor industry in the United States, resulting in
an emission of N2O across the time series of between 0.1 to 0.2 MMT CO2 Eq. In total, coastal wetlands are a net
sink of 6.7 MMT C02 Eq. in 2022.
Table 6-63: Emissions and Removals from Coastal Wetlands Remaining Coastal Wetlands
(MMT C02 Eq.)
Land Use/Carbon Pool
1990
2005
2018
2019
2020
2021
2022
Vegetated Coastal Wetlands Remaining
i
Vegetated Coastal Wetlands
(8.4)
(8.4)
(8.3)
(8.3)
(8.3)
(8.2)
(8.2)
Biomass C Flux
Ml
Will
(+)
(+)
(+)
(+)
(+)
Soil C Flux
(12.5)
(12.6)
(12.5)
(12.5)
(12.5)
(12.5)
(12.5)
Net CH4 Flux
4-21
4-21
4.3
4.3
4.3
4.3
4.3
Vegetated Coastal Wetlands Converted
to Unvegetated Open Water Coastal
Wetlands
1.8
2.6
1.5
1.5
1.5
1.5
1.5
Biomass C Flux
0.1 j
0.l|
0.1
0.1
0.1
0.1
0.1
Dead Organic Matter C Flux
+
+
+
+
+
+
+
Soil C Flux
1.7S
2.511
1.5
1.5
1.5
1.5
1.5
Unvegetated Open Water Coastal
Wetlands Converted to Vegetated
(+)1
(0.1)1
(0.1)
(0.1)
(0.1)
(0.1)
(0.1)
61 See https://coast.noaa.gov/digitalcoast/tools/lca.html; accessed September 2023.
6-110 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
Coastal Wetlands
Biomass C Flux
(+)
(0.1)
(0.1)
(0.1)
(0.1)
(0.1)
(0.1)
Dead Organic Matter C Flux
(+)
(+)
+
+
+
+
+
Soil C Flux
(+)
(+)
(+)
(+)
(+)
(+)
(+)
Net N20 Flux from Aquaculture in
Coastal Wetlands
0.1
0.2
0.1
0.1
0.1
0.1
0.1
Total Biomass C Flux
+
+
(0.1)
(0.1)
(0.1)
(0.1)
(0.1)
Total Dead Organic Matter C Flux
(+)
(+)
+
+
+
+
+
Total Soil C Flux
(10.8)
(10.1)
(11.0)
(11.0)
(11.0)
(11.0)
(11.1)
Total CH4 Flux
4.2
4.2
4.3
4.3
4.3
4.3
4.3
Total N20 Flux
0.1
0.2
0.1
0.1
0.1
0.1
0.1
Total Flux
(6.5)
(5.7)
(6.7)
(6.7)
(6.7)
(6.7)
(6.7)
+ Absolute value does not exceed 0.05 MMT C02 Eq.
Note: Totals may not sum due to independent rounding. Parentheses indicate net sequestration.
Emissions and Removals from Vegetated Coastal Wetlands
Remaining Vegetated Coastal Wetlands
The conterminous United States currently has 2.98 million hectares of intertidal vegetated coastal wetlands
remaining vegetated coastal wetlands comprised of tidally influenced palustrine emergent marsh (663,014 ha),
palustrine scrub shrub (133,582 ha) and estuarine emergent marsh (1,892,507 ha), estuarine scrub shrub (95,225
ha) and estuarine forested wetlands (195,199 ha). Mangroves fall under both estuarine forest and estuarine scrub
shrub categories depending upon height. Dwarf mangroves, found in subtropical states along the Gulf of Mexico,
do not attain the height status to be recognized as forest land, and are therefore always classified within vegetated
coastal wetlands, vegetated coastal wetlands remaining vegetated coastal wetlands are found in cold temperate
(53,968 ha), warm temperate (896,583 ha), subtropical (1,966,101 ha) and Mediterranean (62,874 ha) climate
zones.
Soils are the largest carbon pool in vegetated coastal wetlands remaining vegetated coastal wetlands, reflecting
long-term removal of atmospheric CO2 by vegetation and transfer into the soil pool in the form of both
autochthonous and allochthonous decaying organic matter. Soil carbon emissions are not assumed to occur in
coastal wetlands that remain vegetated. This Inventory includes changes in carbon stocks in both biomass and
soils. Changes in DOM carbon stocks are not included. Methane emissions from decomposition of organic matter
in anaerobic conditions are present at salinity less than half that of sea water. Mineral and organic soils are not
differentiated in terms of carbon stock changes or Cm emissions.
Table 6-64 through Table 6-66 summarize nationally aggregated biomass and soil carbon stock changes and CH4
emissions on vegetated coastal wetlands remaining vegetated coastal wetlands. Intact vegetated coastal wetlands
remaining vegetated coastal wetlands hold a total biomass carbon stock of 35.96 MMT C. Removals from biomass
carbon stocks in 2022 were 0.05 MMT CO2 Eq. (0.01 MMT C), which has increased over the time series (Table 6-64
and Table 6-65). Carbon dioxide emissions from biomass in vegetated coastal wetlands remaining vegetated
coastal wetlands between 2002 and 2011, with very low sequestration between 2002 and 2006 and emissions of
0.21 MMT CO2 Eq. between 2007 and 2011, are not inherently typical and are a result of coastal wetland loss over
time. Most of the coastal wetland loss has occurred in palustrine and estuarine emergent wetlands. Vegetated
coastal wetlands maintain a large carbon stock within the top 1 meter of soil (estimated to be 804 MMT C) to
which carbon accumulated at a rate of 12.5 MMT CO2 Eq. (3.4 MMT C) in 2022, a value that has remained
relatively constant across the reporting period. For vegetated coastal wetlands remaining vegetated coastal
wetlands, methane emissions of 4.3 of MMT CO2 Eq. (154 kt CH4) in 2022 (Table 6-66) offset carbon removals
resulting in a net removal of 8.2 MMT CO2 Eq. in 2022; this rate has been relatively consistent across the reporting
period. Dead organic matter stock changes are not calculated in vegetated coastal wetlands remaining vegetated
coastal wetlands since this stock is considered to be in a steady state when using Tier 1 methods (IPCC 2014). Due
to federal regulatory protection, loss of vegetated coastal wetlands through human activities slowed considerably
Land Use, Land-Use Change, and Forestry 6-111
-------
in the 1970s and the current annual rates of carbon stock change and Cm emissions are relatively constant over
time.
Table 6-64: Net CO2 Flux from Carbon Stock Changes in Vegetated Coastal Wetlands
Remaining Vegetated Coastal Wetlands (MMT CO2 Eq.)
Year
1990
2005
2018
2019
2020
2021
2022
Biomass Flux
(+)l
(+)1
(+)
(+)
(+)
(+)
(+)
Soil Flux
(12.5)1
(12.6)1
(12.5)
(12.5)
(12.5)
(12.5)
(12.5)
Total C Stock Change
(12.6)
(12.6)
(12.5)
(12.5)
(12.5)
(12.5)
(12.5)
+ Absolute value does not exceed 0.05 MMT C02 Eq.
Note: Parentheses indicate net sequestration. Totals may not sum due to independent rounding.
Table 6-65: Net CO2 Flux from Carbon Stock Changes in Vegetated Coastal Wetlands
Remaining Vegetated Coastal Wetlands (MMT C)
Year
1990
2005
2018
2019
2020
2021
2022
Biomass Flux
(+)
+ 1
(+)
(+)
(+)
(+)
(+)
Soil Flux
(3.4)
I (3.4)1
(3.4)
(3.4)
(3.4)
(3.4)
(3.4)
Total C Stock Change
(3.4)
(3.4)
(3.4)
(3.4)
(3.4)
(3.4)
(3.4)
+ Absolute value does not exceed 0.05 MMT C.
Note: Parentheses indicate net sequestration. Totals may not sum due to independent rounding.
Table 6-66: CH4 Emissions from Vegetated Coastal Wetlands Remaining Vegetated Coa:
Wetlands (MMT CO2 Eq. and kt CH4)
Year
1990
2005
2018
2019
2020
2021
2022
Methane Emissions (MMT C02 Eq.)
4.2
4.21
4.3
4.3
4.3
4.3
4.3
Methane Emissions (kt CH4)
149
1511
153
153
154
154
154
Methodology and Time-Series Consistency
The following section includes a description of the methodology used to estimate changes in biomass carbon
stocks, soil carbon stocks and emissions of Cm for vegetated coastal wetlands remaining vegetated coastal
wetlands. Dead organic matter is not calculated for vegetated coastal wetlands remaining vegetated coastal
wetlands since it is assumed to be in steady state (IPCC 2014).
Methodological recalculations were applied to the entire time series to ensure time-series consistency from 1990
through 2022.
Biomass Carbon Stock Changes
Above- and belowground biomass carbon stocks for palustrine (freshwater) and estuarine (saline) marshes are
estimated for vegetated coastal wetlands remaining vegetated coastal wetlands on land below the elevation of
high tides (taken to be mean high water spring tide elevation) and as far seawards as the extent of intertidal
vascular plants according to the national LiDAR dataset, the national network of tide gauges and land use histories
recorded in the 1996, 2001, 2006, 2010, and 2016 NOAA C-CAP surveys (NOAA OCM 2020). C-CAP areas are
calculated at the state/territory level and summed according to climate zone to national values. Federal and non-
federal lands are represented. Trends in land cover change are extrapolated to 1990 and 2022 from these datasets.
Based upon NOAA C-CAP, coastal wetlands are subdivided into palustrine and estuarine classes and further
subdivided into emergent marsh, scrub shrub and forest classes (Table 6-67). Biomass is not sensitive to soil
organic matter content but is differentiated based on climate zone. Aboveground biomass carbon stocks for non-
forested wetlands data are derived from a national assessment combining field plot data and aboveground
biomass mapping by remote sensing (Byrd et al. 2017; Byrd et al. 2018; Byrd et al. 2020). The aboveground
6-112 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
biomass carbon stock for subtropical estuarine forested wetlands (dwarf mangroves that are not classified as
forests due to their stature) is derived from a meta-analysis by Lu and Megonigal (2017). Root to shoot ratios from
the Wetlands Supplement (Table 6-69; IPCC 2014) were used to account for belowground biomass, which were
multiplied by the aboveground carbon stock. Above- and belowground values were summed to obtain total
biomass carbon stocks. Biomass carbon stock changes per year for wetlands remaining wetlands were determined
by calculating the difference in area between that year and the previous year to calculate gain/loss of area for each
climate type, which was multiplied by the mean biomass for that climate type.
Table 6-67: Area of Vegetated Coastal Wetlands Remaining Vegetated Coastal Wetlands,
Vegetated Coastal Wetlands Converted to Unvegetated Open Water Coastal Wetlands, and
Unvegetated Open Water Coastal Wetlands Converted to Vegetated Coastal Wetlands (ha)
Year 1990 2005 2018 2019 2020 2021 2022
Vegetated Coastal Wetlands
Remaining Vegetated Coastal
Wetlands 2,975,477 2,985,783 2,974,523 2,975,789 2,977,055 2,978,322 2,979 588
Vegetated Coastal Wetlands
Converted to Unvegetated Open
Water Coastal Wetlands 1,720
Unvegetated Open Water Coastal
Wetlands Converted to
Vegetated Coastal Wetlands 9521 1,7691 2,406 2,406 2,406 2,406 2,406
Table 6-68: Aboveground Biomass Carbon Stocks for Vegetated Coastal Wetlands (t C ha1)
1
I
2,5151 1,488 1,488 1,488 1,488 1,488
Climate Zone
Wetland Type
Cold Temperate
Warm Temperate
Subtropical
Mediterranean
Palustrine Scrub/Shrub Wetland
3.25
3.17
2.24
4.69
Palustrine Emergent Wetland
3.25
3.17
2.24
4.69
Estuarine Forested Wetland
N/A
N/A
17.83
N/A
Estuarine Scrub/Shrub Wetland
3.05
3.05
2.43
3.44
Estuarine Emergent Wetland
3.05
3.10
2.43
3.44
Source: All data from Byrd et al. (2017, 2018 and 2020) except for subtropical estuarine forested wetlands, which is from
Lu and Megonigal (2017); N/A means there are currently no estuarine forested wetlands that are less than 5 meters tall;
these forested wetlands meet the definition of forest land and are included in the Forest Land section.
Table 6-69: Root to Shoot Ratios for Vegetated Coastal Wetlands
Climate Zone
Wetland Type
Cold Temperate
Warm Temperate
Subtropical
Mediterranean
Palustrine Scrub/Shrub Wetland
1.15
1.15
3.65
3.63
Palustrine Emergent Wetland
1.15
1.15
3.65
3.63
Estuarine Forested Wetland
N/A
N/A
0.96
N/A
Estuarine Scrub/Shrub Wetland
2.11
2.11
3.65
3.63
Estuarine Emergent Wetland
2.11
2.11
3.65
3.63
Source: All values from IPCC (2014); N/A means there are currently no estuarine forested wetlands that are less than 5
meters tall; these forested wetlands meet the definition of forest land and are included in the Forest Land section.
Soil Carbon Stock Changes
Soil carbon stock changes are estimated for vegetated coastal wetlands remaining vegetated coastal wetlands for
both mineral and organic soils. Soil carbon stock changes, stratified by climate zones and wetland classes, are
derived from a synthesis of peer-reviewed literature (Table 6-70; Lynch 1989; Orson et al. 1990; Kearny &
Stevenson 1991; Thorn et al. 1992; Roman et al. 1997; Craft et al. 1998; Orson et al. 1998; Merrill 1999; Weis et al.
Land Use, Land-Use Change, and Forestry 6-113
-------
2001; Hussein et al. 2004; Church et al. 2006; Koster et al. 2007; Drexler et al. 2009; Boyd 2012; Callaway et al.
2012a&b; Bianchi et al. 2013; Drexler et al. 2013; Watson and Byrne 2013; Breithaupt et al. 2014; Crooks et al.
2014; Weston et al. 2014; Smith et al. 2015; Villa & Mitsch 2015; Boyd and Sommerfield 2016; Marchio et al. 2016;
Noe et al. 2016; Arriola and Cable 2017; Boyd et al. 2017; Gerlach et al. 2017; Giblin and Forbrich 2018; Krauss et
al. 2018; Abbott et al. 2019; Drexler et al. 2019; Poppe and Rybczyk 2019; Ensign et al. 2020; Kemp et al. 2020;
Lagomasino et al. 2020; Luk et al. 2020; McTigue et al. 2020; Peck et al. 2020; Vaughn et al. 2020; Weston et al.
2020; Arias-Ortiz et al. 2021; Baustian et al. 2021; Allen et al. 2022; Miller et al. 2022).
Tier 2 estimates of soil carbon removals associated with annual soil carbon accumulation on managed vegetated
coastal wetlands remaining vegetated coastal wetlands were developed with country-specific soil carbon removal
factors multiplied by activity data of land area for vegetated coastal wetlands remaining vegetated coastal
wetlands. The methodology follows Eq. 4.7, Chapter 4 of the Wetlands Supplement, and is applied to the area of
vegetated coastal wetlands remaining vegetated coastal wetlands on an annual basis. To estimate soil carbon
stock changes, no differentiation is made between organic and mineral soils since currently, no statistical evidence
supports disaggregation (Holmquist et al. 2018).
Table 6-70: Annual Soil Carbon Accumulation Rates for Vegetated Coastal Wetlands (t C ha 1
yr1)
Climate Zone
Cold Temperate
Warm Temperate
Subtropical
Mediterranean
Palustrine Scrub/Shrub Wetland
1.010
1.544
0.45
0.845
Palustrine Emergent Wetland
1.010
1.544
0.454
0.845
Estuarine Forested Wetland
N/A
N/A
0.821
N/A
Estuarine Scrub/Shrub Wetland
1.254
1.039
0.821
0.845
Estuarine Emergent Wetland
1.254
1.039
1.587
0.845
Source: All data from CCRCN (2023)62; N/A means there are no estuarine forested wetlands outside of subtropical regions.
Soil Methane Emissions
Tier 1 estimates of CH4 emissions for vegetated coastal wetlands remaining vegetated coastal wetlands are derived
from the same wetland map used in the analysis of wetland soil C fluxes, produced from C-CAP, LiDAR and tidal
data, in combination with default Cl-U emission factors provided in Table 4.14 of the Wetlands Supplement. The
methodology follows Equation 4.9, Chapter 4 of the Wetlands Supplement; Tier 1 emissions factors are multiplied
by the area of freshwater (palustrine) coastal wetlands. The Cl-U fluxes applied are determined based on salinity;
only palustrine wetlands are assumed to emit Cl-U. Estuarine coastal wetlands in the C-CAP classification include
wetlands with salinity less than 18 ppt, a threshold at which methanogenesis begins to occur (Poffenbarger et al.
2011), but the dataset currently does not differentiate estuarine wetlands based on their salinities and, as a result,
Cm emissions from estuarine wetlands are not included at this time.
Uncertainty
Underlying uncertainties in the estimates of soil and biomass carbon stock changes and Cl-U emissions include
uncertainties associated with Tier 2 literature values of soil carbon stocks, biomass carbon stocks and Cl-U flux,
assumptions that underlie the methodological approaches applied and uncertainties linked to interpretation of
remote sensing data. Uncertainty specific to vegetated coastal wetlands remaining vegetated coastal wetlands
include differentiation of palustrine and estuarine community classes, which determines the soil carbon stock and
Cl-U flux applied. Uncertainties for soil and biomass carbon stock data for all subcategories are not available and
thus assumptions were applied using expert judgment about the most appropriate assignment of a carbon stock to
a disaggregation of a community class. Because mean soil and biomass carbon stocks for each available community
62 Coastal Carbon Network (2023). Database: Coastal Carbon Library (Version 1.0.0). Smithsonian Environmental Research
Center. Dataset. https://doi.org/10.25573/serc.21565671. Accessed September 2023.
6-114 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
class are in a fairly narrow range, the same overall uncertainty was assigned to each, respectively (i.e., applying
approach for asymmetrical errors, the largest uncertainty for any soil carbon stock value should be applied in the
calculation of error propagation; IPCC 2000). Uncertainty for root to shoot ratios, which are used for quantifying
belowground biomass, are derived from the 2013 Wetlands Supplement. Uncertainties for Cm flux are the Tier 1
default values reported in the 2013 IPCC Wetlands Supplement. Overall uncertainty of the NOAA C-CAP remote
sensing product is 15 percent. This is in the range of remote sensing methods (±10 to 15 percent; IPCC 2003).
However, there is significant uncertainty in salinity ranges for tidal and non-tidal estuarine wetlands and activity
data used to apply Cm flux emission factors (delineation of an 18 ppt boundary) that will need significant
improvement to reduce uncertainties. Details on the emission/removal trends and methodologies through time
are described in more detail in the introduction and the Methodology section. The combined uncertainty was
calculated using the IPCC Approach 1 method of summing the squared uncertainty for each individual source (C-
CAP, soil, biomass and Cm) and taking the square root of that total.
Uncertainty estimates are presented in Table 6-71 for each subcategory (i.e., soil carbon, biomass carbon and Cm
emissions). The combined uncertainty across all subcategory is 37.0 percent below and above the estimate of -6.4
MMT CO2 Eq, which is primarily driven by the uncertainty in the CH4 estimates because there is high variability in
Cm emissions when the salinity is less than 18 ppt. In 2021, the total flux was -8.2 MMT CO2 Eq., with lower and
upper estimates of -11.3 and -5.2 MMT CO2 Eq.
Table 6-71: IPCC Approach 1 Quantitative Uncertainty Estimates for Carbon Stock Changes
and CH4 Emissions occurring within Vegetated Coastal Wetlands Remaining Vegetated Coastal
Wetlands in 2021 (MMT CO2 Eq. and Percent)
Source/Sink
Gas
2022 Estimate
(MMT CO . Eq.)
Uncertainty Range Relative to Estimate
(MMTCO' Eq.) (%)
Lower
Upper
Lower
Upper
Bound
Bound
Bound
Bound
Biomass C Stock Change
C02
(0.05)
(0.06)
(0.03)
-24.1%
+24.1%
Soil C Stock Change
C02
(12.5)
(14.7)
(10.3)
-17.7
+17.7%
CH4 emissions
ch4
4.3
3.0
5.6
-29.9%
+29.9%
Total Flux
(8.2)
(11.3)
(5.2)
-36.5%
+36.5%
Note: Parentheses indicate net sequestration. Totals may not sum due to independent rounding.
QA/QC and Verification
NOAA provided the National LiDAR Dataset, tide data, and C-CAP land cover and land cover change mapping, all of
which are subject to agency internal QA/QC assessment consistent with the general QC checks outlined in the
Inventory QA/QC Plan. Acceptance of final datasets into archive and dissemination are contingent upon the
product compilation being compliant with mandatory QA/QC requirements (McCombs et al. 2016). QA/QC and
verification of soil carbon stock datasets have been provided by the Smithsonian Environmental Research Center
and coastal wetland inventory team leads who reviewed summary tables against reviewed sources. Biomass
carbon stocks are derived from peer-review literature and reviewed by the U.S. Geological Survey prior to
publishing, by the peer-review process during publishing, and by the coastal wetland inventory team leads before
inclusion in this Inventory. A team of two evaluated and verified there were no computational errors within the
calculation worksheets. Soil and biomass carbon stock change data are based upon peer-reviewed literature and
CH4 emission factors derived from the Wetlands Supplement.
Recalculations Discussion
A recalculation of emission factors for soil carbon accretion rates was performed using the same methodology and
criteria as in Lu and Megonigal (2017) and described above. This new analysis incorporated data published since
2016 and other relevant data that were not previously included. Table 6-70 shows the new values. The updated
synthesis resulted in a general increase in soil carbon accumulation rates for estuarine emergent and scrub/shrub
Land Use, Land-Use Change, and Forestry 6-115
-------
wetlands, which resulted in an annual average increase of removals of 2.3 MMT CO2 Eq. for the entire time series.
For vegetated coastal wetlands remaining vegetated coastal wetlands in 2022, inclusion of the updated values
resulted in an increase of the sink from -5.9 MMT CO2 Eq. to -8.2 MMT CO2 Eq.
Planned Improvements
Harmonization across all spatial datasets used to calculate activity data is underway. Once completed, a better
representation of forested tidal wetlands, palustrine tidal wetlands, and forest land near the tidal boundary will be
obtained.
Work is currently underway to examine the feasibility of incorporating seagrass soil and biomass carbon stocks into
the vegetated coastal wetlands remaining vegetated coastal wetlands estimates. Additionally, investigation into
quantifying the distribution, area, and emissions resulting from impounded waters (i.e., coastal wetlands where
tidal connection to the ocean has been restricted or eliminated completely) is underway.
Box 6-6: State-Level Case Studies for the Estimation of GHG Removals in Seagrasses
North Carolina and Maryland are the first states to include seagrasses within their state-level inventory. North
Carolina has the largest extent of seagrass coverage along the U.S. Atlantic coast, measuring approximately
86,412 acres in 2021. Seagrass mapping efforts occurred in 2007, 2013, and 2020 using a field-validated aerial
image classification. The Tier 1 soil carbon accumulation rate was used and currently, biomass is not included
due to lack of local data. The analysis shows that these high salinity seagrass habitats provided a net carbon sink
to the state, although greenhouse gas removals decreased over time due to loss in seagrass coverage. Overall,
seagrass beds in 2021 sequestered approximately 0.055 MMT CO2 Eq. (55.14 kt CO2 Eq.) in the soils alone.
In Maryland, the state greenhouse gas inventory comprises blue carbon stocks and fluxes from estuarine
wetlands and seagrasses. Maryland currently has long-term monitoring of submerged aquatic vegetation (SAV)
extent and density through annual surveying, and the rate of carbon sequestration and methane emission was a
regional average for coastal wetlands. This study at state-level calculation offers an opportunity to maintain
consistency in reporting across spatial scales and allows positioning SAV in its role as a carbon sink, in addition
to its benefits in water quality and habitat conservation, perpetuating Maryland's role as a leader in blue carbon
accounting.
These two case studies demonstrate the importance of refining emission factor data and harmonizing the
inclusion of this ecosystem in the land representation analysis (reconciling the National Ocean and Atmospheric
Administration [NOAA] Coastal Change Analysis Program [C-CAP] data with the National Resource Inventory,
Forest Inventory Analysis, and the National Land Cover Database).
Emissions from Vegetated Coastal Wetlands Converted to
Unvegetated Open Water Coastal Wetlands
Vegetated coastal wetlands converted to unvegetated open water coastal wetlands is a source of emissions from
soil, biomass, and DOM carbon stocks. An estimated 1,488 ha of vegetated coastal wetlands were converted to
unvegetated open water coastal wetlands in 2022, which largely occurred within estuarine and palustrine
emergent wetlands. Prior to 2006, annual conversion to unvegetated open water coastal wetlands was higher than
current rates: 1,720 between 1990 and 2000 and 2,515 ha between 2001 and 2005. The Mississippi Delta
represents more than 40 percent of the total coastal wetland of the United States, and over 90 percent of the area
of vegetated coastal wetlands converted to unvegetated open water coastal wetlands. The drivers of coastal
wetlands loss include legacy human impacts on sediment supply through rerouting river flow, direct impacts of
channel cutting on hydrology, salinity and sediment delivery, and accelerated subsidence from aquifer extraction.
6-116 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
Each of these drivers directly contributes to wetland erosion and subsidence, while also reducing the resilience of
the wetland to build with sea-level rise or recover from hurricane disturbance. Over recent decades, the rate of
Mississippi Delta wetland loss has slowed, though episodic mobilization of sediment occurs during hurricane
events (Couvillion et al. 2011; Couvillion et al. 2016). The land cover analysis between the 2006 and 2011C-CAP
surveys coincides with two such events, hurricanes Katrina and Rita (both making landfall in the late summer of
2005), that occurred between these C-CAP survey dates. The subsequent 2016 C-CAP survey determined that
erosion rates had slowed.
Shallow nearshore open water within the U.S. land representation is recognized as falling under the coastal
wetlands category within this Inventory. While high resolution mapping of coastal wetlands provides data to
support IPCC Approach 2 methods for tracking land cover change, the depth in the soil profile to which sediment is
lost is less clear. This Inventory adopts the Tier 1 methodological guidance from the Wetlands Supplement for
estimating emissions following the methodology for excavation (see Methodology section, below) when vegetated
coastal wetlands are converted to unvegetated open water coastal wetlands, assuming aim depth of disturbed
soil. This 1 m depth of disturbance is consistent with estimates of wetland carbon loss provided in the literature
and the Wetlands Supplement (Crooks et al. 2009; Couvillion et al. 2011; Delaune and White 2012; IPCC 2014). The
same assumption on depth of soils impacted by erosion has been applied here. It is a reasonable Tier 1
assumption, based on experience, but estimates of emissions are sensitive to the depth to which the assumed
disturbances have occurred (Holmquist et al. 2018). A Tier 1 assumption is also adopted in that all mobilized
carbon is immediately returned to the atmosphere (as assumed for terrestrial land-use categories), rather than
redeposited in long-term carbon storage. The science is currently under evaluation to adopt more refined
emissions factors for mobilized coastal wetland carbon based upon the geomorphic setting of the depositional
environment.
In 2022, there were 1,488 ha of vegetated coastal wetlands converted to unvegetated open water coastal
wetlands (Table 6-67) across all wetland types and climates, which resulted in 1.5 MMT CO2 Eq. (0.4 MMT C) and
0.06 MMT CO2 Eq. (0.02 MMT C) lost through soil and biomass, respectively, with minimal DOM C stock loss (Table
6-72, and Table 6-73). Across the reporting period, the area of vegetated coastal wetlands converted to
unvegetated open water coastal wetlands was greatest between the 2006 to 2011 C-CAP reporting period (11,373
ha) and has decreased since then to current levels (Table 6-67).
Table 6-72: Net CO2 Flux from Carbon Stock Changes in Vegetated Coastal Wetlands
Converted to Unvegetated Open Water Coastal Wetlands (MMT CO2 Eq.)
Year
1990
2005
2018
2019
2020
2021
2022
Biomass Flux
0.11
0.11
0.1
0.1
0.1
0.1
0.1
Dead Organic Matter Flux
+
+ iiiiiii
+
+
+
+
+
Soil Flux
1.7
2.5
1.5
1.5
1.5
1.5
1.5
Total C Stock Change
1.8
2.6
1.5
1.5
1.5
1.5
1.5
+ Absolute value does not exceed 0.05 MMT C02 Eq.
Note: Totals may not sum due to independent rounding.
Table 6-73: Net CO2 Flux from Carbon Stock Changes in Vegetated Coastal Wetlands
Converted to Unvegetated Open Water Coastal Wetlands (MMT C)
Year
1990
2005
2018
2019
2020
2021
2022
Biomass Flux
+
+
+
+
+
+
+
Dead Organic Matter Flux
+1
+
+
+
+
+
Soil Flux
0.5
0.7
0.4
0.4
0.4
0.4
0.4
Total C Stock Change
0.5
0.7
0.4
0.4
0.4
0.4
0.4
+ Absolute value does not exceed 0.05 MMT C.
Note: Totals may not sum due to independent rounding.
Land Use, Land-Use Change, and Forestry 6-117
-------
Methodology and Time-Series Consistency
The following section includes a brief description of the methodology used to estimate changes in soil, biomass
and DOM carbon stocks for vegetated coastal wetlands converted to unvegetated open water coastal wetlands.
Methodological recalculations were applied to the entire time series to ensure time-series consistency from 1990
through 2022.
Biomass Carbon Stock Changes
Biomass carbon stock changes for palustrine and estuarine marshes are estimated for vegetated coastal wetlands
converted to unvegetated open water coastal wetlands on lands below the elevation of high tides (taken to be
mean high water spring tide elevation) within the U.S. land representation according to the national LiDAR dataset,
the national network of tide gauges and land use histories recorded in the 1996, 2001, 2006, 2010, and 2016 NOAA
C-CAP surveys. C-CAP areas are calculated at the state/territory level and summed according to climate zone to
national values. Publicly-owned and privately-owned lands are represented. Trends in land cover change are
extrapolated to 1990 and 2021 from these datasets. The C-CAP database provides peer reviewed country-specific
mapping to support IPCC Approach 3 quantification of coastal wetland distribution, including conversion to and
from open water. Biomass carbon stocks are not sensitive to soil organic content but are differentiated based on
climate zone. Non-forested aboveground biomass carbon stock data are derived from a national assessment
combining field plot data and aboveground biomass mapping by remote sensing (Byrd et al. 2017; Byrd et al. 2018;
Byrd et al. 2020). The aboveground biomass carbon stock for estuarine forested wetlands (dwarf mangroves that
are not classified as forests due to their stature) is derived from a meta-analysis by Lu and Megonigal (20 1763;
Table 6-68). Aboveground biomass carbon stock data for all subcategories are not available and thus assumptions
were applied using expert judgment about the most appropriate assignment of a carbon stock to a disaggregation
of a community class. Root to shoot ratios from the Wetlands Supplement were used to account for belowground
biomass, which were multiplied by the aboveground carbon stock (Table 6-69; IPCC 2014). Above- and
belowground values were summed to obtain total biomass carbon stocks. Conversion to open water results in
emissions of all biomass carbon stocks during the year of conversion; therefore, emissions are calculated by
multiplying the C-CAP derived area of vegetated coastal wetlands lost that year in each climate zone by its mean
biomass.
Dead Organic Matter
Dead organic matter (DOM) carbon stocks, which include litter and dead wood stocks for subtropical estuarine
forested wetlands, are an emission from vegetated coastal wetlands converted to unvegetated open water coastal
wetlands across all years in the time series. Data on DOM carbon stocks are not currently available for either
palustrine or estuarine scrub/shrub wetlands for any climate zone. Data for estuarine forested wetlands in other
climate zones are not included since there is no estimated loss of these forests to unvegetated open water coastal
wetlands across any year based on C-CAP data. For subtropical estuarine forested wetlands, Tier 1 estimates of
mangrove DOM were used (IPCC 2014). Trends in land cover change are derived from the NOAA C-CAP dataset and
extrapolated to cover the entire 1990 through 2021 time series. Conversion to open water results in emissions of
all DOM carbon stocks during the year of conversion; therefore, emissions are calculated by multiplying the C-CAP
derived area of vegetated coastal wetlands lost that year by its Tier 1 DOM carbon stock.
Soil Carbon Stock Changes
Soil carbon stock changes are estimated for vegetated coastal wetlands converted to unvegetated open water
coastal wetlands. Country-specific soil carbon stocks were updated in 2018 based upon analysis of an assembled
dataset of 1,959 cores from across the conterminous United States (Holmquist et al. 2018). This analysis
63 See https://eithub.com/Smithsonian/Coastal-Wetland-NGGl-Data-Public; accessed September 2023.
6-118 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
demonstrated that it was not justified to stratify carbon stocks based upon mineral or organic soil classification,
climate zone, or wetland classes; therefore, a single soil carbon stock of 2701C ha 1 was applied to all classes.
Following the Tier 1 approach for estimating CO2 emissions with extraction provided within the Wetlands
Supplement, soil carbon loss with conversion of vegetated coastal wetlands to unvegetated open water coastal
wetlands is assumed to affect soil carbon stock to one-meter depth (Holmquist et al. 2018) with all emissions
occurring in the year of wetland conversion, and multiplied by activity data of vegetated coastal wetland area
converted to unvegetated open water wetlands. The methodology follows Eq. 4.6 in the Wetlands Supplement.
Soil Methane Emissions
A Tier 1 assumption has been applied that salinity conditions are unchanged and hence CH4 emissions are assumed
to be zero with conversion of vegetated coastal wetlands to unvegetated open water coastal wetlands.
Uncertainty
Underlying uncertainties in estimates of soil and biomass carbon stock changes are associated with country-
specific (Tier 2) literature values of these stocks, while the uncertainties with the Tier 1 estimates are associated
with subtropical estuarine forested wetland DOM stocks. Assumptions that underlie the methodological
approaches applied and uncertainties linked to interpretation of remote sensing data are also included in this
uncertainty assessment. The IPCC default assumption of 1 m of soil erosion with anthropogenic activities was
adopted to provide standardization in U.S. tidal carbon accounting (Holmquist et al. 2018). This depth of
potentially erodible tidal wetland soil has not been comprehensively addressed since most soil cores analyzed
were shallow (e.g., less than 50 cm) and do not necessarily reflect the depth to non-wetland soil or bedrock
(Holmquist et al. 2018). Uncertainty specific to coastal wetlands include differentiation of palustrine and estuarine
community classes, which determines the soil carbon stock applied. Because mean soil and biomass carbon stocks
for each available community class are in a fairly narrow range, the same overall uncertainty was assigned to each
(i.e., applying approach for asymmetrical errors, the largest uncertainty for any soil carbon stock value should be
applied in the calculation of error propagation; IPCC 2000). For aboveground biomass carbon stocks, the mean
standard error was very low and largely influenced by the uncertainty associated with the estimated map area
(Byrd et al. 2018). Uncertainty for root to shoot ratios, which are used for quantifying belowground biomass, are
derived from the Wetlands Supplement. Uncertainty for subtropical estuarine forested wetland DOM stocks was
derived from those listed for the Tier 1 estimates (IPCC 2014). Overall uncertainty of the NOAA C-CAP remote
sensing product is 15 percent. This is in the range of remote sensing methods (+/-10 to 15 percent; IPCC 2003). The
combined uncertainty was calculated by summing the squared uncertainty for each individual source (C-CAP, soil,
biomass, and DOM) and taking the square root of that total.
Uncertainty estimates are presented in Table 6-74 for each subcategory (i.e., soil carbon, biomass carbon, and
DOM emissions). The combined uncertainty across all subcategory is 32.0 percent above and below the estimate
of 1.5 MMT CO2 Eq, which is driven by the uncertainty in the soil carbon estimates. In 2022, the total carbon flux
was 1.5 MMT CO2 Eq., with lower and upper estimates of 1.0 and 2.0 MMT CO2 Eq.
Table 6-74: Approach 1 Quantitative Uncertainty Estimates for CO2 Flux Occurring within
Vegetated Coastal Wetlands Converted to Unvegetated Open Water Coastal Wetlands in
2022 (MMT CO2 Eq. and Percent)
2022 Flux Uncertainty Range Relative to Flux Estimate
Source Estimate
(MMTCO. Eq.)
(MMTCO' Eq.) (%)
Lower
Bound
Upper
Bound
Lower
Bound
Upper
Bound
Biomass C Stock
0.06
0.05
0.08
-24.1%
+24.1%
Dead Organic Matter C Stock
0.0005
0.000
0.001
-25.8%
+25.8%
Soil C Stock
1.5
1.3
1.7
-15.0%
+15.0%
Total Flux
1.5
1.0
2.0
-32.0%
+32.0%
Note: Totals may not sum due to independent rounding.
Land Use, Land-Use Change, and Forestry 6-119
-------
QA/QC and Verification
Data provided by NOAA (i.e., National LiDAR Dataset, NOS Tide Data, and C-CAP land cover and land cover change
mapping) undergo internal agency QA/QC procedures. Acceptance of final datasets into archive and dissemination
are contingent upon assurance that the data product is compliant with mandatory NOAA QA/QC requirements
(McCombs et al. 2016). QA/QC and Verification of the soil carbon stock dataset have been provided by the
Smithsonian Environmental Research Center and by the Coastal Wetlands project team leads who reviewed the
estimates against primary scientific literature. Biomass carbon stocks are derived from peer-review literature and
reviewed by the U.S. Geological Survey prior to publishing, by the peer-review process during publishing, and by
the coastal wetland inventory team leads before inclusion in the Inventory. For subtropical estuarine forested
wetlands, Tier 1 estimates of mangrove DOM were used (IPCC 2014). Land cover estimates were assessed to
ensure that the total land area did not change over the time series in which the inventory was developed, and
were verified by a second QA team. A team of two evaluated and verified there were no computational errors
within the calculation worksheets.
Recalculations Discussion
No recalculations were performed for the current Inventory.
Planned Improvements
The depth of soil carbon affected by conversion of vegetated coastal wetlands converted to unvegetated coastal
wetlands will be updated from the IPCC default assumption of 1 m of soil erosion when mapping and modeling
advancements can quantitatively improve accuracy and precision. Improvements are underway to address this,
first conducting a review of literature publications. Until the time where these more detailed and spatially
distributed data are available, the IPCC default assumption that the top 1 m of soil is disturbed by anthropogenic
activity will be applied. This is a longer-term improvement.
More detailed research is in development that provides a longer-term assessment and more highly refined rates of
wetlands loss across the Mississippi Delta (e.g., Couvillion et al. 2016). The Mississippi Delta is the largest extent of
coastal wetlands in the United States. Higher resolution imagery analysis would improve quantification of
conversation to open water, which occurs not only at the edge of the marsh but also within the interior. Improved
mapping could provide a more refined regional Approach 2-3 land representation to support the national-scale
assessment provided by C-CAP.
An approach for calculating the fraction of remobilized coastal wetland soil carbon returned to the atmosphere as
CO2 is currently under review and may be included in future reports.
Research by USGS is investigating higher resolution mapping approaches to quantify conversion of coastal
wetlands is also underway. Such approaches may form the basis for a full Approach 3 land representation
assessment in future years. C-CAP data harmonization with the National Land Cover Dataset (NLCD) will be
incorporated into a future iteration of the Inventory.
Stock Changes from Unvegetated Open Water Coastal
Wetlands Converted to Vegetated Coastal Wetlands
Open water within the U.S. land base, as described in Section 6.1, is recognized as coastal wetlands within this
Inventory. The appearance of vegetated tidal wetlands on lands previously recognized as open water reflects either
the building of new vegetated marsh through sediment accumulation or the transition from other lands uses
through an intermediary open water stage as flooding intolerant plants are displaced and then replaced by
wetland plants. Biomass, DOM and soil carbon accumulation on unvegetated open water coastal wetlands
converted to vegetated coastal wetlands begins with vegetation establishment.
6-120 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
Within the United States, conversion of unvegetated open water coastal wetlands to vegetated coastal wetlands is
predominantly due to engineered activities, which include active restoration of wetlands (e.g., wetlands
restoration in San Francisco Bay), dam removals or other means to reconnect sediment supply to the nearshore
(e.g., Atchafalaya Delta, Louisiana, Couvillion et al. 2011). Wetland restoration projects have been ongoing in the
United States since the 1970s. Early projects were small, a few hectares in size. By the 1990s, restoration projects,
each hundreds of hectares in size, were becoming common in major estuaries. In several coastal areas e.g., San
Francisco Bay, Puget Sound, Mississippi Delta and south Florida, restoration activities are in planning and
implementation phases, each with the goal of recovering tens of thousands of hectares of wetlands.
In 2022, 2,406 ha of unvegetated open water coastal wetlands were converted to vegetated coastal wetlands
across all wetland types and climates, which has steadily increased over the reporting period (Table 6-66). This
resulted in 0.007 MMT CO2 Eq. (0.002 MMT C) and 0.1 MMT CO2 Eq. (0.03 MMT C) sequestered in soil and
biomass, respectively (Table 6-75 and Table 6-76). The soil carbon stock has increased during the Inventory
reporting period, likely due to increasing vegetated coastal wetland restoration over time. While DOM carbon
stock increases are present, they are minimal in the early part of the time series and zero in the later because
there are no conversions from unvegetated open water coastal wetlands to subtropical estuarine forested
wetlands between 2011 and 2016 (and by proxy through 2022), and that is the only coastal wetland type where
DOM data is currently available.
Throughout the reporting period, the amount of open water coastal wetlands converted to vegetated coastal
wetlands has increased over time, reflecting the increase in engineered restoration activities mentioned above.
Table 6-75: CO2 Flux from Carbon Stock Changes from Unvegetated Open Water Coastal
Wetlands Converted to Vegetated Coastal Wetlands (MMT CO2 Eq.)
Year
1990
2005
2018
2019
2020
2021
2022
Biomass C Flux
(+),
(0.1) j
(0.1)
(0.1)
(0.1)
(0.1)
(0.1)
Dead Organic Matter C Flux
w|
wl
0
0
0
0
0
Soil C Flux
(+)i
(+)1
(+)
(+)
(+)
(+)
(+)
Total C Stock Change
(+)
(0.1)
(0.1)
(0.1)
(0.1)
(0.1)
(0.1)
+ Absolute value does not exceed 0.05 MMT C02 Eq.
Notes: Parentheses indicate net sequestration. Totals may not sum due to independent rounding.
Table 6-76: CO2 Flux from Carbon Stock Changes from Unvegetated Open Water Coastal
Wetlands Converted to Vegetated Coastal Wetlands (MMT C)
Year
1990
2005
2018
2019
2020
2021
2022
Biomass C Flux
(0.01)
(0.02) I
(0.03)
(0.03)
(0.03)
(0.03)
(0.03)
Dead Organic Matter C Flux
(+)
1
(+)
0
0
0
0
0
Soil C Flux
(+)
Wl
(+)
(+)
(+)
(+)
(+)
Total C Stock Change
(0.01)
1
_
(0.02)
(0.03)
(0.03)
(0.03)
(0.03)
(0.03)
+ Absolute value does not exceed 0.005 MMT C.
Notes: Parentheses indicate net sequestration. Totals may not sum due to independent rounding.
Methodology and Time-Series Consistency
The following section includes a brief description of the methodology used to estimate changes in soil, biomass
and DOM carbon stocks, and CFU emissions for unvegetated open water coastal wetlands converted to vegetated
coastal wetlands.
Methodological recalculations were applied to the entire time series to ensure time-series consistency from 1990
through 2022.
Land Use, Land-Use Change, and Forestry 6-121
-------
Biomass Carbon Stock Changes
Quantification of regional coastal wetland biomass carbon stock changes for palustrine and estuarine marsh
vegetation are presented for unvegetated open water coastal wetlands converted to vegetated coastal wetlands
on lands below the elevation of high tides (taken to be mean high water spring tide elevation) according to the
national LiDAR dataset, the national network of tide gauges and land use histories recorded in the 1996, 2001,
2005, 2011, and 2016 NOAA C-CAP surveys. C-CAP areas are calculated at the state/territory level and summed
according to climate zone to national values. Privately-owned and publicly-owned lands are represented. Trends in
land cover change are extrapolated to 1990 and 2022 from these datasets (Table 6-65). C-CAP provides peer
reviewed high resolution level mapping of coastal wetland distribution, including conversion to and from open
water. Biomass carbon stock is not sensitive to soil organic content but differentiated based on climate zone. Data
for non-forested wetlands are derived from a national assessment combining field plot data and aboveground
biomass mapping by remote sensing (Table 6-68; Byrd et al. 2017; Byrd et al. 2018; Byrd et al. 2020). The
aboveground biomass carbon stock for subtropical estuarine forested wetlands (dwarf mangroves that are not
classified as forests due to their stature) is derived from a meta-analysis by Lu and Megonigal (20 1764).
Aboveground biomass carbon stock data for all subcategories are not available and thus assumptions were applied
using expert judgment about the most appropriate assignment of a carbon stock to a disaggregation of a
community class. Root to shoot ratios from the Wetlands Supplement were used to account for belowground
biomass, which were multiplied by the aboveground carbon stock (Table 6-69; IPCC 2014). Above- and
belowground values were summed to obtain total biomass carbon stocks.
Conversion of open water to vegetated coastal wetlands results in the establishment of a standing biomass carbon
stock; therefore, stock changes that occur are calculated by multiplying the C-CAP derived area gained that year in
each climate zone by its mean biomass. While the process of revegetation of unvegetated open water wetlands
can take many years to occur, it is assumed in the calculations that the total biomass is reached in the year of
conversion.
Dead Organic Matter
Dead organic matter (DOM) carbon stocks, which include litter and dead wood stocks, are included for subtropical
estuarine forested wetlands for vegetated coastal wetlands converted to unvegetated open water coastal
wetlands across all years. Tier 1 default or country-specific data on DOM are not currently available for either
palustrine or estuarine scrub/shrub wetlands for any climate zone. Data for estuarine forested wetlands in other
climate zones are not included since there is no estimated loss of these forests to unvegetated open water coastal
wetlands across any year based on C-CAP data. Tier 1 estimates of subtropical estuarine forested wetland DOM
were used (IPCC 2014). Trends in land cover change are derived from the NOAA C-CAP dataset and extrapolated to
cover the entire 1990 through 2021 time series. Dead organic matter removals are calculated by multiplying the C-
CAP derived area gained that year by its Tier 1 DOM C stock. Similar to biomass carbon stock gains, gains in DOM
can take many years to occur, but for this analysis, the total DOM stock is assumed to accumulate during the first
year of conversion.
Soil Carbon Stock Change
Soil carbon stock changes are estimated for unvegetated open water coastal wetlands converted to vegetated
coastal wetlands. Country-specific soil carbon removal factors associated with soil carbon accretion, stratified by
climate zones and wetland classes, are derived from a synthesis of peer-reviewed literature and updated this year
based upon refined review of the dataset (Lynch 1989; Orson et al. 1990; Kearny & Stevenson 1991; Thorn et al.
1992; Roman et al. 1997; Craft et al. 1998; Orson et al. 1998; Merrill 1999; Weis et al. 2001; Hussein et al. 2004;
Church et al. 2006; Koster et al. 2007; Drexler et al. 2009; Boyd 2012; Callaway et al. 2012 a & b; Bianchi et al.
64 See https://doi.org/lQ.25573/serc.21565671; accessed September 2023.
6-122 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
2013; Drexler et al. 2013; Watson and Byrne 2013; Crooks et al. 2014; Breithaupt et al. 2014; Weston et al. 2014;
Smith et al. 2015; Villa & Mitsch 2015; Boyd and Sommerfield 2016; Marchio et al. 2016; Noe et al. 2016; Arriola
and Cable 2017; Boyd et al. 2017; Gerlach et al. 2017; Giblin and Forbrich 2018; Krauss et al. 2018; Abbott et al.
2019; Drexler et al. 2019; Poppe and Rybczyk 2019; Ensign et al. 2020; Kemp et al. 2020; Lagomasino et al. 2020;
Luk et al. 2020; McTigue et al. 2020; Peck et al. 2020; Vaughn et al. 2020; Weston et al. 2020; Arias-Ortiz et al.
2021; Baustian et al. 2021; Allen et al. 2022; Miller et al. 2022). Soil carbon stock changes are stratified based upon
wetland class (Estuarine, Palustrine) and subclass (Emergent Marsh, Scrub Shrub). For soil carbon stock change, no
differentiation is made for soil type (i.e., mineral, organic). Soil carbon removal factors were developed from
literature references that provided soil carbon removal factors disaggregated by climate region and vegetation
type by salinity range (estuarine or palustrine) as identified using NOAA C-CAP as described above (see Table 6-70
for values).
Tier 2 level estimates of carbon stock changes associated with annual soil carbon accumulation in vegetated
coastal wetlands were developed using country-specific soil carbon removal factors multiplied by activity data on
unvegetated coastal wetlands converted to vegetated coastal wetlands. The methodology follows Eq. 4.7, Chapter
4 of the Wetlands Supplement, and is applied to the area of unvegetated coastal wetlands converted to vegetated
coastal wetlands on an annual basis.
Soil Methane Emissions
A Tier 1 assumption has been applied that salinity conditions are unchanged and hence Cm emissions are assumed
to be zero with conversion of vegetated open water coastal wetlands to vegetated coastal wetlands.
Uncertainty
Underlying uncertainties in estimates of soil and biomass carbon stock changes include uncertainties associated
with country-specific (Tier 2) literature values of these carbon stocks, assumptions that underlie the
methodological approaches applied and uncertainties linked to interpretation of remote sensing data. Uncertainty
specific to coastal wetlands include differentiation of palustrine and estuarine community classes that determines
the soil carbon stock applied. Because mean soil and biomass carbon stocks for each available community class are
in a fairly narrow range, the same overall uncertainty was applied to each, respectively (i.e., applying approach for
asymmetrical errors, the largest uncertainty for any soil carbon stock value should be applied in the calculation of
error propagation; IPCC 2000). For aboveground biomass carbon stocks, the mean standard error was very low and
largely influenced by error in estimated map area (Byrd et al. 2018). Uncertainty for root to shoot ratios, which are
used for quantifying belowground biomass (Table 6-69), are derived from the Wetlands Supplement. Uncertainty
for subtropical estuarine forested wetland DOM stocks were derived from those listed for the Tier 1 estimates
(IPCC 2014). Overall uncertainty of the NOAA C-CAP remote sensing product is 15 percent. This is in the range of
remote sensing methods (±10 to 15 percent; IPCC 2003). The combined uncertainty was calculated by summing
the squared uncertainty for each individual source (C-CAP, soil, biomass, and DOM) and taking the square root of
that total.
Uncertainty estimates are presented in Table 6-77 for each subcategory (i.e., soil carbon, biomass carbon and DOM
emissions). The combined uncertainty across all subsources is 33.43 percent above and below the estimate of-0.1
MMT CO2 Eq. In 2022, the total carbon flux was -0.1 MMT CO2 Eq., with lower and upper estimates of-0.1 and -
0.08 MMT CO2 Eq.
Land Use, Land-Use Change, and Forestry 6-123
-------
Table 6-77: Approach 1 Quantitative Uncertainty Estimates for Carbon Stock Changes
Occurring within Unvegetated Open Water Coastal Wetlands Converted to Vegetated Coastal
Wetlands in 2022 (MMT CO2 Eq. and Percent)
Source
2022 Flux Estimate
Uncertainty Range
Relative to Flux Estimate
(MMT CO . Eq.)
(MMT CO
Eq.)
(%)
Lower
Upper
Lower
Upper
Bound
Bound
Bound
Bound
Biomass C Stock Flux
(0.1)
(0.12)
(0.08)
-20.0%
+20.0%
Dead Organic Matter C Stock Flux
0
0
0
-25.8%
+25.8%
Soil C Stock Flux
(0.008)
(0.009)
(0.006)
-17.7%
+17.7%
Total Flux
(0.1)
(0.14)
(0.01)
-33.3%
+33.3%
Notes: Parentheses indicate net sequestration. Totals may not sum due to independent rounding.
QA/QC and Verification
NOAA provided data (i.e., National LiDAR Dataset, NOS Tide Data, and C-CAP land cover and land cover change
mapping), which undergo internal agency QA/QC assessment procedures. Acceptance of final datasets into the
archive for dissemination are contingent upon assurance that the product is compliant with mandatory NOAA
QA/QC requirements (McCombs et al. 2016). QA/QC and Verification of soil carbon stock dataset has been
provided by the Smithsonian Environmental Research Center and Coastal Wetlands project team leads who
reviewed the summary tables against primary scientific literature. Aboveground biomass carbon reference stocks
are derived from an analysis by the Blue Carbon Monitoring project and reviewed by U.S. Geological Survey prior
to publishing, the peer-review process during publishing, and the coastal wetland inventory team leads before
inclusion in the Inventory. Root to shoot ratios and DOM data are derived from peer-reviewed literature and
undergo review as per IPCC methodology. Land cover estimates were assessed to ensure that the total land area
did not change over the time series in which the inventory was developed and verified by a second QA team. A
team of two evaluated and verified there were no computational errors within calculation worksheets. Two
biogeochemists at the USGS, also members of the NASA Carbon Monitoring System Science Team, corroborated
the simplifying assumption that where salinities are unchanged CH4 emissions are constant with conversion of
unvegetated open water coastal wetlands to vegetated coastal wetlands.
Recalculations Discussion
A recalculation of emission factors for soil carbon accretion rates was performed using the same methodology and
criteria as in Lu and Megonigal (2017) and described above. This new analysis incorporated data published since
2016 and other relevant data that were not previously included. Table 6-70 shows the new values. The updated
synthesis resulted in a general increase in soil carbon accumulation rates for estuarine emergent and scrub/shrub
wetlands, which resulted in a minimal annual average increases in removals of 0.001 MMT CO2 Eq. for the entire
time series.
Planned Improvements
The USGS is investigating higher resolution mapping approaches to quantify conversion of coastal wetlands. Such
approaches may form the basis for a full Approach 3 land representation assessment in future years. C-CAP data
harmonization with the National Land Cover Dataset (NLCD) will be incorporated into a future iteration of the
Inventory.
N20 Emissions from Aquaculture in Coastal Wetlands
Shrimp and fish cultivation in coastal areas increases nitrogen loads resulting in direct emissions of N2O. Nitrous
oxide is generated and emitted as a byproduct of the conversion of ammonia (contained in fish urea) to nitrate
6-124 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
through nitrification and nitrate to N2 gas through denitrification (Hu et al. 2012). Nitrous oxide emissions can be
readily estimated from data on fish production (IPCC 2014).
Aquaculture production in the United States has fluctuated slightly from year to year, with resulting N2O emissions
between 0.1 and 0.2 MMT CO2 Eq. between 1990 and 2022 (Table 6-78). Aquaculture production data were
updated through 2019; data through 2022 are not yet available and in this analysis are held constant with 2019
emissions of 0.2 MMT CO2 Eq. (0.5 Kt N2O).
Table 6-78: N2O Emissions from Aquaculture in Coastal Wetlands (MMT CO2 Eq. and kt N2O)
Year
1990 2005
2018
2019
2020
2021
2022
Emissions (MMT C02 Eq.)
O
i-*
O
k>
0.1
0.1
0.1
0.1
0.1
Emissions (kt N20)
0.4 i 0.61
0.5
0.5
0.5
0.5
0.5
Methodology and Time-Series Consistency
The methodology to estimate N2O emissions from aquaculture in coastal wetlands follows the Tier 1 guidance in
the Wetlands Supplement by applying country-specific fisheries production data and the IPCC Tier 1 default
emission factor.
Each year NOAA Fisheries document the status of U.S. marine fisheries in the annual report of Fisheries of the
United States (National Marine Fisheries Service 2022), from which activity data for this analysis is derived.65 The
fisheries report has been produced in various forms for more than 100 years, primarily at the national level, on
U.S. recreational catch and commercial fisheries landings and values. In addition, data are reported on U.S.
aquaculture production, the U.S. seafood processing industry, imports and exports offish-related products, and
domestic supply and per capita consumption of fisheries products. Within the aquaculture chapter, the mass of
production for catfish, striped bass, tilapia, trout, crawfish, salmon and shrimp are reported. While some of these
fisheries are produced on land and some in open water cages within coastal wetlands, all have data on the
quantity of food stock produced, which is the activity data that is applied to the IPCC Tier 1 default emissions
factor to estimate emissions of N2O from aquaculture. It is not apparent from the data as to the amount of
aquaculture occurring above the extent of high tides on river floodplains. While some aquaculture occurs on
coastal lowland floodplains, this is likely a minor component of tidal aquaculture production because of the need
for a regular source of water for pond flushing. The estimation of N2O emissions from aquaculture is not sensitive
to salinity using IPCC approaches, and as such, the location of aquaculture ponds within the boundaries of coastal
wetlands does not influence the calculations.
Other open water shellfisheries for which no food stock is provided, and thus no additional N inputs, are not
applicable for estimating N2O emissions (e.g., clams, mussels, and oysters) and have not been included in the
analysis. The IPCC Tier 1 default emissions factor of 0.00169 kg N2O-N per kg offish/shellfish produced is applied to
the activity data to calculate total N2O emissions.
Methodological recalculations were applied to the entire time series to ensure time-series consistency from 1990
through 2022.
Uncertainty
Uncertainty estimates are based upon the Tier 1 default 95 percent confidence interval provided in Table 4.15,
chapter 4 of the Wetlands Supplement for N2O emissions and on expert judgment of the NOAA Fisheries of the
United States fisheries production data. Given the overestimate of fisheries production from coastal wetland areas
65 See https://www.fisheries.noaa.Eov/resource/document/fisheries-uriited-states-2Q2Q; accessed September August 2023.
Land Use, Land-Use Change, and Forestry 6-125
-------
due to the inclusion offish production in non-coastal wetland areas, this is a reasonable initial first approximation
for an uncertainty range.
Uncertainty estimates for N2O emissions from aquaculture production are presented in Table 6-79 for N2O
emissions. The combined uncertainty is 116 percent above and below the estimate of 0.13 MMT CO2 Eq. In 2022,
the total flux was 0.13 MMT CO2 Eq., with lower and upper estimates of 0.00 and 0.29 MMT CO2 Eq.
Table 6-79: Approach 1 Quantitative Uncertainty Estimates for N2O Emissions from
Aquaculture Production in Coastal Wetlands in 2022 (MMT CO2 Eq. and Percent)
2022 Emissions
Estimate
Uncertainty Range Relative to Emissions Estimate-1
Source
(MMT CO . Eq.)
(MMT CO.
Eq.)
(%)
Lower
Upper
Lower
Upper
Bound
Bound
Bound
Bound
Combined Uncertainty for N20 Emissions
for Aquaculture Production in Coastal
0.13
0.00
0.29
-116%
+116%
Wetlands
a Range of emission estimates predicted by Monte Carlo stochastic simulation for a 95 percent confidence interval.
QA/QC and Verification
NOAA provided internal QA/QC review of reported fisheries data. The coastal wetlands inventory team consulted
with the coordinating lead authors of the coastal wetlands chapter of the Wetlands Supplement to assess which
fisheries production data to include in estimating emissions from aquaculture. It was concluded that N2O emissions
estimates should be applied to any fish production to which food supplement is supplied be they pond or coastal
open water and that salinity conditions were not a determining factor in production of N2O emissions.
Recalculations Discussion
No recalculations were performed for the current Inventory.
Flooded Land Remaining Flooded Land
Flooded lands are defined as water bodies where human activities have 1) caused changes in the amount of
surface area covered by water, typically through water level regulation (e.g., constructing a dam), 2) waterbodies
where human activities have changed the hydrology of existing natural waterbodies thereby altering water
residence times and/or sedimentation rates, in turn causing changes to the natural emission of greenhouse gases,
and 3) waterbodies that have been created by excavation, such as canals, ditches and ponds (IPCC 2019). Flooded
lands include waterbodies with seasonally variable degrees of inundation, but these waterbodies would be
expected to retain some inundated area throughout the year under normal conditions.
Flooded lands are broadly classified as "reservoirs" or "other constructed waterbodies" (IPCC 2019). Other
constructed waterbodies include canals/ditches and ponds (flooded land <8 ha surface area). Reservoirs are
defined as flooded land greater than 8 ha. IPCC guidance (IPCC 2019) provides default emission factors for
reservoirs, ponds, and canals/ditches.
Land that has been flooded for greater than 20 years is defined as flooded land remaining flooded land and land
flooded for 20 years or less is defined as land converted to flooded land. The distinction is based on literature
reports that CFU and CO2 emissions are high immediately following flooding, but decline to a steady background
level approximately 20 years after flooding (Abril et al. 2005; Barros et al. 2011; Teodoru et al. 2012). Emissions of
Cm are estimated for flooded land remaining flooded land, but CO2 emissions are not included as they are
6-126 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
primarily the result of decomposition of organic matter entering the waterbody from the catchment or contained
in inundated soils and are captured in Chapter 6, Land Use, Land-Use Change, and Forestry,
Nitrous oxide emissions from flooded lands are largely related to input of organic or inorganic nitrogen from the
watershed. These inputs from runoff/leaching/deposition are largely driven by anthropogenic activities such as
land-use change, wastewater disposal or fertilizer application in the watershed or application of fertilizer or feed in
aquaculture. These emissions are not included here to avoid double-counting of N2O emissions which are captured
in other source categories, such as indirect N2O emissions from managed soils (Section 5.4, Agricultural Soil
Management) and wastewater management (Section 7.2, Wastewater Treatment and Discharge).
Emissions from Flooded Land Remaining Flooded Land-
Reservoirs
Reservoirs are designed to store water for a wide range of purposes including hydropower, flood control, drinking
water, and irrigation. In 2022, the United States and Puerto Rico hosted 10.2 million ha of reservoir surface area in
the flooded land remaining flooded land category (see Methodology and Time-Series Consistency below for
calculation details). These reservoirs are distributed across all six of the aggregated climate zones used to define
flooded land emission factors (Figure 6-10) (IPCC 2019).
Figure 6-10: U.S. Reservoirs (black polygons) in the Flooded Land Remaining Flooded Land
Category in 2022
Climate Zone
H boreal
Q cool temperate
| tropical dry/montane
¦ tropical moist/wet
| warm temperate dry
H warm temperate moist
Alaska
Hawaii
o
100 mi
1000 mi
500 mi
Note: Colors represent climate zone used to derive IPCC default emission factors.
Methane is produced in reservoirs through the microbial breakdown of organic matter. Per unit area, CH4 emission
rates tend to scale positively with temperature and system productivity (i.e., abundance of algae), but negatively
with system size (i.e., depth, surface area). Methane produced in reservoirs can be emitted from the reservoir
surface or exported from the reservoir when ChU-rich water passes through the dam. This exported CH4 can be
released to the atmosphere as the water passes through hydropower turbines or the downstream river channel.
Methane emitted to the atmosphere via this pathway is referred to as "downstream emissions."
Table 6-80 and Table 6-81 below summarize nationally aggregated CH4 emissions from reservoirs. The increase in
CH4 emissions through the time series is attributable to reservoirs matriculating from the land converted to
flooded land category into the flooded land remaining flooded land category.
Land Use, Land-Use Change, and Forestry 6-127
-------
Table 6-80: CH4 Emissions from Flooded Land Remaining Flooded Land—Reservoirs (MMT
C02 Eq.)
Source 1990
2005
2018
2019
2020
2021
2022
Reservoirs
Surface Emission 26.2
27.7
27.9
27.9
27.9
27.9
27.9
Downstream Emission 2.4
2.5
2.5
2.5
2.5
2.5
2.5
Total 28.6
30.2
30.4
30.4
30.4
30.4
30.4
Note: Totals may not sum to due independent rounding.
Table 6-81: CH4 Emissions from Flooded Land Remaining Flooded Land-
source 1990
2005
2018
2019
2020
2021
2022
Reservoirs
Surface Emission 937
989
997
997
997
997
997
Downstream Emission 84
89
90
90
90
90
90
Total 1,022
1,078
1,086
1,086
1,087
1,087
1,087
Note: Totals may not sum to due independent rounding.
Methane emissions from reservoirs in Texas, Florida, and Louisiana (Figure 6-11, Table 6-82) compose 34 percent
of national CH4 emissions from reservoirs in 2022. Emissions from these states are particularly high due to 1) the
large expanse of reservoirs in these states (Table 6-85) and 2) the high CFU emission factor for the tropical
dry/montane and topical moist climate zones which encompass a majority of the flooded land area in these states
(Figure 6-10, Table 6-83).
Methane emissions from reservoirs in flooded land remaining flooded land increased 6.4 percent from 1990 to
2022 due to the matriculation of reservoirs in land converted to flooded land to flooded land remaining flooded
land.
Figure 6-11: Total CH4 Emissions (Downstream + Surface) from Reservoirs in Flooded Land
Remaining Flooded Land in 2022 (kt CH4)
6-128 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
Table 6-82: Surface and Downstream CH4 Emissions from Reservoirs in Flooded Land
Remaining Flooded Land in 2022 (kt CH4)
State
Surface
Downstream
Total
Alabama
22
2
24
Alaska
1
+
1
Arizona
14
1
16
Arkansas
25
2
27
California
42
4
46
Colorado
7
1
7
Connecticut
3
+
3
Delaware
3
+
3
District of Columbia
+
+
+
Florida
143
13
155
Georgia
35
3
38
Hawaii
1
+
1
Idaho
12
1
13
Illinois
17
2
19
Indiana
7
1
7
Iowa
7
1
7
Kansas
10
1
11
Kentucky
13
1
14
Louisiana
58
5
64
Maine
14
1
15
Maryland
13
1
14
Massachusetts
5
+
5
Michigan
9
1
10
Minnesota
21
2
23
Mississippi
20
2
21
Missouri
17
1
18
Montana
16
1
17
Nebraska
7
1
7
Nevada
17
2
19
New Hampshire
3
+
4
New Jersey
9
1
9
New Mexico
7
1
7
New York
18
2
20
North Carolina
33
3
36
North Dakota
14
1
15
Ohio
7
1
7
Oklahoma
26
2
28
Oregon
14
1
16
Pennsylvania
7
1
8
Puerto Rico
+
+
+
Rhode Island
1
+
1
South Carolina
38
3
41
South Dakota
12
1
14
Tennessee
20
2
21
Texas
138
12
150
Utah
21
2
23
Vermont
5
+
5
Virginia
25
2
27
Washington
23
2
25
West Virginia
3
+
3
Wisconsin
10
1
11
Wyoming
7
1
8
+ Indicates values less than 0.5 kt.
Land Use, Land-Use Change, and Forestry 6-129
-------
Methodology and Time-Series Consistency
Estimates of Cm emission for reservoirs in flooded land remaining flooded land follow the Tier 1 methodology in
the 2019 Refinement to the 2006IPCC Guidelines (IPCC 2019). Methane emissions from the surface of these
flooded lands are calculated as the product of flooded land surface area and a climate-specific emission factor
(Table 6-83). Downstream emissions are calculated as nine percent of the surface emission (Tier 1 default). Total
Cm emissions from reservoirs are calculated as the sum of surface and downstream emissions. National emissions
are calculated as the sum of state emissions.
The IPCC default surface emission factors used in the Tier 1 methodology are derived from model-predicted (G-res
model, Prairie et al. 2017) emission rates for all reservoirs in the Global Reservoir and Dam (GRanD) database
(Lehner et al. 2011). Predicted emission rates were aggregated by the 11 IPCC climate zones (IPCC 2019, Table
7A.2) which were collapsed into six climate zones using a regression tree approach. All six aggregated climate
zones are present in the United States.
Table 6-83: IPCC (2019) Default CH4 Emission Factors for Surface Emission from Reservoirs in
Flooded Land Remaining Flooded Land
Climate
Surface emission factor
(MTCH, ha 1 y ¦)
Boreal
0.0136
Cool Temperate
0.054
Warm Temperate Dry
0.1509
Warm Temperate Moist
0.0803
Tropical Dry/Montane
0.2837
Tropical Moist/Wet
0.1411
Note: downstream CH4 emissions are calculated as 9 percent of surface
emissions. Downstream emissions are not calculated for C02.
Area estimates
U.S. reservoirs were identified from the NHDWaterbody layer in the National Hydrography Dataset Plus V2
(NHD),66the National Inventory of Dams (NID),67 the National Wetlands Inventory (NWI),68 the Navigable
Waterways (NW) network,69 and the EPA's Safe Drinking Water Information System (SDWIS).70 The NHD only
covers the conterminous U.S., whereas the NID, NW and NWI also include Alaska, Hawaii, and Puerto Rico.
Waterbodies in the NHDWaterbody layer that were greater than or equal to 8 ha in surface area, not identified as
canal/ditch in NHD, and met any of the following criteria were considered reservoirs: 1) the waterbody was
classified as "Reservoir" in the NHDWaterbody layer, 2) the waterbody name in the NHDWaterbody layer included
"Reservoir", 3) the waterbody in the NHDWaterbody layer was located in close proximity (up to 100 m) to a dam in
the NID, 4) the NHDWaterbody GNIS name was similar to a nearby NID feature (between 100 m to 1000 m), 5) the
waterbody intersected a public drinking water intake.
66 See https://www.uses.gov/core-science-svstems/nep/national-hvdroeraphv.
67 See https://nid.sec.usace.armv.mil.
68 See https://www.fws.eov/program/national-wetlands-inventorv/data-download.
69 See https://hifld-geoplatform.opendata.arcgis.com/maps/aaa3767c7d2b41f69e7528f99cf2fb76 O/about.
70 See https://www.epa.gov/enviro/sdwis-overview. Not publicly available due to security concerns.
6-130 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
EPA assumes that all features included in the NW network are subject to water-level management to maintain
minimum water depths required for navigation and are therefore managed flooded lands. Navigable Waterway
features greater than 8 ha in surface area are defined as reservoirs,
NWI features were considered "managed" if they had a Special Modifier value indicating the presence of
management activities (Figure 6-12). To be included in the flooded lands inventory, the managed flooded land had
to be wet or saturated for at least one season per year (see "Water Regime" in Figure 6-12). NWI features that met
these criteria, were greater than 8 ha in surface area, and were not a canal/ditch (see emissions from land
converted to flooded land - other constructed waterbodies) were defined as reservoirs.
Any NWI or NHD feature that intersected a drinking water intake point from SDWIS was assumed to be
"managed." The rational being that a waterbody used as a source for public drinking water is typically managed in
some capacity - by flow and/or volume control.
Surface areas for identified flooded lands were taken from the NHD, NWI or NW. If features from the NHD, NWI, or
NW datasets overlapped, duplicated areas were erased. The first step was to take the final NWI flooded lands
features and use it to identify overlapping NHD features. If the NHD feature had its center in a NWI feature, it was
removed from analysis. Next, remaining NHD features were erased from any remaining overlapping NWI features.
Final selections of NHD and NWI features were used to erase any overlapping NW waterbodies.
Reservoir age was determined by assuming the waterbody was created the same year as a nearby (up to 100 m)
NID feature. If no nearby NID feature was identified, it was assumed the waterbody was greater than 20-years old
throughout the time series.
Figure 6-12: Selected Features from NWI that Meet Flooded Lands Criteria
MODIFIERS
In order to more adequately describe the wetland and deepwater habitats, one each of the water regime, water chemistry, soil, or
special modifiers may be applied at the class or lower level in the hierarchy
Water Regime
Special Modifiers
Water Chemistry
Soil
Nontidal
A Temporarily Flooded
B Seasonally Saturated
Saltwater Tidal
:L Subtidal
Freshwater Tidal
Q Regularly Flooded-Fresh Tidal
R Seasonally Flooded-Fresh Tidal
b Beaver
Halinity/Salinity pH Modifiers for
Fresh Water
1 Hyperhaline / Hypersaline a Acid
2 Euhaline / Eusaline t Circumneutral
3 Mixohaline / M ixosaline (Brackish) i Alkaline
4 Polyhaline
5 Mesohaline
6 Oligohaline
0 Fresh
g Organic
n Mineral
M Irregularly Exposed
d Partly Drained/Ditched
f Farmed
m Managed
h Diked/Impounded
r Artificial Substrate
s Spoil
x Excavated |
C Seasonally Flooded
N Regularly Flooded
P Irregularly Flooded
S Temporarily Flooded- Fresh Tidal
D Continuously Saturated
T Semipermanently Flooded-Fresh Tidal
V Permanently Flooded-Fresh Tidal
E Seasonally Flooded /
Saturated
F Semipermanently Flooded
G Intermittently Exposed
H Permanently Fjooded
U Intermittently Flooded
K Artificially Flooded |
I "I Must also meet one selected special modifier (red box) to be included in the flooded lands inventory
I ~1 Included in the flooded lands inventory if it meets water regime qualifier (gold box)
Source (modified): https://www.fws.gov/sites/default/files/documents/wetlands-and-deepwater-map-code-diagram.pdf
IPCC (2019) allows for the exclusion of managed waterbodies from the inventory if the water surface area or
residence time was not substantially changed by the construction of the dam. The guidance does not quantify
what constitutes a "substantial" change, but here EPA excludes the U.S. Great Lakes from the inventory based on
expert judgment that neither the surface area nor water residence time was substantially altered by their
associated dams.
Reservoirs were disaggregated by state (using boundaries from the 2016 U.S. Census Bureau71) and climate zone.
Downstream and surface emissions for cross-state reservoirs were allocated to states based on the surface area
that the reservoir occupied in each state. Only the U.S. portion of reservoirs that cross country borders were
included in the inventory.
71 See https://www.census.gov/geographies/mappine-files/time-series/geo/carto-bouiidarv-file.html.
Land Use, Land-Use Change, and Forestry 6-131
-------
The surface area of reservoirs in flooded land remaining flooded land increased by approximately 6 percent from
1990 to 2022 (Table 6-84) due to reservoirs matriculating into flooded land remaining flooded land when they
reached 20 years of age.
Table 6-84: National Totals of Reservoir Surface Area in Flooded Land Remaining Flooded
Land (millions of ha)
Surface Area (millions of ha)
1990
2005
2018
2019
2020
2021
2022
Reservoir
10.1,
10.2
10.2
10.2
10.2
10.2
Table 6-85: State Breakdown of Reservoir Surface Area in Flooded Land Remaining Flooded
Land (millions of ha)
State
1990
2005
2018
2019
2020
2021
2022
Alabama
0.22
0.23
0.23
0.23
0.23
0.23
0.23
Alaska
0.02
0.021
0.02
0.02
0.02
0.02
0.02
Arizona
0.06
0.06
0.06
0.06
0.06
0.06
0.06
Arkansas
0.281
0.29 =
0.29
0.29
0.29
0.29
0.29
California
0.37
0.39
0.39
0.39
0.39
0.39
0.39
Colorado
0.08 =
0-°9 :
0.09
0.09
0.09
0.09
0.09
Connecticut
0.03
0.03
0.04
0.04
0.04
0.04
0.04
Delaware
0-031
0.03 i
0.03
0.03
0.03
0.03
0.03
District of Columbia
+
+
+
+
+
+
+
Florida
°-98 '
1.01
1.01
1.01
1.01
1.01
1.01
Georgia
0.27
0.29
0.29
0.29
0.29
0.29
0.29
Hawaii
+::
+
+
+
+
+
+
Idaho
0.17
0.19
0.19
0.19
0.19
0.19
0.19
Illinois
0-17|
0.18-
0.22
0.22
0.22
0.22
0.22
Indiana
0.07
0.08
0.08
0.08
0.08
0.08
0.08
Iowa
0.081
o.o9=;
0.10
0.10
0.10
0.10
0.10
Kansas
0.09
0.11
0.11
0.11
0.11
0.11
0.11
Kentucky
0.16
0-16
0.16
0.16
0.16
0.16
0.16
Louisiana
0.40
0.41
0.41
0.41
0.41
0.41
0.41
Maine
0.25
0.26
0.26
0.26
0.26
0.26
0.26
Maryland
0.16
0.16
0.16
0.16
0.16
0.16
0.16
Massachusetts
0-07 ;
0.071
0.07
0.07
0.07
0.07
0.07
Michigan
0.16
0.17
0.17
0.17
0.17
0.17
0.17
Minnesota
0.381
0-38;
0.39
0.39
0.39
0.39
0.39
Mississippi
0.18
0.19
0.19
0.19
0.19
0.19
0.19
Missouri
0.195
0.211
0.21
0.21
0.21
0.21
0.21
Montana
0.27
0.29
0.29
0.29
0.29
0.29
0.29
Nebraska
0.071
0.08 =
0.08
0.08
0.08
0.08
0.08
Nevada
0.09
0.09
0.09
0.09
0.09
0.09
0.09
New Hampshire
0.061
0.06
0.06
0.06
0.06
0.06
0.06
New Jersey
0.11
0.11
0.11
0.11
0.11
0.11
0.11
New Mexico
0.05
liMHi
LO
o
d
0.05
0.05
0.05
0.05
0.05
New York
0.31
0.32
0.32
0.32
0.32
0.32
0.32
North Carolina
0.39!
°-41i
0.41
0.41
0.41
0.41
0.41
North Dakota
0.10
0.25
0.26
0.26
0.26
0.26
0.26
Ohio
o.o8:
0.09
0.09
0.09
0.09
0.09
0.09
Oklahoma
0.24
0.27
0.27
0.27
0.27
0.27
0.27
Oregon
0.20
0.20
0.20
0.20
0.20
0.20
0.20
Pennsylvania
0.09
0.11
0.11
0.11
0.11
0.11
0.11
Puerto Rico
+
+i
+
+
+
+
+
Rhode Island
0.02
0.02
0.02
0.02
0.02
0.02
0.02
South Carolina
0.311
0.32!
0.33
0.33
0.33
0.33
0.33
6-132 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
South Dakota
0.23
0.23
0.23
0.23
0.23
0.23
0.23
Tennessee
0-181
°-24
0.24
0.24
0.24
0.24
0.24
Texas
0.63
0.70
0.71
0.71
0.71
0.71
0.71
Utah
0.18 !
0-181
0.18
0.18
0.18
0.18
0.18
Vermont
0.09
0.09
0.09
0.09
0.09
0.09
0.09
Virginia
0.30 =
!!!!!!;
0.311
0.31
0.31
0.31
0.31
0.31
Washington
0.24
0.24
0.24
0.24
0.24
0.24
0.24
West Virginia
0.03 =
0.03 I
0.04
0.04
0.04
0.04
0.04
Wisconsin
0.18
0.18
0.18
0.18
0.18
0.18
0.18
Wyoming
0.121
0.131
0.14
0.14
0.14
0.14
0.14
Total
9.47
10.10
10.20
10.20
10.20
10.20
10.20
+ Indicates values less than 0.005 million ha.
Note: Totals may not sum due to independent rounding.
Uncertainty
Uncertainty in estimates of Cm emissions from reservoirs in flooded land remaining flooded land (Table 6-86) are
developed using Monte Carlo simulations (IPCC Approach 2) and include uncertainty in the default emission factors
and land areas. Each iteration of the simulation draws surface and downstream emission factors from a statistical
distribution based on the mean and variance in the 2019 Refinement to the 2006 IPCC Guidelines (IPCC 2019). The
Cm emission factors for surface and downstream emissions are modeled using normal and lognormal
distributions, respectively. The 2019 IPCC Refinement does not contain sufficient information to define a normal
distribution for the CO2 emission factor and a uniform distribution bounded by the 95% confidence internal of the
mean is assumed. Uncertainties in the spatial data include 1) uncertainty in area estimates from the NHD, NWI,
and NW, and 2) uncertainty in the location of dams in the NID and drinking water intakes in SDWIS. Overall
uncertainties in these spatial datasets are unknown, but uncertainty for remote sensing products is assumed to be
±10-15 percent based on IPCC guidance (IPCC 2003). An uncertainty range of ± 15 percent for the reservoir area
estimates is assumed and is based on expert judgment. Each iteration of the simulation draws a surface area for
each waterbody from a uniform distribution bounded by ± 15 percent of the estimated surface area.
Table 6-86: Approach 2 Quantitative Uncertainty Estimates for CH4 Emissions from Reservoirs
in Flooded Land Remaining Flooded Land
Source
Gas
2022 Emission Estimate
Uncertainty Range Relative to Emission Estimate-1
(MMT COEq.)
(MMT CO
Eq.)
(%)
Lower
Upper
Lower
Upper
Bound
Bound
Bound
Bound
Reservoir
Surface
ch4
27.9
27.4
28.4
-1.7%
+1.7%
Downstream
ch4
2.5
2.4
3.1
-5.6%
+22.4%
Total
CH,
30.4
29.9
31.3
-1.6%
+2.9%
a Range of emission estimates predicted by Monte Carlo stochastic simulation for a 95 percent confidence interval.
QA/QC and Verification
The National Hydrography Data (NHD) is managed by the USGS in collaboration with many other federal, state, and
local entities. Extensive QA/QC procedures are incorporated into the curation of the NHD. The National Inventory
of Dams (NID) is maintained by the U.S. Army Corps of Engineers (USACE) in collaboration with the Federal
Emergency Management Agency (FEMA) and state regulatory offices. USACE resolves duplicative and conflicting
data from 68 data sources, which helps obtain the more complete, accurate, and updated NID. The Navigable
Waterways (NW) dataset is part of the U.S. Department of Transportation (USDOT)/Bureau of Transportation
Statistics (BTS) National Transportation Atlas Database (NTAD). The NW is a comprehensive network database of
the nation's navigable waterways updated on a continuing basis. U.S. Fish and Wildlife Service is the principal
agency in charge of wetland mapping including the National Wetlands Inventory (NWI). Quality and consistency of
Land Use, Land-Use Change, and Forestry 6-133
-------
the Wetlands Layer is supported by federal wetlands mapping and classification standards, which were developed
under the oversight of the Federal Geographic Data Committee (FGDC) with input by the FGDC Wetlands
Subcommittee. This dataset is part of the FGDC Water-Inland Theme, which is co-chaired by the FWS and the U.S.
Geological Survey. The EPA's Safe Drinking Water Information System (SDWIS) tracks information on drinking
water contamination levels as required by the 1974 Safe Drinking Water Act and its 1986 and 1996 amendments.
General QA/QC procedures were applied to activity data, documentation, and emission calculations consistent
with the U.S. Inventory QA/QC plan, which is in accordance with Vol. 1 Chapter 6 of the 2006IPCC Guidelines (see
Annex 8 for more details). All calculations were executed independently in Excel and R. Ten percent of state and
national totals were randomly selected for comparison between the two approaches to ensure there were no
computational errors.
Recalculations Discussion
The EPA's SDWIS is a new data source used in the current (1990 through 2022) Inventory. The assumption is that
any waterbody used as a public drinking water source is managed in some capacity - by flow and/or volume
control. This data source added 418 reservoirs totaling 736,344 ha.
The National Inventory of Dams (NID) data are updated regularly. The version of NID used for the current Inventory
contains 47 new dams and updated values for "year of dam completion" for 975 dams relative to the previous
(1990 through 2021) Inventory data. Similarly, the National Wetlands Inventory (NWI) is periodically updated. The
NWI version used for the current Inventory has major updates for MS, ND, NM, and MT.
The net effect of these recalculations was an average annual increase in Cm emission estimates from reservoirs of
1.23 MMT CO2 Eq., or 4 percent, over the time series from 1990 to 2021 compared to the previous Inventory.
Planned Improvements
The EPA recently completed a survey of greenhouse gas emissions from 108 reservoirs in the conterminous United
States.72 The data will be used to develop country-specific emission factors for U.S. reservoirs to be used in the
1990 through 2024 Inventory submission.
Emissions from Flooded Land Remaining Flooded Land-Other
Constructed Waterbodies
The IPCC (IPCC 2019) provides emission factors for several types of "other constructed waterbodies" including
freshwater ponds and canals/ditches. IPCC (2019) describes ponds as waterbodies that are "...constructed by
excavation and/or construction of walls to hold water in the landscape for a range of uses, including agricultural
water storage, access to water for livestock, recreation, and aquaculture." Furthermore, the IPCC "Decision tree
for types of Flooded Land" (IPCC 2019, Fig. 7.2) defines a size threshold of 8 ha to distinguish reservoirs from
"other constructed waterbodies." For this Inventory, ponds are defined as managed flooded land that are 1) less
than 8 ha in surface area, and 2) not categorized as canals/ditches. IPCC (2019) further distinguishes saline versus
brackish ponds, with the former supporting lower CFU emissions than the latter. Activity data on pond salinity are
not uniformly available for the conterminous United States and all ponds in the inventory are assumed to be
freshwater. Ponds often receive high organic matter and nutrient loadings, may have low oxygen levels, and are
often sites of substantial CFU emissions from anaerobic sediments.
Canals and ditches (terms are used interchangeably) are linear water features constructed to transport water (i.e.,
stormwater drainage, aqueduct), to irrigate or drain land, to connect two or more bodies of water, or to serve as a
72 See https://www.epa.gov/air-research/research-emissions-us-reservoirs.
6-134 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
waterway for watercraft. The geometry and construction of canals and ditches varies widely and includes narrow
earthen channels (<1 m wide) and concrete lined aqueducts in excess of 50 m wide. Canals and ditches can be
extensive in many agricultural, forest and settlement areas, and may also be significant sources of emissions in
some circumstances.
Methane emissions from freshwater ponds in flooded land remaining flooded land increased by approximately 1
percent from 1990 to 2022. Methane emissions from canals and ditches have remained constant throughout the
time series because age data are not available for canals and ditches, thus they are assumed to be greater than 20-
years old in 1990 and are included in flooded land remaining flooded land throughout the time series. Overall, Cm
emissions from other constructed waterbodies have remained fairly constant since 1990 (Table 6-87 and Table
6-88).
Table 6-87: CH4 Emissions from Other Constructed Waterbodies in Flooded Land Remaining
Flooded Land (MMT CO2 Eq.)
Source
1990
2005
2018
2019
2020
2021
2022
Other Constructed Waterbodies
Canals and Ditches
2.31
2.31
2.3
2.3
2.3
2.3
2.3
Freshwater Ponds
11.4|
11.5|
11.5
11.5
11.5
11.5
11.5
Total
13.7
13.8
13.8
13.8
13.8
13.8
13.8
Note: Totals may not sum due to independent rounding.
Table 6-88: CH4 Emissions from Other Constructed Waterbodies in Flooded Land Remaining
Flooded Land (kt CH4)
Source
1990
2005
2018
2019
2020
2021
2022
Other Constructed Waterbodies
Canals and Ditches
80-9l
80-9l
80.9
80.9
80.9
80.9
80.9
Freshwater Ponds
406.6
411.0|
411.7
411.8
411.8
411.8
411.9
Total
487.5
491.9
492.6
492.6
492.7
492.7
492.7
Note: Totals may not sum due to independent rounding.
Florida and Louisiana have the greatest methane emissions from canals and ditches in the United States (Figure
6-13, Table 6-89). Presumably, most of these canals serve to drain the extensive wetland complexes in these states
(Davis, 1973). California has the third greatest methane emissions from canals and ditches. Canals and ditches in
California primarily serve to convey water from the mountains to urban and agricultural areas. Michigan and
Minnesota have the fourth and fifth largest methane emissions from canals and ditches. These systems serve to
drain historic wetlands to facilitate row-crop agriculture. Texas, Florida, and Georgia have the greatest methane
emissions from freshwater ponds, although states throughout the eastern United States make significant
contributions to the national total. These patterns of emissions are in accordance with the distribution of other
constructed waterbodies in the United States.
Table 6-89: CH4 Emissions from Other Constructed Waterbodies in Flooded Land Remaining
Flooded Land in 2022 (kt CH4)
State
Canals and Ditches
Freshwater Ponds
Total
Alabama
+
10.5
10.6
Alaska
+
+
+
Arizona
1.5
1.0
2.4
Arkansas
3.1
9.4
12.4
California
7.0
9.2
16.2
Colorado
2.9
4.8
7.7
Connecticut
+
1.8
1.8
Delaware
+
0.9
0.9
District of Columbia
+
+
+
Land Use, Land-Use Change, and Forestry 6-135
-------
Florida
15.6
30.7
46.2
Georgia
+
21.0
21.2
Hawaii
+
+
0.5
Idaho
1.7
2.4
4.1
Illinois
1.0
11.7
12.8
Indiana
1.7
10.6
12.3
Iowa
+
11.2
11.6
Kansas
+
15.4
15.5
Kentucky
+
7.7
7.9
Louisiana
9.4
5.9
15.3
Maine
+
3.5
3.5
Maryland
+
2.3
2.7
Massachusetts
+
2.3
2.3
Michigan
5.4
10.0
15.4
Minnesota
4.7
12.7
17.3
Mississippi
1.6
13.4
15.1
Missouri
2.4
20.7
23.1
Montana
2.0
10.5
12.5
Nebraska
2.0
9.1
11.1
Nevada
0.7
0.8
1.5
New Hampshire
+
1.0
1.1
New Jersey
+
3.0
3.4
New Mexico
0.8
2.1
2.9
New York
+
8.3
8.7
North Carolina
2.6
12.2
14.8
North Dakota
0.8
20.6
21.3
Ohio
0.8
8.9
9.7
Oklahoma
+
19.3
19.4
Oregon
1.0
3.6
4.6
Pennsylvania
+
4.1
4.1
Puerto Rico
+
+
+
Rhode Island
+
+
+
South Carolina
1.3
10.4
11.7
South Dakota
+
16.6
16.9
Tennessee
+
6.7
6.8
Texas
4.6
32.1
36.8
Utah
0.8
2.0
2.8
Vermont
+
0.8
0.8
Virginia
0.5
7.3
7.9
Washington
+
2.0
2.5
West Virginia
+
1.5
1.5
Wisconsin
+
3.8
4.2
Wyoming
0.9
4.8
5.7
Total
80.9
411.9
492.7
+ Indicates values less than 0.5 kt.
Note: Totals may not sum due to independent rounding.
6-136 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
Figure 6-13: 2022 CH4 Emissions from A) Ditches and Canals and B) Freshwater Ponds in
Flooded Land Remaining Flooded Land (kt CH4)
B. CH4 Emissions from Freshwater Ponds
. CZ7--
100 mi
500 mi
A. CH4 Emissions from Ditches and Canals
kt CH4 y
Hawaii
1000 mi
Methodology and Time-Series Consistency
Estimates of Cm emissions for other constructed waterbodies in flooded land remaining flooded Land follow the
Tier 1 methodology in IPCC (2019). All calculations are performed at the state level and summed to obtain national
estimates. Based on IPCC guidance, methane emissions from the surface of these flooded lands are calculated as
the product of flooded land surface area and an emission factor (Table 6-90). Although literature data on
greenhouse gas emissions from canals and ditches is relatively sparse, they have the highest default emission
factor of all flooded land types (Table 6-90). Default emission factors for freshwater ponds are on the higher end of
those for reservoirs. There are insufficient data to support climate-specific emission factors for ponds or canals and
ditches. Downstream emissions are not inventoried for other constructed waterbodies because 1) many of these
systems are not associated with dams (e.g., excavated ponds and ditches), and 2) there are insufficient data to
derive downstream emission factors for other constructed waterbodies that are associated with dams (IPCC 2019).
Table 6-90: IPCC (2019) Default CH4 Emission Factors for Surface Emissions from Other
Constructed Waterbodies in Flooded Land Remaining Flooded Land
Other Constructed Waterbody
Surface emission factor
(MT CH4 ha1 y1)
Freshwater ponds 0.183
Canals and ditches 0.416
Area estimates
Other constructed waterbodies were identified from the NHDWaterbody layer in the National Hydrography
Dataset Plus V2 (NHD),73 the National Inventory of Dams (NID),74 the National Wetlands Inventory (NWI),75 the
Navigable Waterways (NW) network,76 and the EPA's Safe Drinking Water Information System (SDWIS).77 The NHD
73 See https://www.uses.eov/core-science-svstems/nep/national-hvdroeraphv.
74 See https://nid.sec.usace.armv.mil.
75 See https://www.fws.eov/proeram/national-wetlands-inventorv/data-download.
76 See https://hifld-eeoplatform.opendata.arceis.com/maps/aaa3767c7d2b41f69e7528f99cf2fb76 0/about.
77 See https://www.epa.eov/enviro/sdwis-overview. Not publicly available due to security concerns.
Land Use, Land-Use Change, and Forestry 6-137
-------
only covers the conterminous United States, whereas the NID, NW and NWI also include Alaska, Hawaii, District of
Columbia, and Puerto Rico. The following paragraphs present the criteria used to identify other constructed
waterbodies in the NHD, NW, and NWI.
Waterbodies in the NHDWaterbody layer that were greater than 20-years old, less than 8 ha in surface area, not
identified as canal/ditch in NHD, and met any of the following criteria were considered freshwater ponds in
flooded land remaining flooded land: 1) the waterbody was classified "Reservoir" in the NHDWaterbody layer, 2)
the waterbody name in the NHDWaterbody layer included "Reservoir", 3) the waterbody in the NHDWaterbody
layer was located in close proximity (up to 100 m) to a dam in the NID, 4) the NHDWaterbody GNIS name was
similar to nearby NID feature (between 100 m to 1000 m), the waterbody intersected a drinking water intake.
EPA assumes that all features included in the NW are subject to water-level management to maintain minimum
water depths required for navigation and are therefore managed flooded lands. NW features that were less than 8
ha in surface area and not identified as canals/ditch (see below) were considered freshwater ponds. Only 2.1
percent of NW features met these criteria, and they were primarily associated with larger navigable waterways,
such as lock chambers on impounded rivers.
NWI features were considered "managed" if they had a special modifier value indicating the presence of
management activities (Figure 6-12). To be included in the flooded lands inventory, the managed flooded land had
to be wet or saturated for at least one season per year (see "Water Regime" in Figure 6-12). NWI features that met
these criteria, were less than 8 ha in surface area, and were not a canal/ditch (see beiow) were defined as
freshwater ponds.
Any NWI or NHD feature that intersected a drinking water intake point from SDWIS was assumed to be
"managed." The rational being that a waterbody used as a source for public drinking water is typically managed in
some capacity - by flow and/or volume control.
Canals and ditches, a subset of other constructed waterbodies, were identified in the NWI by their morphology.
Unlike a natural water body, canals and ditches are typically narrow, linear features with abrupt angular turns.
Figure 6-14 contrasts the unique shape of ditches/canals vs more natural water features.
Figure 6-14: Left: NWI Features Identified as Canals/Ditches (pink) by Unique Narrow,
Linear/Angular Morphology. Right: Non-Canal/Ditches with More Natural Morphology (blue)
This morphology was identified systematically using shape attributes in a decision tree model. A training set of 752
features were identified as either "ditch" or "not ditch" using expert judgment. The training set was used to train a
decision tree which was used to categorize millions of NWI features based on three shape attribute ratios (Figure
6-12).
6-138 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
Table 6-91: Predictors used in Decision Tree to Identify Canal/Ditches
Shape Length : # of Shape Vertices
Shape Area : Shape Length
Shape Area : # of Shape Vertices
The decision tree built a model using 80 percent of the 752 training features and used the 20 percent to validate
the model. The model was 93.1 percent accurate. Below are the validation results (Table 6-92).
Table 6-92: Validation Results for Ditch/Canal Classification Decision Tree
Truth
Prediction
Ditch/Canal Not Ditch/Canal
Ditch/Canal 49 5
Not Ditch/Canal 8 27
The decision tree model was then applied to the entire NWI dataset using the following shape attribute ratios
(Figure 6-15).
Figure 6-15: Structure of Decision Tree Used to Identify Canals/Ditches
NOT DITCH
0.71
L 100%
I— fygs f Area Length < 2.2 -©
DITCH
0.14
27% .
NOT DITCH
0 92
I 73% J
Area Vertices >= 9.8
Length_Vertlces >= 34
Area Vertices < 474
NOT DITCH
092
3%
NOT DITCH
096
67%
Surface areas for other constructed waterbodies were taken from NHD, NWI or the NW. If features from the NHD,
NWI, or the NW datasets overlapped, these areas were erased. The first step was to take the final NWI flooded
lands features and use it to identify overlapping NHD features. If the NHD feature had its center in a NWI feature, it
was removed from analysis. Next, remaining NHD features were erased from any remaining overlapping NWI
features. Final selections of NHD and NWI features were used to erase any overlapping NW waterbodies.
The age of other constructed waterbody features was determined by assuming the waterbody was created the
same year as a nearby (up to 100 m) NID feature. If no nearby NID feature was Identified, it was assumed the
waterbody was greater than 20-years old throughout the time series. No canal/ditch features were associated with
a nearby dam, therefore all canal/ditch features were assumed to be greater than 20-years old through the time
series.
Land Use, Land-Use Change, and Forestry 6-139
-------
For the year 2022, this Inventory contains 2,250,662 ha of freshwater ponds and 194,412 ha of canals and ditches
in flooded land remaining flooded land (Table 6-93). The surface area of freshwater ponds increased by 28,632 ha
(1.3 percent) from 1990 to 2022 due to flooded lands matriculating from land converted to flooded land to flooded
land remaining flooded land. All canals and ditches were assumed to be greater than 20-years old throughout the
time series, thus the surface area of these flooded lands is constant throughout the time series.
Table 6-93: National Surface Area Totals in Flooded Land Remaining Flooded Land - Other
Constructed Waterbodies (ha)
1990
2005
2018
2019
2020
2021
2022
Canals and ditches
194,412
194,412
194,412
194,412
194,412
194,412
194,412
Freshwater ponds
2,222,030
2,245,881
2,249,672
2,250,007
2,250,337
2,250,540
2,250,662
Total
2,416,442
2,440,292
2,444,084
2,444,418
2,444,749
2,444,951
2,445,074
Note: Totals may not sum due to independent rounding.
Canals and ditches in the conterminous United States are most abundant in the Gulf Coast states and California
(Figure 6-16A, Table ). Florida contains 19 percent of all U.S. canal and ditch surface area, most of which were
constructed in the early 1900s for drainage, flood protection, and water storage purposes. Freshwater ponds are
more widely distributed across the United States (Figure 6-16B, Table 6-95). Texas has the greatest surface area of
freshwater ponds, equivalent to 8 percent of all freshwater pond surface area in the United States, closely
followed by Florida.
Figure 6-16: 2022 Surface Area of A) Ditches and Canals and B) Freshwater Ponds in Flooded
Land Remaining Flooded Land (ha)
A. Area of Ditches and Canals B. Area Freshwater Ponds
Table 6-94: State Totals of Surface Area in Flooded Land Remaining Flooded Land— Canals
and Ditches (ha)
State
1990
2005
2018
2019
2020
2021
2022
Alabama
228
228
228
228
228
228
228
Alaska
115
115
115
115
115
115
115
Arizona
3,536
3,536
3,536
3,536
3,536
3,536
3,536
Arkansas
7,349
7,349
7,349
7,349
7,349
7,349
7,349
California
16,725
16,725
16,725
16,725
16,725
16,725
16,725
Colorado
6,874
6,874
6,874
6,874
6,874
6,874
6,874
Connecticut
28
28
28
28
28
28
28
Delaware
130
130
130
130
130
130
130
District of Columbia
1
1
1
1
1
1
1
Florida
37,482
37,482
37,482
37,482
37,482
37,482
37,482
Georgia
352
352
352
352
352
352
352
6-140 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
Hawaii
538
538
538
538
538
538
538
Idaho
4,027
4,027
4,027
4,027
4,027
4,027
4,027
Illinois
2,489 :
2,489 IIIIH'
2,489
2,489
2,489
2,489
2,489
Indiana
4,064
4,064
4,064
4,064
4,064
4,064
4,064
Iowa
867 1
867 it
867
867
867
867
867
Kansas
258
258
258
258
258
258
258
Kentucky
672 ;
672
672
672
672
672
672
Louisiana
22,565
22,565
22,565
22,565
22,565
22,565
22,565
Maine
56:
56
56
56
56
56
Maryland
967
967
967
967
967
967
967
Massachusetts
132 s
132
132
132
132
132
132
Michigan
12,897
12,897
12,897
12,897
12,897
12,897
12,897
Minnesota
11,235 1
11,235
11,235
11,235
11,235
11,235
11,235
Mississippi
3,936
3,936
3,936
3,936
3,936
3,936
3,936
Missouri
5,670
5,670
5,670
5,670
5,670
5,670
5,670
Montana
4,740
4,740
4,740
4,740
4,740
4,740
4,740
Nebraska
4,864
4,864
4,864
4,864
4,864
4,864
4,864
Nevada
1,587
1,587
1,587
1,587
1,587
1,587
1,587
New Hampshire 103 f
103 i
103
103
103
103
103
New Jersey
944
944
944
944
944
944
944
New Mexico
2,002 |
2,002
2,002
2,002
2,002
2,002
2,002
New York
925
¦
925
925
925
925
925
925
North Carolina
6,321. j®
6,321
6,321
6,321
6,321
6,321
6,321
North Dakota
1,819
1,819
1,819
1,819
1,819
1,819
1,819
Ohio
1,819 iii
1,819 ill
1,819
1,819
1,819
1,819
1,819
Oklahoma
278
278
278
278
278
278
278
Oregon
2,498
2,498
2,498
2,498
2,498
2,498
2,498
Pennsylvania
143
143
143
143
143
143
143
Puerto Rico
249 :
249
249
249
249
249
249
Rhode Island
1
1
1
1
1
1
1
South Carolina
3,226
3,226
3,226
3,226
3,226
3,226
3,226
South Dakota
703
703
703
703
703
703
703
Tennessee
442
442
442
442
442
442
442
Texas
11,152
11,152
11,152
11,152
11,152
11,152
11,152
Utah
1,875 1
1,875 :
1,875
1,875
1,875
1,875
1,875
Vermont
95
95
95
95
95
95
95
Virginia
1,306
1,306
1,306
1,306
1,306
1,306
1,306
Washington
1,125 "
1,125
1,125
1,125
1,125
1,125
1,125
West Virginia
28
28
28
28
28
28
28
Wisconsin
887
887
887
887
887
887
887
Wyoming
2,086 |
2,086 1
2,086
2,086
2,086
2,086
2,086
Total
194,412
194,412
194,412
194,412
194,412
194,412
194,412
Table 6-95:
State Totals of Surface Area in Flooded Land Remaining Flooded Land—
Freshwater Ponds (ha)
State
1990
2005
2018
2019
2020
2021
2022
Alabama
57,034
57,342
57,355
57,355
57,355
57,355
57,355
Alaska
2,367 =
2,370 mm
2,370
2,370
2,370
2,370
2,370
Arizona
5,199
5,236
5,249
5,249
5,253
5,253
5,253
Arkansas
50,880 |
51,211
51,211
51,211
51,211
51,211
51,211
California
50,219
50,426
50,499
50,504
50,511
50,513
50,519
Colorado
26,174
26,448 mill
26,478
26,479
26,480
26,480
26,494
Connecticut
9,630
9,697
9,699
9,699
9,699
9,699
9,699
Delaware
4,717
4,7211
4,721
4,721
4,721
4,721
4,721
District of Columbia 16
16
16
16
16
16
16
Florida
167,317 !
167,453
167,496
167,502
167,502
167,508
167,508
Land Use, Land-Use Change, and Forestry 6-141
-------
Georgia
113,254
114,898
114,969
114,972
114,972
114,972
114,972
Hawaii
1,580 :
1,592
1,595
1,595
1,595
1,595
1,595
Idaho
13,220
13,352
13,359
13,359
13,359
13,360
13,360
Illinois
63,516 i;
64,044
64,144
64,149
64,159
64,160
64,169
Indiana
57,593
58,065
58,170
58,175
58,175
58,175
58,185
Iowa
57,450
59,612 III:
60,745
60,929
61,051
61,147
61,168
Kansas
81,828
83,900
83,976
83,985
83,985
84,002
84,004
Kentucky
41,427
41,808
41,837
41,837
41,837
41,837
41,837
Louisiana
32,085
32,210
32,221
32,221
32,226
32,226
32,226
Maine
19,102 !
19,149
19,159
19,159
19,159
19,159
19,159
Maryland
12,569
12,739
12,810
12,810
12,812
12,815
12,818
Massachusetts
12'359 1
12,413
12,457
12,464
12,470
12,472
12,476
Michigan
54,525
54,672
54,701
54,701
54,709
54,709
54,709
Minnesota
68,801
69,082 I
69,173
69,176
69,202
69,210
69,220
Mississippi
72,832
73,209
73,336
73,343
73,363
73,375
73,383
Missouri
109,573 I
112,993 =
113,068
113,071
113,073
113,077
113,079
Montana
56,860
57,246
57,263
57,268
57,269
57,269
57,269
Nebraska
48,051
49,380
49,649
49,667
49,697
49,706
49,709
Nevada
4,452
4,455
4,508
4,509
4,512
4,512
4,515
New Hampshire
5,427 I
5,526
5,585
5,585
5,586
5,587
5,587
New Jersey
16,192
16,232
16,253
16,253
16,253
16,253
16,253
New Mexico
11,379 j
11,394
11,398
11,401
11,401
11,401
11,406
New York
45,224
45,485
45,590
45,592
45,592
45,598
45,598
North Carolina
66,205 1
66,661 ii
66,744
66,744
66,747
66,750
66,751
North Dakota
112,310
112,384
112,469
112,475
112,485
112,489
112,492
Ohio
48,028
48,393 i
48,591
48,605
48,637
48,651
48,656
Oklahoma
103,243
105,224
105,288
105,304
105,318
105,324
105,333
Oregon
19,304 E
1.9,487 1
19,532
19,534
19,539
19,539
19,539
Pennsylvania
22,018
22,256
22,289
22,289
22,289
22,289
22,289
Puerto Rico
708 =
708 ¦
708
708
708
708
708
Rhode Island
2,204
2,213
2,220
2,220
2,220
2,220
2,220
South Carolina
55<794 111
56'456 =
56,673
56,682
56,686
56,686
56,686
South Dakota
90,237
90,447
90,504
90,515
90,516
90,521
90,521
Tennessee
35,927
36,307
36,332
36,337
36,343
36,344
36,344
Texas
172,580
175,497
175,569
175,574
175,575
175,575
175,575
Utah
10,703
10,764 Hi
10,772
10,772
10,773
10,773
10,773
Vermont
4,316
4,381
4,392
4,392
4,392
4,392
4,392
Virginia
39,938
39,996 I
40,000
40,000
40,000
40,000
40,000
Washington
10,943
11,081
11,119
11,119
11,122
11,123
11,123
West Virginia
8,027
8,156 |
8,166
8,166
8,166
8,166
8,166
Wisconsin
20,845
20,989
21,003
21,003
21,003
21,003
21,003
Wyoming
25,851 :
26,106 :
26,243
26,243
26,244
26,246
26,250
Total
2,222,030
2,245,881
2,249,672
2,250,007
2,250,337
2,250,540
2,250,662
Uncertainty
Uncertainty in estimates of Cm emissions from other constructed waterbodies (ponds, canals/ditches) in flooded
land remaining flooded land (Table 6-96) are estimated using IPCC Approach 2 and include uncertainty in the
default emission factors and the flooded land area inventory. Uncertainty in default emission factors is provided in
the 2019 Refinement to the 2006 IPCC Guidelines (IPCC 2019). Uncertainties in the spatial data include 1)
uncertainty in area estimates from the NHD, NWI, and NW, and 2) uncertainty in the location of dams in the NID.
Overall uncertainties in these spatial datasets are unknown, but uncertainty for remote sensing products is
assumed to be ± 10 to 15 percent based on IPCC guidance (IPCC 2003). An uncertainty range of ± 15 percent for
the flooded land area estimates is assumed and is based on expert judgment.
6-142 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
Table 6-96: Approach 2 Quantitative Uncertainty Estimates for CH4 Emissions from Other
Constructed Waterbodies in Flooded Land Remaining Flooded Land
Source
Gas
2022 Emission Estimate
Uncertainty Range Relative to Emission Estimate-1
(MMTCOEq.)
(MMT CO
Eq.)
(%)
Lower Bound
Upper Bound
Lower Bound
Upper Bound
Canals and ditches
ch4
? a
2.1
2.4
-5.1%
+7.0%
Freshwater pond
ch4
11.5
11.5
11.5
-0.04%
+0.04%
Total
ch4
13.8
13.7
13.9
-0.8%
+1.0%
a Range of emission estimates predicted by Monte Carlo stochastic simulation for a 95 percent confidence interval.
QA/QC and Verification
The National Hydrography Data (NHD) is managed by the USGS in collaboration many other federal, state, and
local entities. Extensive QA/QC procedures are incorporated into the curation of the NHD. The National Inventory
of Dams (NID) is maintained by the U.S. Army Corps of Engineers (USACE) in collaboration with the Federal
Emergency Management Agency (FEMA) and state regulatory offices. USACE resolves duplicative and conflicting
data from 68 data sources, which helps obtain the more complete, accurate, and updated NID.78 The Navigable
Waterways (NW) dataset is part of the U.S. Department of Transportation (USDOT)/Bureau of Transportation
Statistics (BTS) National Transportation Atlas Database (NTAD). The NW is a comprehensive network database of
the nation's navigable waterways updated on a continuing basis. U.S. Fish and Wildlife Service is the principal
agency in charge of wetland mapping including the National Wetlands Inventory (NWI). Quality and consistency of
the Wetlands Layer is supported by federal wetlands mapping and classification standards, which were developed
under the oversight of the Federal Geographic Data Committee (FGDC) with input by the FGDC Wetlands
Subcommittee. This dataset is part of the FGDC Water-Inland Theme, which is co-chaired by the FWS and the U.S.
Geological Survey. The EPA's Safe Drinking Water Information System (SDWIS) tracks information on drinking
water contamination levels as required by the 1974 Safe Drinking Water Act and its 1986 and 1996 amendments.
General QA/QC procedures were applied to activity data, documentation, and emission calculations consistent
with the U.S. Inventory QA/QC plan, which is in accordance with Vol. 1 Chapter 6 of 2006IPCC Guidelines (see
Annex 8 for more details). All calculations were executed independently in Excel and R. Ten percent of state and
national totals were randomly selected for comparison between the two approaches to ensure there were no
computational errors.
Recalculations Discussion
The EPA's SDWIS is a new data source used in the current (1990 through 2022) Inventory. The assumption is that
any waterbody used as a public drinking water source is managed in some capacity—by flow and/or volume
control. This data source added 54 features totaling 173 ha of other constructed waterbodies.
The National Inventory of Dams (NID) data are updated regularly. The version of NID used for the current (1990
through 2022) Inventory contains 47 new dams and updated values for "year of dam completion" for 975 dams
relative to the previous (1990 through 2021) Inventory data. Similarly, the National Wetlands Inventory (NWI) is
periodically updated. The NWI version used for the current Inventory has major updates for MS, ND, NM, and MT.
The net effect of these recalculations was an average annual decrease in CH4 emission estimates from other
constructed waterbodies of 2.7 MMT CO2 Eq., or 17 percent, over the time series from 1990 to 2021 compared to
the previous Inventory.
78 See https://www.epa.Eov/national-aauatic-resource-survevs/national-lakes-assessment-2017-qualitv-assurance-proiect-
plan.
Land Use, Land-Use Change, and Forestry 6-143
-------
Planned Improvements
Default emission factors for canals/ditches were derived from a global dataset that include few measurements
from U.S. systems. The EPA plans to conduct a literature survey to determine if sufficient data are available to
derive a country-specific emission factor for the 1990 through 2024 Inventory submission.
Canal and ditch surface area included here may overlap with ditches and canals included in Cm emission estimates
for ditches draining inland organic soils (IPCC 2013, section 2.2.2.1). EPA plans to reconcile ditch/canal surface
areas between the two managed land types (flooded land vs drained inland organic soils) in the next (i.e., 1990
through 2023) Inventory.
Features less than 8 ha in the NW that were not identified as Canal/Ditch were defined as freshwater ponds. Many
of these features are lock chambers connected to an upstream reservoir. These systems likely have emission rates
more similar to a reservoir than freshwater pond. In the next (1990 through 2023) Inventory these systems will be
classified as reservoirs.
6.9 Land Converted to Wetlands (CRT
Source Category 4D2)
Emissions and Removals from Land Converted to Vegetated
Coastal Wetlands
Land converted to vegetated coastal wetlands occurs as a result of inundation of unprotected low-lying coastal
areas with gradual sea-level rise, flooding of previously drained land behind hydrological barriers, and through
active restoration and creation of coastal wetlands through removal of hydrological barriers. Based upon NOAA C-
CAP, wetlands are subdivided into freshwater (Palustrine) and saline (Estuarine) classes and further subdivided
into emergent marsh, scrub shrub and forest classes All other land categories (i.e., forest land, cropland, grassland,
settlements and other lands) are identified as having some area converting to vegetated coastal wetlands. This
Inventory does not include land converted to unvegetated open water coastal wetlands (see Planned
Improvements section below). Between 1990 and 2022 the rate of annual transition for land converted to
vegetated coastal wetlands ranged from 0 to 2,650 ha per year, depending on the type of land converted.79
Conversion rates from forest land were relatively consistent between 1990 and 2010 (ranging between 2,409 and
2,650 ha) and decreased to 625 ha starting in 2011; the majority of these conversions resulted in increases in the
area of palustrine wetlands, which also initiates Cm emissions when lands are inundated with fresh water.80 Little
to no conversion of cropland, grassland, settlement, or other lands to vegetated coastal wetlands occurred during
the reporting period, with converted areas ranging from 0 to 25 ha per year.81
79 Data from C-CAP; see https://coast.noaa.gov/dieitalcoast/tools/. Accessed October 2023.
80 Currently, the C-CAP dataset categorizes coastal wetlands as either palustrine (fresh water) or estuarine (presence of saline
water). This classification does not differentiate between estuarine wetlands with salinity < 18 ppt (when methanogenesis
begins to occur) and those that are >18 ppt (where negligible to no CH4 is produced); therefore, it is not possible at this time to
account for CH4 emissions from estuarine wetlands in the Inventory.
81 At the present stage of Inventory development, coastal wetlands are not explicitly shown in the land representation analysis
(Section 6.1) while work continues harmonizing data from NOAA's Coastal Change Analysis Program (C-CAP) with NRI, FIA and
NLDC data used to compile the land representation (NOAA OCM 2020).
6-144 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
Conversion to coastal wetlands resulted in a biomass carbon stock loss of 0.2 MMT CO2 Eq. (0.03 MMT C) in 2022
(Table 6-97 and Table 6-98). Loss of forest biomass through conversion of forest lands to vegetated coastal
wetlands is the primary driver behind biomass carbon stock change being a source rather than a sink across the
time series. Conversion of cropland, grassland, settlement and other lands result in a net increase in biomass
stocks. Conversion of lands to vegetated coastal wetlands resulted in a DOM loss of 0.03 MMT CO2 Eq. (0.008 MMT
C) in 2022 (Table 6-97 and Table 6-98), which is driven by the loss of DOM when forest land is converted to
vegetated coastal wetlands. This is likely an overestimate of loss because wetlands inherently preserve dead
organic material. Conversion of cropland, grassland, settlement and other land results in a net increase in DOM
Across all time periods, soil carbon accumulation resulting from lands converted to vegetated coastal wetlands is a
carbon sink and has ranged between -0.14 and -0.3 MMT CO2 Eq. (-0.04 and -0.07 MMT C; Table 6-97 and Table
6-98). Conversion of lands to coastal wetlands resulted in CH4 emissions of 0.17 MMT CO2 Eq. (6.1 kt CH4) in 2022
(Table 6-99). Methane emissions due to the conversion of lands to vegetated coastal wetlands are largely the
result of forest land converting to palustrine emergent and scrub shrub coastal wetlands in warm temperate
climates. Emissions were the highest between 1990 and 2002 (0.28 MMT CO2 Eq., 10.0 kt CH4) and have
continually decreased to current levels. This decrease was driven by a reduction in the rate of conversion of forest
land to palustrine scrub-shrubs and emergent wetlands.
Table 6-97: Net CO2 Flux from Carbon Stock Changes in Land Converted to Vegetated Coastal
Wetlands (MMT C02 Eq.)
Land Use/Carbon Pool
1990
2005
2018
2019
2020
2021
2022
Cropland Converted to Vegetated Coastal
i
i
Wetlands
(+)
(+)
(+)
(+)
(+)
(+)
(+)
Biomass C Stock
w I
(+)
(+)
(+)
(+)
(+)
Soil C Stock
(+)
<+)
(+)
(+)
(+)
(+)
(+)
Forest Land Converted to Vegetated
1
iiiiiii
Coastal Wetlands
0.49
0.50
(+)
0.01
0.02
0.03
0.04
Biomass C Stock
0.62
0.62
0.13
0.13
0.13
0.13
0.13
Dead Organic Matter C Flux
0.111
0.12 1
0.03
0.03
0.03
0.03
0.03
Soil C Stock
10.24)
(0.24)
(0.16)
(0.15)
(0.14)
(0.13)
(0.12)
Grassland Converted to Vegetated Coastal
i
Wetlands
(+)
(+)
(+)
(+)
(+)
(+)
(+)
Biomass C Stock
(+)
(+)
(+)
(+)
(+)
(+)
(+)
Soil C Stock
wl
(+) 1
(+)
(+)
(+)
(+)
(+)
Other Land Converted to Vegetated
=
Coastal Wetlands
(0.03)
(0.03)
(0.02)
(0.02)
(0.02)
(0.02)
(0.02)
Biomass C Stock
(0.01) =
(°.°2)
(0.01)
(0.01)
(0.01)
(0.01)
(0.01)
Soil C Stock
(0.02)
(0.02)
(0.02)
(0.02)
(0.02)
(0.02)
(0.02)
Settlements Converted to Vegetated
Coastal Wetlands
(+)
(+)
(+)
(+)
(+)
(+)
(+)
Biomass C Stock
(+) ^
(+)
(+)
(+)
(+)
(+)
(+)
Soil C Stock
(+) 1
(+) 1
(+)
(+)
(+)
(+)
(+)
Total Biomass Flux
0.60
0.60
0.12
0.12
0.12
0.12
0.12
Total Dead Organic Matter Flux
0.11
0.12
0.03
0.03
0.03
0.03
0.03
Total Soil C Flux
(0.25)
(0.25)
(0.18)
(0.17)
(0.16)
(0.15)
(0.14)
Total Flux
0.46
0.47
(0.02)
(0.02)
(0.01)
0.00
0.01
+ Absolute value does not exceed 0.005 MMT C02 Eq.
Note: Totals may not sum due to independent rounding. Parentheses indicate net sequestration.
Table 6-98: Net CO2 Flux from Carbon Stock Changes in Land Converted to Vegetated Coastal
Wetlands (MMT C)
Land Use/Carbon Pool 1990 2005 2018 2019 2020 2021 2022~
Cropland Converted to Vegetated Coastal (+) | (+) | (+) (+) (+) (+) (+)
Land Use, Land-Use Change, and Forestry 6-145
-------
Wetlands
Biomass C Stock
M 1
(+) I
(+)
(+)
(+)
(+)
(+)
Soil C Stock
(+)
(+)
(+)
(+)
(+)
(+)
(+)
Forest Land Converted to Vegetated
1
I
Coastal Wetlands
0.13
0.14
+
+
0.006
0.008
0.01
Biomass C Stock
0.17
0.17
0.04
0.04
0.04
0.04
0.04
Dead Organic Matter C Flux
0.03 f
0.03 =
0.01
0.01
0.01
0.01
0.01
Soil C Stock
(0.06)
(0.06)
(0.04)
(0.04)
(0.04)
(0.04)
(0.03)
Grassland Converted to Vegetated
¦
Coastal Wetlands
(+)
(+) B!
(+)
(+)
(+)
0
(+)
Biomass C Stock
(+)
(+)
(+)
(+)
(+)
(+)
(+)
Soil C Stock
(+) i
(+)
(+)
(+)
(+)
(+)
(+)
Other Land Converted to Vegetated
Coastal Wetlands
(0.01)
(0.01)
(0.01)
(0.01)
(0.01)
(0.01)
(0.01)
Biomass C Stock
(+)
(0.01) =
(+)
(+)
(+)
(+)
(+)
Soil C Stock
(+)
(0.01)
(+)
(+)
(+)
(+)
(+)
Settlements Converted to Vegetated
Coastal Wetlands
(+)
(+)
(+)
(+)
(+)
(+)
(+)
Biomass C Stock
(+)
(+)
(+)
(+)
(+)
(+)
(+)
Soil C Stock
(+) §
(+)
(+)
(+)
(+)
(+)
Total Biomass Flux
0.16
0.16
0.03
0.03
0.03
0.03
0.03
Total Dead Organic Matter Flux
0.03
0.03
0.01
0.01
0.01
0.01
0.01
Total Soil C Flux
(0.07)
(0.07)
(0.05)
(0.05)
(0.04)
(0.04)
(0.04)
Total Flux
0.13
0.13
(0.01)
(+)
(+)
+
+
+ Absolute value does not exceed 0.005 MMT C.
Notes: Totals may not sum due to independent rounding. Parentheses indicate net sequestration.
Table 6-99: CH4 Emissions from Land Converted to Vegetated Coastal Wetlands (MMT CO2
Eq. and kt CH4)
Land Use/Carbon Pool
1990
2005
2018
2019
2020
2021
2022
Cropland Converted to Vegetated Coastal
Wetlands
CH4 Emissions (MMT C02 Eq.)
+:::::
+ =
+
+
+
+
+
CH4 Emissions (kt CH4)
+
0.01
0.04
0.04
0.05
0.05
0.05
Forest Land Converted to Vegetated
1
Coastal Wetlands
1
1
CH4 Emissions (MMT C02 Eq.)
0.28
0.27
0.19
0.18
0.17
0.16
0.15
CH4 Emissions (kt CH4)
9-88 1
9-74 1
6.85
6.48
6.10
5.76
5.41
Grassland Converted to Vegetated Coastal
Wetlands
CH4 Emissions (MMT C02 Eq.)
+ I
IB
+ ¦
+
+
+
+
+
CH4 Emissions (kt CH4)
0.01
0.01
0.07
0.07
0.08
0.08
0.09
Other Land Converted to Vegetated
1
Coastal Wetlands
iiiiiii
1
CH4 Emissions (MMT C02 Eq.)
+
+
0.01
0.01
0.01
0.01
0.02
CH4 Emissions (kt CH4)
0-08 a
0.14 i
0.43
0.47
0.50
0.52
0.54
Settlements Converted to Vegetated
Coastal Wetlands
CH4 Emissions (MMT C02 Eq.)
+1
IIR
+1
+
+
+
+
+
CH4 Emissions (kt CH4)
0.01
+
+
+
+
+
+
Total CH.i Emissions (MMTCO' Eq.)
0.28
0.28
0.21
0.20
0.19
0.18
0.17
Total CH.i Emissions (kt CH.i)
9.98
9.91
7.39
7.06
6.73
6.41
6.09
+ Absolute value does not exceed 0.005 MMT C02 Eq. or 0.005 kt CH4.
Note: Totals may not sum due to independent rounding.
6-146 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
Methodology and Time-Series Consistency
The following section provides a description of the methodology used to estimate changes in biomass, dead
organic matter and soil carbon stocks and Cm emissions for land converted to vegetated coastal wetlands.
Methodological recalculations were applied to the entire time series to ensure time-series consistency from 1990
through 2022.
Biomass Carbon Stock Changes
Biomass carbon stocks for land converted to vegetated coastal wetlands are estimated for palustrine and estuarine
marshes for land below the elevation of high tides (taken to be mean high water spring tide elevation) and as far
seawards as the extent of intertidal vascular plants within the U.S. land representation according to the national
LiDAR dataset, the national network of tide gauges and land use histories recorded in the 1996, 2001, 2005, 2011,
and 2016 NOAA C-CAP surveys (NOAA OCM 2020). Both federal and non-federal lands are represented.
Delineating vegetated coastal wetlands from ephemerally flooded upland grasslands represents a particular
challenge in remote sensing. Moreover, at the boundary between wetlands and uplands, which may be gradual on
low lying coastlines, the presence of wetlands may be ephemeral depending upon weather and climate cycles and
as such, impacts on the emissions and removals will vary over these time frames. Trends in land cover change are
extrapolated to 1990 and 2021 from these datasets using the C-CAP change data closest in date to a given year.
Biomass is not sensitive to soil organic content. Aboveground biomass carbon stocks for non-forested coastal
wetlands are derived from a national assessment combining field plot data and aboveground biomass mapping by
remote sensing (Byrd et al. 2017; Byrd et al. 2018; Byrd et al. 2020). Aboveground biomass carbon removal data
for all subcategories are not available and thus assumptions were applied using expert judgment about the most
appropriate assignment to a disaggregation of a community class. The aboveground biomass carbon stock for
estuarine forested wetlands (dwarf mangroves that are not classified as forests due to their stature) is derived
from a meta-analysis by Lu and Megonigal (201782). Root to shoot ratios from the Wetlands Supplement were used
to account for belowground biomass, which were multiplied by the aboveground carbon stock (IPCC 2014) and
summed with aboveground biomass to obtain total biomass carbon stocks. Aboveground biomass carbon stocks
for forest land, cropland, and grassland that are lost with the conversion to vegetated coastal wetlands were
derived from Tier 1 default values (IPCC 2006; IPCC 2019). Biomass carbon stock changes are calculated by
subtracting the biomass carbon stock values of each land use category (i.e., forest land, cropland, and grassland)
from those of vegetated coastal wetlands in each climate zone and multiplying that value by the corresponding C-
CAP derived area gained that year in each climate zone. The difference between the stocks is reported as the stock
change under the assumption that the change occurred in the year of the conversion. The total coastal wetland
biomass carbon stock change is accounted for during the year of conversion; therefore, no interannual changes are
calculated during the remaining years it is in the category.
Dead Organic Matter
Dead organic matter (DOM) carbon stocks, which include litter and dead wood stocks, are accounted for in
subtropical estuarine forested wetlands for lands converted to vegetated coastal wetlands across all years. Tier 1
estimates of mangrove DOM carbon stocks were used for subtropical estuarine forested wetlands (IPCC 2014).
Neither Tier 1 or 2 data on DOM are currently available for either palustrine or estuarine scrub/shrub wetlands for
any climate zone or estuarine forested wetlands in climates other than subtropical climates. Tier 1 DOM C stocks
for forest land converted to vegetated coastal wetlands were derived from IPCC (2019) to account for the loss of
DOM that occurs with conversion. Changes in DOM are assumed to be negligible for other land use conversions
(i.e., other than forest land) to coastal wetlands based on the Tier 1 method in IPCC (2006). Trends in land cover
change are derived from the NOAA C-CAP dataset and extrapolated to cover the entire 1990 through 2022 time
series. Dead organic matter removals are calculated by multiplying the C-CAP derived area gained that year by the
82 See https://github.com/Smithsonian/Coastal-Wetland-NGGl-Data-Public; accessed October 2023.
Land Use, Land-Use Change, and Forestry 6-147
-------
difference between Tier 1 DOM carbon stocks for vegetated coastal wetlands and forest land. The difference
between the stocks is reported as the stock change under the assumption that the change occurred in the year of
the conversion. The coastal wetland DOM stock is assumed to be in steady state once established in the year of
conversion; therefore, no interannual changes are calculated.
Soil Carbon Stock Changes
Soil carbon removals are estimated for land converted to vegetated coastal wetlands across all years. Soil carbon
stock changes, stratified by climate zones and wetland classes, are derived from a synthesis of peer-reviewed
literature83 (Lynch 1989; Orson et al. 1990; Kearny & Stevenson 1991; Thorn et al. 1992; Roman et al. 1997; Craft
et al. 1998; Orson et al. 1998; Merrill 1999; Weis et al. 2001; Hussein et al. 2004; Church et al. 2006; Koster et al.
2007; Drexler et al. 2009; Boyd 2012; Callaway et al. 2012 a & b; Bianchi et al. 2013; Drexler et al. 2013; Watson
and Byrne 2013; Breithaupt et al. 2014; Crooks et al. 2014; Weston et al. 2014; Smith et al. 2015; Villa & Mitsch
2015; Boyd and Sommerfield 2016; Marchio et al. 2016; Noe et al. 2016; Arriola and Cable 2017; Boyd et al. 2017;
Gerlach et al. 2017; Giblin and Forbrich 2018; Krauss et al. 2018; Abbott et al. 2019; Drexler et al. 2019; Poppe and
Rybczyk 2019; Ensign et al. 2020; Kemp et al. 2020; Lagomasino et al. 2020; Luk et al. 2020; McTigue et al. 2020;
Peck et al. 2020; Vaughn et al. 2020; Weston et al. 2020; Arias-Ortiz et al. 2021; Baustian et al. 2021; Allen et al.
2022; Miller et al. 2022). To estimate soil carbon stock changes, no differentiation is made for soil type (i.e.,
mineral, organic). Soil C removal data for all subcategories are not available and thus assumptions were applied
using expert judgment about the most appropriate assignment to a disaggregation of a community class.
As per IPCC (2014) guidance, land converted to vegetated coastal wetlands is assumed to remain in this category
for up to 20 years before transitioning to vegetated coastal wetlands remaining vegetated coastal wetlands. Tier 2
level estimates of soil carbon stock changes associated with annual soil carbon accumulation from land converted
to vegetated coastal wetlands were developed using country-specific soil carbon removal factors multiplied by
activity data of land area for land converted to vegetated coastal wetlands for a given year in addition to the
previous 19-year cumulative area. Guidance from the Wetlands Supplement allows for the rate of soil carbon
accumulation to be instantaneously equivalent to that in natural settings and that soil carbon accumulation is
initiated when natural vegetation becomes established; this is assumed to occur in the first year of conversion. No
loss of soil carbon as a result of land conversion to coastal wetlands is assumed to occur. Since the C-CAP coastal
wetland area dataset begins in 1996, the area converted prior to 1996 is assumed to be the same as in 1996.
Similarly, the coastal wetland area data for 2017 through 2022 is assumed to be the same as in 2016. The
methodology follows Eq. 4.7, Chapter 4 of the IPCC Wetlands Supplement (IPCC 2014) and is applied to the area of
land converted to vegetated coastal wetlands on an annual basis.
Soil Methane Emissions
Tier 1 estimates of Cm emissions for land converted to vegetated coastal wetlands are derived from the same
wetland map used in the analysis of wetland soil carbon fluxes for palustrine wetlands, and are produced from C-
CAP, LiDAR and tidal data, in combination with default Cm emission factors provided in Table 4.14 of the IPCC
Wetlands Supplement. The methodology follows Eq. 4.9, Chapter 4 of the IPCC Wetlands Supplement and a global
warming potential of 28 is used (IPCC 2013). Because land converted to vegetated coastal wetlands is held in this
category for up to 20 years before transitioning to vegetated coastal wetlands remaining to vegetated coastal
wetlands, Cm emissions in a given year represent the cumulative area held in this category for that year and the
prior 19 years.
Uncertainty
Underlying uncertainties in estimates of soil carbon removal factors, biomass change, DOM, and Cm emissions
83 Coastal Carbon Network (2023). Database: Coastal Carbon Library (Version 1.0.0). Smithsonian Environmental Research
Center. Dataset. https://doi.org/10.25573/serc.21565671. Accessed October 2023
6-148 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
include error in uncertainties associated with Tier 2 literature values of soil carbon removal estimates, biomass
stocks, DOM, and IPCC default Cm emission factors, uncertainties linked to interpretation of remote sensing data,
as well as assumptions that underlie the methodological approaches applied.
Uncertainty specific to coastal wetlands include differentiation of palustrine and estuarine community classes,
which determines what flux is applied. Because mean soil and biomass carbon removal for each available
community class are in a fairly narrow range, the same overall uncertainty was assigned to each, respectively (i.e.,
applying approach for asymmetrical errors, the largest uncertainty for any soil carbon stock value should be
applied in the calculation of error propagation; IPCC 2000). Uncertainties for Cm flux are the Tier 1 default values
reported in the Wetlands Supplement. Overall uncertainty of the NOAA C-CAP remote sensing product is 15
percent. This is in the range of remote sensing methods (±10 to 15 percent; IPCC 2003). However, there is
significant uncertainty in salinity ranges for tidal and non-tidal estuarine wetlands and activity data used to
estimate the Cm flux (e.g., delineation of an 18 ppt boundary), which will need significant improvement to reduce
uncertainties. The combined uncertainty was calculated by summing the squared uncertainty for each individual
source (C-CAP, soil, biomass, and DOM) and taking the square root of that total.
Uncertainty estimates are presented in Table 6-100 for each carbon pool and the Cm emissions. The combined
uncertainty is 42.6 percent above and below the estimate of 0.17 MMT CO2 Eq. In 2022, the total flux was 0.17
MMT CO2 Eq., with lower and upper estimates of 0.10 and 0.24 MMT CO2 Eq.
Table 6-100: Approach 1 Quantitative Uncertainty Estimates for Carbon Stock Changes
occurring within Land Converted to Vegetated Coastal Wetlands in 2022 (MMT CO2 Eq. and
Percent)
Source
2022 Estimate
Uncertainty Range Relative to Estimate-1
(MMTCO. Eq.)
(MMT CO
Eq.)
(%)
Lower
Upper
Lower
Upper
Bound
Bound
Bound
Bound
Biomass C Stock Flux
0.12
0.1
0.15
-20.0%
20.0%
Dead Organic Matter Flux
0.03
0.02
0.03
-25.8%
25.8%
Soil C Stock Flux
(0.14)
(0.2)
(0.1)
-17.7%
17.7%
Methane Emissions
0.17
0.12
0.22
-29.9%
29.9%
Total Uncertainty
0.18
0.11
0.26
-42.2%
42.2%
a Range of flux estimates based on error propagation at 95 percent confidence interval.
Notes: Totals may not sum due to independent rounding. Parentheses indicate net sequestration.
QA/QC and Verification
NOAA provided National LiDAR Dataset, tide data, and C-CAP land cover and land cover change mapping, all of
which are subject to agency internal mandatory QA/QC assessment (McCombs et al. 2016). QA/QC and verification
of soil carbon stock dataset has been provided by the Smithsonian Environmental Research Center and coastal
wetland inventory team leads. Biomass carbon stocks are derived from peer-review literature, reviewed by U.S.
Geological Survey prior to publishing, by the peer-review process during publishing, and by the coastal wetland
inventory team leads prior to inclusion in the Inventory and from IPCC reports. As a QC step, a check was
undertaken confirming that coastal wetlands recognized by C-CAP represent a subset of wetlands recognized by
the NRI for marine coastal states. A team of two evaluated and verified there were no computational errors within
the calculation worksheets. Soil carbon stock, emissions/removals data are based upon peer-reviewed literature
and Cm emission factors are derived from the Wetlands Supplement.
Recalculations Discussion
A recalculation of emission factors for soil carbon accretion rates was performed using the same methodology and
criteria as in Lu and Megonigal (2017) and described above. This new analysis incorporated data published since
Land Use, Land-Use Change, and Forestry 6-149
-------
2016 and other relevant data that were not previously included. The updated synthesis resulted in a general
increase in soil carbon accumulation rates for estuarine emergent and scrub/shrub wetlands, which resulted in a
minimal annual average increase (0.001 MMT CO2 Eq.) for the entire time series.
Planned Improvements
Currently, the only coastal wetland conversion that is reported in the Inventory is lands converted to vegetated
coastal wetlands. The next (1990 through 2023) Inventory submission is expected to include carbon stock change
data for lands converted to unvegetated open water coastal wetlands.
Land Converted to Flooded Land
Flooded lands are defined as water bodies where human activities have 1) caused changes in the amount of
surface area covered by water, typically through water level regulation (e.g., constructing a dam), 2) waterbodies
where human activities have changed the hydrology of existing natural waterbodies thereby altering water
residence times and/or sedimentation rates, in turn causing changes to the natural production of greenhouse
gases, and 3) waterbodies that have been created by excavation, such as canals, ditches and ponds (IPCC 2019).
Flooded lands include waterbodies with seasonally variable degrees of inundation but would be expected to retain
some inundated area throughout the year under normal conditions.
Flooded lands are broadly classified as "reservoirs" or "other constructed waterbodies" (IPCC 2019). Reservoirs are
defined as flooded land greater than 8 ha and includes the seasonally flooded land on the perimeter of
permanently flooded land (i.e., inundation areas). IPCC guidance (IPCC 2019) provides default emission factors for
reservoirs and several types of "other constructed waterbodies" including freshwater ponds and canals/ditches.
Land that has been flooded for 20 years or greater is defined as flooded land remaining flooded land and land
flooded for less than 20 years is defined as land converted to flooded land. The distinction is based on literature
reports that CO2 and CFU emissions are high immediately following flooding as labile organic matter is rapidly
degraded but decline to a steady background level approximately 20 years after flooding (Abril et al. 2005, Barros
et al. 2011, Teodoru et al. 2012). Both CO2 and CFU emissions are estimated for land converted to flooded land.
Nitrous oxide emissions from flooded lands are largely related to inputs of organic or inorganic nitrogen from the
watershed. These inputs from runoff/leaching/deposition are largely driven by anthropogenic activities such as
land-use change, wastewater disposal or fertilizer application in the watershed or application of fertilizer or feed in
aquaculture. These emissions are not included here to avoid double-counting N2O emissions which are captured in
other source categories, such as indirect N2O emissions from managed soils (Section 5.4, Agricultural Soil
Management) and wastewater management (Section 7.2, Wastewater Treatment and Discharge).
Reservoirs are designed to store water for a wide range of purposes including hydropower, flood control, drinking
water, and irrigation. The permanently wetted portion of reservoirs are typically surrounded by periodically
inundated land referred to as a "drawdown zone" or "inundation area." Greenhouse gas emissions from
inundation areas are considered significant and similar per unit area to the emissions from the water surface and
are therefore included in the total reservoir surface area when estimating greenhouse gas emissions from flooded
land. Lakes converted into reservoirs without substantial changes in water surface area or water residence times
are not considered to be managed flooded land (see Area Estimates below) (IPCC 2019).
In 2022, the United States and Puerto Rico contained 72,461 ha of reservoir surface area in land converted to
flooded land (see Methodology and Time-Series Consistency below for calculation details) distributed across all six
of the aggregated climate zones used to define flooded land emission factors (Figure 6-17) (IPCC 2019).
6-150 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
Emissions from Land Converted to Flooded Land-Reservoirs
Figure 6-17: U.S. Reservoirs (black polygons) in the Land Converted to Flooded Land Category
in 2022
Climate Zone
I I boreal
| cool temperate
I I tropical dry/montane
I I tropical moist/wet
| warm temperate dry
dl warm temperate moist
Alaska
100 mi
500 mi
Note: Colors represent climate zone used to derive IPCC default emission factors. Reservoirs (indicated by black
polygons) are sparsely distributed across United States, but can be seen in MN, IL, and IN in this image.
Methane and CO2 are produced in reservoirs through the natural breakdown of organic matter. Per unit area
emission rates tend to scale positively with temperature and system productivity (i.e., abundance of algae).
Greenhouse gases produced in reservoirs can be emitted directly from the water surface and inundation areas or
as greenhouse gas-enriched water passes through the dam and the downstream river. Sufficient information exists
to estimate downstream Cm emissions using Tier 1 IPCC guidance (IPCC 2019), but no guidance is provided for
downstream CO2 emissions. Table 6-101 and Table 6-102 below summarize nationally aggregated CH4 and CO2
emissions from reservoirs in land converted to flooded land. The decrease in CO2 and Cm emissions through the
time series is attributable to reservoirs matriculating from the land converted to flooded land category into the
flooded land remaining flooded land category. Emissions have been stable since 2005, reflecting the low rate of
new flooded land creation over the past 17 years.
Table 6-101: CH4 Emissions from Land Converted to Flooded Land—Reservoirs (MMT CO2 Eq.)
Source
1990
2005
2018
2019
2020
2021
2022
Reservoirs
Surface Emissions 2.5 0.4 0.2 0.2 0.2 0.2 0.2
Downstream Emissions 0.2 + + + + + +
Total
2.7
0.4
0.2
0.2
0.2
0.2
0.2
+lndicates values less than 0.05 MMT C02
Note: Totals may not sum due to independent rounding.
Table 6-102: CH4 Emissions from Land Converted to Flooded Land—Reservoirs (kt CH4)
Source
Total
1990
2005
2018
2019
2020
98
15
8
Note: Totals may not sum due to independent rounding.
2021
2022
Reservoirs
Surface Emissions 90 14 7 7 7 7 7
Downstream Emissions 8 111111
Land Use, Land-Use Change, and Forestry 6-151
-------
Table 6-103: CO2 Emissions from Land Converted to Flooded Land—Reservoirs (MMT CO2)
Source
1990
2005
2018
2019
2020
2021
2022
Reservoir
3.51
0.6
0.3
0.3
0.3
0.3
0.3
Table 6-104: CO2 Emissions from Land Converted to Flooded Land—Reservoirs (MMT C)
Source
1990
2005
2018
2019
2020
2021
2022
Reservoir
0.91
0.2 |
0.1
0.1
0.1
0.1
0.1
Methane and CO2 emissions from reservoirs in Minnesota were 8-fold greater than from any other state (Figure
6-18 and Table 6-105). This is attributed to nineteen reservoirs created in Minnesota after 2001 which impound
54,064 ha of water, 96 percent of which is located in Mille Lacs Lake.
North Dakota is the second largest source of CO2 and Cm from reservoirs in land converted to flooded land. Over
ninety-nine percent of land converted to flooded land reservoir surface area in North Dakota is attributed to Devils
Lake. Both Mille Lacs and Devils Lakes are natural waterbodies provisioned with dams for water level
management.
Figure 6-18: 2022 A) CH4 and B) CO2 Emissions from U.S. Reservoirs in Land Converted to
Flooded Land
A. CH4 Emissions from Reservoirs B. C02 Emissions from Reservoirs
Table 6-105: Methane and CO2 Emissions from Reservoirs in Land Converted to Flooded Land
in 2022 (kt CH4; kt CO2)
State
ch4
co2a
Surface
Downstream
Total
Surface
Alabama
0
0
0
0
Alaska
0
0
0
0
Arizona
0
0
0
0
Arkansas
+
+
+
9
California
+
+
+
3
Colorado
+
+
+
1
Connecticut
+
+
+
+
Delaware
0
0
0
0
District of Columbia
0
0
0
0
Florida
+
+
+
13
Georgia
+
+
+
1
Hawaii
0
0
0
0
Idaho
+
+
+
2
6-152 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
Illinois
+
+
+
4
Indiana
+
+
+
+
Iowa
+
+
+
2
Kansas
+
+
+
+
Kentucky
0
0
0
0
Louisiana
+
+
+
+
Maine
+
+
+
+
Maryland
+
+
+
+
Massachusetts
+
+
+
5
Michigan
+
+
+
+
Minnesota
5
+
5
202
Mississippi
+
+
+
+
Missouri
+
+
+
2
Montana
+
+
+
8
Nebraska
+
+
+
1
Nevada
+
+
+
+
New Hampshire
+
+
+
1
New Jersey
0
0
0
0
New Mexico
+
+
+
1
New York
+
+
+
+
North Carolina
+
+
+
1
North Dakota
+
+
1
22
Ohio
+
+
+
1
Oklahoma
0
0
0
0
Oregon
+
+
+
1
Pennsylvania
+
+
+
1
Puerto Rico
0
0
0
0
Rhode Island
0
0
0
0
South Carolina
0
0
0
0
South Dakota
+
+
+
+
Tennessee
+
+
+
1
Texas
+
+
+
3
Utah
+
+
+
1
Vermont
0
0
0
0
Virginia
0
0
0
0
Washington
+
+
+
3
West Virginia
+
+
+
+
Wisconsin
+
+
+
1
Wyoming
+
+
+
1
+ Indicates values greater than zero and less than 0.5 kt.
a C02: Only surface C02 emissions are included in the Inventory.
Methodology and Time-Series Consistency
Estimates of Cm and CO2 emissions for reservoirs in land converted to flooded land follow the Tier 1 methodology
in the IPCC guidance (IPCC 2019). All calculations are performed at the state level and summed to obtain national
estimates. Emissions from the surface of these flooded lands are calculated as the product of flooded land surface
area and a climate-specific emission factor (Table 6-106). Downstream CH4 emissions are calculated as 9 percent of
the surface Cm emission (Tier 1 default). The IPCC guidance (IPCC 2019) does not address downstream CO2
emissions, presumably because there are insufficient data in the literature to estimate this emission pathway.
The IPCC default surface emission factors are derived from model-predicted (G-res model, Prairie et al. 2017)
emission rates for all reservoirs in the Global Reservoir and Dam (GRanD) database (Lehner et al. 2011). Predicted
emission rates were aggregated by the 11 IPCC climate zones (IPCC 2019, Table 7A.2) which were collapsed into six
climate zones using a regression tree approach. All six aggregated climate zones are present in the United States.
Land Use, Land-Use Change, and Forestry 6-153
-------
Table 6-106: IPCC (2019) Default CH4 and CO2 Emission Factors for Surface Emissions from
Reservoirs in Land Converted to Flooded Land
Surface emission factor
Climate
MT CH.i ha 1 y 1
MT CO ¦ ha 1 y 1
Boreal
0.0277
3.45
Cool Temperate
0.0847
3.74
Warm Temperate Dry
0.1956
6.23
Warm Temperate Moist
0.1275
5.35
Tropical Dry/Montane
0.3923
10.82
Tropical Moist/Wet
0.2516
10.16
Note: Downstream CH4 emissions are calculated as 9 percent of surface emissions.
Downstream emissions are not calculated for C02.
Area estimates
U.S. reservoirs were identified from the NHDWaterbody layer in the National Hydrography Dataset Plus V2
(NHD),84 the National Inventory of Dams (NID),85 the National Wetlands Inventory (NWI),86 and the Navigable
Waterways (NW) network,87 and the EPA's Safe Drinking Water Information System (SDWIS).88 The NHD only
covers the conterminous United States, whereas the NID, NW and NWI also include Alaska, Hawaii, and Puerto
Rico. The following paragraphs present the criteria used to identify other constructed waterbodies in the NHD,
NW, and NWI.
Waterbodies in the NHDWaterbody layer that were less than or equal to 20-years old, greater than or equal to 8
ha in surface area, not identified as canal/ditch in NHD, and met any of the following criteria were considered
reservoirs in land converted to flooded land: 1) the waterbody was classified "Reservoir" in the NHDWaterbody
layer, 2) the waterbody name in the NHDWaterbody layer included "Reservoir", 3) the waterbody in the
NHDWaterbody layer was located in close proximity (up to 100 m) to a dam in the NID, 4) the NHDWaterbody GNIS
name was similar to nearby NID feature (between 100 m to 1000 m).
EPA assumes that all features included in the NW are subject to water-level management to maintain minimum
water depths required for navigation and are therefore managed flooded lands. NW features greater than 8 ha in
surface area are defined as reservoirs.
NWI features were considered "managed" if they had a special modifier value indicating the presence of
management activities (Figure 6-19). To be included in the flooded lands inventory, the managed flooded land had
to be wet or saturated for at least one season per year (see 'Water Regime' in Figure 6-19). NWI features that met
these criteria, were greater than 8 ha in surface area, and were not a canal/ditch (see emissions from land
converted to flooded land-other constructed waterbodies) were defined as reservoirs.
Any NWI or NHD feature that intersected a drinking water intake point from SDWIS was assumed to be
"managed." The rational being that a waterbody used as a source for public drinking water is typically managed in
some capacity - by flow and/or volume control.
Surface areas for identified flooded lands were taken from NHD, NWI or the NW. If features from the NHD, NWI, or
the NW datasets overlapped, duplicate areas were erased. The first step was to take the final NWI flooded lands
84 See https://www.uses.gov/core-science-svstems/nep/national-hvdroeraphv.
85 See https://nid.sec.usace.armv.mil.
86 See https://www.fws.eov/proeram/national-wetlands-inventory/data-download.
87 See https://www.census.eov/eeoeraphies/mappine-files/time-series/eeo/carto-boundary-file.html.
88 See https://www.epa.eov/enviro/sdwis-overview. Not publicly available due to security concerns.
6-154 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
features and use it to identify overlapping NHD features. If the NHD feature had its center in a NWI feature, it was
removed from analysis. Next, remaining NHD features were erased from any remaining overlapping NWI features.
Final selections of NHD and NWI features were used to erase any overlapping NW waterbodies.
Reservoir age was determined by assuming they were created the same year as a nearby (up to 100 m) NID
feature. If no nearby NID feature was identified, it was assumed the feature was greater than 20-years old
throughout the time series. Only reservoirs less than or equal to 20-years old are included in land converted to
flooded land.
Figure 6-19: Selected Features from NWI that meet Flooded Lands Criteria
MODIFIERS
In order to more adequately describe the wetland and deepwater habitats, one each of the water regime, water chemistry, soil, or
special modifiers may be applied at the class or lower level in the hierarchy
Water Regime
Special Modifiers
Water Chemistry
Soil
Nontidal
A Temporanly Flooded
B Seasonally Saturated
Saltwater Tidal
L Subtidal
Freshwater Tidal
Q Regularly Flooded-Fresh Tidal
R Seasonally Flooded-Fresh Tidal
b Beaver
Halinity/Salinity pH Modifiers for
Fresh Water
1 Hyperhaline / Hypersaline a Acid
2 Euhaline / Eusaline t Circumneutral
3 Mixohaline / M ixosaline (Brackish) i Alkaline
4 Polyhaline
5 Mesohaline
6 Oligohaline
0 Fresh
g Organic
n Mineral
M Irreaularlv Exposed
d Partly Drained/Ditched
f Farmed
m Managed
h Diked/Impounded
r Artificial Substrate
s Spoil
x Excavated |
C Seasonally Flooded
N Regularly Flooded
P Irregularly Flooded
S Temporarily Flooded- Fresh Tidal
D Continuously Saturated
T Semipermanently Flooded-Fresh Tidal
V Permanently Flooded-Fresh Tidal
E Seasonally Flooded /
Saturated
F Semipermanently Flooded
G Intermittently Exposed
H Permanently Flooded
J Intermittently Flooded
K Artificially Flooded |
I 1 Must also meet one selected special modifier (red box) to be included in the flooded lands inventory
I ~1 Included in the flooded lands inventory if it meets water regime qualifier (gold box)
Source (modified): https://www.fws.gov/sites/default/files/documents/wetlands-and-deepwater-map-code-diagram.pdf
IPCC (2019) allows for the exclusion of managed waterbodies from the inventory if the water surface area or
residence time was not substantially changed by the construction of the dam. The guidance does not quantify
what constitutes a "substantial" change, but here EPA excludes the U.S. Great Lakes from the inventory based on
expert judgment that neither the surface area nor water residence time was substantially altered by their
associated dams.
Reservoirs were disaggregated by state (using boundaries from the 2016 U.S. Census Bureau89) and climate zone.
Downstream and surface emissions for cross-state reservoirs were allocated to states based on the surface area
that the reservoir occupied in each state. Only the U.S. portion of reservoirs that cross country borders were
included in the Inventory.
The surface area of reservoirs in land converted to flooded land decreased by nearly 90 percent from 1990 to 2022
(Table 6-107). This is due to reservoirs that were less than 20-years old at the beginning of time series entering the
flooded land remaining flooded land category when they exceeded 20 years of age. The rate at which flooded land
has aged out of the land converted to flooded land category has outpaced the rate of new dam construction. New
dam construction has slowed considerably during the time series with oniy nine new dams constructed in 2022,90
versus 552 in 1990 (Figure 6-20).
Table 6-107: National Totals of Reservoir Surface Area in Land Converted to Flooded Land
(thousands of ha)
Surface Area (thousands of ha)
1990
2005
2018
2019
2020
2021
2022
Reservoir
566
115
78
77
75
74
73
89 See https://www.census.gov/geographies/mapping-fiies/tinne-series/geo/carto-boundarv-file.htmi.
90 See https://nid.sec.usace.armv.mil.
Land Use, Land-Use Change, and Forestry 6-155
-------
Figure 6-20: Number of Dams Built per Year from 1990 through 2022
CD O O O
Table 6-108: State Breakdown of Reservoir Surface Area in Land Converted to Flooded Land
(thousands of ha)
State
1990
2005
2018
2019
2020
2021
2022
Alabama
7.5
0.0
0.0
0.0
0.0
0.0
0.0
Alaska
0.6
0.0
0.0
0.0
0.0
0.0
0.0
Arizona
0.0
0.1
0.0
0.0
0.0
0.0
0.0
Arkansas
10.1
2.9
1.8
1.8
1.8
1.8
1.8
California
19.6
2.1
0.5
0.5
0.5
0.5
0.5
Colorado
5.9
1.3
0.3
0.3
0.3
0.3
0.3
Connecticut
2.3
2.2
0.0
0.0
0.0
0.0
0.0
Delaware
0.0
0.0
0.0
0.0
0.0
0.0
0.0
District of Columbia
0.0
0.0
0.0
0.0
0.0
0.0
0.0
Florida
25.7
3.8
1.5
1.5
1.4
1.2
1.2
Georgia
20.6
4.9
0.1
0.1
0.1
0.1
0.1
Hawaii
0.0
0.0
0.0
0.0
0.0
0.0
0.0
Idaho
17.7
1.2
0.5
0.5
0.5
0.5
0.5
Illinois
49.2
39.4
1.4
1.3
1.3
0.8
0.8
Indiana
10.6
0.3
0.1
0.1
0.1
0.1
0.1
Iowa
12.3
3.1
1.0
1.0
0.7
0.7
0.5
Kansas
19.6
0.4
0.2
0.2
0.1
0.1
0.0
Kentucky
1.3
0.1
0.0
0.0
0.0
0.0
0.0
Louisiana
9.4
0.2
0.0
0.0
0.0
0.0
0.0
Maine
10.9
4.4
0.0
0.0
0.0
0.0
0.0
Maryland
0.8
0.1
0.1
0.1
0.1
0.1
0.1
Massachusetts
1.6
0.5
1.5
1.5
1.4
1.4
1.2
Michigan
11.6
0.9
0.1
0.1
0.1
0.1
0.1
Minnesota
9.9
6.4
54.6
54.6
54.3
54.2
54.1
Mississippi
6.2
0.6
0.1
0.1
0.1
0.1
0.1
Missouri
16.4
0.5
0.4
0.4
0.4
0.4
0.4
Montana
14.4
3.9
2.1
2.1
2.1
2.1
2.1
Nebraska
5.8
1.7
0.6
0.6
0.3
0.3
0.3
Nevada
1.6
1.1
0.1
0.1
0.1
0.1
0.1
New Hampshire
0.4
0.2
0.1
0.1
0.1
0.1
0.1
New Jersey
0.7
0.6
0.0
0.0
0.0
0.0
0.0
New Mexico
1.3
0.1
0.1
0.1
0.1
0.1
0.1
New York
4.2
2.6
0.1
0.1
0.1
0.1
0.1
6-156 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
North Carolina
19.7
0-71
0.3
0.2
0.2
0.2
0.2
North Dakota
7.5
3.5
6.3
6.3
6.3
6.2
5.9
Ohio
7.2
13 i
0.3
0.3
0.2
0.2
0.1
Oklahoma
28.7
0.2
0.0
0.0
0.0
0.0
0.0
Oregon
6-2i
0.4 I
0.2
0.2
0.2
0.1
0.1
Pennsylvania
12.6
1.3 "
0.1
0.1
0.1
0.1
0.1
Puerto Rico
o.oi
O
O
¦¦¦I!
0.0
0.0
0.0
0.0
0.0
Rhode Island
0.1;
0.0
0.0
0.0
0.0
0.0
0.0
South Carolina
18.5I
5-8|
0.0
0.0
0.0
0.0
0.0
South Dakota
0.5
3.9
0.8
0.8
0.0
0.0
0.0
Tennessee
11111
1^
00
LO
0-01
0.1
0.1
0.1
0.1
0.1
Texas
74.8
0.9
0.3
0.3
0.2
0.2
0.2
Utah
1.9!
0.11
0.3
0.3
0.3
0.3
0.3
Vermont
0.2
0.1
0.0
0.0
0.0
0.0
0.0
Virginia
un
00
¦¦mil
0.4
0.0
0.0
0.0
0.0
0.0
Washington
5.3
1.6
0.9
0.9
0.5
0.5
0.5
West Virginia
3-1 =
1-6II
0.1
0.0
0.0
0.0
0.0
Wisconsin
1.9
0.4
0.2
0.2
0.2
0.2
0.2
Wyoming
15. ill
6.55
0.4
0.2
0.2
0.2
0.2
Total
565.8
114.6
77.6
77.1
74.7
73.7
72.5
Uncertainty
Uncertainty in estimates of Cm and CO2 emissions from reservoirs on land converted to flooded land were
developed using IPCC Approach 2 and include uncertainty in the default emission factors and the flooded land area
inventory (Table 1-105). Uncertainty in emission factors is provided in the 2019 Refinement to the 2006 IPCC
Guidelines (IPCC 2019). Uncertainties in the spatial data include 1) uncertainty in area estimates from the NHD,
NWI, and NW, and 2) uncertainty in the location of dams in the NID and drinking water intakes in SDWIS. Overall
uncertainties in these spatial datasets are unknown, but uncertainty for remote sensing products is assumed to be
± 10 to 15 percent based on IPCC guidance (IPCC 2003). An uncertainty range of ± 15 percent for the flooded land
area estimates is assumed and is based on expert judgment.
Table 6-109: Approach 2 Quantitative Uncertainty Estimates for CH4 and CO2 Emissions from
Reservoirs in Land Converted to Flooded Land
Source
Gas
2022 Emission Estimate
Uncertainty Range Relative to Emission Estimate"
(MMTCOEq.)
(MMT CO . Eq.)
(%)
Lower Bound Upper Bound
Lower Bound
Upper Bound
Reservoir
Surface
ch4
0.19
0.17 0.21
-11.5%
+11.9%
Surface
C02
0.2
0.26 0.33
-11.7%
+12.4%
Downstream
ch4
+
+ 0.08
-54.1%
+397.0%
Total
0.5
0.44 0.59
-12.2%
+18.8%
+ Indicates values less than 0.05 MMT C02 Eq.
a Range of emission estimates predicted by Monte Carlo stochastic simulation for a 95 percent confidence interval.
QA/QC and Verification
The National Hydrography Data (NHD) is managed by the USGS in collaboration many other federal, state, and
local entities. Extensive QA/QC procedures are incorporated into the curation of the NHD. The National Inventory
of Dams (NID) is maintained by the U.S. Army Corps of Engineers (USACE) in collaboration with the Federal
Emergency Management Agency (FEMA) and state regulatory offices. USACE resolves duplicative and conflicting
data from 68 data sources, which helps obtain the more complete, accurate, and updated NID. The Navigable
Waterways (NW) dataset is part of the U.S. Department of Transportation (USDOT)/Bureau of Transportation
Land Use, Land-Use Change, and Forestry 6-157
-------
Statistics (BTS) National Transportation Atlas Database (NTAD). The NW is a comprehensive network database of
the nation's navigable waterways updated on a continuing basis. U.S. Fish and Wildlife Service is the principal
agency in charge of wetland mapping including the National Wetlands Inventory (NWI). Quality and consistency of
the Wetlands Layer is supported by federal wetlands mapping and classification standards, which were developed
under the oversight of the Federal Geographic Data Committee (FGDC) with input by the FGDC Wetlands
Subcommittee. This dataset is part of the FGDC Water-Inland Theme, which is co-chaired by the FWS and the U.S.
Geological Survey. The EPA's Safe Drinking Water Information System (SDWIS) tracks information on drinking
water contamination levels as required by the 1974 Safe Drinking Water Act and its 1986 and 1996 amendments.
General QA/QC procedures were applied to activity data, documentation, and emission calculations consistent
with the U.S. Inventory QA/QC plan, which is in accordance with Vol. 1 Chapter 6 of the 2006IPCC Guidelines (see
Annex 8 for more details). All calculations were executed independently in Excel and R. Ten percent of state and
national totals were randomly selected for comparison between the two approaches to ensure there were no
computational errors.
Recalculations Discussion
The EPA's SDWIS is a new data source used in the current (1990 through 2022) Inventory. The assumption is that
any waterbody used as a public drinking water source is managed in some capacity—by flow and/or volume
control. This data source added 418 reservoirs totaling 736,344 ha.
The National Inventory of Dams (NID) data are updated regularly. The version of NID used for the current Inventory
contains 47 new dams and updated values for "year of dam completion" for 975 dams relative to the previous
(1990 through 2021) Inventory data. Similarly, the National Wetlands Inventory (NWI) is periodically updated. The
NWI version used for the current 1990 through 2022 Inventory has major updates for MS, ND, NM, and MT.
Overall, the recalculations resulted in substantial increases in methane and carbon dioxide emissions in the first
few years of the time series (e.g., increase of 3.8 MMT CO2 Eq. in 1990), but the differences were minor by 2008
through 2021 (<0.1 MMT C02 Eq.).
Planned Improvements
The EPA recently completed a survey of greenhouse gas emissions from 108 reservoirs in the conterminous United
States.91 The data will be used to develop country-specific emission factors for U.S. reservoirs to be used in the
1990 through 2024 Inventory submission.
Emissions from Land Converted to Flooded Land-Other
Constructed Waterbodies
Freshwater ponds are the only type of flooded lands within the "other constructed waterbodies" subcategory of
land converted to flooded land that are included in this Inventory (see Methodology for details) because age data
are not available for canals and ditches. All canals and ditches are assumed to be greater than 20-years old
throughout the time series and are included in flooded land remaining flooded land.
IPCC (2019) describes ponds as waterbodies that are "...constructed by excavation and/or construction of walls to
hold water in the landscape for a range of uses, including agricultural water storage, access to water for livestock,
recreation, and aquaculture." The IPCC "Decision tree for types of Flooded Land" (IPCC 2019, Fig. 7.2) elaborates
on this description by defining waterbodies less than 8 ha as a subset of "other constructed waterbodies." For this
Inventory, ponds are defined as managed flooded land not identified as "canal/ditch" (see Methods below) with
91 See https://www.epa.gov/air-research/research-emissions-us-reservoirs.
6-158 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
surface area less than 8 ha. IPCC (2019) further distinguishes saline versus brackish ponds, with the former
supporting lower CFU emission rates than the latter. Activity data on pond salinity is not uniformly available for the
United States and all ponds in land converted to flooded land are assumed to be freshwater. Ponds often receive
high organic matter and nutrient loadings, may have low oxygen levels, and are sites of substantial Cm and CO2
emissions from anaerobic sediments.
Methane and CO2 emissions from freshwater ponds decreased 95 percent from 1990 to 2022 due to flooded land
matriculating from land converted to flooded land to flooded land remaining flooded land. In 2022, states in the
Great Plains region generally had the greatest CO2 and CFU emissions from freshwater ponds in land converted to
flooded land (Table 6-110 through Table 6-114, Figure 6-21). Mississippi had the second greatest emissions of all
states, partly due to the relatively high CO2 emission factor for the tropical moist/wet climate zone (Figure 6-17,
Table 6-115).
Table 6-110: CH4 Emissions from Other Constructed Waterbodies in Land Converted to
Flooded Land (MMT CO2 Eq.)
Source
1990 2005 2018
2019
2020
2021
2022
Freshwater Ponds
0.1 1 +
+
+
+
+
+ Indicates values less than 0.05 MMT C02 Eq.
Table 6-111: CH4 Emissions from Other Constructed Waterbodies in Land Converted to
Flooded Land (kt CH4)
Source
1990
2005 2018
2019
2020
2021
2022
Freshwater Ponds
51
ll +
+
+
+
+
+ Indicates values less than 0.5 kt.
Table 6-112: CO2 Emissions from Other Constructed Waterbodies in Land Converted to
Flooded Land (MMT CO2 Eq.)
Source
1990 2005 2018
2019
2020
2021
2022
Freshwater Ponds
+
+
+
+
+
+ Indicates values less than 0.05 MMT C.
Table 6-113: CO2 Emissions from Other Constructed Waterbodies in Land Converted to
Flooded Land (MMT C)
Source
1990
2005
2018
2019
2020
2021
2022
Freshwater Ponds
0.04
0.01
+
+
+
+
+
+ Indicates values less than 0.005 MMT C.
Table 6-114: CH4 and CO2 Emissions from Other Constructed Waterbodies in Land Converted
to Flooded Land in 2022 (MT CO2 Eq.)
Freshwater Ponds
State
CH.,
CO.
Total
Alabama
0
0
0
Alaska
0
0
0
Arizona
0
0
0
Arkansas
1
1
3
California
126
146
272
Colorado
382
290
672
Connecticut
0
0
0
Delaware
0
0
1
District of Columbia
0
0
0
Florida
18
37
55
Georgia
164
293
457
Land Use, Land-Use Change, and Forestry 6-159
-------
Hawaii
0
0
0
Idaho
0
0
0
Illinois
83
74
157
Indiana
66
69
135
Iowa
282
254
535
Kansas
435
456
891
Kentucky
3
3
5
Louisiana
3
6
10
Maine
1
1
2
Maryland
58
60
118
Massachusetts
381
358
738
Michigan
41
30
71
Minnesota
317
232
549
Mississippi
400
612
1,012
Missouri
133
139
271
Montana
509
371
880
Nebraska
620
567
1,186
Nevada
103
80
183
New Hampshire
80
59
139
New Jersey
0
0
0
New Mexico
57
46
103
New York
120
96
215
North Carolina
107
112
219
North Dakota
229
167
396
Ohio
289
285
574
Oklahoma
339
378
717
Oregon
89
71
161
Pennsylvania
29
25
54
Puerto Rico
0
0
0
Rhode Island
0
0
0
South Carolina
47
49
95
South Dakota
455
332
788
Tennessee
11
11
22
Texas
83
138
222
Utah
207
151
359
Vermont
15
11
26
Virginia
10
11
21
Washington
140
132
272
West Virginia
19
19
38
Wisconsin
93
68
162
Wyoming
369
269
639
Total
6,917
6,510
13,427
6-160 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
Figure 6-21: 2022 A) CH4 and B) CO2 Emissions from Other Constructed Waterbodies
(Freshwater Ponds) in Land Converted to Flooded Land (MT CO2 Eq.)
A. CH4 Emissions from Freshwater Ponds B. C02 Emissions from Freshwater Ponds
Methodology and Time-Series Consistency
Estimates of CFU and CO2 emissions for other constructed waterbodies in land converted to flooded land follow the
Tier 1 methodology in IPCC (2019). All calculations are performed at the state level and summed to obtain national
estimates. Greenhouse gas emissions from the surface of these flooded lands are calculated as the product of
flooded land surface area and an emission factor (Table 6-115). Due to a lack of empirical data on CO2 emissions
from recently created ponds, IPCC (2019) states "For all types of ponds created by damming, the methodology
described above to estimate CO2 emissions from land converted to reservoirs may be used." This Inventory uses
IPCC default CO2 emission factors for land converted to reservoirs when estimating CO2 emissions from land
converted to freshwater ponds. IPCC guidance also states that "there is insufficient information available to derive
separate CFU emission factors for recently constructed ponds..." and allows for the use of IPCC default CH4
emission factors for land remaining flooded land. Downstream emissions are not inventoried for other constructed
waterbodies because 1) many of these systems are not associated with dams (e.g., excavated ponds and ditches),
and 2) there are insufficient data to derive downstream emission factors for other constructed waterbodies that
are associated with dams (IPCC 2019).
Table 6-115: IPCC Default Methane and CO2 Emission Factors for Other Constructed
Waterbodies in Land Converted to Flooded Land
Emission Factor
Other Constructed Waterbody
Climate Zone
MT CH4 ha 1 y1 MT COz ha 1 y1
Freshwater ponds Boreal 0.183 3.45
Freshwater ponds Cool Temperate 0.183 3.74
Freshwater ponds
Warm Temperate Dry
0.183
6.23
Freshwater ponds
Warm Temperate Moist
0.183
5.35
Freshwater ponds
Tropical Dry/Montane
0.183
10.82
Freshwater ponds
Tropical Moist/Wet
0.183
10.16
Land Use, Land-Use Change, and Forestry 6-161
-------
Area Estimates
Other constructed waterbodies were identified from the NHDWaterbody layer in the National Hydrography
Dataset Plus V2 (NHD),92 the National Inventory of Dams (NID),93 the National Wetlands Inventory (NWI),94 and
the Navigable Waterways (NW) network95, and the EPA's Safe Drinking Water Information System (SDWIS)96. The
NHD only covers the conterminous United States, whereas the NID, NW and NWI also include Alaska, Hawaii, and
Puerto Rico.
Waterbodies in the NHDWaterbody layer that were less than or equal to 20-years old, less than 8 ha in surface
area, not identified as canal/ditch in NHD, and met any of the following criteria were considered freshwater ponds
in land converted to flooded land: 1) the waterbody was classified "Reservoir" in the NHDWaterbody layer, 2) the
waterbody name in the NHDWaterbody layer included "Reservoir", 3) the waterbody in the NHDWaterbody layer
was located in close proximity (up to 100 m) to a dam in the NID, 4) the NHDWaterbody GNIS name was similar to
nearby NID feature (between 100 m to 1000 m).
EPA assumes that all features included in the NW are subject to water-level management to maintain minimum
water depths required for navigation and are therefore managed flooded lands. NW features that were less than 8
ha in surface area and not identified as canals/ditch (see below) were considered freshwater ponds. Only 2.1
percent of NW features met these criteria, and they were primarily associated with larger navigable waterways,
such as lock chambers on impounded rivers.
NWI features were considered "managed" if they had a special modifier value indicating the presence of
management activities (Figure 6-19). To be included in the flooded lands inventory, the managed flooded land had
to be wet or saturated for at least one season per year (see 'Water Regime' in Figure 6-19). NWI features that met
these criteria, were less than 8 ha in surface area, and were not a canal/ditch were defined as freshwater ponds.
Any NWI or NHD feature that intersected a drinking water intake point from SDWIS was assumed to be
"managed". The rational being that a waterbody used as a source for public drinking water is typically managed in
some capacity - by flow and/or volume control.
Surface areas for other constructed waterbodies were taken from NHD, NWI or the NW. If features from the NHD,
NWI, or the NW datasets overlapped, duplicate areas were erased. The first step was to take the final NWI flooded
lands features and use it to identify overlapping NHD features. If the NHD feature had its center in a NWI feature, it
was removed from analysis. Next, remaining NHD features were erased from any remaining overlapping NWI
features. Final selections of NHD and NWI features were used to erase any overlapping NW waterbodies.
The age of other constructed waterbody features was determined by assuming the waterbody was created the
same year as a nearby (up to 100 m) NID feature. If no nearby NID feature was identified, it was assumed the
waterbody was greater than 20-years old throughout the time series. No canal/ditch features were associated with
a nearby dam, therefore all canal/ditch features were assumed to be greater than 20-years old through the time
series.
For the year 2022, this Inventory contains 1,350 ha of freshwater ponds in land converted to flooded land. The
surface area of freshwater ponds decreased by 95 percent from 1990 to 2022 due to flooded lands aging out of
land converted to flooded land more quickly than new flooded lands entered the category. The greatest reduction
in freshwater pond surface area occurred in Iowa, Kansas, and Georgia (Table 6-117). Freshwater ponds in the
2021 inventory are most abundant in Nebraska, Montana, and Kansas (Figure 6-22).
92 See https://www.usgs.gov/core-science-svstems/ngp/national-hvdrography.
93 See https://nid.sec.usace.armv.mil.
94 See https://www.fws.gov/program/national-wetlands-inventory/data-download.
95 See https://hifld-eeoplatform.opendata.arceis.com/datasets/eeoplatform::navieable-waterway-network-lines-l/about.
96 Not publicly available due to security concerns.
6-162 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
Table 6-116: National Surface Area Totals of Other Constructed Waterbodies in Land
Converted to Flooded Land (ha)
Other Constructed Waterbodies
1990
2005
2018
2019
2020
2021
2022
Freshwater Ponds
25,492
5,357
2,604
2,317
1,983
1,673
1,472
Figure 6-22: Surface Area of Other Constructed Waterbodies in Land Converted to Flooded
Land (ha) in 2022
Table 6-117: State Surface Area Totals of Other Constructed Waterbodies in Land Converted
to Flooded Land (ha)
State
1990
2005
2018
2019
2020
2021
2022
Alabama
317
13
0
0
0
0
0
Alaska
3
0
0
0
0
0
0
Arizona
39
16
4
4
0
0
0
Arkansas
331
0
0
0
0
0
0
California
263
103
45
40
33
31
25
Colorado
279
71
79
78
89
89
75
Connecticut
67
2
0
0
0
0
0
Delaware
4
0
0
0
0
0
0
District of Columbia
0
0
0
0
0
0
0
Florida
154
58
15
10
10
4
4
Georgia
1,686
83
35
32
32
32
32
Hawaii
11
4
0
0
0
0
0
Idaho
133
8
1
1
1
0
0
Illinois
557
133
42
37
27
26
16
Indiana
494
133
28
23
23
23
13
Iowa
2,592
1,580
474
290
172
76
55
Kansas
2,099
147
113
104
103
87
85
Kentucky
394
30
1
1
1
1
1
Louisiana
130
17
7
7
1
1
1
Maine
51
10
0
0
0
0
0
Maryland
226
81
20
20
18
15
11
Land Use, Land-Use Change, and Forestry 6-163
-------
Massachusetts
68 =
79 =1;
93
86
80
78
74
Michigan
162
37
16
16
8
8
8
Minnesota
344
142
103
101
79
71
62
Mississippi
414
200
124
117
98
85
78
Missouri
3,451
104 :
38
34
32
29
26
Montana
400
109
105
100
99
99
99
Nebraska
1,427
3 741
182
164
133
125
121
Nevada
21
64
26
26
22
22
20
New Hampshire
154*
61 ¦
17
17
16
16
16
New Jersey
50
21
0
0
0
0
0
New Mexico
14
]_4 5
19
17
17
17
11
New York
312
124
31
29
29
24
23
North Carolina
498
92
28
28
25
22
21
North Dakota
90
135
67
61
51
48
45
Ohio
431|
293 i
121
107
75
60
56
Oklahoma
2,008
147
111
95
81
75
66
Oregon
2201
69
25
22
18
17
17
Pennsylvania
255
33
6
6
6
6
6
Puerto Rico
Q iiiiiil
0 1
0
0
0
0
0
Rhode Island
9
7
0
0
0
0
0
South Carolina
826;
230
22
13
9
9
9
South Dakota
227
98
105
94
93
89
89
Tennessee
389 i
37 ¦
14
9
2
2
2
Texas
2,950
89
21
16
17
16
16
Utah
681
I9 >
42
42
40
40
40
Vermont
70
11
3
3
3
3
3
Virginia
00
LO
A iiiiiii
4 mm1
2
2
2
2
2
Washington
153
57
31
31
28
27
27
West Virginia
i3o!
101
4
4
4
4
4
Wisconsin
146
21
18
18
18
18
18
Wyoming
316 =
190
79
79
78
75
72
Total
25,492
5 357
3,33'
2,317
1,983
1,673
1,472
1,350
Uncertainty
Uncertainty in estimates of CO2 and Cmemissions from land converted to flooded land-other constructed water
bodies include uncertainty in the default emission factors and the flooded land area inventory. Uncertainty in
emission factors is provided in the 2019 Refinement to the 2006IPCC Guidelines (IPCC 2019). Uncertainties in the
spatial data include 1) uncertainty in area estimates from the NHD and NW, and 2) uncertainty in the location of
dams in the NID and drinking water intakes in SDWIS. Overall uncertainties in the NHD, NWI, NID, and NW are
unknown, but uncertainty for remote sensing products is ±10 to 15 percent (IPCC 2003). EPA assumes an
uncertainty of ± 15 percent for the flooded land area inventory based on expert judgment. These uncertainties do
not include the underestimate of pond surface area discussed above.
Table 6-118: Approach 2 Quantitative Uncertainty Estimates for CH4 and CO2 Emissions from
Other Constructed Waterbodies in Land Converted to Flooded Land
Source
Gas
2022 Emission Estimate
(kt COEq.)
Uncertainty Range Relative to Emission Estimate''
(kt CO' Eq.) (%)
Lower Bound Upper Bound
Lower Bound
Upper Bound
Freshwater ponds
ch4
6.90
6.80 7.10
-2.3%
+2.7%
Freshwater ponds
C02
6.51
6.38 6.62
-2.0%
+1.8%
Total
13.42
13.18 13.70
-1.8%
+2.1%
a Range of emission estimates predicted by Monte Carlo stochastic simulation for a 95 percent confidence interval.
Note: Totals may not sum due to independent rounding.
6-164 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
QA/QC and Verification
The National Hydrography Data (NHD) is managed by the USGS with collaboration from many other federal, state,
and local entities. Extensive QA/QC procedures are incorporated into the curation of the NHD. The National
Inventory of Dams (NID) is maintained by the U.S. Army Corps of Engineers (USACE) in collaboration with the
Federal Emergency Management Agency (FEMA) and state regulatory offices. USACE resolves duplicative and
conflicting data from 68 data sources, which helps obtain the more complete, accurate, and updated NID. The
Navigable Waterways (NW) dataset is part of the U.S. Department of Transportation (USDOT)/Bureau of
Transportation Statistics (BTS) National Transportation Atlas Database (NTAD). The NW is a comprehensive
network database of the nation's navigable waterways updated on a continuing basis. U.S. Fish and Wildlife Service
is the principal agency in charge of wetland mapping including the National Wetlands Inventory. Quality and
consistency of the Wetlands Layer is supported by federal wetlands mapping and classification standards, which
were developed under the oversight of the Federal Geographic Data Committee (FGDC) with input by the FGDC
Wetlands Subcommittee. This dataset is part of the FGDC Water-Inland Theme, which is co-chaired by the FWS and
the U.S. Geological Survey. The EPA's Safe Drinking Water Information System (SDWIS) tracks information on
drinking water contamination levels as required by the 1974 Safe Drinking Water Act and its 1986 and 1996
amendments.
General QA/QC procedures were applied to activity data, documentation, and emission calculations consistent
with the U.S. Inventory QA/QC plan, which is in accordance with Vol. 1 Chapter 6 of the 2006IPCC Guidelines (see
Annex 8 for more details). All calculations were executed independently in Excel and R. Ten percent of state and
national totals were randomly selected for comparison between the two approaches to ensure there were no
computational errors.
Recalculations Discussion
The EPA's SDWIS is a new data source used in the current (1990 through 2022) Inventory. The assumption is that
any waterbody used as a public drinking water source is managed in some capacity - by flow and/or volume
control. This data source added 54 features totaling 173 ha of other constructed waterbodies.
The National Inventory of Dams (NID) data are updated regularly. The version of NID used for the current Inventory
contains 47 new dams and updated values for "year of dam completion" for 975 dams relative to the previous
(1990 through 2021) Inventory data. Similarly, the National Wetlands Inventory (NWI) is periodically updated. The
NWI version used for the current Inventory has major updates for MS, ND, NM, and MT.
The net effect of these recalculations was an average annual increase in CH4 and CO2 emissions from other
constructed waterbodies of 0.03 MMT CO2 Eq., or 51 percent, over the time series from 1990 to 2021 compared to
the previous Inventory.
Planned Improvements
Features < 8 ha in the NW that were not identified as canal/ditch were defined as freshwater ponds. Many of these
features are lock chambers connected to an upstream reservoir. These systems likely have emission rates more
similar to a reservoir than freshwater pond. In the next (i.e., 1990 through 2023) Inventory these systems will be
classified as reservoirs.
Land Use, Land-Use Change, and Forestry 6-165
-------
6.10 Settlements Remaining Settlements
(CRT Category 4E1)
Soil Carbon Stock Changes (CRT Category 4E1)
Soil organic C stock changes for settlements remaining settlements occur in both mineral and organic soils.
However, the United States does not estimate changes in soil organic C stocks for mineral soils in settlements
remaining settlements. This approach is consistent with the assumption of the Tier 1 method in the 2006 IPCC
Guidelines (IPCC 2006) that inputs equal outputs, and therefore the soil organic C stocks do not change in this land
use category. This assumption may be re-evaluated in the future if funding and resources are available to conduct
an analysis of soil organic C stock changes for mineral soils in settlements remaining settlements.
Drainage of organic soils is common when wetland areas have been developed for settlements. Organic soils, also
referred to as Histosols, include all soils with more than 12 to 20 percent organic carbon by weight, depending on
clay content (NRCS 1999; Brady and Weil 1999). The organic layer of these soils can be very deep (i.e., several
meters), and form under inundated conditions that results in minimal decomposition of plant residues. Drainage of
organic soils leads to aeration of the soil that accelerates decomposition rate and CO2 emissions.97 Due to the
depth and richness of the organic layers, carbon loss from drained organic soils can continue over long periods of
time, which varies depending on climate and composition (i.e., decomposability) of the organic matter (Armentano
and Menges 1986).
Settlements remaining settlements includes all areas that have been settlements for a continuous time period of
at least 20 years according to the 2017 United States Department of Agriculture (USDA) National Resources
Inventory (NRI) (USDA-NRCS 2020)98 or according to the National Land Cover Dataset (NLCD) for federal lands
(Yang et al. 2018; Fry et al. 2011; Homer et al. 2007, 2015). There are discrepancies between the current land
representation (see Section 6.1) and the area data that have been used in the Inventory for settlements remaining
settlements. Specifically, Alaska and the small amount of settlements on federal lands are not included in this
Inventory even though these areas are part of the U.S. managed land base. There is a planned improvement to
include CO2 emissions from drainage of organic soils in settlements of Alaska and federal lands as part of a future
Inventory (see Planned Improvements section).
CO2 emissions from drained organic soils in settlements are 15.4 MMT CO2 Eq. (4.2 MMT C) in 2022 (see Table
6-119 and Table 6-120). Although the flux is relatively small, the amount has increased by 56 percent since 1990
due to an increase in area of drained organic soils in settlements.
Table 6-119: Net CO2 Flux from Soil C Stock Changes in Settlements Remaining Settlements
(MMT C02 Eq.)
Soil Type
1990
2005 2018 2019
2020
2021
2022
Organic Soils
9.91
10.l| 14.4 14.6
15.1
15.4
15.4
97 N20 emissions from drained organic soils are included in the N20 Emissions from Settlement Soils section.
98 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 1998. This may have led to an overestimation of Settlements Remaining Settlements
in the early part of the time series to the extent that some areas are converted to settlements between 1971 and 1978.
6-166 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
Table 6-120: Net CO2 Flux from Soil C Stock Changes in Settlements Remaining Settlements
(MMT C)
Soil Type
1990
2005 2018
2019
2020
2021
2022
Organic Soils
2.7
2.7l 3.9
4.0
4.1
4.2
4.2
Methodology and Time-Series Consistency
An IPCC Tier 2 method is used to estimate soil organic C stock changes for organic soils in settlements remaining
settlements (IPCC 2006). Organic soils in settlements remaining settlements are assumed to be losing C at a rate
similar to croplands due to deep drainage, and therefore emission rates are based on country-specific values for
cropland (Ogle et al. 2003).
The land area designated as settlements is based primarily on the 2017 NRI (USDA-NRCS 2020) with additional
information from the NLCD to the extend the time series through 2020 (Yang et al. 2018). Soils are classified as
organic using data from the Soil Survey Geographic Database (SSURGO) (Soil Survey Staff 2020). The areas have
been modified through a process in which the Forest Inventory and Analysis (FIA) survey data are harmonized with
the NRI data (Nelson et al. 2020). This process ensures that the land use areas are consistent across all land use
categories (see Section 6.1 for more information). All settlements occurring on organic soil are assumed to be
drained for the purposes of approximating greenhouse gas emissions. The area of drained organic soils is
estimated from the NRI spatial weights and aggregated to the country (Table 6-121). The area of land on organic
soils in settlements remaining settlements has increased from 216 thousand hectares in 1990 to over 327
thousand hectares in 2020.
Table 6-121: Thousands of Hectares of Drained Organic Soils in Settlements Remaining
Settlements
1990
2005
2014
2015
2016
2017
2018
2019
2020 2021 2022
Area (Thousand Hectares)
216
219|
276
283
291
302
311
317
327 * *
NRI data have not been incorporated into the Inventory after 2020, designated with asterisks (*).
To estimate CO2 emissions from drained organic soils across the time series from 1990 to 2020, the area of organic
soils by climate (i.e., cool temperate, warm temperate, subtropical) in settlements remaining settlements is
multiplied by the appropriate country-specific emission factors for cropland remaining cropland under the
assumption that there is deep drainage of the soils. The emission factors are 11.2 MT C per ha in cool temperate
regions, 14.0 MT C per ha in warm temperate regions, and 14.3 MT C per ha in subtropical regions (see Annex 3.12
for more information).
In order to ensure time-series consistency, the same methods are applied from 1990 to 2020, and a linear
extrapolation method is used to approximate emissions for the remainder of the 2021 to 2020 time series (see Box
6-4 in cropland remaining cropland). The extrapolation is based on a linear regression model with moving-average
(ARMA) errors using the 1990 to 2020 emissions data, and is a standard data splicing method for imputing missing
emissions data in a time series (IPCC 2006). The Tier 2 method described previously will be applied in future
inventories to recalculate the estimates beyond 2020 as new activity data are integrated into the analysis.
Uncertainty
The total uncertainty was quantified with two variance components (Ogle et al. 2010) that are combined using
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 first variance component is associated with
uncertainty in the emission factor, and the second variance component is associated with scaling of the data from
the NRI survey to the entire area of drained organic soils in settlements remaining settlements, and is computed
using a standard variance estimator for a two-stage sample design (Sarndal et al. 1992). There is also additional
uncertainty associated with the fit of the linear regression model for the data splicing methods that was
Land Use, Land-Use Change, and Forestry 6-167
-------
incorporated into the analysis for the latter part of the time series. Soil carbon losses from drained organic soils in
settlements remaining settlements for 2022 are estimated to be between 7.7 and 23.2 MMT CO2 Eq. at a 95
percent confidence level (Table 6-115). This indicates a range of 50 percent below and 50 percent above the 2022
emission estimate of 15.4 MMT CO2 Eq.
Table 6-122: Uncertainty Estimates for CO2 Emissions from Drained Organic Soils in
Settlements Remaining Settlements (MMT CO2 Eq. and Percent)
Uncertainty Range Relative to Emission Estimate"
Source
Gas
2022 Emission Estimate
(MMTCO' Eq.) (%)
(MMT CO . Eq.)
Lower Bound Upper Bound Lower Bound Upper Bound
Organic Soils
C02
15.4
7.7 23.2 -50% +50%
a Range of emission estimates is a 95 percent confidence interval.
QA/QC and Verification
Quality control measures included checking input data, model scripts, and results to ensure data are properly
handled throughout the inventory process consistent with the U.S. Inventory QA/QC plan, which is in accordance
with Volume 1, Chapter 6 of the 2006IPCC Guidelines (see Annex 8 for more details). Inventory reporting forms
and text are reviewed and revised as needed to correct transcription errors. No errors were found in this
Inventory.
Recalculations Discussion
Recalculations are associated with updated land areas for drainage of organic soils in settlements remaining
settlements, and update that was made by incorporating new USDA-NRCS NRI data through 2017 and extending
the time series using NLCD. As a result of this change, C02-equivalent emissions changed annually with an average
annual decrease of 1.8 MMT CO2 Eq., or 14 percent, over the time series from 1990 to 2021 compared to the
previous Inventory.
Planned Improvements
A key improvement is to estimate CO2 emissions from drainage of organic soils in settlements of Alaska and federal
lands. This improvement will resolve most of the differences between the managed land base for settlements
remaining settlements and amount of area currently included in this Inventory as settlements remaining
settlements (see Table 6-123). This improvement will be made pending prioritization of resources to expand the
inventory for this source category.
6-168 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
Table 6-123: Area of Managed Land in Settlements Remaining Settlements that is not
included in the current Inventory (Thousand Hectares)
Year
Area (Thousand Hectares)
SRS Managed Land
Area (Section 6.1)
SRS Area Included
in Inventory
Difference
1990
30,548
30,366
182
1991
30,545
30,364
182
1992
30,543
30,361
182
1993
30,470
30,288
182
1994
30,385
30,203
182
1995
30,322
30,141
182
1996
30,263
30,081
182
1997
30,193
30,011
182
1998
30,127
29,945
182
1999
30,073
29,891
182
2000
30,015
29,834
182
2001
29,963
29,781
182
2002
29,956
29,774
182
2003
30,479
30,298
182
2004
30,973
30,791
182
2005
31,432
31,250
182
2006
31,940
31,758
182
2007
32,397
32,215
182
2008
33,015
32,833
182
2009
33,591
33,410
182
2010
34,166
33,984
182
2011
34,731
34,549
182
2012
35,302
35,120
182
2013
36,224
36,042
182
2014
37,159
36,977
182
2015
38,026
37,844
182
2016
38,938
38,756
182
2017
39,861
39,679
182
2018
40,756
40,574
182
2019
41,602
41,420
182
2020
42,452
42,270
182
2021
43,175
*
*
2022
43,734
*
*
NRI data have not been incorporated into the Inventory after 2020, designated with
asterisks (*).
Changes in Carbon Stocks in Settlement Trees (CRT Category
4E1)
Settlements are land uses where human populations and activities are concentrated. In these areas, the
anthropogenic impacts on tree growth, stocking and mortality are particularly pronounced (Nowak 2012) in
comparison to forest lands where non-anthropogenic forces can have more significant impacts. Estimates included
in this section include net CO2 and carbon flux from trees on settlements remaining settlements and land
converted to settlements as it is not possible to report on these separately at this time.
Trees in settlement areas of the United States are estimated to account for an average annual net sequestration of
118.2 MMT CO2 Eq. (32.2 MMT C) over the period from 1990 through 2022. Net carbon sequestration from
settlement trees in 2022 is estimated to be 138.5 MMT CO2 Eq. (37.8 MMT C) (Table 6-124). Dominant factors
affecting carbon flux trends for settlement trees are changes in the amount of settlement area (increasing
Land Use, Land-Use Change, and Forestry 6-169
-------
sequestration due to more land and trees) and net changes in tree cover (e.g., tree losses versus tree gains through
planting and natural regeneration), with percent tree cover trending downward recently. In addition, changes in
species composition, tree sizes and tree densities affect base carbon flux estimates. Annual sequestration
increased by 43 percent between 1990 and 2022 due to increases in settlement area and changes in total tree
cover.
Trees in settlements often grow faster than forest trees because of their relatively open structure (Nowak and
Crane 2002). Because tree density in settlements is typically much lower than in forested areas, the carbon storage
per hectare of land is in fact smaller for settlement areas than for forest areas. Also, percent tree cover in
settlement areas is less than in forests and this urban tree cover varies significantly across the United States (e.g.,
Nowak and Greenfield 2018a). To quantify the carbon stored in settlement trees, the methodology used here
requires analysis per unit area of tree cover, rather than per unit of total land area (as is done for forest lands).
Table 6-124: Net Flux from Trees in Settlements Remaining Settlements (MMT CO2 Eq. and
MMT C)a
Year
1990
2005
2018
2019
2020
2021
2022
MMTCO2 Eq.
<96'6)|
(117.0)
(134.4)
(135.6)
(136.7)
(137.8)
(138.5)
MMT C
(26.3)1
(31.9)
(36.7)
(37.0)
(37.3)
(37.6)
(37.8)
a These estimates include net C02 and C flux from trees on settlements remaining settlements and
land converted to settlements as it is not possible to report on these separately at this time.
Note: Parentheses indicate net sequestration.
Methodology and Time-Series Consistency
To estimate net carbon sequestration in settlement areas, three types of data are required for each state:
1. Settlement area
2. Percent tree cover in settlement areas
3. Carbon sequestration density per unit of tree cover
Settlement Area
Settlement area is defined in Section 6.1 as a land-use category representing developed areas. The data used to
estimate settlement area within Section 6.1 comes from the latest NRI as updated through 2017, with the
extension of the time series through 2022 based on assuming the settlement area is the same as 2017. The NRI
data is also harmonized with the FIA dataset, which is available through 2022, and the 2019 NLCD dataset. This
process of combining the datasets extends the time series to ensure that there is a complete and consistent
representation of land use data for all source categories in the LULUCF sector. Annual estimates of the net CO2 flux
(Table 6-124) were developed based on estimates of annual settlement area and tree cover derived from NLCD
developed lands. Developed land, which was used to estimate tree cover in settlement areas, is about six percent
higher than the area categorized as settlements in the representation of the U.S. land base developed for this
report.
Percent Tree Cover in Settlement Areas
Percent tree cover in settlement area by state is needed to convert settlement land area to settlement tree cover
area. Converting to tree cover area is essential as tree cover, and thus carbon estimates, can vary widely among
states in settlement areas due to variations in the amount of tree cover (e.g., Nowak and Greenfield 2018a).
However, since the specific geography of settlement area is unknown because they are based on NRI sampling
methods, NLCD developed land was used to estimate the percent tree cover to be used in settlement areas. The
NLCD developed land cover classes 21-24 (developed, open space (21), low intensity (22), medium intensity (23),
and high intensity (24)) were used to estimate percent tree cover in settlement area by state (U.S. Department of
Interior 2018; MRLC 2013).
6-170 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
a) "Developed, Open Space - areas with a mixture of some constructed materials, but mostly vegetation in
the form of lawn grasses. Impervious surfaces account for less than 20 percent of total cover. These areas
most commonly include large-lot single-family housing units, parks, golf courses, and vegetation planted
in developed settings for recreation, erosion control, or aesthetic purposes." Plots designated as either
park, recreation, cemetery, open space, institutional or vacant land were classified as "Developed, Open
Space".
b) "Developed, Low Intensity - areas with a mixture of constructed materials and vegetation. Impervious
surfaces account for 20 to 49 percent of total cover. These areas most commonly include single-family
housing units." Plots designated as single family or low-density residential land were classified as
"Developed, Low Intensity".
c) "Developed, Medium Intensity - areas with a mixture of constructed materials and vegetation.
Impervious surfaces account for 50 to 79 percent of the total cover. These areas most commonly include
single-family housing units." Plots designated as medium density residential, other urban or mixed urban
were classified as "Developed, Medium Intensity".
d) "Developed High Intensity - highly developed areas where people reside or work in high numbers.
Examples include apartment complexes, row houses and commercial/industrial. Impervious surfaces
account for 80 to 100 percent of the total cover." Plots designated as either commercial, industrial, high
density residential, downtown, multi-family residential, shopping, transportation or utility were classified
as "Developed, High Intensity".
As NLCD is known to underestimate tree cover (Nowak and Greenfield 2010), photo-interpretation of tree cover
within NLCD developed lands was conducted for the years of c. 2016 and 2020 using 1,000 random points to
determine an average adjustment factor for NLCD tree cover estimates in developed land and determine recent
tree cover changes. This photo-interpretation of change followed methods detailed in Nowak and Greenfield
(2018b). Percent tree cover (%TC) in settlement areas by state was estimated as:
%TC in state = state NLCD %TC x national photo-interpreted %TC / national NLCD %TC
Percent tree cover in settlement areas by year was set as follows:
• 1990 to 2011: used 2011 NLCD tree cover adjusted with 2011 photo-interpreted values
• 2012 to 2015: used 2011 NLCD tree cover adjusted with photo-interpreted values, which were
interpolated from values between 2011 and 2016
• 2016 to 2020: used 2016 NLCD tree cover adjusted with 2020 photo-interpreted values
Carbon Sequestration Density per Unit of Tree Cover
Methods for quantifying settlement tree biomass, carbon sequestration, and carbon emissions from tree mortality
and decomposition were taken directly from Nowak et al. (2013), Nowak and Crane (2002), and Nowak (1994). In
general, net carbon sequestration estimates followed three steps, each of which is explained further in the
paragraphs below. First, field data from cities and urban areas within entire states were used to estimate carbon in
tree biomass from field data on measured tree dimensions. Second, estimates of annual tree growth and biomass
increment were generated from published literature and adjusted for tree condition, crown competition, and
growing season to generate estimates of gross carbon sequestration in settlement trees for all 50 states and the
District of Columbia. Third, estimates of carbon emissions due to mortality and decomposition were subtracted
from gross carbon sequestration estimates to obtain estimates of net carbon sequestration. Carbon storage, gross
and net sequestration estimates were standardized per unit tree cover based on tree cover in the study area.
Land Use, Land-Use Change, and Forestry 6-171
-------
Settlement tree carbon estimates are based on published literature (Nowak et al. 2013; Nowak and Crane 2002;
Nowak 1994) as well as newer data from the i-Tree database" and U.S. Forest Service urban forest inventory data
(e.g., Nowak et al. 2016, 2017) (Table 6-125). These data are based on collected field measurements in several U.S.
cities between 1989 and 2017. Carbon storage and sequestration in these cities were estimated using the U.S.
Forest Service's i-Tree Eco model (Nowak et al. 2008). This computer model uses standardized field data from
randomly located plots, along with local hourly air pollution and meteorological data, to quantify urban forest
structure, monetary values of the urban forest, and environmental effects, including total carbon stored and
annual carbon sequestration (Nowak et al. 2013).
In each city, a random sample of plots were measured to assess tree stem diameter, tree height, crown height and
crown width, tree location, species, and canopy condition. The data for each tree were used to estimate total dry-
weight biomass using allometric models, a root-to-shoot ratio to convert aboveground biomass estimates to whole
tree biomass, and wood moisture content. Total dry weight biomass was converted to carbon by dividing by two
(50 percent carbon content). An adjustment factor of 0.8 was used for open grown trees to account for settlement
trees having less aboveground biomass for a given stem diameter than predicted by allometric models based on
forest trees (Nowak 1994). Carbon storage estimates for deciduous trees include only carbon stored in wood.
Estimated carbon storage was divided by tree cover in the area to estimate carbon storage per square meter of
tree cover.
Table 6-125: Carbon Storage (kg C/m2 tree cover), Gross and Net Sequestration (kg C/m2 tree
cover/year) and Tree Cover (percent) among Sampled U.S. Cities (see Nowak et al. 2013)
Sequestration
City
Storage
SE
Gross
SE
Net
SE
RatioJ
Tree
Cover
SE
Adrian, Ml
12.17
1.88
0.34
0.04
0.13
0.07
0.36
22.1
2.3
Albuquerque, NM
5.61
0.97
0.24
0.03
0.20
0.03
0.82
13.3
1.5
Arlington, TX
6.37
0.73
0.29
0.03
0.26
0.03
0.91
22.5
0.3
Atlanta, GA
6.63
0.54
0.23
0.02
0.18
0.03
0.76
53.9
1.6
Austin, TX
3.57
0.25
0.17
0.01
0.13
0.01
0.73
30.8
1.1
Baltimore, MD
10.30
1.24
0.33
0.04
0.20
0.04
0.59
28.5
1.0
Boise, ID
7.33
2.16
0.26
0.04
0.16
0.06
0.64
7.8
0.2
Boston, MA
7.02
0.96
0.23
0.03
0.17
0.02
0.73
28.9
1.5
Camden, NJ
11.04
6.78
0.32
0.20
0.03
0.10
0.11
16.3
9.9
Casper, WY
6.97
1.50
0.22
0.04
0.12
0.04
0.54
8.9
1.0
Chester, PA
8.83
1.20
0.39
0.04
0.25
0.05
0.64
20.5
1.7
Chicago (region), IL
9.38
0.59
0.38
0.02
0.26
0.02
0.70
15.5
0.3
Chicago, IL
6.03
0.64
0.21
0.02
0.15
0.02
0.70
18.0
1.2
Corvallis, OR
10.68
1.80
0.22
0.03
0.20
0.03
0.91
32.6
4.1
El Paso, TX
3.93
0.86
0.32
0.05
0.23
0.05
0.72
5.9
1.0
Freehold, NJ
11.50
1.78
0.31
0.05
0.20
0.05
0.64
31.2
3.3
Gainesville, FL
6.33
0.99
0.22
0.03
0.16
0.03
0.73
50.6
3.1
Golden, CO
5.88
1.33
0.23
0.05
0.18
0.04
0.79
11.4
1.5
Grand Rapids, Ml
9.36
1.36
0.30
0.04
0.20
0.05
0.65
23.8
2.0
Hartford, CT
10.89
1.62
0.33
0.05
0.19
0.05
0.57
26.2
2.0
Houston, TX
4.55
0.48
0.31
0.03
0.25
0.03
0.83
18.4
1.0
Indiana15
8.80
2.68
0.29
0.08
0.27
0.07
0.92
20.1
3.2
Jersey City, NJ
4.37
0.88
0.18
0.03
0.13
0.04
0.72
11.5
1.7
Kansas'5
7.42
1.30
0.28
0.05
0.22
0.04
0.78
14.0
1.6
Kansas City (region),
MO/KS
7.79
0.85
0.39
0.04
0.26
0.04
0.67
20.2
1.7
99 See http://www.itreetools.org.
6-172 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
Lake Forest Park, WA
12.76
2.63
0.49
0.07
0.42
0.07
0.87
42.4
0.8
Las Cruces, NM
3.01
0.95
0.31
0.14
0.26
0.14
0.86
2.9
1.0
Lincoln, NE
10.64
1.74
0.41
0.06
0.35
0.06
0.86
14.4
1.6
Los Angeles, CA
4.59
0.51
0.18
0.02
0.11
0.02
0.61
20.6
1.3
Milwaukee, Wl
7.26
1.18
0.26
0.03
0.18
0.03
0.68
21.6
1.6
Minneapolis, MN
4.41
0.74
0.16
0.02
0.08
0.05
0.52
34.1
1.6
Moorestown, NJ
9.95
0.93
0.32
0.03
0.24
0.03
0.75
28.0
1.6
Morgantown, WV
9.52
1.16
0.30
0.04
0.23
0.03
0.78
39.6
2.2
Nebraska15
6.67
1.86
0.27
0.07
0.23
0.06
0.84
15.0
3.6
New York, NY
6.32
0.75
0.33
0.03
0.25
0.03
0.76
20.9
1.3
North Dakotab
7.78
2.47
0.28
0.08
0.13
0.08
0.48
2.7
0.6
Oakland, CA
5.24
0.19
NA
NA
NA
NA
NA
21.0
0.2
Oconomowoc, Wl
10.34
4.53
0.25
0.10
0.16
0.06
0.65
25.0
7.9
Omaha, NE
14.14
2.29
0.51
0.08
0.40
0.07
0.78
14.8
1.6
Philadelphia, PA
8.65
1.46
0.33
0.05
0.29
0.05
0.86
20.8
1.8
Phoenix, AZ
3.42
0.50
0.38
0.04
0.35
0.04
0.94
9.9
1.2
Roanoke, VA
9.20
1.33
0.40
0.06
0.27
0.05
0.67
31.7
3.3
Sacramento, CA
7.82
1.57
0.38
0.06
0.33
0.06
0.87
13.2
1.7
San Francisco, CA
9.18
2.25
0.24
0.05
0.22
0.05
0.92
16.0
2.6
Scranton, PA
9.24
1.28
0.40
0.05
0.30
0.04
0.74
22.0
1.9
Seattle, WA
9.59
0.98
0.67
0.06
0.55
0.05
0.82
27.1
0.4
South Dakotab
3.14
0.66
0.13
0.03
0.11
0.02
0.87
16.5
2.2
Syracuse, NY
9.48
1.08
0.30
0.03
0.22
0.04
0.72
26.9
1.3
Tennessee15
6.47
0.50
0.34
0.02
0.30
0.02
0.89
37.7
0.8
Washington, DC
8.52
1.04
0.26
0.03
0.21
0.03
0.79
35.0
2.0
Woodbridge, NJ
8.19
0.82
0.29
0.03
0.21
0.03
0.73
29.5
1.7
SE (Standard Error)
NA (Not Available)
a Ratio of net to gross sequestration
bStatewide assessment of urban areas
To determine gross sequestration rates, tree growth rates need to be estimated. Base growth rates were
standardized for open-grown trees in areas with 153 days of frost-free length based on measured data on tree
growth (Nowak et al. 2013). These growth rates were adjusted to local tree conditions based on length of frost-
free season, crown competition (as crown competition increased, growth rates decreased), and tree condition (as
tree condition decreased, growth rates decreased). Annual growth rates were applied to each sampled tree to
estimate gross annual sequestration-that is, the difference in carbon storage estimates between year 1 and year (x
+ 1) represents the gross amount of carbon sequestered. These annual gross carbon sequestration rates for each
tree were then scaled up to city estimates using tree population information. Total carbon sequestration was
divided by total tree cover to estimate a gross carbon sequestration density (kg C/m2 of tree cover/year). The area
of assessment for each city or state was defined by its political boundaries; parks and other forested urban areas
were thus included in sequestration estimates.
Where gross carbon sequestration accounts for all carbon sequestered, net carbon sequestration for settlement
trees considers carbon emissions associated with tree death and removals. The third step in the methodology
estimates net carbon emissions from settlement trees based on estimates of annual mortality, tree condition, and
assumptions about whether dead trees were removed from the site. Estimates of annual mortality rates by
diameter class and condition class were obtained from a study of street-tree mortality (Nowak 1986). Different
decomposition rates were applied to dead trees left standing compared with those removed from the site. For
removed trees, different rates were applied to the removed/aboveground biomass in contrast to the belowground
biomass (Nowak et al. 2002). The estimated annual gross carbon emission rates for each plot were then scaled up
to city estimates using tree population information.
The full methodology development is described in the underlying literature, and key details and assumptions were
made as follows. The allometric models applied to the field data for the Nowak methodology for each tree were
taken from the scientific literature (see Nowak 1994, Nowak et al. 2002), but if no allometric model could be found
Land Use, Land-Use Change, and Forestry 6-173
-------
for the particular species, the average result for the genus or botanical relative was used. The adjustment (0.8) to
account for less live tree biomass in open-grown urban trees was based on information in Nowak (1994).
Measured tree growth rates for street (Frelich 1992; Fleming 1988; Nowak 1994), park (deVries 1987), and forest
(Smith and Shifley 1984) trees were standardized to an average length of growing season (153 frost free days) and
adjusted for site competition and tree condition. Standardized growth rates of trees of the same species or genus
were then compared to determine the average difference between standardized street tree growth and
standardized park and forest growth rates. Crown light exposure (CLE) measurements (number of sides and/or top
of tree exposed to sunlight) were used to represent forest, park, and open (street) tree growth conditions. Local
tree base growth rates were then calculated as the average standardized growth rate for open-grown trees
multiplied by the number of frost-free days divided by 153. Growth rates were then adjusted for CLE. The CLE-
adjusted growth rate was then adjusted based on tree condition to determine the final growth rate. Assumptions
for which dead trees would be removed versus left standing were developed specific to each land use and were
based on expert judgment of the authors. Decomposition rates were based on literature estimates (Nowak et al.
2013).
Estimates of gross and net sequestration rates for each of the 50 states and the District of Columbia (Table 6-126)
were compiled in units of carbon sequestration per unit area of tree canopy cover. These rates were used in
conjunction with estimates of state settlement area and developed land percent tree cover data to calculate each
state's annual net carbon sequestration by urban trees. This method was described in Nowak et al. (2013) and has
been modified here to incorporate developed land percent tree cover data.
Net annual carbon sequestration estimates were obtained for all 50 states and the District of Columbia by
multiplying the gross annual emission estimates by 0.73, the average ratio for net/gross sequestration (Table
6-126). However, state specific ratios were used where available.
State Carbon Sequestration Estimates
The gross and net annual carbon sequestration values for each state were multiplied by each state's settlement
area of tree cover, which was the product of the state's settlement area and the state's tree cover percentage
based on NLCD developed land. The model used to calculate the total carbon sequestration amounts for each
state, can be written as follows:
Equation 6-1: Net State Annual Carbon Sequestration
Net state annual C sequestration (t C/yr) = Gross state sequestration rate (t C/ha/yr) x
Net to Gross state sequestration ratio x state settlement Area (ha) x % state tree cover in settlement area
The results for all 50 states and the District of Columbia are given in Table 6-126. This approach is consistent with
the default IPCC Gain-Loss methodology in IPCC (2006), although sufficient field data are not yet available to
separately determine interannual gains and losses in carbon stocks in the living biomass of settlement trees.
Instead, the methodology applied here uses estimates of net carbon sequestration based on modeled estimates of
decomposition, as given by Nowak et al. (2013).
Table 6-126: Estimated Annual Carbon Sequestration, Tree Cover, and Annual Carbon
Sequestration per Area of Tree Cover for settlement areas in the United States by State and
the District of Columbia (2022)
Gross Annual
Net Annual
Gross Annual
Net Annual
Sequestration
Sequestration
Net: Gross
Sequestration
Sequestration
per Area of
per Area of
Annual
(Metric Tons
(Metric Tons
Tree Cover
Tree Cover
Tree Cover
Sequestration
State
C/Year)
C/Year)
(Percent)
(kg C/m '/Year)
(kg C/m/Year)
Ratio
Alabama
2,268,736
1,653,170
53.1
0.376
0.274
0.73
Alaska
150,345
109,553
47.0
0.169
0.123
0.73
6-174 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
Arizona
165,494
120,591
4.5
0.388
0.283
0.73
Arkansas
1,316,764
959,492
48.5
0.362
0.264
0.73
California
2,015,461
1,468,615
16.8
0.426
0.311
0.73
Colorado
142,538
103,864
7.9
0.216
0.157
0.73
Connecticut
649,817
473,505
58.2
0.262
0.191
0.73
Delaware
101,819
74,193
24.2
0.366
0.267
0.73
DC
12,919
9,414
24.9
0.366
0.267
0.73
Florida
4,632,104
3,375,296
39.9
0.520
0.379
0.73
Georgia
3,886,939
2,832,313
55.9
0.387
0.282
0.73
Hawaii
302,023
220,076
41.4
0.637
0.464
0.73
Idaho
59,771
43,553
7.3
0.201
0.146
0.73
Illinois
669,891
488,132
15.4
0.310
0.226
0.73
Indiana
479,505
443,378
17.0
0.274
0.254
0.92
Iowa
177,874
129,612
8.5
0.263
0.191
0.73
Kansas
288,317
224,359
10.7
0.310
0.241
0.78
Kentucky
984,663
717,499
36.5
0.313
0.228
0.73
Louisiana
1,585,823
1,155,549
46.6
0.435
0.317
0.73
Maine
445,519
324,639
55.1
0.242
0.176
0.73
Maryland
857,152
624,585
39.8
0.353
0.257
0.73
Massachusetts
1,093,110
796,521
56.8
0.278
0.203
0.73
Michigan
1,410,284
1,027,638
34.4
0.241
0.175
0.73
Minnesota
325,047
236,853
13.0
0.251
0.183
0.73
Mississippi
1,630,583
1,188,165
56.9
0.377
0.275
0.73
Missouri
878,510
640,148
23.0
0.313
0.228
0.73
Montana
45,414
33,092
4.8
0.201
0.147
0.73
Nebraska
97,770
82,504
7.3
0.261
0.220
0.84
Nevada
35,783
26,074
4.8
0.226
0.165
0.73
New Hampshire
392,480
285,990
58.8
0.238
0.174
0.73
New Jersey
961,860
700,883
40.4
0.321
0.234
0.73
New Mexico
188,804
137,577
10.1
0.288
0.210
0.73
New York
1,606,981
1,170,966
39.6
0.263
0.192
0.73
North Carolina
3,457,794
2,519,606
53.7
0.341
0.249
0.73
North Dakota
18,730
8,900
1.7
0.244
0.116
0.48
Ohio
1,276,930
930,467
28.0
0.271
0.198
0.73
Oklahoma
718,922
523,860
21.9
0.364
0.265
0.73
Oregon
674,472
491,470
39.6
0.265
0.193
0.73
Pennsylvania
1,900,962
1,385,183
39.9
0.267
0.195
0.73
Rhode Island
127,720
93,066
49.6
0.283
0.206
0.73
South Carolina
2,052,656
1,495,718
53.3
0.370
0.269
0.73
South Dakota
29,351
25,453
2.8
0.258
0.224
0.87
Tennessee
1,678,890
1,501,125
40.8
0.332
0.297
0.89
Texas
4,416,309
3,218,052
28.2
0.403
0.294
0.73
Utah
119,794
87,291
11.6
0.235
0.172
0.73
Vermont
188,016
137,002
50.2
0.234
0.170
0.73
Virginia
2,111,293
1,538,445
52.5
0.321
0.234
0.73
Washington
1,139,218
830,119
37.3
0.282
0.206
0.73
West Virginia
774,594
564,427
63.6
0.264
0.192
0.73
Wisconsin
711,938
518,771
25.7
0.246
0.180
0.73
Wyoming
29,558
21,538
4.7
0.199
0.145
0.73
Total
51,287,245
37,768,294
Uncertainty
Uncertainty associated with changes in carbon stocks in settlement trees includes the uncertainty associated with
settlement area, percent tree cover in developed land and how well it represents percent tree cover in settlement
areas, and estimates of gross and net carbon sequestration for each of the 50 states and the District of Columbia. A
Land Use, Land-Use Change, and Forestry 6-175
-------
ten percent uncertainty was associated with settlement area estimates based on expert judgment. Uncertainty
associated with estimates of percent settlement tree coverage for each of the 50 states was based on standard
error associated with the photo-interpretation of national tree cover in developed lands. Uncertainty associated
with estimates of gross and net carbon sequestration for each of the 50 states and the District of Columbia was
based on standard error estimates for each of the state-level sequestration estimates (Table 6-127). These
estimates are based on field data collected in each of the 50 states and the District of Columbia, and uncertainty in
these estimates increases as they are scaled up to the national level.
Additional uncertainty is associated with the biomass models, conversion factors, and decomposition assumptions
used to calculate carbon sequestration and emission estimates (Nowak et al. 2002). These results also exclude
changes in soil carbon stocks, and there is likely some overlap between the settlement tree carbon estimates and
the forest tree carbon estimates (e.g., Nowak et al. 2013). Due to data limitations, settlement soil flux is not
quantified as part of this analysis, while reconciliation of settlement tree and forest tree estimates will be
addressed through the land-representation effort described in the Planned Improvements section of this chapter.
A Monte Carlo (Approach 2) uncertainty analysis was applied to estimate the overall uncertainty of the
sequestration estimate in 2022. The results of this quantitative uncertainty analysis are summarized in Table
6-127. The change in carbon stocks in settlement trees in 2022 was estimated to be between -208.5 and -66.6
MMT CO2 Eq. at a 95 percent confidence level. This analysis indicates a range of 51 percent more sequestration to
52 percent less sequestration than the 2022 flux estimate of -138.5 MMT CO2 Eq.
Table 6-127: Approach 2 Quantitative Uncertainty Estimates for Net CO2 Flux from Changes
in Carbon Stocks in Settlement Trees (MMT CO2 Eq. and Percent)
Uncertainty Range Relative to Flux Estimate-1
Source Gas 2022 Flux Estimate (MMTCO'Eq.) (%)
Lower Upper Lower Upper
(MMTCO'Eq.) Bound Bound Bound Bound
Changes in C Stocks in ~ ~ ~ ~ ~ ~
Settlement Trees ' ' ' ' ' '
a Range of C stock change estimates predicted by Monte Carlo stochastic simulation with a 95 percent confidence
interval.
Note: Parentheses indicate negative values or net sequestration.
QA/QC and Verification
Tier 1 and Tier 2 QA/QC activities were conducted consistent with the U.S. QA/QC plan. Source-specific quality
control measures for settlement trees included checking input data, documentation, and calculations to ensure
data were properly handled through the inventory process. Errors that were found during this process were
corrected as necessary.
Recalculations Discussion
The compilation methods remained the same in the latest Inventory relative to the previous Inventory. New data
from the NRI and NLCD resulted in a small decrease in the settlement area for 2021, leading to no substantial
change in the net carbon sequestration (Table 6-128).
Table 6-128: Recalculations of the Settlement Tree Categories
Category
2021 Estimate, 2021 Estimate, 2022 Estimate,
Previous Inventory Current Inventory Current Inventory
Settlement Area (km2)
Settlement Tree Coverage (km2)
Net C Flux (MMT C)
469,705
151,694
(37.6)
469,600
151,664
(37.6)
471,851
152,442
(37.8)
6-176 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
Net C02 FluxMMT C02 Eg.
(137.8)
(137.8)
(138.5)
Planned Improvements
A consistent representation of the managed land base in the United States is discussed in Section 6.1, and
discusses a planned improvement by the USDA Forest Service to reconcile the overlap between settlement trees
and the forest land categories. Estimates for settlement trees are based on tree cover in settlement areas. Work is
needed to clarify how much of this settlement area tree cover may also be accounted for in "forest" area
assessments as some of these forests may be adjacent to settlement areas. For example, "forest" as defined by the
USDA Forest Service Forest Inventory and Analysis (FIA) program fall within urban areas. Nowak et al. (2013)
estimates that 1.5 percent of forest plots measured by the FIA program fall within land designated as Census
urban, suggesting that approximately 1.5 percent of the carbon reported in the forest source category might also
be counted in the urban areas. The potential overlap with settlement areas is unknown at this time but research is
underway to develop spatially explicit and spatially continuous land representation products which will eliminate
the potential for double counting. Future research may also enable more complete coverage of changes in the
carbon stock of trees for all settlements land.
N20 Emissions from Settlement Soils (CRT Source Category
4E1)
Of the synthetic N fertilizers applied to soils in the United States, approximately 1 to 2 percent are currently
applied to lawns, golf courses, and other landscaping within settlement areas, and contributes to soil N2O
emissions. The area of settlements is considerably smaller than other land uses that are managed with fertilizer,
particularly cropland soils, and therefore, settlements account for a smaller proportion of total synthetic fertilizer
application in the United States. In addition to synthetic N fertilizers, a portion of surface applied biosolids (i.e.,
treated sewage sludge) is used as an organic fertilizer in settlement areas, and drained organic soils (i.e., soils with
high organic matter content, known as histosols) also contribute to emissions of soil N2O.
N additions to soils result in direct and indirect N2O emissions. Direct emissions occur on-site due to the N
additions in the form of synthetic fertilizers and biosolids as well as enhanced mineralization of N in drained
organic soils. Indirect emissions result from fertilizer and biosolids N that is transformed and transported to
another location in a form other than N2O (i.e., volatilization of ammonia [NH3] and nitrogen oxide [NOx], and
leaching/runoff of nitrate [NO3 ]), and later converted into N2O at the off-site location. The indirect emissions are
assigned to settlements because the management activity leading to the emissions occurred in settlements.
Total N2O emissions from soils in settlements remaining settlements100 are 2.5 MMT CO2 Eq. (10 kt of N2O) in
2022. There is an overall increase of 23 percent from 1990 to 2022 due to an expanding settlement area leading to
more synthetic N fertilizer applications that peaked in the mid-2000s. Inter-annual variability in these emissions is
directly attributable to variability in total synthetic fertilizer consumption, area of drained organic soils, and
biosolids applications in the United States. Emissions from this source are summarized in Table 6-129.
Table 6-129: N2O Emissions from Soils in Settlements Remaining Settlements (MMT CO2 Eq.
and kt N2O)
100 Estimates of Soil N20 for settlements remaining settlements include emissions from land converted to settlements because
it was not possible to separate the activity data.
1990 2005 2018 2019 2020 2021 2022
MMT COz Eq.
Direct N20 Emissions from Soils
Synthetic Fertilizers
1.7 2.6 2.1 2.1 2.2 2.2 2.2
0.8( 1.51 0.8 0.8 0.8 0.8 0.8
Land Use, Land-Use Change, and Forestry 6-177
-------
Biosolids
°-2:
0.2 =
0.2
0.2
0.2
0.2
0.2
Drained Organic Soils
0.8
1.0
1.2
1.2
1.2
1.2
1.2
Indirect N20 Emissions from Soils
0.3
0.5
0.3
0.3
0.3
0.3
0.3
Total
2.1
3.1
2.4
2.5
2.5
2.5
2.5
kt N20
Direct N20 Emissions from Soils
7
10
8
8
8
8
8
Synthetic Fertilizers
3
5
3
3
3
3
3
Biosolids
1
11
1
1
1
1
1
Drained Organic Soils
3
4
4
4
4
4
4
Indirect N20 Emissions from Soils
1
2
1
1
1
1
1
Total
8
12
9
9
9
10
10
Note: Totals may not sum due to independent rounding.
Methodology and Time-Series Consistency
For settlement soils, the IPCC Tier 1 approach is used to estimate soil N2O emissions from synthetic N fertilizer,
biosolids additions, and drained organic soils. Estimates of direct N2O emissions from soils in settlements are based
on the amount of N in synthetic commercial fertilizers applied to settlement soils, the amount of N in biosolids
applied to non-agricultural land and surface disposal (see Section 7.2 for a detailed discussion of the methodology
for estimating biosolids available for non-agricultural land application), and the area of drained organic soils within
settlements.
Nitrogen applications to settlement soils are estimated using data compiled by the USGS (Brakebill and Gronberg
2017). The USGS estimated on-farm and non-farm fertilizer use is based on sales records at the county level from
1987 through 2012 (Brakebill and Gronberg 2017). Non-farm N fertilizer is assumed to be applied to settlements
and forest lands; values for 2013 through 2017 are based on 2012 values adjusted for total annual total N fertilizer
sales in the United States (AAPFCO 2016 through 2022) because there are no activity data on non-farm application
after 2012. Settlement application is calculated by subtracting forest application from total non-farm fertilizer use.
Since the total N fertilizer sales is only available through 2017 (AAPFCO 2022), the amount of synthetic fertilization
from 2018 to 2022 is determined using a linear extrapolation method (see Box 6-4 in cropland remaining
cropland). This method is based on a linear regression model with moving-average (ARMA) errors using the 1990
to 2017 fertilization data. To estimate direct N2O for the time series, the total amount of fertilizer N applied to
settlements is multiplied by the IPCC default emission factor (1 percent) (IPCC 2006) for 1990 to 2022.
Biosolids applications are derived from national data on biosolids generation, disposition, and N content (see
Section 7.2 for further detail). The total amount of N resulting from these sources is multiplied by the IPCC default
emission factor for applied N (one percent) to estimate direct N2O emissions (IPCC 2006) for 1990 to 2022.
The IPCC (2006) Tier 1 method is also used to estimate direct N2O emissions due to drainage of organic soils in
settlements at the national scale. Estimates of the total area of drained organic soils are obtained from the 2017
National Resources Inventory (NRI) (USDA-NRCS 2020) using soils data from the Soil Survey Geographic Database
(SSURGO) (Soil Survey Staff 2020). The NRI time series has been extended through 2020 using the National Land
Cover Dataset (Yang et al. 2018). The areas have been modified through a process in which the Forest Inventory
and Analysis (FIA) survey data are harmonized with the NRI data (Nelson et al. 2020). This process ensures that the
land use areas are consistent across all land use categories (see Section 6.1 for more information). All settlements
occurring on organic soil are assumed to be drained for the purposes of approximating greenhouse gas emissions.
To estimate annual emissions from 1990 to 2020, the total area is multiplied by the IPCC default emission factor
for temperate regions (IPCC 2006). The annual emissions for 2021 to 2022 are estimated using a linear
extrapolation method (see Box 6-4 in Cropland Remaining Cropland). This Inventory does not include soil N2O
emissions from drainage of organic soils in Alaska and federal lands, although this is a planned improvement for a
future Inventory.
For indirect emissions, the total N applied from fertilizer and biosolids is multiplied by the IPCC default factors of
10 percent for volatilization and 30 percent for leaching/runoff to calculate the amount of N volatilized and the
6-178 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
amount of N leached/runoff. The amount of N volatilized is multiplied by the IPCC default factor of one percent for
the portion of volatilized N that is converted to N2O off-site and the amount of N leached/runoff is multiplied by
the IPCC default factor of 0.075 percent for the portion of leached/runoff N that is converted to N2O off-site. The
resulting estimates are summed to obtain total indirect emissions from 1990 to 2022 for biosolids and synthetic
fertilization.
In order to ensure time-series consistency, the same methods are applied from 1990 to 2022 for biosolids. For
synthetic fertilizer, a linear extrapolation method is used to approximate fertilizer application for the remainder of
the 2018 to 2022 time series and then used to estimate emissions. For drainage of organic soils, the methods
described above are applied for 1990 to 2020, and a linear extrapolation method is used to approximate emissions
for the remainder of the 2021 to 2022 time series (see Box 6-4 in cropland remaining cropland). The extrapolation
is based on a linear regression model with moving-average (ARMA) errors using the 1990 to 2020 emissions data,
which is a standard data splicing method for imputing missing emissions data in a time series (IPCC 2006). The time
series will be recalculated in a future Inventory with the methods described previously for drainage of organic soils.
Uncertainty
The amount of N2O emitted from settlement soils depends not only on N inputs and area of drained organic soils,
but also on a large number of variables that can influence rates of nitrification and denitrification, including organic
C availability; rate, application method, and timing of N input; oxygen gas partial pressure; soil moisture content;
pH; temperature; and irrigation/watering practices. The effect of the combined interaction of these variables on
N2O emissions is complex and highly uncertain. The IPCC default methodology does not explicitly incorporate these
variables, except variation in the total amount of fertilizer N and biosolids application, which leads to uncertainty
in the results.
Uncertainties exist in both the fertilizer N and biosolids application rates in addition to the emission factors.
Uncertainty in fertilizer N application is assigned a default level of ±50 percent.101 For emissions from drained
organic soils, the total uncertainty was quantified with two variance components (Ogle et al. 2010) that are
combined using 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 first variance component is
associated with uncertainty in the emission factor, and the second variance component is associated with scaling
of the data from the NRI survey to the entire area of drained organic soils in settlements remaining settlements,
and is computed using a standard variance estimator for a two-stage sample design (Sarndal et al. 1992). There is
also additional uncertainty associated with the fit of the linear regression model for the data splicing methods that
was incorporated into the analysis for the latter part of the time series.
Uncertainty is propagated through the calculations of N2O emissions from fertilizer N and drainage of organic soils
based on a Monte Carlo analysis. The results are combined with the uncertainty in N2O emissions from the
biosolids application using simple error propagation methods (IPCC 2006). The results are summarized in Table
6-130. Direct N2O emissions from soils in settlements remaining settlements in 2022 are estimated to be between
1.2 and 3.4 MMT CO2 Eq. at a 95 percent confidence level. This indicates a range of 47 percent below to 54 percent
above the 2022 emission estimate of 2.2 MMT CO2 Eq. Indirect N2O emissions in 2022 are between 0.1 and 1.1
MMT CO2 Eq., ranging from 76 percent below to 218 percent above the estimate of 0.3 MMT CO2 Eq.
101 No uncertainty is provided with the USGS fertilizer consumption data (Brakebill and Gronberg 2017) so a conservative ±50
percent is used in the analysis. Biosolids data are also assumed to have an uncertainty of ±50 percent.
Land Use, Land-Use Change, and Forestry 6-179
-------
Table 6-130: Quantitative Uncertainty Estimates of N2O Emissions from Soils in Settlements
Remaining Settlements (MMT CO2 Eq. and Percent)
Source
Gas
2022 Emissions
(MMT CO . Eq.)
Uncertainty Range Relative to Emission Estimate-1
(MMTCO' Eq.) (%)
Lower
Bound
Upper
Bound
Lower
Bound
Upper
Bound
Settlements Remaining Settlements
Direct N20 Emissions from Soils
N20
2.2
1.2
3.4
-47%
+54%
Indirect N20 Emissions from Soils
n2o
0.3
0.1
1.1
-76%
+218%
a Range of emission estimates is a 95 percent confidence interval.
Note: These estimates include direct and indirect N20 emissions from settlements remaining settlements and land
converted to settlements because it was not possible to separate the activity data.
QA/QC and Verification
The spreadsheet containing fertilizer, drainage of organic soils, and biosolids applied to settlements and
calculations for N2O and uncertainty ranges have been checked consistent with the U.S. Inventory QA/QC plan,
which is in accordance with Volume 1, Chapter 6 of the 2006IPCC Guidelines (see Annex 8 for more details). An
error was found in the initial calculations for emissions from drained organic soils, which was corrected.
Recalculations Discussion
Recalculations are associated with updated land areas for drainage of organic soils in settlements remaining
settlements, by incorporating new USDA-NRCS NRI data through 2017 and extending the time series using CDL and
NLCD for grassland converted to settlements, cropland converted to settlements, other land converted to
settlements and wetlands converted to settlements. In addition, recalculations are associated with revised
fertilizer application data from the AAPFCO report. As a result of these changes, C02-equivalent emissions changed
annually with an average annual increase of 0.36 MMT CO2 Eq., or 17 percent, over the time series from 1990 to
2021 compared to the previous Inventory.
Planned Improvements
This source will be extended to include soil N2O emissions from drainage of organic soils in Alaska and federal lands
in order to provide a complete inventory of emissions for this category. Data on fertilizer amounts from 2018 to
2022 will be updated after data are released for the latter part of the time series. These improvements will be
incorporated into a future Inventory, pending prioritization of resources.
Changes in Yard Trimmings and Food Scrap Carbon Stocks in
Landfills (CRT Category 4E1)
In the United States, yard trimmings (i.e., grass clippings, leaves, and branches) and food scraps account for a
significant portion of the municipal waste stream, and a large fraction of the collected yard trimmings and food
scraps are put in landfills. A portion of the carbon contained in landfilled yard trimmings and food scraps can be
stored for very long periods.
Carbon storage estimates within the Inventory are associated with particular land uses. For example, harvested
wood products are reported under forest land remaining forest land because these wood products originated from
the forest ecosystem. Similarly, carbon stock changes in yard trimmings and food scraps are reported under
settlements remaining settlements because the bulk of the carbon, which comes from yard trimmings, originates
from settlement areas. While the majority of food scraps originate from cropland and grassland, in this Inventory
6-180 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
they are reported with the yard trimmings in the settlements remaining settlements section. Additionally, landfills
are considered part of the managed land base under settlements (see Section 6.1), and reporting these carbon
stock changes that occur entirely within landfills fits most appropriately within the settlements remaining
settlements section. The CFU emissions resulting from anaerobic decomposition of yard trimmings and food scraps
in landfills are reported in the Waste chapter, see Section 7.1.
The estimated amount of yard trimmings collected annually has stagnated since 1990 and the fraction that is
landfilled has been declining since 1990. From 1970 to 1990, yard trimmings collected for disposal increased by
about 51 percent. In 1990, over 53 million metric tons (wet weight) of yard trimmings and food scraps are
estimated to have been generated (i.e., put at the curb for collection to be taken to disposal sites or to composting
facilities) (EPA 2020). Since then, programs banning or discouraging yard trimmings disposal in landfills have led to
an increase in backyard composting and the use of mulching mowers, and consequently a slowing of year-over-
year increases in the tonnage of yard trimmings generated. From 1990 to 2022, yard trimmings collected for
disposal are estimated to have increased 1.1. percent. At the same time, an increase in the number of municipal
composting facilities has reduced the proportion of collected yard trimmings that are discarded in landfills per
year—from 72 percent in 1990 to 30 percent in 2022. The net effect of the slight increase in generation and the
increase in composting is a 58 percent decrease in the quantity of yard trimmings disposed of in landfills since
1990. Composting trends and emissions estimations are presented in the Waste chapter, Section 7.3 composting.
Food scrap generation has grown by an estimated 165 percent since 1990. Though the proportion of total food
scraps generated that are eventually discarded in landfills has decreased from an estimated 82 percent in 1990 to
55 percent in 2020, the tonnage disposed of in landfills has increased considerably (by an estimated 78 percent)
due to the increase in food scrap generation. Although the total tonnage of food scraps disposed of in landfills has
increased from 1990 to 2022, the difference in the amount of food scraps added from one year to the next
generally decreased, and consequently the annual net changes in carbon stock from food scraps have generally
decreased as well (as shown in Table 6-131 and Table 6-132). Landfilled food scraps decompose over time,
producing CFU and CO2. Decomposition happens at a higher rate initially, then decreases. As decomposition
decreases, the carbon stock becomes more stable. Because the cumulative carbon stock left in the landfill from
previous years is (1) not decomposing as much as the carbon introduced from food scraps in a single more recent
year; and (2) is much larger than the carbon introduced from food scraps in a single more recent year, the total
carbon stock in the landfill is primarily driven by the more stable "older" carbon stock, thus resulting in decreasing
annual changes in later years.
Overall, the decrease in the landfill disposal rate of yard trimmings has more than compensated for the increase in
food scrap disposal in landfills, and the net result is a decrease in the annual net change in landfill carbon storage
from 24.5 MMT C02 Eq. (6.7 MMT C) in 1990 to 11.8 MMT C02 Eq. (3.2 MMT C) in 2022 (Table 6-131 and Table
6-132), a decrease of 48 percent over the time series.
Table 6-131: Net Changes in Yard Trimmings and Food Scrap Carbon Stocks in Landfills
(MMT C02 Eq.)
Carbon Pool
1990
2005
2018
2019
2020
2021
2022
Yard Trimmings
(20.1)
(7.5)
(8.3)
(8.2)
(8.2)
(8.2)
(8.1)
Grass
(0.6)1
(0.8)
(0.8)
(0.8)
(0.8)
(0.7)
Leaves
(8.7)
(3.4)
(3.8)
(3.8)
(3.8)
(3.8)
(3.7)
Branches
<9-8)!!l
(3.4)
(3.7)
(3.7)
(3.7)
(3.7)
(3.6)
Food Scraps
(4.4)
(3.9)
(5.2)
(4.8)
(4.5)
(4.3)
(3.7)
Total Net Flux
(24.5)
(11.4)
(13.4)
(13.1)
(12.8)
(12.5)
(11.8)
Notes: Totals may not sum due to independent rounding. Parentheses indicate net sequestration.
Land Use, Land-Use Change, and Forestry 6-181
-------
Table 6-132: Net Changes in Yard Trimmings and Food Scrap Carbon Stocks in Landfills
(MMT C)
Carbon Pool
1990
2005
2018
2019
2020
2021
2022
Yard Trimmings
(5.5)
(2.0)
(2.3)
(2.2)
(2.2)
(2.2)
(2.2)
Grass
(0.5) (
(0.2)
(0.2)
(0.2)
(0.2)
(0.2)
(0.2)
Leaves
(2.4)
(0.9)
(1.0)
(1.0)
(1.0)
(1.0)
(1.0)
Branches
(2-7)I
(0.9)
(1.0)
(1.0)
(1.0)
(1.0)
(1.0)
Food Scraps
(1.2)
(1-1)
(1.4)
(1.3)
(1.2)
(1.2)
(1.0)
Total Net Flux
(6.7)
(3.1)
(3.7)
(3.6)
(3.5)
(3.4)
(3.2)
Notes: Totals may not sum due to independent rounding. Parentheses indicate net sequestration.
Methodology and Time-Series Consistency
When wastes of biogenic origin (such as yard trimmings and food scraps) are landfilled and do not completely
decompose, the carbon that remains is effectively removed from the carbon cycle. Empirical evidence indicates
that yard trimmings and food scraps do not completely decompose in landfills (Barlaz 1998, 2005, 2008; De la Cruz
and Barlaz 2010), and thus the stock of carbon in landfills can increase, with the net effect being removal of carbon
from the atmosphere. Estimates of the net carbon flux resulting from landfilled yard trimmings and food scraps
were developed by estimating the change in landfilled carbon stocks between inventory years and uses a country-
specific methodology based on the methodology for estimating the amount of harvested wood products stored in
solid waste disposal systems that is provided in the Land Use, Land-Use Change, and Forestry sector in IPCC (2003)
and the 2006 IPCC Guidelines for National Greenhouse Gas Inventories (IPCC 2006). Carbon stock estimates were
calculated by determining the mass of landfilled carbon resulting from yard trimmings and food scraps discarded in
a given year; adding the accumulated landfilled carbon from previous years; and subtracting the mass of carbon
that was landfilled in previous years and has since decomposed and been emitted as CO2 and Cm.
To determine the total landfilled carbon stocks for a given year, the following data and factors were assembled:
(1) The composition of the yard trimmings (i.e., the proportion of grass, leaves and branches);
(2) The mass of yard trimmings and food scraps discarded in landfills;
(3) The carbon storage factor of the landfilled yard trimmings and food scraps; and
(4) The rate of decomposition of the degradable carbon.
The composition of yard trimmings was assumed to be 30 percent grass clippings, 40 percent leaves, and 30
percent branches on a wet weight basis (Oshins and Block 2000). The yard trimmings were subdivided, because
each component has its own unique adjusted carbon storage factor (i.e., based on differences in moisture content
and carbon content) and rate of decomposition. The mass of yard trimmings and food scraps disposed of in
landfills was estimated by multiplying the quantity of yard trimmings and food scraps discarded by the proportion
of discards managed in landfills. Data on discards (i.e., the amount generated minus the amount diverted to
centralized composting facilities) for both yard trimmings and food scraps were taken primarily from Advancing
Sustainable Materials Management: Facts and Figures 2018 (EPA 2020), which provides data for 1960,1970,1980,
1990, 2000, 2005, 2010, 2015, 2017 and 2018. To provide data for some of the missing years, detailed backup data
were obtained from the 2012, 2013, and 2014, 2015, and 2017 versions of the Advancing Sustainable Materials
Management: Facts and Figures reports (EPA 2019), as well as historical data tables that EPA developed for 1960
through 2012 (EPA 2016). Remaining years in the time series for which data were not provided were estimated
using linear interpolation. Since the Advancing Sustainable Materials Management: Facts and Figures reports for
2019 through 2022 were unavailable, landfilled material generation, recovery, and disposal data for 2019 through
2022 were proxied equal to 2018 values.
The amount of carbon disposed of in landfills each year, starting in 1960, was estimated by converting the
discarded landfilled yard trimmings and food scraps from a wet weight to a dry weight basis, and then multiplying
6-182 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
by the initial (i.e., pre-decomposition) carbon content (as a fraction of dry weight). The dry weight of landfilled
material was calculated using dry weight to wet weight ratios (Tchobanoglous et al. 1993, cited by Barlaz 1998) and
the initial carbon contents and the carbon storage factors were determined by Barlaz (1998, 2005, 2008).
The amount of carbon remaining in the landfill for each subsequent year was tracked based on a simple model of
carbon fate based on a laboratory experiment simulating decomposition of landfilled biogenic materials by
methanogenic microbes (Barlaz 1998, 2005, 2008). Carbon remaining in landfilled materials is expressed as a
proportion of initial carbon content, shown in the row labeled "C Storage Factor, Proportion of Initial C Stored (%)"
in Table 6-133.
The modeling approach applied to simulate U.S. landfill carbon flows builds on the findings of Barlaz (1998, 2005,
2008). The proportion of carbon stored is assumed to persist in landfills. The remaining portion is assumed to
degrade over time, resulting in emissions of Cm and CO2.102 The degradable portion of the carbon is assumed to
decay according to first-order kinetics. The decay rates for each of the materials are shown in Table 6-133.
The first-order decay rates, k, for each waste component are derived from De la Cruz and Barlaz (2010):
• De la Cruz and Barlaz (2010) calculate first-order decay rates using laboratory data published in Eleazer et
al. (1997), and a correction factor,/, is calculated so that the weighted average decay rate for all
components is equal to the EPA AP-42 default decay rate (0.04) for mixed municipal solid waste (MSW)
for regions that receive more than 25 inches of rain annually (EPA 1995). Because AP-42 values were
developed using landfill data from approximately 1990, De la Cruz and Barlaz used 1990 waste
composition for the United States from EPA's Characterization of Municipal Solid Waste in the United
States: 1990 Update (EPA 1991) to calculate/. De la Cruz and Barlaz multiplied this correction factor by
the Eleazer et al. (1997) decay rates of each waste component to develop field-scale first-order decay
rates.
• De la Cruz and Barlaz (2010) also use other assumed initial decay rates for mixed MSW in place of the AP-
42 default value based on different types of environments in which landfills in the United States are
located, including dry conditions (less than 25 inches of rain annually, k=0.02) and bioreactor landfill
conditions (moisture is controlled for rapid decomposition, /c=0.12).
Similar to the methodology in the Landfills section of the Inventory (Section 7.1), which estimates CH4 emissions,
the overall MSW decay rate is estimated by partitioning the U.S. landfill population into three categories based on
annual precipitation ranges of: (1) Less than 20 inches of rain per year, (2) 20 to 40 inches of rain per year, and (3)
greater than 40 inches of rain per year. These correspond to overall MSW decay rates of 0.020,0.038, and 0.057
year"1, respectively. De la Cruz and Barlaz (2010) calculate component-specific decay rates corresponding to the
first value (0.020 year"1), but not for the other two overall MSW decay rates.
To maintain consistency between landfill-related methodologies across the Inventory, EPA developed correction
factors (/) for decay rates of 0.038 and 0.057 year"1 through linear interpolation. A weighted national average
component-specific decay rate is calculated by assuming that waste generation is proportional to population (the
same assumption used in the landfill methane emission estimate), based on population data from the 2000 U.S.
Census. The percent of census population is calculated for each of the three categories of annual precipitation
(noted in the previous paragraph); the population data are used as a surrogate for the number of landfills in each
annual precipitation category. Precipitation range percentages weighted by population are updated over time as
new Census data are available, to remain consistent with percentages used in the Waste chapter, Section 7.1
landfills. The component-specific decay rates are shown in Table 6-133.
De la Cruz and Barlaz (2010) also use other assumed initial decay rates for mixed MSW in place of the AP-42
default value based on different types of environments in which landfills in the United States are located, including
102 The CH4 emissions resulting from anaerobic decomposition of yard trimmings and food scraps in landfills are reported in the
Waste chapter, Section 7.1—Landfills.
Land Use, Land-Use Change, and Forestry 6-183
-------
dry conditions (less than 25 inches of rain annually, k=0.02) and bioreactor landfill conditions (moisture is
controlled for rapid decomposition, /c=0.12).
For each of the four materials (grass, leaves, branches, food scraps), the stock of carbon in landfills for any given
year is calculated according to Equation 6-2:
Equation 6-2: Total Carbon Stock for Yard Trimmings and Food Scraps in Landfills
t
LFCi t = x (1- MQ) x ICC, x {[C5; x ICC,] + [(l - (CSt x ICC,)) x e^"^]}
n
where,
t = Year for which carbon stocks are being estimated (year),
/ = Waste type for which carbon stocks are being estimated (grass, leaves, branches, food
scraps),
LFQt = Stock of carbon in landfills in year t, for waste / (metric tons),
Wi,n = Mass of waste / disposed of in landfills in year n (metric tons, wet weight),
n = Year in which the waste was disposed of (year, where 1960 < n < t),
MCi = Moisture content of waste / (percent of water),
CSi = Proportion of initial carbon that is stored for waste / (percent),
ICC, = Initial carbon content of waste / (percent),
e = Natural logarithm, and
k = First-order decay rate for waste /', (year-1).
For a given year t, the total stock of carbon in landfills (TLFCt) is the sum of stocks across all four materials (grass,
leaves, branches, food scraps). The annual flux of carbon in landfills (Ft) for year t is calculated in as the change in
carbon stock compared to the preceding year according to Equation 6-3:
Equation 6-3: Carbon Stock Annual Flux for Yard Trimmings and Food Scraps in Landfills
Ft = TLFCt-TLFC(t_1}
Thus, as seen in Equation 6-2, the carbon placed in a landfill in year n is tracked for each year t through the end of
the inventory period. For example, disposal of food scraps in 1960 resulted in depositing about 1,135,000 metric
tons of carbon in landfills. Of this amount, 16 percent (179,000 metric tons) is persistent; the remaining 84 percent
(956,000 metric tons) is degradable. By 1965, more than half of the degradable portion (507,000 metric tons)
decomposes, leaving a total of 628,000 metric tons (the persistent portion, plus the remainder of the degradable
portion).
Continuing the example, by 2022, the total food scraps carbon originally disposed of in 1960 had declined to
179,000 metric tons (i.e., virtually all degradable carbon had decomposed). By summing the carbon remaining
from 1960 with the carbon remaining from food scraps disposed of in subsequent years (1961 through 2021), the
total landfill carbon from food scraps in 2022 was 53.0 million metric tons. This value is then added to the carbon
stock from grass, leaves, and branches to calculate the total landfill carbon stock in 2022, yielding a value of 292.6
million metric tons (as shown in Table 6-134). In the same way total net flux is calculated for forest carbon and
harvested wood products, the total net flux of landfill carbon for yard trimmings and food scraps for a given year
(Table 6-132) is the difference in the landfill carbon stock for the following year and the stock in the current year.
For example, the net change in 2022 shown in Table 6-132 (3.2 MMT C with rounding) is equal to the stock in 2023
(295.9 MMT C) minus the stock in 2022 (292.6 MMT C). The carbon stocks calculated through this procedure are
shown in Table 6-134.
6-184 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
To develop the 2023 carbon stock estimate, estimates of yard trimming and food scrap carbon stocks were
forecasted for 2023, based on data from 1990 through 2022. These forecasted values were used to calculate net
changes in carbon stocks for 2022. Excels FORECAST.ETS function was used to predict a 2023 value using historical
data via an algorithm called "Exponential Triple Smoothing." This method determined the overall trend and
provided appropriate carbon stock estimates for 2023.
Table 6-133: Moisture Contents, Carbon Storage Factors (Proportions of Initial Carbon
Sequestered), Initial C Contents, and Decay Rates for Yard Trimmings and Food Scraps in
Landfills
Variable
Grass
Yard Trimmings
Leaves
Branches
Food Scraps
Moisture Content (% H20)
70
30
10
70
C Storage Factor, Proportion of Initial C
Stored (%)
53
85
77
16
Initial C Content (%)
45
46
49
51
Decay Rate (year1)
0.313
0.179
0.015
0.151
Note: The decay rates are presented as weighted averages based on annual precipitation categories and
population residing in each precipitation category.
Table 6-134: Carbon Stocks in Yard Trimmings and Food Scraps in Landfills (MMT C)
Carbon Pool
1990
2005
2018
2019
2020
2021
2022
2023
Yard Trimmings
156.0
203.1
231.6
233.9
236.1
238.4
240.6
242.8
Branches
14.61
18. l.|
20.7
20.9
21.1
21.3
21.6
21.8
Leaves
66.7
87.4
100.4
101.5
102.5
103.6
104.6
105.6
Grass
74.7
97.71
110.5
111.5
112.5
113.5
114.5
115.4
Food Scraps
17.9
33.2
46.9
48.3
49.6
50.9
52.0
53.0
Total Carbon Stocks
173.9
236.3
278.5
282.2
285.7
289.2
292.6
295.9
a 2023 C stock estimate was forecasted using 1990 to 2022 data.
Note: Totals may not sum due to independent rounding.
Methodological recalculations were applied to the entire time series to ensure time-series consistency from 1990
through 2022. When available, the same data source was used across the entire time series for the analysis. When
data were unavailable, missing values were estimated using linear interpolation or forecasting, as noted above.
Uncertainty
The uncertainty analysis for landfilled yard trimmings and food scraps includes an evaluation of the effects of
uncertainty for the following data and factors: disposal in landfills per year (tons of carbon), initial carbon content,
moisture content, decay rate, and proportion of carbon stored. The carbon storage landfill estimates are also a
function of the composition of the yard trimmings (i.e., the proportions of grass, leaves and branches in the yard
trimmings mixture). There are respective uncertainties associated with each of these factors.
A Monte Carlo (Approach 2) uncertainty analysis was applied to estimate the overall uncertainty of the
sequestration estimate for 2022. The results of the Approach 2 quantitative uncertainty analysis are summarized in
Table 6-135. Total yard trimmings and food scraps CO2 flux in 2022 was estimated to be between -17.3 and -4.9
MMT CO2 Eq. at a 95 percent confidence level. This indicates a range of 47 percent below to 58 percent above the
2022 flux estimate of -11.8 MMT CO2 Eq.
Land Use, Land-Use Change, and Forestry 6-185
-------
Table 6-135: Approach 2 Quantitative Uncertainty Estimates for CO2 Flux from Yard
Trimmings and Food Scraps in Landfills (MMT CO2 Eq. and Percent)
Source
2021 Flux Estimate
Gas
(MMTCO. Eq.)
Uncertainty Range Relative to Flux Estimate-1
(MMTCO' Eq.) (%)
Lower Upper
Lower Upper
Bound Bound
Bound Bound
Yard Trimmings and Food
Scraps
C02 (11.8)
(17.3) (4.9)
-47% +58%
a Range of flux estimates predicted by Monte Carlo stochastic simulation for a 95 percent confidence interval.
Note: Parentheses indicate negative values or net carbon sequestration.
QA/QC and Verification
Tier 1 and Tier 2 QA/QC activities were conducted consistent with the U.S. Inventory QA/QC plan. Source-specific
quality control measures for landfilled yard trimmings and food scraps included checking that input data were
properly transposed within the spreadsheet, checking calculations were correct, and confirming that all activity
data and calculations documentation was complete and updated to ensure data were properly handled through
the inventory process.
Order of magnitude checks and checks of time-series consistency were performed to ensure data were updated
correctly and any changes in emissions estimates were reasonable and reflected changes in activity data. An
annual change trend analysis was also conducted to ensure the validity of the emissions estimates. Errors that
were found during this process were corrected as necessary.
To ensure consistency across the LULUCF and Waste sectors, and the accuracy of emissions, EPA plans to perform
a comparison of the activity data used and carbon inputs between the landfilled yard trimmings and food scraps,
and the Waste chapter, Section 7.1—Landfills categories.
Recalculations Discussion
No recalculations were performed for the current Inventory.
Planned Improvements
EPA notes the following improvements may be implemented or investigated within the next two or three Inventory
cycles pending time and resource constraints:
• MSW data more recent than 2018 have not been released through the Advancing Sustainable Materials
Management reports. EPA will monitor the release schedule for these data and evaluate data for
integration into the Inventory when released. Six new food waste management pathways were
introduced in the 2018 Advancing Sustainable Materials Management report. Time series data for all of
these pathways are not provided prior to 2018 but EPA plans to investigate potential data sources and/or
methods to address time-series consistency and apply these data to the time series.
• EPA has been made aware of inconsistencies in landfilled food scraps data reported to the EPA
Greenhouse Gas Reporting Program (GHGRP) and will evaluate changes to how landfilled and energy
recovery values for yard trimmings and food scraps are calculated.
EPA notes the following improvements will continue to be investigated as time and resources allow, but there are
no immediate plans to implement these improvements until data are available or identified:
• EPA also plans to continue to investigate updates to the decay rate estimates for food scraps, leaves,
grass, and branches, as well as evaluate using decay rates that vary over time based on Census population
6-186 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
and climate data changes over time. Currently the inventory calculations use 2010 U.S. Census data, but
2020 U.S. Census data may be available.
• Other improvements include investigation into yard waste composition to determine if changes need to
be made based on changes in residential practices. A review of available literature will be conducted to
determine if there are changes in the allocation of yard trimmings. For example, leaving grass clippings in
place is becoming a more common practice, thus reducing the percentage of grass clippings in yard
trimmings disposed in landfills. In addition, agronomists may be consulted for determining the mass of
grass per acre on residential lawns to provide an estimate of total grass generation for comparison with
Inventory estimates.
• EPA will continue to evaluate data from recent peer-reviewed literature that may modify the default
carbon storage factors, initial carbon contents, and decay rates for yard trimmings and food scraps in
landfills - particularly updates to population precipitation ranges used to calculate k values. Based upon
this evaluation, changes may be made to the default values.
• Finally, EPA plans to review available data to ensure all types of landfilled yard trimmings and food scraps
are being included in the Inventory estimates, such as debris from road construction and commercial food
waste not included in other Inventory estimates.
6.11 Land Converted to Settlements (CRT
Category 4E2)
Land converted to settlements includes all settlements in an inventory year that had been in another land use(s)
during the previous 20 years (USDA-NRCS 2015).103 For example, cropland, grassland or forest land converted to
settlements during the past 20 years would be reported in this category. Converted lands are retained in this
category for 20 years as recommended by IPCC (2006).
Land use change can lead to large losses of carbon to the atmosphere, particularly conversions from forest land
(Houghton et al. 1983). Moreover, conversion of forest to another land use (i.e., deforestation) is one of the largest
anthropogenic sources of emissions to the atmosphere globally (Schimel 1995), although this source may be
declining globally (Tubiello et al. 2015). IPCC (2006) recommends reporting changes in biomass, dead organic
matter, and soil organic carbon stocks due to land-use change. All soil organic carbon stock changes are estimated
and reported for land converted to settlements, but there is limited reporting of other pools in this Inventory. Loss
of aboveground and belowground biomass, dead wood and litter carbon are reported for forest land converted to
settlements and woodlands associated with grasslands converted to settlements, but not for other land-use
conversions to settlements.
There are discrepancies between the current land representation (see Section 6.1) and the area data that have
been used in the Inventory for land converted to settlements. Specifically, this Inventory includes all settlements in
the conterminous United States and Hawaii, but does not include settlements in Alaska. Areas of drained organic
soils in settlements on federal lands are also not included in this Inventory. These differences lead to discrepancies
between the managed area in land converted to settlements and the settlement area included in the inventory
analysis (Table 6-129). There is a planned improvement to include CO2 emissions from drainage of organic soils in
settlements of Alaska and federal lands as part of a future Inventory (see Planned Improvements section).
103 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 settlements in
the early part of the time series to the extent that some areas are converted to settlements from 1971 to 1978.
Land Use, Land-Use Change, and Forestry 6-187
-------
Forest land converted to settlements is the largest source of emissions from 1990 to 2022, accounting for
approximately 76 percent of the average total loss of carbon among all of the land-use conversions in Land
Converted to Settlements. Total losses of aboveground and belowground biomass, dead wood and litter carbon
losses in 2022 for all conversions are 35.6, 6.2, 6.7, and 9.2 MMT CO2 Eq., respectively (9.7,1.7,1.8, and 2.5 MMT
C). Mineral and organic soils also lost 9.2 and 1.3 MMT CO2 Eq. in 2022 (2.5 and 0.3 MMT C). The total net flux is
68.2 MMT CO2 Eq. in 2022 (18.6 MMT C), which is a 19 percent increase in CO2 emissions compared to the
emissions in the initial reporting year of 1990 (Table 6-136 and Table 6-137). The main driver of net emissions for
this source category is the conversion of forest land to settlements, with large losses of biomass, deadwood and
litter carbon.
Table 6-136: Net CO2 Flux from Soil, Dead Organic Matter and Biomass Carbon Stock Changes
for Land Converted to Settlements (MMT CO2 Eq.)
1990
2005
2018
2019
2020
2021
2022
Cropland Converted to Settlements
2.6
8.2
3.9
3.5
3.1
2.9
2.9
Mineral Soils
2.1
6-9 I
3.4
2.9
2.6
2.5
2.5
Organic Soils
0.5 "
1.2
0.6
0.5
0.4
0.4
0.4
Forest Land Converted to Settlements
49.3
53.9
58.4
58.6
58.6
58.6
58.6
Aboveground Live Biomass
30.0
32.4
35.0
35.2
35.2
35.2
35.2
Belowground Live Biomass
5.2
5.6 §
6.1
6.1
6.1
6.1
6.1
Dead Wood
5.5
6.0
6.5
6.6
6.6
6.6
6.6
Litter
7-6 i
8-2
9.0
9.0
9.0
9.0
9.0
Mineral Soils
1.0
1.5
1.5
1.5
1.5
1.5
1.5
Organic Soils
0.1
0.2 1
0.2
0.2
0.3
0.3
0.3
Grassland Converted to Settlements
5.6
15.8
9.8
8.8
7.9
7.4
7.5
Aboveground Live Biomass
0.4 1
0.4
0.5
0.5
0.5
0.5
0.5
Belowground Live Biomass
0.1
0.1
0.1
0.1
0.1
0.1
0.1
Dead Wood
0.1
0-11
0.2
0.2
0.2
0.2
0.2
Litter
0.2
0.2
0.2
0.2
0.2
0.2
0.2
Mineral Soils
4.3 1
13-7 1
8.2
7.3
6.5
6.0
6.0
Organic Soils
0.5
1.3
0.7
0.6
0.5
0.5
0.6
Other Lands Converted to
Settlements
(0.4)
(1.4)
(1.1)
(1.0)
(0.8)
(0.8)
(0.8)
Mineral Soils
(0.4)
(1.5)
(1.1)
(1.0)
(0.9)
(0.8)
(0.8)
Organic Soils
+ i
0-1 1
+
+
+
+
+
Wetlands Converted to Settlements
+
0.6
0.3
0.3
0.1
+
0.1
Mineral Soils
+ 1
O.i
+
+
+
+
+
Organic Soils
+
0.6
0.3
0.3
+
+
+
Total Aboveground Biomass Flux
30.4
32.8
35.5
35.6
35.6
35.6
35.6
Total Belowground Biomass Flux
5.2
5.7
6.1
6.2
6.2
6.2
6.2
Total Dead Wood Flux
5.6
6.1
6.7
6.7
6.7
6.7
6.7
Total Litter Flux
7.7
8.4
9.2
9.2
9.2
9.2
9.2
Total Mineral Soil Flux
7.0
20.7
12.0
10.8
9.8
9.2
9.2
Total Organic Soil Flux
1.2
3.4
1.8
1.6
1.3
1.2
1.3
Total Net Flux
57.2
77.1
71.4
70.2
68.8
68.2
68.2
+ Absolute value does not exceed 0.05 MMT C02 Eq.
Notes: Totals may not sum due to independent rounding. Parentheses indicate negative values or net sequestration.
Table 6-137: Net CO2 Flux from Soil, Dead Organic Matter and Biomass Carbon Stock Changes
for Land Converted to Settlements (MMT C)
1990
2005
2018
2019
2020
2021
2022
Cropland Converted to Settlements
0.7
2.2
1.1
1.0
0.8
0.8
0.8
Mineral Soils
0.6
1.9 !!!!!!!
0.9
0.8
0.7
0.7
0.7
Organic Soils
0.1
0.3
0.2
0.1
0.1
0.1
0.1
6-188 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
Forest Land Converted to Settlements
13.5
14.7
15.9
16.0
16.0
16.0
16.0
Aboveground Live Biomass
8.2
8.8
9.6
9.6
9.6
9.6
9.6
Belowground Live Biomass
1.4
1.5 1
1.7
1.7
1.7
1.7
1.7
Dead Wood
1.5
1.6
1.8
1.8
1.8
1.8
1.8
Litter
2.1 if
2.2 |
2.5
2.5
2.5
2.5
2.5
Mineral Soils
0.3
0.4
0.4
0.4
0.4
0.4
0.4
Organic Soils
+ ii
0.1 1
0.1
0.1
0.1
0.1
0.1
Grassland Converted to Settlements
1.5
4.3
2.7
2.4
2.2
2.0
2.0
Aboveground Live Biomass
o.i 5
0-11
0.1
0.1
0.1
0.1
0.1
Belowground Live Biomass
+
+
+
+
+
+
+
Dead Wood
+ »
+ i
+
+
+
+
+
Litter
+
+
0.1
0.1
0.1
0.1
0.1
Mineral Soils
1.2
3.7
2.2
2.0
1.8
1.6
1.6
Organic Soils
0.1
0.3
0.2
0.2
0.1
0.1
0.2
Other Lands Converted to
Settlements
(0.1)
(0.4)
(0.3)
(0.3)
(0.2)
(0.2)
(0.2)
Mineral Soils
(0.1)
(0.4)
(0.3)
(0.3)
(0.2)
(0.2)
(0.2)
Organic Soils
+1
+
+
+
+
+
Wetlands Converted to Settlements
+
0.2
0.1
0.1
+
+
+
Mineral Soils
+ gj
+::
+
+
+
+
+
Organic Soils
+
0.2
0.1
0.1
+
+
+
Total Aboveground Biomass Flux
8.3
8.9
9.7
9.7
9.7
9.7
9.7
Total Belowground Biomass Flux
1.4
1.5
1.7
1.7
1.7
1.7
1.7
Total Dead Wood Flux
1.5
1.7
1.8
1.8
1.8
1.8
1.8
Total Litter Flux
2.1
2.3
2.5
2.5
2.5
2.5
2.5
Total Mineral Soil Flux
1.9
5.6
3.3
2.9
2.7
2.5
2.5
Total Organic Soil Flux
0.3
0.9
0.5
0.4
0.4
0.3
0.3
Total Net Flux
15.6
21.0
19.5
19.1
18.8
18.6
18.6
+ Absolute value does not exceed 0.05 MMT C.
Notes: Totals may not sum due to independent rounding. Parentheses indicate negative values or net sequestration
Methodology and Time-Series Consistency
The following section includes a description of the methodology used to estimate carbon stock changes for land
converted to settlements, including (1) loss of aboveground and belowground biomass, dead wood and litter
carbon with conversion to settlements from forest lands and woodlands designated in the grassland, as well as (2)
the impact from all land-use conversions to settlements on soil organic carbon stocks in mineral and organic soils.
Biomass, Dead Wood, and Litter Carbon Stock Changes
A Tier 2 method is applied to estimate biomass, dead wood, and litter carbon stock changes for forest land
converted to settlements. Estimates are calculated in the same way as those in the forest land remaining forest
land category using data from the USDA Forest Service, Forest Inventory and Analysis (FIA) program (USDA Forest
Service 2023) however there is no country-specific data for settlements so the biomass, litter, and dead wood
carbon stocks were assumed to be zero. The difference between the stocks is reported as the stock change under
the assumption that the change occurred in the year of the conversion. Details for each of the carbon attributes
described below are available in Domke et al. (2022) and Westfall et al. (2023).
If FIA plots include data on individual trees, aboveground and belowground carbon density estimates are based on
Woodall et al. (2011) and Westfall et al. (2023). Aboveground and belowground biomass estimates also include live
understory which is a minor component of biomass defined as all biomass of undergrowth plants in a forest,
including woody shrubs and trees less than 2.54 cm dbh. For this Inventory, it was assumed that 10 percent of total
understory carbon mass is belowground (Smith et al. 2006). Estimates of carbon density are based on information
in Birdsey (1996) and biomass estimates from Jenkins et al. (2003).
Land Use, Land-Use Change, and Forestry 6-189
-------
This Inventory also includes estimates of change in dead organic matter for standing dead, deadwood and litter. If
FIA plots include data on standing dead trees, standing dead tree carbon density is estimated following the basic
method applied to live trees (Woodall et al. 2011 and Westfall et al. 2023) with additional modifications for
woodland species to account for decay and structural loss (Domke et al. 2011; Harmon et al. 2011). If FIA plots
include data on downed dead wood, downed dead wood carbon density is estimated based on measurements of a
subset of FIA plots for downed dead wood (Domke et al. 2013; Woodall and Monleon 2008). Downed dead wood is
defined as pieces of dead wood greater than 7.5 cm diameter, at transect intersection, that are not attached to
live or standing dead trees. This includes stumps and roots of harvested trees. To facilitate the downscaling of
downed dead wood carbon estimates from the state-wide population estimates to individual plots, downed dead
wood models specific to regions and forest types within each region are used. Litter carbon is the pool of organic
carbon (also known as duff, humus, and fine woody debris) above the mineral soil and includes woody fragments
with diameters of up to 7.5 cm. A subset of FIA plots is measured for litter carbon. If FIA plots include litter
material, a modeling approach using litter carbon measurements from FIA plots is used to estimate litter carbon
density (Domke et al. 2016).
In order to ensure time-series consistency, the same methods are applied from 1990 to 2020 so that changes
reflect anthropogenic activity and not methodological adjustments. See Annex 3.13 for more information about
reference carbon density estimates for forest land and the compilation system used to estimate carbon stock
changes from forest land.
Soil Carbon Stock Changes
Soil organic carbon stock changes are estimated for land converted to settlements according to land use histories
recorded in the 2017 USDA NRI survey for non-federal lands (USDA-NRCS 2020) and extended through 2020 using
the USDA-NASS Crop Data Layer Product (USDA-NASS 2021; Johnson and Mueller 2010) and National Land Cover
Dataset (NLCD) (Yang et al. 2018; Fry et al. 2011; Homer et al. 2007, 2015)). For federal lands, the land use history
is derived from land cover changes in the NLCD. The areas have been modified through a process in which the
Forest Inventory and Analysis (FIA) survey data are harmonized with the NRI data (Nelson et al. 2020). This process
ensures that the land use areas are consistent across all land use categories (see Section 6.1 for more information).
Land use and some management information were originally collected for each NRI survey location on a 5-year
cycle beginning in 1982. In 1998, the NRI program began collecting annual data, and the annual data have been
incorporated from the NRI into the inventory analysis through 2017 (USDA-NRCS 2020).
NRI survey locations are classified as land converted to settlements in a given year between 1990 and 2020 if the
land use is settlements but had been classified as another use during the previous 20 years. 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 1998. This may have led to an underestimation of land converted to settlements in the
early part of the time series to the extent that some areas are converted to settlement between 1971 and 1978.
Soil Carbon Stock Changes for Mineral Soils
An IPCC Tier 2 method (Ogle et al. 2003) is applied to estimate carbon stock changes for mineral soils on land
converted to settlements from 1990 to 2020. Data on climate, soil types, land use, and land management activity
are used to classify land area and apply appropriate stock change factors (Ogle et al. 2003, 2006). Reference
carbon stocks are estimated using the National Soil Survey Characterization Database (USDA-NRCS 1997) with
cultivated cropland as the reference condition, rather than native vegetation as used in IPCC (2006). Soil
measurements under agricultural management are much more common and easily identified in the National Soil
Survey Characterization Database (USDA-NRCS 1997) than are soils under a native condition, and therefore
cultivated cropland provide a more robust sample for estimating the reference condition. Country-specific carbon
stock change factors are derived from published literature to determine the impact of management practices on
soil organic carbon storage (Ogle et al. 2003; Ogle et al. 2006). However, there are insufficient data to estimate a
set of land use, management, and input factors for settlements. Moreover, the 2017 NRI survey data (USDA-NRCS
2020) do not provide the information needed to assign different land use subcategories to settlements, such as
6-190 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
turf grass and impervious surfaces, which is needed to apply the Tier 1 factors from the IPCC guidelines (2006).
Therefore, the United States has adopted a land use factor of 0.7 to represent a net loss of soil organic carbon with
conversion to settlements under the assumption that there are additional soil organic carbon losses with land
clearing, excavation and other activities associated with development. More specific factor values can be derived
in future Inventories as data become available. See Annex 3.12 for additional discussion of the Tier 2 methodology
for mineral soils.
In order to ensure time-series consistency, the same methods are applied from 1990 to 2020 so that changes
reflect anthropogenic activity and not methodological adjustments. Soil organic carbon stock changes from 2021
to 2022 are estimated using a linear extrapolation method described in Box 6-4 of the Methodology section in
Cropland Remaining Cropland. The extrapolation is based on a linear regression model with moving-average
(ARMA) errors using the 1990 to 2020 emissions data, which is a standard data splicing method for imputing
missing emissions data in a time series (IPCC 2006). The Tier 2 method described previously will be applied to
recalculate the 2021 to 2022 emissions in a future Inventory.
Soil Carbon Stock Changes for Organic Soils
Annual carbon emissions from drained organic soils in land converted to settlements are estimated using the Tier 2
method provided in IPCC (2006). The Tier 2 method assumes that organic soils are losing carbon at a rate similar to
croplands, and therefore uses the country-specific values for cropland (Ogle et al. 2003). To estimate CO2
emissions from 1990 to 2020, the area of organic soils in land converted to settlements is multiplied by the Tier 2
emission factor, which is 11.2 MT C per ha in cool temperate regions, 14.0 MT C per ha in warm temperate regions
and 14.3 MT C per ha in subtropical regions (see Annex 3.12 for more information).
In order to ensure time-series consistency, the same methods are applied from 1990 to 2020, and a linear
extrapolation method is used to approximate emissions for the remainder of the 2021 to 2022 time series (see Box
6-4 of the Methodology section in Cropland Remaining Cropland). The extrapolation is based on a linear regression
model with moving-average (ARMA) errors using the 1990 to 2020 emissions data, and is a standard data splicing
method for imputing missing emissions data in a time series (IPCC 2006). Estimates will be recalculated in future
Inventories when new activity data are incorporated into the analysis.
Uncertainty
The uncertainty analysis for carbon losses with forest land converted to settlements is conducted in the same way
as the uncertainty assessment for forest ecosystem carbon flux in the forest land remaining forest land category.
Sample and model-based error are combined using 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. For additional details, see the Uncertainty Analysis in Annex 3.13.
Sources of uncertainty for mineral soil organic carbon stock changes and annual carbon emission estimates from
drained organic soils include emission factors and variance associated with the NRI sample. The total uncertainty
was quantified with two variance components (Ogle et al. 2010) that are combined using 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. For the first variance component, a Monte Carlo analysis was used to
propagate uncertainties in the Tier 2 methods for the land use area and the country-specific factors or mineral and
organic soils. The second variance component is associated with scaling of the data from the NRI survey to the
entire area of land converted to settlements, and is computed using a standard variance estimator for a two-stage
sample design (Sarndal et al. 1992).
Uncertainty estimates are presented in Table 6-138 for each sub-source (i.e., biomass carbon, dead wood, litter,
soil organic carbon in mineral soils and organic soils) and the method applied in the inventory analysis (i.e., Tier 2
and Tier 3). Uncertainty estimates are combined from the forest land converted to settlements and other land use
conversions to settlements using the simple error propagation methods provided by the IPCC (2006). There are
Land Use, Land-Use Change, and Forestry 6-191
-------
also additional uncertainties propagated through the analysis associated with the data splicing methods applied to
estimate soil organic carbon stock changes from 2021 to 2022. The combined uncertainty for total carbon stock
changes in land converted to settlements ranges from 36 percent below to 36 percent above the 2022 stock
change estimate of 68.2 MMT CO2 Eq.
Table 6-138: Approach 2 Quantitative Uncertainty Estimates for Soil, Dead Organic Matter
and Biomass Carbon Stock Changes occurring within Land Converted to Settlements (MMT
CO2 Eq. and Percent)
Source
2022 Flux Estimate
Uncertainty Range Relative to Flux Estimate-'
(MMT CO . Eq.)
(MMT CO.
Eq.)
(%)
Lower
Upper
Lower
Upper
Bound
Bound
Bound
Bound
Cropland Converted to Settlements
2.9
1.3
4.5
-56%
56%
Mineral Soil C Stocks
2.5
0.9
4.1
-64%
63%
Federal Mineral Soil C Stocks
0.0
(0.0)
0.0
-146%
146%
Organic Soil C Stocks
0.4
0.1
0.7
-80%
80%
Forest Land Converted to Settlements
58.6
35.3
81.9
-40%
40%
Aboveground Biomass C Stocks
35.2
13.3
57.1
-62%
62%
Belowground Biomass C Stocks
6.1
2.3
9.9
-62%
62%
Dead Wood
6.6
2.5
10.6
-62%
62%
Litter
9.0
3.4
14.7
-62%
62%
Mineral Soil C Stocks
1.5
1.2
1.7
-19%
19%
Federal Mineral Soil C Stocks
0.0
0.0
0.0
0%
0%
Organic Soil C Stocks
0.3
0.1
0.4
-70%
70%
Grassland Converted to Settlements
7.4
5.4
9.5
-27%
27%
Aboveground Biomass C Stocks
0.5
0.2
0.8
-62%
60%
Belowground Biomass C Stocks
0.1
0.0
0.1
-45%
66%
Dead Wood
0.2
0.1
0.3
-51%
70%
Litter
0.2
0.1
0.3
-61%
57%
Mineral Soil C Stocks
6.0
4.1
7.9
-32%
32%
Federal Mineral Soil C Stocks
0.0
(0.0)
0.0
-194%
194%
Organic Soil C Stocks
0.5
(0.0)
1.1
-108%
108%
Other Lands Converted to Settlements
-0.8
(1.1)
(0.4)
-44%
44%
Mineral Soil C Stocks
-0.8
(1.1)
(0.5)
-38%
38%
Federal Mineral Soil C Stocks
0.0
(0.2)
0.1
-738%
738%
Organic Soil C Stocks
0.0
(0.1)
0.1
-511%
511%
Wetlands Converted to Settlements
0.1
(0.3)
0.3
-557%
533%
Mineral Soil C Stocks
0.0
0.0
0.1
-69%
69%
Federal Mineral Soil C Stocks
0.0
0.0
0.0
0%
0%
Organic Soil C Stocks
0.0
(0.3)
0.3
-1,269%
1,214%
Total: Land Converted to Settlements
68.2
43.5
92.9
-36%
36%
Aboveground Biomass C Stocks
35.6
13.3
57.1
-62%
62%
Belowground Biomass C Stocks
6.2
2.3
9.9
-62%
62%
Dead Wood
6.7
2.5
10.6
-62%
62%
Litter
9.2
3.4
14.7
-62%
62%
Mineral Soil C Stocks
9.2
6.6
11.7
-28%
28%
Organic Soil C Stocks
1.2
(6.6)
9.1
-625%
625%
+ Does not exceed 0.05 MMT C02 Eq.
a Range of C stock change estimates is a 95 percent confidence interval.
Note: Totals may not sum due to independent rounding. Parentheses indicate negative values or net sequestration.
6-192 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
QA/QC and Verification
Quality control measures included checking input data, model scripts, and results to ensure data are properly
handled throughout the inventory process. Inventory reporting forms and text are reviewed and revised as needed
to correct transcription errors. No errors were found in this Inventory.
Recalculations are associated with new FIA data from 1990 to 2022 on biomass, dead wood and litter carbon
stocks in forest land converted to settlements and woodland conversion associated with grassland converted to
settlements. Additional recalculations are associated with incorporating new USDA-NRCS NRI data through 2017
and extending the time series using CDL and NLCD for grassland converted to settlements, cropland converted to
settlements, other land converted to settlements and wetlands converted to settlements. As a result, land
converted to settlements has an estimated smaller carbon loss of 7.7 MMT CO2 Eq. on average over the time
series. This represents a 19 percent decrease in carbon stock changes for land converted to settlements compared
to the previous Inventory.
A key improvement is to develop an inventory of mineral soil organic carbon stock changes in Alaska and losses of
carbon from drained organic soils in federal lands. These improvements will resolve most of the differences
between the managed land base for land converted to settlements and amount of area currently included in the
Inventory for land converted to settlements (see Table 6-139).
There are plans to improve classification of trees in settlements and to include transfer of biomass from forest land
to those areas in this category. There are also plans to extend the inventory to included carbon losses associated
with drained organic soils in settlements occurring on federal lands.
These improvements will be made pending prioritization of resources to expand the inventory for this source
category.
Table 6-139: Area of Managed Land in Land Converted to Settlements that is not included in
the current Inventory (Thousand Hectares)
Recalculations Discussion
Planned Improvements
Area (Thousand Hectares)
Year
LCS Managed Land
Area (Section 6.1)
LCS Area Included
in Inventory
Included in Inventory
LCS Area Not
1990
1991
1992
1993
1994
1995
1996
1997
1998
1999
2000
2001
2002
2003
2004
2005
2,865
3,213
3,573
4,138
4,703
5,262
5,833
6,409
6,929
7,446
7,952
8,362
8,696
8,705
8,710
8,727
2,865
3,213
3,573
4,138
4,702
5,261
5,832
6,408
6,928
7,446
7,952
8,361
8,695
8,704
8,708
8,724
2
2
Land Use, Land-Use Change, and Forestry 6-193
-------
2006
8,691
8,688
3
2007
8,672
8,668
3
2008
8,501
8,497
4
2009
8,309
8,305
5
2010
8,130
8,124
5
2011
7,930
7,925
6
2012
7,717
7,711
6
2013
7,325
7,318
6
2014
6,942
6,935
7
2015
6,530
6,523
7
2016
6,112
6,105
7
2017
5,715
5,708
7
2018
5,201
5,194
7
2019
4,696
4,689
7
2020
4,175
4,168
7
2021
3,771
*
*
2022
3,437
*
*
NRI data have not been incorporated into the Inventory after 2020, designated with asterisks (*).
6.12 Other Land Remaining Other Land (CRT
Category 4F1)
Land use is constantly occurring, and areas under a number of differing land-use types remain in their respective
land-use type each year, just as other land can remain as other land. While the magnitude of other land remaining
other land is known (see Table 6-4), research is ongoing to track carbon pools in this land use. Until such time that
reliable and comprehensive estimates of carbon for other land remaining other land can be produced, it is not
possible to estimate CO2, Cm or N2O fluxes on other land remaining other land at this time.
6.13 Land Converted to Other Land (CRT
Category 4F2)
Land-use change is constantly occurring, and areas under a number of differing land-use types are converted to
other land each year, just as other land is converted to other uses. While the magnitude of these area changes is
known (see Table 6-4), research is ongoing to track carbon across other land remaining other land and land
converted to other land. Until such time that reliable and comprehensive estimates of carbon across these land-
use and land-use change categories can be produced, it is not possible to separate CO2, Cm or N2O fluxes on land
converted to other land from fluxes on other land remaining other land at this time.
6-194 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
7. Waste
Waste management and treatment activities are sources of greenhouse gas emissions (see Figure 7-1 and Figure
7-2). Landfills were the third largest source of anthropogenic methane (Cm) in the United States in 2022,
accounting for approximately 17.1 percent of total U.S. Cm emissions. Additionally, wastewater treatment and
discharge, composting of organic waste, and anaerobic digestion at biogas facilities accounted for approximately
3.0 percent, 0.4 percent, and less than 0.1 percent of U.S. CFU emissions, respectively. Nitrous oxide (N2O)
emissions resulting from the discharge of wastewater treatment effluents into aquatic environments were
estimated, along with the wastewater treatment process itself and composting. Together, these waste activities
account for 6.1 percent of total U.S. N2O emissions. Nitrogen oxides (NOx), carbon monoxide (CO), and non-CFU
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 sector is presented in Table 7-1 and Table
7-2. Overall, in 2022, waste activities generated emissions of 166.9 MMT CO2 Eq., or 2.6 percent of total U.S.
greenhouse gas emissions.
Emissions from landfills contributed 71.8 percent of Waste sector emissions in 2022 (see Figure 7-1) and are
primarily composed of CH4 emissions from municipal solid waste landfills. Landfill emissions decreased by 2.3 MMT
CO2 Eq. (1.9 percent) since 2021. Emissions from wastewater treatment were the second largest source of waste-
related emissions in 2022, accounting for 25.6 percent of sector emissions. The remaining two sources of
emissions, composting and anaerobic digestion at biogas facilities, account for 2.6 percent and less than 0.1
percent of Waste sector emissions in 2022, respectively.
Figure 7-1: 2022 Waste Sector Greenhouse Gas Sources
Wastewater Treatment
Anaerobic Digestion at
Biogas Facilities
Composting
Landfills
0 10 20 30 40 50 60 70 80 90 100 110 120 130
MMT CO2 Eq.
Waste 7-1
-------
Figure 7-2: Trends in Waste Sector Greenhouse Gas Sources
250
200
CO
m co
rsi rM
o
fN
Anaerobic Digestion at Biogas Facilities
I Composting
i Wastewater Treatment
Landfills
fN
CO
i-n
r-v
O
u
150
100
50
o[ BIBBBIIBIBIIIIBIBBBIIBBBIIIBBIB
OT-Hf\jm'rLov£)rN.oocnoT-HfNm^-Lri^orvoo<^OT-((NmTru->^rvcochOT-HrM
cncr.cric^cncncncr.cr.a^oooooooooO'-HT-HT-i^HT-H^HT-H^-iT-HHrvjrsirsj
crio^cncncrtchcricncricrtooooooooooooooooooooooo
HHHHHHHHHH(MrNfNfNfNfNfNfNfNJOJfN(N(NfNfNfN(NfNfM(NfNfNfN
Table 7-1: Emissions from Waste (MMT CO2 Eq.)
Gas/Source
1990
2005
2018
2019
2020
2021
2022
ch4
220.9
172.4
150.2
152.4
147.6
145.3
143.2
Landfills
197.8
147.7
126.3
128.7
124.1
122.0
119.8
Wastewater T reatment
22.7
22.7
21.4
21.1
21.0
20.7
20.8
Composting
0.4
2.1
2.5
2.5
2.6
2.6
2.6
Anaerobic Digestion at Biogas
Facilities
+
+
+
+
+
+
+
n2o
15.1
19.5
23.0
23.4
24.1
23.9
23.7
Wastewater T reatment
14.8
18.1
21.2
21.6
22.3
22.1
21.9
Composting
0.3
1.5
1.8
1.8
1.8
1.8
1.8
Total
235.9
192.0
173.2
175.8
171.7
169.2
166.9
+ Does not exceed 0.05 MMT C02 Eq.
Note: Totals may not sum due to independent rounding.
Table 7-2: Emissions from Waste (kt)
Gas/Source
1990
2005
2018
2019
2020
2021
2022
ch4
7,889
6,159
5,365
5,441
5,271
5,189
5,113
Landfills
7,063
5,275
4,512
4,595
4,431
4,359
4,277
Wastewater T reatment
811
809
763
755
748
738
743
Composting
15
75
90
91
92
92
92
Anaerobic Digestion at Biogas
Facilities
+
+
+
1
+
+
+
n2o
57
74
87
88
91
90
89
Wastewater T reatment
56
68
80
81
84
83
83
Composting
1
6
7
7
7
7
7
+ Does not exceed 0.5 kt.
Note: Totals by gas may not sum due to independent rounding.
7-2 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
Carbon dioxide (CO2), CH4, and N2O emissions from the incineration of waste are accounted for in the Energy
sector rather than in the Waste sector because almost all incineration of municipal solid waste (MSW) in the
United States occurs at waste-to-energy facilities where useful energy is recovered. Similarly, the Energy sector
also includes an estimate of emissions from burning waste tires and hazardous industrial waste, because virtually
all of the combustion occurs in industrial and utility boilers that recover energy. The incineration of waste in the
United States in 2022 resulted in 12.7 MMT CO2 Eq. emissions, more than half of which is attributable to the
combustion of plastics. For more details on emissions from the incineration of waste, see Section 7.5. Greenhouse
Gas Precursor Emissions from the Waste sector are presented in Section 7.6.
Each year, some emission and sink estimates in the Inventory are recalculated and revised with improved methods
and/or data. In general, recalculations are made to the U.S. greenhouse gas emission estimates either to
incorporate new methodologies or, most commonly, to update recent historical data. These improvements are
implemented consistently across the previous Inventory's time series (i.e., 1990 to 2021) to ensure that the trend
is accurate. For the current Inventory, minor improvements were implemented beyond routine activity data
updates, including changes to MSW and industrial waste landfill activity data, updates to production activity
affecting wastewater influent, and methodological changes for CH4 emissions from anaerobic digesters processing
food waste. In total, the methodological and historic data improvements made to the Waste sector in this
Inventory resulted in an average decrease in greenhouse gas emissions across the time series by 0.06 MMT CO2 Eq.
(0.03 percent). For more information on specific methodological updates, please see the Recalculations Discussion
section for each category in this chapter.
Due to lack of data availability, EPA is not able to estimate emissions associated with sludge generated from the
treatment of industrial wastewater. Emissions reported in the Waste chapter for landfills, wastewater treatment,
and anaerobic digestion at biogas facilities include those from all 50 states, including Hawaii and Alaska, the
District of Columbia, and U.S. Territories. Emissions from landfills include modern, managed sites in most U.S.
Territories except for outlying Pacific Islands. Emissions from domestic wastewater treatment include most U.S.
Territories except for outlying Pacific Islands. Those emissions are likely insignificant as those outlying Pacific
Islands (e.g., Baker Island) have no permanent population. No industrial wastewater treatment emissions are
estimated for U.S. Territories, due to lack of data availability. However, industrial wastewater treatment emissions
are not expected for outlying Pacific Islands and assumed to be small for other U.S. Territories. Emissions for
composting include Puerto Rico and all states except Alaska. Some composting operations in Alaska are known, but
these consist of aerated composting facilities. Composting emissions are not included from the remaining U.S.
Territories, and these are assumed to be small. Similarly, EPA is not aware of any anaerobic digestion at biogas
facilities in U.S. Territories but will review this on an ongoing basis to include these emissions if they are occurring.
See Annex 5 for more information on EPA's assessment of the sources not included in this Inventory.
Box 7-1: Methodological Approach for Estimating and Reporting U.S. Emissions and
Removals, including Relationship to Greenhouse Gas Reporting Data
Consistent with Article 13.7(a) of the Paris Agreement and Article 4.1(a) of the UNFCCC as well as relevant
decisions under those agreements, the emissions and removals 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) in the 2006IPCC Guidelines for National Greenhouse Gas
Inventories (2006 IPCC Guidelines) and its supplements and refinements. Additionally, the calculated emissions
and removals in a given year for the United States are presented in a common format in line with the reporting
guidelines for the reporting of inventories under the Paris Agreement and the UNFCCC. The Parties' use of
consistent methods to calculate emissions and removals for their inventories helps to ensure that these reports
are comparable. The presentation of emissions and sinks provided in the Waste chapter do not preclude
alternative examinations, but rather, this chapter presents emissions and removals in a common format
consistent with how Parties are to report inventories under the Paris Agreement and the UNFCCC. The report
itself, and this chapter, follows this common format, and provides an explanation of the application of methods
used to calculate emissions and removals from waste management and treatment activities.
Waste 7-3
-------
EPA also collects greenhouse gas emissions data from individual facilities and suppliers of certain fossil fuels and
industrial gases through its Greenhouse Gas Reporting Program (GHGRP). The GHGRP applies to direct
greenhouse gas emitters, fossil fuel suppliers, industrial greenhouse gas suppliers, and facilities that inject CO2
underground for sequestration or other reasons and requires reporting by sources or suppliers in 41 industrial
categories. Annual 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. See
Annex 9 "Use of EPA Greenhouse Gas Reporting Program in Inventory" for more information.
Waste Data from EPA's Greenhouse Gas Reporting Program
EPA uses annual GHGRP facility-level data in the Landfills category to compile the national estimate of emissions
from Municipal Solid Waste (MSW) landfills (see Section 7.1 of this chapter for more information). EPA uses
directly reported GHGRP data for net CH4 emissions from MSW landfills for the years 2010 to 2022 of the
Inventory. MSW landfills subject to the GHGRP began collecting data in 2010. These data are also used to
recalculate emissions from MSW landfills for the years 2005 to 2009 to ensure time-series consistency.
7.1 Landfills (CRT Source Category 5A1)
In In the United States, solid waste is managed by landfilling, recovery through recycling or composting, and
combustion through waste-to-energy facilities. Disposing of solid waste in modern, managed landfills is the most
used waste management technique in the United States. More information on how solid waste data are collected
and managed in the United States is provided in Box 7-3. The municipal solid waste (MSW) and industrial waste
landfills referred to in this section are all modern landfills that must comply with a variety of regulations as
discussed in Box 7-2. Disposing of waste in illegal dumping sites is not considered to have occurred in years later
than 1980 and these sites are not considered to contribute to net emissions in this section for the timeframe of
1990 to the current Inventory year. MSW landfills, or sanitary landfills, are sites where MSW is managed to prevent
or minimize health, safety, and environmental impacts. Waste is deposited in different cells and covered daily with
soil; many have environmental monitoring systems to track performance, collect leachate, and collect landfill gas.
Industrial waste landfills are constructed in a similar way as MSW landfills, but are used to dispose of industrial
solid waste, such as RCRA Subtitle D wastes (e.g., non-hazardous industrial solid waste defined in Title 40 of the
Code of Federal Regulations [CFR] in section 257.2), commercial solid wastes, or conditionally exempt small-
quantity generator wastes (EPA 2016a).
After being placed in a landfill, organic waste (such as paper, food scraps, and yard trimmings) is initially
decomposed by aerobic bacteria. After the oxygen has been depleted, the remaining waste is available for
consumption by anaerobic bacteria, which break down organic matter into substances such as cellulose, amino
acids, and sugars. These substances are further broken down through fermentation into gases and short-chain
organic compounds that form the substrates for the growth of methanogenic bacteria. These CH4 producing
anaerobic bacteria convert the fermentation products into stabilized organic materials and biogas consisting of
approximately 50 percent biogenic carbon dioxide (CO2) and 50 percent CH4, by volume. Landfill biogas also
contains trace amounts of non-methane organic compounds (NMOC) and volatile organic compounds (VOC) that
either result from decomposition byproducts or volatilization of biodegradable wastes (EPA 2008).
Box 7-2: Description of a Modern, Managed Landfill in the United States
Modern, managed landfills are well-engineered facilities that are located, designed, operated, and monitored to
ensure compliance with federal, state, and tribal regulations. A modern, managed landfill is EPA's interpretation
of the IPCC's terminology of a managed solid waste disposal site. Municipal solid waste (MSW) landfills must be
designed to protect the environment from contaminants which may be present in the solid waste stream.
7-4 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
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.
Specific federal regulations that affected MSW landfills must comply with include the 40 CFR Part 258 (Subtitle
D of RCRA), or equivalent state regulations and the NSPS 40 CFR Part 60 Subparts WWW and XXX.1 Additionally,
state and tribal requirements may exist.
Methane and CO2 are the primary constituents of landfill gas generation and emissions. Net carbon dioxide flux
from carbon stock changes of materials of biogenic origin in landfills are estimated and reported under the Land
Use, Land-Use Change, and Forestry (LULUCF) sector (see Chapter 6 of this Inventory). Nitrous oxide (N2O)
emissions from the disposal and application of sewage sludge on landfills are also not explicitly modeled as part of
greenhouse gas emissions from landfills. Nitrous oxide emissions from sewage sludge applied to landfills as a daily
cover or for disposal are expected to be relatively small because the microbial environment in an anaerobic landfill
is not very conducive to the nitrification and denitrification processes that result in N2O emissions. Furthermore,
the 2006IPCC Guidelines did not include a methodology for estimating N2O emissions from solid waste disposal
sites "because they are not significant." Therefore, only CH4 generation and emissions are estimated for landfills
under the Waste sector.
Methane generation and emissions from landfills are a function of several factors, including: (1) the total amount
and composition of waste-in-place, which is the total waste landfilled annually over the operational lifetime of a
landfill; (2) the characteristics of the landfill receiving waste (e.g., size, climate, cover material); (3) the amount of
Cm that is recovered and either flared or used for energy purposes; and (4) the amount of CH4 oxidized as the
landfill gas-that is not collected by a gas collection system - passes through the cover material into the
atmosphere. Each landfill has unique characteristics, but all managed landfills employ similar operating practices,
including the application of a daily and intermediate cover material over the waste being disposed of in the landfill
to prevent odor and reduce risks to public health. Based on recent literature, the specific type of cover material
used can affect the rate of oxidation of landfill gas (RTI 2011). The most used cover materials are soil, clay, and
sand. Some states also permit the use of green waste, tarps, waste derived materials, sewage sludge or biosolids,
and contaminated soil as a daily cover. Methane production typically begins within the first year after the waste is
1 For more information regarding federal MSW landfill regulations, see
http://www.epa.gov/osw/nonhaz/municipal/landfill/msw regs.htm.
Waste 7-5
-------
disposed of in a landfill and will continue for 10 to 50 or more years as the degradable waste decomposes over
time.
In 2022, landfill Cm emissions were approximately 119.8 MMT CO2 Eq. (4,277 kt), representing the third largest
source of CH4 emissions in the United States, behind enteric fermentation and natural gas systems. Emissions from
MSW landfills accounted for approximately 84 percent of total landfill emissions (100.9 MMT CO2 Eq.), while
industrial waste landfills accounted for the remainder (18.9 MMT CO2 Eq.). Nationally, there are significantly less
industrial waste landfills compared to MSW landfills, which contributes to the lower national estimate of Cm
emissions for industrial waste landfills. Additionally, the average organic content of waste streams disposed in
industrial waste landfills is lower than MSW landfills. Estimates of operational MSW landfills in the United States
have ranged from 1,700 to 2,000 facilities (EPA 2023a; EPA 2023b; EPA 2020c; Waste Business Journal [WBJ] 2016;
WBJ 2010). The Environment Research & Education Foundation (EREF) conducted a nationwide analysis of MSW
management and counted 1,540 operational MSW landfills in 2013 (EREF 2016). Conversely, there are
approximately 3,200 MSW landfills in the United States that have been closed since 1980 (for which a closure data
is known, (EPA 2023b; WBJ 2010). While the number of active MSW landfills has decreased significantly over the
past 20 years, from approximately 6,326 in 1990 to as few as 1,540 in 2013, the average landfill size has increased
(EPA 2023a; EREF 2016; BioCycle 2010). Larger landfills may have deeper cells where a greater amount of area will
be anaerobic (more Cm is generated in anaerobic versus aerobic areas) and larger landfills tend to generate more
Cm compared to a smaller landfill (assuming the same waste composition and age of waste). Regarding industrial
waste landfills, the WBJ database includes approximately 1,100 landfills accepting industrial and/or construction
and demolition debris for 2021 (WBJ 2021). Only 169 facilities with industrial waste landfills met the reporting
threshold under Subpart TT (Industrial Waste Landfills) in the first year (2011) of EPA's Greenhouse Gas Reporting
Program for this subpart (GHGRP codified in 40 CFR Part 98), indicating that there may be several hundred
industrial waste landfills that are not required to report under EPA's GHGRP. Less industrial waste landfills meet
the GHGRP eligibility threshold because they typically accept waste streams with low to no organic content, which
will not decompose and generate CH4 when disposed.
The annual amount of MSW generated and subsequently disposed in MSW landfills varies annually and depends
on several factors (e.g., the economy, consumer patterns, recycling and composting programs, inclusion in a
garbage collection service). The estimated annual quantity of waste placed in MSW landfills increased 10 percent
from approximately 205 MMT in 1990 to 226 MMT in 2000, then decreased by 11 percent to 202 MMT in 2010,
and then increased by 7 percent to approximately 217 MMT in 2022 (see Annex 3.14, Table A-233). Emissions
decreased between 1990 to 2022 largely because of increased use of landfill gas collection and control systems,
closure of older landfills, better management practices, and increased diversion of organics through state and local
policy and regulations. The total amount of MSW generated is expected to increase as the U.S. population
continues to grow. The impacts of the coronavirus (COVID-19) pandemic with respect to landfilled waste cannot be
quantified as data sources such as the EPA's Advancing Sustainable Materials Management: Facts and Figures
report have not been published for 2019 through 2022. The quantities of waste landfilled for 2019 to 2022
(presented in Annex 3.14) are extrapolated based on population growth and the last national assessment of MSW
landfilled from 2013 (EREF 2016). Net CH4 emissions from MSW landfills have decreased since 1990 (see Table 7-3
and Table 7-4).
The estimated quantity of waste placed in industrial waste landfills (from the pulp and paper, and food processing
sectors) has remained relatively steady since 1990, ranging from 9.7 MMT in 1990 to 11.0 MMT in 2022 (see Annex
3.14, Table A-219). CH4 emissions from industrial waste landfills have also remained at similar levels recently,
ranging from 16.1 MMT CO2 Eq. in 2005 to 18.9 MMT CO2 Eq. in 2022 when accounting for both CH4 generation
and oxidation. The EPA has focused the industrial waste landfills source category on industrial sectors known to
generate and dispose of by-products that are organic and contribute to CH4 generation, which are the pulp and
paper and food processing sectors. Construction and demolition (C&D) landfills, another type of industrial waste
landfill, may accept waste that could degrade (e.g., treated wood), but these waste streams are unlikely to
generate significant amounts of CH4 and are therefore not as relevant to the purpose of national greenhouse gas
emissions estimate. There is also a general lack of data on annual quantities of waste disposed in industrial waste
7-6 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
landfills, and the GHGRP Subpart TT (Industrial Waste Landfills) dataset has confirmed C&D landfills, for example,
are insignificant Cm generators.
EPA's Landfill Methane Outreach Program (LMOP) collects information on landfill gas energy projects currently
operational or under construction throughout the United States. LMOP's Landfill and Landfill Gas Energy Database
contains certain information on the gas collection and control systems in place at landfills provided by
organizations that are a part of the program, which can include the amount of landfill gas collected and flared. In
2022, LMOP identified 9 new landfill gas-to-energy (LFGE) projects (EPA 2023b) that began operation.
Landfill gas collection and control is not accounted for at industrial waste landfills in this chapter (see the
Methodology discussion for more information).
Table 7-3: CH4 Emissions from Landfills (MMT CO2 Eq.)
Activity
1990
2005
2018
2019
2020
2021
2022
MSW CH4 Generation3
230.0
303.7
332.0
340.9
340.9
335.9
331.4
Industrial CH4 Generation
13.6
1
17.9
i
20.8
20.9
21.0
21.0
21.0
MSW CH4 Recovered3
(23.8)
(148.4)
(195.2)
(201.4)
(206.3)
(203.3)
(199.8)
MSW CH4 Oxidized3
(20.6)
1
1
(23.6)
i
(29.2)
(29.6)
(29.4)
(29.5)
(30.7)
Industrial CH4 Oxidized
(1.4)
(1.8)
1
(2.1)
(2.1)
(2.1)
(2.1)
(2.1)
MSW net CH4 Emissions
185.5
1
131.6
107.7
109.9
105.2
103.1
100.9
Industrial CH4 Emissions'5
12.2
16.1
18.7
18.8
18.9
18.9
18.9
Total
197.8
147.7
126.3
128.7
124.1
122.0
119.8
a For years 1990 to 2004, the Inventory methodology for MSW landfills uses the first order decay methodology. A
methodological change occurs in year 2005. For years 2005 to 2022, directly reported net CH4 emissions from the GHGRP
data plus a scale-up factor are used to account for emissions from landfill facilities that are not subject to the GHGRP. More
details on the scale-up factor and how it was developed can be found in Annex 3.14. These data incorporate CH4 recovered
and oxidized for MSW landfills. As such, CH4 generation, CH4 oxidation, and CH4 recovery are not calculated separately and
totaled to net CH4 emissions. See the Methodology and Time-Series Consistency section of this chapter for more
information.
b Methane recovery is not calculated for industrial landfills because this is not a common practice in the United States. Only
1 landfill of 167 that report to Subpart TT (Industrial Waste Landfills) of the GHGRP had an active gas collection and control
system during the year 2021 (EPA 2023a).
Notes: Parentheses indicate negative values. Totals may not sum due to independent rounding.
Table 7-4: CH4 Emissions from Landfills (kt CH4)
Activity
1990
2005
2018
2019
2020
2021
2022
MSW CH4 Generation3
8,214
10,845
11,857
12,175
12,174
11,997
11,834
Industrial CH4 Generation
484
cn
u>
00
iiiiiii
741
745
748
750
750
MSW CH4 Recovered3
(851)
(5,301)
(6,970)
(7,193)
(7,367)
(7,262)
(7,135)
MSW CH4 Oxidized3
(736) 1
(843) S
(1,041)
(1,058)
(1,050)
(1,052)
(1,097)
Industrial CH4 Oxidized
(48) "
(64) "
(74)
(75)
(75)
(75)
(75)
MSW net CH4 Emissions
6'627 '
4,701 '
3,845
3,924
3,757
3,683
3,602
Industrial net CH4 Emissions'5
436
574
667
671
674
675
675
Total
7,063
5,275
4,512
4,595
4,431
4,359
4,277
a For years 1990 to 2004, the Inventory methodology for MSW landfills uses the first order decay methodology. A
methodological change occurs in year 2005. For years 2005 to 2022, directly reported net CH4 emissions from the GHGRP
data plus a scale-up factor are used to account for emissions from landfill facilities that are not subject to the GHGRP. More
details on the scale-up factor and how it was developed can be found in Annex 3.14. These data incorporate CH4 recovered
and oxidized for MSW landfills. As such, CH4 generation, CH4 oxidation, and CH4 recovery are not calculated separately and
totaled to net CH4 emissions. See the Methodology and Time-Series Consistency section of this chapter for more
information.
b Methane recovery is not calculated for industrial landfills because this is not a common practice in the United States. Only
1 landfill of 167 that report to Subpart TT (Industrial Waste Landfills) of the GHGRP had an active gas collection and control
system during the year 2021 (EPA 2023a).
Notes: Parentheses indicate negative values. Totals may not sum due to independent rounding.
Waste 7-7
-------
Methodology and Time-Series Consistency
Methodology Applied for MSW Landfills
A combination of IPCC Tier 2 and 3 approaches (IPCC 2006) are used over the reported time series to calculate
emissions from MSW Landfills, using two primary methods in accordance with IPCC methodological decision trees
based on available data. The first method uses the first order decay (FOD) model as described by the 2006 IPCC
Guidelines to estimate CFU generation. The amount of Cm recovered and combusted from MSW landfills is
subtracted from the CFU generation and is then adjusted with an oxidation factor. The oxidation factor represents
the amount of Cm in a landfill that is oxidized to CO2 as it passes through the landfill cover (e.g., soil, clay,
geomembrane). This method is presented below.
Equation 7-1: Landfill Methane Generation
^4,MSW = ((*CH4 ~ Zjn—1 nn
where,
CH4msw = (gCH4 - IZ-M * (1 - OX)
Cm,msw = Net CH4 emissions from solid waste
Gch4,msw = CH4 generation from MSW landfills, using emission factors for DOC, k, MCF, F from IPCC
(2006) and other peer-reviewed sources
R = CH4 recovered and combusted
Ox = CH4 oxidized from MSW landfills before release to the atmosphere, using Ox values from
IPCC (2006) and other peer-reviewed or scientifically validated literature (40 CFR Part 98)
The second method used to calculate CH4 emissions from landfills, also called the back-calculation method, is
based on directly measured amounts of recovered CH4 from the landfill gas and is expressed below and by
Equation HH-8 in 40 CFR Part 98.343. The two parts of the equation consider the portion of CFUin the landfill gas
that is not collected by the landfill gas collection system, and the portion that is collected. First, the recovered CH4
is adjusted with the collection efficiency of the gas collection and control system and the fraction of hours the
recovery system operated in the calendar year. This quantity represents the amount of CH4 in the landfill gas that is
not captured by the collection system; this amount is then adjusted for oxidation. The second portion of the
equation adjusts the portion of CH4 in the collected landfill gas with the efficiency of the destruction device(s), and
the fraction of hours the destruction device(s) operated during the year.
The current Inventory uses both methods to estimate CH4 emissions across the time series within EPA's Waste
Model, as summarized in Figure 7-3 below. This chapter provides a summary of the methods, activity data, and
parameters used. Additional stepwise explanations to generate the net emissions are provided in Annex 3.14.
Equation 7-2: Net Methane Emissions from MSW Landfills
CH4,soiid waste = ~ R) xC1 ~ OX) + R x (1 - (DE X fDest))]
where,
CH4 ,solid waste — Net CH4 emissions from solid waste
R = Quantity of recovered CH4 from Equation HH-4 of EPA's GHGRP
CE = Collection efficiency estimated at the landfill, considering system coverage, operation, and
cover system materials from Table HH-3 of EPA's GHGRP. If area by soil cover type
information is not available, the default value of 0.75 should be used (percent)
fREc = fraction of hours the recovery system was operating (percent)
OX = oxidation factor (percent)
DE = destruction efficiency (percent)
fDest = fraction of hours the destruction device was operating (fraction)
7-8 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
Figure 7-3: Methodologies Used Across the Time Series to Compile the U.S. Inventory of
Emission Estimates for MSW Landfills
1990 - 2004
2005 - 2009
2010 - 2016
2017 - Present
U.S.-specific first-order
decay (FOD) model
Back-casted EPA
GHGRP reported net
methane emissions
EPA GHGRP
reported net
methane emissions
EPA GHGRP
reported net
methane emissions
Annex Steps 1-3
Annex Step 4
Annex Step 5
Annex Step 6
IPCC2006 Emission Factors:
• DOC =0.20
• MCF = 1
• DOC, = 0.5
• OX = 0.10
• DE = 0.99
Activity Data:
• National waste generation
data multiplied by the
national disposal factor
• Back-casted GHGRP
emissions plus a 9%
scale-up factor1-2
• Recovery calculated from
four CH4 recovery
databases
• Back-calculated CH4
generation 3
• Weighted average
oxidation factor based on
GHGRP data3
• Net GHGRP emissions
plus a 9% scale-up
factor2
• GHGRP CH4 recovery plus
a 9% scale-up factor
• Back-calculated CH4
generation3
• Weighted average
oxidation factor based on
GHGRP data 3
• Net GHGRP emissions
plus an 11% scale-up
factor2
• GHGRP CH4 recovery plus
an 11% scale-up factor
• Back-calculated CH4
generation3
• Weighted average
oxidation factor based on
GHGRP data 3
1 The intent of the scale-up factor is to estimate emissions from landfills that do not report to the GHGRP. More details on the
scale-up factor and how it was developed can be found in Annex 3.14. The back-casted emissions are calculated using directly
reported net methane emissions for GHGRP reporting years 2010 to 2016. The back-casted emissions are subject to change in
each Inventory based on new reporting year reports and resubmitted greenhouse gas reports for previous years. This method
is compatible with the 2.006 IPCC Guidelines because facilities reporting to the GHGRP either use the FOD method, or directly
measured methane recovery data with default emission factors either directly included in the 2006 IPCC Guidelines or
scientifically validated through peer review.
2 Emission factors used by facilities reporting to GHGRP Subpart HH are facility-specific defaults derived from peer-reviewed
literature and the 2006 IPCC Guidelines.
3 Methane generation is back-calculated from the net MSW emissions, estimated methane recovery data, and the weighted
average oxidation factor based on GHGRP Subpart HH reported data of 0.18 between 2010 to 2016, and 0.21 between 2017 to
2020, and 0.23 in 2021 and 2022.
The Waste Model is a spreadsheet developed by the IPCC for purposes of estimating methane emissions from solid
waste disposal sites, adapted to the United States by the inclusion and usage of U.S.-specific parameters. The
Waste Model contains activity and waste generation information from both the MSW and Industrial landfill sectors
and estimates the amount of ChU emissions from each sector for each year of the time series, using both methods.
Prior to the 1990 through 2015 Inventory, only the FOD method was used. Methodological changes were made to
the 1990 through 2015 Inventory to incorporate higher tier data (i.e., ChU emissions as directly reported to EPA's
GHGRP), which cannot be directly applied to earlier years in the time series without significant bias. The technique
used to merge the directly reported GHGRP data with the previous methodology is described as the overlap
technique in the Time-Series Consistency chapter of the 2006 IPCC Guidelines. Additional details on the technique
used is included in Annex 3.14, and a technical memorandum (RTI 2017).
Supporting information, including details on the techniques used to ensure time-series consistency by
incorporating the directly reported GHGRP emissions is presented in Annex 3.14.
Methodology Applied for Industrial Waste Landfills
Emissions from industrial waste landfills are estimated using a Tier 2 approach (IPCC 2006) and a tailored (country-
specific) IPCC waste model in accordance with IPCC methodological decision trees based on available data. Activity
Waste 7-9
-------
data used are industrial production data (ERG 2023) for two sectors (pulp and paper manufacturing, and food and
beverage manufacturing) to which country-specific default waste disposal factors are applied (a separate disposal
factor for each sector). The disposal factors, as described below, are based on scientifically reviewed data, and are
the same across the entire time series. The emission factors are based on those recommended by the 2006IPCC
Guidelines and are the same across the entire time series.
The FOD equation from IPCC (2006) is used via the waste model to estimate methane emissions:
Equation 7-3: Net Methane Emissions from Industrial Waste Landfills
ch4,ind = (gCH4 - * (1 - OX)
where,
CH4 ,Solid Waste
GcH4,lnd
R
OX
Net Cm emissions from solid waste
Cm generation from industrial waste landfills, using production data multiplied by a
disposal factor and emission factors for DOC, k, MCF, F (IPCC 2006)
Cm recovered and combusted (no recovery is assumed for industrial waste landfills)
Cm oxidized from industrial waste landfills before release to the atmosphere (using the
2006 IPCC Guidelines value for OX of 0.10)
The activity data used in the emission calculations are production data (e.g., the amount of meat, poultry,
vegetables processed; the amount of paper produced) versus disposal data. There are currently no facility-specific
data sources that track and report the amount and type of waste disposed of in the universe of industrial waste
landfills in the United States. Based on this limited information, the Inventory methodology assumes most of the
organic waste placed in industrial waste landfills originates from the food processing (meat, vegetables, fruits) and
pulp and paper sectors, thus estimates of industrial landfill emissions focused on these two sectors.
A waste disposal factor is applied to the annual quantities of key food products generated. A waste disposal factor
of 4.86 percent is used for 1990 to 2009 and a factor of 6 percent is used for 2010 to the current year. The 4.86
percent disposal factor is based on available data from a 1993 Report to Congress (EPA 1993). The 6 percent waste
disposal factor is derived from recent surveys of the food and beverage industry where approximately 94 percent
of food waste generated is repurposed (FWRA 2016). The composition of waste disposed of in industrial waste
landfills is expected to be more consistent in terms of composition and quantity than that disposed of in MSW
landfills. The amount of waste landfilled is assumed to be a fraction of production that is held constant over the
time series as explained in Annex 3.14.
Landfill Cm recovery is not accounted for in industrial waste landfills and is believed to be minimal based on
available data collected under EPA's GHGRP for industrial waste landfills (Subpart TT), which shows that only one
of the 167 facilities, or 1 percent of facilities, have active gas collection systems (EPA 2023a). The amount of Cm
oxidized by the landfill cover at industrial waste landfills is assumed to be 10 percent of the CH4 generated (IPCC
2006; Mancinelli and McKay 1985; Czepiel et al. 1996) for all years.
Additionally, the 2019 Refinement to the 2006 IPCC Guidelines for National Greenhouse Gas Inventories (IPCC 2019)
were reviewed to determine if any revisions were required to emission factors, methodologies, and assumptions
underlying emission estimates for all source categories. None of the 2019 Refinements are applicable to the
country-specific methodology applied for the landfills source category.
Box 7-3: Nationwide Municipal Solid Waste Data Sources
Municipal solid waste (MSW) generated in the United States can be managed through a variety of methods. MSW
that is not recycled, composted, combusted with energy recovery, or digested is assumed to be landfilled. In
addition to these management pathways, waste or excess food from the food manufacturing and processing
sector may be disposed through the sewerage network, used for animal feed, land application, donated for
human consumption, and rendered or recycled into biofuels in the case of animal by-products, fats, oils and
greases.
7-10 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
There have been three main sources for nationwide solid waste management data in the United States that the
Inventory has used (see Annex 3.14, Box A-3 for comparison of estimates from these data sources):
• The BioCycle and Earth Engineering Center of Columbia University's SOG in America surveys [no longer
published];
• The EPA's Advancing Sustainable Materials Management: Facts and Figures reports; and
• The EREF's MSW Generation in the United States reports.
The SOG surveys and, most recently EREF, collected state-reported data on the amount of waste generated and
the amount of waste managed via different management options: landfilling, recycling, composting, and
combustion. These data sources used a 'bottom-up' method. The survey asked 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 was not available, the survey asked for
total tons landfilled. The data were adjusted for imports and exports across state lines so that the principles of
mass balance were adhered to for completeness, whereby the amount of waste managed did not exceed the
amount of waste generated. The SOG and EREF reports present survey data aggregated to the state level.
The EPA Advancing Sustainable Materials Management: Facts and Figures report characterizes national post-
consumer municipal solid waste (MSW) generation and management using a top-down materials flow (mass
balance) methodology. It captures an annual snapshot of MSW generation and management in the United States
for specific products. Data are gathered from U.S. Government (e.g., U.S. Census Bureau and U.S. Department of
Commerce), state environmental agencies, industry and trade groups, and sampling studies. The materials flow
methodology develops MSW waste generation estimates of quantities of MSW products in the marketplace (using
product sales and replacement data) and assessing waste generation by component material based on product
lifespans. The data are used to estimate tons of materials and products generated, recycled, combusted with
energy recovery, managed via other food waste management pathways, or landfilled nationwide. MSW that is
not recycled or composted is assumed to be combusted or landfilled, except for wasted food, which uses a
different methodology and includes nine different management pathways. The 2018 Facts and Figures Report
(EPA 2020) uses a methodology that expanded the number of management pathways to include: animal feed;
bio-based materials and/or biochemical processing (i.e., rendering); co-digestion and/or anaerobic digestion;
composting/aerobic processes; combustion; donation; land application; landfill; and sewer or wastewater
treatment.
In this Inventory, emissions from solid waste management are presented separately by waste disposal option,
except for recycling of waste materials.
• Recycling: 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. The emissions estimates for recycling are not called out separately.
• Landfill Disposal: Emissions from solid waste disposal in landfills and the composting of solid waste
materials are presented in the Landfills and Composting sections in the Waste sector of this report.
• Anaerobic Digestion: Emissions from anaerobic digesters are presented in three different sections
depending on the digester category:
o Emissions from on-farm digesters are included in the Agriculture sector.
o Emissions from digesters at wastewater treatment plants are included in the Waste sector, and
o Emissions from stand-alone digesters are also included in the Waste sector.
• Waste Incineration: Emissions from waste incineration are accounted for in the Incineration chapter of
the Energy sector of this report because, in the United States, almost all incineration of MSW occurs at
waste-to-energy (WTE) facilities or industrial facilities where useful energy is recovered.s
Waste 7-11
-------
Uncertainty
Several types of uncertainty are associated with the estimates of Cm emissions from MSW and industrial waste
landfills when the FOD method is applied directly for 1990 to 2004 in the Waste Model and, to some extent, in the
GHGRP methodology. The approach used in the MSW emission estimates assumes that the Cm generation
potential (L0) and the rate of decay that produces Cmfrom MSW, as determined from several studies of Cm
recovery at MSW landfills, are representative of conditions at U.S. MSW landfills. When this top-down approach is
applied at the nationwide level, the uncertainties are assumed to be less than when applying this approach to
individual landfills and then aggregating the results to the national level. In other words, the FOD method as
applied in this Inventory is not facility-specific modeling and while this approach may over- or underestimate CFU
generation at some landfills if used at the facility-level, the result is expected to balance out because it is being
applied nationwide.
There is a high degree of uncertainty associated with the FOD model, particularly when a homogeneous waste
composition and hypothetical decomposition rates are applied to heterogeneous landfills (IPCC 2006). There is less
uncertainty in EPA's GHGRP data because this methodology is facility-specific, uses directly measured Cm recovery
data (when applicable), and allows for a variety of landfill gas collection efficiencies, destruction efficiencies,
and/or oxidation factors to be used.
Uncertainty also exists in the scale-up factors (both 9 percent and 11 percent) applied for years 2005 to 2016 and
2017 to 2022, respectively, and in the back-casted emissions estimates for 2005 to 2009. As detailed in RTI (2018),
limited information is available for landfills that do not report to the GHGRP. RTI developed an initial list of landfills
that do not report to the GHGRP with the intent of quantifying the total waste-in-place for these landfills that
would add up to the scale-up factor. Input was provided by industry, LMOP, and additional EPA support. However,
many gaps existed in the initial development of this Non-Reporting Landfills Database. Assumptions were made for
hundreds of landfills to estimate their waste-in-place and the subsequent scale-up factors. The waste-in-place
estimated for each landfill is likely not 100 percent accurate and should be considered a reasonable estimate.
Additionally, a simple methodology was used to back-cast emissions for 2005 to 2009 using the GHGRP-reported
emissions from 2010 to 2022. This methodology does not factor in annual landfill to landfill changes in landfill CH4
generation and recovery. Because of this, an uncertainty factor of 25 percent is applied to the scale-up factor and
years (emission estimates) the scale-up factor is applied to.
Aside from the uncertainty in estimating landfill CH4 generation, uncertainty also exists in the estimates of the
landfill gas oxidized at MSW landfills. Facilities directly reporting to EPA's GHGRP can use oxidation factors ranging
from 0 to 35 percent, depending on their facility-specific CH4 flux. As recommended by the 2006 IPCC Guidelines
for managed landfills, a 10 percent default oxidation factor is applied in the Inventory for both MSW landfills
(those not reporting to the GHGRP and for the years 1990 to 2004 when GHGRP data are not available) and
industrial waste landfills regardless of climate, the type of cover material, and/or presence of a gas collection
system.
Another significant source of uncertainty lies with the estimates of CH4 recovered by flaring and gas-to-energy
projects at MSW landfills that are sourced from the Inventory's CH4 recovery databases (used for years 1990 to
2004). Four CH4 recovery databases are used to estimate nationwide CH4 recovery for MSW landfills for 1990 to
2009. The GHGRP MSW landfills database was added as a fourth recovery database starting with the 1990 to 2013
Inventory report (two years before the full GHGRP data set started being used for net CH4 emissions for the
Inventory). Relying on multiple databases for a complete picture introduces uncertainty because the coverage and
characteristics of each database differs, which increases the chance of double counting avoided emissions. The
methodology and assumptions that go into each database differ. For example, the flare database assumes the
midpoint of each flare capacity at the time it is sold and installed at a landfill; the flare may be achieving a higher
capacity, in which case the flare database would underestimate the amount of CH4 recovered. Additionally, two
databases, the EIA database and flare vendor database, could no longer be updated for the entire time series due
to external factors. For example, the EIA database has not been updated since 2006 because the EIA stopped
collecting landfill recovery data. The EIA database has, for the most part, been replaced by the GHGRP MSW
7-12 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
landfills database. The flare database was populated annually until 2015, but decreasing, voluntary participation
from flare vendors sharing their flare sales data for several years prior to 2015.
To avoid double counting and to use the most relevant estimate of Cm recovery for a given landfill, a hierarchical
approach is used among the four databases. GHGRP data and the EIA data are given precedence because facility
data were directly reported; the LFGE data are given second priority because Cm recovery is estimated from
facility-reported LFGE system characteristics; and the flare data are given the lowest priority because this database
contains minimal information about the flare, no site-specific operating characteristics, and includes smaller
landfills not included in the other three databases (Bronstein et al. 2012). The coverage provided across the
databases most likely represents the complete universe of landfill Cm gas recovery; however, the number of
unique landfills between the four databases does differ.
The 2006IPCC Guidelines default value of 10 percent for uncertainty in recovery estimates was used for two of the
four recovery databases in the uncertainty analysis where metering of landfill gas was in place (for about 64
percent of the Cm estimated to be recovered). This 10 percent uncertainty factor applies to the LFGE database; 12
percent to the EIA database; and 1 percent for the GHGRP MSW landfills dataset because of the supporting
information provided and rigorous verification process. For flaring without metered recovery data (the flare
database), a much higher uncertainty value of 50 percent is used. The compounding uncertainties associated with
the four databases in addition to the uncertainties associated with the FOD method and annual waste disposal
quantities leads to the large upper and lower bounds for MSW landfills presented in Table 7-5.
The lack of landfill-specific information regarding the number and type of industrial waste landfills in the United
States is a primary source of uncertainty with respect to the industrial waste generation and emission estimates.
The approach used here assumes that most of the organic waste disposed of in industrial waste landfills that
would result in CFU emissions consists of waste from the pulp and paper and food processing sectors. However,
because waste generation and disposal data are not available in an existing data source for all U.S. industrial waste
landfills, a straight disposal factor is applied over the entire time series to the amount produced to determine the
amounts disposed. Industrial waste facilities reporting under EPA's GHGRP do report detailed waste stream
information, and these data have been used to improve, for example, the DOC value used in the Inventory
methodology for the pulp and paper sector. A 10 percent oxidation factor is also applied to CH4 generation
estimates for industrial waste landfills and carries the same amount of uncertainty as with the factor applied to
CH4 generation for MSW landfills. The specified probability density functions (PDFs) are assumed to be normal for
most activity data and emission factors, and due to lack of data, are based on expert judgement (RTI 2004).
The results of the 2006 IPCC Guidelines Approach 2 quantitative uncertainty analysis are summarized in Table 7-5.
There is considerable uncertainty for the MSW landfills estimates due to the many data sources used, each with its
own uncertainty factor.
Table 7-5: Approach 2 Quantitative Uncertainty Estimates for CH4 Emissions from Landfills
(MMT CO2 Eq. and Percent)
Source
Gas
2022 Emission Estimate
(MMTCO. Eq.)
Uncertainty Range Relative to Emission Estimate-'
(MMTCO' Eq.) (%)
Lower
Upper
Lower Upper
Bound
Bound
Bound Bound
Total Landfills
ch4
119.8
109.9
137.2
-8% +15%
MSW
ch4
100.9
98.8
121.2
-2% +20%
Industrial
ch4
18.9
13.1
23.7
-31% +25%
a Range of emission estimates predicted by Monte Carlo stochastic simulation for a 95 percent confidence interval.
Individual uncertainty factors are applied to activity data and emission factors in the Monte Carlo analysis.
Waste 7-13
-------
QA/QC and Verification
General quality assurance/quality control (QA/QC) procedures were applied consistent with the U.S. Inventory
QA/QC Plan, which is in accordance with Vol. 1, Chapter 6 of the 2006IPCC Guidelines (see Annex 8 for more
details). QA/QC checks are performed for the transcription of the published data set (e.g., EPA's GHGRP dataset)
used to populate the Inventory data set in terms of completeness and accuracy against the reference source.
Additionally, all datasets used for this category have been checked to ensure they are of appropriate quality and
are representative of U.S. conditions. The primary calculation spreadsheet is tailored from the 2006 IPCC
Guidelines waste model and has been verified previously using the original, peer-reviewed IPCC waste model. All
model input values and calculations were verified by secondary QA/QC review. Stakeholder engagements sessions
in 2016 and 2017 were used to gather input on methodological improvements and facilitate an external expert
review on the methodology, activity data, and emission factors.
Category-specific checks include the following:
• Evaluation of the secondary data sources used as inputs to the Inventory dataset to ensure they are
appropriately collected and are reliable;
• Cross-checking the data (activity data and emissions estimates) with previous years to ensure the data are
reasonable, and that any significant variation can be explained through the activity data;
• Conducting literature reviews to evaluate the appropriateness of country-specific emission factors (e.g.,
DOC values, precipitation zones with respect to the application of the k values) given findings from recent
peer-reviewed studies; and
• Reviewing secondary datasets to ensure they are nationally complete and supplementing where
necessary (e.g., using a scale-up factor to account for emissions from landfills that do not report to EPA's
GHGRP).
A primary focus of the QA/QC checks in past Inventories was to ensure that Cm recovery estimates were not
double-counted and that all LFGE projects and flares were included in the respective project databases. QA/QC
checks performed in the past for the recovery databases were not performed in this Inventory, because new data
were not added to the recovery databases in this Inventory year.
For the GHGRP data, EPA verifies annual facility-level reports through a multi-step process (e.g., combination of
electronic checks and manual reviews by staff) to identify potential errors and ensure that data submitted to EPA
are accurate, complete, and consistent.2 Based on the results of the verification process, EPA follows up with
facilities to resolve mistakes that may have occurred. The post-submittals checks are consistent with several
general and category-specific QC procedures, including range checks, statistical checks, algorithm checks, and year-
to-year checks of reported data and emissions. For the MSW Landfills sector, under Subpart HH of the GHGRP,
MSW Landfills with gas collection are required to report emissions from their site using both a forward- (using a
first order decay model as a basis) and back-calculating (using parameters specific to the landfill itself, such as
measured recovery and collection efficiency of the landfill gas) methodology. Details on the forward- and back-
calculation approach can be found in Annex 3.14 and 40 CFR Subpart HH of Part 98. Reporters can choose which of
these two methodologies they believe best represents the emissions at their landfill and are required to submit
that value as their total Subpart HH emissions. Facilities are generally not expected to switch between the two
equations each year, as the emissions calculated using each method can vary greatly and can have a significant
effect on emission trends for that landfill, and potentially the entire MSW Landfill sector under the GHGRP. Key
checks are in place to assure that emissions are trending in a sensible way year over year for each reporting
landfill.
2 See https://www.epa.gov/sites/production/files/2015-07/documents/eherp verification factsheet.pdf.
7-14 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
Recalculations Discussion
Revisions to the individual facility reports submitted to EPA's GHGRP can be made at any time and a portion of
facilities have revised their reports since 2010 for various reasons, resulting in changes to the total net Cm
emissions for MSW landfills. Each Inventory year, the back-casted emissions for 2005 to 2009 will be recalculated
using the most recently verified data from the GHGRP. Changes in these data result in changes to the back-casted
emissions. The impact of the revisions to the GHGRP Subpart HH annual greenhouse gas reports resubmitted for
2017 to 2021 slightly decreased total Subpart HH reported net emissions, which decreased net MSW emissions by
an average of 0.5 percent. A change in net Subpart HH reported emissions results in the same percentage change
in the Inventory emissions for that year.
Slight revisions were also made to the food and beverage sector production data for nearly every year of the time
series. The production data increased by 1 to 2 MMT per year between 1990 to 2017 and decreased by a few MMT
per year between 2019 to 2021. These revisions decreased net emissions from industrial waste landfills by less
than 0.1 percent between 1990 to 2010. Emissions increased slightly between 2011 to 2017 (ranging from 0.3
percent in 2011 to a high of 1.3 percent in 2017). The revisions to the production data also slightly increased
emissions by an average of 0.6 percent between 2018 to 2021.
The combined changes to the MSW and industrial waste landfills activity data resulted in annual increases ranging
from 0.005 percent to 0.01 percent to net emissions between 2005 to 2010. A slight decrease in net emissions is
observed between 2011 to 2016 (-0.04 percent to -0.2 percent), and an increase, averaging 0.36 percent of
emissions, is observed between 2017 to 2021. A 0.6 percent increase is observed for 2020, and a 0.5 percent
increase is observed for 2021. Between 2005 to 2020, on average, the impact or change was very small (less than
0.1 percent).
Planned Improvements
EPA received recommendations from industry stakeholders regarding the DOC values and decay rates (k values)
required to be used in the GHGRP calculations. Stakeholders have suggested that newer, more up-to-date default
values considering recent trends in the composition of waste disposed in MSW landfills for both k and DOC in the
GHGRP should be developed and reflected in the 2005 and later years of the Inventory. In response, EPA
developed a multivariate analysis using publicly available Subpart HH GHGRP data, solving for optimized DOC and k
values across the more than 1,100 landfills reporting to the program. The results of this analysis could help inform
a current GHGRP rulemaking (87 FR 36920) where changes could be made to the default DOC and k values
contained within Subpart HH, which could then be carried over to the Inventory emissions estimates for MSW
landfills upon promulgation of any revisions to 40 CFR Part 98. This potential improvement may be long-term.
With respect to the scale-up factor, EPA received comments on revisions made to the scale-up for the 1990 to
2020 Inventory from a total waste-in-place approach to a time-based threshold of 50 years. Commenters noted
that this time-based threshold approach does not adjust for the non-linearity of methane production of landfill
gas. In response, EPA will further investigate how best to account for emissions from MSW landfills that do not
report to the GHGRP, including using the FOD model for these landfills based on estimated annual waste disposed
for this subset of landfills between 2005 to 2022, reverting to the total waste-in-place approach, or modifying the
time-based threshold approach. Any methodological revisions to accounting for emissions from this subset of
landfills will be made in the next (1990 to 2023) Inventory.
Relatedly, EPA will periodically assess the impact to the waste-in-place and emissions data from GHGRP facilities
that have resubmitted annual reports during any reporting years, are new reporting facilities, and from facilities
that have stopped reporting to the GHGRP to ensure national estimates are as complete as possible. Facilities may
stop reporting to the GHGRP when they meet the "off-ramp" provisions (reported less than 15,000 metric tons of
CO2 equivalent emissions for 3 consecutive years or less than 25,000 metric tons of CO2 equivalent emissions for 5
consecutive years). If warranted, EPA will revise the scale-up factor to reflect newly acquired information to ensure
completeness of the Inventory. EPA considered public comments received on the 1990 through 2019 Inventory
Waste 7-15
-------
specific to using a time-based threshold to calculate the scale-up factor instead of a total waste-in-place approach.
The rationale supporting the comments was that older, closed landfills with large quantities of waste-in-place are
driving up the scale-up factor but have little impact on total methane generation. EPA assessed two time-based
scenarios for developing the scale-up factor - one scenario looking at the past 30 years of waste disposed, and the
second looking at the past 50 years of waste disposed. The 50-year time-based threshold was applied and resulted
in the 11 percent scale-up factor used between 2017 and 2022.
EPA is planning to account for unmanaged landfills in Puerto Rico and other U.S. Territories to the landfill
emissions estimates. Data limitations for historical waste received at these sites make this challenging. Presently,
emissions from managed sites in Puerto Rico and Guam are accounted for in 2005 to present as part of the GHGRP
Subpart HH dataset.
Box 7-4: Overview of U.S. Solid Waste Management Trends
As shown in Figure 7-4 and Figure 7-5 landfilling of MSW is currently and has been the most common waste
management practice. A large portion of materials in the waste stream are recovered for recycling and
composting, which is becoming an increasingly prevalent trend throughout the country. Materials that are
composted and recycled would have previously been disposed in a landfill.
Figure 7-4: Management of Municipal Solid Waste in the United States, 2018
Other Food Management
6%
Note: 2018 is the latest year of available data. Data taken from Table 35 of EPA (2020a). MSW to WTE is combustion with
energy recovery (WTE = waste-to-energy).
Source: EPA (2020b)
7-16 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
Figure 7-5: MSW Management Trends from 1990 to 2018
Note: 2018 is the latest year of available data. Only one year of data (2018) is available for the "Other Food Management"
category.
Source: EPA (2020b). The EPA Advancing Sustainable Materials Management reports only present data for select years, thus
several reports were used in the compilation of this figure. All data were taken from Table 35 in EPA 2020b for 1990, 2000,
2015, 2017 and 2018. Data were taken from Table 35 in EPA (2019) for 2010 and 2016. Data were taken from EPA (2018)
for 2014. Data were taken from Table 35 of EPA (2016b) for 2012 and 2013. Data were taken from Table 30 of EPA (2014)
for 2008 and 2011. The reports with data available for years prior to EPA (2012) can be provided upon request but are no
longer on the EPA's Advancing Sustainable Materials Management web site.3
Table 7-6 presents the national-level material composition of waste disposed across typical MSW landfills 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 (i.e.,
those types known to generate Cm as they break down in a modern MSW landfill), is important for estimating
greenhouse gas emissions. Increased diversion of degradable materials so that they are not disposed of in
landfills reduces the CFU generation potential and Cm emissions from landfills. For certain degradable waste
types (i.e., paper and paperboard), the amounts discarded have decreased over time due to an increase in
waste diversion through recycling and composting (see Table 7-6 and Figure 7-6). As shown in Figure 7-6, the
diversion of food scraps has been consistently low since 1990 because most cities and counties do not practice
curbside collection of these materials, although the quantity has been slowly increasing in recent years. Neither
Table 7-6 nor Figure 7-6 reflect the frequency of backyard composting of yard trimmings and food waste
because this information is largely not collected nationwide and is hard to estimate.
Table 7-6: Materials Discarded in the Municipal Waste Stream by Waste Type from 1990 to
2018 (Percent)
Waste Type
1990
2005
2015
2016
2017
2018
Paper and Paperboard
30.0%
24.7%
13.3%
12.7%
13.1%
11.8%
Glass
6.0%
5.8%
5.0%
4.9%
4.9%
5.2%
Metals
7.2%
7.9%
9.5%
9.8%
9.9%
9.5%
Plastics
9.5%
16.4%
18.9%
18.9%
19.2%
18.5%
Rubber and Leather
3.2%
2.9%
3.3%
3.4%
3.5%
3.4%
Waste 7-17
-------
Textiles
2.9%
5.3%
7.7%
8.0%
8.0%
7.7%
Wood
6.9%
7.5%
8.0%
8.8%
8.7%
8.3%
Other
1.4%
1.8%
2.2%
2.2%
2.2%
2.0%
Food Scraps
13.6%
18.5%
22.0%
22.1%
22.0%
24.1%
Yard Trimmings
17.6%
7.0%
7.8%
6.9%
6.2%
7.2%
Miscellaneous Inorganic Wastes
1.7%
2.2%
2.3%
2.3%
2.3%
2.3%
Source: EPA (2020b)
Figure 7-6: Percent of Degradable Materials Diverted from Landfills from 1990 to 2018
(Percent)
%
c
JS
1
.<£
"O
3
-
I
¦d
3
~o
£
CL
O
(U
-
Q
a.
1990 1992 1994 1996 1998 2000 2002 2004 2006 2008 2010 2012 2014 2016 2018
Note: The data shown in this chart are for recycling of paper and paperboard, composting of food scraps and yard trimmings,
and alternative management pathways for the Other Food Management (non-disposal) category. The Other Food
Management (non-disposal) category is a new addition and only one year of data are available for 2018 (28 percent of the
food waste generated was beneficially reused or managed using a method that was not landfilling, recycling, or
composting). The Other Food Management pathways include animal feed, bio-based materials/biochemical processing, co-
digestion/anaerobic digestion, donation, land application, and sewer/wastewater treatment.
Source: EPA (2020b). The EPA Advancing Sustainable Materials reports only present data for select years, thus several
reports were used in the compilation of this figure. All data were taken from Table 35 in EPA (2020b) for 1990, 2000, 2015,
2017 and 2018. Data were taken from Table 35 in EPA (2019) for 2010 and 2016. Data were taken from EPA (2018) for
2014. Data were taken from Table 35 of EPA (2016b) for 2012 and 2013. Data were taken from Table 30 of EPA (2014) for
2008 and 2011. The reports with data available for years prior to EPA (2012) can be provided upon request, but are not
longer on the EPA's Advancing Sustainable Materials Management website.4
3 See https://www.epa.gov/facts-and-figures-about-materials-waste-and-recvcling/advancing-sustainable-materials-
management.
4 See https://www.epa.gov/facts-and-figures-about-materials-waste-and-recvcling/advancing-sustainable-materials-
management.
7-18 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
7.2 Wastewater Treatment and Discharge
(CRT Source Category 5D)
Wastewater treatment and discharge processes are sources of anthropogenic methane (Cm) and nitrous oxide
(N2O) emissions. Wastewater from domestic and industrial sources is treated to remove soluble organic matter,
suspended solids, nutrients, pathogenic organisms, and chemical contaminants.5 Treatment of domestic
wastewater may either occur on site, most commonly through septic systems, or off site at centralized treatment
systems, most commonly at publicly owned treatment works (POTWs). In the United States, approximately 16
percent of domestic wastewater is treated in septic systems or other on-site systems, while the rest is collected
and treated centrally (U.S. Census Bureau 2021a). Treatment of industrial wastewater may occur at the industrial
plant using package or specially designed treatment plants or be collected and transferred off site for co-treatment
with domestic wastewater in centralized treatment systems.
Centralized Treatment. Centralized wastewater treatment systems use sewer systems to collect and transport
wastewater to the treatment plant. Sewer collection systems provide an environment conducive to the formation
of Cm, which can be substantial depending on the configuration and operation of the collection system (Guisasola
et al. 2008). Recent research has shown that at least a portion of CH4 formed within the collection system enters
the centralized system where it contributes to CH4 emissions from the treatment system (Foley et al. 2015).
The treatment plant may include a variety of processes, ranging from physical separation of material that readily
settles out (typically referred to as primary treatment), to treatment operations that use biological processes to
convert and remove contaminants (typically referred to as secondary treatment), to advanced treatment for
removal of targeted pollutants, such as nutrients (typically referred to as tertiary treatment). Not all wastewater
treatment plants conduct primary treatment prior to secondary treatment, and not all plants conduct advanced or
tertiary treatment (EPA 2010).
Soluble organic matter is generally removed using biological processes in which microorganisms consume the
organic matter for maintenance and growth. Microorganisms can biodegrade soluble organic material in
wastewater under aerobic or anaerobic conditions, where the latter condition produces CH4. The resulting biomass
(sludge) is removed from the wastewater (effluent) prior to discharge to the receiving stream and may be further
biodegraded under aerobic or anaerobic conditions, such as anaerobic sludge digestion. Sludge can be produced
from both primary and secondary treatment operations. Some wastewater may also be treated using constructed
(or semi-natural) wetland systems, though this is much less common in the United States and represents a
relatively small portion of wastewater treated centrally (<0.1 percent) (ERG 2016). Constructed wetlands are a
coupled anaerobic-aerobic system and may be used as the primary method of wastewater treatment, or are more
commonly used as a final treatment step following settling and biological treatment. Constructed wetlands
develop natural processes that involve vegetation, soil, and associated microbial assemblages to trap and treat
incoming contaminants (IPCC 2014). Constructed wetlands do not produce secondary sludge (sewage sludge).
Emissions from flooded lands or constructed waterbodies (not used for wastewater treatment) and lands
converted to flooded lands (not used for wastewater treatment) are estimated and reported in Chapter 6, under
Sections 6.8 Wetlands Remaining Wetlands and 6.9 Lands Converted to Wetlands.
The generation of N2O may also result from the treatment of wastewater during both nitrification and
denitrification of the nitrogen (N) present, usually in the form of urea, proteins, and ammonia in wastewater.
Ammonia N is converted to nitrate (NO3) through the aerobic process of nitrification. Denitrification occurs under
anoxic/anaerobic conditions, whereby anaerobic or facultative organisms reduce oxidized forms of nitrogen (e.g.,
5 Throughout the Inventory, emissions from domestic wastewater also include any commercial and industrial wastewater
collected and co-treated with domestic wastewater.
Waste 7-19
-------
nitrite, nitrate) in the absence of free oxygen to produce nitrogen gas (N2). Nitrous oxide is generated as a by-
product of nitrification, or as an intermediate product of denitrification. No matter where N2O is formed it is
typically stripped (i.e., transferred from the liquid stream to the air and ultimately emitted to the atmosphere) in
aerated parts of the treatment process. Stripping also occurs in non-aerated zones at rates lower than in aerated
zones.
On-site Treatment. The vast majority of on-site systems in the United States are septic systems composed of a
septic tank, generally buried in the ground, and a soil dispersion system. Solids and dense materials contained in
the incoming wastewater (influent) settle in the septic tank as sludge. Floatable material (scum) is also retained in
the tank. The sludge that settles on the bottom of the tank undergoes anaerobic digestion. Partially treated water
is discharged in the soil dispersal system. The solid fraction accumulates and remains in the tank for several years,
during which time it degrades anaerobically. The gas produced from anaerobic sludge digestion (mainly Cm and
biogenic CO2) rises to the liquid surface and is typically released through vents. The gas produced in the effluent
dispersal system (mainly N2O and biogenic CO2) is released through the soil.
Discharge. Dissolved Cm and N2O that is present in wastewater discharges to aquatic environments has the
potential to be released into the atmosphere (Short et al. 2014; Short et al. 2017). In addition, the presence of
organic matter or nitrogen in wastewater discharges is generally expected to increase CH4 and N2O emissions from
these aquatic environments. Where organic matter is released to slow-moving aquatic systems, such as lakes,
estuaries, and reservoirs, CH4 emissions are expected to be higher. Similarly, in the case of discharge to nutrient-
impacted or hypoxic waters, N2O emissions can be significantly higher.
In summary, the principal factor in determining the CH4 generation potential of wastewater is the amount of
degradable organic material in the wastewater. Common parameters used to measure the organic component of
the wastewater are the biochemical oxygen demand (BOD) and chemical oxygen demand (COD). Under the same
conditions, wastewater with higher COD (or BOD) concentrations will generally yield more CH4 than wastewater
with lower COD (or BOD) concentrations. BOD represents the amount of oxygen that would be required to
completely consume the organic matter contained in the wastewater through aerobic decomposition processes,
while COD measures the total material available for chemical oxidation (both biodegradable and non-
biodegradable). The BOD value is most commonly expressed in milligrams of oxygen consumed per liter of sample
during 5 days of incubation at 20°C, or BODs. Throughout the rest of this chapter, the term "BOD" refers to BODs.
Because BOD is an aerobic parameter, it is preferable to use COD to estimate CH4 production, since Cm is
produced only in anaerobic conditions. Where present, biogas recovery and flaring operations reduce the amount
of Cm generated that is actually emitted. Per IPCC guidelines (IPCC 2019), emissions from anaerobic sludge
digestion, including biogas recovery and flaring operations, where the digester's primary use is for treatment of
wastewater treatment solids, are estimated and reported under wastewater treatment. The principal factor in
determining the N2O generation potential of wastewater is the amount of N in the wastewater. The variability of N
in the influent to the treatment system, as well as the operating conditions of the treatment system itself, also
impact the N2O generation potential. The methods and underlying data sources to estimate emissions from are
described in further detail in the "Methodology and Time Series Consistency" section below for treatment of
domestic and industrial wastewater.
Overall, treatment of wastewater emitted 42.7 MMT CO2 Eq. in 2022. Total methane (CH4) emissions from
wastewater treatment and discharge were estimated to be 20.8 MMT CO2 Eq. (743 kt CH4). Methane (CH4)
emissions from domestic wastewater treatment and discharge were estimated to be 11.6 MMT CO2 Eq. (413 kt
CH4) and 2.0 MMT CO2 Eq. (72 kt CH4), respectively, totaling 13.6 MMT CO2 Eq. (485 kt CH4) in 2022. Emissions
remained fairly steady from 1990 through 2002 but have decreased since that time due to decreasing percentages
of wastewater being treated in anaerobic systems, generally including reduced use of on-site septic systems and
central anaerobic treatment systems (EPA 1992,1996, 2000, and 2004; U.S. Census Bureau 2021a). In 2022, CH4
emissions from industrial wastewater treatment and discharge were estimated to be 6.7 MMT CO2 Eq. (239 kt CH4)
and 0.5 MMT CO2 Eq. (19 kt CH4), respectively, totaling 7.2 MMT CO2 Eq. (258 kt CH4). Industrial emissions from
wastewater treatment have generally increased across the time series through 1999 and then fluctuated up and
correspond with production changes from the pulp and paper manufacturing, meat and poultry processing, fruit
7-20 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
and vegetable processing, starch-based ethanol production, petroleum refining, and brewery industries. Industrial
wastewater emissions have generally seen an uptick since 2016. Table 7-7 and Table 7-8 provide Cm emission
estimates from domestic and industrial wastewater treatment.
Nitrous oxide (N2O) emissions from wastewater treatment and discharge in 2022 totaled 21.9 MMT CO2 Eq. (83 kt
N2O). In 2022, domestic treatment and discharge were estimated to be 17.0 MMT CO2 Eq. (64 kt N2O) and 4.4
MMT CO2 Eq. (16 kt N2O), respectively, totaling 21.4 MMT CO2 Eq. (81 kt N2O). Domestic emission sources have
gradually increased across the time series because of an increasing U.S. population and protein consumption. In
2022, N2O emissions from industrial wastewater treatment and discharge were estimated to be 0.4 MMT CO2 Eq.
(1.5 kt N2O) and 0.1 MMT CO2 Eq. (0.3 kt N2O), respectively, totaling 0.5 MMT CO2 Eq. (1.8 kt N2O). Industrial
emission sources have gradually increased across the time series with production changes associated with the
treatment of wastewater namely from meat and poultry processing and petroleum refining, but also with
contributions from pulp and paper manufacturing and brewery industries. Table 7-7 and Table 7-8 provide N2O
emission estimates from domestic wastewater treatment.
Table 7-7: CH4 and N2O Emissions from Domestic and Industrial Wastewater Treatment
(MMT C02 Eq.)
Activity
1990
2005
2018
2019
2020
2021
2022
ch4
22.7
22.7
21.4
21.1
21.0
20.7
20.8
Domestic T reatment
15.1
14.6 I
12.3
11.9
11.7
11.4
11.6
Domestic Effluent
1.4
1.4 1
2.0
2.0
2.1
2.1
2.0
Industrial Treatment3
5.5
6.1
6.5
6.6
6.6
6.7
6.7
Industrial Effluent3
0.7
0.6 1
0.6
0.5
0.5
0.5
0.5
n2o
14.8
18.1
21.2
21.6
22.3
22.1
21.9
Domestic T reatment
10.5
13.7 I
16.2
16.6
17.2
17.1
17.0
Domestic Effluent
3.9
3.9 1
4.5
4.5
4.6
4.5
4.4
Industrial Treatment15
0.3
0.4 1
0.4
0.4
0.4
0.4
0.4
Industrial Effluentb
0.1
0.1
0.1
0.1
0.1
0.1
0.1
Total
37.5
40.7
42.5
42.7
43.2
42.7
42.7
a Industrial activity for CH4 includes the pulp and paper manufacturing, meat and poultry processing, fruit and vegetable
processing, starch-based ethanol production, petroleum refining, and breweries industries.
b Industrial activity for N20 includes the pulp and paper manufacturing, meat and poultry processing, starch-based ethanol
production, and petroleum refining.
Note: Totals may not sum due to independent rounding.
Table 7-8: CH4 and N2O Emissions from Domestic and Industrial Wastewater Treatment (kt)
Activity
1990
2005
2018
2019
2020
2021
2022
ch4
811
809
763
755
748
738
743
Domestic T reatment
539
521
438
426
419
407
413
Domestic Effluent
49
49
73
73
74
74
72
Industrial Treatment3
196
216 :
232
236
236
238
239
Industrial Effluent3
27
22
20
19
19
19
19
N20
56
!!!!!»
68
80
81
84
83
83
Domestic Treatment
40
52
61
63
65
65
64
Domestic Effluent
15 =
15
17
17
17
17
16
Industrial Treatment15
1
1
2
2
1
1
1
Industrial Effluentb
+ 5
+ i
+
+
+
+
+
+ Does not exceed 0.5 kt.
a Industrial activity for CH4 includes the pulp and paper manufacturing, meat and poultry processing, fruit and vegetable
processing, starch-based ethanol production, petroleum refining, and breweries industries.
b Industrial activity for N20 includes the pulp and paper manufacturing, meat and poultry processing, starch-based ethanol
production, and petroleum refining.
Note: Totals by gas may not sum due to independent rounding
Waste 7-21
-------
Methodology and Time-Series Consistency
The methodologies presented in IPCC (2019) form the basis of the Cm and N2O emission estimates for both
domestic and industrial wastewater treatment and discharge.6 Domestic wastewater treatment follows the IPCC
Tier 2 methodology for key pathways, while domestic wastewater discharge follows IPCC Tier 2 discharge
methodology and emission factors in accordance with IPCC methodological decision trees based on available data
for treatment and discharge. Default factors from IPCC (2019) or IPCC (2006) are used when there are insufficient
U.S.-specific data to develop a U.S.-specific factor, though IPCC default factors are often based in part on data from
or representative of U.S. wastewater treatment systems. Industrial wastewater treatment follows IPCC Tier 1 and
wastewater treatment discharge follows Tier 1 or Tier 2 methodologies, again in accordance with methodological
decision trees and available data, depending on the industry. EPA will continue to implement the Tier 2 discharge
methodology for more industries as data are investigated and time and resource constraints allow (see the
Planned Improvements section below). Similar to domestic wastewater, IPCC default factors are used when there
are insufficient U.S.-specific data to develop a U.S.-specific factor.
Methodological approaches were applied to the entire time series to ensure consistency in emissions from 1990
through 2022. In the following cases, the source used to capture activity data changed over the time series. EPA
transitioned to these newer data sources to continue estimating emissions in a way that ensured both accuracy
and continuity. For example:
• Starch-based ethanol production data: the source used for 1990 to 2017 production was no longer
available after 2017. A new, publicly available source was identified and is used for production in 2015-
2022. However, this source does not have sufficient data for the earlier timeseries. EPA confirmed with
experts familiar with the sources that combining these two sources to populate the time series was
accurate (ERG 2019; Lewis 2019) and does not present any significant discontinuities in the time series.
• Brewery production data: the source used for production changed in 2007 to publish craft brewery
production broken out by size but does not include data prior to 2007. Therefore, rather than estimating
total production data prior to 2007 with this source, another data source was used to ensure accuracy of
production data through the time series (ERG 2018b).
Refer to the Recalculations Discussion section below for details on updates implemented to improve accuracy,
consistency and/or completeness of the time series.
Domestic Wastewater CH4 Emission Estimates
Domestic wastewater Cm emissions originate from both septic systems and from centralized treatment systems.
Within these centralized systems, Cm emissions can arise from aerobic systems that liberate dissolved Cm that
formed within the collection system or that are designed to have periods of anaerobic activity (e.g., constructed
wetlands and facultative lagoons), anaerobic systems (anaerobic lagoons and anaerobic reactors), and from
anaerobic sludge digesters when the captured biogas is not completely combusted. Emissions will also result from
the discharge of treated effluent from centralized wastewater plants to waterbodies where carbon accumulates in
sediments (typically slow-moving systems, such as lakes, reservoirs, and estuaries). The systems with emissions
estimates are:
6 IPCC (2019) updates, supplements, and elaborates the 2006 IPCC Guidelines where gaps or out-of-date science have been
identified. EPA used these methodologies to improve completeness and include sources of greenhouse gas emissions that have
not been estimated prior to the 1990 to 2019 Inventory, such as N20 emissions from industrial wastewater treatment, and to
improve emission estimates for other sources, such as emissions from wastewater discharge and centralized wastewater
treatment.
7-22 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
• Septic systems (A);
• Centralized treatment aerobic systems (B), including aerobic systems (other than constructed wetlands)
(Bl), constructed wetlands only (B2), and constructed wetlands used as tertiary treatment (B3);
• Centralized anaerobic systems (C);
• Anaerobic sludge digesters (D); and
• Centralized wastewater treatment effluent (E).
Methodological equations for each of these systems are presented in the subsequent subsections; total domestic
Cm emissions are estimated as follows:
Equation 7-4: Total Domestic CH4 Emissions from Wastewater Treatment and Discharge
Total Domestic CH4 Emissions from Wastewater Treatment and Discharge (kt) = A+ B + C + D + E
Table 7-9 presents domestic wastewater CH4 emissions for both septic and centralized systems, including
anaerobic sludge digesters and emissions from centralized wastewater treatment effluent, in 2022.
Table 7-9: Domestic Wastewater CH4 Emissions from Septic and Centralized Systems (2022,
kt, MMT CO2 Eq. and Percent)
CH.i Emissions (MMT
% of Domestic
CHi Emissions (kt)
COEq.)
Wastewater CH.i
Septic Systems (A)
215
6.0
44.4%
Centrally-Treated Aerobic Systems (B)
77
2.2
15.9%
Centrally-Treated Anaerobic Systems (C)
113
3.2
23.2%
Anaerobic Sludge Digesters (D)
8
0.2
1.7%
Centrally-Treated Wastewater Effluent (E)
72
2.0
14.9%
Total
485
13.6
100%
Notes: Totals may not sum due to independent rounding.
Emissions from Septic Systems:
Methane emissions from septic systems were estimated by multiplying the U.S. population by the percent of
wastewater treated in septic systems (about 16 percent in 2022; U.S. Census Bureau 2021a) and an emission factor
and then converting the result to kt/year. The method was selected in accordance with IPCC methodological
decision trees in based on available data for septic systems.
U.S. population data were taken from historic U.S. Census Bureau national population totals data and include the
populations of the United States and Puerto Rico (U.S. Census Bureau 2002; U.S. Census Bureau 2011; U.S. Census
Bureau 2022 and 2023; Instituto de Estadisticas de Puerto Rico 2021). Population data for American Samoa, Guam,
Northern Mariana Islands, and the U.S. Virgin Islands were taken from the U.S. Census Bureau International
Database (U.S. Census Bureau 2023). Table 7-10 presents the total U.S. population for 1990 through 2022. The
fraction of the U.S. population using septic systems or centralized treatment systems is based on data from the
American Housing Surveys (U.S. Census Bureau 2021a).
Methane emissions for septic systems are estimated as follows:
Waste 7-23
-------
Equation 7-5: CH4 Emissions from Septic Systems
Emissions from Septic Systems (U. S. Specific) = A
= USp0p x (TSBP7-/c) x (EFSBP7-/c) x 1/109 x 365.25
Table 7-10: Variables and Data Sources for CH4 Emissions from Septic Systems
Inventory Years: Source of
Variable
Variable Description
Units
Value
United States and Puerto
Rico:
1990-1999: U.S. Census
Bureau (2002); Instituto de
Estadisticas de Puerto Rico
(2021)
2000-2009: U.S. Census
USpop
U.S. population3
Persons
Bureau (2011)
2010-2019: U.S. Census
Bureau (2021b)
2020-2022: U.S. Census
Bureau (2022)
U.S. Territories other than
Puerto Rico:
1990-2022: U.S. Census
Bureau (2023)
Odd years from 1989 through
2021: U.S. Census Bureau
(2021a)
Tseptic
Percent treated in septic systems3
%
Data for intervening years
obtained by linear
interpolation
2022: Forecasted from the
rest of the time series
EFseptic
Methane emission factor - septic systems
(10.7)
g CH4/capita/day
1990-2022: Leverenz et al.
(2010)
1/109
Conversion factor
g to kt
Standard conversion
365.25
Conversion factor
Days in a year
Standard conversion
a Value of activity data varies over the Inventory time series.
Emissions from Centrally Treated Aerobic and Anaerobic Systems:
Methane emissions from POTWs depend on the total organics in wastewater. Table 7-12 presents the total
organically degradable material in wastewater, or TOW, for 1990 through 2022. The TOW was determined using
BOD generation rates per capita weighted average both with and without kitchen scraps as well as an estimated
percent of housing units that utilize kitchen garbage disposals. Households with garbage disposals (with kitchen
scraps or ground up food scraps) typically have wastewater with higher BOD than households without garbage
disposals due to increased organic matter contributions (ERG 2018a). The equations are as follows:
Equation 7-6: Total Wastewater BOD5 Produced per Capita (U.S.-Specific [ERG 2018a])
BODgenrate (k§/capita/day) HODwlth0ut str,ips x (1 %kitchen disposal) -I- BO scraps x
(%kitchen disposal)
7-24 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
Equation 7-7: Total Organically Degradable Material in Domestic Wastewater (IPCC 2019 [Eq.
6.3])
T0W PfSr) = USP°P X B0Baenrate X 365.25 X ^
Table 7-11: Variables and Data Sources for Organics in Domestic Wastewater
Variable
Variable Description
Units
Inventory Years: Source of
Value
BODgen rate
Total wastewater BOD produced per
capita
kg/capita/day
1990-2022: Calculated
BODwithout scrap
Wastewater BOD produced per capita
without kitchen scraps3
kg/capita/day
1990-2003: Metcalf & Eddy
(2003)
2004-2013: Linear
interpolation
2014-2022: Metcalf & Eddy
(2014)
BODwith scraps
Wastewater BOD produced per capita
with kitchen scraps3
kg/capita/day
% kitchen disposal
Percent of housing units with kitchen
disposal3
%
1990-2013: U.S. Census
Bureau (2013)
2014-2022: Forecasted from
the rest of the time series
TOW
Total wastewater BOD Produced per
Capita3
Gg BOD/year
1990-2022: Calculated, ERG
(2018a)
USpop
U.S. population3
Persons
United States and Puerto
Rico:
1990-1999: U.S. Census
Bureau (2002); Instituto de
Estadisticas de Puerto Rico
(2021)
2000-2009: U.S. Census
Bureau (2011)
2010-2019: U.S. Census
Bureau (2021b)
2020-2022: U.S. Census
Bureau (2022)
U.S. Territories other than
Puerto Rico:
1990-2022: U.S. Census
Bureau (2023)
365.25
Conversion factor
Days in a year
Standard conversion
1/106
Conversion factor
kg to Gg
Standard conversion
a Value of activity data varies over the Inventory time series.
Table 7-12: U.S. Population (Millions) and Domestic Wastewater TOW (kt)
Activity
1990
2005
2018
2019
2020
2021
2022
Population
253
300
330
332
335
336
337
TOW
8,131 I
9,624 1
9,958
10,019
10,132
10,163
10,216
Sources: U.S. Census Bureau (2002); U.S. Census Bureau (2011); U.S. Census Bureau (2021b
and 2022); Instituto de Estadisticas de Puerto Rico (2021); U.S. Census Bureau (2023); ERG
(2018a).
Methane emissions from POTWs were estimated by multiplying the total organics in centrally treated wastewater
(total BODs) produced per capita in the United States by the percent of wastewater treated centrally, or percent
collected (about 84 percent in 2022), the correction factor for additional industrial BOD discharged to the sewer
system, the relative percentage of wastewater treated by aerobic systems (other than constructed wetlands),
Waste 7-25
-------
constructed wetlands only, and anaerobic systems, and the emission factor7 for aerobic systems, constructed
wetlands only, and anaerobic systems. Methane emissions from constructed wetlands used as tertiary treatment
were estimated by multiplying the flow from treatment to constructed wetlands, wastewater BOD concentration
entering tertiary treatment, constructed wetlands emission factor, and then converting to kt/year.
In the United States, the removal of sludge8 from wastewater reduces the biochemical oxygen demand of the
wastewater that undergoes aerobic treatment. The amount of this reduction (S) is estimated using the default IPCC
(2019) methodology and multiplying the amount of sludge removed from wastewater treatment in the United
States by the default factors in IPCC (2019) to estimate the amount of BOD removed based on whether the
treatment system has primary treatment with no anaerobic sludge digestion (assumed to be zero by expert
judgment), primary treatment with anaerobic sludge digestion, or secondary treatment without primary
treatment. The organic component removed from anaerobic wastewater treatment and the amount of Cm
recovered or flared from both aerobic and anaerobic wastewater treatment were set equal to the IPCC default of
zero.
The methodological equations for Cm emissions from aerobic and anaerobic systems are:
Equation 7-8: Total Domestic CH4 Emissions from Centrally Treated Aerobic Systems
Emissions from Centrally Treated Aerobic Systems (other than Constructed Wetlands)(B 1) +
Emissions from Centrally Treated Aerobic Systems (Constructed Wetlands Only)(B2) +
Emissions from Centrally Treated Aerobic Systems (Constructed Wetlands used as Tertiary Treatment) (B3)
B
where,
Equation 7-9: Total Organics in Centralized Wastewater Treatment [IPCC 2019 (Eq. 6.3A)]
TOWcentralized ( ®ear ) = TOW X TCENXRALIZED X IcOLLECTED
Table 7-13: Variables and Data Sources for Organics in Centralized Domestic Wastewater
Variable
Variable Description Units
Inventory Years: Source of Value
Centrally Treated Organics (Gg BOD/year)
TOWcentralized
Total organics in centralized
wastewater treatment3
Gg BOD/year
1990-2022: Calculated
TOW
Total wastewater BOD Produced per
Capita3
Gg
BOD/capita/year
1990-2022: Calculated, ERG (2018a)
Tcentrauzed
Percent collected3
%
1990-2019: U.S. Census Bureau
(2021a)
Data for intervening years obtained
by linear interpolation
2020-2022: Forecasted from the rest
of the time series
IcOLLECTED
Correction factor for additional
industrial BOD discharged (1.25)
No units
1990-2022: IPCC (2019) Eq. 6.3a
7 Emission factors are calculated by multiplying the maximum CH4-producing capacity of domestic wastewater (B0, 0.6 kg
CH4/kg BOD) and the appropriate methane correction factors (MCF) for aerobic (0.03) and anaerobic (0.8) systems (IPCC 2019,
Table 6.3) and constructed wetlands (0.4) (IPCC 2014, Table 6.4).
8 Throughout this document, the term "sludge" refers to the solids separated during the treatment of municipal wastewater.
The definition includes domestic septage. "Biosolids" refers to treated sewage sludge that meets the EPA pollutant and
pathogen requirements for land application and surface disposal.
7-26 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
a Value of this activity data varies over the time series.
Equation 7-10: Organic Component Removed from Aerobic Wastewater Treatment (IPCC
2019 [Eq. 6.3B])
Saerobic (^) = Smass x [(% aerobic-^ x Krem,aerprim) + (% aerobic ^primary x Krem,aernoprim) +
(%aerobic + digestion x Kremaeraigest)] x 1000
Equation 7-11: CH4 Emissions from Centrally Treated Aerobic Systems (other than
Constructed Wetlands) (IPCC 2019 [Eq. 6.1])
Bl(kt CH4/year) = [(TOWcentralized) x (% aerobic0Tcw)~ ^aerobic] ^ ^^aerobic ^aerobic
Table 7-14: Variables and Data Sources for CH4 Emissions from Centrally Treated Aerobic
Systems (Other than Constructed Wetlands)
Variable Variable Description
Units
Inventory Years: Source of Value
Emissions from Centrally Treated Aerobic Systems (Other than Constructed Wetlands) (kt CH i/year)
Saerobic
Organic component removed from
aerobic wastewater treatment3
Gg
BOD/year
1990-2022: Calculated
Smass
Raw sludge removed from wastewater
treatment as dry mass3
Tg dry
weight/year
1988: EPA (1993c); EPA (1999)
1990-1995: Calculated based on
sewage sludge production change
per year EPA (1993c); EPA (1999);
Beecher et al. (2007)
1996: EPA (1999)
2004: Beecher et al. (2007)
Data for intervening years obtained
by linear interpolation
2005-2017: Interpolated
2018: NEBRA (2022), as described in
ERG (2023)
2019-2022: Forecasted from the rest
of the time series.
Methodology for estimating sludge
generated from the U.S. territories
provided in ERG (2023).
% aerobicoTcw
Percent of flow to aerobic systems, other
than wetlands3
%
1990,1991: Set equal to 1992
1992, 1996, 2000, 2004: EPA (1992,
1996, 2000, 2004), respectively
Data for intervening years obtained
by linear interpolation.
2005-2022: Forecasted from the rest
of the time series
% aerobic
w/primary
Percent of aerobic systems with primary
treatment and no anaerobic sludge
digestion (0)
%
% aerobic w/out
primary
Percent of aerobic systems without
primary treatment3
%
%aerobic+digestion
Percent of aerobic systems with primary
and anaerobic sludge digestion3
%
Krem,aer_prim
Sludge removal factor for aerobic
treatment plants with primary treatment
(mixed primary and secondary sludge,
untreated or treated aerobically) (0.8)
kg BOD/kg
sludge
1990-2022: IPCC (2019) Table 6.6a
Krem,aer_noprim
Sludge removal factor for aerobic
wastewater treatment plants without
separate primary treatment (1.16)
kg BOD/kg
sludge
Waste 7-27
-------
Variable
Variable Description
Units
Inventory Years: Source of Value
Krem,aer_digest
Sludge removal factor for aerobic
treatment plants with primary treatment
and anaerobic sludge digestion (mixed
primary and secondary sludge, treated
anaerobically) (1)
kg BOD/kg
sludge
EF aerobic
Emission factor - aerobic systems (0.018)
kg CH4/kg
BOD
1990-2022: IPCC (2019) Table 6.3
Raerobic
Amount CH4 recovered or flared from
aerobic wastewater treatment (0)
kg CH4/year
1990-2022: IPCC (2019) Eq. 6.1
1000
Conversion factor
metric tons
to kilograms
Standard conversion
a Value of this activity data varies over the time series.
Constructed wetlands provide aerobic treatment but also exhibit partially anaerobic conditions; however, they are
referred to in this chapter as aerobic systems. Constructed wetlands may be used as the sole treatment unit at a
centralized wastewater treatment plant or may serve as tertiary treatment after simple settling and biological
treatment. Emissions from all constructed wetland systems were included in the estimates of emissions from
centralized wastewater treatment plant processes and effluent from these plants. Methane emissions equations
from constructed wetlands used as sole treatment were previously described. Methane emissions from
constructed wetlands used as tertiary treatment were estimated by multiplying the flow from treatment to
constructed wetlands, wastewater BOD concentration entering tertiary treatment, constructed wetlands emission
factor, and then converting to kt/year.
For constructed wetlands, an IPCC default emission factor for surface flow wetlands was used. This is the most
conservative factor for constructed wetlands and was recommended by IPCC (2014) when the type of constructed
wetland is not known. A median BODs concentration of 9.1 mg/Lwas used for wastewater entering constructed
wetlands used as tertiary treatment based on U.S. secondary treatment standards for POTWs. This median value is
based on plants generally utilizing simple settling and biological treatment (EPA 2013). Constructed wetlands do
not have secondary sludge removal.
Equation 7-12: CH4 Emissions from Centrally Treated Aerobic Systems (Constructed Wetlands
Only) [IPCC 2014 (Eq. 6.1)]
B2 ( year4) = [(T0Wcenxralized) x (% aerobiccw)] x (EFCW)
Equation 7-13: CH4 Emissions from Centrally Treated Aerobic Systems (Constructed Wetlands
used as Tertiary Treatment) (U.S. Specific)
B3 (tSt)= [(P0TWflowcw) x (B0Dcw,inf)x 3.785 x (EFcw)]x^x 365.25
Table 7-15: Variables and Data Sources for CH4 Emissions from Centrally Treated Aerobic
Systems (Constructed Wetlands)
Variable Variable Description
Units
Inventory Years: Source of Value
Emissions from Constructed Wetlands Only (kt CH.i/year)
TOWcentrauzed
Total organics in centralized
wastewater treatment3
Gg
BOD/year
1990-2022: Calculated
% aerobiccw
Flow to aerobic systems,
constructed wetlands used as sole
treatment / total flow to POTWs.3
%
1990,1991: Set equal to 1992
1992, 1996, 2000, 2004, 2008, 2012:
EPA (1992, 1996, 2000, 2004, 2008,
and 2012)
7-28 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
Variable
Variable Description
Units
Inventory Years: Source of Value
Data for intervening years obtained
by linear interpolation.
2013-2022: Forecasted from the rest
of the time series
EFcw
Emission factor for constructed
wetlands (0.24)
kg CH4/kg
BOD
1990-2022: IPCC (2014)
Emissions from Constructed Wetlands used as Tertiary Treatment (kt CH4/year)
POTW_flow_CW
Wastewater flow to POTWs that
use constructed wetlands as
tertiary treatment3
MGD
1990,1991: Set equal to 1992
1992, 1996, 2000, 2004, 2008, 2012:
EPA (1992, 1996, 2000, 2004, 2008,
and 2012)
Data for intervening years obtained
by linear interpolation.
2013-2022: Forecasted from the rest
of the time series
BODcw.inf
BOD concentration in wastewater
entering the constructed wetland
(9.1)
mg/L
1990-2022: EPA (2013)
3.785
Conversion factor
liters to
gallons
Standard conversion
EFcw
Emission factor for constructed
wetlands (0.24)
kg CH4/kg
BOD
1990-2022: IPCC (2014)
1/106
Conversion factor
kg to kt
Standard conversion
365.25
Conversion factor
Days in a
year
Standard conversion
a Value of this activity data varies over the time series.
Data sources and methodologies for centrally treated anaerobic systems are similar to those described for aerobic
systems, other than constructed wetlands. See discussion above.
Equation 7-14: CH4 Emissions from Centrally Treated Anaerobic Systems (IPCC 2019 (Eq. 6.1])
C ( yeai.4) = [(TOWcenxralized) x (% anaerobic)-
^anaerobic] * ^^anaerobic ^-anaerobic
Table 7-16: Variables and Data Sources for CH4 Emissions from Centrally Treated Anaerobic
Systems
Variable
Variable Description
Units
Inventory Years: Source of Value
Emissions from Centrally Treated Anaerobic Systems (kt CH i/year)
TOWcentrauzed
Total organics in centralized
wastewater treatment3
Gg
BOD/year
1990-2022: Calculated
% anaerobic
Percent centralized wastewater that
is anaerobically treated3
%
1990,1991: Set equal to 1992
1992, 1996, 2000, 2004: EPA
(1992, 1996, 2000, 2004),
respectively
Data for intervening years
obtained by linear interpolation.
2005-2022: Forecasted from the
rest of the time series
Sanaerobic
Organic component removed from
anaerobic wastewater treatment (0)
Gg/year
1990-2022: IPCC (2019) Table 6.3
EF anaerobic
Emission factor for anaerobic
reactors/deep lagoons (0.48)
kg CH4/kg
BOD
Waste 7-29
-------
Variable
Variable Description
Units
Inventory Years: Source of Value
Ranaerobic
Amount CH4 recovered or flared
from anaerobic wastewater
treatment (0)
kg CH4/year
a Value of this activity data varies over the time series.
Emissions from Anaerobic Sludge Digesters:
Total Cm emissions from anaerobic sludge digesters were estimated by multiplying the wastewater influent flow
to POTWs with anaerobic sludge digesters, the cubic feet of digester gas generated per person per day divided by
the flow to POTWs, the fraction of CH4 in biogas, the density of CH4, one minus the destruction efficiency from
burning the biogas in an energy/thermal device and then converting the results to kt/year.
Equation 7-15: CH4 Emissions from Anaerobic Sludge Digesters (U.S. Specific)
D (]Sr) = [(PQTWfl0WAD) x '"f;/6"] X 0.0283 x (FRACCH4) x 365.25 x (662) x (1 - DE) x ^
Table 7-17: Variables and Data Sources for Emissions from Anaerobic Sludge Digesters
Variable
Variable Description
Units
Inventory years: Source of Value
Emissions from Anaerobic Sludge Digesters (kt CH i/year)
POTW_flow_AD
POTW Flow to Facilities with
Anaerobic Sludge Digesters3
MGD
1990,1991: Set equal to 1992
1992, 1996, 2000, 2004: EPA
(1992, 1996, 2000, and 2004),
respectively
Data for intervening years
obtained by linear interpolation.
2005-2022: Forecasted from the
rest of the time series
biogas gen
Gas Generation Rate (1.0)
ft3/capita/day
1990-2022: Metcalf & Eddy (2014)
100
Per Capita POTW Flow (100)
gal/capita/day
1990-2022: Ten-State Standards
(2004)
0.0283
Conversion factor
ft3 to m3
Standard Conversion
FRACCh4
Proportion of Methane in Biogas
(0.65)
No units
1990-2022: Metcalf & Eddy (2014)
365.25
Conversion factor
Days in a year
Standard conversion
662
Density of Methane (662)
g CFU/m3 ch4
1990-2022: EPA (1993a)
DE
Destruction Efficiency (99%
converted to fraction)
No units
1990-2022: EPA (1998); CAR
(2011); Sullivan (2007); Sullivan
(2010); and UNFCCC(2012)
1/109
Conversion factor
g to kt
Standard conversion
a Value of this activity data varies over the time series.
Emissions from Discharge of Centralized Treatment Effluent:
Methane emissions from the discharge of wastewater treatment effluent were estimated by multiplying the total
BOD of the discharged wastewater effluent by an emission factor associated with the location of the discharge.
The BOD in treated effluent was determined by multiplying the total organics in centrally treated wastewater by
the percent of wastewater treated in primary, secondary, and tertiary treatment, and the fraction of organics
remaining after primary treatment (one minus the fraction of organics removed from primary treatment,
secondary treatment, and tertiary treatment).
7-30 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
Equation 7-16: CH4 Emissions from Centrally Treated Systems Discharge (U.S.-Specific)
E (tSt) = (T0WRLE X EFRLE) + (TOWother X EFother)
where,
Equation 7-17: Total Organics in Centralized Treatment Effluent (IPCC 2019 [Eq. 6.3D])
TOWEFFtreatCENXRALIZED ( ®ear )
= [TOWcenxralized x % primary x (l - TOWrem PRIMARY)] + [TOWcenxralized x % secondary x (l —
TOWrem SECOndary
)] + [TOWcenxralized x % tertiary x (l — TOWrem
.tertiary)]
Equation 7-18: Total Organics in Effluent Discharged to Reservoirs, Lakes, or Estuaries (U.S.-
Specific)
TOWrle ( ®ear ) = TOWEFFtreatCENXRALIZED x PercentRLE
Equation 7-19: Total Organics in Effluent Discharged to Other Waterbodies (U.S.-Specific)
TOWother ( year ) = TOWEFFtreatCENXRALIZED x Percentother
Table 7-18: Variables and Data Sources for CH4 Emissions from Centrally Treated Systems
Discharge
Variable
Variable Description
Units
Source of Value
TOW EFFtreat.CENTRALIZED
Total organics in centralized treatment effluent3
Gg
BOD/year
1990-2022:
Calculated
TOWcentrauzed
Total organics in centralized wastewater treatment3
Gg
BOD/year
1990-2022:
Calculated
% primary
Percent of primary domestic centralized treatment3
%
1990, 1991: Set
equal to 1992.
1992, 1996, 2000,
2004, 2008, 2012:
EPA (1992, 1996,
2000, 2004, 2008,
and 2012),
respectively
Data for
intervening years
obtained by linear
interpolation.
2013-2022:
Forecasted from
the rest of the time
series
% secondary
Percent of secondary domestic centralized treatment3
%
% tertiary
Percent of tertiary domestic centralized treatment3
%
TOWrem.PRIMARY
Fraction of organics removed from primary domestic
centralized treatment (0.4)
No units
1990-2022: IPCC
(2019) Table 6.6B
TOWrem.SECONDARY
Fraction of organics removed from secondary domestic
centralized treatment (0.85)
No units
TOWrem.TERTIARY
Fraction of organics removed from tertiary domestic
centralized treatment (0.90)
No units
TOWrle
Total organics in effluent discharged to reservoirs, lakes, and
estuaries3
Gg
BOD/year
1990-2022:
Calculated
Waste 7-31
-------
Variable
Variable Description
Units
Source of Value
TOWother
Total organics in effluent discharge to other waterbodies3
Gg
BOD/year
EFrle
Emission factor (discharge to reservoirs/lakes/estuaries)
(0.114)
kg CH4/kg
BOD
1990-2022: IPCC
(2019) Table 6.8
EFother
Emission factor (discharge to other waterbodies) (0.021)
kg CH4/kg
BOD
PercentRLE
% discharged to reservoirs, lakes, and estuaries3
%
1990-2010: Set
equal to 2010
2010: ERG (2021a)
2011: Obtained by
linear interpolation
2012: ERG (2021a)
2013-2022: Set
equal to 2012
Percenter
% discharged to other waterbodies3
%
a Value of this activity data varies over the time series.
Industrial Wastewater CH4 Emission Estimates
Industrial wastewater Cm emissions originate from on-site treatment systems, typically comprised of biological
treatment operations. The collection systems at an industrial plant are not as extensive as domestic wastewater
sewer systems; therefore, it is not expected that dissolved Cm will form during collection. However, some
treatment systems are designed to have anaerobic activity (e.g., anaerobic reactors or lagoons), or may
periodically have anaerobic conditions form (facultative lagoons or large stabilization basins). Emissions will also
result from discharge of treated effluent to waterbodies where carbon accumulates in sediments (typically slow-
moving systems, such as lakes, reservoirs, and estuaries).
Industry categories that are likely to produce significant Cm emissions from wastewater treatment were identified
and included in the Inventory. The main criteria used to identify U.S. industries likely to generate Cm from
wastewater treatment are whether an industry generates high volumes of wastewater, whether there is a high
organic wastewater load, and whether the wastewater is treated using methods that result in Cm emissions. The
top six industries that meet these criteria are pulp and paper manufacturing; meat and poultry processing;
vegetables, fruits, and juices processing; starch-based ethanol production; petroleum refining; and breweries.
Wastewater treatment and discharge emissions for these sectors for 2022 are displayed in Table 7-19 below.
Further discussion of wastewater treatment for each industry is included below.
Table 7-19: Total Industrial Wastewater CH4 Emissions by Sector (2022, MMT CO2 Eq. and
Percent)
CH.i Emissions
% of Industrial
Industry
(MMTCO. Eq.)
Wastewater CH.i
Meat & Poultry
5.7
79.0%
Pulp & Paper
0.8
11.6%
Fruit & Vegetables
0.2
3.3%
Ethanol Refineries
0.2
2.3%
Breweries
0.1
2.0%
Petroleum Refineries
0.1
1.7%
Total
7.2
100%
Note: Totals may not sum due to independent rounding.
7-32 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
Emissions from Industrial Wastewater Treatment Systems:
Equation 7-20 presents the general IPCC equation (Equation 6.4, IPCC 2019) to estimate CFU emissions from each
type of treatment system used for each industrial category.
Equation 7-20: Total CH4 Emissions from Industrial Wastewater
CH4 (industrial sector) = [(TOW; - Sj) x EF — RJ
where,
Cm (industrial sector) = Total CH4 emissions from industrial sector wastewater treatment (kg/year)
I = Industrial sector
TOWi = Total organics in wastewater for industrial sector / (kg COD/year)
Si = Organic component removed from aerobic wastewater treatment for industrial
sector / (kg COD/year)
EF = System-specific emission factor (kg Cm/kg COD)
Ri = Methane recovered for industrial sector / (kg Cm/year)
Equation 7-21 presents the general IPCC equation to estimate the total organics in wastewater (TOW) for each
industrial category.
Equation 7-21: TOW in Industry Wastewater Treatment Systems
TOWj = P;xWjX CODj
where,
TOWi = Total organically degradable material in wastewater for industry / (kg COD/yr)
i = Industrial sector
Pi = Total industrial product for industrial sector / (t/yr)
Wi = Wastewater outflow (m3/t product)
CODi = Chemical oxygen demand (industrial degradable organic component in wastewater) (kg
COD/m3)
The annual industry production is shown in Table 7-20, and the average wastewater outflow and the organics
loading in the outflow is shown in Table 7-21.
For some industries, U.S.-specific data on organics loading is reported as BOD rather than COD. In those cases, an
industry-specific COD:BOD ratio is used to convert the organics loading to COD.
The amount of organics treated in each type of wastewater treatment system was determined using the percent of
wastewater in the industry that is treated on site and whether the treatment system is anaerobic, aerobic or
partially anaerobic. Table 7-22 presents the industrial wastewater treatment activity data used in the calculations
and described in detail in ERG (2008a), ERG (2013a), ERG (2013b), and ERG (2021a). For CH4 emissions, wastewater
treated in anaerobic lagoons or reactors was categorized as "anaerobic", wastewater treated in aerated
stabilization basins or facultative lagoons were classified as "ASB" (meaning there may be pockets of anaerobic
activity), and wastewater treated in aerobic systems such as activated sludge systems were classified as
"aerobic/other."
The amount of organic component removed from aerobic wastewater treatment as a result of sludge removal
(Saerobic) was either estimated as an industry-specific percent removal, if available, or as an estimate of sludge
produced by the treatment system and IPCC default factors for the amount of organic component removed (Krem),
using one of the following equations. Table 7-23 presents the sludge variables used for industries with aerobic
wastewater treatment operations (i.e., pulp and paper, fruit/vegetable processing, and petroleum refining).
Waste 7-33
-------
Equation 7-22: Organic Component Removed from Aerobic Wastewater Treatment - Pulp,
Paper, and Paperboard
Spuip.ash = TOWpUip x % removal w/primary
where,
Spuip.asb = Organic component removed from pulp and paper wastewater during primary
treatment before treatment in aerated stabilization basins (Gg COD/yr)
TOWpuip = Total organically degradable material in pulp and paper wastewater (Gg
COD/yr)
% removal w/primary = Percent reduction of organics in pulp and paper wastewater associated with
sludge removal from primary treatment (%)
Equation 7-23: Organic Component Removed from Aerobic Treatment Plants
c _ c w 17 w 1 r\—6
^aerobic Jmass Rrem -L*J
where,
Saerobic = Organic component removed from fruit and vegetable or petroleum refining wastewater
during primary treatment before treatment in aerated stabilization basins (Gg COD/yr)
Smass = Raw sludge removed from wastewater treatment as dry mass (kg sludge/yr)
Krem = Sludge factor (kg BOD/kg sludge)
10"6 = Conversion factor, kilograms to Gigagrams
Equation 7-24: Raw Sludge Removed from Wastewater Treatment as Dry Mass
Smass = (Sprim + Saer) X P X W
where,
Smass = Raw sludge removed from wastewater treatment as dry mass (kg sludge/yr)
Sprim = Sludge production from primary sedimentation (kg sludge/m3)
Saer = Sludge production from secondary aerobic treatment (kg sludge/m3)
P = Production (t/yr)
W = Wastewater outflow (m3/t)
Default emission factors9 from IPCC (2019) were used. Information on methane recovery operations varied by
industry. See industry descriptions below.
9 Emission factors are calculated by multiplying the maximum CH4-producing capacity of wastewater (B0, 0.25 kg CH4/kg COD)
and the appropriate methane correction factors (MCF) for aerobic (0), partially anaerobic (0.2), and anaerobic (0.8) systems
(IPCC 2019), Table 6.3.
7-34 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
Table 7-20: U.S. Pulp and Paper, Meat, Poultry, Vegetables, Fruits and Juices, Ethanol,
Breweries, and Petroleum Refining Production (MMT)
Activity
1990
2005
2018
2019
2020
2021
2022
Pulp and Paper3
83.6
92.4
78.7
76.3
74.7
75.5
73.9
Meat (Live Weight Killed)
27.3 I
31-4 i:
36.4
37.4
37.8
38.1
37.9
Poultry (Live Weight Killed)
14.6
25.1
29.4
30.1
30.5
30.5
31.1
Vegetables, Fruits and Juices
40.8 »
45.3 !
42.3
41.8
40.6
39.2
38.4
Ethanol Production
2.5
11.7
48.1
47.1
41.6
44.8
48.3
Breweries
23.9 i
23.1
21.5
21.1
21.1
21.2
21.6
Petroleum Refining
702.4
818.6
951.7
940
806.5
858.3
892.1
a Pulp and paper production is the sum of market pulp production plus paper and paperboard production.
Sources: Pulp and Paper - FAO (2023a) and FAO (2023b); Meat, Poultry, and Fruits and Vegetables - USDA (2023a,2023b,
2023c, 2023d, 2022a, and 2022b), ERG (2023); Ethanol - Cooper (2018) and RFA (2023a and 2023b); Breweries - Beer
Institute (2011) and TTB (2022); Petroleum Refining - EIA (2023).
Table 7-21: U.S. Industrial Wastewater Characteristics Data (2022)
Industry
Wastewater
Wastewater
Wastewater
Outflow (m'/ton)
BOD (g/L)
COD (kg/m')
COD:BOD Ratio
Pulp and Paper
See Table 7-25
0.3
-
2.5
Meat Processing
5.3
2.8
-
3
Poultry Processing
12.5
1.5
-
3
Fruit/Vegetable Processing
See Table 7-26
-
1.5
Ethanol Production - Wet Mill
10a
1.5
-
2
Ethanol Production - Dry Mill
1.25a
3b
-
2
Petroleum Refining
0.8
-
0.45
2.5
Breweries - Craft
3.21
-
17.6
1.67
Breweries - NonCraft
1.69
-
17.6
1.67
a Units are gallons per gallons ethanol produced.
b Units are COD (g/L).
Sources: Pulp and Paper (BOD, COD:BOD) - Malmberg (2018); Meat and Poultry (Outflow, BOD) - ERG (2006a); Meat and
Poultry (COD:BOD) - EPA (1997a); Fruit/Vegetables (Outflow, BOD) - CAST (1995), EPA (1974), EPA (1975); Fruit/Vegetables
(COD:BOD) - EPA (1997a); Ethanol Production - Wet Mill (Outflow) - Donovan (1996), NRBP (2001), Ruocco (2006b);
Ethanol Production - Wet Mill (BOD) - White and Johnson (2003); Ethanol Production - Dry Mill (Outflow and COD) -
Merrick (1998), Ruocco (2006a); Ethanol Production (Dry and Wet, COD:BOD) - EPA (1997a); Petroleum Refining (Outflow)
- ERG (2013b); Petroleum Refining (COD) - Benyahia et al. (2006); Petroleum Refining (COD:BOD) - EPA (1982); Breweries -
Craft BIER (2021); ERG (2018b); Breweries - NonCraft ERG (2018b); Brewers Association (2016a); Breweries (Craft and
NonCraft; COD and COD:BOD) - Brewers Association (2016b).
Waste 7-35
-------
Table 7-22: U.S. Industrial Wastewater Treatment Activity Data
Industry
% Wastewater
Treated On-Site
% Treated
Anaerobically
% Treated
Aerobically
% Treated Aerobically
% Treated in
ASBs
% Treated in
Other Aerobic
Pulp and Paperb
60
5.2
75.9
38.5
37.4
Meat Processing
33
33a
33
0
33
Poultry Processing
25
25a
25
0
25
Fruit/Vegetable
Processing
11
0
11
5.5
5.5
Ethanol Production -
Wet Mill
33.3
33.3
66.7
0
0
Ethanol Production -
Dry Mill
75
75
25
0
0
Petroleum Refining
62.1
0
62.1
23.6
38.5
Breweries-Craft
0.5
0.5
0
0
0
Breweries - NonCraft
100
99
1
0
1
a Wastewater is pretreated in anaerobic lagoons prior to aerobic treatment.
b Remaining onsite treated in other treatment assumed to be non-emissive and not shown here.
Note: Due to differences in data availability and methodology, zero values in the table are for calculation purposes only and
may indicate unavailable data.
Sources: ERG (2008a, 2008b); ERG (2013a); ERG (2013b); ERG (2021a).
Table 7-23: Sludge Variables for Aerobic Treatment Systems
Industry
Variable
Pulp and
Paper
Fruit/Vegetable
Processing
Petroleum
Refining
Organic reduction associated with sludge removal (%)
Sludge Production (kg/m3)
Primary Sedimentation
Aerobic Treatment
Sludge Factor (kg BOD/kg dry mass sludge)
Aerobic Treatment w/Primary Sedimentation and No Anaerobic
Sludge Digestion
Aerobic Treatment w/out Primary Sedimentation
58
0.15
0.096
0.E
0.096
1.16
Sources: Organic reduction (pulp) - ERG (2008a); Sludge production - Metcalf & Eddy (2003); Sludge factors - IPCC (2019),
Table 6.6a.
Emissions from Discharge of Industrial Wastewater Treatment Effluent:
Methane emissions from discharge of industrial wastewater treatment effluent are estimated via a Tier 1 method
for all industries except for pulp, paper, and paperboard in accordance with IPCC methodological decision trees in
based on available data for treatment and discharge. Emissions from discharge of pulp, paper, and paperboard
treatment effluent is estimated via a Tier 2 method and is described in the industry-specific data section. Tier 1
emissions from effluent are estimated by multiplying the total organic content of the discharged wastewater
effluent by an emission factor associated with the discharge:
Equation 7-25: CH4 Emissions from Industrial Wastewater Treatment Discharge
CH4 EffluentIND = T0WEFFLUENXIND x EFeffluenx
where,
Cm EffluentiND = CH4 emissions from industrial wastewater discharge for inventory year (kg Cm/year)
7-36 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
TOWeffluent.ind = Total organically degradable material in wastewater effluent from industry for inventory
year (kg COD/year or kg BOD/year)
EFeffluent = Tier 1 emission factor for wastewater discharged to aquatic environments (0.028 kg
Cm/kg COD or 0.068 kg CH4/kg BOD) (IPCC 2019)
The COD or BOD in industrial treated effluent (TOWeffluent.ind) was determined by multiplying the total organics in
the industry's untreated wastewater that is treated on site by an industry-specific percent removal where available
or a more general percent removal based on biological treatment for other industries. Table 7-22 presents the
percent of wastewater treated onsite, while Table 7-24 presents the fraction of TOW removed during treatment.
Equation 7-26: TOW in Industrial Wastewater Effluent
TOWEFFLUent,ind = T0Wind x % onsite x (1 — T0WREM)
where,
TOWeffluent.ind = Total organically degradable material in wastewater effluent from industry for inventory
year (kg COD/year or kg BOD/year)
TOWind = Total organics in untreated wastewater for industry for inventory year (kg COD/year)
%onsite = Percent of industry wastewater treated on site (%)
TOWrem = Fraction of organics removed during treatment
Table 7-24: Fraction of TOW Removed During Treatment by Industry
Industry
TOWrem
Source
Pulp, Paper, and Paperboard
Red Meat and Poultry
Fruits and Vegetables
Ethanol Production
Biomethanator T reatment
Other Treatment
Petroleum Refining
Breweries
0.91 Malmberg (2018)
0.85 IPCC (2019), Table 6.6b
0.85 IPCC (2019), Table 6.6b
0.90 ERG (2008a), ERG (2006b)
0.85 IPCC (2019), Table 6.6b
0.93 Kenari, Sarrafzadeh, and Tavakoli (2010)
0.85 IPCC (2019), Table 6.6b
Discussion of Industry-Specific Data:
Pulp, Paper, and Paperboard Manufacturing Wastewater Treatment. Wastewater treatment for the pulp, paper,
and paperboard manufacturing (hereinafter referred to as "pulp and paper") industry typically includes
neutralization, screening, sedimentation, and flotation/hydrocycloning to remove solids (World Bank 1999;
Nemerow and Dasgupta 1991). Secondary treatment (storage, settling, and biological treatment) mainly consists of
lagooning. About 60 percent of pulp and paper mills have on-site treatment with primary treatment and about half
of these also have secondary treatment (ERG 2008). In the United States, primary treatment is focused on solids
removal, equalization, neutralization, and color reduction (EPA 1993b). The vast majority of pulp and paper mills
with on-site treatment systems use mechanical clarifiers to remove suspended solids from the wastewater. About
10 percent of pulp and paper mills with treatment systems use settling ponds for primary treatment and these are
more likely to be located at mills that do not perform secondary treatment (EPA 1993b).
Approximately 42 percent of the BOD passes on to secondary treatment, which consists of activated sludge,
aerated stabilization basins, or non-aerated stabilization basins. Pulp and paper mill wastewater treated using
anaerobic ponds or lagoons or unaerated ponds were classified as anaerobic (with an MCF of 0.8). Wastewater
flow treated in systems with aerated stabilization basins or facultative lagoons was classified as partially anaerobic
(with an MCF of 0.2, which is the 2006 IPCC Guidelines-suggested MCF for shallow lagoons). Wastewater flow
treated in systems with activated sludge systems or similarly aerated biological systems was classified as aerobic.
A time series of CFU emissions for 1990 through 2022 was developed based on paper and paperboard production
data and market pulp production data. Market pulp production values were available directly for 1998, 2000
Waste 7-37
-------
through 2003, and 2010 through 2021. Where market pulp data were unavailable, a percent of woodpulp that is
market pulp was applied to woodpulp production values from FAOSTAT to estimate market pulp production (FAO
2023a). The percent of woodpulp that is market pulp for 1990 to 1997 was assumed to be the same as 1998,1999
was interpolated between values for 1998 and 2000, 2000 through 2009 were interpolated between values for
2003 and 2010, and 2022 was forecasted from the rest of the time series. A time series of the overall wastewater
outflow in units of cubic meters of wastewater per ton of total production (i.e., market pulp plus woodpulp) is
presented in Table 7-25. Data for 1990 through 1994 varies based on data outlined in ERG (2013a) to reflect
historical wastewater flow. Wastewater generation rates for 1995, 2000, and 2002 were estimated from the 2014
American Forest and Paper Association (AF&PA) Sustainability Report (AF&PA 2014). Wastewater generation rates
for 2004, 2006, 2008, 2010, 2012, and 2014 were estimated from the 2016 AF&PA Sustainability Report (AF&PA
2016). Data for 2005 and 2016 were obtained from the 2018 AF&PA Sustainability Report (AF&PA 2018), data for
2018 were obtained from the 2020 AF&PA Sustainability Report (AF&PA 2020), and data for 2020 were obtained
from a 2022 AF&PA sustainability update (AF&PA 2022). Data for intervening years were obtained by linear
interpolation, while 2021-2022 were set equal to 2020. The average BOD concentration in raw wastewater was
estimated to be 0.4 grams BOD/liter for 1990 to 1998, while 0.3 grams BOD/liter was estimated for 2014 through
2022 (EPA 1997b; EPA 1993b; World Bank 1999; Malmberg 2018). Data for intervening years were obtained by
linear interpolation.
Table 7-25: Wastewater Outflow (m3/ton) for Pulp, Paper, and Paperboard Mills
Wastewater Outflow (m '/ton)
1990
2005
2018
2019
2020
2021
2022
Pulp and Paper
68
43 i
40
39
39
39
39
Sources: ERG (2013a), AF&PA (2014), AF&PA (2016), AF&PA (2018), AF&PA (2020); AF&PA (2022)
Pulp, Paper, and Paperboard Wastewater Treatment Effluent. Methane emissions from pulp, paper, and
paperboard wastewater treatment effluent were estimated by multiplying the total BOD of the discharged
wastewater effluent by an emission factor associated with the location of the discharge.
Equation 7-27: CH4 Emissions from Pulp and Paper Discharge (U.S. Specific)
Emissions from Pulp and Paper Discharge (u. S. Specific,
= (T0WRLEpulp X EFrle) + (TOWother,pulp X EFother)
Equation 7-28: Total Organics in Pulp and Paper Effluent Discharged to Reservoirs, Lakes, Or
Estuaries (U.S. Specific)
T0WRLE pulp ( 8 ) - T0WEFFLUENXIND x PercentRLE
year
pulp
Equation 7-29: Total Organics in Pulp and Paper Effluent Discharged to Other Waterbodies
(U.S. Specific)
TOW,
Other,pulp ( year ) TOWEffluENT,IND a reIut:ULOther,pulp
x Percentr
where,
TOWrle.puIp
EFrle
TOWother,pulp
E Fother
Total organics in pulp, paper, and paperboard manufacturing wastewater treatment
effluent discharged to reservoirs, lakes, or estuaries (Gg BOD/year)
Emission factor (discharge to reservoirs/lakes/estuaries) (0.114 kg Cm/kg BOD) (IPCC
2019)
Total organics in pulp, paper, and paperboard manufacturing wastewater treatment
effluent discharged to other waterbodies (Gg BOD/year)
Emission factor (discharge to other waterbodies) (0.021 kg Cm/kg BOD) (IPCC 2019)
7-38 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
TOWeffluent.ind = Total organically degradable material in pulp, paper, and paperboard manufacturing
wastewater effluent for inventory year (Gg BOD/year)
PercentRLE.puip = Percent of wastewater effluent discharged to reservoirs, lakes, and estuaries (ERG
2021b)
Percentother.puip = Percent of wastewater effluent discharged to other waterbodies (ERG 2021b)
The percent of pulp, paper, and paperboard wastewater treatment effluent routed to reservoirs, lakes, or
estuaries (3 percent) and other waterbodies (97 percent) were obtained from discussions with NCASI (ERG 2021b).
Data for 2019 were assumed the same as the rest of the time series due to lack of available data. Default emission
factors for reservoirs, lakes, and estuaries (0.114 kg Cm/kg BOD) and other waterbodies (0.021 kg Cm/kg BOD)
were obtained from IPCC (2019).
Meat and Poultry Processing. The meat and poultry processing industry makes extensive use of anaerobic lagoons
in sequence with screening, fat traps, and dissolved air flotation when treating wastewater on site. Although all
meat and poultry processing facilities conduct some sort of treatment on site, about 33 percent of meat processing
operations (EPA 2002) and 25 percent of poultry processing operations (U.S. Poultry 2006) perform on-site
treatment in anaerobic lagoons. The IPCC default emission factor of 0.2 kg Cm/kg COD for anaerobic lagoons were
used to estimate the Cm produced from these on-site treatment systems.
Vegetables, Fruits, and Juices Processing. Treatment of wastewater from fruits, vegetables, and juices processing
includes screening, coagulation/settling, and biological treatment (lagooning). The flows are frequently seasonal,
and robust treatment systems are preferred for on-site treatment. About half of the operations that treat and
discharge wastewater use lagoons intended for aerobic operation, but the large seasonal loadings may develop
limited anaerobic zones. In addition, some anaerobic lagoons may also be used (Nemerow and Dasgupta 1991).
Wastewater treated in partially anaerobic systems were assigned the IPCC default emission factor of 0.12 kg
Cm/kg BOD. Outflow and BOD data, presented in Table 7-26, were obtained from CAST (1995) for apples, apricots,
asparagus, broccoli, carrots, cauliflower, cucumbers (for pickles), green peas, pineapples, snap beans, and spinach;
EPA (1974) for potato and citrus fruit processing; and EPA (1975) for all other commodities.
Table 7-26: Wastewater Outflow (m3/ton) and BOD Production (g/L) for U.S. Vegetables,
Fruits, and Juices Production
Wastewater Outflow Organic Content in Untreated
Commodity (m /ton) Wastewater (g BOD/L)
Vegetables
Potatoes 10.27
Other Vegetables 9.85
Fruit
Apples 9.08
Citrus Fruits 10.11
Non-citrus Fruits 12.59
Grapes (for wine) 2.78
Sources: CAST (1995); EPA (1974); EPA (1975).
Ethanol Production. Ethanol, or ethyl alcohol, is produced primarily for use as a fuel component, but is also used in
industrial applications and in the manufacture of beverage alcohol. Ethanol can be produced from the
fermentation of sugar-based feedstocks (e.g., molasses and beets), starch- or grain-based feedstocks (e.g., corn,
sorghum, and beverage waste), and cellulosic biomass feedstocks (e.g., agricultural wastes, wood, and bagasse).
Ethanol can also be produced synthetically from ethylene or hydrogen and carbon monoxide. However, synthetic
ethanol comprises a very small percent of ethanol production in the United States. Currently, ethanol is mostly
made from sugar and starch crops, but with advances in technology, cellulosic biomass is increasingly used as
ethanol feedstock (DOE 2013).
1.765
0.751
8.16
0.317
1.226
1.831
Waste 7-39
-------
Ethanol is produced from corn (or other sugar or starch-based feedstocks) primarily by two methods: wet milling
and dry milling. Historically, the majority of ethanol was produced by the wet milling process, but now the majority
is produced by the dry milling process. The dry milling process is cheaper to implement and is more efficient in
terms of actual ethanol production (Rendleman and Shapouri 2007). The wastewater generated at ethanol
production facilities is handled in a variety of ways. Dry milling facilities often combine the resulting evaporator
condensate with other process wastewaters, such as equipment wash water, scrubber water, and boiler blowdown
and anaerobically treat this wastewater using various types of digesters. Wet milling facilities often treat their
steepwater condensate in anaerobic systems followed by aerobic polishing systems. Wet milling facilities may treat
the stillage (or processed stillage) from the ethanol fermentation/distillation process separately or together with
steepwater and/or wash water. Methane generated in anaerobic sludge digesters is commonly collected and
either flared or used as fuel in the ethanol production process (ERG 2006b).
About 33 percent of wet milling facilities and 75 percent of dry milling facilities treat their wastewater
anaerobically (ERG 2006b). A default emission factor of 0.2 kg Cm/kg COD for anaerobic treatment was used to
estimate the Cm produced from these on-site treatment systems. The amount of Cm recovered through the use
of biomethanators was estimated, and a 99 percent destruction efficiency was used. Biomethanators are
anaerobic reactors that use microorganisms under anaerobic conditions to reduce COD and organic acids and
recover biogas from wastewater (ERG 2006b). For facilities using biomethanators, approximately 90 percent of
BOD is removed during on-site treatment (ERG 2006b, 2008a). For all other facilities, the removal of organics was
assumed to be equivalent to secondary treatment systems, or 85 percent (IPCC 2019).
Petroleum Refining. Petroleum refining wastewater treatment operations have the potential to produce Cm
emissions from anaerobic wastewater treatment. EPA's Office of Air and Radiation performed an Information
Collection Request (ICR) for petroleum refineries in 2011.10 Facilities that reported using non-aerated surface
impoundments or other biological treatment units (trickling filter, rotating biological contactor), which have the
potential to lead to anaerobic conditions, were assigned the IPCC default emission factor of 0.05 kg Cm/kg COD. In
addition, the wastewater generation rate was determined to be 26.4 gallons per barrel of finished product, or 0.8
m3/ton (ERG 2013b).
Breweries. Since 2010, the number of breweries has increased from less than 2,000 to more than 8,000 (Brewers
Association 2021). This increase has primarily been driven by craft breweries, which have increased by over 250
percent during that period. Craft breweries were defined as breweries producing less than six million barrels of
beer per year, and non-craft breweries produce greater than six million barrels. With their large amount of water
use and high strength wastewater, breweries generate considerable CFU emissions from anaerobic wastewater
treatment. However, because many breweries recover their CH4, their emissions are much lower.
The Alcohol and Tobacco Tax and Trade Bureau (TTB) provides total beer production in barrels per year for
different facility size categories from 2007 to 2021 (TTB 2022). Because data were unavailable for 2022, EPA
extrapolated from 1990 to 2021 values. For years prior to 2007 where TTB data were not readily available, the
Brewers Almanac (Beer Institute 2011) was used, along with an estimated percent of craft and non-craft breweries
based on the breakdown of craft and non-craft for the years 2007 through 2020.
To determine the overall amount of wastewater produced, data on water use per unit of production and a
wastewater-to-water ratio were used from the Benchmarking Report (Brewers Association 2016a) for both craft
and non-craft breweries. Since brewing is a batch process, and different operations have varying organic loads,
full-strength brewery wastewater can vary widely on a day-to-day basis. However, the organic content of brewery
wastewater does not substantially change between craft and non-craft breweries. Some breweries may collect and
discharge high strength wastewater from particular brewing processes (known as "side streaming") to a POTW,
10 Available online at https://www.epa.Eov/stationarv-sources-air-pollution/comprehensive-data-collected-petroleum-refining-
sector.
7-40 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
greatly reducing the organics content of the wastewater that is treated on site. Subsequently, the MCF for
discharge to a POTW was assumed to be zero (ERG 2018b).
Breweries may treat some or all of their wastewater on site prior to discharge to a POTW or receiving water. On-
site treatment operations can include physical treatment (e.g., screening, settling) which are not expected to
contribute to Cm emissions, or biological treatment, which may include aerobic treatment or pretreatment in
anaerobic reactors (ERG 2018b). The IPCC default emission factor of 0.2 kg Cm/kg COD for anaerobic treatment
and 0 for aerobic treatment were used to estimate the Cm produced from these on-site treatment systems (IPCC
2006). The amount of Cm recovered through anaerobic wastewater treatment was estimated, and a 99 percent
destruction efficiency was used (ERG 2018b; Stier J. 2018). Very limited activity data are available on the number
of U.S. breweries that are performing side streaming or pretreatment of wastewater prior to discharge.
Domestic Wastewater N2O Emission Estimates
Domestic wastewater N2O emissions originate from both septic systems and POTWs. Within these centralized
systems, N2O emissions can result from aerobic systems, including systems like constructed wetlands. Emissions
will also result from discharge of centrally treated wastewater to waterbodies with nutrient-impacted/eutrophic
conditions. The systems with emission estimates are:
• Septic systems (A);
• Centralized treatment aerobic systems (B), including aerobic systems (other than constructed wetlands)
(Bl), constructed wetlands only (B2), and constructed wetlands used as tertiary treatment (B3);
• Centralized anaerobic systems (C); and
• Centralized wastewater treatment effluent (D).
Methodological equations for each of these systems are presented in the subsequent subsections; total domestic
N2O emissions are estimated as follows:
Equation 7-30: Total Domestic N2O Emissions from Wastewater Treatment and Discharge
Total Domestic N2O Emissions from Wastewater Treatment and Discharge (kt) = A+ B + C + D
Table 7-27 presents domestic wastewater N2O emissions for both septic and centralized systems, including
emissions from centralized wastewater treatment effluent, in 2022.
Table 7-27: Domestic Wastewater N2O Emissions from Septic and Centralized Systems (2022,
kt, MMT CO2 Eq. and Percent)
N'O Emissions (kt)
N'O Emissions
(MMT CO . Eq.)
% of Domestic
Wastewater N O
Septic Systems
3
0.8
3.6%
Centrally-Treated Aerobic Systems
61
16.3
76.0%
Centrally-Treated Anaerobic Systems
+
+
+
Centrally-Treated Wastewater Effluent
16
4.4
20.4%
Total
81
21.4
100%
+ Does not exceed 0.5 kt, 0.05 MMT C02 Eq., or 0.5 percent.
Note: Totals may not sum due to independent rounding.
Emissions from Septic Systems:
Nitrous oxide emissions from domestic treatment depend on the nitrogen present, in this case, in the form of
protein. Per capita protein consumption (kg protein/person/year) was determined by multiplying per capita annual
food availability data and its protein content. Those data are then adjusted using a factor to account for the
fraction of protein actually consumed. The methodological equations are:
Waste 7-41
-------
Equation 7-31: Annual per Capita Protein Supply (U.S. Specific)
ProteinSUPPLY (kg/person/year) = Proteinpercapita/1000 x 365.25
Equation 7-32: Consumed Protein (IPCC 2019 [Eq. 6.10A])
Protein (kg/person/year) = ProteinSUPPLY x FPC
Table 7-28: Variables and Data Sources for Protein Consumed
Variable
Variable Description
Units
Inventory Years: Source of
Value
Protein
ProteinsuppLY
Annual per capita protein supply3
kg/person/year
1990-2022: Calculated
PrOteirlper capita
Daily per capita protein supply3
g/person/day
1990-2022: USDA (2015)
1000
Conversion factor
gto kg
Standard conversion
365.25
Conversion factor
Days in a year
Standard conversion
FPC
Fraction of Protein Consumed3
kg protein
consumed / kg
protein available
1990-2010: USDA (2015)
2011-2020: FAO (2022) and
scaling factor
2021-2022: Forecasted
from the rest of the time
series
a Value of this activity data varies over the Inventory time series.
Nitrous oxide emissions from septic systems were estimated by multiplying the U.S. population by the percent of
wastewater treated in septic systems (about 16 percent in 2022; U.S. Census Bureau 2019), consumed protein per
capita (kg protein/person/year), the fraction of N in protein, the correction factor for additional nitrogen from
household products, the factor for industrial and commercial co-discharged protein into septic systems, the factor
for non-consumed protein added to wastewater and an emission factor and then converting the result to kt/year.
The method selected is in accordance with IPCC methodological decision trees and available data. All factors were
obtained from IPCC (2019).
U.S. population data were taken from historic U.S. Census Bureau national population totals data and include the
populations of the United States and Puerto Rico (U.S. Census Bureau 2002; U.S. Census Bureau 2011; U.S. Census
Bureau 2021b and 2022, Instituto de Estadisticas de Puerto Rico 2021). Population data for American Samoa,
Guam, Northern Mariana Islands, and the U.S. Virgin Islands were taken from the U.S. Census Bureau International
Database (U.S. Census Bureau 2023). Table 7-12 presents the total U.S. population for 1990 through 2022. The
fraction of the U.S. population using septic systems, as well as centralized treatment systems (see below), is based
on data from American Housing Survey (U.S. Census Bureau 2021a). The methodological equations are:
Equation 7-33: Total Nitrogen Entering Septic Systems (IPCC 2019 [Eq. 6.10])
TNDOmseptic
= (USP0P x Tsepxic) X Protein X FNPR X NHH X FNON-CON_septic X FlND-COM_septic
7-42 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
Equation 7-34: N2O Emissions from Septic Systems (IPCC 2019 [Eq. 6.9])
A ( —) = TNDOm_septic x (EFseptic) x 44/28 x 1/106
Table 7-29: Variables and Data Sources for N2O Emissions from Septic System
Variable
Variable Description
Units
Inventory Years: Source of
Value
Emissions from Septic Systems
TNdom septic
Total nitrogen entering septic systems
kg N/year
1990-2022: Calculated
USpop
U.S. population3
Persons
United States and Puerto Rico:
1990-1999: U.S. Census Bureau
2002; Instituto de Estadisticas
de Puerto Rico 2021
2000-2009: U.S. Census Bureau
2011
2010-2019: U.S. Census Bureau
(2021b)
2020-2022: U.S. Census Bureau
(2022)
U.S. Territories other than
Puerto Rico:
1990-2022: U.S. Census Bureau
(2023)
Tseptic
Percent treated in septic systems3
%
Odd years from 1989 through
2021: U.S. Census Bureau
(2021a)
Data for intervening years
obtained by linear
interpolation
2022: Forecasted from the rest
of the time series
Fnpr
Fraction of nitrogen in protein (0.16)
kg N/kg
protein
1990-2022: IPCC (2019) Eq.
6.10
Nhh
Additional nitrogen from household products (1.17)
No units
1990-2022: IPCC (2019) Table
6.10a
FNON-CON_septic
Factor for Non-Consumed Protein Added to
Wastewater (1.13)
No units
F|ND-COM_septic
Factor for Industrial and Commercial Co-Discharged
Protein, septic systems (1)
No units
1990-2022: IPCC (2019)
EFseptic
Emission factor, septic systems (0.0045)
kg N20-N/kg
N
1990-2022: IPCC (2019) Table
6.8a
44/28
Conversion factor
Molecular
weight ratio
of N20 to N2
Standard conversion
1/106
Conversion factor
kg to kt
Standard conversion
a Value of this activity data varies over the Inventory time series.
Waste 7-43
-------
Emissions from Centrally Treated Aerobic and Anaerobic Systems:
Nitrous oxide emissions from POTWs depend on the total nitrogen entering centralized wastewater treatment. The
total nitrogen entering centralized wastewater treatment was estimated by multiplying the U.S. population by the
percent of wastewater collected for centralized treatment (about 84 percent in 2022), the consumed protein per
capita, the fraction of N in protein, the correction factor for additional N from household products, the factor for
industrial and commercial co-discharged protein into wastewater treatment, and the factor for non-consumed
protein added to wastewater.
Equation 7-35: Total Nitrogen Entering Centralized Systems (IPCC 2019 [Eq. 10])
TNdomcentral Q
= (USpop x Tcenxralized) x Protein x FNPR x NHH x FNON-con x Find-com
Table 7-30: Variables and Data Sources for Non-Consumed Protein and Nitrogen Entering
Centralized Systems
Inventory Years: Source
Variable
Variable Description
Units
of Value
United States and
Puerto Rico:
1990-1999: U.S. Census
Bureau (2002); Instituto
de Estadisticas de
Puerto Rico (2021)
2000-2009: U.S. Census
USpop
U.S. population3
Persons
Bureau 2011
2010-2019: U.S. Census
Bureau (2021b)
2020-2022: U.S. Census
Bureau (2022)
U.S. Territories other
than Puerto Rico:
1990-2022: U.S. Census
Bureau (2023)
Odd years from 1989
through 2021: U.S.
Census Bureau (2021a)
Data for intervening
Tcentrauzed
Percent collected3
%
years obtained by linear
interpolation
2022: Forecasted from
the rest of the time
series
Protein
Consumed protein per capita3
kg/person/year
1990-2022: Calculated
Fnpr
1990-2022: IPCC (2019),
Fraction of nitrogen in protein (0.16)
kg N/kg protein
Eq. 6.10
Factor for additional nitrogen from household
Nhh
products (1.17)
No units
1990-2022: IPCC (2019),
Fnon-con
Factor for U.S. specific non-consumed protein
(1.13)
No units
Table 6.10a
7-44 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
Variable
Variable Description
Units
Inventory Years: Source
of Value
Find-com
Factor for Industrial and Commercial Co-
Discharged Protein (1.25)
No units
1990-2022: IPCC (2019)
Table 6.11
a Value of this activity data varies over the Inventory time series.
Nitrous oxide emissions from POTWs were estimated by multiplying the total nitrogen entering centralized
wastewater treatment, the relative percentage of wastewater treated by aerobic systems (other than constructed
wetlands) and anaerobic systems, aerobic systems with constructed wetlands as the sole treatment, the respective
emission factors for aerobic systems and anaerobic systems, and the conversion from N2 to N2O.
Table 7-34 presents the data for U.S. population, population served by centralized wastewater treatment plants,
available protein, and protein consumed. The methodological equations are:
Equation 7-36: Total Domestic N2O Emissions from Centrally Treated Aerobic Systems
Emissions from Centrally Treated Aerobic Systems (other than Constructed Wetlands)(B 1)
+ Emissions from Centrally Treated Aerobic Systems (Constructed Wetlands Only)(B2)
+ Emissions from Centrally Treated Aerobic Systems (Constructed Wetlands used as Tertiary Treatment)
(63) = B (kt N20/year)
where,
Equation 7-37: N2O Emissions from Centrally Treated Aerobic Systems (other than
Constructed Wetlands) (IPCC 2019 [Eq. 6.9])
B1 (kt N20/year) = [(TNDOm_central) x (% aer°bicoTcw)] x EFaerobiC x 44/28 x 1/106
Table 7-31: Variables and Data Sources for N2O Emissions from Centrally Treated Aerobic
Systems (Other than Constructed Wetlands)
Variable
Variable Description
Units
Inventory Years: Source
of Value
Emissions from Centrally Treated Aerobic Systems (Other than Constructed Wetlands) (kt N20/year)
TNdom central
Total nitrogen entering centralized systems3
kg N/year
1990-2022: Calculated
% aerobicoTcw
Flow to aerobic systems, other than constructed
wetlands only / total flow to POTWs3
%
1990,1991: Set equal to
1992
1992,1996, 2000, 2004:
EPA (1992, 1996, 2000,
2004), respectively
Data for intervening
years obtained by linear
interpolation.
2005-2022: Forecasted
from the rest of the time
series
EF aerobic
U.S.-specific emission factor - aerobic systems
(0.015)
kg N20-N/kg N
1990-2022: IPCC (2022)
44/28
Conversion factor
Molecular
weight ratio of
N2O to N2
Standard conversion
1/106
Conversion factor
kg to kt
Standard conversion
a Value of this activity data varies over the Inventory time series.
Nitrous oxide emissions from constructed wetlands used as sole treatment include similar data and processes as
aerobic systems other than constructed wetlands. See description above. Nitrous oxide emissions from
Waste 7-45
-------
constructed wetlands used as tertiary treatment were estimated by multiplying the flow to constructed wetlands
used as tertiary treatment, wastewater N concentration entering tertiary treatment, constructed wetlands
emission factor, and converting to kt/year.
Equation 7-38: N2O Emissions from Centrally Treated Aerobic Systems (Constructed Wetlands
Only) (IPCC 2014 [Eq. 6.9])
B2 ( yea2r ) = [(TNDOm_central) x (% aerobiccw)] x EFCW x 44/28 x 1/106
Equation 7-39: N2O Emissions from Centrally Treated Aerobic Systems (Constructed Wetlands
used as Tertiary Treatment) (U.S.-Specific)
B3 (UMi) = [(POTW_flow_CW) x (Ncw,inf) x 3.785 x (EFCW)] x 1/106 x 365.25
Table 7-32: Variables and Data Sources for N2O Emissions from Centrally Treated Aerobic
Systems (Constructed Wetlands)
Inventory Years: Source of
Variable
Variable Description
Units
Value
Emissions from Constructed Wetlands Only (kt N/O/year)
TNdom central
Total nitrogen entering centralized treatment3
kg N/year
1990-2022: Calculated
1990,1991: Set equal to
1992
1992,1996, 2000, 2004,
2008, 2012: EPA (1992,
1996, 2000, 2004, 2008,
% aerobiccw
Flow to aerobic systems, constructed wetlands
used as sole treatment / total flow to POTWs3
%
and 2012)
Data for intervening years
obtained by linear
interpolation.
2013-2022: Forecasted
from the rest of the time
series
EFcw
Emission factor for constructed wetlands
(0.0013)
kg N20-N/kg N
1990-2022: IPCC (2014)
Table 6.7
Molecular
44/28
Conversion factor
weight ratio of
IM2O to N2
Standard conversion
1/106
Conversion factor
kg to kt
Standard conversion
Emissions from Constructed Wetlands used as Tertiary Treatment (kt N20/year)
1990,1991: Set equal to
1992
1992, 1996, 2000, 2004,
2008, 2012: EPA (1992,
1996, 2000, 2004, 2008,
POTW_flow_CW
Wastewater flow to POTWs that use constructed
MGD
and 2012)
wetlands as tertiary treatmenta
Data for intervening years
obtained by linear
interpolation.
2013-2022: Forecasted
from the rest of the time
series
7-46 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
Variable
Variable Description
Units
Inventory Years: Source of
Value
Ncw.inf
BOD concentration in wastewater entering the
constructed wetland (25)
mg/L
1990-2022: Metcalf & Eddy
(2014)
3.785
Conversion factor
liters to gallons
Standard conversion
EFcw
Emission factor for constructed wetlands
(0.0013)
kg NjO-N/kg N
1990-2022: IPCC (2014)
Table 6.7
1/106
Conversion factor
mg to kg
Standard conversion
365.25
Conversion factor
Days in a year
Standard conversion
a Value of this activity data varies over the Inventory time series.
Data sources and methodologies are similar to those described for aerobic systems, other than constructed
wetlands. See discussion above.
Equation 7-40: N2O Emissions from Centrally Treated Anaerobic Systems (IPCC 2019 [Eq. 6.9])
kt N,0
C ( ) - [(TNDOm_central) x (% anaerobic)] x EFanaerobic x 44/28 x 1/10
Table 7-33: Variables and Data Sources for N2O Emissions from Centrally Treated Anaerobic
Systems
Variable
Variable Description
Units
Inventory Years: Source of
Value
Emissions from Centrally Treated Anaerobic Systems
TNdom_central
Total nitrogen entering centralized
treatment3
kg N/year
1990-2022: Calculated
% anaerobic
Percent centralized wastewater that
is anaerobically treated3
%
1990,1991: Set equal to 1992
1992, 1996, 2000, 2004: (EPA
1992, 1996, 2000, 2004),
respectively
Data for intervening years
obtained by linear
interpolation.
2005-2022: Forecasted from
the rest of the time series
EF anaerobic
Emission factor for anaerobic
reactors/deep lagoons (0)
kg N20-N/kg N
1990-2022: IPCC (2019) Table
6.8A
44/28
Conversion factor
Molecular weight
ratio of IM2O to N2
Standard conversion
1/106
Conversion factor
mg to kg
Standard conversion
a Value of this activity data varies over the Inventory time series.
Table 7-34: U.S. Population (Millions) Fraction of Population Served by Centralized
Wastewater Treatment (percent), Protein Supply (kg/person-year), and Protein Consumed
(kg/person-year)
Year
1990
2005
2018
2019
2020
2021
2022
Population
253
300
330
332
335
336
337
Centralized WWT Population (%)
75.6 1
78.8
82.9
83.6
84.2
84.8
83.6
Protein Supply
43.1
44.9
45.5
46
46.9
46.3
46.3
Protein Consumed
33.2 1
34.7
35.1
35.5
36.2
35.7
35.7
Sources: Population - U.S. Census Bureau 2002; U.S. Census Bureau 2011; U.S. Census Bureau (2021b); Instituto de
Estadisticas de Puerto Rico (2021); U.S. Census Bureau (2022); U.S. Census Bureau (2023); WWTP Population - U.S. Census
Bureau (2021a); Available Protein - USDA (2015), FAO (2022); Protein Consumed - FAO (2022).
Waste 7-47
-------
Emissions from Discharge of Centralized Treatment Effluent:
Nitrous oxide emissions from the discharge of wastewater treatment effluent were estimated by multiplying the
total nitrogen in centrally treated wastewater effluent by the percent of wastewater treated in primary,
secondary, and tertiary treatment and the fraction of nitrogen remaining after primary, secondary, or tertiary
treatment and then multiplying by the percent of wastewater volume routed to waterbodies with nutrient-
impaired/eutrophic conditions and all other waterbodies (ERG 2021a) and emission factors for discharge to
impaired waterbodies and other waterbodies from IPCC (2019). The methodological equations are:
Equation 7-41: N2O Emissions from Centrally Treated Systems Discharge (U.S.-Specific)
D ( yea2r ) = [(NEFfluent,imp x EFimp) + (Neffluenx NOnimp x EFNOnimp)] x 44/28 x 1/106
where,
Equation 7-42: Total Organics in Centralized Treatment Effluent (IPCC 2019 [Eq. 6.8])
N
1NEFFLUENT,DOM ^yearj
= [TNdqm.central11 x % primary x (1 - Nrem PRIMARY)] + [TND0M_central x % secondary x (1 -
Nrem.sECONDARy)] + [TNDOm_central x % tertiary x (1 — Nrem XERXIARY)]
Equation 7-43: Total Nitrogen in Effluent Discharged to Impaired Waterbodies (U.S.-Specific)
Neffluent.imp (kg N/year) = (NEFFLUENXDOm x PercentIMP)/1000
Equation 7-44: Total Nitrogen in Effluent Discharged to Nonimpaired Waterbodies (U.S.-
Specific)
Neffluent.nonimp (kg N year) = NEFFLUENX DOm x PercentNONiMp)/1000
Table 7-35: Variables and Data Sources for N2O Emissions from Centrally Treated Systems
Discharge
Variable
Variable Description
Units
Source of Value
Neffulent.dom
Total organics in centralized treatment effluent3
kg N/year
1990-2022: Calculated
44/28
Conversion factor
Molecular
weight ratio of
N20 to N2
Standard conversion
1/106
Conversion factor
kg to kt
Standard conversion
TNdom central
Total nitrogen entering centralized treatment3
kg N/year
1990-2022: Calculated
1000
Conversion factor
kg to kt
Standard Conversion
% primary
Percent of primary domestic centralized treatment3
%
1990,1991: Set equal to
1992.
1992, 1996, 2000,
2004, 2008, 2012: EPA
(1992, 1996, 2000,
2004, 2008, and 2012),
respectively
% secondary
Percent of secondary domestic centralized treatment3
%
% tertiary
Percent of tertiary domestic centralized treatment3
%
11 See emissions from centrally treated aerobic and anaerobic systems for methodological equation calculating TNdom_central-
7-48 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
Variable
Variable Description
Units
Source of Value
Data for intervening
years obtained by
linear interpolation.
2013-2022: Forecasted
from the rest of the
time series
Nrem.PRIMARY
Fraction of nitrogen removed from primary domestic
centralized treatment (0.1)
No units
1990-2022: IPCC (2019)
Table 6.10c
Nrem.SECONDARY
Fraction of nitrogen removed from secondary domestic
centralized treatment (0.4)
No units
Nrem.TERTIARY
Fraction of nitrogen removed from tertiary domestic
centralized treatment (0.9)
No units
Neffluent.imp
Total nitrogen in effluent discharged to impaired waterbodies
kg N/year
1990-2022: Calculated
Neffluent.nonimp
Total nitrogen in effluent discharged to nonimpaired
waterbodies
kg N/year
EFimp
Emission factor (discharge to impaired waterbodies) (0.19)
kg N20-N/kg N
1990-2022: IPCC (2019)
Table 6.8a
EFlMONIMPr
Emissions factor (discharge to nonimpaired waterbodies)
(0.005)
kg N20-N/kg N
PercentiMP
Percent of wastewater discharged to impaired waterbodies3
%
1990-2010: Set equal to
2010
2010: ERG (2021a)
2011: Obtained by
linear interpolation
2012: ERG (2021a)
2013-2022: Set equal to
2012
PercentNONiMP
Percent of wastewater discharged to nonimpaired
waterbodies3
%
a Value for this activity data varies over the Inventory time series.
Industrial Wastewater N2O Emission Estimates
Nitrous oxide emission estimates from industrial wastewater are estimated according to the methodology
described in the 2019 Refinement. U.S. industry categories that are likely to produce significant N2O emissions
from wastewater treatment were identified based on whether they generate high volumes of wastewater,
whether there is a high nitrogen wastewater load, and whether the wastewater is treated using methods that
result in N2O emissions. The top four industries that meet these criteria and were added to the Inventory are meat
and poultry processing; petroleum refining; pulp and paper manufacturing; and breweries (ERG 2021a).
Wastewater treatment and discharge emissions for these sectors for 2022 are displayed in Table 7-36 below. Table
7-20 contains production data for these industries.
Table 7-36: Total Industrial Wastewater N2O Emissions by Sector (2022, MMT CO2 Eq. and
Percent)
N O Emissions
% of Industrial
Industry
(MMTCO. Eq.)
Wastewater N<0
Meat & Poultry
0.2
47.3%
Petroleum Refineries
0.1
30.5%
Pulp & Paper
0.1
21.4%
Breweries
+
0.7%
Total
0.5
100%
+ Does not exceed 0.5 MMT C02 Eq.
Note: Totals may not sum due to independent rounding.
Emissions from Industrial Wastewater Treatment Systems:
Waste 7-49
-------
More recent research has revealed that emissions from nitrification or nitrification-denitrification processes at
wastewater treatment, previously judged to be a minor source, may in fact result in more substantial emissions
(IPCC 2019). N2O is generated as a by-product of nitrification, or as an intermediate product of denitrification.
Therefore, N2O emissions are primarily expected to occur from aerobic treatment systems. To estimate these
emissions, the total nitrogen entering aerobic wastewater treatment for each industry must be calculated. Then,
the emission factor provided by the 2019 Refinement is applied to the portion of wastewater that undergoes
aerobic treatment.
The total nitrogen that enters each industry's wastewater treatment system is a product of the total amount of
industrial product produced, the wastewater generated per unit of product, and the nitrogen expected to be
present in each meter cubed of wastewater (IPCC equation 6.13).
Equation 7-45: Total Nitrogen in Industrial Wastewater
TNINDi = PixWix TNt
where,
TN iNDi — total nitrogen in wastewater for industry /' for inventory year, kg TN/year
/ = industrial sector
Pi = total industrial product for industrial sector / for inventory year, t/year
Wi = wastewater generated per unit of production for industrial sector / for inventory year,
m3/t product
Tni = total nitrogen in untreated wastewater for industrial sector / for inventory year, kg TN/m3
For the four industries of interest, the total production and the total volume of wastewater generated has already
been calculated for CH4 emissions. For these new N2O emission estimates, the total nitrogen in the untreated
wastewater was determined by multiplying the annual industry production, shown in Table 7-20, by the average
wastewater outflow, shown in Table 7-23 and the nitrogen loading in the outflow shown in Table 7-37.
Table 7-37: U.S. Industrial Wastewater Nitrogen Data
Industry
Wastewater Total N
(kg N/ m')
Source for Total N
Pulp and Paper
0.30a
Cabrera (2017)
Meat Processing
0.19
IPCC (2019), Table 6.12
Poultry Processing
0.19
IPCC (2019), Table 6.12
Petroleum Refining
0.051
Kenari et al. (2010)
Breweries - Craft
0.055
IPCC (2019), Table 6.12
Breweries - NonCraft
0.055
IPCC (2019), Table 6.12
a Units are kilograms N per air-dried metric ton of production.
Nitrous oxide emissions from industry wastewater treatment are calculated by applying an emission factor to the
percent of wastewater (and therefore nitrogen) that undergoes aerobic treatment (IPCC Equation 6.11).
Equation 7-46: N2O Emissions from Industrial Wastewater Treatment Plants
N20 PlantsIND = j x EFx TNINDiJ\ x —
where,
N2O PlantsiND = N2O emissions from industrial wastewater treatment plants for inventory year, kg
INhO/year
TN iNDi — total nitrogen in wastewater from industry /' for inventory year, kg N/year
Ti,j = degree of utilization of treatment/discharge pathway or system j, for each industry / for
inventory year
7-50 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
/ = industrial sector
j = each treatment/discharge pathway or system
EFi.j = emission factor for treatment/discharge pathway or system j, kg N20-N/kg N. 0.015 kg
N20-N/kg N (IPCC 2022)
44/28 = conversion of kg N2O-N into kg N2O
For each industry, the degree of utilization (Ti,j)-the percent of wastewater that undergoes each type of
treatment-was previously determined for CFU emissions and presented in Table 7-22.
Emissions from Industrial Wastewater Treatment Effluent:
Nitrous oxide emissions from industrial wastewater treatment effluent are estimated by multiplying the total
nitrogen content of the discharged wastewater effluent by an emission factor associated with the location of the
discharge. Where wastewater is discharged to aquatic environments with nutrient-impacted/eutrophic conditions
(i.e., water bodies which are rich in nutrients and very productive in terms of aquatic animal and plant life), or
environments where carbon accumulates in sediments such as lakes, reservoirs, and estuaries, the additional
organic matter in the discharged wastewater is expected to increase emissions.
Equation 7-47: N2O Emissions from Industrial Wastewater Treatment Effluent
N20 EffluentIND = Neffluentind x EFeffluent x 44/28
where,
N2O EffluentiND = N2O emissions from industrial wastewater discharge for inventory year (kg INhO/year)
Neffluent.ind = Total nitrogen in industry wastewater effluent discharged to aquatic environments for
inventory year (kg N/year)
EFeffluent = Tier 1 emission factor for wastewater discharged to aquatic environments (0.005 kg
INhO-N/kg N) (IPCC 2019)
44/28 = Conversion of kg N2O-N into kg N2O
The total N in treated effluent was determined through use of a nutrient estimation tool developed by EPA's Office
of Water (EPA 2019). The Nutrient Tool uses known nutrient discharge data within defined industrial sectors or
subsectors, as reported on Discharge Monitoring Reports, to estimate nutrient discharges for facilities within that
sector or subsector that do not have reported nutrient discharges but are likely to discharge nutrients. The
estimation considers, within each sector or subsector, elements such as the median nutrient concentration and
flow, as well as the percent of facilities within the sector or subsector that have reported discharges. Data from
2018 are available for the pulp, paper, and paperboard, meat and poultry processing, and petroleum refining
industries. To complete the time series, an industry-specific percent removal of nitrogen was calculated using the
total nitrogen in untreated wastewater. See Table 7-38.
Because data for breweries was not available, the removal of nitrogen was assumed to be equivalent to secondary
treatment, or 40 percent (IPCC 2019). The Tier 1 emission factor (0.005 kg N20/kg N) from IPCC (2019) was used.
Table 7-38: Industrial Wastewater Nitrogen Discharged in 2018 by Sector (kg N)
Industry
N Effluent^, (kg N)
Industry-Specific N
Removal Factor
Meat & Poultry
12,078,919
0.082
Petroleum Refineries
1,698,953
0.045
Pulp & Paper
18,809,623
1.08
Breweries3
1,604,878
NA
a Nitrogen discharged by breweries was estimated as 60 percent of untreated wastewater
nitrogen.
NA (Not Available)
Source: ERG (2021a).
Waste 7-51
-------
Uncertainty
The overall uncertainty associated with both the 2022 Cm and N2O emission estimates from wastewater
treatment and discharge was calculated using the 2006IPCC Guidelines Approach 2 methodology (IPCC 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 emissions from wastewater from pulp and
paper manufacturing, meat and poultry processing, fruits and vegetable processing, ethanol production,
petroleum refining, and breweries. Uncertainty associated with the parameters used to estimate N2O emissions
include that of numerous input variables used to model emissions from domestic wastewater and emissions from
wastewater from pulp and paper manufacturing, meat and poultry processing, petroleum refining, and breweries.
Uncertainty associated with centrally treated constructed wetlands parameters including U.S. population served by
constructed wetlands, and emission and conversion factors are from IPCC (2014), whereas uncertainty associated
with POTW flow to constructed wetlands and influent BOD and nitrogen concentrations were based on expert
judgment (ERG 2021b). The specified probability density functions (PDFs) are assumed to be normal for most
activity data and emission factors, and due to lack of data, are based on expert judgement (ERG 2021c).
The results of this Approach 2 quantitative uncertainty analysis are summarized in Table 7-39. For 2022, methane
emissions from wastewater treatment were estimated to be between 14.9 and 27.7 MMT CO2 Eq. at the 95
percent confidence level (or in 19 out of 20 Monte Carlo stochastic simulations). This indicates a range of
approximately 29 percent below to 33 percent above the 2022 emissions estimate of 20.8 MMT CO2 Eq. Nitrous
oxide emissions from wastewater treatment were estimated to be between 13.9 and 64.0 MMT CO2 Eq., which
indicates a range of approximately 36 percent below to 192 percent above the 2022 emissions estimate of 21.9
MMTCChEq.
Table 7-39: Approach 2 Quantitative Uncertainty Estimates for 2022 Emissions from
Wastewater Treatment (MMT CO2 Eq. and Percent)
2022 Emission Estimate Uncertainty Range Relative to Emission Estimate-1
(MMTCO' Eq.) (MMTCO Eq.) (%)
Lower
Upper
Lower
Upper
Bound
Bound
Bound
Bound
Wastewater T reatment
ch4
20.8
14.9
27.7
-29%
+33%
Domestic
ch4
13.6
00
00
19.3
-35%
+42%
Industrial
ch4
7.2
4.2
11.4
-42%
+57%
Wastewater Treatment
n2o
21.9
13.9
64.0
-36%
+192%
Domestic
n2o
21.4
13.0
63.2
-39%
+195%
Industrial
n2o
0.5
0.5
1.4
-0.7%
+201%
a Range of emission estimates predicted by Monte Carlo stochastic simulation for a 95 percent confidence interval.
QA/QC and Verification
General QA/QC procedures were applied to activity data, documentation, and emission calculations consistent
with the U.S. Inventory QA/QC Plan, which is in accordance with Vol. 1 Chapter 6 of the 2006 IPCC Guidelines (see
Annex 8 for more details). This effort included a general or Tier 1 analysis, including the following checks:
• Checked for transcription errors in data input;
• Ensured references were specified for all activity data used in the calculations;
• Checked a sample of each emission calculation used for the source category;
• Checked that parameter and emission units were correctly recorded and that appropriate conversion
factors were used;
7-52 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
• Checked for temporal consistency in time series input data for each portion of the source category;
• Confirmed that estimates were calculated and reported for all portions of the source category and for all
years;
• Investigated data gaps that affected trends of emission estimates; and
• Compared estimates to previous estimates to identify significant changes.
Calculation-related QC (category-specific, Tier 2) was performed for a portion of the domestic wastewater
treatment discharges methodology, which included assessing available activity data to ensure the most complete
publicly data set was used and checking historical trends in the data to assist determination of best methodology
for filling in the time series for data that are not available annually.
All transcription errors identified were corrected and documented. The QA/QC analysis did not reveal any systemic
inaccuracies or incorrect input values.
Recalculations Discussion
Population data were updated using the same and latest data sources as the state-level emissions inventory to
create consistency across Inventory estimates. These changes affected the years 2020 and 2021. Protein data were
updated to reflect available protein values available for 2018 through 2020 (FAO 2022). Pulp, paper, and
paperboard production data were updated to reflect revised values for 2021 (FAO 2023a). Updated red meat
production values for 2021, were updated based on revised data (USDA 2023a). Fruits and vegetables production
values were updated for the time series (ERG 2022; USDA 2023c). Ethanol production values for 2021 were based
on revised data (RFA 2023a; RFA 2023b). Petroleum refining production values for 2021 were revised based on EIA
(2023). Updated values for non-craft brewery wastewater generation were included for the years 2015 and 2020,
affecting the values for 2016, 2018, 2019, and 2021 (BIER 2021).
Compared to the previous Inventory the cumulative effect of all these recalculations had a minor impact on the
overall wastewater treatment emission estimates:
• Domestic wastewater treatment and discharge Cm emissions decreased on average 0.2 percent over the
timeseries, with 1990 through 2019 not changing and the largest decrease of 3.1 percent (0.4MMT CO2
Eq.) in 2021.
• Domestic wastewater treatment and discharge N2O emissions increased an average 5.6 percent over the
timeseries, with 1990 through 2017 not changing and the largest increase of 6.8 percent (1.4 MMT CO2
Eq.) in 2020.
• Industrial wastewater treatment and discharge CH4 emissions decreased on average 0.01 percent over the
timeseries, with the smallest decrease of 0.003 percent (0.0 MMT CO2 Eq.) in 2017 and largest decrease of
0.2 percent (0.01 MMT C02 Eq.) in 2020.
• Industrial wastewater treatment and discharge N2O emissions increase an average 0.02 percent over the
timeseries, with the smallest increase of 0.0 percent (0.0 MMT CO2 Eq.) in 1990 to the largest increase of
0.6 percent (0.003 MMT C02 Eq.) in 2021.
Over the time series, the total emissions on average increased by 0.1 percent from the previous Inventory. The
changes ranged from the smallest increase, 0.0005 percent (0.0002 MMT CO2 Eq.), in 2017, to the largest
decrease, 2.4 percent (1.0 MMT CO2 Eq.), in 2020.
Planned Improvements
EPA notes the following improvements will continue to be investigated as time and resources allow, but there are
no immediate plans to implement them until data are available or identified:
Waste 7-53
-------
• Continue to investigate anaerobic sludge digester and biogas data compiled by the Water Environment
Federation (WEF) in collaboration with other entities as a potential source of updated activity data. Due
to lack of these data, the United States continues to use another method for estimating biogas produced.
This method uses the standard 100 gallons/capita/day wastewater generation factor for the United States
(Ten-State Standards), which EPA believes is reasonable to estimate national emissions. However, based
on stakeholder input, some regions of the United States use markedly less water due to water
conservation efforts so EPA plans to investigate updated sources for this method as well.
• Investigate additional sources for estimating wastewater volume discharged and discharge location for
both domestic and industrial sources. For domestic wastewater, the goal would be to provide additional
data points along the time series, while the goal for industrial wastewater would be to update the Tier 1
discharge methodology to a Tier 2 methodology.
• Investigate additional sources for domestic wastewater treatment type in place data.
• Continue to review whether sufficient data exist to develop U.S.-specific Cm or N2O emission factors for
domestic wastewater treatment systems, including whether emissions should be differentiated for
systems that incorporate biological nutrient removal operations.
• Investigate additional data sources for improving the uncertainty of the estimate of N entering municipal
treatment systems.
• Evaluate literature provided by expert review commenters for potential inclusion as updates, in particular
focusing on the industrial N2O emission factor for pulp and paper wastewater treatment.
• Evaluate the use of POTW BOD effluent discharge data from ICIS-NPDES.12 Currently only half of POTWs
report organics as BODsso EPA would need to determine a hierarchy of parameters to appropriately sum
all loads. Using these data could potentially improve the current methane emission estimates from
domestic discharge, or at least provide a comparison to the current method for QA/QC.
• Evaluate the use of POTW N effluent discharge data from ICIS-NPDES. Currently only about 80 percent of
POTWs report a form of N so EPA would need to determine an appropriate method to scale to the total
POTW population. EPA is aware of a method for industrial sources and plans to determine if this method
is appropriate for domestic sources. Using these data could potentially improve the current nitrous oxide
emissions estimates from domestic discharge, or at least provide a comparison to the current method for
QA/QC.
7.3 Composting (CRT Source Category 5B1)
Composting of organic waste, such as food waste, garden (yard) and park waste, and wastewater treatment sludge
and/or biosolids, is common in the United States. Composting reduces the amount of methane-generating waste
entering landfills, destroys pathogens in the waste, sequesters carbon, and provides a source of organic matter.
Composting can also generate a saleable product and reduce the need for chemical fertilizers when the end
product is used as a fertilizer or soil amendment. This source category assumes all composting facilities are
commercial, large-scale anaerobic windrow composting facilities with yard trimmings as the main waste stream
composted, which aligns with findings from full-scale compost infrastructure survey data published by BioCycle
(2017, 2023). Of 200 major food waste composting facilities in the United States, 75 (38 percent) use the windrow
method, 45 (23 percent) use the aerated static pile method, and the remainder use other methods. The BioCycle
2023 survey received responses from facilities using aerobic composting methods (e.g., aerated static piles, in-
12 ICIS-NPDES refers to EPA's Integrated Compliance Information System - National Pollutant Discharge Elimination System.
7-54 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
vessel composting) are operational in the United States, however national estimates of the material processed by
these facilities are not readily available; therefore, emissions estimates by composting method are not included in
this source category. Residential backyard composting is also not included in this source category.
Composting naturally converts a large fraction of the degradable organic carbon in the waste material into carbon
dioxide (CO2) through aerobic processes without anthropogenic influence. With anthropogenic influences (e.g., at
commercial or large on-site composting operations), anaerobic conditions can be created in sections of the
compost pile when there is excessive moisture or inadequate aeration (or mixing) of the compost pile, resulting in
the formation of methane (CH4). Methane in aerobic sections of a windrow pile is generally oxidized by
microorganisms, which convert the CH4 to CO2 emissions. Even though CO2 emissions are generated, they are not
included in net greenhouse gas emissions for composting. Consistent with the 2006IPCC Guidelines, net CChflux
from carbon stock changes in waste material are estimated and reported under the LULUCF sector. The estimated
Cm released into the atmosphere ranges from less than 1 percent to a few percent of the initial carbon content in
the material (IPCC 2006). Depending on how well the compost pile is managed, nitrous oxide (N2O) emissions can
also be produced. The formation of N2O depends on the initial nitrogen content of the material and is mostly due
to nitrogen oxide (NOx) denitrification during the thermophilic and secondary mesophilic stages of composting
(Cornell 2007). Emissions vary and range from less than 0.5 percent to 5 percent of the initial nitrogen content of
the material (IPCC 2006). Animal manures are typically expected to generate more N2O than, for example, yard
waste, however data are limited.
From 1990 to 2022, the amount of waste composted in the United States increased from 3,810 kt to 23,042 kt (see
Table 7-42). There was some fluctuation in the amount of waste composted between 2006 to 2009 where a peak
of 20,063 kt composted was observed in 2008, which decreased to 18,838 kt composted the following year,
presumably driven by the economic crisis of 2009 (data not shown). Since 2009, the amount of waste composted
has gradually increased, and when comparing 2010 to 2022, a 26 percent increase in waste composted is
observed. Emissions of CH4 and N2O from composting from 2010 to 2022 have increased by the same percentage.
In 2022, CH4 emissions from composting (see Table 7-40 and Table 7-41) were 2.6 MMT CO2 Eq. (92 kt), and N2O
emissions from composting were 1.8 MMT CO2 Eq. (7 kt), representing consistent emissions trends over the past
several years. Composted material primarily includes yard trimmings (grass, leaves, and tree and brush trimmings)
and food scraps from the residential and commercial sectors (such as grocery stores; restaurants; and school,
business, and factory cafeterias). The composted waste quantities reported here do not include small-scale
backyard composting and agricultural composting mainly due to the lack of consistent and comprehensive national
data. Additionally, it is assumed that backyard composting tends to be a more naturally managed process with less
chance of generating anaerobic conditions and CH4 and N2O emissions. Agricultural composting is accounted for in
Chapter 5, Section 5.4 (Agricultural Soil Management) of this Inventory, as most agricultural composting operations
are assumed to land-apply the resultant compost to soils.
The growth in composting since the 1990s and specifically over the past decade may be attributable to the
following factors: (1) the enactment of legislation by state and local governments that discouraged or banned the
disposal of yard trimmings and/or food waste in landfills, (2) an increase in yard trimming collection and yard
trimming drop off sites operated by local solid waste management districts/divisions, (3) an increased awareness of
the environmental benefits of composting, and (4) loans or grant programs to establish or expand composting
infrastructure.
Most bans or diversion laws on the disposal of yard trimmings were initiated in the early 1990s by state or local
governments (U.S. Composting Council 2010). California, for example, enacted a waste diversion law for organics
including yard trimmings and food scraps in 1999 (AB939) that required jurisdictions to divert 50 percent of the
waste stream by 2000, or be subjected to fines. Currently, 20 states representing up to 42 percent of the nation's
population have enacted legislation banning yard waste from landfill disposal (U.S. Composting Council 2022).
Additional initiatives at the metro and municipal level also exist across the United States. Roughly 4,713
composting facilities exist in the United States with most (57.2 percent) composting yard trimmings only (BioCycle
2017).
Waste 7-55
-------
In the last decade, bans and diversions for food waste have also become more common. As of 2022, eight states
(California, Connecticut, Massachusetts, New Jersey, New York, Orgon, Vermont, Washington) and seven local
governments (Austin, TX; Boulder, CO; Hennepin County, MN; Portland, OR; New York City, NY; San Francisco, CA;
Seattle, WA) had implemented organic waste bans or mandatory recycling laws to help reduce organic waste
entering landfills, with most having taken effect after 2013 (U.S. Composting Council 2022). In most cases, organic
waste reduction in landfills is accomplished by following recycling guidelines, donating excess food for human
consumption, or by sending waste to organics processing facilities (Harvard Law School and CET 2019). An example
of an organic waste ban as implemented by California is the California Mandatory Recycling Law (AB1826), which
requires companies to comply with organic waste recycling procedures if they produce a certain amount of organic
waste and took effect on January 1, 2015 (Harvard Law School and CET 2019). In 2017, BioCycle released a report
in which 27 of 43 states that responded to their organics recycling survey noted that food waste (collected
residential, commercial, institutional, and industrial food waste) was recycled via anaerobic digestion and/or
composting. These 27 states reported an estimated total of 1.8 million tons of food waste diverted from landfills in
2016 (BioCycle 2018b). A growing number of initiatives to encourage households and businesses to compost or
beneficially reuse food waste also exist.
Table 7-40: CH4 and N2O Emissions from Composting (MMT CO2 Eq.)
Activity
1990
2005
2018
2019
2020
2021
2022
ch4
0.4
2.1
2.5
2.5
2.6
2.6
2.6
n2o
0.3
1.5
1.8
1.8
1.8
1.8
1.8
Total
0.7
3.6
4.3
4.3
4.4
4.4
4.4
Note: Totals may not sum due to independent rounding.
Table 7-41: CH4 and N2O Emissions from Composting (kt)
Activity
1990
2005
2018
2019
2020
2021
2022
ch4
15
75
90
91
92
92
92
n2o
1 1
6
7
7
7
7
7
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 7-40 and Table 7-41 were estimated using the IPCC default (Tier 1) methodology
(IPCC 2006) in accordance with IPCC methodological decision trees and available data. Using this method,
emissions are 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):
Equation 7-48: Greenhouse Gas Emission Calculation for Composting
Per IPCC Tier 1 methodology defaults, the emission factors for CH4 and N2O assume a moisture content of 60
percent in the wet waste (IPCC 2006). While the moisture content of composting feedstock can vary significantly
Methodology
E; = M X EF,
where,
Ei
M
EFi
CH4 or N2O emissions from composting, kt CH4or N2O
mass of organic waste composted in kt
emission factor for composting, 41 CH4/kt of waste treated (wet basis) and
0.31 N20/kt of waste treated (wet basis) (IPCC 2006)
designates either CH4or N2O
7-56 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
by type, composting as a process ideally proceeds between 40 to 65 percent moisture (University of Maine 2016;
Cornell 1996).
Estimates of the quantity of waste composted (M, wet weight as generated) are presented in Table 7-42 for select
years. Estimates of the quantity composted for 1990 and 2005 were taken from EPA's Advancing Sustainable
Materials Management: Facts and Figures 2015 (EPA 2018); estimates of the quantities composted for 2017 to
2018 were taken from EPA's Advancing Sustainable Materials Management: 2018 Tables and Figures (EPA 2020);
the estimate of the quantity composted for 2019 to 2022 were extrapolated using the 2018 quantity composted
and a ratio of the U.S. population growth for each year between 2018 and 2022 (U.S. Census Bureau 2021; U.S.
Census Bureau 2022; U.S Census Bureau 2023). Estimates of waste composted by commercial facilities in Puerto
Rico were provided for select years by EPA Region 2 (Kijanka 2020). This data includes amount of waste composted
at three facilities in Puerto Rico for 2017, 2018, and 2019, ranging from approximately 1,200 kt to a high of 15,000
kt. The average waste composted for these years was used as the annual amount composted for the respective
facility for years the facility was operational. The annual quantity of composted waste in Puerto Rico was
forecasted for 2020, 2021, and 2022 using available data from prior years, assumed metro area population data
near where each facility is located, and the Microsoft Excel FORECAST function to obtain annual composting
estimates.
Table 7-42: U.S. Waste Composted (kt)
Activity
1990 2005 2018 2019 2020 2021 2022
Waste Composted
3,810 I 18,655 I 22,594 22,698 22,918 22,946 23,042
Uncertainty
The major uncertainty drivers are the assumption that all composting emissions come from commercial windrow
facilities and the use of default emission factors (IPCC 2006) which is tied to a homogenous mixture of waste
processed across the country (largely yard trimmings). Data presented by BioCycle (BioCycle 2017, 2023) confirm
most composting operations use the windrow method and yard trimmings are the largest share of material
composted across the country, but there are other composting methods used and waste characteristics will vary at
a facility level. Additionally, there are composting operations in Puerto Rico and U.S. territories that are not
explicitly included in the national quantity of material composted as reported in the EPA Sustainable Materials
Management Reports because the methodological scope does not include Puerto Rico and U.S. territories. EPA
took steps to include emissions from Puerto Rico and U.S. Territories beginning in the 1990 to 2020 Inventory and
will continue to seek out additional data in future Inventories.
The estimated uncertainty from the 2006 IPCC Guidelines is ±58 percent for the Tier 1 methodology and considers
the individual emission factors applied to the default emission factors and activity data.
Emissions from composting in 2022 were estimated to range between 1.8 and 7.0 MMT CO2 Eq., which indicates a
range of 58 percent below to 58 percent above the 2022 emission estimate of each gas (see Table 7-43).
Table 7-43: Tier 1 Quantitative Uncertainty Estimates for Emissions from Composting (MMT
CO2 Eq. and Percent)
Source
Gas
2022 Emission Estimate
Uncertainty Range Relative to Emission Estimate
(MMTCO. Eq.)
(MMTCO. Eq.)
(%)
Lower Upper
Lower
Upper
Bound Bound
Bound
Bound
Composting
ch4
2.6
1.1 4.1
-58%
+58%
Composting
n2o
1.8
0.8 2.9
-58%
+58%
Composting
Total
4.4
O
00
1
-58%
+58%
Waste 7-57
-------
QA/QC and Verification
General QA/QC procedures were applied to data gathering and input, documentation, and calculations consistent
with the U.S. Inventory QA/QC Plan, which is in accordance with Vol. 1 Chapter 6 of the 2006IPCC Guidelines (see
Annex 8 for more details). No errors were found for the current Inventory.
Recalculations Discussion
No recalculations were performed for the 1990 to 2022 Inventory.
Planned Improvements
EPA recently completed a literature search on emission factors and composting systems and management
techniques that were documented in a draft technical memorandum. The purpose of this literature review was to
compile all published emission factors specific to various composting systems and composted materials in the
United States to determine whether the emission factors used in the current methodology can be revised or
expanded to account for geographical differences and/or differences in composting systems used. For example,
outdoor composting processes in arid regions typically require the addition of moisture compared to similar
composting processes in wetter climates. In general, there is a lack of facility-specific data on the management
techniques and sum of material composted to enable the use of different emission factors. EPA will continue to
seek out more detailed data on composting facilities to enable this improvement in the future.
Relatedly, EPA has received comments during previous Inventory cycles recommending that calculations for the
composting sector be based on waste subcategories (i.e., leaves, grass and garden debris, food waste) and
category-specific moisture contents. At this time, EPA is not aware of any available datasets which would enable
estimations to be performed at this level of granularity. EPA will continue to search for data which could lead to
the development of subcategory-specific composting emission factors to be used in future Inventory cycles.
EPA will also continue to seek out activity data including processing capacity and years of operation for commercial
composting facilities in Puerto Rico (for additional years), Guam, and other U.S. Territories for inclusion in a future
Inventory.
7.4 Anaerobic Digestion at Biogas Facilities
(CRT Source Category 5B2)
Anaerobic digestion is a series of biological processes in the absence of oxygen in which microorganisms break
down organic matter, producing biogas and digestate. The biogas primarily consists of Cm, biogenic CO2, and trace
amounts of other gases such as N2O (IPCC 2006) and is often combusted to produce heat and power, or further
processed into renewable natural gas or for use as a transportation fuel. Digester gas contains approximately 65
percent CFU (a normal range is 55 percent to 65 percent) and approximately 35 percent CO2 (WEF 2012; EPA 1993).
Methane emissions may result from a fraction of the biogas that is lost during the process due to leakages and
other unexpected events (0 to 10 percent of the amount of CH4 generated, IPCC 2006), collected biogas that is not
completely combusted, and entrained gas bubbles and residual gas potential in the digestate. Carbon dioxide
emissions are biogenic in origin and should be reported as an informational item in the Energy Sector (IPCC 2006).
Volume 5 Chapter 4 of the 2006 IPCC Guidelines notes that at biogas plants where unintentional CH4 emissions are
flared, CFU emissions are likely to be close to zero.
Anaerobic digesters differ based on the operating temperature, feedstock type and moisture content, and mode of
operation. The operating temperature dictates the microbial communities that live in the digester. Mesophilic
7-58 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
microbes are present at temperatures ranging from 85 to 100 degrees Fahrenheit while thermophilic microbes
thrive at temperatures ranging from 122 to 140 degrees Fahrenheit (WEF 2012). Digesters may process one or
more types of feedstock, including food waste; municipal wastewater solids; livestock manure; industrial
wastewater and residuals; fats, oils, and grease; and other types of organic waste streams. Co-digestion (multiple
feedstocks) is employed to increase methane production in cases where an organic matter type does not break
down easily. In co-digestion, various organic wastes are decomposed in a singular anaerobic digester by using a
combination of wastewater solids or manure and food waste from restaurants or food processing industry, a
combination of manure and waste from energy crops or crop residues (EPA 2016), or alternative combinations of
feedstock. The moisture content of the feedstock (wet or dry) impacts the amount of biogas generation. Wet
anaerobic digesters process feedstock with a solids content of less than 15 percent while dry anaerobic digesters
process feedstock with a solids content greater than 15 percent (EPA 2020). Digesters may also operate in batch or
continuous mode, which affects the feedstock loading and removal. Batch anaerobic digesters are manually loaded
with feedstock all at once and then manually emptied while continuous anaerobic digesters are continuously
loaded and emptied with feedstock (EPA 2020).
The three main categories of anaerobic digestion facilities included in national greenhouse gas inventories include
the following:
• Anaerobic digestion at biogas facilities, or stand-alone digesters, can be industry-dedicated digesters that
process waste from on industry or industrial facility (typically food of beverage waste from
manufacturing), or multi-source digesters that process feedstocks from various sources (e.g., municipal
food scraps, manure, food processing waste). Some stand-alone digesters also co-digest other organics
such as yard waste.
• On-farm digesters manage organic matter and reduce odor generated by farm animals or crops. On-farm
digesters are found mainly at dairy, swine, and poultry farms where there is the highest potential for
methane production to energy conversion. On-farm digesters may also accept food waste as feedstock for
co-digestion.
• Digesters at water resource recovery facilities (WRRF) produce biogas through the treatment and
reduction of wastewater solids. Some WRRF facilities may also accept and co-digest food waste.
This section focuses on stand-alone anaerobic digestion at biogas facilities. Emissions from on-farm digesters are
included Chapter 5 (Agriculture) and AD facilities at WRRFs are included in Section 7.2.
From 1990 to 2022, the estimated amount of waste managed by stand-alone digesters in the United States
increased from approximately 988 kt to 11,947 kt, an increase of 1,109 percent. As described in the Uncertainty
section, no data sources present the annual amount of waste managed by these facilities prior to 2015 when the
EPA began a comprehensive data collection survey. Thus, the emission estimates between 1990 and 2014, and for
2020 to 2022 are general estimates extrapolated from data collected for years 2015 to 2019 via the EPA surveys
(EPA 2018, 2019, 2021, and 2023). The steady increase in the amount of waste processed over the time series is
likely driven by increasing interest in using biogas produced from waste as a renewable energy source and other
organics diversion goals.
In 2022, emissions from stand-alone anaerobic digestion at biogas facilities were approximately 13,380 MT CO2 Eq.
(0.5 kt) (see Table 7-44 and Table 7-45).
Table 7-44: CH4 Emissions from Anaerobic Digestion at Biogas Facilities (MT CO2 Eq.)
Activity
1990
2005
2018
2019
2020
2021
2022
CH4 Generation
22,129 1
66,388
186,507
348,699
239,720
267,603
267,603
CH4 Recovery
(21,023) 1
(63,069)
(177,182)
(331,264)
(227,734)
(254,223)
(254,223)
CH4 Emissions
1,106
3,319
9,325
17,435
11,986
13,380
13,380
Notes: Parentheses indicate negative values. Totals may not sum due to independent rounding.
Waste 7-59
-------
Table 7-45: CH4 Emissions from Anaerobic Digestion at Biogas Facilities (kt CH4)
Activity
1990
2005
2018
2019
2020
2021
2022
CH4 Generation
1 1
2 1
7
12
9
10
10
CH4 Recovery
(1) i
(2) i
(6)
(12)
(8)
(9)
(9)
CH4 Emissions
+
+
+
1
+
+
+
+ Does not exceed 0.5 kt CH4.
Notes: Parentheses indicate negative values. Totals may not sum due to independent rounding.
Methodology
Methane emissions from anaerobic digestion depend on factors such as the type of waste managed, 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), aeration during the digestion process, unintentional leakages, and how the biogas
generated is used/combusted (e.g., flared, used on-site, used off-site).
The emissions presented in Table 7-44 were estimated using the IPCC default (Tier 1) methodology (Volume 5,
Chapter 4, IPCC 2006) given in Equation 7-49 below, which applies a default leakage factor of 5 percent to the Cm
generated, which is the product of an emission factor and the mass of organic waste processed (Equation 7-50).
Only Cm emissions are estimated because N2O emissions are considered negligible (IPCC 2006). Some Tier 2 data
are available (annual quantity of waste digested) for the later portion of the time series (2015 and later). The
methods were selected in accordance with IPCC methodological decision trees and available data on organic waste
processed.
Per the 2006 IPCC Guidelines, emissions of CH4 from anaerobic digestion facilities due to unintentional leakages
during process disturbances or other unexpected events are generally between 0 to 10 percent of the amount of
CH4 generated. When facility-specific information or data are unavailable, a 5 percent leakage factor is
recommended (IPCC 2006).
Equation 7-49: Methane Emissions Calculation for Anaerobic Digestion
CH4 Emissions = L x (GCH4)
where,
CH4 Emissions = total CH4 emissions in inventory year, Gg CH4
L = leakage factor, default assumed 5 percent (IPCC 2006)
Gch4 = total CH4 generation in inventory year, Gg CH4
Equation 7-50: Methane Generation Calculation for Anaerobic Digestion
Gch4 = li(Mi x EFd x 10"3
where,
M, = mass of organic waste treated by biological treatment type /', Gg, see Table 7-46
EF = emission factor for treatment /', g Cl-U/kg waste treated, 0.8 Mg/Gg CH4
i = anaerobic digestion
Per IPCC Tier 1 methodology defaults, the emission factor for CH4 assumes a moisture content of 60 percent in the
wet waste (IPCC 2006). Both liquid and solid wastes are processed by stand-alone digesters and the moisture
content entering a digester may be higher. One emission factor, 0.8 Mg/Gg CH4 is applied for the entire time series
(IPCC 2006 Volume 5, Chapter 4, Table 4.1).
The annual quantity of waste digested is sourced from EPA surveys of anaerobic digestion facilities (EPA 2018,
2019, 2021, and 2023). The EPA was granted the authority to survey anaerobic digestion facilities that process food
waste annually through an Information Collection Request (ICR No. 2533.01). The scope includes stand-alone and
7-60 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
co-digestion facilities (on-farm and water resource recovery facilities [WRRF]). Four reports with survey results
have been published to date:
• Anaerobic Digestion Facilities Processing Food Waste in the United States in 2015: Survey Results (EPA
2018)
• Anaerobic Digestion Facilities Processing Food Waste in the United States in 2016: Survey Results (EPA
2019)
• Anaerobic Digestion Facilities Processing Food Waste in the United States in (2017 & 2018): Survey Results
(EPA 2021)
• Anaerobic Digestion Facilities Processing Food Waste in the United States in (2019): Survey Results (EPA
2023).
These reports present aggregated survey data including the annual quantity of waste processed by digester type
(i.e., stand-alone, on-farm, and WRRF); waste types accepted; biogas generation and end use; and more. The
amount of waste digested as reported in the survey reports were assumed to be in wet weight; the majority of
stand-alone digesters were found to be wet and mesophilic (EPA 2019).
The aggregated data presented in the EPA reports are underestimates of the actual amount of processed waste
and biogas produced because (1) surveys rarely achieve a 100 percent response rate and some fraction of facilities
in each survey year did not respond to the survey; (2) EPA focused the surveys on facilities that primarily process
food waste, although non-food waste quantities processed were also collected and reported; and (3) while the EPA
has done due diligence to identify all stand-alone digesters that process food waste, EPA may not have identified
all facilities across the United States and its territories.
The annual quantity of waste digested at stand-alone digesters for 1990 to 2014 (only 1990 and 2005 are shown in
Table 7-46) was estimated by multiplying the count of estimated operating facilities (as presented in Table 7-47) by
the weighted average of waste digested in 2015 to 2019 collected through EPA's survey data (EPA 2018, 2019,
2021, 2023). Masked survey responses of food and non-food waste processed were shared with the Inventory
team by the EPA team leading the EPA AD Data Collection Surveys. This provided an accurate count of the number
of facilities that provided annual quantities of digested waste, which matters for the weighted average. The
weighted average applied to the current Inventory is calculated as follows for 1990 to 2014:
Equation 7-50: Weighted Average of Waste Processed
Weighted Average Waste Processed = ^ year — year
t—iyear Sum Of All Fac
where,
year = the year of data for the average waste processed and count of facilities in the numerator
W = total average waste processed in the respective survey year, food and non-food waste (short
tons).
Fac = the number of facilities that reported an amount of waste processed in the respective
survey year. Note the number of facilities that provided an annual quantity of waste
processed data was internally shared and differs from the total number of facilities that
responded to the EPA surveys as presented in EPA (2018, 2019, 2021, and 2023).
The number of facilities that reported annual quantities of waste digested to the EPA survey varies by year. The
masked data provided by the EPA AD survey data collection team include data for 41, 44, 42, 43, and 18 facilities
between 2015 to 2019, respectively. This data was used to calculate the weighted average of waste digested of
239,709 short tons.
Estimates of the quantity of waste digested for 1990 to 2014 are calculated by multiplying the weighted average of
waste digested from the masked survey data by the count of operating facilities in each year. This calculation
Waste 7-61
-------
assumes that each facility operates continuously from the first year of operation for the remainder of the time
series. Additional efforts will be made to quantify the number of operating facilities and estimates of the total
waste digested by year for future Inventories as described in the Planned Improvements section.
Estimates of the quantity digested for 2015 to 2019 were taken from EPA's AD survey data (EPA 2018, EPA 2019,
EPA 2021, and EPA 2023). In the 1990 to 2022 Inventory, the quantity of liquid, non-food waste was converted to
tons using a general conversion factor of 3.8 lbs/gallon.
The EPA (2023) report provides a significant increase in data granularity for stand-alone digesters compared to
earlier reports because food waste processed by the beverage sector is included as tons of food waste processed
as opposed to gallons of food waste processed in prior survey years. Detail on the sources and types of the liquid
food and non-food waste was not available in the 2015 to 2018 data to reliably convert the data to tons. However,
the 2019 data point provides some assurance that using a general conversion factor to convert liquid waste to tons
yields a more comprehensive estimate of total waste processed at stand-alone AD facilities.
The estimate of waste digested for 2020 to 2022 were extrapolated using the average of the waste digested from
the 2017 to 2019 survey data (EPA 2021, 2023) as a proxy. The average did not include data from 2015 and 2016
because there is a drop in the amount of waste digested by nearly 1 million tons between 2016 and 2017. The
quantities digested between 2015 and 2016 are similar, and quantities digested between 2017 and 2018 are
similar. The quantity digested for 2019 is nearly twice the amount of prior EPA survey years because food waste
from the beverage sector were able to be accurately converted to tons. Estimates for 2020 to 2022 will be updated
as future EPA survey reports are published.
Table 7-46: Estimated U.S. Waste Digested (kt) from 1990-2022
Activity
1990 2005
2018
2019
2020
2021
2022
Waste Digested
9881 2,964
8,326
15,567
10,702
11,947
11,947
The estimated count of operating facilities is calculated by summing the count of digesters that began operating by
year over the time series. The year a digester began operating is sourced from EPA (2021). This assumes all
facilities are in operation from their first year of operation throughout the remainder of the time series, including
facilities prior to 1990. This is likely an overestimate of facilities operating per year but does not necessarily
translate to an overestimate in the amount of waste processed because a weighted average of waste processed for
the surveyed facilities is applied to these years. The number of facilities in 1990 to 2014 are directly used in
calculating the emissions for those years.
Table 7-47: Estimated Number of Stand-Alone AD Facilities Operating from 1990-2022
Year
1990
2005
2018
2019
2020
2021
2022
Estimated Count of Operational Facilities
4
12
68
68
68
68
68
Uncertainty
The methodology applied for the 1990 to 2014 emissions estimates should be considered a starting point to build
on in future years if additional historical data become available. Five years of facility-provided data are available
(2015 to 2019) while the rest of the time series is estimated based on an assumption of facility counts and the
2015 to 2019 weighted average annual waste digested as calculated from survey data. The major limitations, and
uncertainty drivers in the emissions estimates, are related to the uncertainty in assumptions to ensure
completeness across the time series and the limitations in the EPA AD survey data, as described below:
1. The EPA AD surveys (EPA 2018; EPA 2019; EPA 2021; EPA 2023) did not receive a 100 percent response
rate, meaning that the survey data represent a portion, albeit the majority, of stand-alone digesters, and
annual waste processed. The methodology applied here did not attempt to estimate waste digested by
facilities that did not respond to the survey, which likely underestimates the quantity of waste digested
and Cl-Uemissions.
7-62 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
2. The EPA AD survey data (EPA 2018; EPA 2019) present both food and non-food waste digested. The non-
food waste was reported as liquid (gallons) and solid (tons). The quantity of liquid waste managed for
2015 and 2016, which is used as a proxy for 1990 to 2014, was converted to tons using a general
conversion factor of 3.8 lbs/gallon. This may slightly over- or underestimate the quantity of waste
digested and Cl-Uemissions between 1990 to 2018. This conversion was not made by EPA in the survey
report (EPA 2018). However, EPA (2021) did convert the liquid waste managed to tons for 2017 and 2018
using the general conversion factor of 3.8 lbs/gallon.
3. The assumption required to estimate the activity data for 1990 to 2014 may overestimate the number of
facilities in operation because it assumes that each facility operates from its start year for the entire time
series (i.e. facility closures are not taken into account). This introduces a large amount of uncertainty in
the estimates compared to years where there is directly reported survey data. It is unclear whether this
under- or over-estimates the quantity of waste digested and Cm emissions.
4. The most recent EPA AD survey data (EPA 2023) includes waste processed by the beverage sector, which
was not presented in prior survey years. No attempts were made to separately estimate and include this
waste stream in years prior to 2019 (i.e. the EPA 2023 survey). This means that annual Cm estimates for
1990 to 2018 are underestimated.
The estimated uncertainty from the 2006IPCC Guidelines is ±54 percent for the Approach 1 methodology.
Emissions from anaerobic digestion at stand-alone biogas facilities in 2022 were estimated to be between 6,175
and 20,586 MT CO2 Eq., which indicates a range of 54 percent below to 54 percent above the 2022 emission
estimate of Cm (see Table 7-48). A ±20 percent uncertainty factor is applied to the annual amount of material
digested (i.e., the activity data), which was developed with expert judgment (Bronstein 2021). A ±50 percent
default uncertainty factor is applied to the CH4 emission factor (IPCC 2006). Using the IPCC's error propagation
equation (Equation 3.1 in IPCC 2006 Volume 1, Chapter 3), the combined uncertainty percentage is ±54 percent.
Table 7-48: Approach 1 Quantitative Uncertainty Estimates for Emissions from Anaerobic
Digestion (MT CO2 Eq. and Percent)
Source
Gas
2022 Emission Estimate
(MTCO. Eq.)
Uncertainty Range Relative to Emission Estimate
(MTCOEq.) (%)
Lower Upper
Bound Bound
Lower Upper
Bound Bound
Anaerobic Digestion
at Biogas Facilities
ch4
13,380
6,175 20,586
-54% +54%
QA/QC and Verification
General QA/QC procedures were applied to data gathering and input, documentation, and calculations consistent
with the U.S. Inventory QA/QC Plan, which is in accordance with Vol. 1, Chapter 6 of the 2006 IPCC Guidelines (see
Annex 8 for more details). No errors were found for the current Inventory.
Recalculations Discussion
For the current Inventory, a methodological change was made whereby the CH4 emissions are considered equal to
leakage from the digester network of pipes. A leakage factor of 5 percent as recommended in IPCC 2006 is applied
to the CH4 generation estimate for all years in the time series. This methodological change applies to every year in
the time series and significantly reduces annual CH4 emissions estimates. Previously the EPA AD Survey data of
amount of biogas produced at AD facilities was used for the amount of gas recovered, with the remaining gas
assumed to be leaked or emitted. This method calculated higher emissions estimates, which showed most of the
gas generated at an AD was emitted, instead of used in biogas projects. This was inconsistent with the EPA AD
Survey findings that approximately 95 percent of stand-alone AD facilities use some or all biogas onsite and the
Waste 7-63
-------
IPCC guidance on default leakage from AD facilities. EPA will further investigate the survey data for the biogas
produced data point, since it indicates very low gas utilized as compared to this revised methodology.
The weighted average of waste digested was recalculated for the current Inventory to incorporate EPA AD survey
data from 2017 to 2019. The recalculation increased the weighted average annual waste digested from 216,494
short tons to 272,249 short tons, an approximately 25 percent increase. The weighted average is applied to the
estimated count of stand-along digesters operating between 1990 to 2014 and resulted in a 26 percent increase in
the amount of waste digested annually.
Additional recalculations were also made for the waste processed in 2019 to 2021. In the previous Inventory, the
amount of waste processed for 2019 to 2021 was extrapolated based on available survey data. With the
publication of survey data for 2019 (EPA 2023), the values for 2019 were replaced and the extrapolation for years
2020 to 2022 were updated. Recalculations for the amount of waste processed resulted in increases of 88 percent
in 2019, 30 percent in 2020, and 45 percent in 2021.
Despite the increase of waste processed across the time series, recalculations for this Inventory resulted in
significant decreases to the emissions estimates as compared to the previous 1990 through 2021 Inventory.
Emissions estimates were reduced by 93 percent annually between 1990 to 2014, and between 90 to 95 percent
between 2015 to 2021. For example, the net emissions estimate in 2021 decreased from 6.1 kt to 0.48 kt. The
decrease in emissions is driven by the methodological change described in the first paragraph.
Planned Improvements
EPA will continue to incorporate updated survey data from future EPA AD Data Collection Surveys when the survey
data are published. These revisions will change the estimated emissions for 2020 to 2022. Additionally, quality
control checks on the default emission factor used to determine Cm generation is in process.
EPA will also reassess how best to estimate annual waste processed using proxy data for years between the EPA
AD Data Collection Survey reports as needed (e.g., for 2020, 2021, 2022). The methodology described here
assumes the same average amount of waste is processed each year for 2020 through 2022.
EPA continues to seek out data sources to confirm the estimated number of operational facilities by year prior to
2015 and consider how best to estimate the quantity of waste processed per year by these facilities with the goal
of better estimating the annual quantity of waste digested between 1990 to 2014. Available data will also be
compiled where available for facilities that did not directly respond to the EPA AD Data Collection surveys for
completeness.
7.5 Waste Incineration (CRT Source
Category 5C1)
As stated earlier in this chapter, carbon dioxide (CO2), nitrous oxide (N2O), and methane (CH4) emissions from the
combustion of waste are accounted for in the Energy sector rather than in the Waste sector because almost all
combustion of municipal solid waste (MSW) in the United States occurs at waste-to-energy facilities where useful
energy is recovered. Similarly, the Energy sector also includes an estimate of emissions from burning waste tires
and hazardous industrial waste, because virtually all of the combustion occurs in industrial and utility boilers that
recover energy. The combustion of waste in the United States in 2022 resulted in 12.7 MMT CO2 Eq. of emissions.
For more details on emissions from the combustion of waste, see Section 3.3 of the Energy chapter.
Additional sources of emissions from waste combustion include non-hazardous industrial waste incineration and
medical waste incineration. As described in Annex 5 of this report, data are not readily available for these sources
and emission estimates are not provided.
7-64 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
An analysis of the likely level of medical waste incineration emissions was conducted based on a 2009 study of
hospital/ medical/ infectious waste incinerator (HMIWI) facilities in the United States (RTI 2009). Based on that
study's information of waste throughput and an analysis of the fossil-based composition of the waste, it was
determined that annual greenhouse gas emissions for medical waste incineration would be below 500 kt CO2 Eq.
per year and considered insignificant for the purposes of inventory reporting under the Paris Agreement and the
UNFCCC. More information on this analysis is provided in Annex 5.
Furthermore, an analysis was conducted on the likely level of sewage sludge incineration emissions based on the
total amount of sewage sludge generated and assumed percent incineration. Based on assumed amount of sludge
incinerated and non-CC>2 factors for solid biomass it was determined that annual greenhouse gas emissions for
sewage sludge incineration would be below 500 kt CO2 Eq. per year and considered insignificant for the purposes
of inventory reporting under the Paris Agreement and the UNFCCC. More information on this analysis is provided
in Annex 5.
7.6 Waste Sources of Precursor
Greenhouse Gases
In addition to the main greenhouse gases addressed above, waste generating and handling processes are also
sources of precursors to greenhouse gases. The reporting requirements of the Paris Agreement and the UNFCCC13
request that information should be provided on precursor emissions, which include carbon monoxide (CO),
nitrogen oxides (NOx), non-methane volatile organic compounds (NMVOCs), and sulfur dioxide (SO2). These gases
are not direct greenhouse gases, but can indirectly impact Earth's radiative balance by altering the concentrations
of other greenhouse gases (e.g., tropospheric ozone) and atmosphere aerosol (e.g., particulate sulfate). Total
emissions of NOx, CO, NMVOCs, and SO2 from waste sources for the years 1990 through 2022 are provided in Table
7-49.
Table 7-49: Emissions of NOx, CO, NMVOC, and SO2 from Waste (kt)
Gas/Source
1990
2005
2018
2019
2020
2021
2022
NOx
84
51 1
73
73
76
76
75
CO
1,028 !
1,178
1,182
1,182
1,342
1,343
1,343
NMVOCs
870
152
156
157
173
172
172
S02
36
20
23
23
33
32
31
Methodology and Time-Series Consistency
Emission estimates for 1990 through 2022 were obtained from data published on the National Emissions Inventory
(NEI) Air Pollutant Emissions Trends Data website (EPA 2023a). For Table 7-49, NEI reported emissions of CO, NOx,
SO2, and NMVOCs are recategorized from NEI Emissions Inventory System (EIS) sectors. The EIS sectors are
mapped to categories more closely aligned with reporting sectors and categories under the Paris Agreement and
the UNFCCC, based on discussions between the EPA Inventory and NEI staff (see crosswalk documented in Annex
6.3).14 EIS sectors mapped to the waste sector categories in this report include: waste disposal and recycling
13 See paragraph 51 of Annex to 18/CMA.l available online at:
https://unfccc.int/sites/default/files/resource/CMA2018 03a02E.pdf.
14 The NEI estimates and reports emissions from six criteria air pollutants (CAPs) and 187 hazardous air pollutants (HAPs) in
support of National Ambient Air Quality Standards. EPA reported CAP emission trends are grouped into 60 sectors and 15 Tier 1
Waste 7-65
-------
(landfills; publicly owned treatment works; industrial wastewater; treatment, storage, and disposal facilities; waste
incineration; and other).15 As described in the NEI Technical Support Documentation (TSD) (EPA 2023b), emissions
are estimated through a combination of emissions data submitted directly to the EPA by state, local, and tribal air
agencies, as well as additional information added by the Agency from EPA emissions programs, such as the
emission trading program, Toxics Release Inventory (TRI), and data collected during rule development or
compliance testing. Within the NEI, there is only one EIS sector for waste generating and handling processes, so
precursor estimates are aggregated in Table 7-49 for consistency with NEI reporting. Future presentations of this
data may disaggregate emissions so it better maps to reporting categories under the Paris Agreement and the
UNFCCC.
Methodological recalculations were applied to the entire time series to ensure time-series consistency from 1990
through 2022, which are described in detail in the NEI's TSD (EPA 2021). No quantitative estimates of uncertainty
were calculated for this source category.
source categories, which broadly cover similar source categories to those presented in this chapter. For reporting precursor
emissions in the common reporting tables (CRT), EPA has mapped and regrouped emissions of greenhouse gas precursors (CO,
NOx, S02, and NMVOCs) from NEI's EIS sectors to better align with NIR source categories, and to ensure consistency and
completeness to the extent possible. See Annex 6.3 for more information on this mapping.
15 Precursor emissions from waste incineration were reported in the Energy sector in the previous Inventory but are not
disaggregated from the Waste sector in this report.
7-66 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
8. Other
The United States does not report any greenhouse gas emissions under the Intergovernmental Panel on Climate
Change (IPCC) "Other" sector.
Other 8-1
-------
9. Recalculations and Improvements
Each year, many emission and sink estimates in the Inventory of U.S. Greenhouse Gas Emissions and Sinks are
recalculated and revised through the use of better methods and/or data with the goal of improving inventory
quality and reducing uncertainties, including improving the transparency, completeness, consistency, and overall
usefulness of the report. In this effort, the United States follows the 2006IPCC Guidelines (IPCC 2006) and its
refinements/supplements, which state:
"Both methodological changes and refinements over time are an essential part of improving inventory
quality. It is good practice to change or refine methods when available data have changed; the previously
used method is not consistent with the IPCC guidelines for that category; a category has become key; the
previously used method is insufficient to reflect mitigation activities in a transparent manner; the capacity
for inventory preparation has increased; improved inventory methods become available; and/or for
correction of errors."
When methodological changes have been implemented, the previous Inventory's time series (i.e., 1990 to 2021) is
assessed and potentially recalculated to reflect the change, per guidance in IPCC (2006, 2019). Changes in historical
data are often the result of changes in statistical data supplied by other agencies, and these changes do not
necessarily impact the entire time series.
The results of all methodological changes and historical data updates made in the current Inventory in calculating
CCh-equivalent U.S. greenhouse gas emissions and sinks are presented in Figure 9-2, while impacts on both total
and net emissions by gas are presented in Table 9-1 and Table 9-2. Collectively, these changes resulted in an
average annual decrease of 114.8 million metric tons of carbon dioxide equivalent (MMT CO2 Eq.) (1.9 percent) in
net total emissions relative to the previously published Inventory (i.e., the 1990 to 2021 report) in units of MMT
CO2 Eq.
Recalculations and Improvements 9-1
-------
Figure 9-1: Impacts of Recalculations on Net Emissions
Table 9-1: Overall Impact of Recalculations by Gas Compared to Previous Inventory
Average
Annual
Gas/Source
1990
2005
2018
2019
2020
2021
Change
C02
10.7
(5.3)
(15.6)
(27.7)
(25.7)
(15.0)
(4.6)
CH4a
3.0
4.4
(2.6)
(13.4)
(6.9)
(7.0)
0.5
N2Oa
1.9
3.4
9.3
6.1
2.3
4.9
3.9
HFCs and PFCs
34.0
14.8
6.8
7.2
6.1
5.7
21.2
Total Gross Emissions (Sources)
49.6
17.3
(2.1)
(27.8)
(24.2)
(11.4)
21.0
Change in LULUCF Total Net Fluxb
(95.8)
(123.0)
(148.8)
(153.4)
(120.2)
(151.4)
(133.6)
LULUCF Emissions0
0.0
(3.5)
(1.6)
(6.2)
(8.0)
(4.9)
(2.2)
ch4
(0.4)
(2.8)
(1.8)
(4.4)
(6.1)
(3.8)
(2.0)
n2o
0.4
(0.7)
0.2
(1.8)
(1.9)
(1.1)
(0.2)
Change in LULUCF Sector Net Totald
(95.7)
(126.5)
(150.4)
(159.5)
(128.2)
(156.3)
(135.8)
Net Emissions (Sources and Sinks)
(46.2)
(109.3)
(152.4)
(187.3)
(152.4)
(167.8)
(114.8)
a Does not include CH4 and N20 emissions from LULUCF.
b LULUCF carbon stock change includes any C stock gains and losses from all land use and land use conversion categories
c LULUCF emissions include the CH4 and N20 emissions reported for peatlands remaining peatlands, forest fires, drained
organic soils, grassland fires, and coastal wetlands remaining coastal wetlands; CH4 emissions from land converted to
coastal wetlands; and N20 emissions from forest soils and settlement soils.
d The LULUCF sector net total is the net sum of all CH4 and N20 emissions to the atmosphere plus net carbon stock
changes. More detail on the impacts of recalculations on the LULUCF sector can be found in Table 9-5.
Notes: Parentheses indicate negative values. Totals may not sum due to independent rounding.
9-2 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
Table 9-2: Overall Impact of Recalculations by Sector Compared to Previous Inventory
Average
Annual
Gas/Source
1990
2005
2018
2019
2020
2021
Change
Energy
16.81
1.6
(13.4)
(32.1)
(25.6)
(16.7)
(0.2)
IPPU
33.5
15.3
5.0
5.1
4.7
5.2
20.4
Agriculture
3.1
4.1
12.9
5.6
2.4
6.7
5.3
LULUCF
(95.8)
(126.5)
(150.4)
(159.5)
(128.2)
(156.3)
(135.8)
Waste
0.0
(0.1)
(0.5)
(0.2)
0.2
0.0
(0.1)
Total Gross Emissions (Sources)
49.6
17.3
(2.1)
(27.8)
(24.2)
(11.4)
21.0
Net Emissions (Sources and Sinks)
(46.2)
(109.3)
(152.4)
(187.3)
(152.4)
(167.8)
(114.8)
Notes: Parentheses indicate negative values. Totals may not sum due to independent rounding.
Table 9-4 and Table 9-5 include the category-level quantitative effects of methodological changes and historical
data updates made in the current Inventory in calculating CC>2-equivalent U.S. greenhouse gas emissions by gas
across all sectors. To understand the details of any specific recalculation or methodological improvement, see the
Recalculations sections within each source/sink categories' section found in Chapters 3 through 7 of this report. A
discussion of Inventory improvements in response to review processes is described in Annex 8.
Key Recalculations and Improvements for 1990-2022 Inventory
The current Inventory includes new categories that improve completeness of the national estimates. Specifically,
the current report includes CO2 emissions from ceramics production and non-metallurgical magnesia production
within other process use of carbonates category, fluorinated gases from fluorochemical production other than
HCFC-22 within the fluorochemical production category, and managed forest land in Hawaii and several U.S.
Territories.1 The report also now includes SF6 and PFCs from product uses.
The following source and sink categories underwent the most significant methodological and historical data
changes. A brief summary of the recalculations and/or improvements undertaken are provided for these
categories.
Table 9-3: Key Recalculations
Impact of
Average Impact over Time Series
Reason for Recalculation or
Recalculation on
Sector
Category
Improvement
2021 Value
Percent (MMT CO¦ Eq.)
Accuracy. Use of new and
Forest Land
updated data and methods from
LULUCF
Remaining Forest
Land (C02)
the USFS Forest Inventory and
Analysis program, see details in
Chapter 6.2.
148.8
+21.7% 159.1
LULUCF
Land Converted
to Grassland
(C02)
Accuracy. Use of new activity
data and methods from FIA,
USDA-NRCS NRI and DayCent
49.2
-237.1% 53.1
model, see details in Chapter 6.7.
Accuracy. Use of new activity
LULUCF
Land Converted
data and methods from FIA,
21.6
-36.6% 20.7
to Cropland (C02)
USDA-NRCS NRI and DayCent
model, see details in Chapter 6.5.
1 American Samoa, Guam, Northern Marianas Islands, U.S. Virgin Islands, and Puerto Rico.
Recalculations and Improvements 9-3
-------
LULUCF
Grassland
Remaining
Grassland (CO2)
Accuracy. Use of new data from
USDA-NRCS NRI, and re-
calibration of the DayCent
model, see details in Chapter 6.6.
0.6
+1,850.1%
10.7
Accuracy and Consistency. Use of
new, updated, and recategorized
Energy
Non-Energy Use
data from U.S. International
28.6
-8.2%
10.1
of Fuels (C02)
Trade Commission, EIA and other
data sources (ACC), see details in
Chapter 3.2.
LULUCF
Land Converted
to Settlements
(C02)
Accuracy. Use of new data from
USDA-NRCS NRI, FIA, and
extended time series, see details
12.8
-9.3%
7.7
in Chapter 6.11.
Completeness. Inclusion of new
Fluorochemical
subcategory fluorochemical
IPPU
Production
(HFCs)
production other than HCFC-22,
see details in Chapter 4.14 and
4.15.
1.7
+58.7%
6.8
Accuracy and Consistency. Use of
Energy
Fossil Fuel
updated data and alignment of
15.2
+0.1%
Combustion (C02)
methodology from EIA, see
details in Chapter 3.1.
Accuracy. Use of new data from
Cropland
USDA-NRCS NRI and the OpTIS
LULUCF
Remaining
Cropland (C02)
remote-sensing data, and
methods to extend time series,
see details in Chapter 6.4.
13.0
+26.4%
4.2
Accuracy. Use of updated time
series data for land
Agricultural Soil
representation, re-calibration of
Agriculture
Management
(N20)
DayCent model, and updated
cropland management
parameters, see details in
Chapter 5.4.
3.9
+1.1%
3.3
Energy
Petroleum
Systems (CH4)
Accuracy. Use of additional data
from GHGRP, see details in
Chapter 3.6.
1.5
-4.6%
2.5
Wetlands
Accuracy. Use of new data and
LULUCF
Remaining
Wetlands (C02)
updated emissions factors, see
details in Chapter 1.8.
2.3
+2.2%
2.4
9-4 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
Figure 9-2: Impacts of Recalculations to U.S. Greenhouse Gas Emissions and Sinks by Sector
o
u
100
80
60
40
20
0
-20
-40
-60
-80
-100
-120
-140
-160
-180
- Change in Net Total Emissions
¦ Agriculture
¦ Energy
¦ Industrial Processes and Product Use
¦ LULUCF Sector Net Total
¦ Waste
-200
Oi-»rsimrrLovDrv
ChO^CTiO^O^O^G^O^
cr»cr>crtcr»cr>o^cr»
CO at
crt cn
cr» cn
H (M m t LT) UD N
o o o o o o o
o o o o o
co cr>
o o
o o
OHrNni-in^Ncoo^
oooooooo
o
cn rsi
HHHHHHHHHHfNfNOJfN(NrNfNf\rN(NfNfNl(NfNfMfNrMfN|
o o o o
(N IN IN (N
Table 9-4: Revisions to U.S. Greenhouse Gas Emissions (MMT CO2 Eq.)
Gas/Source
1990
2005
2018
2019
2020
2021
Average
Annual
Change
co2
10.7
(5-3)
(15.6)
(27.7)
(25.7)
(15.0)
(4.6)
Fossil Fuel Combustion
24.0
(3.2)
(1.7)
(3.3)
(3.2)
15.2
5.8
Electric Power Sector
NC
NC
NC
NC
+
+
+
Transportation
NC
NC
0.2
0.2
0.3
1.1
0.1
Industrial
24.0
(3,2)
(3.1)
(6.1)
(5.9)
4.8
5.2
Residential
(+)
+
0.7
1.5
1.6
4.7
0.3
Commercial
(+)
+
0.5
1.0
0.8
4.5
0.2
U.S. Territories
(+)
(+)
(+)
(+)
(+)
(+)
(+)
Non-Energy Use of Fuels
(13,3)
(3,9)
(11.1)
(21.1)
(21.5)
(28.6)
(10.1)
Natural Gas Systems
0.2
1.3
(+)
(+)
0.3
(0.4)
0.4
Cement Production
NC
NC
NC
NC
NC
NC
NC
Recalculations and Improvements 9-5
-------
Lime Production
NC
NC I
NC
NC
NC
NC
NC
Other Process Uses of Carbonates3
0.9
1.0
0.6
0.6
0.6
0.6
0.8
Glass Production
03:
(+):
NC
NC
NC
NC
+
Soda Ash Production
NC
NC
NC
NC
NC
NC
NC
Carbon Dioxide Consumption
NC I
nc =
NC
NC
NC
NC
NC
Incineration of Waste
NC
NC
NC
NC
NC
NC
NC
Titanium Dioxide Production
NC I
nc :
NC
(0.1)
0.1
NC
+
Aluminum Production
NC
NC
NC
NC
NC
NC
NC
Iron and Steel Production & Metallurgical Coke
I
I
Production
+
+
+
+
+
0.2
+
Ferroalloy Production
NC
NC
NC
NC
NC
NC
NC
Ammonia Production
NC 1
NC
NC
NC
NC
(+)
(+)
Urea Consumption for Non-Agricultural Purposes
NC
NC
+
(+)
(+)
1.6
0.1
Phosphoric Acid Production
NC
NC |
NC
NC
NC
(+)
(+)
Petrochemical Production
(1-5)
(0.5)
(2.1)
(2.2)
(1.9)
(2.5)
(1.4)
Carbide Production and Consumption
NC
NC
NC
NC
NC
NC
NC
Lead Production
NC
NC
NC
NC
(+)
(+)
(+)
Zinc Production
nc =
nc !!!!
NC
NC
NC
+
+
Petroleum Systems
0.1
(+)
(1.3)
(1.4)
(0.1)
(0.5)
(0.1)
Abandoned Oil and Gas Wells
IE
+ !::
+
+
+
+
+
Magnesium Production and Processing
+
+
NC
NC
NC
(+)
+
Liming
NC :
NC 1
NC
NC
(+)
(0.7)
(+)
Urea Fertilization
NC
NC
(+)
+
+
+
(+)
Coal Mining
<+> 1
(+>:
(+)
(+)
(+)
(+)
(+)
Substitution of Ozone Depleting Substances
NC
NC
NC
NC
NC
NC
NC
Biomass and Biodiesel Consumptionb
NC I
NC j
M
M
(9.9)
(12.3)
(0.7)
International Bunker Fuelsc
NC
NC
NC
NC
NC
NC
NC
n
z
3.0
4.4
(2.6)
(13.4)
(6.9)
(7.0)
0.5
Stationary Combustion
+
(+)
+
+
(0.8)
(0.9)
(+)
Mobile Combustion
+ II
(o.D 1:
(+)
(+)
(+)
(+)
(0.1)
Coal Mining
NC
(0.4)
NC
(+)
NC
(+)
(+)
Abandoned Underground Coal Mines
nc ;
NC =
NC
NC
NC
(0.1)
(+)
Natural Gas Systems
3.7
6.9
(4.0)
(4.9)
(5.0)
(6.8)
1.9
Petroleum Systems
(1.9) 1:
(2.7) I
(1.6)
(7.7)
(1.2)
(1.5)
(2.5)
Abandoned Oil and Gas Wells
0.1
0.2
0.2
0.1
0.2
0.3
0.2
Petrochemical Production
(0.2) I
(o.D 1:
(0.3)
(0.4)
(0.3)
(0.4)
(0.2)
Carbide Production and Consumption
NC
NC
NC
NC
NC
NC
NC
Iron and Steel Production & Metallurgical Coke
I
NC I
in:
1
Production
NC I
NC
NC
+
(+)
+
Ferroalloy Production
NC
NC
NC
NC
NC
NC
NC
Enteric Fermentation
-
IS
+
(+)
(+)
0.1
1.5
0.1
Manure Management
0.1
0.1
1.2
0.9
0.2
0.4
0.3
Rice Cultivation
1.0 1
0.4 j
2.5
(1.2)
1.0
1.5
1.0
Field Burning of Agricultural Residues
0.1
0.2
0.1
0.2
0.1
0.1
0.1
Landfills
NC
w =:
(0.4)
(0.3)
(0.7)
(0.6)
(+)
Wastewater Treatment
(+)
(+)
(+)
(+)
(0.4)
(0.4)
(+)
Composting
nc :::
NCl
NC
NC
NC
NC
NC
Anaerobic Digestion at Biogas Facilities
(+)
(+)
(0.2)
(0.2)
(0.2)
(0.2)
(0.1)
Incineration of Waste
NC i
NC '
NC
NC
NC
NC
NC
International Bunker Fuelsc
NC
NC
NC
NC
NC
NC
NC
N2Od
1.9
3.4
9.3
6.1
2.3
4.9
3.9
Stationary Combustion
0.1
(+)
(+)
(+)
(0.1)
(0.1)
+
Mobile Combustion
+ |
+|l
0.1
0.1
+
0.1
0.1
Adipic Acid Production
NC
NC
(+)
NC
NC
NC
(+)
9-6 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
Nitric Acid Production
NC
NC 1
NC
NC
NC
NC
NC
Manure Management
1.0
0.7
(0.6)
(0.6)
(0.6)
(0.3)
0.5
Agricultural Soil Management
0-8 1
2-6 |
9.7
6.3
1.6
3.9
3.3
Field Burning of Agricultural Residues
+
+
+
+
+
+
+
Wastewater Treatment
NC i
NC I
+
0.3
1.4
1.1
0.1
N20 from Product Uses
NC
NC
NC
NC
NC
NC
NC
Caprolactam, Glyoxal, and Glyoxylic Acid
I
I
Production
NC 1
NC !
NC
NC
(+)
(+)
(+)
Incineration of Waste
NC
NC
NC
NC
NC
NC
NC
Composting
nc
NC
NC
NC
NC
NC
NC
Electronics Industry
NC
NC
+
+
+
+
+
Natural Gas Systems
+ si
+::
+
+
+
+
+
Petroleum Systems
(+)
(+>
+
(+)
(+)
(+)
(+)
International Bunker Fuelsc
nc :
NC
NC
NC
NC
NC
NC
HFCs, PFCs, SF6 and NF3
34.0
14.8
6.8
7.2
6.1
5.7
21.2
HFCs
8.7
5.4
3.0
2.7
2.1
1.9
6.9
Substitution of Ozone Depleting Substances
NC
0.1
0.1
0.1
0.1
0.2
0.1
Fluorochemical Production6
8.7
5-3 1
2.9
2.6
2.0
1.7
6.8
Electronics Industry
(+)
(+)
(+)
+
(+)
+
+
Magnesium Production and Processing
nc
NCl
NC
NC
NC
NC
NC
PFCs
17.7
4.2
3.1
3.3
2.7
2.8
9.3
Aluminum Production
NC 1
NC 1
NC
NC
NC
+
+
Fluorochemical Production
17.5*
4.0*
2.9*
3.0*
2.5*
2.6*
9.1*
Electronics Industry
+ iiiii
0.1
0.1
0.1
+
+
Substitution of Ozone Depleting Substances
NC
+
(+)
(+)
(+)
(+)
(+)
SF6 and PFCs from Other Product Use
0.1*
O.i* !
0.2*
0.2*
0.2*
0.1*
0.1*
Electrical Equipment
+
(+)
NC
(+)
+
(+)
(+)
sf6
7.4
4.7
!!!!!!!
0.5
0.6
0.6
0.4
4.6
Electrical Equipment
NC
+
(0.2)
(+)
+
+
(+)
SF6 and PFCs from Other Product Use
1.3*
1.3* ;
0.8*
0.6*
0.5*
0.4*
1.2*
Fluorochemical Production
5.8*
3.3*
+
+
+
+
3.3*
Electronics Industry
NC =
+
+
+
(+)
+
Magnesium Production and Processing
0.2
0.1
NC
NC
NC
+
0.1
nf3
Q 3 """l
0.6
0.1
0.6
0.7
0.5
0.5
Electronics Industry
NC
NC
+
+
+
(+)
+
Fluorochemical Production
0.3* i
0.6* i
0.1*
0.6*
0.7*
0.5*
0.5*
Total Gross Emissions (Sources)
49.6
17.3
(2.1)
(27.8)
(24.2)
(11.4)
21.0
Percent Change in Total Emissions
0.8%
0.2%
0.0%
-0.4%
-0.4%
(+)
0.3%
Change in LULUCF Sector Net Total'
(95.8)
(126.5)
(150.4)
(159.5)
(128.2)
(156.3)
(135.8)
Net Emissions (Sources and Sinks)
(46.2)
(109.3)
(152.4)
(187.3)
(152.4)
(167.8)
(114.8)
Percent Change in Net Emissions
-0.8%
-1.6%
-2.5%
-3.2%
-2.9%
-3.0%
-1.9%
NC (No Change)
+ Absolute value does not exceed 0.05 MMT C02 Eq. or 0.05 percent.
* Indicates a new source for the current Inventory year. Emissions from new sources are captured in net emissions and
percent change totals.
a Category includes emissions from ceramics production (new subcategory, not estimated in the previous Inventory), other
uses of soda ash, and non-metallurgical magnesia (new subcategory, not estimated in the previous Inventory) in the current
Inventory.
b Emissions from biomass and biofuel 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 LULUCF.
c Emissions from international bunker fuels are not included in totals.
d LULUCF emissions of CH4 and N20 are reported separately from gross emissions totals in Table 9-4. LULUCF emissions
include the CH4 and N20 emissions reported for peatlands remaining peatlands, forest fires, drained organic soils, grassland
Recalculations and Improvements 9-7
-------
fires, and coastal wetlands remaining coastal wetlands; CH4 emissions from land converted to coastal wetlands; and N20
emissions from forest soils and settlement soils.
e This category was reported as HCFC-22 production in the 1990 to 2021 Inventory.
f The LULUCF sector net total is the net sum of all CH4 and N20 emissions to the atmosphere plus net carbon stock changes.
More detail on the impacts of recalculations on the LULUCF sector can be found in Table 9-5.
Notes: Totals may not sum due to independent rounding. Parentheses indicate negative values.
Table 9-5: Revisions to U.S. Greenhouse Gas Emissions and Removals (Net Flux) from Land
Use, Land-Use Change, and Forestry (MMT CO2 Eq.)
Land-Use Category
1990
2005
2017
2018
2019
2020
Average
Annual
Change
Forest Land Remaining Forest Land
(153.1)
(164.6)
(170.4)
(169.0)
(162.4)
(153.4)
(160.7)
Changes in Forest Carbon Stocks3
(153.3)
(161.8) 1
(169.1)
(163.8)
(154.6)
(148.8)
(159.1)
Non-C02 Emissions from Forest Firesb
0.3
(2.8)
(1.3)
(5.1)
(7.7)
(4.5)
(1.6)
N20 Emissions from Forest Soils0
+
(+) 5
(+)
(+)
(+)
(+)
(+)
Non-C02 Emissions from Drained Organic Soilsd
NC
NC
NC
NC
NC
NC
NC
Land Converted to Forest Land
(1.8)
(1.7)
(2.1)
(2.0)
(2.0)
(2.1)
(1.8)
Changes in Forest Carbon Stocks6
(1.8)
(1.7)
(2.1)
(2.0)
(2.0)
(2.1)
(1.8)
Cropland Remaining Cropland
18.1
(2.6)
(1.2)
(4.9)
14.5
(13.0)
(4.2)
Changes in Mineral and Organic Soil Carbon
Stocks
18.1
(2.6)
(1.2)
(4.9)
14.5
(13.0)
(4.2)
Land Converted to Cropland
(9.4)
(20.1)
(24.4)
(24.9)
(27.4)
(21.6)
(20.7)
Changes in all Ecosystem Carbon Stocks'
(9.4)
(20.1)
(24.4)
(24.9)
(27.4)
(21.6)
(20.7)
Grassland Remaining Grassland
15.7
13.2
17.8
14.2
10.5
0.9
10.8
Changes in Mineral and Organic Soil Carbon
Stocks
15.7
13.0
17.3
14.5
10.0
0.6
10.7
Non-C02 Emissions from Grassland Fires8
0.1
0.11
0.5
(0.3)
0.4
0.3
0.1
Land Converted to Grassland
41.9
61.9
49.4
48.7
54.6
49.2
53.1
Changes in all Ecosystem Carbon Stocks'
41.9
61-9 1
49.4
48.7
54.6
49.2
53.1
Wetlands Remaining Wetlands
(4.7)
(3.7)
(3.7)
(3.7)
(3.7)
(3.7)
(3.9)
Changes in Organic Soil Carbon Stocks in
I
iiiiiii
I
Peatlands
nc :
NC 1
(0.1)
(0.1)
(0.1)
(0.2)
(+)
Changes in Biomass, DOM, and Soil Carbon Stocks
in Coastal Wetlands
(2.4)
(2.4)
(2.3)
(2.3)
(2.3)
(2.3)
(2.4)
CH4 Emissions from Coastal Wetlands Remaining
!!!!!!!
mm!
1
I
Coastal Wetlands
NC
NC 1
NC
NC
NC
NC
NC
N20 Emissions from Coastal Wetlands Remaining
Coastal Wetlands
NC
NC
NC
NC
NC
NC
NC
Non-C02 Emissions from Peatlands Remaining
I
I!!!!!:
1
1
Peatlands
NC |
NC
mill!
(+)
(+)
(+)
(+)
(+)
CH4 Emissions from Flooded Land Remaining
Flooded Land
(2.3)
(1.3)
(1.2)
(1.2)
(1.2)
(1.2)
(1.5)
Land Converted to Wetlands
3.9
0.4
(0.1)
(0.1)
0.1
0.1
1.1
Changes in Biomass, DOM, and Soil Carbon Stocks
in Land Converted to Coastal Wetlands
(+)
(+)
(+)
(+)
(+)
(+)
(+)
CH4 Emissions from Land Converted to Coastal
I
Wetlands
NC
NC ¦
NC
NC
NC
NC
NC
Changes in Land Converted to Flooded Land
2.2
0.3
(+)
(+)
0.1
0.1
0.6
CH4 Emissions from Land Converted to Flooded
1
mm;
1
Land
1.8
0.2 1
(+)
(+)
+
+
0.5
Settlements Remaining Settlements
(1.4)
(1.4)
(6.0)
(7.1)
(0.2)
0.2
(1.8)
Changes in Organic Soil Carbon Stocks
(1.4) 1
(2.1)5
(1.6)
(1.3)
(0.7)
(0.4)
(1.8)
9-8 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
Changes in Settlement Tree Carbon Stocks
(0.2)
0.4
(4.9)
(6.2)
0.1
+
(0.3)
Changes in Yard Trimming and Food Scrap Carbon
!!!!!!!
mm!
1
NC 1
Stocks in Landfills
NC
NC
NC
NC
0.1
+
N20 Emissions from Settlement Soilsh
0.3
0.3
0.5
0.5
0.5
0.5
0.4
Land Converted to Settlements
(5.2)
(7.9)
(9.7)
(10.9)
(12.2)
(12.8)
(7.7)
Changes in all Ecosystem Carbon Stocks'
(5.2)
(7.9)
(9.7)
(10.9)
(12.2)
(12.8)
(7.7)
Change in LULUCF Total Net Flux'
(95.8)
(123.0)
(148.8)
(153.4)
(120.2)
(151.4)
(133.6)
Change in LULUCF Emissions1
+
(3.6)
(1.6)
(6.2)
(8.0)
(4.9)
(2.2)
CH4 (0.4) (2.8) (1.8) (4.4) (6.1) (3.8) (2.0)
N20 0.4 (0.7) 0.2 (1.8) (1.9) (1.1) (0.2)
Change in LULUCF Sector Net Total1 (95.87 (126.6) (150.4) (159.5) (128.2) (156.3) (135.8)"
Percent Change in LULUCF Sector Net Total -10.9% -16.2% -19.7% -22.7% -16.5% -20.7% -17.0%
NC (No Change)
+ Absolute value does not exceed 0.05 MMT C02 Eq. or 0.05 percent.
a Includes the net changes to carbon stocks stored in all forest ecosystem pools (estimates include carbon stock changes from
drained organic soils from both forest land remaining forest land and land converted to forest land) and harvested wood
products.
b Estimates include CH4 and N20 emissions from fires on both forest land remaining forest land and land converted to forest
land.
c Estimates include N20 emissions from N fertilizer additions on both forest land remaining forest land and land
converted to forest land.
d Estimates include CH4 and N20 emissions from drained organic soils on both forest land remaining forest land and land
converted to forest land. Carbon stock changes from drained organic soils are included with the forest land remaining forest
land forest ecosystem pools.
e Includes the net changes to carbon stocks stored in all forest ecosystem pools.
f Includes changes in mineral and organic soil carbon stocks for all land use conversions to cropland, grassland, and
settlements, respectively. Also includes aboveground/belowground biomass, dead wood, and litter carbon stock changes for
conversion of forest land to cropland, grassland, and settlements.
g Estimates include CH4 and N20 emissions from fires on both grassland remaining grassland and land converted to grassland.
h Estimates include N20 emissions from N fertilizer additions on both settlements remaining settlements and land
converted to settlements because it is not possible to separate the activity data at this time.
' LULUCF carbon stock change includes any C stock gains and losses from all land use and land use conversion categories.
' LULUCF emissions include the CH4 and N20 emissions reported for peatlands remaining peatlands, forest fires, drained
organic soils, grassland fires, and coastal wetlands remaining coastal wetlands; CH4 emissions from land converted to coastal
wetlands, flooded land remaining flooded land, and land converted to flooded land; and N20 emissions from forest soils and
settlement soils.
k The LULUCF sector net total is the net sum of all LULUCF CH4 and N20 emissions to the atmosphere plus LULUCF net carbon
stock changes in units of MMT C02 Eq.
Notes: Totals may not sum due to independent rounding. Parentheses indicate negative values.
Recalculations and Improvements 9-9
-------
10. References and Abbreviations
Executive Summary
BEA (2024) 2022 Comprehensive Revision of the National Income and Product Accounts: Current-dollar and "real"
GDP, 1929-2022. Bureau of Economic Analysis (BEA), U.S. Department of Commerce, Washington, D.C. Available
online at: http://www.bea.gOv/national/index.htm#gdp.
EIA (2024) Monthly Energy Review, February 2024. Energy Information Administration, U.S. Department of Energy,
Washington, D.C. DOE/EIA-0035(2023/11).
IEA (2022) Energy related CO2 emissions, 2022, International Energy Agency, Paris. Available online at:
https://www.iea.org/reports/co2-emissions-in-2022.
IPCC (2021) Climate Change 2021: The Physical Science Basis. Contribution of Working Group I to the Sixth
Assessment Report of the Intergovernmental Panel on Climate Change. [Masson-Delmotte, V., P. Zhai, A. Pirani, S.L
Connors, C. Pean, S. Berger, N. Caud, Y. Chen, L. Goldfarb, M.I. Gomis, M. Huang, K. Leitzell, E. Lonnoy, J.B.R.
Matthews, T.K. Maycock, T. Waterfield, O. Yelekgi, R. Yu, and B. Zhou (eds.)]. Cambridge University Press,
Cambridge, United Kingdom and New York, NY, USA, 2391 pp. doi:10.1017/9781009157896.
IPCC (2019) 2019 Refinement to the 2006 IPCC Guidelines for National Greenhouse Gas Inventories. Calvo Buendia,
E., Tanabe, K., Kranjc, A., Baasansuren, J., Fukuda, M., Ngarize, S., Osako, A., Pyrozhenko, Y., Shermanau, P. and
Federici, S. (eds). Published: IPCC, Switzerland.
IPCC (2013) Climate Change 2013: The Physical Science Basis. Contribution of Working Group I to the Fifth
Assessment Report of the Intergovernmental Panel on Climate Change. [Stocker, T.F., D. Qin, G.-K., Plattner, M.
Tignor, S.K. Allen, J. Boschung, A. Nauels, Y. Xia, V. Bex and P.M. Midgley (eds.)]. Cambridge University Press,
Cambridge, United Kingdom and New York, NY, USA, 1535 pp.
IPCC (2007) Climate Change 2007: The Physical Science Basis. Contribution of Working Group I to the Fourth
Assessment Report of the Intergovernmental Panel on Climate Change. [S. Solomon, D. Qin, M. Manning, Z. Chen,
M. Marquis, K.B. Averyt, M. Tignor and H.L Miller (eds.)]. Cambridge University Press. Cambridge, United Kingdom
996 pp.
IPCC (2006) 2006 IPCC Guidelines for National Greenhouse Gas Inventories. The National Greenhouse Gas
Inventories Programme, The Intergovernmental Panel on Climate Change. [H.S. Eggleston, L. Buendia, K. Miwa, T.
Ngara, and K. Tanabe (eds.)]. Hayama, Kanagawa, Japan.
National Academies of Sciences, Engineering, and Medicine (2018) Improving characterization of anthropogenic
methane emissions in the United States. Washington, DC: The National Academies Press. Available online at:
https://doi.org/10.17226/24987.
References and Abbreviations 10-1
-------
National Research Council (2010) Verifying greenhouse gas emissions: methods to support international climate
agreements. Washington, DC: The National Academies Press. Available online at: https://doi.org/10.17226/12883.
NOAA/ESRL (2024a) Trends in Atmospheric Carbon Dioxide. Available online at: https://gml.noaa.gov/ccgg/trends/.
05 January 2024.
NOAA/ESRL (2024b) Trends in Atmospheric Methane. Available online at: https://gml.noaa.gov/ccgg/trends ch4/.
05 January 2024.
NOAA/ESRL (2024c) Trends in Atmospheric Nitrous Oxide. Available online at:
https://gml.noaa.gov/ccgg/trends n2o/. 05 January 2024.
UNFCCC (2014) Report of the Conference of the Parties on its Nineteenth Session, Held in Warsaw from 11 to 23
November 2013. (FCCC/CP/2013/10/Add.3). January 31, 2014. Available online at:
http://unfccc.int/resource/docs/2013/copl9/eng/10a03.pdf.
U.S. Census Bureau (2024) U.S. Census Bureau International Database (IDB). Available online at:
https://www.census.gov/programs-survevs/intemational-programs.html.
Introduction
IPCC (2021) Climate Change 2021: The Physical Science Basis. Contribution of Working Group I to the Sixth
Assessment Report of the Intergovernmental Panel on Climate Change. [Masson-Delmotte, V., P. Zhai, A. Pirani, S.L
Connors, C. Pean, S. Berger, N. Caud, Y. Chen, L. Goldfarb, M.I. Gomis, M. Huang, K. Leitzell, E. Lonnoy, J.B.R.
Matthews, T.K. Maycock, T. Waterfield, O. Yelekgi, R. Yu, and B. Zhou (eds.)]. Cambridge University Press,
Cambridge, United Kingdom and New York, NY, USA, 2391 pp. doi:10.1017/9781009157896.
IPCC (2014) Climate Change 2014: Mitigation of Climate Change. Contribution of Working Group III to the Fifth
Assessment Report of the Intergovernmental Panel on Climate Change [Edenhofer, O., R. Pichs-Madruga, Y.
Sokona, J. Minx, E. Farahani, S. Kadner, K. Seyboth, A. Adler, I. Baum, S. Brunner, P. Eickemeier, B. Kriemann, J.
Savolainen, S. Schlomer, C. von Stechow, and T. Zwickel (eds.)]. Cambridge University Press, Cambridge, United
Kingdom and New York, NY, USA, 1435 pp.
IPCC (2013) Climate Change 2013: The Physical Science Basis. Contribution of Working Group I to the Fifth
Assessment Report of the Intergovernmental Panel on Climate Change [Stocker, T.F., D. Qin, G.-K. Plattner, M.
Tignor, S.K. Allen, J. Boschung, A. Nauels, Y. Xia, V. Bex and P.M. Midgley (eds.)]. Cambridge University Press,
Cambridge, United Kingdom and New York, NY, USA, 1535 pp.
IPCC (2007) Climate Change 2007: The Physical Science Basis. Contribution of Working Group I to the Fourth
Assessment Report of the Intergovernmental Panel on Climate Change. [S. Solomon, D. Qin, M. Manning, Z. Chen,
M. Marquis, K.B. Averyt, M. Tignor and H.L Miller (eds.)]. Cambridge University Press. Cambridge, United Kingdom
996 pp.
IPCC (2006) 2006 IPCC Guidelines for National Greenhouse Gas Inventories. The National Greenhouse Gas
Inventories Programme, The Intergovernmental Panel on Climate Change. [H.S. Eggleston, L. Buendia, K. Miwa, T.
Ngara, and K. Tanabe (eds.)]. Hayama, Kanagawa, Japan.
IPCC (2001) Climate Change 2001: The Scientific Basis. Intergovernmental Panel on Climate Change. [J.T. Houghton,
Y. Ding, D.J. Griggs, M. Noguer, P.J. van der Linden, X. Dai, C.A. Johnson, and K. Maskell (eds.)]. Cambridge
University Press. Cambridge, United Kingdom.
IPCC/TEAP (2005) Special Report: Safeguarding the Ozone Layer and the Global Climate System, Chapter 4:
Refrigeration. 2005. Available online at: https://www.ipcc.ch/site/assets/uploads/2018/03/sroc04-l.pdf.
10-2 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
Meinhausen, M., et al. (2017) Historical Greenhouse Gas Concentrations for Climate Modeling (CMIP6). Available
online at: https://gmd.copemicus.org/articles/10/2057/2017/.
NOAA (2017) Vital Signs of the Planet. Available online at: http://climate.nasa.gov/causes/. Accessed on 9 January
2017.
NOAA/ESRL (2024a) Trends in Atmospheric Carbon Dioxide. Available online at:
https://gml.noaa.gov/ccgg/trends/gr.html. March 26, 2024.
NOAA/ESRL (2024b) Trends in Atmospheric Methane. Available online at: https://gml.noaa.gov/ccgg/trends ch4/.
March 26, 2024.
NOAA/ESRL (2024c) Trends in Atmospheric Nitrous Oxide. Available online at:
https://gml.noaa.gov/ccgg/trends n2o/. March 26, 2024.
NOAA/ESRL (2024d) Trends in Atmospheric Sulfur Hexafluoride. Available online at:
https://gml.noaa.gov/ccgg/trends sf6/. March 26, 2024.
Rigby, M., et al. (2010) History of Atmospheric SFe from 1973 to 2008. Available online at: www.atnios-eheni-
phvs.net/10/10305/2010. March 23, 2024.
UNEP/WMO (1999) Information Unit on Climate Change. Framework Convention on Climate Change. Available
online at: http://unfccc.int.
UNFCCC (2014) Report of the Conference of the Parties on its nineteenth session, held in Warsaw from 11 to 23
November 2013. (FCCC/CP/2013/10/Add.3). January 31, 2014. Available online at:
http://unfccc.int/resource/docs/2013/copl9/eng/10a03.pdf.
USGCRP (2017) Climate Science Special Report: Fourth National Climate Assessment, Volume I. [Wuebbles, D.J.,
D.W. Fahey, K.A. Hibbard, D.J. Dokken, B.C. Stewart, and T.K. Maycock (eds.)]. U.S. Global Change Research
Program, Washington, DC, USA, 470 pp, doi: 10.7930/J0J964J6. Available online at:
https://science2017.globalchange.gov/.
WMO/UNEP (2018) Scientific Assessment of Ozone Depletion: 2018. Available online at:
https://csl.noaa.gov/assessments/ozone/2018.
Trends in Greenhouse Gas Emissions
BEA (2024) 2022 Comprehensive Revision of the National Income and Product Accounts: Current-dollar and "real"
GDP, 1929-2022. Bureau of Economic Analysis (BEA), U.S. Department of Commerce, Washington, D.C. Available
online at: http://www.bea.gOv/national/index.htm#gdp.
EIA (2024) Monthly Energy Review, February 2024. Energy Information Administration, U.S. Department of Energy,
Washington, D.C. DOE/EIA-0035(2023/11).
EIA (2020) Fuel Oil and Kerosene Sales. Energy Information Administration, U.S. Department of Energy.
Washington, D.C. Available online at: http://www.eia.gov/petroleum/fueloilkerosene.
EPA (2023a) "Criteria pollutants National Tier 1 for 1970 - 2023." National Emissions Inventory (NEI) Air Pollutant
Emissions Trends Data. Office of Air Quality Planning and Standards, March 2024. Available online at:
https://www.epa.gov/air-emissions-inventories/air-pollutant-emissions-trends-data.
EPA (2023b) 2023 EPA Automotive Trends Report. Office of Transportation and Air Quality, U.S. Environmental
Protection Agency. Available online at: https://www.epa.gov/automotive-trends/download-automotive-trends-
report.
References and Abbreviations 10-3
-------
IPCC (2021) Climate Change 2021: The Physical Science Basis. Contribution of Working Group I to the Sixth
Assessment Report of the Intergovernmental Panel on Climate Change. [Masson-Delmotte, V., P. Zhai, A. Pirani, S.L
Connors, C. Pean, S. Berger, N. Caud, Y. Chen, L. Goldfarb, M.I. Gomis, M. Huang, K. Leitzell, E. Lonnoy, J.B.R.
Matthews, T.K. Maycock, T. Waterfield, O. Yelekgi, R. Yu, and B. Zhou (eds.)]. Cambridge University Press,
Cambridge, United Kingdom and New York, NY, USA, 2391 pp. doi:10.1017/9781009157896.
IPCC (2019) 2019 Refinement to the 2006 IPCC Guidelines for National Greenhouse Gas Inventories. Calvo Buendia,
E., Tanabe, K., Kranjc, A., Baasansuren, J., Fukuda, M., Ngarize, S., Osako, A., Pyrozhenko, Y., Shermanau, P. and
Federici, S. (eds). Published: IPCC, Switzerland.
IPCC (2013) Climate Change 2013: The Physical Science Basis. Contribution of Working Group I to the Fifth
Assessment Report of the Intergovernmental Panel on Climate Change. [Stocker, T.F., D. Qin, G.-K., Plattner, M.
Tignor, S.K. Allen, J. Boschung, A. Nauels, Y. Xia, V. Bex and P.M. Midgley (eds.)]. Cambridge University Press,
Cambridge, United Kingdom and New York, NY, USA, 1535 pp.
IPCC (2007) Climate Change 2007: The Physical Science Basis. Contribution of Working Group I to the Fourth
Assessment Report of the Intergovernmental Panel on Climate Change. [S. Solomon, D. Qin, M. Manning, Z. Chen,
M. Marquis, K.B. Averyt, M. Tignor and H.L Miller (eds.)]. Cambridge University Press. Cambridge, United Kingdom
996 pp.
IPCC (2006) 2006 IPCC Guidelines for National Greenhouse Gas Inventories. The National Greenhouse Gas
Inventories Programme, The Intergovernmental Panel on Climate Change. [H.S. Eggleston, L. Buendia, K. Miwa, T.
Ngara, and K. Tanabe (eds.)]. Hayama, Kanagawa, Japan.
U.S. Census Bureau (2024) U.S. Census Bureau International Database (IDB). Available online at:
https://www.census.gov/programs-surveys/intemational-programs.html.
U.S. Department of Agriculture, National Agricultural Statistics Service (USDA/NASS) (2023) Farm Production
Expenditures Annual Summary. National Agricultural Statistics Service, U.S. Department of Agriculture, Washington
DC. Available online at: hitps://usda.librarv.cornell.edu/concern/publications/qz20ss48r?locale=en.
Energy
EIA (2024) Monthly Energy Review, February 2024, Energy Information Administration, U.S. Department of Energy,
Washington, DC. DOE/EIA-0035(2024/02).
IEA (2022) Energy related CO2 emissions, 2022, International Energy Agency, Paris. Available online at:
https://www.iea.org/reports/co2-emissions-in-2022.
Carbon Dioxide Emissions from Fossil Fuel Combustion
AAR (2008 through 2022) Railroad Facts. Policy and Economics Department, Association of American Railroads,
Washington, D.C. Private communication with Dan Keen.
AISI (2004 through 2021) Annual Statistical Report, American Iron and Steel Institute, Washington, D.C.
APTA (2007 through 2020) Public Transportation Fact Book. American Public Transportation Association,
Washington, D.C. Available online at: http://www.apta.com/resources/statistics/Pages/transitstats.aspx.
APTA (2006) Commuter Rail National Totals. American Public Transportation Association, Washington, D.C.
BEA (2024) 2022 Comprehensive Revision of the National Income and Product Accounts: Current-dollar and "real"
GDP, 1929-2022. Bureau of Economic Analysis (BEA), U.S. Department of Commerce, Washington, D.C. Available
online at: http://www.bea.gOv/national/index.htm#gdp.
10-4 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
BEA (1991 through 2015) Unpublished BE-36 survey data. Bureau of Economic Analysis, U.S. Department of
Commerce. Washington, D.C.
Benson, D. (2002 through 2004) Unpublished data. Upper Great Plains Transportation Institute, North Dakota State
University and American Short Line & Regional Railroad Association.
Browning (2022a) Addressing the Time Series Inconsistency in FHWA Data. Memorandum from ICF to Sarah
Roberts, Office of Transportation and Air Quality, U.S. Environmental Protection Agency. September 2022.
Browning (2022b) Updated Methodology for Estimating Cm and N2O Emissions from Highway Vehicle Alternative
Fuel Vehicles. Memorandum from ICF to Sarah Roberts, Office of Transportation and Air Quality, U.S.
Environmental Protection Agency. November 2022.
Browning, L (2020) GHG Inventory EF Development Using Certification Data. Technical Memo, September 2020.
Browning, L (2019) Updated On-highway Cm and N2O Emission Factors for GHG Inventory. Memorandum from ICF
to Sarah Roberts, Office of Transportation and Air Quality, U.S. Environmental Protection Agency. September 2019.
Browning, L (2018a) Updated Methodology for Estimating Electricity Use from Highway Plug-In Electric Vehicles.
Technical Memo, October 2018.
Browning, L (2018b) Updated Non-Highway CH4 and N2O Emission Factors for U.S. GHG Inventory. Technical
Memo, November 2018.
Browning, L (2017) Updated Methodology for Estimating CH4 and N2O Emissions from Highway Vehicle Alternative
Fuel Vehicles. Technical Memo, October 2017.
Coffeyville Resources Nitrogen Fertilizers (2012) Nitrogen Fertilizer Operations. Available online at:
http://coffevvillegroup.com/NitrogenFertilizerOperations/index.html.
Dakota Gasification Company (2006) CO2 Pipeline Route and Designation Information. Bismarck, ND.
DHS (2008) Email Communication. Elissa Kay, Department of Homeland Security and Joe Aamidor, ICF
International. January 11, 2008.
DLA Energy (2022) Unpublished data from the Fuels Automated System (FAS). Defense Logistics Agency Energy,
U.S. Department of Defense. Washington, D.C.
DOC (1991 through 2022) Unpublished Report of Bunker Fuel Oil Laden on Vessels Cleared for Foreign Countries.
Form-563. Foreign Trade Division, Bureau of the Census, U.S. Department of Commerce. Washington, D.C.
DOE (1991 through 2020) Transportation Energy Data Book. Edition 40. Office of Transportation Technologies,
Center for Transportation Analysis, Energy Division, Oak Ridge National Laboratory. ORNL-6978. Personal
Communication between Stacy Davis (DOE) and Deep Shah (ICF) for sharing selected tables from the pre-release
version.
DOE (2012) 2010 Worldwide Gasification Database. National Energy Technology Laboratory and Gasification
Technologies Council. Available online at:
http://www.netl.doe.gov/technologies/coalpower/gasification/worlddatabase/index.html. Accessed on 15 March
2012.
DOT (1991 through 2023) Airline Fuel Cost and Consumption. U.S. Department of Transportation, Bureau of
Transportation Statistics, Washington, D.C. DAI-10. Available online at: http://www.transtats.bts.gov/fuel.asp.
Eastman Gasification Services Company (2011) Project Data on Eastman Chemical Company's Chemicals-from-Coal
Complex in Kingsport, TN.
EIA (2024a) Monthly Energy Review, February 2024, Energy Information Administration, U.S. Department of
Energy, Washington, DC. DOE/EIA-0035 (2024/02).
References and Abbreviations 10-5
-------
EIA (2024b) International Energy Statistics 1980-2022. Energy Information Administration, U.S. Department of
Energy. Washington, D.C. Available online at: https://www.eia.gov/beta/international/.
EIA (2024c) Quarterly Coal Report: January - September 2023. Energy Information Administration, U.S. Department
of Energy. Washington, D.C. DOE/EIA-0121.
EIA (2023a) Natural Gas Annual 2022. Energy Information Administration, U.S. Department of Energy. Washington,
D.C. DOE/EIA-0131(20).
EIA (2023b). Petroleum Supply Annual 2022. Energy Information Administration, U.S. Department of Energy.
Washington, D.C. Available online at: https://www.eia.gov/petroleum/.
EIA (2023c) Form EIA-923 detailed data with previous form data (EIA-906/920), Energy Information Administration,
U.S. Department of Energy. Washington, DC. DOE/EIA. November 2023.
EIA (2023d) Annual Coal Report 2022. Energy Information Administration, U.S. Department of Energy. Washington,
D.C. DOE/EIA-0584.
EIA (2022) "Energy use in homes." Use of energy explained. Available online at:
https://www.eia.gov/energyexplained/use-of-energv/homes.php.
EIA (2020a) Glossary. Energy Information Administration, U.S. Department of Energy, Washington, D.C. Available
online at: https://www.eia.gov/tools/glossary/?id=electricity.
EIA (2020b) "Natural gas prices, production, consumption, and exports increased in 2019." Today in Energy.
Available online at: https://www.eia.gov/todavinenergy/detail.php?id=37892.
EIA (2018) "Both natural gas supply and demand have increased from year-ago levels." Today in Energy. Available
online at: https://www.eia.gov/todavinenergy/detail.php?id=37193.
EIA (2009a) Emissions of Greenhouse Gases in the United States 2008, Draft Report. Office of Integrated Analysis
and Forecasting, Energy Information Administration, U.S. Department of Energy. Washington, D.C. DOE-EIA-0573
(2009).
EIA (2009b) Manufacturing Consumption of Energy (MECS) 2006. U.S. Department of Energy, Energy Information
Administration, Washington, D.C. Released July 2009.
EIA (2008) Historical Natural Gas Annual, 1930 - 2008. Energy Information Administration, U.S. Department of
Energy. Washington, D.C.
EIA (2007) Personal Communication. Joel Lou, Energy Information Administration and Aaron Beaudette, ICF
International. Residual and Distillate Fuel Oil Consumption for Vessel Bunkering (Both International and Domestic)
for American Samoa, U.S. Pacific Islands, and Wake Island. October 24, 2007.
EIA (2002) Alternative Fuels Data Tables. Energy Information Administration, U.S. Department of Energy.
Washington, D.C. Available online at: https://www.eia.gov/renewable/.
EIA (2001) U.S. Coal, Domestic and International Issues. Energy Information Administration, U.S. Department of
Energy. Washington, D.C. March 2001.
EIA (1990-2001) State Energy Data System. Energy Information Administration, U.S. Department of Energy.
Washington, D.C.
Environment and Climate Change Canada (2022) Personal Communication between Environment and Climate
Change Canada and Vincent Camobreco for imported CO2. March 2022.
EPA (2024) Acid Rain Program Dataset 1996-2022. Office of Air and Radiation, Office of Atmospheric Programs,
U.S. Environmental Protection Agency, Washington, D.C.
10-6 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
EPA (2023) The 2023 EPA Automotive Trends Report: Greenhouse Gas Emissions, Fuel Economy, and Technology
since 1975. Office of Transportation and Air Quality, U.S. Environmental Protection Agency. Available online at:
https://www.epa.gov/automotive-trends.
EPA (2022) Motor Vehicle Emissions Simulator (M0VES3). Office of Transportation and Air Quality, U.S.
Environmental Protection Agency, Washington, D.C. Available online at: https://www.epa.gov/moves.
EPA (2021) The Emissions & Generation Resource Integrated Database (eGRID) 2019 Technical Support Document.
Clean Air Markets Division, Office of Atmospheric Programs, U.S. Environmental Protection Agency, Washington,
D.C. Available Online at: https://www.epa.gov/sites/default/files/2021-
02/documents/egrid2019 technical guide.pdf
EPA (2020) EPA Inventory of U.S. Greenhouse Gas Emissions and Sinks 1990-2019: Updated Gasoline and Diesel
Fuel CO2 Emission Factors - Memo.
EPA (2010) Carbon Content Coefficients Developed for EPA's Mandatory Reporting Rule. Office of Air and
Radiation, Office of Atmospheric Programs, U.S. Environmental Protection Agency, Washington, D.C.
Erickson, T. (2003) Plains CO2 Reduction (PCOR) Partnership. Presented at the Regional Carbon Sequestration
Partnership Meeting Pittsburgh, Pennsylvania, Energy and Environmental Research Center, University of North
Dakota. November 3, 2003.
FAA (2024) Personal Communication between FAA and John Steller, Mausami Desai, and Vincent Camobreco for
aviation emissions estimates from the Aviation Environmental Design Tool (AEDT). March 2024.
FHWA (1996 through 2023) Highway Statistics. Federal Highway Administration, U.S. Department of
Transportation, Washington, D.C. Report FHWA-PL-96-023-annual. Available online at:
http://www.fhwa.dot.gov/policy/ohpi/hss/hsspubs.htm.
FHWA (2015) Off-Highway and Public-Use Gasoline Consumption Estimation Models Used in the Federal Highway
Administration, Publication Number FHWA-PL-17-012. Available online at:
https://www.fhwa.dot.gov/policyinformation/pubs/pll7012.pdf.
Fitzpatrick, E. (2002) The Weyburn Project: A Model for International Collaboration.
FRB (2022) Industrial Production and Capacity Utilization. Federal Reserve Statistical Release, G.17, Federal
Reserve Board. Available online at: http://www.federalreserve.gov/releases/G17/tablel 2.htm.
Gaffney, J. (2007) Email Communication. John Gaffney, American PublicTransportation Association and Joe
Aamidor, ICF International. December 17, 2007.
IPCC (2013) Climate Change 2013: The Physical Science Basis. Contribution of Working Group I to the Fifth
Assessment Report of the Intergovernmental Panel on Climate Change. [Stocker, T.F., D. Qin, G.K. Plattner, M.
Tignor, S.K. Allen, J. Boschung, A. Nauels, Y. Xia, V. Bex and P.M. Midgley (eds.)]. Cambridge University Press,
Cambridge, United Kingdom and New York, NY, USA, 1535 pp.
IPCC (2007) Climate Change 2007: The Physical Science Basis. Contribution of Working Group I to the Fourth
Assessment Report of the Intergovernmental Panel on Climate Change. [S. Solomon, D. Qin, M. Manning, Z. Chen,
M. Marquis, K.B. Averyt, M. Tignor and H.L Miller (eds.)]. Cambridge University Press. Cambridge, United Kingdom,
996 pp.
IPCC (2006) 2006 IPCC Guidelines for National Greenhouse Gas Inventories. The National Greenhouse Gas
Inventories Programme, The Intergovernmental Panel on Climate Change. [H.S. Eggleston, L. Buendia, K. Miwa, T.
Ngara, and K. Tanabe (eds.)]. Hayama, Kanagawa, Japan.Marland, G. and A. Pippin (1990) "United States Emissions
of Carbon Dioxide to the Earth's Atmosphere by Economic Activity." Energy Systems and Policy, 14(4):323.
NREL (2023) "NREL Researchers Reveal How Buildings Across United States Do—and Could—Use Energy." Available
online at: https://www.nrel.gov/news/features/2023/nrel-researdiers-reveal-how-buildings-across-the-united-
References and Abbreviations 10-7
-------
states-do-and-could-use-
energv.html#r:text=Buildings%20are%20responsible%20for%2040,building%20stock%20is%20also%20essential.
SAIC/EIA (2001) Monte Carlo Simulations of Uncertainty in U.S. Greenhouse Gas Emission Estimates. Final Report.
Prepared by Science Applications International Corporation (SAIC) for Office of Integrated Analysis and Forecasting,
Energy Information Administration, U.S. Department of Energy. Washington, D.C. June 22, 2001.
U.S. Aluminum Association (USAA) (2008 through 2021) U.S. Primary Aluminum Production. U.S. Aluminum
Association, Washington, D.C.
USAF (1998) Fuel Logistics Planning. U.S. Air Force: AFPAM23-221. May 1,1998.
U.S. Census Bureau (2001 through 2011) Current Industrial Reports Fertilizer Materials and Related Products:
Annual Summary. Available online at: https://www.census.gov/data/tables/time-series/econ/cir/mq325b.html.
United States Geological Survey (USGS) (2020a) 2020 Mineral Commodity Summaries: Aluminum. U.S. Geological
Survey, Reston, VA.
USGS (2021b) 2021 Mineral Commodity Summary: Titanium and Titanium Dioxide. U.S. Geological Survey, Reston,
VA.
USGS (2019) 2017Mineral Yearbook: Aluminum. U.S. Geological Survey, Reston, VA
USGS (2014 through 2021a) Mineral Industry Surveys: Silicon. U.S. Geological Survey, Reston, VA.
USGS (2014 through 2021b) Mineral Commodity Summary, Lead. U.S. Geological Survey, Reston, VA.
USGS (2014 through 2019) Minerals Yearbook: Nitrogen [Advance Release], Available online at:
http://minerals.usgs.gov/minerals/pubs/commodity/nitrogen/.
USGS (1991 through 2020) Minerals Yearbook - Iron and Steel Scrap. U.S. Geological Survey, Reston, VA.
USGS (1991 through 2015a) Minerals Yearbook: Manufactured Abrasives Annual Report. U.S. Geological Survey,
Reston, VA. Available online at: http://minerals.usgs.gov/minerals/pubs/commodity/abrasives/.
USGS (1991 through 2015b) Minerals Yearbook: Titanium. U.S. Geological Survey, Reston, VA.
USGS (1991 through 2015c) Minerals Yearbook: Silicon Annual Report. U.S. Geological Survey, Reston, VA. Available
online at: http://minerals.usgs.gov/minerals/pubs/commodity/silicon/.
USGS (1996 through 2013) Minerals Yearbook: Silicon. U.S. Geological Survey, Reston, VA.
USGS (1995 through 2013) Minerals Yearbook: Lead Annual Report. U.S. Geological Survey, Reston, VA.
USGS (1995,1998, 2000, 2001, 2002, 2007) Minerals Yearbook: Aluminum Annual Report. U.S. Geological Survey,
Reston, VA.
Stationary Combustion (excluding C02)
EIA (2024a) Monthly Energy Review, February 2024. Energy Information Administration, U.S. Department of
Energy. Washington, D.C. DOE/EIA-0035(2024/02).
EIA (2024b) International Energy Statistics 1980-2022. Energy Information Administration, U.S. Department of
Energy. Washington, D.C. Available online at: https://www.eia.gov/international/data/world.
EPA (2024) Acid Rain Program Dataset 1996-2022. Office of Air and Radiation, Office of Atmospheric Programs,
U.S. Environmental Protection Agency, Washington, D.C.
EPA (2022) MOtor Vehicle Emissions Simulator (MOVES3). Office of Transportation and Air Quality, U.S.
Environmental Protection Agency, Washington, D.C. Available online at: https://www.epa.gov/moves.
10-8 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
EPA (1997) Compilation of Air Pollutant Emission Factors, AP-42. Office of Air Quality Planning and Standards, U.S.
Environmental Protection Agency. Research Triangle Park, NC. October 1997.
FHWA (1996 through 2023) Highway Statistics. Federal Highway Administration, U.S. Department of
Transportation, Washington, D.C. Report FHWA-PL-96-023-annual. Available online at:
http://www.fhwa.dot.gov/policy/ohpi/hss/hsspubs.htm.
IPCC (2013) Climate Change 2013: The Physical Science Basis. Contribution of Working Group I to the Fifth
Assessment Report of the Intergovernmental Panel on Climate Change. [Stocker, T.F., D. Qin, G.K. Plattner, M.
Tignor, S.K. Allen, J. Boschung, A. Nauels, Y. Xia, V. Bex and P.M. Midgley (eds.)]. Cambridge University Press,
Cambridge, United Kingdom and New York, NY, USA, 1535 pp.
IPCC (2007). Climate Change 2007: The Physical Science Basis. Contribution of Working Group I to the Fourth
Assessment Report of the Intergovernmental Panel on Climate Change. [S. Solomon, D. Qin, M. Manning, Z. Chen,
M. Marquis, K.B. Averyt, M. Tignor and H.L Miller (eds.)]. Cambridge University Press. Cambridge, United Kingdom,
996 pp.
IPCC (2006) 2006 IPCC Guidelines for National Greenhouse Gas Inventories. The National Greenhouse Gas
Inventories Programme, The Intergovernmental Panel on Climate Change. [H.S. Eggleston, L. Buendia, K. Miwa, T.
Ngara, and K. Tanabe (eds.)]. Hayama, Kanagawa, Japan.SAIC/EIA (2001) Monte Carlo Simulations of Uncertainty in
U.S. Greenhouse Gas Emission Estimates. Final Report. Prepared by Science Applications International Corporation
(SAIC) for Office of Integrated Analysis and Forecasting, Energy Information Administration, U.S. Department of
Energy. Washington, D.C. June 22, 2001.
Mobile Combustion (excluding C02)
AAR (2008 through 2023) Railroad Facts. Policy and Economics Department, Association of American Railroads,
Washington, D.C. Private communication with Dan Keen.
ANL (2022) The Greenhouse Gases, Regulated Emissions, and Energy Use in Transportation Model (GREET2022).
Argonne National Laboratory. October 2022. Available online at: https://greet.es.anl.gov.
APTA (2007 through 2023) Public Transportation Fact Book. American Public Transportation Association,
Washington, D.C. Available online at: http://www.apta.com/resources/statistics/Pages/transitstats.aspx.
APTA (2006) Commuter Rail National Totals. American Public Transportation Association, Washington, D.C.
Available online at: http://www.apta.com/research/stats/rail/crsum.cfm.
BEA (1991 through 2015) Unpublished BE-36 survey data. Bureau of Economic Analysis, U.S. Department of
Commerce. Washington, D.C.
Benson, D. (2002 through 2004) Personal communication. Unpublished data developed by the Upper Great Plains
Transportation Institute, North Dakota State University and American Short Line & Regional Railroad Association.
Browning (2022a) Addressing the Time Series Inconsistency in FHWA Data. Memorandum from ICF to Sarah
Roberts, Office of Transportation and Air Quality, U.S. Environmental Protection Agency. September 2022.
Browning (2022b) Updated Methodology for Estimating CH4 and N2O Emissions from Highway Vehicle Alternative
Fuel Vehicles. Memorandum from ICF to Sarah Roberts, Office of Transportation and Air Quality, U.S.
Environmental Protection Agency. November 2022.
Browning (2020) GHG Inventory EF Development Using Certification Data. Memorandum from ICF to Sarah
Roberts, Office of Transportation and Air Quality, U.S. Environmental Protection Agency. September 2020.
Browning, L. (2019) Updated On-highway CH4 and N2O Emission Factors for GHG Inventory. Memorandum from ICF
to Sarah Roberts and Justine Geidosch, Office of Transportation and Air Quality, U.S. Environmental Protection
Agency. September 2019.
References and Abbreviations 10-9
-------
Browning, L (2018) Updated Non-Highway CFU and N2O Emission Factors for U.S. GHG Inventory. Technical
Memorandum from ICF International to Sarah Roberts and Justine Geidosch, Office of Transportation and Air
Quality, U.S. Environmental Protection Agency. November 2018.
Browning, L (2005) Personal communication with Lou Browning, "Emission control technologies for diesel highway
vehicles specialist," ICF International.
BTS (2023) Amtrak Fuel Consumption and Travel. Bureau of Transportation Statistics, Washington, DC. Available
online at: https://www.bts.gov/content/amtrak-fuel-consumption-and-travel-l.
DLA Energy (2022) Unpublished data from the Defense Fuels Automated Management System (DFAMS). Defense
Energy Support Center, Defense Logistics Agency, U.S. Department of Defense. Washington, D.C.
DOC (1991 through 2022) Unpublished Report of Bunker Fuel Oil Laden on Vessels Cleared for Foreign Countries.
Form-563. Foreign Trade Division, Bureau of the Census, U.S. Department of Commerce. Washington, D.C.
DOE (1993 through 2022) Transportation Energy Data Book Edition 40. Office of Transportation Technologies,
Center for Transportation Analysis, Energy Division, Oak Ridge National Laboratory. Personal Communication
between Stacy Davis (DOE) and Deep Shah (ICF) for sharing selected tables from the pre-release version.
DOT (1991 through 2023) Airline Fuel Cost and Consumption. U.S. Department of Transportation, Bureau of
Transportation Statistics, Washington, D.C. DAI-10. Available online at: http://www.transtats.bts.gov/fuel.asp.
EIA (2024) Monthly Energy Review, February 2024, Energy Information Administration, U.S. Department of Energy,
Washington, D.C. DOE/EIA-0035(2024/02).
EIA (2023) Natural Gas Annual 2022. Energy Information Administration, U.S. Department of Energy, Washington,
D.C. DOE/EIA-O131(22).
EIA (1991 through 2022) Fuel Oil and Kerosene Sales. Energy Information Administration, U.S. Department of
Energy. Washington, D.C. Available online at: http://www.eia.gov/petroleum/fyeloilkerosene.
EIA (2007) Personal Communication. Joel Lou, Energy Information Administration and Aaron Beaudette, ICF
International. Residual and Distillate Fuel Oil Consumption for Vessel Bunkering (Both International and Domestic)
for American Samoa, U.S. Pacific Islands, and Wake Island. October 24, 2007.
EIA (2002) Alternative Fuels Data Tables. Energy Information Administration, U.S. Department of Energy,
Washington, D.C. Available online at: http://www.eia.doe.gov/fuelrenewable.html.
EPA (2023) Annual Certification Test Results Report. Office of Transportation and Air Quality, U.S. Environmental
Protection Agency. Available online at: https://www.epa.gov/compliance-and-fuel-economy-data/annual-
certification-test-data-vehicles-and-engines.
EPA (2022a) Motor Vehicle Emissions Simulator (MOVES3). Office of Transportation and Air Quality, U.S.
Environmental Protection Agency. Available online at: https://www.epa.gov/moves.
EPA (2022b) Confidential Engine Family Sales Data Submitted to EPA by Manufacturers. Office of Transportation
and Air Quality, U.S. Environmental Protection Agency.
EPA (2004) Mobile6.2 Vehicle Emission Modeling Software. Office of Mobile Sources, U.S. Environmental
Protection Agency, Ann Arbor, Michigan.
EPA (1999) Emission Facts: The History of Reducing Tailpipe Emissions. Office of Mobile Sources. May 1999. EPA
420-F-99-017. Available online at: https://www.epa.gov/nscep.
EPA (1998) Emissions of Nitrous Oxide from Highway Mobile Sources: Comments on the Draft Inventory of U.S.
Greenhouse Gas Emissions and Sinks, 1990-1996. Office of Mobile Sources, Assessment and Modeling Division,
U.S. Environmental Protection Agency. August 1998. EPA420-R-98-009.
EPA (1994a) Automobile Emissions: An Overview. Office of Mobile Sources. August 1994. EPA 400-F-92-007.
Available online at: https://www.epa.gov/nscep.
10-10 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
EPA (1994b) Milestones in Auto Emissions Control. Office of Mobile Sources. August 1994. EPA 400-F-92-014.
Available online at: https://www.epa.gov/nscep.
Esser, C. (2003 through 2004) Personal Communication with Charles Esser, Residual and Distillate Fuel Oil
Consumption for Vessel Bunkering (Both International and Domestic) for American Samoa, U.S. Pacific Islands, and
Wake Island.
FAA (2022) Personal Communication between FAA and John Steller, Mausami Desai and Vincent Camobreco for
aviation emission estimates from the Aviation Environmental Design Tool (AEDT). March 2022.
FHWA (1996 through 2023) Highway Statistics. Federal Highway Administration, U.S. Department of
Transportation, Washington, D.C. Report FHWA-PL-96-023-annual.
FTA (2023) National Transit Database "Fuel and Energy by Mode and TOS" table. Available online at:
https://data.transportation.gov/.
Gaffney, J. (2007) Email Communication. John Gaffney, American PublicTransportation Association and Joe
Aamidor, ICF International. December 17, 2007.
IPCC (2006) 2006 IPCC Guidelines for National Greenhouse Gas Inventories. The National Greenhouse Gas
Inventories Programme, The Intergovernmental Panel on Climate Change. [H.S. Eggleston, L. Buendia, K. Miwa, T.
Ngara, and K. Tanabe (eds.)]. Hayama, Kanagawa, Japan.
ICF (2006) Revised Gasoline Vehicle EFsfor LEV and Tier 2 Emission Levels. Memorandum from ICF International to
John Davies, Office of Transportation and Air Quality, U.S. Environmental Protection Agency. November 2006.
ICF (2004) Update of Methane and Nitrous Oxide Emission Factors for On-Highway Vehicles. Final Report to U.S.
Environmental Protection Agency. February 2004.
Raillnc (2014 through 2023) Raillnc Short line and Regional Traffic Index. Carloads Originated Year-to-Date.
November 2023. Available online at: https://public.railinc.com/.
Whorton, D. (2006 through 2014) Personal communication, Class II and III Rail energy consumption, American
Short Line and Regional Railroad Association.
Carbon Emitted from Non-Energy Uses of Fossil Fuels
ACC (2023a) "U.S. Resin Production & Sales 2022 vs. 2021." Available online at:
https://www.americanchemistry.com/chemistry-in-america/data-industry-statistics/statistics-on-the-plastic-
resins-industrv/resources/pips-resin-sales-and-production~cy-figures-2022-vs-2021.
ACC (2023b) Guide to the Business of Chemistry, 2023, American Chemistry Council. Available online at:
https://www.americanchemistry.com/chemistry-in-america/data-industry-statistics/resources/2023-guide-to-the-
business-of-chemistry.
ACC (2022) "U.S. Resin Production & Sales 2021 vs. 2020." Available online at:
https://www.americanchemistry.com/chemistry-in-america/data-industry-statistics/statistics-on-the-plastic-
resins-industry.
ACC (2021) "U.S. Resin Production & Sales 2020 vs. 2019." Available online at:
https://www.americanchemistrv.com/chemistrv-in-america/chemistrv-in-everydav-products/plastics.
ACC (2020) "U.S. Resin Production & Sales 2019 vs. 2018." Available online at:
https://www.americanchemistrv.com/chemistrv-in-america/chemistrv-in-evervdav-products/plastics.
ACC (2019) "U.S. Resin Production & Sales 2018 vs. 2017." Available online at:
https://www.americanchemistrv.com/chemistrv-in-america/chemistrv-in-everydav-products/plastics.
References and Abbreviations 10-11
-------
ACC (2018) "U.S. Resin Production & Sales 2017 vs. 2016." Available online at:
https://www.americanchemistrv.com/chemistrv-in-america/chemistrv-in-everydav-products/plastics.
ACC (2017) "U.S. Resin Production & Sales 2016 vs. 2015."
ACC (2016) "U.S. Resin Production & Sales 2015 vs. 2014."
ACC (2015) "PIPS Year-End Resin Statistics for 2014 vs. 2013: Production, Sales and Captive Use." Available online
at: https://www.americanchemistrv.com/chemistrv-in-america/data-industry-statistics/statistics-on-the-plastic-
resins-industry/resin-report-subscriptions.
ACC (2014) "U.S. Resin Production & Sales: 2013 vs. 2012," American Chemistry Council. Available online at:
http://www.americanchemistry.com/Jobs/EconomicStatistics/Plastics-Statistics/Production-and-Sales-Data-by-
Resin.pdf.
ACC (2013) "U.S. Resin Production & Sales: 2012 vs. 2011," American Chemistry Council. Available online at:
http://www.americanchemistry.com/Jobs/EconomicStatistics/Plastics-Statistics/Production-and-Sales-Data-by-
Resin.pdf.
ACC (2003-2011) "PIPS Year-End Resin Statistics for 2010: Production, Sales and Captive Use." Available online at:
http://www.americanchemistry.com/Jobs/EconomicStatistics/Plastics-Statistics/Production-and-Sales-Data-by-
Resin.pdf.
Bank of Canada (2023) Financial Markets Department Year Average of Exchange Rates. Available online at:
https://www.bankofcanada.ca/rates/exchange/annual-average-exchange-rates/#download.
Bank of Canada (2022) Financial Markets Department Year Average of Exchange Rates. Available online at:
https://www.bankofcanada.ca/rates/exchange/annual-average-exchange-rates/#download.
Bank of Canada (2021) Financial Markets Department Year Average of Exchange Rates. Available online at:
https://www.bankofcanada.ca/rates/exchange/annual-average-exchange-rates/#download.
Bank of Canada (2020) Financial Markets Department Year Average of Exchange Rates. Available online at:
https://www.bankofcanada.ca/rates/exchange/annual-average-exchange-rates/#download.
Bank of Canada (2019) Financial Markets Department Year Average of Exchange Rates. Available online at:
https://www.bankofcanada.ca/rates/exchange/annual-average-exchange-rates/#download.
Bank of Canada (2018) Financial Markets Department Year Average of Exchange Rates. Available online at:
https://www.bankofcanada.ca/rates/exchange/annual-average-exchange-rates/.
Bank of Canada (2017) Financial Markets Department Year Average of Exchange Rates. Available online at:
https://www.bankofcanada.ca/rates/exchange/legacv-noon-and-closing-rates/.
Bank of Canada (2016) Financial Markets Department Year Average of Exchange Rates. Available online at:
https://www.bankofcanada.ca/rates/exchange/legacv-noon-and-closing-rates/.
Bank of Canada (2014) Financial Markets Department Year Average of Exchange Rates. Available online at:
https://www.bankofcanada.ca/rates/exchange/legacv-noon-and-closing-rates/.
Bank of Canada (2013) Financial Markets Department Year Average of Exchange Rates. Available online at:
https://www.bankofcanada.ca/rates/exchange/legacv-noon-and-closing-rates/.
Bank of Canada (2012) Financial Markets Department Year Average of Exchange Rates. Available online at:
https://www.bankofcanada.ca/rates/exchange/legacy-noon-and-closing-rates/.
CIAC (2022). 2022 Economic Review of Chemistry. Available online at: https://canadianchemistry.ca/wp-
content/uploads/2022/06/2022-Economic-Review-of-Chemistry23732 removed.pdf.
10-12 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
EIA (2024) Monthly Energy Review, February 2024. Energy Information Administration, U.S. Department of Energy,
Washington, D.C. DOE/EIA-0035 (2024/02). Available online at:
https://www.eia.gov/totalenergy/data/monthly/pdf/mer.pdf.
EIA (2021) EIA Manufacturing Consumption of Energy (MECS) 2018. U.S. Department of Energy, Energy Information
Administration, Washington, D.C.
EIA (2020) Glossary. Energy Information Administration, U.S. Department of Energy, Washington, D.C. Available
online at: https://www. eia.gov/tools/glossary/index. php?id=N#nat Gas Liquids.
EIA (2019) Personal communication between EIA and ICF on November 11, 2019.
EIA (2017) EIA Manufacturing Consumption of Energy (MECS) 2014. U.S. Department of Energy, Energy Information
Administration, Washington, D.C.
EIA (2013) EIA Manufacturing Consumption of Energy (MECS) 2010. U.S. Department of Energy, Energy Information
Administration, Washington, D.C.
EIA (2010) EIA Manufacturing Consumption of Energy (MECS) 2006. U.S. Department of Energy, Energy Information
Administration, Washington, D.C.
EIA (2005) EIA Manufacturing Consumption of Energy (MECS) 2002. U.S. Department of Energy, Energy Information
Administration, Washington, D.C.
EIA (2001) EIA Manufacturing Consumption of Energy (MECS) 1998. U.S. Department of Energy, Energy Information
Administration, Washington, D.C.
EIA (1997) EIA Manufacturing Consumption of Energy (MECS) 1994. U.S. Department of Energy, Energy Information
Administration, Washington, D.C.
EIA (1994) EIA Manufacturing Consumption of Energy (MECS) 1991. U.S. Department of Energy, Energy Information
Administration, Washington, D.C.
EPA (2023) EPA's Emissions Inventory System (EIS) to National Inventory Report (NIR) Mapping file
EIS_NIR_mapping.xlsx. U.S. Environmental Protection Agency. Washington, D.C.
EPA (2021) Resource Conservation and Recovery Act (RCRA) Info, Biennial Report, GM Form (Section 2- Onsite
Management) and WR Form.
EPA (2019) Advancing Sustainable Materials Management: 2016 and 2017 Data Tables. Office of Land and
Emergency Management, U.S. Environmental Protection Agency. Washington, D.C. Available online at:
https://www.epa.gov/sites/production/files/2019-
ll/documents/2016 and 2017 facts and figures data tables O.pdf.
EPA (2018a) Advancing Sustainable Materials Management: Facts and Figures 2015, Assessing Trends in Material
Generation, Recycling and Disposal in the United States. Washington, D.C.
EPA (2018b) Resource Conservation and Recovery Act (RCRA) Info, Biennial Report, GM Form (Section 2- Onsite
Management) and WR Form.
EPA (2017) EPA's Pesticides Industry Sales and Usage, 2008 - 2012 Market Estimates. Available online at:
https://www.epa.gov/sites/production/files/2017-01/documents/pesticides-industry-sales-usage-2016 O.pdf.
Accessed September 2017.
EPA (2016a) Advancing Sustainable Materials Management: 2014 Facts and Figures Fact Sheet. Office of Solid
Waste and Emergency Response, U.S. Environmental Protection Agency, Washington, D.C. Available online at:
https://www.epa.gov/sites/production/files/2016-ll/documents/2014 smmfactsheet 5Q8.pdf.
EPA (2016b) Resource Conservation and Recovery Act (RCRA) Info, Biennial Report, GM Form (Section 2- Onsite
Management) and WR Form.
References and Abbreviations 10-13
-------
EPA (2015) Resource Conservation and Recovery Act (RCRA) Info, Biennial Report, GM Form (Section 2- Onsite
Management) and WR Form.
EPA (2014a) Municipal Solid Waste in the United States: 2012 Facts and Figures. Office of Solid Waste and
Emergency Response, U.S. Environmental Protection Agency, Washington, D.C. Available online at:
https://www.epa.gov/sites/default/files/2015-09/documents/2012 msw dat tbls.pdf.
EPA (2014b) Chemical Data Access Tool (CDAT). U.S. Environmental Protection Agency, June 2014. Available online
at: https://edg.epa.gov/metadata/catalog/search/resource/details.page?uuid=%7B2D73C764-6919-4Q4D-8C9B-
61869B3330D6%7D. Accessed January 2015.
EPA (2013a) Municipal Solid Waste in the United States: 2011 Facts and Figures. Office of Solid Waste and
Emergency Response, U.S. Environmental Protection Agency, Washington, D.C. Available online at:
http://www.epa.gov/epaoswer/non-hw/muncpl/msw99.litm.
EPA (2013b) Resource Conservation and Recovery Act (RCRA) Info, Biennial Report, GM Form (Section 2- Onsite
Management) and WR Form.
EPA (2011) EPA's Pesticides Industry Sales and Usage, 2006 and 2007 Market Estimates. Available online at:
https://www.epa.gov/pesticides/pesticides-industry-sales-and-usage-2006-and-2007-market-estimates. Accessed
January 2012.
EPA (2009) Biennial Reporting System (BRS) Database. U.S. Environmental Protection Agency, Envirofacts
Warehouse. Washington, D.C. Available online at: https://www.epa.gov/enviro/br-search. Data for 2001-2007 are
current as of Sept. 9, 2009.
EPA (2004) EPA's Pesticides Industry Sales and Usage, 2000 and 2001 Market Estimates. Available online at:
https://nepis.epa.gov/Exe/Z.yPURL.cgi?Dockev=3QQ0659P.TXT. Accessed September 2006.
EPA (2002) EPA's Pesticides Industry Sales and Usage, 1998 and 1999 Market Estimates, Table 3.6. Available online
at https://nepis.epa.gov/Exe/ZyPURL.cgi?Dockey=200001G5.TXT. Accessed July 2003.
EPA (2001) AP 42, Volume I, Fifth Edition. Chapter 11: Mineral Products Industry. Available online at:
http://www.epa.gov/ttn/chief/ap42/chll/index.html.
EPA (2000a) Biennial Reporting System (BRS). U.S. Environmental Protection Agency, Envirofacts Warehouse.
Washington, D.C. Available online at: https://www.epa.gov/enviro/br-search.
EPA (2000b) Toxics Release Inventory, 1998. U.S. Environmental Protection Agency, Office of Environmental
Information, Office of Information Analysis and Access, Washington, D.C. Available online at:
https://enviro.epa.gov/triexplorer/tri release.chemical.
EPA (1999) EPA's Pesticides Industry Sales and Usage, 1996-1997 Market Estimates. Available online at:
https://nepis.epa.gov/Exe/ZvPURL.cgi?Dockey=2000011LTXT.
EPA (1998) EPA's Pesticides Industry Sales and Usage, 1994-1995 Market Estimates. Available online at:
http://www.epa.gov/oppbeadl/pestsales/95pestsales/market estimatesl995.pdf.
FEB (2013) Fiber Economics Bureau, as cited in C&EN (2013) Lackluster Year for Chemical Output: Production
stayed flat or dipped in most world regions in 2012. Chemical &Engineering News, American Chemical Society, 1
July. Available online at: http://www.cen-online.org.
FEB (2012) Fiber Economics Bureau, as cited in C&EN (2012) Too Quiet After the Storm: After a rebound in 2010,
chemical production hardly grew in 2011. Chemical & Engineering News, American Chemical Society, 2 July.
Available online at: http://www.cen-online.org.
FEB (2011) Fiber Economics Bureau, as cited in C&EN (2011) Output Ramps up in all Regions. Chemical Engineering
News, American Chemical Society, 4 July. Available online at: http://www.cen-online.org.
10-14 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
FEB (2010) Fiber Economics Bureau, as cited in C&EN (2010) Output Declines in U.S., Europe. Chemical &
Engineering News, American Chemical Society, 6 July. Available online at: http://www.cen-online.org.
FEB (2009) Fiber Economics Bureau, as cited in C&EN (2009) Chemical Output Slipped In Most Regions Chemical &
Engineering News, American Chemical Society, 6 July. Available online at: http://www.cen-online.org.
FEB (2007) Fiber Economics Bureau, as cited in C&EN (2007) Gains in Chemical Output Continue. Chemical &
Engineering News, American Chemical Society. July 2, 2007. Available online at: http://www.cen-online.org.
FEB (2005) Fiber Economics Bureau, as cited in C&EN (2005) Production: Growth in Most Regions Chemical &
Engineering News, American Chemical Society, 11 July. Available online at: http://www.cen-online.org.
FEB (2003) Fiber Economics Bureau, as cited in C&EN (2003) Production Inches Up in Most Countries, Chemical &
Engineering News, American Chemical Society, 7 July. Available online at: http://www.cen-online.org.
FEB (2001) Fiber Economics Bureau, as cited in ACS (2001) Production: slow gains in output of chemicals and
products lagged behind U.S. economy as a whole Chemical & Engineering News, American Chemical Society, 25
June. Available online at: http://pubs.acs.org/cen.
Financial Planning Association (2006) Canada/US Cross-Border Tools: US/Canada Exchange Rates. Available online
at: http://www.fpanet.org/global/planners/US Canada ex rates.cfm. Accessed on August 16, 2006.
Gosselin, Smith, and Hodge (1984) "Clinical Toxicology of Commercial Products." Fifth Edition, Williams & Wilkins,
Baltimore.
ICIS (2016) "Production issues force US melamine plant down" Available online at:
https://www.icis.com/resources/news/2016/05/03/9994556/production-issues-force-us-melamine-plant-down/.
ICIS (2008) "Chemical profile: Melamine" Available online at:
https://www.icis.com/resources/news/2008/12/01/9174886/chemical-profile-melamine/. Accessed November
2017.
IISRP (2003) "IISRP Forecasts Moderate Growth in North America to 2007" International Institute of Synthetic
Rubber Producers, Inc. New Release. Available online at: http://www.iisrp.com/press-releases/20Q3-Press-
Releases/IIS RP-NA-Forecast-03-07.html.
IISRP (2000) "Synthetic Rubber Use Growth to Continue Through 2004, Says IISRP and RMA" International Institute
of Synthetic Rubber Producers press release.
INEGI (2006) Produccion bruta total de las unidades economicas manufactureras por Subsector, Rama, Subrama y
Clase de actividad. Available online at:
http://www.inegi.gob.mx/est/contenidos/espanol/proyectos/censos/ce2004/tb manufacturas.asp. Accessed on
August 15, 2006.
IPCC (2006) 2006IPCC Guidelines for National Greenhouse Gas Inventories. The National Greenhouse Gas
Inventories Programme, The Intergovernmental Panel on Climate Change. [H.S. Eggleston, L. Buendia, K. Miwa, T.
Ngara, and K. Tanabe (eds.)]. Hayama, Kanagawa, Japan.
Marland, G., and R.M. Rotty (1984) "Carbon dioxide emissions from fossil fuels: A procedure for estimation and
results for 1950-1982," Tellus 36b:232-261.
NPRA (2002) North American Wax - A Report Card. Available online at:
http://www.npra.org/members/publications/papers/lubes/LVV-02-126.pdf.
U.S. Census Bureau (2021) 2017 Economic Census. Available online at:
https://www.census.gov/data/tables/2017/econ/economic-census/naics-sector-31-33.html. Accessed October
2021.
U.S. Census Bureau (2014) 2012 Economic Census. Available online at:
http://www.census.gov/econ/census/schedule/whats been released.html. Accessed November 2014.
References and Abbreviations 10-15
-------
U.S. Census Bureau (2009) Soap and Other Detergent Manufacturing: 2007.
U.S. Census Bureau (2004) Soap and Other Detergent Manufacturing: 2002. Issued December 2004. EC02-31I-
325611 (RV). Available online at: http://www.census.gov/prod/ec02/ec0231i325611.pdf.
U.S. Census Bureau (1999) Soap and Other Detergent Manufacturing: 1997. Available online at:
http://www.census.gov/epcd/www/ec97stat.htm.
U.S. International Trade Commission (2023) "Interactive Tariff and Trade DataWeb: Quick Query." Available online
at: http://dataweb.usitc.gov/. Accessed September 2023.
USTMA (2022) "2021 U.S. Scrap Tire Management Summary." U.S. Tire Manufacturers Association, Washington,
DC. October 2022. Available online at:
https://www.ustires.org/sites/default/files/21%20US%20Scrap%20Tire%20Management%20Report%20101722.pd
f.
USTMA (2020) "2019 U.S. Scrap Tire Management Summary." U.S. Tire Manufacturers Association, Washington,
DC. October 2020. Available online at:
https://www.ustires.org/sites/default/files/2019%20USTMA%20Scrap%20Tire%20Management%20Summary%20
Report.pdf.
USTMA (2018) "2017 U.S. Scrap Tire Management Summary." U.S. Tire Manufacturers Association, Washington,
DC. July 2018. Available online at: https://www.tyrepress.com/wp-
content/uploads/2013/07/USTMA scraptire sornm 2017 07 11 2018.pdf.
USTMA (2016) "2015 U.S. Scrap Tire Management Summary." U.S. Tire Manufacturers Association. August 2016.
Available online at: https://www.ustires.org/sites/default/files/MAR 028 USTMA.pdf.
USTMA (2014) "2013 U.S. Scrap Tire Management Summary." U.S. Tire Manufacturers Association. November
2014. Available online at: https://www.ustires.org/sites/default/files/MAR 027 USTMA.pdf.
USTMA (2013) "U.S. Scrap Tire Management Summary 2005-2009." U.S. Tire Manufacturers Association. October
2011; Updated September 2013. Available online at:
https://www.ustires.org/sites/default/files/MAR 025 USTMA.pdf.
USTMA (2012) "Scrap Tire Markets: Facts and Figures - Scrap Tire Characteristics." U.S. Tire Manufacturers
Association. Accessed 18 on January 2012.
Incineration of Waste
ArSova, Ljupka, Rob van Haaren, Nora Goldstein, Scott M. Kaufman, and Nickolas J. Themelis (2008) "16th Annual
BioCycle Nationwide Survey: The State of Garbage in America" BioCycle, JG Press, Emmaus, PA. December.
Bahor, B (2009) Covanta Energy's public review comments re: Draft Inventory of U.S. Greenhouse Gas Emissions
and Sinks: 1990-2007. Submitted via email on April 9, 2009 to Leif Hockstad, U.S. EPA.
De Soete, G.G. (1993) "Nitrous Oxide from Combustion and Industry: Chemistry, Emissions and Control." In A. R.
Van Amstel, (ed.) Proc. of the International Workshop Methane and Nitrous Oxide: Methods in National Emission
Inventories and Options for Control, Amersfoort, NL. February 3-5,1993.
Energy Recovery Council (2018) Energy Recovery Council. 2018 Directory of Waste to Energy Facilities. Ted
Michaels and Karunya Krishnan. October 2018. Available online at: http://energyrecoverycouncil.org/wp-
content/uploads/2019/10/ERC-2018-directory.pdf.
Energy Recovery Council (2009) "2007 Directory of Waste-to-Energy Plants in the United States." Accessed on
September 29, 2009.
EIA (2019) EIA St. Louis Federal Reserve's Economic Data (FRED) Consumer Price Index for All Urban Consumers:
Education and Communication (CPIEDUSL). Available online at: https://www.eia.gov/opendata/excel/.
10-16 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
EIA (2017) MSW Incineration for Heating or Electrical Generation, December 2017, Energy Information
Administration, U.S. Department of Energy, Washington, DC. DOE/EIA-0035. Available online at:
https://www.eia.gov/opendata/?src=-f3.
EPA (2022) Greenhouse Gas Reporting Program (GHGRP). 2022 Envirofacts. Available online at:
https://ghgdata.epa.gov/ghgp/main.do.
EPA (2020a) Advancing Sustainable Materials Management: 2018 Data Tables. Office of Land and Emergency
Management, U.S. Environmental Protection Agency. Washington, D.C. Available online at:
https://www.epa.gov/sites/production/files/2020-ll/documents/2018 ff fact sheet.pdf.
EPA (2020b) Greenhouse Gas Reporting Program (GHGRP). 2020 Envirofacts. Available online at:
https://ghgdata.epa.gov/ghgp/main.do.
EPA (2019) Advancing Sustainable Materials Management: 2016 and 2017 Data Tables. Office of Land and
Emergency Management, U.S. Environmental Protection Agency. Washington, D.C. Available online at:
https://www.epa.gov/sites/production/files/2019-
ll/documents/2016 and 2017 facts and figures data tables O.pdf.
EPA (2018a) Advancing Sustainable Materials Management: 2015 Data Tables. Office of Land and Emergency
Management, U.S. Environmental Protection Agency. Washington, D.C. Available online at:
https://www.epa.gov/sites/production/files/2018-
07/documents/smm 2015 tables ; res 07252018 fnl 508 O.pdf.
EPA (2018b) Greenhouse Gas Reporting Program Data. Washington, DC: U.S. Environmental Protection Agency.
Available online at: https://www.epa.gov/ghgreporting/ghg-reporting-program-data-sets.
EPA (2016) Advancing Sustainable Materials Management: 2014 Fact Sheet. Office of Land and Emergency
Management, U.S. Environmental Protection Agency. Washington, D.C. Available online at:
https://www.epa.gov/sites/production/files/2016-ll/documents/2014 smmfactsheet 508.pdf.
EPA (2015) Advancing Sustainable Materials Management: Facts and Figures 2013 - Assessing Trends in Material
Generation, Recycling and Disposal in the United States. Office of Solid Waste and Emergency Response, U.S.
Environmental Protection Agency. Washington, D.C. Available online at:
http://www3.epa.gov/epawaste/nonhaz/municipal/pubs/2013 advncng smm rpt.pdf.
EPA (2007, 2008, 2011, 2013, 2014) Municipal Solid Waste in the United States: Facts and Figures. Office of Solid
Waste and Emergency Response, U.S. Environmental Protection Agency. Washington, D.C. Available online at:
http://www.epa.gov/osw/nonhaz/municipal/msw99.html.
EPA (2006) Solid Waste Management and Greenhouse Gases: A Life-Cycle Assessment of Emissions and Sinks.
Office of Solid Waste and Emergency Response, U.S. Environmental Protection Agency. Washington, D.C.
EPA (2000) Characterization of Municipal Solid Waste in the United States: Source Data on the 1999 Update. Office
of Solid Waste, U.S. Environmental Protection Agency. Washington, D.C. EPA530-F-00-024.
Goldstein, N. and C. Madtes (2001) "13th Annual BioCycle Nationwide Survey: The State of Garbage in America."
BioCycle, JG Press, Emmaus, PA. December 2001.
IPCC (2013) Climate Change 2013: The Physical Science Basis. Contribution of Working Group I to the Fifth
Assessment Report of the Intergovernmental Panel on Climate Change [Stocker, T.F., D. Qin, G.K. Plattner, M.
Tignor, S.K. Allen, J. Boschung, A. Nauels, Y. Xia, V. Bex and P.M. Midgley (eds.)]. Cambridge University Press,
Cambridge, United Kingdom and New York, NY, USA, 1535 pp.
IPCC (2007) Climate Change 2007: The Physical Science Basis. Contribution of Working Group I to the Fourth
Assessment Report of the Intergovernmental Panel on Climate Change. [S. Solomon, D. Qin, M. Manning, Z. Chen,
M. Marquis, K.B. Averyt, M. Tignor and H.L Miller (eds.)]. Cambridge University Press. Cambridge, United Kingdom,
996 pp.
References and Abbreviations 10-17
-------
IPCC (2006) 2006IPCC Guidelines for National Greenhouse Gas Inventories. The National Greenhouse Gas
Inventories Programme, The Intergovernmental Panel on Climate Change. [H.S. Eggleston, L. Buendia, K. Miwa, T.
Ngara, and K. Tanabe (eds.)]. Hayama, Kanagawa, Japan.
Kaufman, et al. (2004) "14th Annual BioCycle Nationwide Survey: The State of Garbage in America 2004" Biocycle,
JG Press, Emmaus, PA. January 2004.
Schneider, S. (2007) E-mail between Shelly Schneider of Franklin Associates (a division of ERG) and Sarah Shapiro of
ICF International, January 10, 2007.
Shin, D. (2014) Generation and Disposition of Municipal Solid Waste (MSW) in the United States-A National
Survey. Thesis. Columbia University, Department of Earth and Environmental Engineering, January 3, 2014.
Simmons, et al. (2006) "15th Nationwide Survey of Municipal Solid Waste Management in the United States: The
State of Garbage in America." BioCycle, JG Press, Emmaus, PA. April 2006.
USTMA (2022) "2021 U.S. Scrap Tire Management Summary." U.S. Tire Manufacturers Association, Washington,
DC. October 2022. Available online at:
https://www.ustires.org/sites/default/files/21%20US%20Scrap%20Tire%20Management%20Report%20101722.pd
f.
USTMA (2020) "2019 U.S. Scrap Tire Management Summary." U.S. Tire Manufacturers Association, Washington,
DC. October 2020. Available online at:
https://www.ustires.org/sites/default/files/2019%20USTMA%20Scrap%20Tire%20Management%20Summary%20
Report.pdf.
USTMA (2018) "2017 U.S. Scrap Tire Management Summary." U.S. Tire Manufacturers Association, Washington,
DC. July 2018. Available online at: https://www.tyrepress.com/wp-
content/uploads/2018/07/USTMA scraptire suroro 2017 07 11 2018.pdf.
USTMA (2016) "2015 U.S. Scrap Tire Management Summary." U.S. Tire Manufacturers Association. August 2016.
Available online at: https://www.ustires.org/sites/default/files/MAR 028 USTMA.pdf.
USTMA (2014) "2013 U.S. Scrap Tire Management Summary." U.S. Tire Manufacturers Association. November
2014. Available online at: https://www.ustires.org/sites/default/files/MAR 027 USTMA.pdf.
USTMA (2013) "U.S. Scrap Tire Management Summary 2005-2009." U.S. Tire Manufacturers Association. October
2011; Updated September 2013. Available online at:
https://www.ustires.org/sites/default/files/MAR 025 USTMA.pdf.
USTMA (2012a) "Rubber FAQs." U.S. Tire Manufacturers Association. Accessed on 19 November 2014.
USTMA (2012b) "Scrap Tire Markets: Facts and Figures - Scrap Tire Characteristics." U.S. Tire Manufacturers
Association. Accessed 18 on January 2012.
van Haaren, Rob, Themelis, N., and Goldstein, N. (2010) "The State of Garbage in America." BioCycle, October
2010. Volume 51, Number 10, pg. 16-23.
Coal Mining
AAPG (1984) Coalbed Methane Resources of the United States. AAPG Studies in Geology Series #17.
Creedy, D.P. (1993) Methane Emissions from Coal Related Sources in Britain: Development of a Methodology.
Chemosphere, 26: 419-439.
DMME (2023) DGO Data Information System. Department of Mines, Minerals and Energy of Virginia. Available
online at https://www.dmme.virginia.gov/dgoinquiry/frmmain.aspx.
10-18 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
EIA (2023) Annual Coal Report 2022. Table 1. Energy Information Administration, U.S. Department of Energy.
Washington, D.C. DOE/EIA-0584.
El Paso (2009) Shoal Creek Mine Plan, El Paso Exploration & Production.
EPA (2023) Greenhouse Gas Reporting Program (GHGRP) 2022 Subpart FF: Underground Coal Mines.
EPA (2005) Surface Mines Emissions Assessment. Draft. U.S. Environmental Protection Agency.
EPA (1996) Evaluation and Analysis of Gas Content and Coal Properties of Major Coal Bearing Regions of the United
States. EPA/600/R-96-065. U.S. Environmental Protection Agency.
ERG (2023). Correspondence between ERG and Buchanan Mine.
Geological Survey of Alabama State Oil and Gas Board (GSA) (2023) Well Records Database. Available online at
http://www.gsa.state.al.us/ogb/database.aspx.
IEA (2022) Coal 2022, International Energy Agency, Paris, License: CC BY 4.0. Available online at:
https://www.iea.org/reports/coal-2022.
IPCC (2019) 2019 Refinement to the 2006IPCC Guidelines for National Greenhouse Gas Inventories. Calvo Buendia,
E., Tanabe, K., Kranjc, A., Baasansuren, J., Fukuda, M., Ngarize S., Osako, A., Pyrozhenko, Y., Shermanau, P. and
Federici, S. (eds). Published: IPCC, Switzerland.
IPCC (2013) Climate Change 2013: The Physical Science Basis. Contribution of Working Group I to the Fifth
Assessment Report of the Intergovernmental Panel on Climate Change [Stocker, T.F., D. Qin, G.-K. Plattner, M.
Tignor, S.K. Allen, J. Boschung, A. Nauels, Y. Xia, V. Bex and P.M. Midgley (eds.)]. Cambridge University Press,
Cambridge, United Kingdom and New York, NY, USA, 1535 pp.
IPCC (2011) Use of Models and Facility-Level Data in Greenhouse Gas Inventories. Report of IPCC Expert Meeting on
Use of Models and Measurements in Greenhouse Gas Inventories 9-11 August 2010, Sydney, Australia. Eds:
Eggleston H.S., Srivastava N., Tanabe K., Baasansuren J., Fukuda M. IGES.
IPCC (2007) Climate Change 2007: The Physical Science Basis. Contribution of Working Group I to the Fourth
Assessment Report of the Intergovernmental Panel on Climate Change [Solomon, S., D. Qin, M. Manning, Z. Chen,
M. Marquis, K.B. Averyt, M. Tignor and H.L Miller (eds.)]. Cambridge University Press, Cambridge, United Kingdom
and New York, NY, USA, 996 pp.
JWR (2010) No. 4 & 7 Mines General Area Maps. Walter Energy: Jim Walter Resources.
King, Brian (1994) Management of Methane Emissions from Coal Mines: Environmental, Engineering, Economic and
Institutional Implication of Options. Neil and Gunter Ltd.
McElroy OVS (2023) Marshall County VAM Abatement Project Offset Verification Statement submitted to
California Air Resources Board, August 2023.
MSHA (2023) Data Transparency at MSHA. Mine Safety and Health Administration. Available online at
http://www.msha.gov/.
Mutmansky, Jan M. and Yanbei Wang (2000) Analysis of Potential Errors in Determination of Coal Mine Annual
Methane Emissions. Mineral Resources Engineering, 9(4).
Saghafi, Abouna (2013) Estimation of Fugitive Emissions from Open Cut Coal Mining and Measurable Gas Content.
13th Coal Operators' Conference, University of Wollongong, The Australian Institute of Mining and Metallurgy &
Mine Managers Association of Australia. 306-313.
USBM (1986) Results of the Direct Method Determination of the Gas Contents of U.S. Coal Basins. Circular 9067.
U.S. Bureau of Mines.
West Virginia Geological & Economic Survey (WVGES) (2023) Oil & Gas Production Data. Available online at
http://www.wvgs.wvnet.edu/www/datastat/datastat.htm.
References and Abbreviations 10-19
-------
Abandoned Underground Coal Mines
CMOP (2023) EPA's Coalbed Methane Outreach Program, Map of US Coal Mine Methane Current Projects and
Potential Opportunities. Available online at: https://www.epa.gov/cmop/map-us-coal-mine-methane-current-
projects-and-potential-opportunities.
COGIS (2018) Colorado Oil and Gas Information System. Colorado Oil and Gas Commission, Department of Natural
Resources. Available online at https://cogcc.state.co.us/data.html.
EPA (2004) Methane Emissions Estimates & Methodology for Abandoned Coal Mines in the U.S. Draft Final Report.
Washington, D.C. April 2004.
GMI (2021) Global Methane Initiative, International Coal Mine Methane Database. Available online at:
https://www.globalmethane.org/resources/details.aspx?resourceid=1981.
IPCC (2013) Climate Change 2013: The Physical Science Basis. Contribution of Working Group I to the Fifth
Assessment Report of the Intergovernmental Panel on Climate Change [Stocker, T.F., D. Qin, G.-K. Plattner, M.
Tignor, S.K. Allen, J. Boschung, A. Nauels, Y. Xia, V. Bex and P.M. Midgley (eds.)]. Cambridge University Press,
Cambridge, United Kingdom and New York, NY, USA, 1535 pp.
IPCC (2007) Climate Change 2007: The Physical Science Basis. Contribution of Working Group I to the Fourth
Assessment Report of the Intergovernmental Panel on Climate Change [Solomon, S., D. Qin, M. Manning, Z. Chen,
M. Marquis, K.B. Averyt, M. Tignor and H.L Miller (eds.)]. Cambridge University Press, Cambridge, United Kingdom
and New York, NY, USA, 996 pp.
MSHA (2023) U.S. Department of Labor, Mine Health & Safety Administration, Mine Data Retrieval System.
Available online at: https://www.msha.gov/mine-data-retrieval-system..
Petroleum Systems
API (1992) Global Emissions of Methane from Petroleum Sources. American Petroleum Institute, Health and
Environmental Affairs Department, Report No. DR140, February 1992.
BOEM (2023a) BOEM Platform Structures Online Query. Available online at:
https://www.data.boem.gov/Platform/PlatformStructures/Default.aspx.
BOEM (2023b) BOEM Oil and Gas Operations Reports - Part A (OGOR-A). Production Data for 1947 to 2022.
Download "Production Data" online at: https://www.data.boem.gov/Main/RawData.aspx.
BOEM (2023c) BOEM Oil and Gas Operations Reports - Part A (OGOR-A). Production Data for 1996 to 2022.
Available online at: https://www.data.boem.gov/Main/OGOR-A.aspx.
BOEM (2023d) BOEM Oil and Gas Operations Reports - Part B (OGOR-B). Flaring volumes for 1996 to 2022.
Available online at: https://www.data.boem.gov/Main/OGOR-B.aspx.
EIA (2023) Crude Oil Production. Energy Information Administration.
Enverus (2023) September 2023 Download. Enverus, Inc.
EPA (2023) Greenhouse Gas Reporting Program. U.S. Environmental Protection Agency. Data reported as of August
18, 2023.
EPA (2017) 2017 Nonpoint Oil and Gas Emission Estimation Tool, Version 1.2. Prepared for U.S. Environmental
Protection Agency by Eastern Research Group, Inc. (ERG). October 2019.
EPA (1999) Estimates of Methane Emissions from the U.S. Oil Industry (Draft Report). Prepared by ICF International.
Office of Air and Radiation, U.S. Environmental Protection Agency. October 1999.
10-20 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
EPA (1997) Compilation of Air Pollutant Emission Factors, AP-42. Office of Air Quality Planning and Standards, U.S.
Environmental Protection Agency. Research Triangle Park, NC. October 1997.
EPA/GRI (1996) Methane Emissions from the Natural Gas Industry. Prepared by Radian. U.S. Environmental
Protection Agency. April 1996.
IPCC (2006) 2006IPCC Guidelines for National Greenhouse Gas Inventories. The National Greenhouse Gas
Inventories Programme, The Intergovernmental Panel on Climate Change. [H.S. Eggleston, L. Buendia, K. Miwa, T.
Ngara, and K. Tanabe (eds.)]. Hayama, Kanagawa, Japan.
Natural Gas Systems
AHS (2022) U.S. Census Bureau's American Housing Survey (AHS). https://www.census.gov/programs-
surveys/ahs.html.
CBECS (2023) Energy Information Administration's Commercial Buildings Energy Consumption Survey (CBECS).
https://www.eia.gov/consumption/commercial.
CenSARA (2012) 2011 Oil and Gas Emission Inventory Enhancement Project for CenSARA States. Prepared by
ENVIRON International Corporation and Eastern Research Group, Inc. (ERG). Central States Air Resources Agencies
(CenSARA). December 2012.
Cusworth, D.H., Duren, R.M., Thorpe, A.K., Pandey S., Maasakkers, J.D., Aben, I., et al. (2021). Multisatellite
imaging of a gas well blowout enables quantification of total methane emissions. Geophysical Research Letters, 48,
e2020GL090864. https://doi.org/10.1029/2020GL090864.
EIA (2023a) Natural Gas Gross Withdrawals and Production. Energy Information Administration.
EIA (2023b) October 2021 Monthly Energy Review. Energy Information Administration.
https://www.eia.gov/totalenergv/data/monthly/archive/00352110.pdf.
Enverus (2023) September 2023 Download. Enverus, Inc.
EPA (2023a) MOVES3. https://www.epa.gov/moves/latest-version-motor-vehicle-emission-simulator-moves.
EPA (2023b) Greenhouse Gas Reporting Program- Subpart W-Petroleum and Natural Gas Systems. Environmental
Protection Agency. Data reported as of August 18, 2023.
EPA (2022) Nonpoint Oil & Gas Emission estimation Tool.
EPA (1977) Atmospheric Emissions from Offshore Oil and Gas Development and Production. Office of Air Quality
Planning and Standards, Research Triangle Park, NC. PB272268. June 1977.
Evans, D.J. & Chadwick, R.A. (2009) (eds) "Underground Gas Storage: Worldwide Experiences and Future
Development in the UK and Europe." The Geological Society, London, Special Publications, 313: 173-216.
https://doi.org/10.1144/SP313.12.
FERC (2023) Form No. 2, Major Natural Gas Pipeline Annual Report. Federal Energy Regulatory Commission.
https://ferc.gov/industries-data/natural-gas/industry-forms.
Fischer et al. (2018) "An Estimate of Natural Gas Methane Emissions from California Homes." Environmental
Science & Technology 2018, 52 (17), 10205-10213. https://pubs.acs.org/doi/10.1021/acs.est.8b03217.
GRI/EPA (1996) Methane Emissions from the Natural Gas Industry. Prepared by Harrison, M., T. Shires, J. Wessels,
and R. Cowgill, eds., Radian International LLC for National Risk Management Research Laboratory, Air Pollution
Prevention and Control Division, Research Triangle Park, NC. EPA-600/R-96-080a.
GTI (2019) Classification of Methane Emissions from Industrial Meters, Vintage vs Modern Plastic Pipe, and Plastic-
lined Steel and Cast-Iron Pipe. June 2019. Gas Technology Institute and U.S. Department of Energy GTI Project
Number 22070. DOE project Number ED-FE0029061.
References and Abbreviations 10-21
-------
GTI (2001) Gas Resource Database: Unconventional Natural Gas and Gas Composition Databases. Second Edition.
GRI-01/0136.
Illinois Office of Oil and Gas Resource Management (2022) State-level natural gas production quantities.
Indiana Division of Oil & Gas (2022) State-level natural gas production quantities.
Kansas Department of Health and Environment (2022) County-level produced water quantities.
Lamb, et al. (2015) "Direct Measurements Show Decreasing Methane Emissions from Natural Gas Local
Distribution Systems in the United States." Environmental Science & Technology, Vol. 49 5161-5169.
Lavoie et al. (2017) "Assessing the Methane Emissions from Natural Gas-Fired Power Plants and Oil Refineries."
Environmental Science & Technology. 2017 Mar 21;51(6):3373-3381. doi: 10.1021/acs.est.6b05531.
Li, H Z et al. (2022) "A national estimate of U.S. underground natural gas storage incident emissions."
Environmental Research Letters. 17: 084013. https://doi.org/10.1088/1748-9326/ac8Q69.
Maasakkers, Joannes D., Mark Omara, Ritesh Gautam, Alba Lorente, Sudhanshu Pandey, Paul Tol, Tobias Borsdorff,
Sander Houweling, Use Aben (2022). Reconstructing and quantifying methane emissions from the full duration of a
38-day natural gas well blowout using space-based observations. Remote Sensing of Environment.
https://doi.Org/10.1016/i.rse.2021.112755.
Ohio Environmental Protection Agency (2022) Well-level produced water quantities.
Oklahoma Department of Environmental Quality (2022) Well-level produced water quantities.
Pandey, S., Gautam, R., Houweling, S., van der Gon, H. D., Sadavarte, P., Borsdorff, T., et al. (2019). Satellite
observations reveal extreme methane leakage from a natural gas well blowout. Proceedings of the National
Academy of Sciences, 116, 26376- 26381. https://doi.org/10.1073/pnas.1908712116.
PHMSA (2022a) Gas Distribution Annual Data. Pipeline and Hazardous Materials Safety Administration, U.S.
Department of Transportation, Washington, DC. Available online at: https://www.phmsa.dot.gov/data-and-
statistics/pipeline/annual-report-mileage-gas-distribution-systems.
PHMSA (2022b) Underground Natural Gas STAR, Part C. Pipeline and Hazardous Materials Safety Administration,
U.S. Department of Transportation, Washington, DC. https://www.phmsa.dot.gov/data-and-statistics/pipeline/gas-
distribution-gas-gathering-gas-transmission-hazardous-liquids.
West Virginia Department of Environmental Protection (2020) State-level natural gas production quantities.
Zimmerle et al. (2019) "Characterization of Methane Emissions from Gathering Compressor Stations." October
2019. Available at https://mountainscholar.org/handle/10217/195489.
Zimmerle et al. (2015) "Methane Emissions from the Natural Gas Transmission and Storage System in the United
States." Environmental Science and Technology, Vol. 49 9374-9383.
Abandoned Oil and Gas Wells
Alaska Oil and Gas Conservation Commission, Available online at:
https://www.commerce.alaska.gov/web/aogcc/Data.aspx.
Arkansas Geological & Conservation Commission, "List of Oil & Gas Wells - Data From November 1,1936 to January
1,1955."
The Derrick's Handbook of Petroleum: A Complete Chronological and Statistical Review of Petroleum
Developments From 1859 to 1898 (V.l), (1898-1899) (V.2).
Enverus (2023) October 2023 Download. Enverus, Inc.
10-22 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
Florida Department of Environmental Protection - Oil and Gas Program, Available online at:
https://floridadep.gov/water/oil-gas.
Geological Survey of Alabama, Oil & Gas Board, Available online at: https://www.gsa.state.al.us/ogb/.
GRI/EPA (1996) Methane Emissions from the Natural Gas Industry. Prepared by Harrison, M., T. Shires, J. Wessels,
and R. Cowgill, eds., Radian International LLC for National Risk Management Research Laboratory, Air Pollution
Prevention and Control Division, Research Triangle Park, NC. EPA-600/R-96-080a.
GTI (2001) Gas Resource Database: Unconventional Natural Gas and Gas Composition Databases. Second Edition.
GRI-01/0136.
Interstate Oil and Gas Compact Commission (2021). IDLE AND ORPHAN OIL AND GAS WELLS: STATE AND
PROVINCIAL REGULATORY STRATEGIES 2021. Available online at:
https://iogcc.ok.gOv/sites/g/files/gmc836/f/iogcc idle and orphan wells 2021 final web.pdf.
Kang, et al. (2016) "Identification and characterization of high methane-emitting abandoned oil and gas wells."
PNAS, vol. 113 no. 48, 13636-13641, doi: 10.1073/pnas.l605913113.
Oklahoma Geological Survey. "Oklahoma Oil: Past, Present, and Future." Oklahoma Geology Notes, v. 62 no. 3,
2002 pp. 97-106.
Pennsylvania Department of Environmental Protection, Oil and Gas Reports - Oil and Gas Operator Well Inventory.
Available online at:
http://www.depreportingservices.state.pa.us/ReportServer/Pages/ReportViewer.aspx7/Oil Gas/OG Well Invento
II-
International Bunker Fuels
Anderson, B.E., et al. (2011) Alternative Aviation Fuel Experiment (AAFEX), NASA Technical Memorandum, in press.
ASTM (1989) Military Specification for Turbine Fuels, Aviation, Kerosene Types, NATO F-34 (JP-8) and NATO F-35.
February 10,1989.
DHS (2008) Personal Communication with Elissa Kay, Residual and Distillate Fuel Oil Consumption (International
Bunker Fuels). Department of Homeland Security, Bunker Report. January 11, 2008.
DLA Energy (2023) Unpublished data from the Defense Fuels Automated Management System (DFAMS). Defense
Energy Support Center, Defense Logistics Agency, U.S. Department of Defense. Washington, D.C.
DOC (1991 through 2022) Unpublished Report of Bunker Fuel Oil Laden on Vessels Cleared for Foreign Countries.
Form-563. Foreign Trade Division, Bureau of the Census, U.S. Department of Commerce. Washington, D.C.
DOT (1991 through 2013) Fuel Cost and Consumption. Federal Aviation Administration, Bureau of Transportation.
Statistics, U.S. Department of Transportation. Washington, D.C. DAI-10.
EIA (2024) Monthly Energy Review, February 2024, Energy Information Administration, U.S. Department of Energy,
Washington, D.C. DOE/EIA-0035(2024/02).
EPA (2020) EPA Inventory of U.S. Greenhouse Gas Emissions and Sinks 1990-2019: Updated Gasoline and Diesel
Fuel CO2 Emission Factors - Memo.
FAA (2024) Personal Communication between FAA and John Steller, Mausami Desai, and Vincent Camobreco for
aviation emissions estimates from the Aviation Environmental Design Tool (AEDT). March 2024.
IPCC (2013) Climate Change 2013: The Physical Science Basis. Contribution of Working Group I to the Fifth
Assessment Report of the Intergovernmental Panel on Climate Change. [Stocker, T.F., D. Qin, G.K. Plattner, M.
Tignor, S.K. Allen, J. Boschung, A. Nauels, Y. Xia, V. Bex and P.M. Midgley (eds.)]. Cambridge University Press,
Cambridge, United Kingdom and New York, NY, USA, 1535 pp.
References and Abbreviations 10-23
-------
IPCC (2007) Climate Change 2007: The Physical Science Basis. Contribution of Working Group I to the Fourth
Assessment Report of the Intergovernmental Panel on Climate Change. [S. Solomon, D. Qin, M. Manning, Z. Chen,
M. Marquis, K.B. Averyt, M. Tignor and H.L Miller (eds.)]. Cambridge University Press. Cambridge, United Kingdom,
996 pp.
IPCC (2006) 2006 IPCC Guidelines for National Greenhouse Gas Inventories. The National Greenhouse Gas
Inventories Programme, The Intergovernmental Panel on Climate Change. [H.S. Eggleston, L. Buendia, K. Miwa, T.
Ngara, and K. Tanabe (eds.)]. Hayama, Kanagawa, Japan.USAF (1998) Fuel Logistics Planning. U.S. Air Force
pamphlet AFPAM23-221, May 1,1998.
IPCC/UNEP/OECD/IEA (1997) Revised 1996 IPCC Guidelines for National Greenhouse Gas Inventories. 31
Intergovernmental Panel on Climate Change, United Nations Environment Programme, Organization for Economic
32 Co-Operation and Development, International Energy Agency, Paris, France.
Wood Biomass and Biofuel Consumption
EIA (2024a) Monthly Energy Review, February 2024. Energy Information Administration, U.S. Department of
Energy. Washington, D.C. DOE/EIA-0035(2024/02).
EIA (2024b) Biofuels explained: Use of biomass-based diesel fuel. Energy Information Administration, U.S.
Department of Energy. Washington, D.C. Available online at: https://www.eia.gov/energyexplained/biofuels/use-
of-biodiesel.php.
EPA (2023). Greenhouse Gas Reporting Program (GHGRP). 2022 Envirofacts. Available online at:
https://ghgdata.epa.gov/ghgp/main.do.
EPA (2010) Carbon Content Coefficients Developed for EPA's Mandatory Reporting Rule. Office of Air and
Radiation, Office of Atmospheric Programs, U.S. Environmental Protection Agency, Washington, D.C.
Lindstrom, P. (2006) Personal Communication. Perry Lindstrom, Energy Information Administration and Jean Kim,
ICF International.
Energy Sources of Precursor Greenhouse Gases
EPA (2023a) EPA's Emissions Inventory System (EIS) to National Inventory Report (NIR) Mapping file
EIS_NIR_mapping.xlsx. U.S. Environmental Protection Agency. Washington, D.C.
EPA (2023b) "Criteria pollutants National Tier 1 for 1970 - 2023." National Emissions Inventory (NEI) Air Pollutant
Emissions Trends Data. Office of Air Quality Planning and Standards, March 2024. Available online at:
https://www.epa.gov/air-emissions-inventories/air-pollutant-emissions-trends-data.
EPA (2023c) "2020 National Emissions Inventory Technical Support Document: Introduction." Office of Air Quality
Planning and Standards, March 2023. Available online at: https://www.epa.gov/system/files/documents/2023-
01/NE12020 TSD Section 1 lntroduction.pdf.
Industrial Processes and Product Use
EPA (2014) Greenhouse Gas Reporting Program. Developments on Publication of Aggregated Greenhouse Gas
Data, November 25, 2014. See http://www.epa.gov/ghgreporting/confidential-business-information-ghg-
reporting.
EPA (2002) Quality Assurance/Quality Control and Uncertainty Management Plan for the U.S. Greenhouse Gas
Inventory: Procedures Manual for Quality Assurance/Quality Control and Uncertainty Analysis, U.S. Greenhouse
10-24 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
Gas Inventory Program, U.S. Environmental Protection Agency, Office of Atmospheric Programs, EPA430-R-02-
007B, June 2002.
IPCC (2011) Use of Models and Facility-Level Data in Greenhouse Gas Inventories (Report of IPCC Expert Meeting
on Use of Models and Measurements in Greenhouse Gas Inventories 9-11 August 2010, Sydney, Australia) eds.:
Eggleston H.S., Srivastava N., Tanabe K., Baasansuren J., Fukuda M., Pub. IGES, Japan 2011.
IPCC (2007) Climate Change 2007: The Physical Science Basis. Contribution of Working Group I to the Fourth
Assessment Report of the Intergovernmental Panel on Climate Change. [S. Solomon, D. Qin, M. Manning, Z. Chen,
M. Marquis, K.B. Averyt, M. Tignor and H.L Miller (eds.)]. Cambridge University Press. Cambridge, United Kingdom
996 pp.
Cement Production
IPCC (2006) 2006 IPCC Guidelines for National Greenhouse Gas Inventories. The National Greenhouse Gas
Inventories Programme, The Intergovernmental Panel on Climate Change. [H.S. Eggleston, L. Buendia, K. Miwa, T.
Ngara, and K. Tanabe (eds.)]. Hayama, Kanagawa, Japan.
U.S. Bureau of Mines (1990 through 1993) Minerals Yearbook: Cement Annual Report. U.S. Department of the
Interior, Washington, D.C.
U.S. Environmental Protection Agency (EPA) (2015) Greenhouse Gas Reporting Program Report Verification.
Available online at https://www.epa.gov/sites/production/files/2015-
07/documents/ghgrp verification factsheet.pdf.
U.S. EPA (2023) Greenhouse Gas Reporting Program (GHGRP). Aggregation of Reported Facility Level Data under
Subpart H -National Level Clinker Production from Cement Production for Calendar Years 2014 through 2022.
Office of Air and Radiation, Office of Atmospheric Programs, U.S. Environmental Protection Agency, Washington,
D.C.
United States Geological Survey (USGS) (2023a) 2021 Minerals Yearbook - Cement (Advance Release Tables). U.S.
Geological Survey, Reston, VA. July 2023.
USGS (2023b) Mineral Commodity Summaries: Cement. U.S. Geological Survey, Reston, VA. January 2023.
USGS (2023c) Mineral Industry Surveys, Cement in December 2022. U.S. Geological Survey, Reston, VA. (March
2023).
USGS (1995 through 2014) Minerals Yearbook - Cement. U.S. Geological Survey, Reston, VA.
Van Oss (2013a) 1990 through 2012 Clinker Production Data Provided by Hendrik van Oss (USGS) via email on
November 8, 2013.
Van Oss (2013b) Personal communication. Hendrik van Oss, Commodity Specialist of the U.S. Geological Survey
and Gopi Manne, Eastern Research Group, Inc. October 28, 2013.
Lime Production
IPCC (2006) 2006 IPCC Guidelines for National Greenhouse Gas Inventories. The National Greenhouse Gas
Inventories Programme, The Intergovernmental Panel on Climate Change. [H.S. Eggleston, L. Buendia, K. Miwa, T.
Ngara, and K. Tanabe (eds.)]. Hayama, Kanagawa, Japan.
Males, E. (2003) Memorandum from Eric Males, National Lime Association to William N. Irving & Leif Hockstad,
Environmental Protection Agency. March 6, 2003.
Miner, R. and B. Upton (2002) Methods for estimating greenhouse gas emissions from lime kilns at kraft pulp mills.
Energy. Vol. 27 (2002), p. 729-738.
References and Abbreviations 10-25
-------
RTI (2023) Expert judgment, RTI International. March 30, 2023.
Seeger (2013) Memorandum from Arline M. Seeger, National Lime Association to Leif Hockstad, Environmental
Protection Agency. March 15, 2013.
U.S. Environmental Protection Agency (EPA) (2023) Greenhouse Gas Reporting Program (GHGRP). Aggregation of
Reported Facility Level Data under Subpart S-National Lime Production for Calendar Years 2010 through
2022. Office of Air and Radiation, Office of Atmospheric Programs, U.S. Environmental Protection Agency,
Washington, D.C.
United States Geological Survey (USGS) (2023a) (1996 through 2023) Mineral Commodities Summary: Lime. U.S.
Geological Survey, Reston, VA (January 2023). Latest edition was updated in 2023 for 2022. Applicable editions are
available at: https://www.usgs.gov/centers/national-minerals-information-center/lime-statistics-and-information.
USGS (2023b) (2002 through 2021) Minerals Yearbook Annual Tables: Lime. U.S. Geological Survey, Reston, VA
(January 2023). Latest edition was updated in 2023 for 2021 tables. Applicable editions are available at:
https://www.usgs.gov/centers/national-minerals-information-center/lime-statistics-and-information.
USGS (2021) (1991 through 2018) Minerals Yearbook: Lime. U.S. Geological Survey, Reston, VA (October 2021).
Latest edition was updated in 2021 for 2018. Applicable editions are available at:
https://www.usgs.gov/centers/national-minerals-information-center/lime-statistics-and-information. See
"Archive" for editions prior to 1993.
USGS (2012) 2012 Expert Elicitation. Michael Miller, U.S. Geological Survey (2012).
Glass Production
Federal Reserve (2023) Board of Governors of the Federal Reserve System (US), Industrial Production:
Manufacturing: Durable Goods: Glass and Glass Product (NAICS = 3272) [IPG3272N], retrieved from FRED, Federal
Reserve Bank of St. Louis. Available at: https://fred.stlouisfed.org/series/lPG3272N. Accessed on November 21,
2023.
IPCC (2006) 2006IPCC Guidelines for National Greenhouse Gas Inventories. The National Greenhouse Gas
Inventories Programme, The Intergovernmental Panel on Climate Change. [H.S. Eggleston, L. Buendia, K. Miwa, T.
Ngara, and K. Tanabe (eds.)]. Hayama, Kanagawa, Japan.
RTI (2023) Expert judgment, RTI International. March 30, 2023.
RTI (2022) Expert judgment. Melissa Icenhour, RTI International. November 16, 2022.
U.S. Bureau of Mines (1991 and 1993a) Minerals Yearbook: Crushed Stone Annual Report. U.S. Department of the
Interior. Washington, D.C.
U.S. Department of Energy (DOE) (2002) Glass Industry of the Future: Energy and Environmental Profile of the U.S.
Glass Industry. Office of Industrial Technologies, U.S. Department of Energy. Washington, D.C.
U.S. Environmental Protection Agency (EPA) (2023) Greenhouse Gas Reporting Program (GHGRP). Aggregation of
Reported Facility Level Data under Subpart N -National Glass Production for Calendar Years 2010 through 2022.
Office of Air and Radiation, Office of Atmospheric Programs, U.S. Environmental Protection Agency, Washington,
D.C.
U.S. EPA (2009) Technical Support Document for the Glass Manufacturing Sector: Proposed Rule for Mandatory
Reporting of Greenhouse Gases. U.S. Environmental Protection Agency, Washington, D.C.
United States Geological Survey (USGS) (1995 through 2015b) Minerals Yearbook: Soda Ash Annual Report. U.S.
Geological Survey, Reston, VA.
10-26 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
USGS (2023) Mineral Industry Surveys: Soda Ash in January 2023. U.S. Geological Survey, Reston, VA. November
2023. Available online at: https://www.usgs.gov/centers/national-minerals-information-center/soda-ash-statistics-
and-information.
USGS (2022) Mineral Industry Surveys: Soda Ash in June 2022. U.S. Geological Survey, Reston, VA. November 2022.
Available online at: Index of /minerals-information-archives/soda ash (usgs.gov).
USGS (2021) Mineral Industry Surveys: Soda Ash in April 2021. U.S. Geological Survey, Reston, VA. July 2021.
Available online at: Index of /minerals-information-archives/soda ash (usgs.gov).
USGS (2020) Mineral Industry Surveys: Soda Ash in April 2020. U.S. Geological Survey, Reston, VA. July 2020.
Available online at: Index of /minerals-information-archives/soda ash (usgs.gov).
USGS (2019) Mineral Industry Surveys: Soda Ash in December 2018. U.S. Geological Survey, Reston, VA. March
2019. Available online at: Index of /minerals-information-archives/soda ash (usgs.gov).
USGS (2018) Mineral Industry Surveys: Soda Ash in February 2018. U.S. Geological Survey, Reston, VA. 2018.
Available online at: Index of /minerals-information-archives/soda ash (usgs.gov).
USGS (2017) Minerals Industry Surveys: Soda Ash in January 2017. U.S. Geological Survey, Reston, VA. March 2017.
Available online at: Index of /minerals-information-archives/soda ash (usgs.gov).
Other Process Uses of Carbonates
AISI (2018 through 2021) Annual Statistical Report. American Iron and Steel Institute.
IPCC (2006) 2006IPCC Guidelines for National Greenhouse Gas Inventories. The National Greenhouse Gas
Inventories Programme, The Intergovernmental Panel on Climate Change. [H.S. Eggleston, L. Buendia, K. Miwa, T.
Ngara, and K. Tanabe (eds.)]. Hayama, Kanagawa, Japan.
Kostick, D. S. (2012) Personal communication. Dennis S. Kostick, U.S. Geological Survey, Soda Ash Commodity
Specialist and Gopi Manne and Bryan Lange of Eastern Research Group, Inc. October 2012.
McNeece (2023) Personal communication, Steve McNeece, Nevada Department of Environmental Quality and
Amanda Chiu, U.S. Environmental Protection Agency. November 28, 2023.
RTI (2023) Expert judgment, RTI International. March 30, 2023.
Simmons (2024) Personal communication, Kristi Simmons, U.S. Geological Survey and Amanda Chiu, U.S.
Environmental Protection Agency. February 9, 2024.
U.S. Bureau of Mines (1991 and 1993a) Minerals Yearbook: Crushed Stone Annual Report. U.S. Department of the
Interior. Washington, D.C.
U.S. Environmental Protection Agency (EPA) (2023). Greenhouse Gas Reporting Program (GHGRP). Dataset as of
August 18, 2023. Available online at: https://ghgdata.epa.gov/ghgp/
United States Geological Survey (USGS) (2023a) Mineral Industry Surveys: Soda Ash in September 2023. U.S.
Geological Survey, Reston, VA. November 2023.
USGS (2023b) 2021 Minerals Yearbook: Stone, Crushed [Advanced Release]. U.S. Geological Survey, Reston, VA.
June 2023.
USGS (2022a) Mineral Industry Surveys: Soda Ash in August 2022. U.S. Geological Survey, Reston, VA. November
2022.
USGS (2022b) 2020 Minerals Yearbook: Stone, Crushed [Advanced Release]. U.S. Geological Survey, Reston, VA.
August 2022.
References and Abbreviations 10-27
-------
USGS (2022c) 2018 Minerals Yearbook: Stone, Crushed [Advanced Release], U.S. Geological Survey, Reston, VA.
August 2022.
USGS (2022d) 2019 Minerals Yearbook: Stone, Crushed [Advanced Release], U.S. Geological Survey, Reston, VA.
June 2022.
USGS (2022e) 2018 Minerals Yearbook: Magnesium Compounds [Advanced Release], U.S. Geological Survey,
Reston, VA. May 2022.
USGS (2022f) 2018 Minerals Yearbook: Clay and Shale [Advanced Release], U.S. Geological Survey, Reston, VA.
March 2022.
USGS (2022g) 2018 Minerals Yearbook: Soda Ash [Advanced Release], U.S. Geological Survey, Reston, VA. January
2022.
USGS (2021a) Minerals Yearbook 2019: Soda Ash [Advanced Data Release of the 2019 Annual Tables], U.S.
Geological Survey, Reston, VA. August 2021.
USGS (2021b) Mineral Industry Surveys: Soda Ash in April 2021. U.S. Geological Survey, Reston, VA. July 2021.
USGS (2021c) 2017 Minerals Yearbook: Stone, Crushed [Advanced Release], U.S. Geological Survey, Reston, VA.
June 2021.
USGS (2021d) 2020 Mineral Commodity Summaries: Stone (Crushed). U.S. Geological Survey, Reston, VA. January
2021.
USGS (2020a) Minerals Yearbook 2017: Stone, Crushed [Advanced Data Release of the 2017 Annual Tables]. U.S.
Geological Survey, Reston, VA. August 2020.
USGS (2020b) Mineral Industry Surveys: Soda Ash in April 2020. U.S. Geological Survey, Reston, VA. July 2020.
USGS (2020c) 2016 Minerals Yearbook: Stone, Crushed [Advanced Release]. U.S. Geological Survey, Reston, VA.
January 2020.
USGS (2019) Mineral Industry Surveys: Soda Ash in April 2019. U.S. Geological Survey, Reston, VA. July 2019.
USGS (2018) Mineral Industry Surveys: Soda Ash in February 2018. U.S. Geological Survey, Reston, VA. 2018.
USGS (2017) Mineral Industry Surveys: Soda Ash in January 2017. U.S. Geological Survey, Reston, VA. March 2017.
USGS (1995a through 2017) Minerals Yearbook: Crushed Stone Annual Report. U.S. Geological Survey, Reston, VA.
USGS (1994 through 2015b) Minerals Yearbook: Soda Ash Annual Report. U.S. Geological Survey, Reston, VA.
USGS (1990 through 2002) Minerals Yearbook: Magnesium Compounds Annual Report. U.S. Geological Survey,
Reston, VA.
USGS (1948) Reports: Magnesite and brucite deposits at Gabbs, Nye County, Nevada. U.S. Geological Survey,
Reston, VA
Willett (2023) Personal communication, Jason Christopher Willett, U.S. Geological Survey and Amanda Chiu, U.S.
Environmental Protection Agency. November 21, 2023.
Willett (2017) Personal communication, Jason Christopher Willett, U.S. Geological Survey and Mausami Desai and
John Steller, U.S. Environmental Protection Agency. March 9, 2017.
Ammonia Production
ACC (2023) Business of Chemistry (Annual Data). American Chemistry Council, Arlington, VA.
Coffeyville Resources Energy, Inc. (CVR) (2008) CVR Energy, Inc. 2008 Annual Report. Available online at:
https://cvrenergv.gcs-web.com/annual-report-and-proxy-archive.
10-28 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
CVR (2022) CVR Energy, Inc. 2022 Annual Report on Form 10-K. Available online at: https://cvrenergy.gcs-
web.com/annual-report-proxy-information.
CVR (2021) CVR Energy, Inc. 2021 CVI Annual Report on Form 10-K. Available online at: https://cvrenergy.gcs-
web.com/annual-report-and-proxy-archive.
CVR (2020) CVR Energy, Inc. 2020 CVI Annual Report on Form 10-K --Final. Available online at:
https://cvrenergy.gcs-web.com/annual-report-and-proxy-archive.
CVR (2019) CVR Energy, Inc. 2019 CVI Form 10-K - Final. Available online at https://cvrenergy.gcs-web.com/annual-
report-and-proxy-archive.
CVR (2018) CVR Energy, Inc. 2018 CVI Annual Report on Form 10-K --Final. Available online at:
https://cvrenergy.gcs-web.com/annual-report-and-proxy-archive.
CVR (2017) CVR Energy, Inc. 2017 CVI Annual Report on Form 10-K (Web). Available online at:
https://cvrenergy.gcs-web.com/annual-report-and-proxy-archive.
CVR (2016) CVR Energy, Inc. 2016 CVI Annual Report on Form 10-K (Web). Available online at:
https://cvrenergy.gcs-web.com/annual-report-and-proxy-archive.
CVR (2014) CVR Energy, Inc. 2014 Annual Report. Available online at: https://cvrenergy.gcs-web.com/annual-
report-and-proxy-archive.
CVR (2015) CVR Energy, Inc. 2015 Annual Report. Available online at: https://cvrenergy.gcs-web.com/annual-
report-and-proxy-archive.
CVR (2013) CVR Energy, Inc. 2013 Annual Report. Available online at: https://cvrenergy.gcs-web.com/annual-
report-and-proxy-archive.
CVR (2012) CVR Energy, Inc. 2012 Annual Report. Available online at: https://cvrenergy.gcs-web.com/annual-
report-and-proxy-archive.
CVR (2011) CVR Energy, Inc. 2011 Annual Report. Available online at: https://cvrenergy.gcs-web.com/annual-
report-and-proxy-archive.
CVR (2010) CVR Energy, Inc. 2010 Annual Report. Available online at: https://cvrenergy.gcs-web.com/annual-
report-and-proxy-archive.
CVR (2009) CVR Energy, Inc. 2009 Annual Report. Available online at: https://cvrenergy.gcs-web.com/annual-
report-and-proxy-archive.
EFMA (2000) Best Available Techniques for Pollution Prevention and Control in the European Fertilizer Industry.
Booklet No. 5 of 8: Production of Urea and Urea Ammonium Nitrate. Available online at:
http://fertilizerseurope.com/site/index.php?id=390.
IPCC (2006) 2006IPCC Guidelines for National Greenhouse Gas Inventories. The National Greenhouse Gas
Inventories Programme, The Intergovernmental Panel on Climate Change. [H.S. Eggleston, L. Buendia, K. Miwa, T.
Ngara, and K. Tanabe (eds.)]. Hayama, Kanagawa, Japan.
RTI (2023) Expert judgment, RTI International. March 30, 2023.
United States Census Bureau (2011) Current Industrial Reports Fertilizer Materials and Related Products: 2010
Summary. Available online at: http://www.census.gov/manufacturing/cir/historical data/mq325b/index.html.
U.S. Census Bureau (2010) Current Industrial Reports Fertilizer Materials and Related Products: 2009 Summary.
Available online at: http://www.census.gov/manufacturing/cir/historical data/mq325b/index.html.
U.S. Census Bureau (2009) Current Industrial Reports Fertilizer Materials and Related Products: 2008 Summary.
Available online at: http://www.census.gov/manufacturing/cir/historical data/mq325b/index.html.
References and Abbreviations 10-29
-------
U.S. Census Bureau (2008) Current Industrial Reports Fertilizer Materials and Related Products: 2007Summary.
Available online at: http://www.census.gov/cir/www/325/mq325b/mq325b075.xls.
U.S. Census Bureau (2007) Current Industrial Reports Fertilizer Materials and Related Products: 2006 Summary.
Available online at: http://www.census.gOv/industry/l/mq325b065.pdf.
U.S. Census Bureau (2006) Current Industrial Reports Fertilizer Materials and Related Products: 2005 Summary.
Available online at: http://www.census.gov/cir/www/325/mq325b.html.
U.S. Census Bureau (2004, 2005) Current Industrial Reports Fertilizer Materials and Related Products: Fourth
Quarter Report Summary. Available online at: http://www.census.gov/cir/www/325/mq325b.html.
U.S. Census Bureau (1998 through 2003) Current Industrial Reports Fertilizer Materials and Related Products:
Annual Reports Summary. Available online at: http://www.census.gov/cir/www/325/mq325b.html.
U.S. Census Bureau (1991 through 1994) Current Industrial Reports Fertilizer Materials Annual Report. Report No.
MQ28B. U.S. Census Bureau, Washington, D.C.
United States EIA (2023) Monthly Energy Review, February 2023, Energy Information Administration, U.S.
Department of Energy, Washington, DC. DOE/EIA-0035(2023/2).
United States Environmental Protection Agency (EPA) (2023) Greenhouse Gas Reporting Program. Aggregation of
Reported Facility Level Data under Subpart G -Annual Urea Production from Ammonia Manufacturing for Calendar
Years 2017-2022. Office of Air and Radiation, Office of Atmospheric Programs, U.S. Environmental Protection
Agency, Washington, D.C.
U.S. EPA (2018) Greenhouse Gas Reporting Program. Aggregation of Reported Facility Level Data under Subpart G -
Annual Urea Production from Ammonia Manufacturing for Calendar Years 2011-2016. Office of Air and Radiation,
Office of Atmospheric Programs, U.S. Environmental Protection Agency, Washington, D.C
United States Geological Survey (USGS) (2023) 2023 Mineral Commodity Summaries: Nitrogen (Fixed) - Ammonia.
January 2023. Available online at: https://pubs.usgs.gov/periodicals/mcs2023/mcs2023-nitrogen.pdf.
USGS (1994-2009) Minerals Yearbook: Nitrogen. Available online at:
http://minerals.usgs.gov/minerals/pubs/commodity/nitrogen/
Urea Consumption for Non-Agricultural Purposes
EFMA (2000) Best Available Techniques for Pollution Prevention and Control in the European Fertilizer Industry.
Booklet No. 5 of 8: Production of Urea and Urea Ammonium Nitrate.
IPCC (2006) 2006IPCC Guidelines for National Greenhouse Gas Inventories. The National Greenhouse Gas
Inventories Programme, The Intergovernmental Panel on Climate Change. [H.S. Eggleston, L. Buendia, K. Miwa, T.
Ngara, and K. Tanabe (eds.)]. Hayama, Kanagawa, Japan.
RTI (2023) Expert judgment, RTI International. March 30, 2023.
TFI (2002) U.S. Nitrogen Imports/Exports Table. The Fertilizer Institute. Available online at:
http://www.tfi.org/statistics/usnexim.asp. August 2002.
United States Census Bureau (2001 through 2011) Current Industrial Reports Fertilizer Materials and Related
Products: Annual Summary. Available online at:
http://www.census.gov/manufacturing/cir/historical data/index.html.
United States Department of Agriculture (2012) Economic Research Service Data Sets, Data Sets, U.S. Fertilizer
Imports/Exports: Standard Tables. Available online at: http://www.ers.usda.gov/data-products/fertilizer-
importsexports/standard-tables.aspx.
10-30 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
U.S. EPA (2023a) Greenhouse Gas Reporting Program. Aggregation of Reported Facility Level Data under Subpart G
-Annual Urea Production from Ammonia Manufacturing for Calendar Years 2017-2022. Office of Air and Radiation,
Office of Atmospheric Programs, U.S. Environmental Protection Agency, Washington, D.C.
U.S. EPA (2023b). Greenhouse Gas Reporting Program. Dataset as of August 18, 2023. Available online at:
https://ghgdata.epa.gov/ghgp/.
U.S. EPA (2018) Greenhouse Gas Reporting Program. Aggregation of Reported Facility Level Data under Subpart G -
Annual Urea Production from Ammonia Manufacturing for Calendar Years 2011-2016. Office of Air and Radiation,
Office of Atmospheric Programs, U.S. Environmental Protection Agency, Washington, D.C.
United States International Trade Commission (ITC) (2002) United States International Trade Commission
Interactive Tariff and Trade DataWeb, Version 2.5.0. Available online at: http://dataweb.usitc.gov/. August 2002.
United States Geological Survey (USGS) (1994 through 2023a) Minerals Yearbook: Nitrogen. Available online at:
http://minerals.usgs.gov/minerals/pubs/commodity/nitrogen/.
USGS (2023b) Minerals Commodity Summaries: Nitrogen (Fixed)-Ammonia. January 2023. Available online at:
https://pubs.usgs.gov/periodicals/mcs2023/mcs2023-nitrogen.pdf.
Nitric Acid Production
IPCC (2006) 2006IPCC Guidelines for National Greenhouse Gas Inventories. The National Greenhouse Gas
Inventories Programme, The Intergovernmental Panel on Climate Change. [H.S. Eggleston, L. Buendia, K. Miwa, T.
Ngara, and K. Tanabe (eds.)]. Hayama, Kanagawa, Japan.
Icenhour (2020) Personal communication, Melissa Icenhour, RTI International and Amanda Chiu, U.S.
Environmental Protection Agency. December 3, 2020.
RTI (2023) Expert judgment, RTI International. March 30, 2023.
United States Census Bureau (2010a) Current Industrial Reports. Fertilizers and Related Chemicals: 2009. 'Table 1:
Summary of Production of Principle Fertilizers and Related Chemicals: 2009 and 2008." June, 2010. MQ325B(08)-5.
Available online at: http://www.census.gov/manufacturing/cir/historical data/mq325b/index.html.
U.S. Census Bureau (2010b) Personal communication between Hilda Ward (of U.S. Census Bureau) and Caroline
Cochran (of ICF International). October 26, 2010 and November 5, 2010.
U.S. Census Bureau (2009) Current Industrial Reports. Fertilizers and Related Chemicals: 2008. 'Table 1: Shipments
and Production of Principal Fertilizers and Related Chemicals: 2004 to 2008." June, 2009. MQ325B(08)-5. Available
online at: http://www.census.gov/manufacturing/cir/historical data/mq325b/index.html.
U.S. Census Bureau (2008) Current Industrial Reports. Fertilizers and Related Chemicals: 2007. 'Table 1: Shipments
and Production of Principal Fertilizers and Related Chemicals: 2003 to 2007." June, 2008. MQ325B(07)-5. Available
online at: http://www.census.gov/manufacturing/cir/historical data/mq325b/index.html.
U.S. EPA (2023) Greenhouse Gas Reporting Program. Aggregation of Reported Facility Level Data under Subpart V -
National Nitric Acid Production for Calendar Years 2017 through 2022. Office of Air and Radiation, Office of
Atmospheric Programs, U.S. Environmental Protection Agency, Washington, D.C.
U.S. EPA (2018) Greenhouse Gas Reporting Program. Aggregation of Reported Facility Level Data under Subpart V -
National Nitric Acid Production for Calendar Years 2010 through 2016. Office of Air and Radiation, Office of
Atmospheric Programs, U.S. Environmental Protection Agency, Washington, D.C.
U.S. EPA (2015) Greenhouse Gas Reporting Program Report Verification. Available online at
https://www.epa.gov/sites/production/files/2015-07/documents/ghgrp verification factsheet.pdf.
References and Abbreviations 10-31
-------
U.S. EPA (2010) Available and Emerging Technologies for Reducing Greenhouse Gas Emissions from the Nitric Acid
Production Industry. Office of Air Quality Planning and Standards, U.S. Environmental Protection Agency. Research
Triangle Park, NC. December 2010. Available online at: http://www.epa.gov/nsr/ghgdocs/nitricacid.pdf.
U.S. EPA (1998) Compilation of Air Pollutant Emission Factors, AP-42. Office of Air Quality Planning and Standards,
U.S. Environmental Protection Agency. Research Triangle Park, NC. February 1998.
Adipic Acid Production
ACC (2023) Business of Chemistry (Annual Data). American Chemistry Council, Arlington, VA.
Ard (2024) Personal communication, Howard Ard, Florida Department of Environmental Protection and Amanda
Chiu, U.S. Environmental Protection Agency. February 6, 2024.
Ascend (2023) "Ascend reaches GHG abatement milestone." Available online at
https://www.ascendmaterials.com/news/ascend-reaches-ghg-abatement-milestone. December 1, 2023.
C&EN (1995) "Production of Top 50 Chemicals Increased Substantially in 1994." Chemical & Engineering News,
73(15):17. April 10,1995.
C&EN (1994) 'Top 50 Chemicals Production Rose Modestly Last Year." Chemical & Engineering News, 72(15):13.
April 11,1994.
C&EN (1993) 'Top 50 Chemicals Production Recovered Last Year." Chemical & Engineering News, 71(15):11. April
12,1993.
C&EN (1992) "Production of Top 50 Chemicals Stagnates in 1991." Chemical & Engineering News, 70(15): 17. April
13,1992.
CMR (2001) "Chemical Profile: Adipic Acid." Chemical Market Reporter. July 16, 2001.
CMR (1998) "Chemical Profile: Adipic Acid." Chemical Market Reporter. June 15,1998.
CW (2005) "Product Focus: Adipic Acid." Chemical Week. May 4, 2005.
CW (1999) "Product Focus: Adipic Acid/Adiponitrile." Chemical Week, p. 31. March 10,1999.
Desai (2010, 2011) Personal communication. Mausami Desai, U.S. Environmental Protection Agency and Adipic
Acid Plant Engineers. 2010 and 2011.
ICIS (2007) "Adipic Acid." ICIS Chemical Business Americas. July 9, 2007.
IPCC (2006) 2006IPCC Guidelines for National Greenhouse Gas Inventories. The National Greenhouse Gas
Inventories Programme, The Intergovernmental Panel on Climate Change. [H.S. Eggleston, L. Buendia, K. Miwa, T.
Ngara, and K. Tanabe (eds.)]. Hayama, Kanagawa, Japan.
Reimer, R.A., Slaten, C.S., Seapan, M., Koch, T.A. and Triner, V.G. (1999) "Implementation of Technologies for
Abatement of N2O Emissions Associated with Adipic Acid Manufacture." Proceedings of the 2nd Symposium on
Non-C02 Greenhouse Gases (NCGG-2), Noordwijkerhout, The Netherlands, 8-10 Sept. 1999, Ed. J. van Ham et al.,
Kluwer Academic Publishers, Dordrecht, pp. 347-358.
RTI (2023) Expert judgment, RTI International. March 30, 2023.
Thiemens, M.H., and W.C. Trogler (1991) "Nylon production; an unknown source of atmospheric nitrous oxide."
Science 251:932-934.
United States Environmental Protection Agency (EPA) (2021 through 2023) Greenhouse Gas Reporting Program.
Subpart E Data. Office of Air and Radiation, Office of Atmospheric Programs, U.S. Environmental Protection
Agency, Washington, D.C. Available online at: https://www.epa.gov/ghgreporting/ghg-reporting-program-data-
sets.
10-32 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
U.S. EPA (2019, 2020) Greenhouse Gas Reporting Program. Subpart E, S-CEMS, BB, CC, LL Data Set (XLSX) (Adipic
Acid Tab). Office of Air and Radiation, Office of Atmospheric Programs, U.S. Environmental Protection Agency,
Washington, D.C. Available online at: https://www.epa.gov/ghgreporting/ghg-reporting-program-data-sets.
U.S. EPA (2015) Greenhouse Gas Reporting Program Report Verification. Available online at
https://www.epa.gov/sites/production/files/2015-07/documents/ghgrp verification factsheet.pdf.
U.S. EPA (2014 through 2018) Greenhouse Gas Reporting Program. Subpart E, S-CEMS, BB, CC, LL Data Set (XLSX)
(Adipic Acid Tab). Office of Air and Radiation, Office of Atmospheric Programs, U.S. Environmental Protection
Agency, Washington, D.C. Available online at: http://www2.epa.gov/ghgreporting/ghg-reporting-program-data-
sets.
U.S. EPA (2010 through 2013) Analysis of Greenhouse Gas Reporting Program data - Subpart E (Adipic Acid), Office
of Air and Radiation, Office of Atmospheric Programs, U.S. Environmental Protection Agency, Washington, D.C.
Caprolactam, Glyoxal and Glyoxylic Acid Production
American Chemistry Council (ACC) (2023) Business of Chemistry (Annual Data). American Chemistry Council,
Arlington, VA.
AdvanSix (2023) AdvanSix's Hopewell Facility Fact Sheet. Retrieved from:
https://www.advansix.com/hopewell/about-us/ on September 13, 2023.
BASF (2023) Welcome to BASF in Freeport Texas. Retrieved from https://www.basf.com/us/en/who-we-
are/organization/locations/featured-sites/Freeport.html on September 13, 2023.
ChemView (2021). Compilation of data submitted under TSCA in 2012 and 2016. Accessed April 2021. Available at
https://chemview.epa.gov/chemview.
Cline, D. (2019) Firm to Clean Up and Market Former Fibrant Site. The Augusta Chronicle. September 9, 2019.
Retrieved from https://www.augustachronicle.com.
Ecofys, et al. (2009). Methodology for the free allocation of emission allowances in the EU ETS post 2012: Sector
Report for the Chemical Industry. Prepared by Ecofys, Fraunhofer Institute for Systems and Innovation Research,
and Oko-lnstitut for the European Commission. November 2009. Available at
https://ec.europa.eu/clima/system/files/2016-ll/bm study-chemicals en.pdf.
ICIS (2006) Chemical Profile - Caprolactam. October 15, 2006. Available online at:
https://www.icis.com/explore/resources/news/2006/10/18/2016832/chemical-profile-caprolactam/.
ICIS (2004) Chemical Profile - Caprolactam. January 5, 2004. Available online at:
https://www.icis.com/explore/resources/news/20Q5/12/02/547244/chemical-profile-caprolactam/.
IPCC (2013) Climate Change 2013: The Physical Science Basis. Contribution of Working Group I to the Fifth
Assessment Report of the Intergovernmental Panel on Climate Change. [Stocker, T.F., D. Qin, G.-K. Plattner, M.
Tignor, S.K. Allen, J. Boschung, A. Nauels, Y. Xia, V. Bex and P.M. Midgley (eds.)]. Cambridge University Press,
Cambridge, United Kingdom and New York, NY, USA, 1535 pp.
IPCC (2007) Climate Change 2007: The Physical Science Basis. Contribution of Working Group I to the Fourth
Assessment Report of the Intergovernmental Panel on Climate Change. [S. Solomon, D. Qin, M. Manning, Z. Chen,
M. Marquis, K.B. Averyt, M. Tignor and H.L Miller (eds.)]. Cambridge University Press. Cambridge, United Kingdom
996 pp.
IPCC (2006) 2006 IPCC Guidelines for National Greenhouse Gas Inventories. The National Greenhouse Gas
Inventories Programme, The Intergovernmental Panel on Climate Change. [H.S. Eggleston, L. Buendia, K. Miwa, T.
Ngara, and K. Tanabe (eds.)]. Hayama, Kanagawa, Japan.
RTI (2023) Expert judgment, RTI International. March 30, 2023.
References and Abbreviations 10-33
-------
Shaw Industries Group, Inc. (Shaw) (2015) "Shaw Carpet Recycling Facility Successfully Processes Nylon and
Polyester". July 13, 2015. Available online at: https://shawinc.com/Newsroom/Press-Releases/Shaw-Carpet-
Recycling-Facility-Successfully-Proces/.
Textile World (2000) "Evergreen Makes Nylon Live Forever". Textile World. October 1, 2000. Available online at:
https://www.textileworld.com/textile-world/textile-news/2000/10/evergreen-makes-nylon-live-forever/.
Carbide Production and Consumption
IPCC (2013) Climate Change 2013: The Physical Science Basis. Contribution of Working Group I to the Fifth
Assessment Report of the Intergovernmental Panel on Climate Change. [Stocker, T.F., D. Qin, G.-K. Plattner, M.
Tignor, S.K. Allen, J. Boschung, A. Nauels, Y. Xia, V. Bex and P.M. Midgley (eds.)]. Cambridge University Press,
Cambridge, United Kingdom and New York, NY, USA, 1535 pp.
IPCC (2007) Climate Change 2007: The Physical Science Basis. Contribution of Working Group I to the Fourth
Assessment Report of the Intergovernmental Panel on Climate Change. [S. Solomon, D. Qin, M. Manning, Z. Chen,
M. Marquis, K.B. Averyt, M. Tignor and H.L Miller (eds.)]. Cambridge University Press. Cambridge, United Kingdom
996 pp.
IPCC (2006) 2006 IPCC Guidelines for National Greenhouse Gas Inventories. The National Greenhouse Gas
Inventories Programme, The Intergovernmental Panel on Climate Change. [H.S. Eggleston, L. Buendia, K. Miwa, T.
Ngara, and K. Tanabe (eds.)]. Hayama, Kanagawa, Japan.
Biscay, Nicolas & Henry, Lucile & Adschiri, Tadafumi & Yoshimura, Masahiro & Aymonier, Cyril. (2021). Behavior of
Silicon Carbide Materials under Dry to Hydrothermal Conditions. Nanomaterials. 11. 1351. doi:
10.3390/nanoll051351.
Environment and Climate Change Canada (ECCC) (2022), Personal Communication between Genevieve Leblanc-
Power, Environment and Climate Change Canada and Mausami Desai and Amanda Chiu, U.S. Environmental
Protection Agency. April 12, 2022.
RTI (2023) Expert judgment, RTI International. March 30, 2023.
United States Census Bureau (1990 through 2022) USITC Trade DataWeb. Available online at:
http://dataweb.usitc.gov/.
USGS (1991a through 2021) Minerals Yearbook: Manufactured Abrasives Annual Report. U.S. Geological Survey,
Reston, VA. Available online at: https://www.usgs.gov/centers/national-minerals-information-
center/manufactured-abrasives-statistics-and-information.
USGS (1991b through 2021) Minerals Yearbook: Silicon Annual Report. U.S. Geological Survey, Reston, VA.
Available online at: http://minerals.usgs.gov/minerals/pubs/commodity/silicon/.
USGS (2023a) 2022 Minerals Yearbook: Manufactured Abrasives (2022 advanced-release tables), September 27,
2023. U.S. Geological Survey, Reston, VA. Available online at: https://www.usgs.gov/centers/national-minerals-
information-center/manufactured-abrasives-statistics-and-information
USGS (2023b) Mineral Commodity Summaries: Abrasives (Manufactured). U.S. Geological Survey, Reston, Va.
January 2023.
USGS (2023c) 2022 Minerals Yearbook: Silicon (2022 advanced-release tables), November 27, 2023. U.S. Geological
Survey, Reston, VA. Available online at: https://www.usgs.gov/centers/national-minerals-information-
center/silicon-statistics-and-information
USGS (2021a) Mineral Commodity Summaries: Abrasives (Manufactured). U.S. Geological Survey, Reston, Va.
January 2021. Available online at: https://pubs.usgs.gov/periodicals/mcs2021/mcs2021-abrasives.pdf.
10-34 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
Washington Mills (2023), North Grafton, MA. Available online at: https://www.washingtonmills.com/silicon-
carbide/sic-industries. Accessed on April 4, 2023.
Titanium Dioxide Production
Gambogi, J. (2002) Telephone communication. Joseph Gambogi, Commodity Specialist, U.S. Geological Survey and
Philip Groth, ICF International. November 2002.
IPCC (2006) 2006IPCC Guidelines for National Greenhouse Gas Inventories. The National Greenhouse Gas
Inventories Programme, Intergovernmental Panel on Climate Change. [H.S. Eggleston, L. Buendia, K. Miwa, T.
Ngara, and K. Tanabe (eds.)]. Hayama, Kanagawa, Japan.
RTI (2023) Expert judgment, RTI International. March 30, 2023.
United States Geological Survey (USGS) (2023a) 2020 Minerals Yearbook: Titanium, 2020 tables-only release, Table
1. U.S. Geological Survey, Reston, Va. March 2023.
USGS (2023b) Mineral Commodity Summaries: Titanium and Titanium Dioxide. U.S. Geological Survey, Reston, Va.
January 2023.
USGS (1991 through 2022) Minerals Yearbook: Titanium. U.S. Geological Survey, Reston, VA.
Soda Ash Production
IPCC (2006) 2006 IPCC Guidelines for National Greenhouse Gas Inventories. The National Greenhouse Gas
Inventories Programme, The Intergovernmental Panel on Climate Change. [H.S. Eggleston, L. Buendia, K. Miwa, T.
Ngara, and K. Tanabe (eds.)]. Hayama, Kanagawa, Japan.
RTI (2023) Expert judgment, RTI International. March 30, 2023.
United States Geological Survey (USGS) (2023a) Mineral Commodity Summary: Soda Ash. U.S. Geological Survey,
Reston, VA. January 2023.
USGS (2023b) Mineral Industry Surveys: Soda Ash in June 2023. U.S. Geological Survey, Reston, VA. August 2023.
United States Geological Survey (USGS) (2022a) Mineral Commodity Summary: Soda Ash. U.S. Geological Survey,
Reston, VA. January 2022.
USGS (2022b) Mineral Industry Surveys: Soda Ash in June 2022. U.S. Geological Survey, Reston, VA. August 2022.
USGS (2021) Mineral Industry Surveys: Soda Ash in April 2021. U.S. Geological Survey, Reston, VA. July 2021.
USGS (2020) Mineral Industry Surveys: Soda Ash in April 2020. U.S. Geological Survey, Reston, VA. July 2020.
USGS (2019) Mineral Industry Surveys: Soda Ash in April 2019. U.S. Geological Survey, Reston, VA. July 2019.
USGS (2018a) Mineral Industry Surveys: Soda Ash in February 2018. U.S. Geological Survey, Reston, VA. Accessed
September 2018.
USGS (2017) Mineral Industry Surveys: Soda Ash in January 2017. U.S. Geological Survey, Reston, VA. March 2017.
USGS (2016) Mineral Industry Surveys: Soda Ash in November 2016. U.S. Geological Survey, Reston, VA. January
2017.
USGS (2015a) Mineral Industry Surveys: Soda Ash in July 2015. U.S. Geological Survey, Reston, VA. September
2015.
USGS (1994 through 2015b, 2018b) Minerals Yearbook: Soda Ash Annual Report. U.S. Geological Survey, Reston,
VA.
References and Abbreviations 10-35
-------
USGS (1995c) Trona Resources in the Green River Basin, Southwest Wyoming. U.S. Department of the Interior, U.S.
Geological Survey. Open-File Report 95-476. Wiig, Stephen, Grundy, W.D., Dyni, John R.
Petrochemical Production
ACC (2023) Business of Chemistry (Annual Data). American Chemistry Council, Arlington, VA.
AN (2014) About Acrylonitrile: Production. AN Group, Washington, D.C.
IPCC (2006) 2006IPCC Guidelines for National Greenhouse Gas Inventories. The National Greenhouse Gas
Inventories Programme, The Intergovernmental Panel on Climate Change. [H.S. Eggleston, L. Buendia, K. Miwa, T.
Ngara, and K. Tanabe (eds.)]. Hayama, Kanagawa, Japan.
Johnson, G. L. (2005 through 2010) Personal communication. Greg Johnson of Liskow & Lewis, on behalf of the
International Carbon Black Association (ICBA) and Caroline Cochran, ICF International. September 2010.
Johnson, G. L. (2003) Personal communication. Greg Johnson of Liskow & Lewis, on behalf of the International
Carbon Black Association (ICBA) and Caren Mintz, ICF International. November 2003.
United States Environmental Protection Agency (EPA) (2023) Greenhouse Gas Reporting Program. Aggregation of
Reported Facility Level Data under Subpart X -National Petrochemical Production for Calendar Years 2010 through
2022. Office of Air and Radiation, Office of Atmospheric Programs, U.S. Environmental Protection Agency,
Washington, D.C.
U.S. EPA (2015) Greenhouse Gas Reporting Program Report Verification. Available online at
https://www.epa.gov/sites/production/files/2015-07/documents/ghgrp verification factsheet.pdf.
U.S. EPA (2008) Technical Support Document for the Petrochemical Production Sector: Proposed Rule for
Mandatory Reporting of Greenhouse Gases. U.S. Environmental Protection Agency. September 2008.
U.S. EPA (2000) Economic Impact Analysis for the Proposed Carbon Black Manufacturing NESHAP, U.S.
Environmental Protection Agency. Research Triangle Park, NC. EPA-452/D-00-003. May 2000.
HCFC-22 Production
ARAP (2010) Electronic mail communication from Dave Stirpe, Executive Director, Alliance for Responsible
Atmospheric Policy to Deborah Ottinger of the U.S. Environmental Protection Agency. September 10, 2010.
ARAP (2009) Electronic mail communication from Dave Stirpe, Executive Director, Alliance for Responsible
Atmospheric Policy to Deborah Ottinger of the U.S. Environmental Protection Agency. September 21, 2009.
ARAP (2008) Electronic mail communication from Dave Stirpe, Executive Director, Alliance for Responsible
Atmospheric Policy to Deborah Ottinger of the U.S. Environmental Protection Agency. October 17, 2008.
ARAP (2007) Electronic mail communication from Dave Stirpe, Executive Director, Alliance for Responsible
Atmospheric Policy to Deborah Ottinger of the U.S. Environmental Protection Agency. October 2, 2007.
ARAP (2006) Electronic mail communication from Dave Stirpe, Executive Director, Alliance for Responsible
Atmospheric Policy to Sally Rand of the U.S. Environmental Protection Agency. July 11, 2006.
ARAP (2005) Electronic mail communication from Dave Stirpe, Executive Director, Alliance for Responsible
Atmospheric Policy to Deborah Ottinger of the U.S. Environmental Protection Agency. August 9, 2005.
ARAP (2004) Electronic mail communication from Dave Stirpe, Executive Director, Alliance for Responsible
Atmospheric Policy to Deborah Ottinger of the U.S. Environmental Protection Agency. June 3, 2004.
ARAP (2003) Electronic mail communication from Dave Stirpe, Executive Director, Alliance for Responsible
Atmospheric Policy to Sally Rand of the U.S. Environmental Protection Agency. August 18, 2003.
10-36 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
ARAP (2002) Electronic mail communication from Dave Stirpe, Executive Director, Alliance for Responsible
Atmospheric Policy to Deborah Ottinger of the U.S. Environmental Protection Agency. August 7, 2002.
ARAP (2001) Electronic mail communication from Dave Stirpe, Executive Director, Alliance for Responsible
Atmospheric Policy to Deborah Ottinger of the U.S. Environmental Protection Agency. August 6, 2001.
ARAP (2000) Electronic mail communication from Dave Stirpe, Executive Director, Alliance for Responsible
Atmospheric Policy to Sally Rand of the U.S. Environmental Protection Agency. August 13, 2000.
ARAP (1999) Facsimile from Dave Stirpe, Executive Director, Alliance for Responsible Atmospheric Policy to
Deborah Ottinger Schaefer of the U.S. Environmental Protection Agency. September 23,1999.
ARAP (1997) Letter from Dave Stirpe, Director, Alliance for Responsible Atmospheric Policy to Elizabeth Dutrow of
the U.S. Environmental Protection Agency. December 23,1997.
EPA (2015) Greenhouse Gas Reporting Program Report Verification. Available online at
https://www.epa.gov/sites/production/files/2015-07/documents/ghgrp verification factsheet.pdf.
IPCC (2007) Climate Change 2007: The Physical Science Basis. Contribution of Working Group I to the Fourth
Assessment Report of the Intergovernmental Panel on Climate Change. [S. Solomon, D. Qin, M. Manning, Z. Chen,
M. Marquis, K.B. Averyt, M. Tignor and H.L Miller (eds.)]. Cambridge University Press. Cambridge, United Kingdom
996 pp.
IPCC (2006) 2006 IPCC Guidelines for National Greenhouse Gas Inventories. The National Greenhouse Gas
Inventories Programme, The Intergovernmental Panel on Climate Change. [H.S. Eggleston, L. Buendia, K. Miwa, T.
Ngara, and K. Tanabe (eds.)]. Hayama, Kanagawa, Japan.
RTI (2008) "Verification of Emission Estimates of HFC-23 from the Production of HCFC-22:Emissions from 1990
through 2006." Report prepared by RTI International for the Climate Change Division. March 2008.
RTI (1997) "Verification of Emission Estimates of HFC-23 from the Production of HCFC-22: Emissions from 1990
through 1996." Report prepared by Research Triangle Institute for the Cadmus Group. November 25,1997; revised
February 16,1998.
UNFCCC (2014) Report of the Conference of the Parties on its nineteenth session, held in Warsaw from 11 to 23
November 2013. United Nations Framework Convention on Climate Change, Warsaw. (FCCC/CP/2013/10/Add.3).
January 31, 2014. Available online at: http://unfccc.int/resource/docs/2013/copl9/eng/10a03.pdf.
Production of Fluorochemicals Other Than HCFC-22
3M (2024) 3M Global EHS Laboratory Response to EPA Data Request on Fluorochemical Emissions. February 2024.
Daikin (2013) Major Source Operating Permit, Daikin America, Alabama Department of Environmental
Management, August 1, 2013. http://lf.adem.alabama.gov/WebLink/DocView.aspx?id=29951882&dbid=0. (p. 11-
1).
Honeywell (2012) Part 70 Operating Permit, Geismar Plant, Honeywell International Inc., Louisiana, Louisiana
Department of Environmental Quality, Page 13, January 28, 2011.
https://edms.deq.louisiana.gov/app/doc/view?doc=7812895.
Honeywell (2011) Part 70 Operating Permit, Baton Rouge Plant Honeywell International Inc., Louisiana Department
of Environmental Quality, Page 25, October 16, 2012.
https://edms.deq.louisiana.gov/app/doc/view?doc=8579001.
ICI Americas (1993) New Permit, KLEA - 134a Plant, ICI Americas, St. Gabriel, Louisiana, Louisiana Department of
Environmental Quality, Page 44, May 28,1993. https://edms.deq.louisiana.gov/app/doc/view?doc= 1309650.
IPCC (2021) Climate Change 2021: The Physical Science Basis. Contribution of Working Group I to the Sixth
Assessment Report of the Intergovernmental Panel on Climate Change. [Masson-Delmotte, V., P. Zhai, A. Pirani, S.L
References and Abbreviations 10-37
-------
Connors, C. Pean, S. Berger, N. Caud, Y. Chen, L. Goldfarb, M.I. Gomis, M. Huang, K. Leitzell, E. Lonnoy, J.B.R.
Matthews, T.K. Maycock, T. Waterfield, O. Yelekgi, R. Yu, and B. Zhou (eds.)]. Cambridge University Press,
Cambridge, United Kingdom and New York, NY, USA, 2391 pp. doi:10.1017/9781009157896. Available from
www.ipcc.ch. The AR6 GWPs are listed in Table 7.SM.7, which appears on page 16 of the Supplementary Material.
IPCC (2019) 2019 Refinement to the 2006IPCC Guidelines for National Greenhouse Gas Inventories, Calvo Buendia,
E., Tanabe, K., Kranjc, A., Baasansuren, J., Fukuda, M., Ngarize S., Osako, A., Pyrozhenko, Y., Shermanau, P. and
Federici, S. (eds). Published: IPCC, Switzerland.
IPCC (2013) Climate Change 2013: The Physical Science Basis. Contribution of Working Group I to the Fifth
Assessment Report of the Intergovernmental Panel on Climate Change [Stocker, T.F., D. Qin, G.-K. Plattner, M.
Tignor, S.K. Allen, J. Boschung, A. Nauels, Y. Xia, V. Bex and P.M. Midgley (eds.)]. Cambridge University Press,
Cambridge, United Kingdom and New York, NY, USA, 1535 pp. The GWPs are listed in Table 8.A.1 of Appendix 8.A:
Lifetimes, Radiative Efficiencies and Metric Values, which appears on pp. 731-737 of Chapter 8, "Anthropogenic
and Natural Radiative Forcing."
IPCC (2007) Climate Change 2007: The Physical Science Basis. Contribution of Working Group I to the Fourth
Assessment Report of the Intergovernmental Panel on Climate Change [Core Writing Team, Pachauri, R.K and
Reisinger, A. (eds.)]. IPCC, Geneva, Switzerland, 104 pp.
McKenna (2022) A 3M Plant in Illinois Was the Country's Worst Emitter of a Climate-Killing 'Immortal' Chemical in
2021, Phil McKenna, Inside Climate News, December 29, 2022. [3M Cordova IL facility.]
https://insideclimatenews.org/news/29122022/3m-cordova-illinois-pfas-cf4-pollution/.
Perkins (1982) Perkins, B. L, Evaluation of Environmental Control Technologies for Commercial Nuclear Fuel
Conversion (UFS) Facilities, LA-9397-MS, October 1982 [030000442.pdf],
Rand (2007) 2004-2006 SF6 Data Summary, Project Memorandum Prepared by D. Knopman and K. Smythe, RAND
Corporation, for the National Electrical Manufacturers Association, June 2007.
SRI Consulting (2004) Chemical Economics Handbook (CEH) Market Research Report: Fluorocarbons, R. Will, A.
Kishi, S. Schlag. SRI Consulting, 2004.
U.S. EPA (2008) Survey of Producers of HFCs, PFCs, SF6 and NF3, 2008. Office of Air and Radiation, Office of
Atmospheric Programs, U.S. Environmental Protection Agency.
U.S. EPA (2023a) GHGRP Data Relevant to the AIM Act, Greenhouse Gas Reporting Program.
https://www.epa.gov/ghgreporting/ghgrp-data-relevant-aim-act. Last accessed 11/16/2023.
U.S. EPA (2023b) Vintaging Model for HFCs. 2023. Office of Atmospheric Programs, U.S. Environmental Protection
Agency.
U.S. EPA (2023c) Estimated layer-weighted substrate production by the semiconductor industry. Office of
Atmospheric Programs, Office of Atmospheric Programs, U.S. Environmental Protection Agency.
U.S. EPA (2023d) SF6 Consumption by Users. 2023. SF6 consumption for 3 industries, Electric Transmission and
Distribution, Semiconductor Manufacturing, and Magnesium Production. Office of Atmospheric Protection, U.S.
Environmental Protection Agency.
U.S. EPA (2015) Greenhouse Gas Reporting Program Report Verification. Available online at
https://www.epa.gov/sites/production/files/2015-07/documents/ghgrp verification factsheet.pdf.
U.S. EPA (1995) Protocol for Equipment Leak Emission Estimates. Office of Air and Radiation, Office of Air Quality
Planning and Standards, U.S. Environmental Protection Agency. Publication No. EPA-453/R-95-017. November
1995.
Vollmer et al. (2019) Vollmer, M. K., Bernard, F., Mitrevski, B., Steele, L. P., Trudinger, C. M., Reimann, S.,
Langenfelds, R. L, Krummel, P. B., Fraser, P. J., Etheridge, D. M., Curran, M. A. J., and Burkholder, J. B.:
Abundances, emissions, and loss processes of the long-lived and potent greenhouse gas octafluorooxolane
10-38 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
(octafluorotetrahydrofuran, C-C4F80) in the atmosphere, Atmos. Chem. Phys., 19, 3481-3492,
https://doi.org/10.5194/acp-19-3481-2019,2019..
Carbon Dioxide Consumption
ARI (1990 through 2010) CO2 Use in Enhanced Oil Recovery. Deliverable to ICF International under Task Order 102,
July 15, 2011.
ARI (2007) CO2-EOR: An Enabling Bridge for the Oil Transition. Presented at "Modeling the Oil Transition—a
DOE/EPA Workshop on the Economic and Environmental Implications of Global Energy Transitions." Washington,
D.C. April 20-21, 2007.
ARI (2006) CO2-EOR: An Enabling Bridge for the Oil Transition. Presented at "Modeling the Oil Transition—a
DOE/EPA Workshop on the Economic and Environmental Implications of Global Energy Transitions." Washington,
D.C. April 20-21, 2006.
Broadhead (2003) Personal communication. Ron Broadhead, Principal Senior Petroleum Geologist and Adjunct
faculty, Earth and Environmental Sciences Department, New Mexico Bureau of Geology and Mineral Resources,
and Robin Petrusak, ICF International. September 5, 2003.
COGCC (2014) Monthly CO2 Produced by County (1999-2009). Available online at:
http://cogcc.state.co.us/COGCCReports/production.aspx?id=MonthlyCQ2ProdByCounty. Accessed October 2014.
Denbury Resources Inc. (2002 through 2010) Annual Report: 2001 through 2009, Form 10-K. Available online at:
http://www.denbury.com/investor-relations/SEC-Filings/SEC-Filings~Details/default.aspx?Filingld=9823015.
Accessed September 2014.
IPCC (2006) 2006IPCC Guidelines for National Greenhouse Gas Inventories. The National Greenhouse Gas
Inventories Programme, The Intergovernmental Panel on Climate Change. [H.S. Eggleston, L. Buendia, K. Miwa, T.
Ngara, and K. Tanabe (eds.)]. Hayama, Kanagawa, Japan.
New Mexico Bureau of Geology and Mineral Resources (2006) Natural Accumulations of Carbon Dioxide in New
Mexico and Adjacent Parts of Colorado and Arizona: Commercial Accumulation of CO2. Available online at:
http://geoinfo.nmt.edU/staff/broadhead/C02.html#commercial.
RTI (2023) Expert judgment, RTI International. March 30, 2023.
U.S. Environmental Protection Agency (EPA) (2023) Greenhouse Gas Reporting Program (GHGRP). Aggregation of
Reported Facility Level Data under Subpart PP -National Level CO2 Transferred for Food & Beverage Applications
for Calendar Years 2010 through 2022. Office of Air and Radiation, Office of Atmospheric Programs, U.S.
Environmental Protection Agency, Washington, D.C.
U.S. EPA (2015) Greenhouse Gas Reporting Program Report Verification. Available online at
https://www.epa.gov/sites/production/files/2015-07/documents/ghgrp verification factsheet.pdf.
Phosphoric Acid Production
EFMA (2000) "Production of Phosphoric Acid." Best Available Techniques for Pollution Prevention and Control in
the European Fertilizer Industry. Booklet 4 of 8. European Fertilizer Manufacturers Association. Available online at:
http://www.efma.org/Publications/BAT%202000/Bat04/sectionQ4.asp.
Florida Institute of Phosphate Research (FIPR) (2003a) "Analyses of Some Phosphate Rocks." Facsimile Gary
Albarelli, Florida Institute of Phosphate Research, Bartow, Florida, to Robert Lanza, ICF International. July 29, 2003.
FIPR (2003b) Florida Institute of Phosphate Research. Personal communication between Michael Lloyd (Laboratory
Manager, FIPR, Bartow, Florida) aand Robert Lanza (ICF International) on August 2003.
References and Abbreviations 10-39
-------
Golder Associates and M3 Engineering, Bayovar 12 Phosphate Project: Nl 43-101 Updated Pre-Feasibility Study,
Issued June 28, 2016. Available at:
https://www.sec.gov/Archives/edgar/data/1471603/000121716016000634/focusiune2016bayovar techrep.htm.
Accessed on October 7, 2020.
IPCC (2006) 2006IPCC Guidelines for National Greenhouse Gas Inventories. The National Greenhouse Gas
Inventories Programme, The Intergovernmental Panel on Climate Change. [H.S. Eggleston, L. Buendia, K. Miwa, T.
Ngara, and K. Tanabe (eds.)]. Hayama, Kanagawa, Japan.
NCDENR (2013) North Carolina Department of Environment and Natural Resources, Title V Air Permit Review for
PCS Phosphate Company, Inc. - Aurora. Available online at:
http://www.ncair.org/permits/permit reviews/PCS rev 08282012.pdf. Accessed on January 25, 2013.
United States Environmental Protection Agency (EPA) (2023) Greenhouse Gas Reporting Program. Review of
Reported Facility Level Data under Subpart Z-Annual Phosphoric Acid Production from Phosphate Rock for
Calendar Year 2022. Office of Air and Radiation, Office of Atmospheric Programs, U.S. Environmental Protection
Agency, Washington, D.C.
United States Geological Survey (USGS) (2023) Mineral Commodity Summaries: Phosphate Rock 2023. January
2023. U.S. Geological Survey, Reston, VA. Available online at: https://www.usgs.gov/centers/nmic/phosphate-
rock-statistics-and-information
USGS (2022) Mineral Commodity Summaries: Phosphate Rock 2022. January 2022. U.S. Geological Survey, Reston,
VA. Available online at: https://www.usgs.gov/centers/nmic/phosphate-rock-statistics-and-information
USGS (2021a) Mineral Commodity Summaries: Phosphate Rock 2021. January 2021. U.S. Geological Survey, Reston,
VA. Available online at: https://www.usgs.gov/centers/nmic/phosphate-rock-statistics-and-information.
USGS (2021b) Personal communication between Stephen Jasinski (USGS) and Amanda Chiu (EPA) on August 25,
2021.
USGS (2020) Mineral Commodity Summaries: Phosphate Rock 2020. January 2020. U.S. Geological Survey, Reston,
VA. Available online at: https://www.usgs.gov/centers/nmic/phosphate-rock-statistics-and-information.
USGS (2019a) Mineral Commodity Summaries: Phosphate Rock 2019. February 2019. U.S. Geological Survey,
Reston, VA. Available online at: https://www.usgs.gov/centers/nmic/phosphate-rock-statistics-and-information.
USGS (2019b) Communication between Stephen Jasinski (USGS) and John Steller (EPA) on November 15, 2019.
USGS (2018) Mineral Commodity Summaries: Phosphate Rock 2018. January 2018. U.S. Geological Survey, Reston,
VA. Available online at: https://www.usgs.gov/centers/nmic/phosphate-rock-statistics-and-information.
USGS (2017) Mineral Commodity Summaries: Phosphate Rock 2017. January 2017. U.S. Geological Survey, Reston,
VA. Available online at: https://www.usgs.gov/centers/nmic/phosphate-rock-statistics-and-information.
USGS (2016) Mineral Commodity Summaries: Phosphate Rock 2016. January 2016. U.S. Geological Survey, Reston,
VA. Available online at: https://www.usgs.gov/centers/nmic/phosphate-rock-statistics-and-information.
USGS (1994 through 2015b) Minerals Yearbook. Phosphate Rock Annual Report. U.S. Geological Survey, Reston, VA.
USGS (2012) Personal communication between Stephen Jasinski (USGS) and Mausami Desai (EPA) on October 12,
2012.
Iron and Steel Production and Metallurgical Coke Production
American Coke and Coal Chemicals Institute (ACCCI) (2021) U.S. Coke Plants as of November 2021, ACCCI,
Washington, D.C. November 2021.
10-40 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
American Iron and Steel Institute (AISI) (2004 through 2023) Annual Statistical Report, American Iron and Steel
Institute, Washington, D.C.
Carroll (2017) Personal communication, Colin P. Carroll, Director of Environment, Health and Safety, American Iron
and Steel Institute and John Steller, U.S. Environmental Protection Agency, November 2017.
Carroll (2016) Personal communication, Colin P. Carroll, Director of Environment, Health and Safety, American Iron
and Steel Institute and Mausami Desai, U.S. Environmental Protection Agency, December 2016.
IPCC (2013) Climate Change 2013: The Physical Science Basis. Contribution of Working Group I to the Fifth
Assessment Report of the Intergovernmental Panel on Climate Change. [Stocker, T.F., D. Qin, G.-K. Plattner, M.
Tignor, S.K. Allen, J. Boschung, A. Nauels, Y. Xia, V. Bex and P.M. Midgley (eds.)]. Cambridge University Press,
Cambridge, United Kingdom and New York, NY, USA, 1535 pp.
IPCC (2007) Climate Change 2007: The Physical Science Basis. Contribution of Working Group I to the Fourth
Assessment Report of the Intergovernmental Panel on Climate Change. [S. Solomon, D. Qin, M. Manning, Z. Chen,
M. Marquis, K.B. Averyt, M. Tignor and H.L Miller (eds.)]. Cambridge University Press. Cambridge, United
Kingdom996 pp.IPCC (2006) 2006 IPCC Guidelines for National Greenhouse Gas Inventories. The National
Greenhouse Gas Inventories Programme, The Intergovernmental Panel on Climate Change. [H.S. Eggleston, L.
Buendia, K. Miwa, T. Ngara, and K. Tanabe (eds.)]. Hayama, Kanagawa, Japan.
IPCC (2006) 2006 IPCC Guidelines for National Greenhouse Gas Inventories. The National Greenhouse Gas
Inventories Programme, The Intergovernmental Panel on Climate Change. [H.S. Eggleston, L. Buendia, K. Miwa, T.
Ngara, and K. Tanabe (eds.)]. Hayama, Kanagawa, Japan.
IPCC/UNEP/OECD/IEA (1995) "Volume 3: Greenhouse Gas Inventory Reference Manual. Table 2-2." IPCC Guidelines
for National Greenhouse Gas Inventories. Intergovernmental Panel on Climate Change, United Nations
Environment Programme, Organization for Economic Co-Operation and Development, International Energy
Agency. IPCC WG1 Technical Support Unit, United Kingdom.
RTI (2024) Expert judgment, RTI International. April 2024.
RTI (2023) Expert judgment, RTI International. March 30, 2023.
Steiner (2008) Personal communication, Bruce Steiner, Technical Consultant with the American Iron and Steel
Institute and Mausami Desai, U.S. Environmental Protection Agency, November 2008.
Tuck (2023a) Personal communication, CrisTuck, Commodity Specialist, U.S. Geological Survey and Amanda Chiu,
U.S. Environmental Protection Agency, January 24, 2023.
Tuck (2023b) Personal communication, Candice Tuck, Commodity Specialist, U.S. Geological Survey and Amanda
Chiu, U.S. Environmental Protection Agency, November 30, 2023.
United States Department of Energy (DOE) (2000) Energy and Environmental Profile of the U.S. Iron and Steel
Industry. Office of Industrial Technologies, U.S. Department of Energy. August 2000. DOE/EE-0229.EIA.
United States Energy Information Administration (EIA) (1998 through 2019) Quarterly Coal Report: October-
December, Energy Information Administration, U.S. Department of Energy, Washington, D.C.
U.S. EIA (2021 through 2023) Quarterly Coal Report: January - March, Energy Information Administration, U.S.
Department of Energy. Washington, D.C.
U.S. EIA (2020) Natural Gas Annual 2019. Energy Information Administration, U.S. Department of Energy.
Washington, D.C. September 2020.
U.S. EIA (2017b) Monthly Energy Review, December 2017, Energy Information Administration, U.S. Department of
Energy, Washington, D.C. DOE/EIA-0035(2015/12).
U.S. EIA (1992) Coal and lignite production. EIA State Energy Data Report 1992, Energy Information Administration,
U.S. Department of Energy, Washington, D.C.
References and Abbreviations 10-41
-------
United States Environmental Protection Agency (EPA) (2023). Greenhouse Gas Reporting Program. Dataset as of
August 18, 2023. Available online at: https://ghgdata.epa.gov/ghgp/.
EPA (2010) Carbon Content Coefficients Developed for EPA's Mandatory Reporting Rule. Office of Air and
Radiation, Office of Atmospheric Programs, U.S. Environmental Protection Agency, Washington, D.C.
United States Geological Survey (USGS) (2023a) 2022 Mineral Commodities Summaries: Iron and Steel. U.S.
Geological Survey, Reston, VA. January 2023.
USGS (2023b) 2021 USGS Minerals Yearbook - Iron and Steel Scrap (tables-only release). U.S. Geological Survey,
Reston, VA.
USGS (2022) 2020 USGS Minerals Yearbook - Iron and Steel Scrap (tables-only release). U.S. Geological Survey,
Reston, VA.
USGS (2021a) 2021 Mineral Commodities Summaries: Iron and Steel. U.S. Geological Survey, Reston, VA. January
2021.
USGS (2021b) 2019 USGS Minerals Yearbook - Iron and Steel Scrap (tables-only release). U.S. Geological Survey,
Reston, VA.
USGS (2020a) 2018 USGS Minerals Yearbook - Iron and Steel Scrap (tables-only release). U.S. Geological Survey,
Reston, VA.
USGS (2020b) 2017 USGS Minerals Yearbook - Iron and Steel. U.S. Geological Survey, Reston, VA.
USGS (1991 through 2020) USGS Minerals Yearbook - Iron and Steel Scrap. U.S. Geological Survey, Reston, VA.
Ferroalloy Production
IPCC (2013) Climate Change 2013: The Physical Science Basis. Contribution of Working Group I to the Fifth
Assessment Report of the Intergovernmental Panel on Climate Change. [Stocker, T.F., D. Qin, G.-K. Plattner, M.
Tignor, S.K. Allen, J. Boschung, A. Nauels, Y. Xia, V. Bex and P.M. Midgley (eds.)]. Cambridge University Press,
Cambridge, United Kingdom and New York, NY, USA, 1535 pp.
IPCC (2007) Climate Change 2007: The Physical Science Basis. Contribution of Working Group I to the Fourth
Assessment Report of the Intergovernmental Panel on Climate Change. [S. Solomon, D. Qin, M. Manning, Z. Chen,
M. Marquis, K.B. Averyt, M. Tignor and H.L Miller (eds.)]. Cambridge University Press. Cambridge, United Kingdom
996 pp.
IPCC (2006) 2006 IPCC Guidelines for National Greenhouse Gas Inventories. The National Greenhouse Gas
Inventories Programme, The Intergovernmental Panel on Climate Change. [H.S. Eggleston, L. Buendia, K. Miwa, T.
Ngara, and K. Tanabe (eds.)]. Hayama, Kanagawa, Japan.
Onder, H., and E.A. Bagdoyan (1993) Everything You've Always Wanted to Know about Petroleum Coke. Allis
Mineral Systems.
RTI (2023) Expert judgment, RTI International. March 30, 2023.
United States Geological Survey (USGS) (2023a) Mineral Industry Survey: Silicon in June 2023. U.S. Geological
Survey, Reston, VA. September 2023.
USGS (2023b) 2022 Mineral Commodity Summaries: Silicon. U.S. Geological Survey, Reston, VA. January 2023.
USGS (2022a) 2021 Minerals Yearbook: Silicon (tables-only release). U.S. Geological Survey, Reston, VA. September
2022.
USGS (2022b) 2020 Minerals Yearbook: Ferroalloys (tables-only release). U.S. Geological Survey, Reston, VA. May
2023.
10-42 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
USGS (2022c) 2021 Mineral Commodity Summaries. U.S. Geological Survey, Reston, VA. January 2022.
USGS (2021) 2020 Mineral Commodity Summaries: Silicon. U.S. Geological Survey, Reston, VA. January 2021.
USGS (2020) 2019 Mineral Commodity Summaries: Silicon. U.S. Geological Survey, Reston, VA. January 2020.
USGS (2013a) 2013 Minerals Yearbook: Chromium. U.S. Geological Survey, Reston, VA. March 2016.
USGS (1996 through 2022) Minerals Yearbook: Silicon. U.S. Geological Survey, Reston, VA.
Aluminum Production
EPA (2023) Greenhouse Gas Reporting Program (GHGRP). Envirofacts, Subpart: F Aluminum Production. Available
online at: https://www.epa.gov/enviro/greenhouse-gas-subpart-f
EPA (2015) Greenhouse Gas Reporting Program Report Verification. Available online at
https://www.epa.gov/sites/production/files/2015-07/documents/ghgrp verification factsheet.pdf.
IPCC (2019) 2019 Refinement to the 2006IPCC Guidelines for National Greenhouse Gas Inventories. The National
Greenhouse Gas Inventories Programme, The Intergovernmental Panel on Climate Change. [Calvo Buendia, E.,
Tanabe, K., Kranjc, A., Baasansuren, J., Fukuda, M., Ngarize, S., Osako, A., Pyrozhenko, Y., Shermanau, P. and
Federici, S. (eds.)]. Switzerland.
IPCC (2006) 2006 IPCC Guidelines for National Greenhouse Gas Inventories. The National Greenhouse Gas
Inventories Programme, The Intergovernmental Panel on Climate Change. [H.S. Eggleston, L. Buendia, K. Miwa, T.
Ngara, and K. Tanabe (eds.)]. Hayama, Kanagawa, Japan.
USGS (2023) Minerals Yearbook: Aluminum Annual Report. U.S. Geological Survey, Reston, VA.
USGS (2022) Mineral Commodity Summaries 2022. U.S. Geological Survey, Reston VA.
USGS (2021) 2020 Mineral Commodity Summaries: Aluminum. U.S. Geological Survey, Reston, VA.
USGS (2021) 2019 Mineral Commodity Summaries: Aluminum. U.S. Geological Survey, Reston, VA.
USGS (2020) Minerals Yearbook: Aluminum Annual Report. U.S. Geological Survey, Reston, VA.
USGS (2020) Mineral Industry Surveys: Aluminum in December 2020. U.S. Geological Survey, Reston VA. December
2020
USGS (2020) 2019 Mineral Commodity Summaries: Aluminum. U.S. Geological Survey, Reston, VA.
USGS (2019) Minerals Yearbook: Aluminum Annual Report. U.S. Geological Survey, Reston, VA.
USGS (2018) Minerals Yearbook: Aluminum Annual Report. U.S. Geological Survey, Reston, VA.
USGS (2017) Minerals Yearbook: Aluminum Annual Report. U.S. Geological Survey, Reston, VA.
USGS (2016) Minerals Yearbook: Aluminum Annual Report. U.S. Geological Survey, Reston, VA.
USGS (2015) Minerals Yearbook: Aluminum Annual Report. U.S. Geological Survey, Reston, VA.
USGS (2014) Minerals Yearbook: Aluminum Annual Report. U.S. Geological Survey, Reston, VA.
USGS (2013) Minerals Yearbook: Aluminum Annual Report. U.S. Geological Survey, Reston, VA.
USGS (2012) Minerals Yearbook: Aluminum Annual Report. U.S. Geological Survey, Reston, VA.
USGS (2011) Minerals Yearbook: Aluminum Annual Report. U.S. Geological Survey, Reston, VA.
USGS (2010) Minerals Yearbook: Aluminum Annual Report. U.S. Geological Survey, Reston, VA.
USGS (2009) Minerals Yearbook: Aluminum Annual Report. U.S. Geological Survey, Reston, VA.
References and Abbreviations 10-43
-------
USGS
2008
Minerals
Yearbook:
Aluminum Annual
Report.
U.S.
Geologica
Survey,
Reston,
VA.
USGS
2007
Minerals
Yearbook:
Aluminum Annual
Report.
U.S.
Geologica
Survey,
Reston,
VA.
USGS
2006
Minerals
Yearbook:
Aluminum Annual
Report.
U.S.
Geologica
Survey,
Reston,
VA.
USGS
2005
Minerals
Yearbook:
Aluminum Annual
Report.
U.S.
Geologica
Survey,
Reston,
VA.
USGS
2004
Minerals
Yearbook:
Aluminum Annual
Report.
U.S.
Geologica
Survey,
Reston,
VA.
USGS
2003
Minerals
Yearbook:
Aluminum Annual
Report.
U.S.
Geologica
Survey,
Reston,
VA.
USGS
2002
Minerals
Yearbook:
Aluminum Annual
Report.
U.S.
Geologica
Survey,
Reston,
VA.
USGS
2001
Minerals
Yearbook:
Aluminum Annual
Report.
U.S.
Geologica
Survey,
Reston,
VA.
USGS
2000
Minerals
Yearbook:
Aluminum Annual
Report.
U.S.
Geologica
Survey,
Reston,
VA.
USGS
1999
Minerals
Yearbook:
Aluminum Annual
Report.
U.S.
Geologica
Survey,
Reston,
VA.
USGS
1998
Minerals
Yearbook:
Aluminum Annual
Report.
U.S.
Geologica
Survey,
Reston,
VA.
USGS
1997
Minerals
Yearbook:
Aluminum Annual
Report.
U.S.
Geologica
Survey,
Reston,
VA.
USGS
1996
Minerals
Yearbook:
Aluminum Annual
Report.
U.S.
Geologica
Survey,
Reston,
VA.
USGS
1995
Minerals
Yearbook:
Aluminum Annual
Report.
U.S.
Geologica
Survey,
Reston,
VA.
USGS
1994
Minerals
Yearbook:
Aluminum Annual
Report.
U.S.
Geologica
Survey,
Reston,
VA.
USGS
1993
Minerals
Yearbook:
Aluminum Annual
Report.
U.S.
Geologica
Survey,
Reston,
VA.
USGS
1992
Minerals
Yearbook:
Aluminum Annual
Report.
U.S.
Geologica
Survey,
Reston,
VA.
USGS
1991
Minerals
Yearbook:
Aluminum Annual
Report.
U.S.
Geologica
Survey,
Reston,
VA.
USGS
1990
Minerals
Yearbook:
Aluminum Annual
Report.
U.S.
Geologica
Survey,
Reston,
VA.
Magnesium Production and Processing
ARB (2015) "Magnesium casters successfully retool for a cleaner future." California Air Resources Board News
Release. Release # 15-07. February 5, 2015. Accessed October 2017. Available online at:
https://www.arb.ca.gov/newsrel/newsrelease.php7ich704.
Bartos S., C. Laush, J. Scharfenberg, and R. Kantamaneni (2007) "Reducing greenhouse gas emissions from
magnesium die casting." Journal of Cleaner Production, 15: 979-987, March.
EPA (2020) Envirofacts. Greenhouse Gas Reporting Program (GHGRP), Subpart T: Magnesium Production and
Processing. Available online at: https://www.epa.gov/enviro/greenhouse-gas-customized-search. Accessed on
October 2020.
EPA (2015) Greenhouse Gas Reporting Program Report Verification. Available online at
https://www.epa.gov/sites/production/files/2015-07/documents/ghgrp verification factsheet.pdf.
Gjestland, H. and D. Magers (1996) "Practical Usage of Sulphur [Sulfur] Hexafluoride for Melt Protection in the
Magnesium Die Casting Industry." #13,1996 Annual Conference Proceedings, International Magnesium
Association. Ube City, Japan.
IPCC (2006) 2006IPCC Guidelines for National Greenhouse Gas Inventories. The National Greenhouse Gas
Inventories Programme, The Intergovernmental Panel on Climate Change. [H.S. Eggleston, L. Buendia, K. Miwa, T.
Ngara, and K. Tanabe (eds.)]. Hayama, Kanagawa, Japan.
Kramer Deborah A. (2000) "Magnesium" U.S. Geological Survey Minerals Yearbook -2000.
10-44 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
RAND (2002) RAND Environmental Science and Policy Center, "Production and Distribution of SF6 by End-Use
Applications" Katie D. Smythe. International Conference on SF6 and the Environment: Emission Reduction
Strategies. San Diego, CA. November 21-22, 2002.
USGS (1995 through 2023) Minerals Yearbook: Magnesium Annual Report. U.S. Geological Survey, Reston, VA.
Available online at: http://minerals.usgs.gOv/minerals/pubs/commoditv/magnesium/index.html#mis.
USGS (2010b) Mineral Commodity Summaries: Magnesium Metal. U.S. Geological Survey, Reston, VA. Available
online at: http://minerals.usgs.gov/minerals/pubs/commodity/magnesium/mcs-2010-mgmet.pdf.
USGS (2005b) Personal Communication between Deborah Kramer of the USGS and Jeremy Scharfenberg of ICF
Consulting.
Lead Production
Dutrizac, J.E., V. Ramachandran, and J.A. Gonzalez (2000) Lead-Zinc 2000. The Minerals, Metals, and Materials
Society.
IPCC (2006) 2006IPCC Guidelines for National Greenhouse Gas Inventories. The National Greenhouse Gas
Inventories Programme, The Intergovernmental Panel on Climate Change. [H.S. Eggleston, L. Buendia, K. Miwa, T.
Ngara, and K. Tanabe (eds.)]. Hayama, Kanagawa, Japan.
Morris, D., F.R. Steward, and P. Evans (1983) Energy Efficiency of a Lead Smelter. Energy 8(5):337-349.
RTI (2023) Expert judgment, RTI International. March 30, 2023.
Sjardin, M. (2003) CO2 Emission Factors for Non-Energy Use in the Non-Ferrous Metal, Ferroalloys and Inorganics
Industry. Copernicus Institute. Utrecht, the Netherlands.
Ullman (1997) Ullman's Encyclopedia of Industrial Chemistry: Fifth Edition. Volume A5. John Wiley and Sons.
United States Geological Survey (USGS) (2023a) 2022 Mineral Commodity Summary, Lead. U.S. Geological Survey,
Reston, VA. January 2023.
USGS (2023b) 2020 Minerals Yearbook, Lead - Advance Data Release. U.S. Geological Survey, Reston, VA. July
2023.
USGS (2023c) 2022 Mineral Industries Surveys: Lead in September 2022. U.S. Geological Survey, Reston, VA.
December 2022.
USGS (2022a) 2019 Minerals Yearbook, Lead - Advance Data Release. U.S. Geological Survey, Reston, VA. October
2022.
USGS (2022b) 2021 Mineral Commodity Summary, Lead. U.S. Geological Survey, Reston, VA. January 2022.
USGS (2021a) 2017 Minerals Yearbook, Lead - Advance Release. U.S. Geological Survey, Reston, VA. July 2021.
USGS (2021b) 2020 Mineral Commodity Summary, Lead. U.S. Geological Survey, Reston, VA. February 2021.
USGS (2020) 2019 Mineral Commodity Summary, Lead. U.S. Geological Survey, Reston, VA. February 2020.
USGS (2019) 2018 Mineral Commodity Summary, Lead. U.S. Geological Survey, Reston, VA. February 2019.
USGS (2018) 2017 Mineral Commodity Summary, Lead. U.S. Geological Survey, Reston, VA. January 2018.
USGS (2017) 2016 Mineral Commodity Summary, Lead. U.S. Geological Survey, Reston, VA. January 2017.
USGS (2016) 2015 Mineral Commodity Summary, Lead. U.S. Geological Survey, Reston, VA. January 2016.
USGS (2015) 2014 Mineral Commodity Summary, Lead. U.S. Geological Survey, Reston, VA. January 2015.
USGS (2014) 2013 Mineral Commodity Summary, Lead. U.S. Geological Survey, Reston, VA. February 2014.
References and Abbreviations 10-45
-------
USGS (1994 through 2013) Minerals Yearbook: Lead Annual Report. U.S. Geological Survey, Reston, VA..
Zinc Production
American Zinc Recycling (AZR) (2021) Summary of Company History. Available online at https://azr.com/our-
history/. Accessed on March 16, 2021.
AZR (2020) Personal communication. Erica Livingston, American Zinc Recycling and Amanda Chiu, U.S.
Environmental Protection Agency. October 29, 2020.
American Zinc Products (AZP) (2021) American Zinc Products Marks First Anniversary of Zinc Production. Available
online at https://americanzincproducts.com/american-zinc-products-marks-first-anniversary-of-zinc-production/.
Accessed on March 1, 2022.
Befesa (2023) Personal communication. Eric Hunsberger, Befesa Zinc US Inc. and Amanda Chiu, U.S. Environmental
Protection Agency. September 19, 2023.
Befesa (2022) Personal communication. Eric Hunsberger, Befesa Zinc US Inc. and Amanda Chiu, U.S. Environmental
Protection Agency. November 8, 2022.
Horsehead Corp. (2016) Form 10-K, Annual Report for the Fiscal Year Ended December 31, 2015. Available online
at: https://www.sec.gov/Archives/edgar/data/1385544/0Q0119312516725704/d236839dl0k.htm. Submitted on
January 25, 2017.
Horsehead Corp. (2015) Form 10-K, Annual Report for the Fiscal Year Ended December 31, 2014. Available online
at: http://www.sec.gov/Archives/edgar/data/1385544/000138554415000Q05/zinc-2014123110k.htm. Submitted
on March 2, 2015.
Horsehead Corp. (2014) Form 10-K, Annual Report for the Fiscal Year Ended December 31, 2013. Available online
at: http://www.sec.gov/Archives/edgar/data/1385544/00Q138554414000Q03/zinc-2013123110k.htm. Submitted
on March 13, 2014.
Horsehead Corp. (2013) Form 10-K, Annual Report for the Fiscal Year Ended December 31, 2012. Available online
at: http://www.sec.gov/Archives/edgar/data/1385544/000119312513110431/0001193125-13-110431-index.htm.
Submitted March 18, 2013.
Horsehead Corp. (2012a) Form 10-K, Annual Report for the Fiscal Year Ended December 31, 2011. Available online
at: http://www.sec.gov/Archives/edgar/data/1385544/00Q119312512107345/d293011dl0k.htm. Submitted on
March 9, 2012.
Horsehead Corp. (2012b) Horsehead's New Zinc Plant and its Impact on the Zinc Oxide Business. February 22, 2012.
Available online at: http://www.horsehead.net/downloadAttachmentNDO.php?ID=118. Accessed on September
10, 2015.
Horsehead Corp. (2011) 10-K Annual Report for the Fiscal Year Ended December 31, 2010. Available online at:
http://google.brand.edgar-online.com/default.aspx?sym=zinc. Submitted on March 16, 2011.
Horsehead Corp. (2010a) 10-K Annual Report for the Fiscal Year Ended December 31, 2009. Available online at:
http://google.brand.edgar-online.com/default.aspx?sym=zinc. Submitted on March 16, 2010.
Horsehead Corp. (2010b) Horsehead Holding Corp. Provides Update on Operations at its Monaco, PA Plant. July 28,
2010. Available online at: http://www.horsehead.net/pressreleases.php?showall=no&news=&ID=65.
Horsehead Corp (2009) 10-K Annual Report for the Fiscal Year Ended December 31, 2008. Available online at:
https://www.sec.gov/Archives/edgar/data/1385544/0000950152090Q2674/l35087aelQvk.htm. Submitted on
March 16, 2009.
Horsehead Corp (2008) 10-K Annual Report for the Fiscal Year Ended December 31, 2007. Available online at:
http://google.brand.edgar-online.com/default.aspx?sym=zinc. Submitted on March 31, 2008.
10-46 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
Horsehead Corp (2007) Registration Statement (General Form) S-l. Available online at http://google.brand.edgar-
online.com/default.aspx?sym=zinc. Submitted on April 13, 2007.
IPCC (2006) 2006IPCC Guidelines for National Greenhouse Gas Inventories. The National Greenhouse Gas
Inventories Programme, The Intergovernmental Panel on Climate Change. [H.S. Eggleston, L. Buendia, K. Miwa, T.
Ngara, and K. Tanabe (eds.)]. Hayama, Kanagawa, Japan.
Nyrstar (2017) 2016 Clarksville Fact Sheet. Available online at:
http://www.nvrstar.eom/~/media/Files/N/Nyrstar/operations/melting/fact-sheet-clarksville-en.pdf. Accessed on
September 27, 2017.
PIZO (2021) Personal communication. Thomas Rheaume, Arkansas Department of Energy and Environment and
Amanda Chiu, U.S. Environmental Protection Agency. February 16, 2021.
PIZO (2012) Available online at http://pizotech.com/index.html. Accessed on October 10, 2012.
Recycling Today (2020) "AZR to restart for zinc recycling plant in North Carolina." March 6, 2020.
https://www.recvclingtodav.com/article/american-zinc-recvcling-restarting-north-carolina-plant-2020/. Accessed
October 10, 2020.
Recycling Today (2017) "Horsehead announces corporate name change to American Zinc Recycling." May 3, 2017.
https://www.recyclingtodav.com/article/horsehead-changes-name-american-zinc-recycling/. Accessed September
19, 2022.
Steel Dust Recycling (SDR) (2023) Personal communication. Jeremy Whitten, Steel Dust Recycling LLC and Amanda
Chiu, U.S. Environmental Protection Agency. September 20 and 25, 2023.
SDR (2022) Personal communication. Jeremy Whitten, Steel Dust Recycling LLC and Amanda Chiu, U.S.
Environmental Protection Agency. October 10, 2022.
SDR (2021) Personal communication. Jeremy Whitten, Steel Dust Recycling LLC and Amanda Chiu, U.S.
Environmental Protection Agency. January 8, 2021.
SDR (2018) Personal communication. Jeremy Whitten, Steel Dust Recycling LLC and John Steller, U.S.
Environmental Protection Agency. October 25, 2018.
SDR (2017) Personal communication. Jeremy Whitten, Steel Dust Recycling LLC and John Steller, U.S.
Environmental Protection Agency. January 26, 2017.
SDR (2015) Personal communication. Jeremy Whitten, Steel Dust Recycling LLC and Gopi Manne, Eastern Research
Group, Inc. September 22, 2015.
SDR (2014) Personal communication. Art Rowland, Steel Dust Recycling LLC and Gopi Manne, Eastern Research
Group, Inc. December 9, 2014.
SDR (2013) Available online at http://steeldust.com/home.htm. Accessed on October 29, 2013.
SDR (2012) Personal communication. Art Rowland, Steel Dust Recycling LLC and Gopi Manne, Eastern Research
Group, Inc. October 5, 2012.
Sjardin (2003) CO2 Emission Factors for Non-Energy Use in the Non-Ferrous Metal, Ferroalloys and Inorganics
Industry. Copernicus Institute. Utrecht, the Netherlands.
United States Environmental Protection Agency (EPA) (1992) "Applications Analysis Report: Horsehead Resource
Development Company Inc., Flame Reactor Technology" EPA/540/A5-91/005. May 1992.
United States Geological Survey (USGS) (2023) 2023 Mineral Commodity Summary: Zinc. U.S. Geological Survey,
Reston, VA. January 2023. Available online at: https://pubs.usgs.gov/periodicals/mcs2023/mcs2023-zinc.pdf
USGS (2022) 2022 Mineral Commodity Summary: Zinc. U.S. Geological Survey, Reston, VA. January 2022.
USGS (2021) 2021 Mineral Commodity Summary: Zinc. U.S. Geological Survey, Reston, VA. January 2021.
References and Abbreviations 10-47
-------
USGS (2020) 2020 Mineral Commodity Summary: Zinc. U.S. Geological Survey, Reston, VA. January 2020.
USGS (2019) 2019 Mineral Commodity Summary: Zinc. U.S. Geological Survey, Reston, VA. January 2019.
USGS (2018) 2018 Mineral Commodity Summary: Zinc. U.S. Geological Survey, Reston, VA. January 2018.
USGS (2017) 2017 Mineral Commodity Summary: Zinc. U.S. Geological Survey, Reston, VA. January 2017.
USGS (2016) 2016 Mineral Commodity Summary: Zinc. U.S. Geological Survey, Reston, VA. January 2016.
USGS (2015) 2015 Mineral Commodity Summary: Zinc. U.S. Geological Survey, Reston, VA. January 2015.
USGS (1994 through 2014) Minerals Yearbook: Zinc Annual Report. U.S. Geological Survey, Reston, VA.
Viklund-White (2000) The use ofLCAforthe environmental evaluation of the recycling of galvanized steel. ISIJ
International, Vol. 40. No. 3, pp 292-299.
Electronics Industry
Burton, C.S., and R. Beizaie (2001) "EPA's PFC Emissions Model (PEVM) v. 2.14: Description and Documentation"
prepared for Office of Global Programs, U. S. Environmental Protection Agency, Washington, DC. November 2001.
Citigroup Smith Barney (2005) Global Supply/Demand Model for Semiconductors. March 2005.
DisplaySearch (2010) DisplaySearch Q4'09 Quarterly FPD Supply/Demand and Capital Spending Report.
DisplaySearch, LLC.
Doering, R. and Nishi, Y (2000) "Handbook of Semiconductor Manufacturing Technology", Marcel Dekker, New
York, USA, 2000.
EPA (2023) Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2021. U.S. Environmental Protection
Agency, Washington, DC. EPA-430-R-23-002.
EPA (2010) Technical Support Document for Process Emissions from Electronics Manufacture (e.g., Micro-Electro-
Mechanical Systems, Liquid Crystal Displays, Photovoltaics, and Semiconductors). U.S. Environmental Protection
Agency, Washington, DC.
EPA (2006) Uses and Emissions of Liquid PFC Heat Transfer Fluids from the Electronics Sector. U.S. Environmental
Protection Agency, Washington, DC. EPA-430-R-06-901.
EPA Greenhouse Gas Reporting Program (GHGRP) Envirofacts. Subpart I: Electronics Manufacture. Available online
at: https://enviro.epa.gov/facts/ghg/search.html.
IPCC (2019) 2019 Refinement to the 2006IPCC Guidelines for National Greenhouse Gas Inventories. The National
Greenhouse Gas Inventories Programme, The Intergovernmental Panel on Climate Change. Calvo Buendia, E.,
Tanabe, K., Kranjc, A., Baasansuren, J., Fukuda, M., Ngarize, S., Osako, A., Pyrozhenko, Y., Shermanau, P. and
Federici, S. (eds). Published: IPCC, Switzerland.
IPCC (2013) Climate Change 2013: The Physical Science Basis. Contribution of Working Group I to the Fifth
Assessment Report of the Intergovernmental Panel on Climate Change [Stocker, T.F., D. Qin, G.-K. Plattner, M.
Tignor, S.K. Allen, J. Boschung, A. Nauels, Y. Xia, V. Bex and P.M. Midgley (eds.)]. Cambridge University Press,
Cambridge, United Kingdom and New York, NY, USA, 1535 pp.
IPCC (2006) 2006 IPCC Guidelines for National Greenhouse Gas Inventories. The National Greenhouse Gas
Inventories Programme, The Intergovernmental Panel on Climate Change. [H.S. Eggleston, L. Buendia, K. Miwa, T.
Ngara, and K. Tanabe (eds.)]. Hayama, Kanagawa, Japan.
ITRS (2007, 2008, 2011, 2013) International Technology Roadmap for Semiconductors: 2006 Update, January 2007;
International Technology Roadmap for Semiconductors: 2007 Edition, January 2008; International Technology
Roadmap for Semiconductors: 2011, January 2012; Update, International Technology Roadmap for
10-48 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
Semiconductors: 2013 Edition, Available online at: https://www.semiconductors.org/resources/20Q7-intemational-
technology-roadmap-for-semiconductors-itrs/. These and earlier editions and updates are available online at:
https://www.semiconductors.org/resources/7fwp resource types=utilization-reports&fwp paged=2. Information
about the number of interconnect layers for years 1990-2010 is contained in Burton and Beizaie, 2001. PEVM is
updated using new editions and updates of the ITRS, which are published annually.
Platzer, Michaela D. (2015) U.S. Solar Photovoltaic Manufacturing: Industry Trends, Global Competition, Federal
Support. Congressional Research Service. January 27, 2015. https://fas.org/sgp/crs/misc/R42509.pdf.
SEMI - Semiconductor Equipment and Materials Industry (2021) World Fab Forecast, June 2021 Edition.
SEMI - Semiconductor Equipment and Materials Industry (2018) World Fab Forecast, June 2018 Edition.
SEMI - Semiconductor Equipment and Materials Industry (2017) World Fab Forecast, August 2018 Edition.
SEMI - Semiconductor Equipment and Materials Industry (2016) World Fab Forecast, May 2017 Edition.
SEMI - Semiconductor Equipment and Materials Industry (2013) World Fab Forecast, May 2013 Edition.
SEMI - Semiconductor Equipment and Materials Industry (2012) World Fab Forecast, August 2012 Edition.
Semiconductor Industry Association (SIA) (2009-2011) STATS: SICAS Capacity and Utilization Rates Q1-Q4 2008, Ql-
Q4 2009, Q1-Q4 2010. Available online at:
http://www.semiconductors.org/industry statistics/semiconductor capacity utilization sicas reports/.
United States Census Bureau (USCB) (2011, 2012, 2014, 2015, 2016, 2017, 2018, 2019, 2020, 2021, 2022) Historical
Data: Quarterly Survey of Plant Capacity Utilization. Available online at: https://www.census.gov/programs-
surveys/qpc.html.
VLSI Research, Inc. (2012) Worldwide Silicon Demand. August 2012.
Substitution of Ozone Depleting Substances
EPA (2023a) Summary of Updates to the Unitary Air-conditioning End-uses in the Vintaging Model. Prepared for
U.S. EPA's Stratospheric Protection Division by ICF under EPA Contract Number 68HERH19D0029.
EPA (2023b) Summary of Updates to the Window Units End-use in the Vintaging Model. Prepared for U.S. EPA's
Stratospheric Protection Division by ICF under EPA Contract Number 68HERH19D0029.
EPA (2023c) Proposed Addition of Small and Large Ductless Mini-Split and Multi-Split Air Conditioning End-uses to
U.S. EPA's Vintaging Model. Prepared for U.S. EPA's Stratospheric Protection Division by ICF under EPA Contract
Number 68HERH19D0029.
EPA (2023d) Proposed Updates to the Streaming Agent End-uses in U.S. EPA's Vintaging Model. Prepared for U.S.
EPA's Stratospheric Protection Division by ICF under EPA Contract Number 68HERH19D0029.
EPA (2018) EPA's Vintaging Model of ODS Substitutes: A Summary of the 2017 Peer Review. Office of Air and
Radiation. Document Number EPA-400-F-18-001. Available online at:
https://www.epa.gov/sites/production/files/2018-09/documents/epas-vintaging-model-of-ods-substitutes-peer-
review-factsheet.pdf.
IPCC (2013) Climate Change 2013: The Physical Science Basis. Contribution of Working Group I to the Fifth
Assessment Report of the Intergovernmental Panel on Climate Change [Stocker, T.F., D. Qin, G.-K. Plattner, M.
Tignor, S.K. Allen, J. Boschung, A. Nauels, Y. Xia, V. Bex and P.M. Midgley (eds.)]. Cambridge University Press,
Cambridge, United Kingdom and New York, NY, USA, 1535 pp.
IPCC (2007) Climate Change 2007: The Physical Science Basis. Contribution of Working Group I to the Fourth
Assessment Report of the Intergovernmental Panel on Climate Change. [S. Solomon, D. Qin, M. Manning, Z. Chen,
References and Abbreviations 10-49
-------
M. Marquis, K.B. Averyt, M. Tignor and H.L Miller (eds.)]. Cambridge University Press. Cambridge, United Kingdom
996 pp.
IPCC (2006) 2006IPCC Guidelines for National Greenhouse Gas Inventories. The National Greenhouse Gas
Inventories Programme, The Intergovernmental Panel on Climate Change. [H.S. Eggleston, L. Buendia, K. Miwa, T.
Ngara, and K. Tanabe (eds.)]. Hayama, Kanagawa, Japan.
Electrical Equipment
CARB (2023). California Greenhouse Gas Emission Inventory - 2023 Edition. Accessed March 2024. Available online
at: https://ww2.arb.ca.gov/ghg-inventory-data.
EPA (2022) Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2020. EPA 430-R-22-003. Available online
at: https://www.epa.gov/ghgemissions/inventorv-us-greenhouse-gas-emissions-and-sinks-1990-2020.
Harnisch and Eisenhauer, "Natural CF4 and SF6 on Earth," GEOPHYSICAL RESEARCH LETTERS, VOL. 25, N0.13,
PAGES 2401-2404, JULY 1,1998. https://agupubs.onlinelibrarv.wiley.com/doi/pdf/10.1029/98GL01779.
HIFLD (2022) Federal Energy Regulatory Commission. Homeland Infrastructure Foundation-Level Data (HIFLD).
2021. Accessed September 2023. Available online at: https://hifld-
geoplatform.opendata.arcgis.com/datasets/geoplatform::transmission-lines/explore?location=38.924381%2C-
122.290494%2C4.78^
HIFLD (2021) Federal Energy Regulatory Commission. Homeland Infrastructure Foundation-Level Data (HIFLD).
2021. Accessed September 2022. Available online at: https://hifld-
geoplatform.opendata.arcgis.com/datasets/electric-power-transmission-lines.
HIFLD (2020) Federal Energy Regulatory Commission. Homeland Infrastructure Foundation-Level Data (HIFLD).
2020. Accessed October 2021. Available online at: https://hifld-
geoplatform.opendata.arcgis.com/datasets/electric-power-transmission-lines/explore?showTable=true.
HIFLD (2019) Federal Energy Regulatory Commission. Homeland Infrastructure Foundation-Level Data (HIFLD).
2019. Accessed March 2021. Available online at: https://hifld-geoplatform.opendata.arcgis.com/datasets/electric-
power-transmission-lines.
Hu, L, Ottinger, D., Bogle, S., Montzka, S., DeCola, P., Dlugokencky, E., Andrews, A.,Thoning, K., Sweeney, C.,
Dutton, G., Aepli, L., and Crotwell, A. (2022) "Declining, seasonal-varying emissions of sulfur hexafluoride from the
United States point to a new mitigation opportunity." EGUsphere [preprint]. Available online at:
https://doi.org/10.5194/egusphere-2022-862.
IPCC (2013) Climate Change 2013: The Physical Science Basis. Contribution of Working Group I to the Fifth
Assessment Report of the Intergovernmental Panel on Climate Change [Stocker, T.F., D. Qin, G.-K. Plattner, M.
Tignor, S.K. Allen, J. Boschung, A. Nauels, Y. Xia, V. Bex and P.M. Midgley (eds.)]. Cambridge University Press,
Cambridge, United Kingdom and New York, NY, USA, 1535 pp.
IPCC (2007) Climate Change 2007: The Physical Science Basis. Contribution of Working Group I to the Fourth
Assessment Report of the Intergovernmental Panel on Climate Change. S. Solomon, D. Qin, M. Manning, Z. Chen,
M. Marquis, K.B. Averyt, M. Tignor and H.L. Miller (eds.). Cambridge University Press. Cambridge, United Kingdom
996 pp.
IPCC (2006) 2006 IPCC Guidelines for National Greenhouse Gas Inventories. The National Greenhouse Gas
Inventories Programme, The Intergovernmental Panel on Climate Change. H.S. Eggleston, L. Buendia, K. Miwa, T.
Ngara, and K. Tanabe (eds.). Hayama, Kanagawa, Japan.
IPCC (1996) Climate Change 1995: The Science of Climate Change. Intergovernmental Panel on Climate Change, J.T.
Houghton, L.G. Meira Filho, B.A. Callander, N. Harris, A. Kattenberg, and K. Maskell (eds.). Cambridge University
Press. Cambridge, United Kingdom.
10-50 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
Levin et al. (2010) "The Global SF6Source Inferred from Long-term High Precision Atmospheric Measurements and
its Comparison with Emission Inventories." Atmospheric Chemistry and Physics, 10: 2655-2662.
Middleton, B. (2000) Cold Weather Applications of Gas Mixture (SF6/N2, SF6/CF4) Circuit Breakers: A User Utility's
Perspective [Conference Presentation], The U.S. Environmental Protection Agency's Conference on SF6 and the
Environment: Emission Reduction Strategies, San Diego, CA, United States. Available online at:
https://www.epa.gov/sites/default/files/2016-02/documents/conf00 roiddletort.pdf
O'Connell, P., F. Heil, J. Henriot, G. Mauthe, H. Morrison, L. Neimeyer, M. Pittroff, R. Probst, J.P. Tailebois (2002)
SFs in the Electric Industry, Status 2000, CIGRE. February 2002.
Ottinger D, Averyt, M. & Harris, D. (2014). Trends in emissions offluorinated GHGs reported under the Greenhouse
Gas Reporting Program: Patterns and potential causes. Submitted to the Seventh International Symposium on
Non-CC>2 Greenhouse Gases (NCGG-7), Amsterdam, Netherlands.
RAND (2004) 'Trends in SF6 Sales and End-Use Applications: 1961-2003," Katie D. Smythe. International Conference
on SFe and the Environment: Emission Reduction Strategies. RAND Environmental Science and Policy Center,
Scottsdale, AZ. December 1-3, 2004.
UDI (2017) 2017 UDI Directory of Electric Power Producers and Distributors, 125th Edition, Platts.
UDI (2013) 2013 UDI Directory of Electric Power Producers and Distributors,121st Edition, Platts.
UDI (2010) 2010 UDI Directory of Electric Power Producers and Distributors, 118th Edition, Platts.
UDI (2007) 2007 UDI Directory of Electric Power Producers and Distributors, 115th Edition, Platts.
UDI (2004) 2004 UDI Directory of Electric Power Producers and Distributors, 112th Edition, Platts.
UDI (2001) 2001 UDI Directory of Electric Power Producers and Distributors, 109th Edition, Platts.
UNFCCC (2014) Report of the Conference of the Parties on its nineteenth session, held in Warsaw from 11 to 23
November 2013. United Nations Framework Convention on Climate Change, Warsaw. (FCCC/CP/2013/10/Add.3).
January 31, 2014. Available online at: http://unfccc.int/resource/docs/2013/copl9/eng/10a03.pdf.
SF6 and PFCs from Other Product Use
IPCC (2019) 2019 Refinement to the 2006IPCC Guidelines for National Greenhouse Gas Inventories. The National
Greenhouse Gas Inventories Programme, The Intergovernmental Panel on Climate Change. Calvo Buendia, E.,
Tanabe, K., Kranjc, A., Baasansuren, J., Fukuda, M., Ngarize, S., Osako, A., Pyrozhenko, Y., Shermanau, P. and
Federici, S. (eds). Published: IPCC, Switzerland.
IPCC (2006) 2006 IPCC Guidelines for National Greenhouse Gas Inventories. The National Greenhouse Gas
Inventories Programme, The Intergovernmental Panel on Climate Change. [H.S. Eggleston, L. Buendia, K. Miwa, T.
Ngara, and K. Tanabe (eds.)]. Hayama, Kanagawa, Japan.
Workman, R.L. et al. Particle Data Group). Prog. Theor. Exp. Phys. 2022, 083C01. Available online at:
https://pdg.lbl.gov/2022/reviews/rpp2022-rev-particle-detectors-accel.pdf
U.S. Department of Energy (DOE). 2022. Federal Energy Management Program (FEMP). Available online at:
https://www.energv.gov/femp/federal-comprehensive-annual-energv-reporting-requirements.
Nitrous Oxide from Product Use
CGA (2003) "CGA Nitrous Oxide Abuse Hotline: CGA/NWSA Nitrous Oxide Fact Sheet." Compressed Gas
Association. November 3, 2003.
CGA (2002) "CGA/NWSA Nitrous Oxide Fact Sheet." Compressed Gas Association. March 25, 2002.
References and Abbreviations 10-51
-------
Heydorn, B. (1997) "Nitrous Oxide—North America." Chemical Economics Handbook, SRI Consulting. May 1997.
IPCC (2013) Climate Change 2013: The Physical Science Basis. Contribution of Working Group I to the Fifth
Assessment Report of the Intergovernmental Panel on Climate Change. [Stocker, T.F., D. Qin, G.K. Plattner, M.
Tignor, S.K. Allen, J. Boschung, A. Nauels, Y. Xia, V. Bex and P.M. Midgley (eds.)]. Cambridge University Press,
Cambridge, United Kingdom and New York, NY, USA, 1535 pp.
IPCC (2007) Climate Change 2007: The Physical Science Basis. Contribution of Working Group I to the Fourth
Assessment Report of the Intergovernmental Panel on Climate Change. [S. Solomon, D. Qin, M. Manning, Z. Chen,
M. Marquis, K.B. Averyt, M. Tignor and H.L Miller (eds.)]. Cambridge University Press. Cambridge, United Kingdom
996 pp.
IPCC (2006) 2006 IPCC Guidelines for National Greenhouse Gas Inventories. The National Greenhouse Gas
Inventories Programme, The Intergovernmental Panel on Climate Change. [H.S. Eggleston, L. Buendia, K. Miwa, T.
Ngara, and K. Tanabe (eds.)]. Hayama, Kanagawa, Japan.
Ottinger (2021) Personal communication. Deborah Ottinger, U.S. Environmental Protection Agency and Amanda
Chiu, U.S. Environmental Protection Agency. January 7, 2021.
RTI (2023) Expert judgment, RTI International. March 30, 2023.
Tupman, M. (2002) Personal communication. Martin Tupman, Airgas Nitrous Oxide and Laxmi Palreddy, ICF
International. July 3, 2002.
Industrial Processes and Product Use Sources of Precursor
Greenhouse Gases
EPA (2023a) "Criteria pollutants National Tier 1 for 1970 - 2023." National Emissions Inventory (NEI) Air Pollutant
Emissions Trends Data. Office of Air Quality Planning and Standards, March 2024. Available online at:
https://www.epa.gov/air-emissions-inventories/air-pollutant-emissions-trends-data.
EPA (2023b) EPA's Emissions Inventory System (EIS) to National Inventory Report (NIR) Mapping file
EIS_NIR_mapping.xlsx. U.S. Environmental Protection Agency. Washington, D.C."
EPA (2023c) "2020 National Emissions Inventory Technical Support Document: Introduction." Office of Air Quality
Planning and Standards, March 2023. Available online at: https://www.epa.gov/system/files/documents/2023-
01/NE12020 TSD Sectionl lntroduction.pdf.
Agriculture
Enteric Fermentation
Archibeque, S. (2011) Personal Communication. Shawn Archibeque, Colorado State University, Fort Collins,
Colorado and staff at ICF International.
Crutzen, P.J., I. Aselmann, and W. Seiler (1986) Methane Production by Domestic Animals, Wild Ruminants, Other
Herbivores, Fauna, and Humans. Tellus, 38B:271-284.
Donovan, K. (1999) Personal Communication. Kacey Donovan, University of California at Davis and staff at ICF
International.
Donovan, K. and L Baldwin. (1999) "Results of the AAMOLLY Model Runs for the Enteric Fermentation Model";
University of California, Davis, 1999.
10-52 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
Doren, P.E., J. F. Baker, C. R. Long and T. C. Cartwright (1989) Estimating Parameters of Growth Curves of Bulls, J
Animal Science 67:1432-1445.
Enns, M. (2008) Personal Communication. Dr. Mark Enns, Colorado State University and staff at ICF International.
EPA (2002) Quality Assurance/Quality Control and Uncertainty Management Plan for the U.S. Greenhouse Gas
Inventory: Procedures Manual for Quality Assurance/Quality Control and Uncertainty Analysis, U.S. Greenhouse
Gas Inventory Program, U.S. Environmental Protection Agency, Office of Atmospheric Programs, EPA430-R-02-
007B, June 2002.
ERG (2021) Updated Other Animal Population Distribution Methodology. ERG, Lexington, MA.
ERG (2016) Development of Methane Conversion Rate Scaling Factor and Diet-Related Inputs to the Cattle Enteric
Fermentation Model for Dairy Cows, Dairy Heifers, and Feedlot Animals. ERG, Lexington, MA. December 2016.
Galyean and Gleghorn (2001) Summary of the 2000 Texas Tech University Consulting Nutritionist Survey. Texas
Tech University. Available online at http://www.depts.ttu.edu/afs/bumett center/progress reports/bcl2.pdf.
June 2009.
Holstein Association (2010) History of the Holstein Breed (website). Available online at:
http://www.holsteinusa.com/holstein breed/breed history, htm I. Accessed September 2010.
ICF (2006) Cattle Enteric Fermentation Model: Model Documentation. Prepared by ICF International for the
Environmental Protection Agency. June 2006.
ICF (2003) Uncertainty Analysis of 2001 Inventory Estimates of Methane Emissions from Livestock Enteric
Fermentation in the U.S. Memorandum from ICF International to the Environmental Protection Agency. May 2003.
IPCC (2019) 2019 Refinement to the 2006IPCC Guidelines for National Greenhouse Gas Inventories. The National
Greenhouse Gas Inventories Programme, The Intergovernmental Panel on Climate Change. [CalvoBuendia, E.,
Tanabe, K., Kranjc, A., Baasansuren, J., Fukuda, M., Ngarize S., Osako, A., Pyrozhenko, Y., Shermanau, P. and
Federici, S. (eds)]. Switzerland.
IPCC (2013) Climate Change 2013: The Physical Science Basis. Contribution of Working Group I to the Fifth
Assessment Report of the Intergovernmental Panel on Climate Change [Stocker, T.F., D. Qin, G.-K. Plattner, M.
Tignor, S.K. Allen, J. Boschung, A. Nauels, Y. Xia, V. Bex and P.M. Midgley (eds.)]. Cambridge University Press,
Cambridge, United Kingdom and New York, NY, USA, 1535 pp.
IPCC (2007) Climate Change 2007: The Physical Science Basis. Contribution of Working Group I to the Fourth
Assessment Report of the Intergovernmental Panel on Climate Change. S. Solomon, D. Qin, M. Manning, Z. Chen,
M. Marquis, K.B. Averyt, M. Tignor and H.L Miller (eds.). Cambridge University Press. Cambridge, United Kingdom
996 pp.
IPCC (2006) 2006 IPCC Guidelines for National Greenhouse Gas Inventories. The National Greenhouse Gas
Inventories Programme, The Intergovernmental Panel on Climate Change. H.S. Eggleston, L. Buendia, K. Miwa, T.
Ngara, and K. Tanabe (eds.). Hayama, Kanagawa, Japan.
Johnson, D. (2002) Personal Communication. Don Johnson, Colorado State University, Fort Collins, and ICF
International.
Johnson, D. (1999) Personal Communication. Don Johnson, Colorado State University, Fort Collins, and David
Conneely, ICF International.
Kebreab E., K. A. Johnson, S. L. Archibeque, D. Pape, and T. Wirth (2008) Model for estimating enteric methane
emissions from United States dairy and feedlot cattle. J. Anim. Sci. 86: 2738-2748.
Lippke, H., T. D. Forbes, and W. C. Ellis. (2000) Effect of supplements on growth and forage intake by stocker steers
grazing wheat pasture. J. Anim. Sci. 78:1625-1635.
References and Abbreviations 10-53
-------
National Bison Association (1999) Total Bison Population—1999. Report provided during personal email
communication with Dave Carter, Executive Director, National Bison Association, July 19, 2011.
Pinchak, W.E., D. R. Tolleson, M. McCloy, L. J. Hunt, R. J. Gill, R. J. Ansley, and S. J. Bevers (2004) Morbidity effects
on productivity and profitability of stocker cattle grazing in the southern plains. J. Anim. Sci. 82:2773-2779.
Platter, W. J., J. D. Tatum, K. E. Belk, J. A. Scanga, and G. C. Smith (2003) Effects of repetitive use of hormonal
implants on beef carcass quality, tenderness, and consumer ratings of beef palatability. J. Anim. Sci. 81:984-996.
Preston, R.L (2010) What's The Feed Composition Value of That Cattle Feed? Beef Magazine, March 1, 2010.
Available at: http://beefmagazine.com/nutrition/feed-composition-tables/feed-composition-value-cattle-0301.
Skogerboe, T. L, L Thompson, J. M. Cunningham, A. C. Brake, V. K. Karle (2000) The effectiveness of a single dose
of doramectin pour-on in the control of gastrointestinal nematodes in yearling stocker cattle. Vet. Parasitology
87:173-181.
Soliva, C.R. (2006) Report to the attention of IPCC about the data set and calculation method used to estimate
methane formation from enteric fermentation of agricultural livestock population and manure management in
Swiss agriculture. On behalf of the Federal Office for the Environment (FOEN), Berne, Switzerland.
U.S. Department of Agriculture (USDA) (2023) Quick Stats: Agricultural Statistics Database. National Agriculture
Statistics Service, U.S. Department of Agriculture. Washington, D.C. Available online at
http://quickstats.nass.usda.gov/. Accessed May-June 2023.
USDA (2022) Economic Research Service "Supply and allocation of milk fat and skim solids by product". Available
online at: https://www.ers.usda.gov/data-products/dairy-data/. Accessed May 2023.
USDA (2021) Economic Research Service Dairy Data. Available online at: https://www.ers.usda.gov/data-
products/dairy-data/. Accessed May 2021.
USDA (2019) 1987,1992,1997, 2002, 2007, 2012, and 2017 Census of Agriculture. National Agriculture Statistics
Service, U.S. Department of Agriculture. Washington, D.C. Available online at:
https://www.nass.usda.gov/AgCensus/index.php. May 2019.
USDA (1996) Beef Cow/Calf Health and Productivity Audit (CHAPA): Forage Analyses from Cow/Calf Herds in 18
States. National Agriculture Statistics Service, U.S. Department of Agriculture. Washington, D.C. Available online at
http://www.aphis.usda.gov/vs/ceah/cahm. March 1996.
USDA:APHIS:VS (2010) Beef 2007-08, Part V: Reference of Beef Cow-calf Management Practices in the United
States, 2007-08. USDA-APHIS-VS, CEAH. Fort Collins, CO.
USDA:APHIS:VS (2002) Reference of 2002 Dairy Management Practices. USDA-APHIS-VS, CEAH. Fort Collins, CO.
Available online at http://www.aphis.usda.gov/vs/ceah/cahm.
USDA:APHIS:VS (1998) Beef'97, Parts l-IV. USDA-APHIS-VS, CEAH. Fort Collins, CO. Available online at
http://www.aphis.usda.gov/animal health/nahms/beefcowcalf/index.shtml#beef97.
USDA:APHIS:VS (1996) Reference of 1996 Dairy Management Practices. USDA-APHIS-VS, CEAH. Fort Collins, CO.
Available online at http://www.aphis.usda.gov/vs/ceah/cahm.
USDA:APHIS:VS (1994) Beef Cow/Calf Health and Productivity Audit. USDA-APHIS-VS, CEAH. Fort Collins, CO.
Available online at http://www.aphis.usda.gov/vs/ceah/cahm.
USDA:APHIS:VS (1993) Beef Cow/Calf Health and Productivity Audit. USDA-APHIS-VS, CEAH. Fort Collins, CO.
August 1993. Available online at http://www.aphis.usda.gov/vs/ceah/cahm.
Vasconcelos and Galyean (2007) Nutritional recommendations of feedlot consulting nutritionists: The 2007 Texas
Tech University Study. J. Anim. Sci. 85:2772-2781.
10-54 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
Manure Management
ASAE (1998) ASAE Standards 1998, 45th Edition. American Society of Agricultural Engineers. St. Joseph, Ml.
Bryant, M.P., V.H. Varel, R.A. Frobish, and H.R. Isaacson (1976) In H.G. Schlegel (ed.)]; Seminar on Microbial Energy
Conversion. E. Goltz KG. Gottingen, Germany.
Bush, E. (1998) Personal communication with Eric Bush, Centers for Epidemiology and Animal Health, U.S.
Department of Agriculture regarding National Animal Health Monitoring System's (NAHMS) Swine '95 Study.
EPA (2023) AgSTAR Anaerobic Digester Database. Available online at: https://www.epa.gov/agstar/livestock-
ariaerobic-digester-database. Accessed August 2023.
EPA (2008) Climate Leaders Greenhouse Gas Inventory Protocol Offset Project Methodology for Project Type
Managing Manure with Biogas Recovery Systems.
EPA (2005) National Emission Inventory—Ammonia Emissions from Animal Agricultural Operations, Revised Draft
Report. U.S. Environmental Protection Agency. Washington, D.C. April 22, 2005.
EPA (2002a) Development Document for the Final Revisions to the National Pollutant Discharge Elimination System
(NPDES) Regulation and the Effluent Guidelines for Concentrated Animal Feeding Operations (CAFOS). U.S.
Environmental Protection Agency. EPA-821-R-03-001. December 2002.
EPA (2002b) Cost Methodology for the Final Revisions to the National Pollutant Discharge Elimination System
Regulation and the Effluent Guidelines for Concentrated Animal Feeding Operations. U.S. Environmental
Protection Agency. EPA-821-R-03-004. December 2002.
EPA (1992) Global Methane Emissions from Livestock and Poultry Manure, Office of Air and Radiation, U.S.
Environmental Protection Agency. February 1992.
ERG (2023) Summary of Data Processing and Proposed Integration of 2018 Beef Feedlot and Poultry Waste
Management System Data into the Manure Management Greenhouse Gas Inventory. Memorandum to EPA from
ERG, December 2023.
ERG (2021) Updated Other Animal Population Distribution Methodology. Memorandum to EPA from ERG.
ERG (2019) "Incorporation of USDA 2016 ARMS Dairy Data into the Manure Management Greenhouse Gas
Inventory." Memorandum to USDA OCE and EPA from ERG, December 2019.
ERG (2018) "Incorporation of USDA 2009 ARMS Swine Data into the Manure Management Greenhouse Gas
Inventory." Memorandum to USDA OCE and EPA from ERG, November 2018.
ERG (2010a) "Typical Animal Mass Values for Inventory Swine Categories." Memorandum to EPA from ERG. July 19,
2010.
ERG (2010b) Telecon with William Boyd of USDA NRCS and Cortney Itle of ERG Concerning Updated VS and Nex
Rates. August 8, 2010.
ERG (2010c) "Updating Current Inventory Manure Characteristics new USDA Agricultural Waste Management Field
Handbook Values." Memorandum to EPA from ERG. August 13, 2010.
ERG (2008) "Methodology for Improving Methane Emissions Estimates and Emission Reductions from Anaerobic
Digestion System for the 1990-2007 Greenhouse Gas Inventory for Manure Management." Memorandum to EPA
from ERG. August 18, 2008.
ERG (2003a) "Methodology for Estimating Uncertainty for Manure Management Greenhouse Gas Inventory."
Contract No. GS-10F-0036, Task Order 005. Memorandum to EPA from ERG, Lexington, MA. September 26, 2003.
ERG (2003b) "Changes to Beef Calves and Beef Cows Typical Animal Mass in the Manure Management Greenhouse
Gas Inventory." Memorandum to EPA from ERG, October 7, 2003.
References and Abbreviations 10-55
-------
ERG (2001) Summary of development of MDP Factor for methane conversion factor calculations. ERG, Lexington,
MA. September 2001.
ERG (2000a) Calculations: Percent Distribution of Manure for Waste Management Systems. ERG, Lexington, MA.
August 2000.
ERG (2000b) Discussion of Methodology for Estimating Animal Waste Characteristics (Summary of Bo Literature
Review). ERG, Lexington, MA. June 2000.
Groffman, P.M., R. Brumme, K. Butterbach-Bahl, K.E. Dobbie, A.R. Mosier, D. Ojima, H. Papen, W.J. Parton, K.A.
Smith, and C. Wagner-Riddle (2000) "Evaluating annual nitrous oxide fluxes at the ecosystem scale." Global
Biogeochemical Cycles, 14(4): 1061-1070.
Hashimoto, A.G. (1984) "Methane from Swine Manure: Effect of Temperature and Influent Substrate Composition
on Kinetic Parameter (k)." Agricultural Wastes, 9:299-308.
Hashimoto, A.G., V.H. Varel, and Y.R. Chen (1981) "Ultimate Methane Yield from Beef Cattle Manure; Effect of
Temperature, Ration Constituents, Antibiotics and Manure Age." Agricultural Wastes, 3:241-256.
Hill, D.T. (1984) "Methane Productivity of the Major Animal Types." Transactions of the ASAE, 27(2):530-540.
Hill, D.T. (1982) "Design of Digestion Systems for Maximum Methane Production." Transactions of the ASAE,
25(l):226-230.
IPCC (2019) 2019 Refinement to the 2006IPCC Guidelines for National Greenhouse Gas Inventories. The National
Greenhouse Gas Inventories Programme, The Intergovernmental Panel on Climate Change. [CalvoBuendia, E.,
Tanabe, K., Kranjc, A., Baasansuren, J., Fukuda, M., Ngarize S., Osako, A., Pyrozhenko, Y., Shermanau, P. and
Federici, S. (eds)]. Switzerland.
IPCC (2013) Climate Change 2013: The Physical Science Basis. Contribution of Working Group I to the Fifth
Assessment Report of the Intergovernmental Panel on Climate Change [Stocker, T.F., D. Qin, G.-K. Plattner, M.
Tignor, S.K. Allen, J. Boschung, A. Nauels, Y. Xia, V. Bex and P.M. Midgley (eds.)]. Cambridge University Press,
Cambridge, United Kingdom and New York, NY, USA, 1535 pp.IPCC (2006) 2006 IPCC Guidelines for National
Greenhouse Gas Inventories. The National Greenhouse Gas Inventories Programme, The Intergovernmental Panel
on Climate Change. [H.S. Eggleston, L. Buendia, K. Miwa, T. Ngara, and K. Tanabe (eds.)]. Hayama, Kanagawa,
Japan.
IPCC (2006) 2006 IPCC Guidelines for National Greenhouse Gas Inventories. The National Greenhouse Gas
Inventories Programme, The Intergovernmental Panel on Climate Change. H.S. Eggleston, L. Buendia, K. Miwa, T.
Ngara, and K. Tanabe (eds.). Hayama, Kanagawa, Japan.
Morris, G.R. (1976) Anaerobic Fermentation of Animal Wastes: A Kinetic and Empirical Design Fermentation. M.S.
Thesis. Cornell University.
National Bison Association (1999) Total Bison Population—1999. Report provided during personal email
communication with Dave Carter, Executive Director, National Bison Association July 19, 2011.
Ott, S.L (2000) Dairy '96 Study. Stephen L. Ott, Animal and Plant Health Inspection Service, U.S. Department of
Agriculture. June 19, 2000.
Robertson, G. P. and P. M. Groffman (2015) Nitrogen transformations. Soil Microbiology, Ecology, and
Biochemistry, pages 421-446. Academic Press, Burlington, Massachusetts, USA.
Safley, L.M., Jr. (2000) Personal Communication. Deb Bartram, ERG and LM. Safley, President, Agri-Waste
Technology. June and October 2000.
Sweeten, J. (2000) Personal Communication. John Sweeten, Texas A&M University and Indra Mitra, ERG. June
2000.
10-56 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
UEP (1999) Voluntary Survey Results—Estimated Percentage Participation/Activity. Caged Layer Environmental
Management Practices, Industry data submissions for EPA profile development, United Egg Producers and National
Chicken Council. Received from John Thorne, Capitolink. June 2000.
USDA (2023a) Quick Stats: Agricultural Statistics Database. National Agriculture Statistics Service, U.S. Department
of Agriculture. Washington, D.C. Available online at: http://quickstats.nass.usda.gov/.
USDA (2023b) Chicken and Eggs 2022 Summary. National Agriculture Statistics Service, U.S. Department of
Agriculture. Washington, D.C. February 2023. Available online at: https://www.nass.usda.gov/Publications/.
USDA (2023c) Poultry - Production and Value 2022 Summary. National Agriculture Statistics Service, U.S.
Department of Agriculture. Washington, D.C. April 2023. Available online at:
https://www.nass.usda.gov/Publications/.
USDA (2021a) Chicken and Eggs 2020 Summary. National Agriculture Statistics Service, U.S. Department of
Agriculture. Washington, D.C. February 2021. Available online at:
http://www.nass.usda.gov/Publications/index.asp.
USDA (2021b) Poultry - Production and Value 2020 Summary. National Agriculture Statistics Service, U.S.
Department of Agriculture. Washington, D.C. April 2021. Available online at:
http://www.nass.usda.gov/Publications/index.asp.
USDA (2019a) Chicken and Eggs 2018 Summary. National Agriculture Statistics Service, U.S. Department of
Agriculture. Washington, D.C. February 2019. Available online at:
http://www.nass.usda.gov/Publications/index.php.
USDA (2019b) Poultry - Production and Value 2018 Summary. National Agriculture Statistics Service, U.S.
Department of Agriculture. Washington, D.C. April 2019. Available online at:
http://www.nass.usda.gov/Publications/index.php.
USDA (2019c) Chicken and Eggs 2013-2017 Summary. National Agriculture Statistics Service, U.S. Department of
Agriculture. Washington, D.C. June 2019. Available online at: http://www.nass.usda.gov/Publications/index.php.
USDA (2019d) 1987,1992,1997, 2002, 2007, 2012, and 2017 Census of Agriculture. National Agriculture Statistics
Service, U.S. Department of Agriculture. Washington, D.C. Available online at:
https://www.nass.usda.gov/AgCensus/index.php. May 2019.
USDA (2018) Poultry - Production and Value 2017 Summary. National Agriculture Statistics Service, U.S.
Department of Agriculture. Washington, D.C. April 2018. Available online at:
http://www.nass.usda.gov/Publications/index.php.
USDA (2017) Poultry - Production and Value 2016 Summary. National Agriculture Statistics Service, U.S.
Department of Agriculture. Washington, D.C. April 2017. Available online at:
http://www.nass.usda.gov/Publications/index.php.
USDA (2016) Poultry - Production and Value 2015 Summary. National Agriculture Statistics Service, U.S.
Department of Agriculture. Washington, D.C. April 2016. Available online at:
http://www.nass.usda.gov/Publications/index.php.
USDA (2015) Poultry - Production and Value 2014 Summary. National Agriculture Statistics Service, U.S.
Department of Agriculture. Washington, D.C. April 2015. Available online at:
http://www.nass.usda.gov/Publications/index.php.
USDA (2014) Poultry - Production and Value 2013 Summary. National Agriculture Statistics Service, U.S.
Department of Agriculture. Washington, D.C. April 2014. Available online at:
http://www.nass.usda.gov/Publications/index.php.
References and Abbreviations 10-57
-------
USDA (2013a) Chicken and Eggs 2012 Summary. National Agriculture Statistics Service, U.S. Department of
Agriculture. Washington, D.C. February 2013. Available online at:
http://www.nass. usda.gov/Publications/index.php.
USDA (2013b) Poultry - Production and Value 2012 Summary. National Agriculture Statistics Service, U.S.
Department of Agriculture. Washington, D.C. April 2013. Available online at:
http://www.nass.usda.gov/Publications/index.php.
USDA (2012a) Chicken and Eggs 2011 Summary. National Agriculture Statistics Service, U.S. Department of
Agriculture. Washington, D.C. February 2012. Available online at:
http://www.nass.usda.gov/Publications/index.php.
USDA (2012b) Poultry - Production and Value 2011 Summary. National Agriculture Statistics Service, U.S.
Department of Agriculture. Washington, D.C. April 2012. Available online at:
http://www.nass.usda.gov/Publications/index.php.
USDA (2011a) Chicken and Eggs 2010 Summary. National Agriculture Statistics Service, U.S. Department of
Agriculture. Washington, D.C. February 2011. Available online at:
http://www.nass.usda.gov/Publications/index.php.
USDA (2011b) Poultry - Production and Value 2010 Summary. National Agriculture Statistics Service, U.S.
Department of Agriculture. Washington, D.C. April 2011. Available online at:
http://www.nass.usda.gov/Publications/index.php.
USDA (2010a) Chicken and Eggs 2009 Summary. National Agriculture Statistics Service, U.S. Department of
Agriculture. Washington, D.C. February 2010. Available online at:
http://www.nass.usda.gov/Publications/index.php.
USDA (2010b) Poultry - Production and Value 2009 Summary. National Agriculture Statistics Service, U.S.
Department of Agriculture. Washington, D.C. April 2010. Available online at:
http://www.nass.usda.gov/Publications/index.php.
USDA (2009a) Chicken and Eggs 2008 Summary. National Agriculture Statistics Service, U.S. Department of
Agriculture. Washington, D.C. February 2009. Available online at:
http://www.nass.usda.gov/Publications/index.php.
USDA (2009b) Poultry - Production and Value 2008 Summary. National Agriculture Statistics Service, U.S.
Department of Agriculture. Washington, D.C. April 2009. Available online at:
http://www.nass.usda.gov/Publications/index.php.
USDA (2009c) Chicken and Eggs - Final Estimates 2003-2007. National Agriculture Statistics Service, U.S.
Department of Agriculture. Washington, D.C. March 2009. Available online at:
https://www.nass.usda.gov/Publications/index.php.
USDA (2009d) Poultry Production and Value—Final Estimates 2003-2007. National Agriculture Statistics Service,
U.S. Department of Agriculture. Washington, D.C. May 2009. Available online at:
https://www.nass.usda.gov/Publications/index.php.
USDA (2008) Agricultural Waste Management Field Handbook, National Engineering Handbook (NEH), Part 651.
Natural Resources Conservation Service, U.S. Department of Agriculture.
USDA (2004a) Chicken and Eggs—Final Estimates 1998-2003. National Agriculture Statistics Service, U.S.
Department of Agriculture. Washington, D.C. April 2004. Available online at:
https://www.nass.usda.gov/Publications/index.php.
USDA (2004b) Poultry Production and Value—Final Estimates 1998-2002. National Agriculture Statistics Service,
U.S. Department of Agriculture. Washington, D.C. April 2004. Available online at:
https://www.nass.usda.gov/Publications/index.php.
10-58 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
USDA (1999) Poultry Production and Value—Final Estimates 1994-97. National Agriculture Statistics Service, U.S.
Department of Agriculture. Washington, D.C. March 1999. Available online at:
https://www.nass.usda.gov/Publications/index.php.
USDA (1998) Chicken and Eggs—Final Estimates 1994-97. National Agriculture Statistics Service, U.S. Department
of Agriculture. Washington, D.C. December 1998. Available online at:
https://www.nass.usda.gov/Publications/index.php.
USDA (1996) Agricultural Waste Management Field Handbook, National Engineering Handbook (NEH), Part 651.
Natural Resources Conservation Service, U.S. Department of Agriculture. July 1996.
USDA (1995a) Poultry Production and Value—Final Estimates 1988-1993. National Agriculture Statistics Service,
U.S. Department of Agriculture. Washington, D.C. March 1995. Available online at:
https://www.nass.usda.gov/Publications/index.php.
USDA (1995b) Chicken and Eggs—Final Estimates 1988-1993. National Agriculture Statistics Service, U.S.
Department of Agriculture. Washington, D.C. December 1995. Available online at:
https://www.nass.usda.gov/Publications/index.php.
USDA (1994) Sheep and Goats—Final Estimates 1989-1993. National Agriculture Statistics Service, U.S. Department
of Agriculture. Washington, D.C. January 31,1994. Available online at:
https://www.nass.usda.gov/Publications/index.php.
USDA APHIS (2003) Sheep 2001, Part I: Reference of Sheep Management in the United States, 2001 and Part IV:
Baseline Reference of 2001 Sheep Feedlot Health and Management. USDA-APHIS-VS. Fort Collins, CO. #N356.0702.
Available online at http://www.aphis.usda.gov/animal health/nahrris/sheep/index.shtrol#sheep2001.
USDA APHIS (2000) Layers '99—Part II: References of 1999 Table Egg Layer Management in the U.S. USDA-APHIS-
VS. Fort Collins, CO. Available online at
http://www.aphis.usda.gov/animal health/nahms/poultry/downloads/layers99/Layers99 dr Partll.pdf.
USDA APHIS (1996) Swine '95: Grower/Finisher Part II: Reference of 1995 U.S. Grower/Finisher Health &
Management Practices. USDA-APHIS-VS. Fort Collins, CO. Available online at:
http://www.aphis.usda.gov/animal health/nahms/swine/downloads/swine95/Swine95 dr Partll.pdf.
Rice Cultivation
Baicich, P. (2013) The Birds and Rice Connection. Bird Watcher's Digest. Available online at:
http://www.usarice.com/doclib/194/6867.pdf.
Brockwell, P.J., and R.A. Davis (2016) Introduction to time series and forecasting. Springer.
Cantens, G. (2004 through 2005) Personal Communication. Janet Lewis, Assistant to Gaston Cantens, Vice
President of Corporate Relations, Florida Crystals Company and ICF International.
Cheng, K., S.M. Ogle, W.J. Parton, G. Pan. (2014) "Simulating greenhouse gas mitigation potentials for Chinese
croplands using the DAYCENT ecosystem model." Global Change Biology 20:948-962.
Cheng, K., S.M. Ogle, W.J. Parton and G. Pan. (2013) "Predicting methanogenesis from rice paddies using the
DAYCENT ecosystem model." Ecological Modelling 261-262:19-31.
Del Grosso, S.J., S.M. Ogle, W.J. Parton, and F.J. Breidt (2010) "Estimating Uncertainty in N2O Emissions from U.S.
Cropland Soils." Global Biogeochemical Cycles, 24, GB1009, doi:10.1029/2009GB003544.
Deren, C. (2002) Personal Communication and Dr. Chris Deren, Everglades Research and Education Centre at the
University of Florida and Caren Mintz, ICF International. August 15, 2002.
Fitzgerald, G.J., K. M. Scow & J. E. Hill (2000) "Fallow Season Straw and Rice Management Effects on Methane
Emissions in California Rice." Global biogeochemical cycles, 14 (3), 767-776.
References and Abbreviations 10-59
-------
Fleskes, J.P., Perry, W.M., Petrik, K.L., Spell, R., and Reid, F. (2005) Change in area of winter-flood and dry rice in
the northern Central Valley of California determined by satellite imagery. California Fish and Game, 91: 207-215.
Gonzalez, R. (2007 through 2014) Email correspondence. Rene Gonzalez, Plant Manager, Sem-Chi Rice Company
and ICF International.
Hardke, J.T. (2015) Trends in Arkansas rice production, 2014. B.R. Wells Arkansas Rice Research Studies 2014.
Norman, R.J. and Moldenhauer, K.A.K. (Eds.). Research Series 626, Arkansas Agricultural Experiment Station,
University of Arkansas.
Hardke, J. (2014) Personal Communication. Dr. Jarrod Hardke, Rice Extension Agronomist at the University of
Arkansas Rice Research and Extension Center and Kirsten Jaglo, ICF International. September 11, 2014.
Hardke, J. (2013) Email correspondence. Dr. Jarrod Hardke, Rice Extension Agronomist at the University of
Arkansas Rice Research and Extension Center and Cassandra Snow, ICF International. July 15, 2013.
Hardke, J.T., and Wilson, C.E. Jr., (2014) Trends in Arkansas rice production, 2013. B.R. Wells Arkansas Rice
Research Studies 2013. Norman, R.J., and Moldenhauer, K.A.K., (Eds.). Research Series 617, Arkansas Agricultural
Experiment Station, University of Arkansas.
Hardke, J.T., and Wilson, C.E. Jr., (2013) Trends in Arkansas rice production. B.R. Wells Arkansas Rice Research
Studies 2012. Norman, R.J., and Moldenhauer, K.A.K., (Eds.). Research Series 609, Arkansas Agricultural Experiment
Station, University of Arkansas.
Hollier, C. A. (ed), (1999) Louisiana rice production handbook. Louisiana State University Agricultural Center. LCES
Publication Number 2321. 116 pp.
Holzapfel-Pschorn, A., R. Conrad, and W. Seiler (1985) "Production, Oxidation, and Emissions of Methane in Rice
Paddies." FEMS Microbiology Ecology, 31:343-351.
IPCC (2013) Climate Change 2013: The Physical Science Basis. Contribution of Working Group I to the Fifth
Assessment Report of the Intergovernmental Panel on Climate Change [Stocker, T.F., D. Qin, G.-K. Plattner, M.
Tignor, S.K. Allen, J. Boschung, A. Nauels, Y. Xia, V. Bex and P.M. Midgley (eds.)]. Cambridge University Press,
Cambridge, United Kingdom and New York, NY, USA, 1535 pp.
IPCC (2006) 2006 IPCC Guidelines for National Greenhouse Gas Inventories. The National Greenhouse Gas
Inventories Programme, The Intergovernmental Panel on Climate Change. [H.S. Eggleston, L. Buendia, K. Miwa, T.
Ngara, and K. Tanabe (eds.)]. Hayama, Kanagawa, Japan.
Johnson, D.M., and R. Mueller (2010) The 2009 Cropland Data Layer. Photogrammetric engineering and remote
sensing 76:1201-1205.
Kirstein, A. (2003 through 2004, 2006) Personal Communication. Arthur Kirstein, Coordinator, Agricultural
Economic Development Program, Palm Beach County Cooperative Extension Service, FL and ICF International.
Klosterboer, A. (1997,1999 through 2003) Personal Communication. Arlen Klosterboer, retired Extension
Agronomist, Texas A&M University and ICF International. July 7, 2003.
Lindau, C.W. and P.K. Bollich (1993) "Methane Emissions from Louisiana First and Ratoon Crop Rice." Soil Science,
156:42-48.
Linquist, B.A., M.A. Adviento-Borbe, C.M. Pittelkow, C.v. Kessel, et al. (2012) Fertilizer management practices and
greenhouse gas emissions from rice systems: A quantitative review and analysis. Field Crops Research, 135:10-21.
Linscombe, S. (1999, 2001 through 2014) Email correspondence. Steve Linscombe, Professor with the Rice
Research Station at Louisiana State University Agriculture Center and ICF International.
LSU, (2015) Louisiana ratoon crop and conservation: Ratoon & Conservation Tillage Estimates. Louisiana State
University, College of Agriculture AgCenter. Online at: www.lsuagcenter.com.
10-60 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
Miller, M.R., Garr, J.D., and Coates, P.S., (2010) Changes in the status of harvested rice fields in the Sacramento
Valley, California: Implications for wintering waterfowl. Wetlands, 30: 939-947.
Nelson, Mark D.; Riitters, Kurt H.; Coulston, John W.; Domke, Grant M.; Greenfield, Eric J.; Langner, Linda L.;
Nowak, David J.; O'Dea, Claire B.; Oswalt, Sonja N.; Reeves, Matthew C.; Wear, David N. 2020. Defining the United
States land base: a technical document supporting the USDA Forest Service 2020 RPA assessment. Gen. Tech. Rep.
NRS-191. Madison, Wl: U.S. Department of Agriculture, Forest Service, Northern Research Station. 70 p.
https://doi.org/10.2737/NRS-GTR-191.
Neue, H.U., R. Wassmann, H.K. Kludze, W. Bujun, and R.S. Lantin (1997) "Factors and processes controlling
methane emissions from rice fields." Nutrient Cycling in Agroecosystems 49: 111-117.
Ogle, S.M., F.J. Breidt, M. Easter, S. Williams and K. Paustian. (2007) "An empirically based approach for estimating
uncertainty associated with modeling carbon sequestration in soils." Ecological Modelling 205:453-463.
Ogle, S.M., F.J. Breidt, M. Easter, S. Williams, K. Killian, and K. Paustian (2010) "Scale and uncertainty in modeled
soil organic carbon stock changes for U.S. croplands using a process-based model." Global Change Biology 16:810-
822.
Ogle, S.M., S. Spencer, M. Hartman, L. Buendia, L Stevens, D. du Toit, J. Witi (2016) "Developing national baseline
GHG emissions and analyzing mitigation potentials for agriculture and forestry using an advanced national GHG
inventory software system." In Advances in Agricultural Systems Modeling 6, Synthesis and Modeling of
Greenhouse Gas Emissions and Carbon Storage in Agricultural and Forestry Systems to Guide Mitigation and
Adaptation, S. Del Grosso, LR. Ahuja and W.J. Parton (eds.), American Society of Agriculture, Crop Society of
America and Soil Science Society of America, pp. 129-148.
Parton, W.J., M.D. Hartman, D.S. Ojima, and D.S. Schimel (1998) "DAYCENT: Its Land Surface Submodel: Description
and Testing". Glob. Planet. Chang. 19: 35-48.
Parton, W.J., D.S. Schimel, C.V. Cole, D.S. Ojima (1987) "Analysis of factors controlling soil organic matter levels in
Great Plains grasslands." Soil Science Society of America Journal 51:1173-1179.
Sarndal C-E, Swensson B, Wretman, J (1992) Model Assisted Survey Sampling. Springer, New York.
Sass, R. L. (2001) Cm Emissions from Rice Agriculture. Good Practice Guidance and Uncertainty Management in
National Greenhouse Gas Inventories. 399-417. Available online at: http://www.ipcc-
nggip.iges.or.ip/public/gp/bgp/4 7 CH4 Rice Agriculture.pdf.
Sass, R.L, F.M. Fisher, P.A. Harcombe, and F.T. Turner (1990) "Methane Production and Emissions in a Texas Rice
Field." Global Biogeochemical Cycles, 4:47-68.
Sass, R.L, F.M. Fisher, S.T. Lewis, M.F. Jund, and F.T. Turner. (1994) "Methane emissions from rice fields: effect of
soil texture." Global Biogeochemical Cycles 8:135-140.
Schueneman, T. (1997,1999 through 2001) Personal Communication. Tom Schueneman, Agricultural Extension
Agent, Palm Beach County, FLand ICF International.
Slaton, N. (1999 through 2001) Personal Communication. Nathan Slaton, Extension Agronomist—Rice, University
of Arkansas Division of Agriculture Cooperative Extension Service and ICF International.
Stansel, J. (2004 through 2005) Email correspondence. Dr. Jim Stansel, Resident Director and Professor Emeritus,
Texas A&M University Agricultural Research and Extension Center and ICF International.
TAMU (2015) Texas Rice Crop Survey. Texas A&M AgriLIFE Research Center at Beaumont. Online at:
https://beaumont.tamu.edu/.
Texas Agricultural Experiment Station (2007 through 2014) Texas Rice Acreage by Variety. Agricultural Research
and Extension Center, Texas Agricultural Experiment Station, Texas A&M University System. Available online at:
http://beaumont.tamu.edu/CropSurvev/CropSurveyReport.aspx.
References and Abbreviations 10-61
-------
Texas Agricultural Experiment Station (2006) 2005 - Texas Rice Crop Statistics Report. Agricultural Research and
Extension Center, Texas Agricultural Experiment Station, Texas A&M University System, p. 8. Available online at:
http://beaumont.tamu.edu/eLibrary/TRRFReport default.htm.
University of California Cooperative Extension (UCCE) (2015) Rice Production Manual. Revised (2015) UCCE, Davis,
in collaboration with the California Rice Research Board.
USDA (2005 through 2015) Crop Production Summary. National Agricultural Statistics Service, Agricultural Statistics
Board, U.S. Department of Agriculture, Washington, D.C. Available online at: http://usda.mannlib.cornell.edu.
USDA (2012) Summary of USDA-ARS Research on the Interrelationship of Genetic and Cultural Management
Factors That Impact Grain Arsenic Accumulation in Rice. News and Events. Agricultural Research Service, U.S.
Department of Agriculture, Washington, D.C. Available online at:
http://www.ars.usda.gov/is/pr/2012/120919.htm. September 2013.
USDA (2003) Field Crops, Final Estimates 1997-2002. Statistical Bulletin No. 982. National Agricultural Statistics
Service, Agricultural Statistics Board, U.S. Department of Agriculture, Washington, D.C. Available online at:
http://usda.mannlib.cornell.edu/usda/reports/general/sb/. September 2005.
USDA (1998) Field Crops Final Estimates 1992-1997. Statistical Bulletin Number 947 a. National Agricultural
Statistics Service, Agricultural Statistics Board, U.S. Department of Agriculture, Washington, D.C. Available online
at: http://usda.mannlib.cornell.edu/. July 2001.
USDA (1994) Field Crops Final Estimates 1987-1992. Statistical Bulletin Number 896. National Agricultural Statistics
Service, Agricultural Statistics Board, U.S. Department of Agriculture, Washington, D.C. Available online at:
http://usda.mannlib.cornell.edu/. July 2001.
USDA-NASS (2021) Published crop data layer. Available at https://nassgeodata.gmu.edu/CropScape/, Accessed July
2021, USDA-NASS, Washington, DC.
USDA-NRCS (2020) Summary Report: 2017 National Resources Inventory. Natural Resources Conservation Service,
Washington, DC, and Center for Survey Statistics and Methodology, Iowa State University, Ames, Iowa.
https://www.nrcs.usda.gov/wps/portal/nrcs/main/national/technical/nra/nri/results/.
van Bodegom, P.M., R. Wassmann, T.M. Metra-Corton (2001) "A process based model for methane emission
predictions from flooded rice paddies." Global Biogeochemical Cycles 15: 247-263.
Wang, J.J., S.K. Dodla, S. Viator, M. Kongchum, S. Harrison, S. D. Mudi, S. Liu, Z. Tian (2013) Agriculture Field
Management Practices and Greenhouse Gas Emissions from Louisiana Soils. Louisiana Agriculture, Spring 2013: 8-
9. Available online at: http://www.lsuagcenter.com/NR/rdonlyres/78D8B61A-96A8-49El-B2EF-
BAlD4CE4E698/93016/v56no2Spring2013.pdf.
Wassmann, R. H.U. Neue, R.S. Lantin, K. Makarim, N. Chareonsil5, LV. Buendia, and H. Rennenberg (2000a)
Characterization of methane emissions from rice fields in Asia II. Differences among irrigated, rainfed, and
deepwater rice." Nutrient Cycling in Agroecosystems, 58(1): 13-22.
Wassmann, R., R.S. Lantin, H.U. Neue, LV. Buendia, et al. (2000b) "Characterization of Methane Emissions from
Rice Fields in Asia. III. Mitigation Options and Future Research Needs." Nutrient Cycling in Agroecosystems,
58(l):23-36.
Way, M.O., McCauley, G.M., Zhou, X.G., Wilson, L.T., and Morace, B. (Eds.), (2014) 2014 Texas Rice Production
Guidelines. Texas A&M AgriLIFE Research Center at Beaumont.
Wilson, C. (2002 through 2007, 2009 through 2012) Personal Communication. Dr. Chuck Wilson, Rice Specialist at
the University of Arkansas Cooperative Extension Service and ICF International.
Wilson, C.E. Jr., and Branson, J.W., (2006) Trends in Arkansas rice production. B.R. Wells Arkansas Rice Research
Studies 2005. Norman, R.J., Meullenet, J.-F., and Moldenhauer, K.A.K., (Eds.). Research Series 540, Arkansas
Agricultural Experiment Station, University of Arkansas.
10-62 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
Wilson, C.E. Jr., and Branson, J.W., (2005) Trends in Arkansas rice production. B.R. Wells Arkansas Rice Research
Studies 2004. Norman, R.J., Meullenet, J.-F., and Moldenhauer, K.A.K., (Eds.). Research Series 529, Arkansas
Agricultural Experiment Station, University of Arkansas.
Wilson, C.E. Jr., Runsick, S.K., and Mazzanti, R., (2010) Trends in Arkansas rice production. B.R. Wells Arkansas Rice
Research Studies 2009. Norman, R.J., and Moldenhauer, K.A.K., (Eds.). Research Series 581, Arkansas Agricultural
Experiment Station, University of Arkansas.
Wilson, C.E. Jr., Runsick, S.K., Mazzanti, R., (2009) Trends in Arkansas rice production. B.R. Wells Arkansas Rice
Research Studies (2008) Norman, R.J., Meullenet, J.-F., and Moldenhauer, K.A.K., (Eds.). Research Series 571,
Arkansas Agricultural Experiment Station, University of Arkansas.
Wilson, C.E. Jr., and Runsick, S.K., (2008) Trends in Arkansas rice production. B.R. Wells Arkansas Rice Research
Studies 2007. Norman, R.J., Meullenet, J.-F., and Moldenhauer, K.A.K., (Eds.). Research Series 560, Arkansas
Agricultural Experiment Station, University of Arkansas.
Wilson, C.E. Jr., and Runsick, S.K., (2007) Trends in Arkansas rice production. B.R. Wells Arkansas Rice Research
Studies 2006. Norman, R.J., Meullenet, J.-F., and Moldenhauer, K.A.K., (Eds.). Research Series 550, Arkansas
Agricultural Experiment Station, University of Arkansas.
Yan, X., H. Akiyana, K. Yagi, and H. Akimoto (2009) "Global estimations of the inventory and mitigation potential of
methane emissions from rice cultivation conducted using the 2006 Intergovernmental Panel on Climate Change
Guidelines." Global Biogeochemical Cycles, 23, DOI: 0.1029/2008GB003299.
Yang, L, et al. (2018). "A new generation of the United States National Land Cover Database: Requirements,
research priorities, design, and implementation strategies." ISPRS Journal of Photogrammetry and Remote Sensing
146: 108-123.
Young, M. (2013) Rice and Ducks. Ducks Unlimited, Memphis, TN. Available online at:
http://www.ducks.org/conservation/farm-bill/rice-and-ducks—by-matt-young.
Agricultural Soil Management
AAPFCO (2008 through 2022) Commercial Fertilizers: 2008-2017. Association of American Plant Food Control
Officials. University of Missouri. Columbia, MO.
AAPFCO (1995 through 2000a, 2002 through 2007) Commercial Fertilizers: 1995-2007. Association of American
Plant Food Control Officials. University of Kentucky. Lexington, KY.
Brockwell, Peter J., and Richard A. Davis (2016) Introduction to time series and forecasting. Springer.
Cibrowski, P. (1996) Personal Communication. Peter Cibrowski, Minnesota Pollution Control Agency and Heike
Mainhardt, ICF Incorporated. July 29,1996.
Cheng, B., and D.M. Titterington (1994) "Neural networks: A review from a statistical perspective." Statistical
science 9: 2-30.
Claassen, R., M. Bowman, J. McFadden, D. Smith, and S. Wallander (2018) Tillage intensity and conservation
cropping in the United States, EIB 197. United States Department of Agriculture, Economic Research Service,
Washington, D.C.
CTIC (2004) 2004 Crop Residue Management Survey. Conservation Technology Information Center. Available at
http://www.ctic.purdue.edu/CRM/.
Del Grosso, S. J., S. M. Ogle, C. Nevison, R. Gurung, W. J. Parton, C. Wagner-Riddle, W. Smith, W. Winiwarter, B.
Grant, M. Tenuta, E. Marx, S. Spencer, and S. Williams. 2022. A gap in nitrous oxide emission reporting complicates
long-term climate mitigation. Proceedings of the National Academy of Sciences 119:e2200354119.
References and Abbreviations 10-63
-------
Del Grosso, S.J., A.R. Mosier, W.J. Parton, and D.S. Ojima (2005) "DAYCENT Model Analysis of Past and
Contemporary Soil N2O and Net Greenhouse Gas Flux for Major Crops in the USA." Soil Tillage and Research, 83: 9-
24. doi: 10.1016/j.still.2005.02.007.
Del Grosso, S.J., S.M. Ogle, W.J. Parton, and F.J. Breidt (2010) "Estimating Uncertainty in N2O Emissions from U.S.
Cropland Soils." Global Biogeochemical Cycles, 24, GB1009, doi:10.1029/2009GB003544.
Del Grosso, S.J., T. Wirth, S.M. Ogle, W.J. Parton (2008) Estimating agricultural nitrous oxide emissions. EOS 89,
529-530.
Del Grosso, S.J., W.J. Parton, A.R. Mosier, M.D. Hartman, J. Brenner, D.S. Ojima, and D.S. Schimel (2001) "Simulated
Interaction of Carbon Dynamics and Nitrogen Trace Gas Fluxes Using the DAYCENT Model." In Schaffer, M., L. Ma,
S. Hansen, (eds.). Modeling Carbon and Nitrogen Dynamics for Soil Management. CRC Press. Boca Raton, Florida.
303-332.
Del Grosso, S.J., W.J. Parton, C.A. Keough, and M. Reyes-Fox. (2011) Special features of the DAYCENT modeling
package and additional procedures for parameterization, calibration, validation, and applications, in Methods of
Introducing System Models into Agricultural Research, L.R. Ahuja and Liwang Ma, editors, p. 155-176, American
Society of Agronomy, Crop Science Society of America, Soil Science Society of America, Madison, Wl. USA.
Delgado, J.A., S.J. Del Grosso, and S.M. Ogle (2009) "15N isotopic crop residue cycling studies and modeling suggest
that IPCC methodologies to assess residue contributions to N2O-N emissions should be reevaluated." Nutrient
Cycling in Agroecosystems, DOI 10.1007/sl0705-009-9300-9.
Edmonds, L, N. Gollehon, R.L. Kellogg, B. Kintzer, L. Knight, C. Lander, J. Lemunyon, D. Meyer, D.C. Moffitt, and J.
Schaeffer (2003) "Costs Associated with Development and Implementation of Comprehensive Nutrient
Management Plans." Part 1. Nutrient Management, Land Treatment, Manure and Wastewater Handling and
Storage, and Recordkeeping. Natural Resource Conservation Service, U.S. Department of Agriculture.
EPA (2003) Clean Watersheds Needs Survey 2000—Report to Congress, U.S. Environmental Protection Agency.
Washington, D.C. Available online at: http://www.epa.gov/owm/mtb/cwns/2000rtc/toc.htm.
EPA (1999) Biosolids Generation, Use and Disposal in the United States. Office of Solid Waste, U.S. Environmental
Protection Agency. Available online at: http://biosolids.policy.net/relatives/18941.PDF.
EPA (1993) Federal Register. Part II. Standards for the Use and Disposal of Sewage Sludge; Final Rules. U.S.
Environmental Protection Agency, 40 CFR Parts 257, 403, and 503.
Firestone, M. K., and E.A. Davidson, Ed. (1989) Microbiological basis of NO and N2O production and consumption in
soil. Exchange of trace gases between terrestrial ecosystems and the atmosphere. New York, John Wiley & Sons.
Friedman, J.H. (2001) "Greedy function approximation: A gradient boosting machine." Ann. Statist. 29 (5) 1189 -
1232.
Hagen, S. C., G. Delgado, P. Ingraham, I. Cooke, R. Emery, J. P. Fisk, L. Melendy, T. Olson, S. Patti, N. Rubin, B. Ziniti,
H. Chen, W. Salas, P. Elias, and D. Gustafson. 2020. Mapping Conservation Management Practices and Outcomes in
the Corn Belt Using the Operational Tillage Information System (OpTIS) and the Denitrification-Decomposition
(DNDC) Model. Land 9:408.
ILENR (1993) Illinois Inventory of Greenhouse Gas Emissions and Sinks: 1990. Office of Research and Planning,
Illinois Department of Energy and Natural Resources. Springfield, IL.
IPCC (2014) 2013 Supplement to the 2006 IPCC Guidelines for National Greenhouse Gas Inventories: Wetlands. The
Intergovernmental Panel on Climate Change. [T, Hiraishi, T. Krug, K. Tanabe, N. Srivastava, B. Jamsranjav, M.
Fukuda and T. Troxler (eds.)]. Hayama, Kanagawa, Japan.
IPCC (2013) Climate Change 2013: The Physical Science Basis. Contribution of Working Group I to the Fifth
Assessment Report of the Intergovernmental Panel on Climate Change [Stocker, T.F., D. Qin, G.-K. Plattner, M.
10-64 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
Tignor, S.K. Allen, J. Boschung, A. Nauels, Y. Xia, V. Bex and P.M. Midgley (eds.)]. Cambridge University Press,
Cambridge, United Kingdom and New York, NY, USA, 1535 pp.
IPCC (2006) 2006IPCC Guidelines for National Greenhouse Gas Inventories. The National Greenhouse Gas
Inventories Programme, The Intergovernmental Panel on Climate Change. [H.S. Eggleston, L. Buendia, K. Miwa, T.
Ngara, and K. Tanabe (eds.)]. Hayama, Kanagawa, Japan.
Johnson, D.M., and R. Mueller (2010) The 2009 Cropland Data Layer. Photogrammetric engineering and remote
sensing 76:1201-1205.
Little, R. (1988) "Missing-data adjustments in large surveys." Journal of Business and Economic Statistics 6: 287-
296.
McFarland, M.J. (2001) Biosolids Engineering, New York: McGraw-Hill, p. 2.12.
McGill, W.B., and C.V. Cole (1981) Comparative aspects of cycling of organic C, N, S and P through soil organic
matter. Geoderma 26:267-286.
Metherell, A.K., L.A. Harding, C.V. Cole, and W.J. Parton (1993) "CENTURY Soil Organic Matter Model
Environment." Agroecosystem version 4.0. Technical documentation, GPSRTech. Report No. 4, USDA/ARS, Ft.
Collins, CO.
Mosier A., C. Kroeze, C. Nevison, O. Oenema, S. Seitzinger, and O. van Cleemput (1998) Closing the global
atmospheric N2O budget: Nitrous oxide emissions through the agricultural nitrogen cycle, Nutrient Cycling in
Agroecosystems, 52, 225-248.
NEBRA (2007) A National Biosolids Regulation, Quality, End Use & Disposal Survey. North East Biosolids and
Residuals Association, July 21, 2007.
Nelson, Mark D.; Riitters, Kurt H.; Coulston, John W.; Domke, Grant M.; Greenfield, Eric J.; Langner, Linda L.;
Nowak, David J.; O'Dea, Claire B.; Oswalt, Sonja N.; Reeves, Matthew C.; Wear, David N. 2020. Defining the United
States land base: a technical document supporting the USDA Forest Service 2020 RPA assessment. Gen. Tech. Rep.
NRS-191. Madison, Wl: U.S. Department of Agriculture, Forest Service, Northern Research Station. 70 p.
https://doi.org/10.2737/NRS-GTR-191.
Noller, J. (1996) Personal Communication. John Noller, Missouri Department of Natural Resources and Heike
Mainhardt, ICF Incorporated. July 30,1996.
Ogle, S.M., F.J. Breidt, M. Easter, S. Williams, K. Killian, and K. Paustian (2010) "Scale and uncertainty in modeled
soil organic carbon stock changes for U.S. croplands using a process-based model." Global Change Biology 16:810-
822.
Ogle, S.M., F.J. Breidt, M. Easter, S. Williams and K. Paustian (2007) "Empirically-Based Uncertainty Associated with
Modeling Carbon Sequestration Rates in Soils." Ecological Modeling 205:453-463.
Oregon Department of Energy (1995) Report on Reducing Oregon's Greenhouse Gas Emissions: Appendix D
Inventory and Technical Discussion. Oregon Department of Energy. Salem, OR.
Parton, W.J., M.D. Hartman, D.S. Ojima, and D.S. Schimel (1998) "DAYCENT: Its Land Surface Submodel: Description
and Testing." Glob. Planet. Chang. 19: 35-48.
Potter, C., S. Klooster, A. Huete, and V. Genovese (2007) Terrestrial carbon sinks for the United States predicted
from MODIS satellite data and ecosystem modeling. Earth Interactions 11, Article No. 13, DOI 10.1175/EI228.1.
Potter, C. S., J.T. Randerson, C.B. Fields, P.A. Matson, P.M. Vitousek, H.A. Mooney, and S.A. Klooster (1993)
"Terrestrial ecosystem production: a process model based on global satellite and surface data." Global
Biogeochemical Cycles 7:811-841.
PRISM Climate Group (2022) PRISM Climate Data, Oregon State University, http://prism.oregonstate.edu,
downloaded January 2022.
References and Abbreviations 10-65
-------
Pukelsheim, F. (1994) 'The 3-Sigma-Rule." American Statistician 48:88-91.
Ruddy B.C., D.L. Lorenz, and D.K. Mueller (2006) County-level estimates of nutrient inputs to the land surface of
the conterminous United States, 1982-2001. Scientific Investigations Report 2006-5012. U.S Department of the
Interior.
Sarndal C-E, Swensson B, Wretman, J (1992) Model Assisted Survey Sampling. Springer, New York.
Scheer, C., S.J. Del Grosso, W.J. Parton, D.W. Rowlings, P.R. Grace (2013) Modeling Nitrous Oxide Emissions from
Irrigated Agriculture: Testing DAYCENT with High Frequency Measurements, Ecological Applications, in press.
Available online at: http://dx.doi.Org/10.1890/13-0570.l.
Soil Survey Staff (2020) Gridded Soil Survey Geographic (gSSURGO) Database for the Conterminous United States.
United States Department of Agriculture, Natural Resources Conservation Service, Accessed February 2020
(FY2020 official release), Available online at https://gdg.sc.egov.usda.gov/.
Towery, D. (2001) Personal Communication. Dan Towery regarding adjustments to the CTIC (1998) tillage data to
reflect long-term trends, Conservation Technology Information Center, West Lafayette, IN, and Marlen Eve,
National Resource Ecology Laboratory, Fort Collins, CO. February 2001.
TVA (1991 through 1992a, 1993 through 1994) Commercial Fertilizers. Tennessee Valley Authority, Muscle Shoals,
AL.
USDA-ERS (2020) Agricultural Resource Management Survey (ARMS) Farm Financial and Crop Production Practices:
Tailored Reports. Available online at: https://www.ers.usda.gov/data-products/arms-farm-financial-and-crop-
production-practices/.
USDA-ERS (1997) Cropping Practices Survey Data —1995. Economic Research Service, United States Department of
Agriculture. Available online at: http://www.ers.usda.gov/data/archive/93018/.
USDA-NASS (2023) Quick Stats. National Agricultural Statistics Service, United States Department of Agriculture,
Washington, D.C., Accessed August 2023, http://quickstats.nass.usda.gov/.
USDA-NASS (2021) Published crop data layer. Available at https://nassgeodata.gmu.edu/CropScape/, Accessed July
2021, USDA-NASS, Washington, DC.
USDA-NASS (2017) 2017 Census of Agriculture. USDA National Agricultural Statistics Service, Complete data
available at http://www.nass.usda.gov/AgCensus.
USDA-NASS (2012) 2012 Census of Agriculture. USDA National Agricultural Statistics Service, Complete data
available at http://www.nass.usda.gov/AgCensus.
USDA-NASS (2004) Agricultural Chemical Usage: 2003 Field Crops Summary. Report AgChl(04)a. National
Agricultural Statistics Service, U.S. Department of Agriculture, Washington, D.C. Available online at:
http://usda.mannlib.cornell.edu/reports/nassr/other/pcu-bb/agcs0504.pdfH.
USDA-NASS (1999) Agricultural Chemical Usage: 1998 Field Crops Summary. Report AgCHl(99). National
Agricultural Statistics Service, U.S. Department of Agriculture, Washington, DC. Available online at:
http://usda.mannlib.cornell.edu/reports/nassr/other/pcu-bb/agch0599.pdf.
USDA-NASS (1992) Agricultural Chemical Usage: 1991 Field Crops Summary. Report AgChl(92). National
Agricultural Statistics Service, U.S. Department of Agriculture, Washington, D.C. Available online at:
http://usda.mannlib.cornell.edu/reports/nassr/other/pcu-bb/agch0392.txtH.
USDA-NRCS (2022) Conversation practice on cultivated croplands: A comparison of CEAP I and CEAP II survey data
and modeling. United States Department of Agriculture, Natural Resources Conservation Service,
https://www.nrcs.usda.gov/sites/default/files/2022-09/CEAP-Croplands-
ConservationPracticesonCultivatedCroplands-Report-March2022.pdf.
10-66 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
USDA-NRCS (2020) Summary Report: 2017 National Resources Inventory. Natural Resources Conservation Service,
Washington, DC, and Center for Survey Statistics and Methodology, Iowa State University, Ames, Iowa.
https://www.nrcs.usda.gov/wps/portal/nrcs/main/national/technical/nra/nri/results/.
USDA-NRCS (2018) CEAP Cropland Farmer Surveys. USDA Natural Resources Conservation Service.
https://www.nrcs.usda.gov/wps/portal/nrcs/detail/national/technical/nra/ceap/na/?cid=nrcsl43 014163.
USDA-NRCS (2012) Assessment of the Effects of Conservation Practices on Cultivated Cropland in the Upper
Mississippi River Basin. U.S. Department of Agriculture, Natural Resources Conservation Service,
https://www.nrcs.usda.gov/lnternet/FSE DOCUMENTS/stelprdb 1042093.pdf.
USFS (2019) Forest Inventory and Analysis Program. United States Department of Agriculture, U.S. Forest Service,
https://www.fia.fs.fed.us/tools-data/default.asp.
Van Buuren, S. (2012) "Flexible imputation of missing data." Chapman & Hall/CRC, Boca Raton, FL
Wagner-Riddle, C., Congreves, K. A., Abalos, D., Berg, A. A., Brown, S. E., Ambadan, J. T., Gao, X. & Tenuta, M.
(2017) "Globally important nitrous oxide emissions from croplands induced by freeze-thaw cycles." Nature
Geosciences 10(4): 279-283.
Wisconsin Department of Natural Resources (1993) Wisconsin Greenhouse Gas Emissions: Estimates for 1990.
Bureau of Air Management, Wisconsin Department of Natural Resources, Madison, Wl.
Yang, L, Jin, S., Danielson, P., Homer, C., Gass, L., Bender, S. M., Case, A., Costello, C., Dewitz, J., Fry, J., Funk, M.,
Granneman, B., Liknes, G. C., Rigge, M. & Xian, G. (2018) "A new generation of the United States National Land
Cover Database: Requirements, research priorities, design, and implementation strategies." ISPRS Journal of
Photogrammetry and Remote Sensing 146:108-123.
Liming
IPCC (2006) 2006IPCC Guidelines for National Greenhouse Gas Inventories. The National Greenhouse Gas
Inventories Programme, The Intergovernmental Panel on Climate Change. [H.S. Eggleston, L. Buendia, K. Miwa, T.
Ngara, and K. Tanabe (eds.)]. Hayama, Kanagawa, Japan.
Tepordei, V.V. (1994 through 2015) "Crushed Stone," In Minerals Yearbook. U.S. Department of the Interior/U.S.
Geological Survey. Washington, D.C. Available online at: http://minerals.usgs.gov/minerals/.
Tepordei, V.V. (2003b) Personal communication. Valentin Tepordei, U.S. Geological Survey and ICF Consulting,
August 18, 2003.
USGS (2023) Mineral Industry Surveys: Crushed Stone and Sand and Gravel in the First Quarter of 2023, U.S.
Geological Survey, Reston, VA. Available online at:
http://minerals.usgs.gov/minerals/pubs/commodity/stone crushed/index.html#mis.
USGS (2022) Mineral Industry Surveys: Crushed Stone and Sand and Gravel in the First Quarter of 2022, U.S.
Geological Survey, Reston, VA. Available online at:
http://minerals.usgs.gov/minerals/pubs/commodity/stone crushed/index.html#mis.
USGS (2021) Mineral Industry Surveys: Crushed Stone and Sand and Gravel in the Fourth Quarter of 2021, U.S.
Geological Survey, Reston, VA. Available online at:
http://minerals.usgs.gov/minerals/pubs/commodity/stone crushed/index.html#mis.
USGS (2020) Mineral Industry Surveys: Crushed Stone and Sand and Gravel in the Fourth Quarter of 2020, U.S.
Geological Survey, Reston, VA. Available online at:
http://minerals.usgs.gov/minerals/pubs/commodity/stone crushed/index.html#mis.
West, T.O., and A.C. McBride (2005) "The contribution of agricultural lime to carbon dioxide emissions in the
United States: dissolution, transport, and net emissions," Agricultural Ecosystems & Environment 108:145-154.
References and Abbreviations 10-67
-------
West, T.O. (2008) Email correspondence. Tristram West, Environmental Sciences Division, Oak Ridge National
Laboratory, U.S. Department of Energy and Nikhil Nadkarni, ICF International on suitability of liming emission
factor for the entire United States. June 9, 2008.
Willett, J.C. (2023a) "Crushed Stone," In Minerals Yearbook 2021. U.S. Department of the Interior/U.S. Geological
Survey, Washington, D.C. Available online at: https://www.usgs.gov/centers/national-minerals-information-
center/crushed-stone-statistics-and-information. Accessed November 2023
Willett, J.C. (2023b) Personal communication. Jason Willett. Preliminary data tables from "Crushed Stone," In 2022
Minerals Yearbook. U.S. Department of the Interior/U.S. Geological Survey. Washington, D.C. November, 2023.
Willett, J.C. (2022a) "Crushed Stone," In Minerals Yearbook 2018. U.S. Department of the Interior/U.S. Geological
Survey, Washington, D.C. Available online at: https://www.usgs.gov/centers/national-minerals-information-
center/crushed-stone-statistics-and-information. Accessed October 2022.
Willett, J.C. (2022b) "Crushed Stone," In Minerals Yearbook 2019. U.S. Department of the Interior/U.S. Geological
Survey, Washington, D.C. Available online at: https://www.usgs.gov/centers/national-minerals-information-
center/crushed-stone-statistics-and-information. Accessed October 2022
Willett, J.C. (2022c) "Crushed Stone," In Minerals Yearbook 2020. U.S. Department of the Interior/U.S. Geological
Survey, Washington, D.C. Available online at: https://www.usgs.gov/centers/national-minerals-information-
center/crushed-stone-statistics-and-information. Accessed October 2022
Willett, J.C. (2022d) Personal communication. Jason Willett. Preliminary data tables from "Crushed Stone," In 2021
Minerals Yearbook. U.S. Department of the Interior/U.S. Geological Survey. Washington, D.C. October, 2022.
Willett, J.C. (2020a) "Crushed Stone," In Minerals Yearbook 2016. U.S. Department of the Interior/U.S. Geological
Survey, Washington, D.C. Available online at:
http://minerals.usgs.gov/minerals/pubs/commodity/stone crushed/index.html#mis. Accessed November 2020.
Urea Fertilization
AAPFCO (2008 through 2022) Commercial Fertilizers. Association of American Plant Food Control Officials.
University of Missouri. Columbia, MO.
AAPFCO (1995 through 2000a, 2002 through 2007) Commercial Fertilizers. Association of American Plant Food
Control Officials. University of Kentucky. Lexington, KY.
AAPFCO (2000b) 1999-2000 Commercial Fertilizers Data, ASCII files. Available from David Terry, Secretary, AAPFCO.
EPA (2000) Preliminary Data Summary: Airport Deicing Operations (Revised). EPA-821-R-00-016. August 2000.
IPCC (2006) 2006IPCC Guidelines for National Greenhouse Gas Inventories. The National Greenhouse Gas
Inventories Programme, The Intergovernmental Panel on Climate Change. [H.S. Eggleston, L. Buendia, K. Miwa, T.
Ngara, and K. Tanabe (eds.)]. Hayama, Kanagawa, Japan.
Itle, C. (2009) Email correspondence. Cortney Itle, ERG and Tom Wirth, U.S. Environmental Protection Agency on
the amount of urea used in aircraft deicing. January 7, 2009.
TVA (1991 through 1994) Commercial Fertilizers. Tennessee Valley Authority, Muscle Shoals, AL.
TVA (1992b) Fertilizer Summary Data 1992. Tennessee Valley Authority, Muscle Shoals, AL.
Field Burning of Agricultural Residues
Akintoye, H.A., Agbeyi, E.O., and Olaniyan, A.B. (2005) 'The effects of live mulches on tomato (Lycopersicon
esculentum) yield under tropical conditions." Journal of Sustainable Agriculture 26: 27-37.
10-68 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
Bange, M.P., Milroy, S.P., and Thongbai, P. (2004) "Growth and yield of cotton in response to waterlogging." Field
Crops Research 88:129-142.
Beyaert, R.P. (1996) The effect of cropping and tillage management on the dynamics of soil organic matter. PhD
Thesis. University ofGuelph.
Bouquet, D.J., and Breitenbeck, G.A. (2000) "Nitrogen rate effect on partitioning of nitrogen and dry matter by
cotton." Crop Science 40:1685-1693.
Brockwell, Peter J., and Richard A. Davis (2016) Introduction to time series and forecasting. Springer. Cantens, G.
(2004 through 2005) Personal Communication. Janet Lewis, Assistant to Gaston Cantens, Vice President of
Corporate Relations, Florida Crystals Company and ICF International.
Brouder, S.M., and Cassman, K.G (1990) "Root development of two cotton cultivars in relation to potassium uptake
and plant growth in a vermiculitic soil." Field Crops Res. 23: 187-203.
Costa, L.D., and Gianquinto, G. (2002) "Water stress and watertable depth influence yield, water use efficiency,
and nitrogen recovery in bell pepper: lysimeter studies." Aust. J. Agric. Res. 53: 201-210.
Crafts-Brandner, S.J., Collins, M., Sutton, T.G., and Burton, H.R. (1994) "Effect of leaf maleic hydrazide
concentration on yield and dry matter partitioning in burley tobacco (Nicotiana tabacum L.)." Field Crops Research
37: 121-128.
De Pinheiro Henriques, A.R., and Marcelis, L.F.M. (2000) "Regulation of growth at steady-state nitrogen nutrition in
lettuce (Lactuca sativa L.): Interactive effects of nitrogen and irradiance." Annals of Botany 86: 1073-1080.0.
Diaz-Perez, J.C., Silvoy, J., Phatak,S.C., Ruberson, J., and Morse, R. (2008) Effect of winter cover crops and co-till on
the yield of organically-grown bell pepper (Capsicum annum L). Acta Hort. 767:243-247.
Dua, K.L., and Sharma, V.K. (1976) "Dry matter production and energy contents often varieties of sugarcane at
Muzaffarnagar (Western Uttar Pradesh)." Tropical Ecology 17: 45-49.
Fritschi, F.B., Roberts, B.A., Travis, R.L, Rains, D.W., and Hutmacher, R.B. (2003) "Seasonal nitrogen concentration,
uptake, and partitioning pattern of irrigated Acala and Pima cotton as influenced by nitrogen fertility level." Crop
Science 44:516-527.
Gerik, T.J., K.L Faver, P.M. Thaxton, and K.M. El-Zik. (1996) "Late season water stress in cotton: I. Plant growth,
water use, and yield." Crop Science 36: 914-921.
Gibberd, M.R., McKay, A.G., Calder, T.C., and Turner, N.C. (2003) "Limitations to carrot (Daucus carota L.)
productivity when grown with reduced rates of frequent irrigation on a free-draining, sandy soil ."Australian
Journal of Agricultural Research 54: 499-506.
Giglio, L., I. Csiszar, and C.O. Justice (2006) "Global distribution and seasonality of active fires as observed with the
Terra and Aqua Moderate Resolution Imaging Spectroradiometer (MODIS) sensors" J. Geophys. Res. Ill, G02016,
doi:10.1029/2005JG000142.
Giglio, L., Justice, C., Boschetti, L., Roy, D. (2015) "MCD64A1 MODIS/Terra+Aqua Burned Area Monthly L3 Global
500m SIN Grid V006 [Data set]". NASA EOSDIS Land Processes Distributed Active Archive Center. Accessed 2023-
12-05 from https://doi.org/10.5067/IVIODlS/IVICD64Al.006.
Halevy, J. (1976) "Growth rate and nutrient uptake of two cotton cultivars grown under irrigation." Agronomy
Journal 68: 701-705.
Halvorson, A.D., Follett, R.F., Bartolo, M.E., and Schweissing, F.C. (2002) "Nitrogen fertilizer use efficiency of
furrow-irrigated onion and corn." Agronomy Journal 94: 442-449.
Heitholt, J.J., Pettigrew, W.T., and Meredith, W.R. (1992) "Light interception and lint yield of narrow-row cotton."
Crop Science 32: 728-733.
References and Abbreviations 10-69
-------
Hollifield, C.D., Silvertooth, J.C., and Moser, H. (2000) "Comparison of obsolete and modern cotton cultivars for
irrigated production in Arizona." 2000 Arizona Cotton Report, University of Arizona College of Agriculture,
http://ag.arizona.edu/pubs/crops/azll70/.
Hopkinson, J.M. (1967) "Effects of night temperature on the growth of Nicotiana tabacum." Australian Journal of
Experimental Agriculture and Animal Husbandry 7: 78-82.
Huett, D.O., and Dettman, E.B. (1991) Effect of nitrogen on growth, quality and nutrient uptake of cabbages grown
in sand culture. Australian Journal of Experimental Agriculture 29: 875-81.
Huett, D.O., and Dettman, B. (1989) "Nitrogen response surface models of zucchini squash, head lettuce and
potato." Plant and Soil 134: 243-254.
IPCC (2006) 2006IPCC Guidelines for National Greenhouse Gas Inventories. The National Greenhouse Gas
Inventories Programme, The Intergovernmental Panel on Climate Change. [H.S. Eggleston, L. Buendia, K. Miwa, T.
Ngara, and K. Tanabe (eds.)]. Hayama, Kanagawa, Japan.
IPCC/UNEP/OECD/IEA (1997) Revised 1996 IPCC Guidelines for National Greenhouse Gas Inventories.
Intergovernmental Panel on Climate Change, United Nations Environment Programme, Organization for Economic
Co-Operation and Development, International Energy Agency, Paris, France.
Jacobs, J.L, Ward, G.N., and Kearney, G. (2004) "Effects of irrigation strategies and nitrogen fertilizer on turnip dry
matter yield, water use efficiency, nutritive characteristics and mineral content in western Victoria." Australian
Journal of Experimental Agriculture 44:13-26.
Jacobs, J.L, Ward, G.N., McDowell, A.M., and Kearney, G. (2002) "Effect of seedbed cultivation techniques, variety,
soil type and sowing time, on brassica dry matter yields, water use efficiency and crop nutritive characteristics in
western Victoria." Australian Journal of Experimental Agriculture 42: 945-952.
Jacobs, J.L, Ward, G.N., McDowell, A.M., and Kearney, G.A. (2001) "A survey on the effect of establishment
techniques, crop management, moisture availability and soil type on turnip dry matter yields and nutritive
characteristics in western Victoria." Australian Journal of Experimental Agriculture 41: 743-751.
Kage, H., Alt, C., and Stutzel, H. (2003) "Aspects of nitrogen use efficiency of cauliflower II. Productivity and
nitrogen partitioning as influenced by N supply." Journal of Agricultural Science 141: 17-29.
Kumar, A., Singh, D.P., and Singh, P. (1994) "Influence of water stress on photosynthesis, transpiration, water-use
efficiency and yield of Brassica juncea L" Field Crops Research 37: 95-101.
LANDFIRE (2008) Existing Vegetation Type Layer, LANDFIRE 1.1.0, U.S. Department of the Interior, Geological
Survey. Accessed 28 October 2010 at http://landfire.cr.usgs.gov/viewer/.
MacLeod, LB., Gupta, U.C., and Cutcliffe, J.A. (1971) "Effect of N, P, and K on root yield and nutrient levels in the
leaves and roots of rutabagas grown in a greenhouse." Plant and Soil 35: 281-288.
Mahrani, A., and Aharonov, B. (1964) "Rate of nitrogen absorption and dry matter production by upland cotton
grown under irrigation." Israel J. Agric. Res. 14: 3-9.
Marcussi, F.F.N., Boas, R.L.V., de Godoy, L.J.G., and Goto, R. (2004) "Macronutrient accumulation and partitioning
in fertigated sweet pepper plants." Sci. Agric. (Piracicaba, Braz.) 61: 62-68.
McCarty, J.L. (2011) "Remote Sensing-Based Estimates of Annual and Seasonal Emissions from Crop Residue
Burning in the Contiguous United States." Journal of the Air & Waste Management Association, 61:1, 22-34, DOI:
10.3155/1047-3289.61.1.22.
McCarty, J.L. (2010) Agricultural Residue Burning in the Contiguous United States by Crop Type and State.
Geographic Information Systems (GIS) Data provided to the EPA Climate Change Division by George Pouliot,
Atmospheric Modeling and Analysis Division, EPA. Dr. McCarty's research was supported by the NRI Air Quality
Program of the Cooperative State Research, Education, and Extension Service, USDA, under Agreement No.
20063511216669 and the NASA Earth System Science Fellowship.
10-70 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
McCarty, J.L. (2009) Seasonal and Interannual Variability of Emissions from Crop Residue Burning in the Contiguous
United States. Dissertation. University of Maryland, College Park.
McPharlin, I.R., Aylmore, P.M., and Jeffery, R.C. (1992) "Response of carrots (Daucus carota L.) to applied
phosphorus and phosphorus leaching on a Karrakatta sand, under two irrigation regimes." Australian Journal of
Experimental Agriculture 32:225-232.
Mondino, M.H., Peterlin, O.A., and Garay, F. (2004) "Response of late-planted cotton to the application of growth
regulator (chlorocholine chloride, CYCOCEL75)." Expl Agric. 40: 381-387.
Moustakas, N.K., and Ntzanis, H. (2005) "Dry matter accumulation and nutrient uptake in flue-cured tobacco
(Nicotiana tabacum L.)." Field Crops Research 94:1-13.
Peach, L., Benjamin, L.R., and Mead, A. (2000) "Effects on the growth of carrots (Daucus carota L.), cabbage
(Brassica oleracea var. capitata L.) and onion (Allium cepa L) of restricting the ability of the plants to intercept
resources." Journal of Experimental Botany 51: 605-615.
Pettigrew, W.T., and Meredith, W.R., Jr. (1997) "Dry matter production, nutrient uptake, and growth of cotton as
affected by potassium fertilization." / Plant Nutr. 20:531-548.
Pettigrew, W.T., Meredith, W.R., Jr., and Young, LD. (2005) "Potassium fertilization effects on cotton lint yield,
yield components, and reniform nematode populations." Agronomy Journal 97: 1245-1251.
PRISM Climate Group (2015) PRISM Climate Data. Oregon State University. July 24, 2015. Available online at:
http://prism.oregonstate.edu.
Reid, J.B., and English, J.M. (2000) "Potential yield in carrots (Daucus carota L.): Theory, test, and an application."
Annals of Botany 85: 593-605.
Sadras, V.O., and Wilson, LJ. (1997) "Growth analysis of cotton crops infested with spider mites: II. Partitioning of
dry matter." Crop Science 37: 492-497.
Scholberg, J., McNeal, B.L., Jones, J.W., Boote, K.J., Stanley, C.D., and Obreza, T.A. (2000a) "Growth and canopy
characteristics of field-grown tomato." Agronomy Journal 92: 152-159.
Scholberg, J., McNeal, B.L., Boote, K.J., Jones, J.W., Locasio, S.J., and Olson, S.M. (2000b) "Nitrogen stress effects on
growth and nitrogen accumulation by field-grown tomato." Agronomy Journal 92:159-167.
Singels, A. and Bezuidenhout, C.N. (2002) "A new method of simulating dry matter partitioning in the Canegro
sugarcane model." Field Crops Research 78: 151 - 164.
Sitompul, S.M., Hairiah, K., Cadisch, G., and Van Noordwuk, M. (2000) "Dynamics of density fractions of macro-
organic matter after forest conversion to sugarcane and woodlots, accounted for in a modified Century model."
Netherlands Journal of Agricultural Science 48: 61-73.
Stirling, G.R., Blair, B.L, Whittle, P.J.L., and Garside, A.L. (1999) "Lesion nematode (Pratylenchus zeae) is a
component of the yield decline complex of sugarcane." In: Magarey, R.C. (Ed.), Proceedings of the First
Australasian Soilborne Disease Symposium. Bureau of Sugar Experiment Stations, Brisbane, pp. 15-16.
Tan, D.K.Y., Wearing, A.H., Rickert, K.G., and Birch, C.J. (1999) "Broccoli yield and quality can be determined by
cultivar and temperature but not photoperiod in south-east Queensland." Australian Journal of Experimental
Agriculture 39: 901-909.
Tadesse, T., Nichols, M.A., and Fisher, K.J. (1999) Nutrient conductivity effects on sweet pepper plants grown using
a nutrient film technique. 1. Yield and fruit quality. New Zealand Journal of Crop and Horticultural Science, 27: 229-
237.
Torbert, H.A., and Reeves, D.W. (1994) "Fertilizer nitrogen requirements for cotton production as affected by
tillage and traffic." Soil Sci. Soc. Am. J. 58:1416-1423.
References and Abbreviations 10-71
-------
USDA-NASS (2021) Published crop data layer. Available at https://nassgeodata.gmu.edu/CropScape/, Accessed July
2021, USDA-NASS, Washington, DC.
USDA-NASS (2019) Quick Stats: U.S. & All States Data; Crops; Production and Area Harvested; 1990 - 2018. National
Agricultural Statistics Service, U.S. Department of Agriculture. Washington, D.C. U.S. Department of Agriculture,
National Agricultural Statistics Service. Washington, D.C., Available online at: http://quickstats.nass.usda.gov/.
USDA-NRCS (2020) Summary Report: 2017 National Resources Inventory. Natural Resources Conservation Service,
Washington, DC, and Center for Survey Statistics and Methodology, Iowa State University, Ames, Iowa.
https://www.nrcs.usda.gov/wps/portal/nrcs/main/national/technical/nra/nri/results/.
USDA-NRCS (2018) Summary Report: 2015 National Resources Inventory, Natural Resources Conservation Service,
Washington, D.C., and Center for Survey Statistics and Methodology, Iowa State University, Ames, Iowa.
https://www.nrcs.usda.gov/lnternet/FSE DOCUMENTS/nrcseprd 1422028.pdf.
Valantin, M., Gary, C., Vaissiere, B.E., and Frossard, J.S. (1999) "Effect of fruit load on partitioning of dry matter and
energy in cantaloupe (Cucumis melo L)." Annals of Botany 84:173-181.
Wallach, D., Marani, A., and Kletter, E. (1978) 'The relation of cotton crop growth and development to final yield."
Field Crops Research 1: 283-294.
Wells, R., and Meredith, W.R., Jr. (1984) "Comparative growth of obsolete and modern cultivars. I. Vegetative dry
matter partitioning." Crop Science 24: 858-872.4.
Wiedenfels, R.P. (2000) "Effects of irrigation and N fertilizer application on sugarcane yield and quality." Field Crops
Research 43:101-108.
Land Use, Land-Use Change, and Forestry
IPCC (2006) 2006IPCC Guidelines for National Greenhouse Gas Inventories. The National Greenhouse Gas
Inventories Programme, The Intergovernmental Panel on Climate Change. [H.S. Eggleston, L. Buendia, K. Miwa, T.
Ngara, and K. Tanabe (eds.)]. Hayama, Kanagawa, Japan.
UNFCCC (2014) Report of the Conference of the Parties on its nineteenth session, held in Warsaw from 11 to 23
November 2013. United Nations Framework Convention on Climate Change, Warsaw. (FCCC/CP/2013/10/Add.3).
January 31, 2014. Available online at: http://unfccc.int/resource/docs/2013/copl9/eng/10a03.pdf.
Representation of the U.S. Land Base
Alaska Department of Natural Resources (2006) Alaska Infrastructure 1:63,360. Available online at:
http://dnr.alaska.gov/SpatialUtility/SUC?cmd=extract&layerid=75.
Alaska Interagency Fire Management Council (1998) Alaska Interagency Wildland Fire Management Plan. Available
online at: http://agdc.usgs.gov/data/blm/fire/index.html.
Alaska Oil and Gas Conservation Commission (2009) Oil and Gas Information System. Available online at:
http://doa.alaska.gov/ogc/publicdb.html.
Dewitz, J., 2023, National Land Cover Database (NLCD) 2021 Products: U.S. Geological Survey data release,
https://doi.org/10.5066/P9JZ7AQ3.
EIA (2011) Coal Production and Preparation Report Shapefile. Available online at: http://www.eia.gov/state/notes-
sources.cfmtfmaps.
ESRI (2008) ESRI Data & Maps. Redlands, CA: Environmental Systems Research Institute. [CD-ROM],
10-72 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
Fry, J., Xian, G., Jin, S., Dewitz, J., Homer, C., Yang, L, Barnes, C., Herold, N., and J. Wickham. (2011) Completion of
the 2006 National Land Cover Database for the Conterminous United States, PE&RS, Vol. 77(9):858-864.
Homer, C., J. Dewitz, J. Fry, M. Coan, N. Hossain, C. Larson, N. Herold, A. McKerrow, J.N. VanDriel and J. Wickham.
(2007) Completion of the 2001 National Land Cover Database for the Conterminous United States,
Photogrammetric Engineering and Remote Sensing, Vol. 73, No. 4, pp 337-341.
Homer, C.G., Dewitz, J.A., Yang, L., Jin, S., Danielson, P., Xian, G., Coulston, J., Herold, N.D., Wickham, J.D., and
Megown, K. (2015) Completion of the 2011 National Land Cover Database for the conterminous United States-
Representing a decade of land cover change information. Photogrammetric Engineering and Remote Sensing, v.
81, no. 5, p. 345-354.
IPCC (2019) 2019 Refinement to the 2006IPCC Guidelines for National Greenhouse Gas Inventories, Calvo Buendia,
E., Tanabe, K., Kranjc, A., Baasansuren, J., Fukuda, M., Ngarize S., Osako, A., Pyrozhenko, Y., Shermanau, P. and
Federici, S. (eds). Published: IPCC, Switzerland.
IPCC (2014) 2013 Supplement to the 2006 IPCC Guidelines for National Greenhouse Gas Inventories: Wetlands.
Hiraishi, T., Krug, T., Tanabe, K., Srivastava, N., Baasansuren, J., Fukuda, M. and Troxler, T.G. (eds.). Published: IPCC,
Switzerland.
IPCC (2010) Revisiting the use of managed land as a proxy for estimating national anthropogenic emissions and
removals. [Eggleston HS, Srivastava N, Tanabe K, Baasansuren J, (eds.)]. Institute for Global Environmental Studies,
Intergovernmental Panel on Climate Change, Hayama, Kanagawa, Japan.
IPCC (2006) 2006 IPCC Guidelines for National Greenhouse Gas Inventories. The National Greenhouse Gas
Inventories Programme, The Intergovernmental Panel on Climate Change. [H.S. Eggleston, L. Buendia, K. Miwa, T.
Ngara, and K. Tanabe (eds.)]. Hayama, Kanagawa, Japan.
Jin, S., L. Yang, P. Danielson, C. Homer, J. Fry, and G. Xian. (2013) A comprehensive change detection method for
updating the National Land Cover Database to circa 2011. Remote Sensing of Environment, 132: 159-175.
Nelson, M.D., Riitters, K.H., Coulston, J.W., Domke, G.M., Greenfield, E.J., Langner, L.L., Nowak, D.J., O'Dea, C.B.,
Oswalt, S.N., Reeves, M.C. and Wear, D.N. (2020) Defining the United States land base: a technical document
supporting the USDA Forest Service 2020 RPA assessment. Gen. Tech. Rep. NRS-191., 191, pp.1-70.
NOAA Coastal Change Analysis Program (C-CAP) Regional Land Cover Database. Data collected 1995-present
Charleston, SC: National Oceanic and Atmospheric Administration (NOAA) Coastal Services Center. Data accessed
at: www.csc.noaa.gov/landcover.
Nusser, S.M. and J.J. Goebel (1997) "The national resources inventory: a long-term multi-resource monitoring
programme." Environmental and Ecological Statistics 4:181-204.
Ogle, S.M., G. Domke, W.A. Kurz, M.T. Rocha, T. Huffman, A. Swan, J.E. Smith, C. Woodall, T. Krug (2018)
Delineating managed land for reporting greenhouse gas emissions and removals to the United Nations Framework
Convention on Climate Change. Carbon Balance and Management 13:9.
U.S. Census Bureau (2010) Topological^ Integrated Geographic Encoding and Referencing (TIGER) system
shapefiles. U.S. Census Bureau, Washington, D.C. Available online at: http://www.census.gov/geo/www/tiger.
U.S. Department of Agriculture (2015) County Data - Livestock, 1990-2014. U.S. Department of Agriculture,
National Agriculture Statistics Service, Washington, D.C.
U.S. Department of Agriculture, Forest Service. (2023) Timber Product Output (TPO) Reports.
https://www.fia.fs.usda.gov/program-features/tpo/
U.S. Department of Interior (2005) Federal Lands of the United States. National Atlas of the United States, U.S.
Department of the Interior, Washington D.C. Available online at:
http://nationalatlas.gov/atlasftp.html?openChapters=chpbound#chpbound.
References and Abbreviations 10-73
-------
United States Geological Survey (USGS), Gap Analysis Program (2012) Protected Areas Database of the United
States (PADUS), version 1.3 Combined Feature Class. November 2012.
USGS (2012) Alaska Resource Data File. Available online at: http://ardf.wr.usgs.gov/.
USGS (2005) Active Mines and Mineral Processing Plants in the United States in 2003. U.S. Geological Survey,
Reston, VA.
Yang, L, Jin, S., Danielson, P., Homer, C., Gass, L., Bender, S. M., Case, A., Costello, C., Dewitz, J., Fry, J., Funk, M.,
Granneman, B., Liknes, G. C., Rigge, M. & Xian, G. (2018) A new generation of the United States National Land
Cover Database: Requirements, research priorities, design, and implementation strategies. ISPRS Journal of
Photogrammetry and Remote Sensing 146:108-123.
Forest Land Remaining Forest Land: Changes in Forest Carbon
Stocks
AF&PA (2006a and earlier) Statistical roundup. (Monthly). Washington, D.C. American Forest & Paper Association.
AF&PA (2006b and earlier) Statistics of paper, paperboard and wood pulp. Washington, D.C. American Forest &
Paper Association.
AF&PA (2021) 2020 Statistics - Paper Industry - Annual Summary Data Through 2020. Washington, D.C.: American
Forest and Paper Association, 54 p.
AF&PA (2023) Capacity & Fiber Consumption Survey - Paper Industry - 63rd Annual 2022-2023. Washington, D.C.:
American Forest and Paper Association, 44 p.
Amichev, B.Y. and J.M. Galbraith (2004) "A Revised Methodology for Estimation of Forest Soil Carbon from Spatial
Soils and Forest Inventory Data Sets." Environmental Management 33(Suppl. 1):S74-S86.
Bechtold, W.A.; Patterson, P.L (2005) The enhanced forest inventory and analysis program—national sampling
design and estimation procedures. Gen. Tech. Rep. SRS-80. Asheville, NC: U.S. Department of Agriculture Forest
Service, Southern Research Station. 85 p.
Birdsey, R. (1996) "Carbon Storage for Major Forest Types and Regions in the Conterminous United States." In R.N.
Sampson and D. Hair, (eds.). Forest and Global Change, Volume 2: Forest Management Opportunities for
Mitigating Carbon Emissions. American Forests. Washington, D.C., 1-26 and 261-379 (appendices 262 and 263).
Coulston, J.W., Wear, D.N., and Vose, J.M. (2015) Complex forest dynamics indicate potential for slowing carbon
accumulation in the southeastern United States. Scientific Reports. 5: 8002.
Deenik J, McClellan AT (2007) Soils of Hawai'i. Soil and Crop Management, SCM-20. College of Tropical Agriculture
and Human Resources, University of Hawai'i at Manoa, Honolulu.
Domke, G.M., Perry, C.H., Walters, B.F., Woodall, C.W., and Smith, J.E. (2016) A framework for estimating litter
carbon stocks in forests of the United States. Science of the Total Environment 557-558: 469-478.
Domke, G.M., Perry, C.H., Walters, B.F., Woodall, C.W., Nave, L., Swanston, C. (2017) Toward inventory-based
estimates of soil organic carbon in forests of the United States. Ecological Applications. 27(4), 1223-1235.
Domke, G.M., Walters, B.F., Smith, J.E., Woodall, C.W. (2022) Chapter 6: FIA Carbon Attributes. In Sampling and
estimation documentation for the Enhanced Forest Inventory and Analysis Program: 2022. Westfall, J.A., Coulston,
J.W., Moisen, G.G., Andersen, H.-E., eds. Gen. Tech. Rep. NRS-GTR-207, Madison, Wl: U.S. Department of
Agriculture, Forest Service, Northern Research Station. 129 p. https://doi.org/10.2737/NRS-GTR-207.
Domke, G.M., Woodall, C.W., Smith, J.E., Westfall, J.A., McRoberts, R.E. (2012) Consequences of alternative tree-
level biomass estimation procedures on U.S. forest carbon stock estimates. Forest Ecology and Management. 270:
108-116.
10-74 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
Domke, G.M., Woodall, C.W., Walters, B.F., Smith, J.E. (2013) From models to measurements: comparing down
dead wood carbon stock estimates in the U.S. forest inventory. PLoS ONE 8(3): e59949.
EPA (2006) Municipal solid waste in the United States: 2005 Facts and figures. Office of Solid Waste, U.S.
Environmental Protection Agency. Washington, D.C. (5306P) EPA530-R-06-011. Available online at:
http://www.epa.gov/msw/msw99.htm.
FAO. Forestry Production and Trade. License: CC BY-NC-SA 3.0 IGO. Extracted from:
https://www.fao.Org/faostat/en/#data/FQ. Data of Access: 13-09-2023.
Frayer, W.E., and G.M. Furnival (1999) "Forest Survey Sampling Designs: A History." Journal of Forestry 97(12): 4-
10.
Freed, R. (2004) Open-dump and Landfill timeline spreadsheet (unpublished). ICF International. Washington, D.C.
Hair, D. (1958) "Historical forestry statistics of the United States." Statistical Bull. 228. U.S. Department of
Agriculture Forest Service, Washington, D.C.
Hair. D. and A.H. Ulrich (1963) The Demand and price situation for forest products - 1963. U.S. Department of
Agriculture Forest Service, Misc Publication No. 953. Washington, D.C.
Harmon, M.E., C.W. Woodall, B. Fasth, J. Sexton, M. Yatkov. (2011) Differences between standing and downed
dead tree wood density reduction factors: A comparison across decay classes and tree species. Res. Paper. NRS-15.
Newtown Square, PA: U.S. Department of Agriculture, Forest Service, Northern Research Station. 40 p.
Howard, J. L. and Liang, S. (2019) U.S. timber production, trade, consumption, and price statistics 1965 to 2017.
Res. Pap. FPL-RP-701. Madison, Wl: USDA, Forest Service, Forest Products Laboratory.
Howard, J. L. and Jones, K.C. (2016) U.S. timber production, trade, consumption, and price statistics 1965 to 2013.
Res. Pap. FPL-RP-679. Madison, Wl: USDA, Forest Service, Forest Products Laboratory.
Howard, J. L. (2007) U.S. timber production, trade, consumption, and price statistics 1965 to 2005. Res. Pap. FPL-
RP-637. Madison, Wl: USDA, Forest Service, Forest Products Laboratory.
Howard, J. L. (2003) U.S. timber production, trade, consumption, and price statistics 1965 to 2002. Res. Pap. FPL-
RP-615. Madison, Wl: USDA, Forest Service, Forest Products Laboratory. Available online at:
http://www.fpl.fs.fed.us/documnts/fplrp/fplrp615/fplrp615.pdf.
IPCC (2014) 2013 Supplement to the 2006IPCC Guidelines for National Greenhouse Gas Inventories: Wetlands.
[Hiraishi, T., Krug, T., Tanabe, K., Srivastava, N., Baasansuren, J., Fukuda, M., and Troxler, T.G. (eds.)]. Switzerland.
IPCC (2007) Climate Change 2007: The Physical Science Basis. Contribution of Working Group I to the Fourth
Assessment Report of the Intergovernmental Panel on Climate Change [Solomon, S., D. Qin, M. Manning, Z. Chen,
M. Marquis, K.B. Averyt, M. Tignor and H.L Miller (eds.)]. Cambridge University Press, Cambridge, United Kingdom
and New York, NY, USA, 996 pp.
IPCC (2006) 2006 IPCC Guidelines for National Greenhouse Gas Inventories. The National Greenhouse Gas
Inventories Programme, The Intergovernmental Panel on Climate Change. [H.S. Eggleston, L. Buendia, K. Miwa, T.
Ngara, and K. Tanabe (eds.)]. Hayama, Kanagawa, Japan.
Jandl, R., Rodeghiero, M., Martinez, C., Cotrufo, M. F., Bampa, F., van Wesemael, B., Harrison, R.B., Guerrini, I.A.,
deB Richter Jr., D., Rustad, L., Lorenz, K., Chabbi, A., Miglietta, F. (2014) Current status, uncertainty and future
needs in soil organic carbon monitoring. Science of the Total Environment, 468, 376-383.
Jenkins, J.C., D.C. Chojnacky, L.S. Heath, and R.A. Birdsey (2003) "National-scale biomass estimators for United
States tree species." Forest Science 49(l):12-35.
Johnson, K. Domke, G.M., Russell, M.B., Walters, B.F., Horn, J., Peduzzi, A., Birdsey, R., Dolan, K., Huang, W. (2017)
Estimating aboveground live understory vegetation carbon in the United States. Environmental Research Letters.
References and Abbreviations 10-75
-------
Johnson, M.G. and Kern, J.S., (2002.) Quantifying the organic carbon held in forested soils of the United States and
Puerto Rico. The potential of US forest soils to sequester and mitigate the greenhouse effect, Lewis, Boca Raton,
pp.47-72.
Nelson, M.D., Riitters, K.H., Coulston, J.W., Domke, G.M., Greenfield, E.J., Langner, L.L., Nowak, D.J., O'Dea, C.B.,
Oswalt, S.N., Reeves, M.C. and Wear, D.N. (2020) Defining the United States land base: a technical document
supporting the USDA Forest Service 2020 RPA assessment. Gen. Tech. Rep. NRS-191., 191, pp.1-70.
Ogle, S. M., G. M. Domke, W. A. Kurz, M. T. Rocha, T. Huffman, A. Swan, J. E. Smith, C. W. Woodall, and T. Krug.
(2018) Delineating managed land for reporting national greenhouse gas emissions and removals to the United
Nations framework convention on climate change. Carbon Balance and Management 13:9.
O'Neill, K.P., Amacher, M.C., Perry, C.H. (2005) Soils as an indicator of forest health: a guide to the collection,
analysis, and interpretation of soil indicator data in the Forest Inventory and Analysis program. Gen. Tech. Rep. NC-
258. St. Paul, MN: U.S. Department of Agriculture, Forest Service, North Central Research Station. 53 p.
Oswalt, S.N., Brandeis, T.J. and Woodall, C.W., (2008). Contribution of dead wood to biomass and carbon stocks in
the Caribbean: St. John, US Virgin Islands. Biotropica, 40(1), pp.20-27.
Oswalt, S.N., Smith, W.B., Miles, P.D. and Pugh, S.A. (2019) Forest resources of the United States, 2017: Atechnical
document supporting the Forest Service 2020 RPA Assessment. Gen. Tech. Rep. WO-97. Washington, DC: U.S.
Department of Agriculture, Forest Service, Washington Office., 97.
Perry, C.H., C.W. Woodall, and M. Schoeneberger (2005) Inventorying trees in agricultural landscapes: towards an
accounting of "working trees". In: "Moving Agroforestry into the Mainstream." Proc. 9th N. Am. Agroforestry
Conf., Brooks, K.N. and Folliott, P.F. (eds.). 12-15 June 2005, Rochester, MN [CD-ROM], Dept. of Forest Resources,
Univ. Minnesota, St. Paul, MN, 12 p. Available online at: http://cinram.umn.edu/afta2005/. (verified 23 Sept 2006).
Russell, M.B.; D'Amato, A.W.; Schulz, B.K.; Woodall, C.W.; Domke, G.M.; Bradford, J.B. (2014) Quantifying
understory vegetation in the U.S. Lake States: a proposed framework to inform regional forest carbon stocks.
Forestry. 87: 629-638.
Russell, M.B.; Domke, G.M.; Woodall, C.W.; D'Amato, A.W. (2015) Comparisons of allometric and climate-derived
estimates of tree coarse root carbon in forests of the United States. Carbon Balance and Management. 10: 20.
Selmants, P.C., Giardina, C.P., Jacobi, J.D., and Zhu, Zhiliang, eds. (2017) Baseline and projected future carbon
storage and carbon fluxes in ecosystems of Hawai'i: U.S. Geological Survey Professional Paper 1834,134 p.,
https://doi.org/10.3133/ppl834.
Skog, K.E. (2008) Sequestration of carbon in harvested wood products for the United States. Forest Products
Journal 58:56-72.
Smith, J.E., Domke, G.M. and Woodall, C.W. (2022) Predicting downed woody material carbon stocks in forests of
the conterminous United States. Science of the Total Environment, 803, p.150061.
Smith, J.E., L.S. Heath, and M.C. Nichols (2010) U.S. Forest Carbon Calculation Tool User's Guide: Forestland Carbon
Stocks and Net Annual Stock Change. General Technical Report NRS-13 revised, U.S. Department of Agriculture
Forest Service, Northern Research Station, 34 p.
Smith, J.E.; Heath, L.S.; Skog, K.E.; Birdsey, R.A. (2006) Methods for calculating forest ecosystem and harvested
carbon with standard estimates for forest types of the United States. Gen. Tech. Rep. NE-343. Newtown Square,
PA: U.S. Department of Agriculture, Forest Service, Northeastern Research Station. 216 p.
Smith, W. B., P. D. Miles, C. H. Perry, and S. A. Pugh (2009) Forest Resources of the United States, 2007. General
Technical Report WO-78, U.S. Department of Agriculture Forest Service, Washington Office.
Soil Survey Staff (2020a) Gridded National Soil Survey Geographic (gNATSGO) Database for the Conterminous
United States. United States Department of Agriculture, Natural Resources Conservation Service. Available online
at https://nrcs.app.box.eom/v/soils.
10-76 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
Soil Survey Staff (2020b) Gridded National Soil Survey Geographic (gNATSGO) Database for Alaska. United States
Department of Agriculture, Natural Resources Conservation Service. Available online at
https://nrcs.app.box.eom/v/soils.
Steer, Henry B. (1948) Lumber production in the United States. Misc. Pub. 669, U.S. Department of Agriculture
Forest Service. Washington, D.C.
Ulrich, Alice (1985) U.S. Timber Production, Trade, Consumption, and Price Statistics 1950-1985. Misc. Pub. 1453,
U.S. Department of Agriculture Forest Service. Washington, D.C.
Ulrich, Alice (1989) U.S. Timber Production, Trade, Consumption, and Price Statistics, 1950-1987. USDA
Miscellaneous Publication No. 1471, U.S. Department of Agriculture Forest Service. Washington, D.C., 77.
United Nations Framework Convention on Climate Change (UNFCC) (2013) Report on the individual review of the
inventory submission of the United States of America submitted in 2012. FCCC/ARR/2012/USA. 42 p.
USDA Forest Service (2022a) PNW-FIA Hawai'i Inventory Database. Last accessed: 28 September 2023.
https://www.fs.usda.gov/research/pnw/products/dataandtools/tools/pnw-fia-hawaii-inventory-database
USDA Forest Service (2022b) PNW-FIA Pacific Islands Inventory Database. Last accessed: 28 September 2023.
https://www.fs.usda.gov/research/pnw/products/dataandtools/tools/pnw-fia-pacific-islands-database
https://www.fs.usda.gov/research/pnw/products/dataandtools/tools/pnw-fia-hawaii-inventory-database
USDA Forest Service (2023a) Forest Inventory and Analysis National Program: Program Features. U.S. Department
of Agriculture Forest Service. Washington, D.C. Available online at: https://www.fia.fs.usda.gov/program-
features/index.php. Accessed 28 September 2023.
USDA Forest Service. (2023b) Forest Inventory and Analysis National Program: FIA Data Mart. U.S. Department of
Agriculture Forest Service. Washington, D.C. Available online at:
https://apps.fs.usda.gov/fia/datamart/datamart.html. Accessed on 28 September 2023.
USDA Forest Service. (2023c) Forest Inventory and Analysis National Program, FIA library: Field Guides, Methods
and Procedures. U.S. Department of Agriculture Forest Service. Washington, D.C. Available online at:
https://www.fia.fs.usda.gov/library/field-guides-methods-proc/index.php. Accessed on 28 September 2023.
USDA Forest Service (2023d) Forest Inventory and Analysis National Program, FIA library: Database
Documentation. U.S. Department of Agriculture, Forest Service, Washington Office. Available online at:
https://www.fia.fs.usda.gov/library/database-documentation/index.php. Accessed on 28 September 2023.
U.S. Census Bureau (1976) Historical Statistics of the United States, Colonial Times to 1970, Vol. 1. Washington,
D.C.
Wear, D.N., Coulston, J.W. (2015) From sink to source: Regional variation in U.S. forest carbon futures. Scientific
Reports. 5:16518.
Westfall, J.A., Coulston, J.W., Gray, A.N., Shaw, J.D., Radtke, P.J., Walker, D.M., Weiskittel, A.R., MacFarlane, D.W.,
Affleck, D.L.R., Zhao, D., Temesgen, H., Poudel, K.P., Frank, J.M., Prisley, S.P., Wang, Y., Sanchez Meador, A.J., Auty,
D., and Domke, G.M. In press . A national-scale tree volume, biomass, and carbon modeling system for the United
States. Gen. Tech. Rep. WO-104. Washington, DC: U.S. Department of Agriculture, Forest Service. 60 p.
https//doi.org/10.2737/WO-GTR-104.
Westfall, J.A., Woodall, C.W., Hatfield, M.A. (2013) A statistical power analysis of woody carbon flux from forest
inventory data. Climatic Change. 118: 919-931.
Woodall, C.W., Amacher, M.C., Bechtold, W.A., Coulston, J.W., Jovan, S., Perry, C.H., Randolph, K.C., Schulz, B.K.,
Smith, G.C., Tkacz, B., Will-Wolf, S. (2011b) "Status and future of the forest health indicators program of the United
States." Environmental Monitoring and Assessment. 177: 419-436.
References and Abbreviations 10-77
-------
Woodall, C.W., and V.J. Monleon (2008) Sampling protocol, estimation, and analysis procedures for the down
woody materials indicator of the FIA program. Gen. Tech. Rep. NRS-22. Newtown Square, PA: U.S. Department of
Agriculture, Forest Service, Northern Research Station. 68 p.
Woodall, C.W., Coulston, J.W., Domke, G.M., Walters, B.F., Wear, D.N., Smith, J.E., Anderson, H.-E., Clough, B.J.,
Cohen, W.B., Griffith, D.M., Hagan, S.C., Hanou, I.S.; Nichols, M.C., Perry, C.H., Russell, M.B., Westfall, J.A., Wilson,
B.T. (2015a) The U.S. Forest Carbon Accounting Framework: Stocks and Stock change 1990-2016. Gen. Tech. Rep.
NRS-154. Newtown Square, PA: U.S. Department of Agriculture, Forest Service, Northern Research Station. 49 pp.
Woodall, C.W., LS. Heath, G.M. Domke, and M.C. Nichols (2011a) Methods and equations for estimating
aboveground volume, biomass, and carbon for trees in the U.S. forest inventory, 2010. Gen. Tech. Rep. NRS-88.
Newtown Square, PA: U.S. Department of Agriculture, Forest Service, Northern Research Station. 30 p.
Woodall, C.W., Walters, B.F., Coulston, J.W., D'Amato, A.W., Domke, G.M., Russell, M.B., Sowers, P.A. (2015b)
Monitoring network confirms land use change is a substantial component of the forest carbon sink in the eastern
United States. Scientific Reports. 5: 17028.
Woodall, C.W., Walters, B.F., Oswalt, S.N., Domke, G.M., Toney, C., Gray, A.N. (2013) Biomass and carbon
attributes of downed woody materials in forests of the United States. Forest Ecology and Management 305: 48-59.
Zhu, Zhiliang, and McGuire, A.D., eds. (2016) Baseline and projected future carbon storage and greenhouse-gas
fluxes in ecosystems of Alaska: U.S. Geological Survey Professional Paper 1826,196 p., Available online at:
http://dx.doi.org/10.3133/ppl826.
Forest Land Remaining Forest Land: Non-C02 Emissions from
Forest Fires
Eidenshink, J., Schwind, B., Brewer, K., Zhu, Z.L, Quayle, B. and Howard, S. (2007) A project for monitoring trends
in burn severity. Fire ecology, 3(1), pp.3-21.
French, N.H.F., W.J. de Groot, L.K. Jenkins, B.M. Rogers, E.C. Alvarado, B. Amiro, B. de Jong, S. Goetz, E. Hoy, E.
Hyer, R. Keane, D. McKenzie, S.G. McNulty, B.E Law, R. Ottmar, D.R. Perez-Salicrup, J. Randerson, K.M. Robertson,
and M. Turetsky (2011) "Model comparisons for estimating carbon emissions from North American wildland fire."
Journal of Geophysical Research 116. 10.1029/2010JG001469
French, N.H.F., D. McKenzie, T. Erickson, B. Koziol, M. Billmire, K.A. Endsley, N.K.Y. Scheinerman, L. Jenkins, M.E.
Miller, R. Ottmar, and S. Prichard (2014) "Modeling regional-scale fire emissions with the Wildland Fire Emissions
Information System." Earth Interactions 18, no. 16
Giglio, L., Boschetti, L., Roy, D. P., Humber, M. L., and Justice, C. O. (2018) The Collection 6 MODIS burned area
mapping algorithm and product. Remote Sensing of Environment, 217, 72-85.
IPCC (2019) Chapter 2: Generic Methodologies Applicable to Multiple Land-Use Categories. Refinement to the
2006 IPCC Guidelines for National Greenhouse Gas Inventories, Calvo Buendia, E., Tanabe, K., Kranjc, A.,
Baasansuren, J., Fukuda, M., Ngarize, S., Osako, A., Pyrozhenko, Y., Shermanau, P. and Federici, S. (eds). Published:
IPCC, Switzerland.
IPCC (2013) Climate Change 2013: The Physical Science Basis. Contribution of Working Group I to the Fifth
Assessment Report of the Intergovernmental Panel on Climate Change [Stocker, T.F., D. Qin, G.-K. Plattner, M.
Tignor, S.K. Allen, J. Boschung, A. Nauels, Y. Xia, V. Bex and P.M. Midgley (eds.)]. Cambridge University Press,
Cambridge, United Kingdom and New York, NY, USA, 1535 pp.
IPCC (2006) 2006 IPCC Guidelines for National Greenhouse Gas Inventories. The National Greenhouse Gas
Inventories Programme, The Intergovernmental Panel on Climate Change. [H.S. Eggleston, L. Buendia, K. Miwa, T.
Ngara, and K. Tanabe (eds.)]. Hayama, Kanagawa, Japan.
10-78 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
Larkin, N. K., S. Raffuse, and T. T. Strand (2014) Wildland fire emissions, carbon, and climate: U.S. emissions
inventories. For. Ecol. Manage. 317:61-69. doi:10.1016/j.foreco.2013.09.012.
MTBS Data Access: Fire Level Geospatial Data (August 2023) MTBS Project (USDA Forest Service/U.S. Geological
Survey). Available online at: http://mtbs.gov/direct-download. Accessed on 15 August 2023.
Ogle, S. M., G. M. Domke, W. A. Kurz, M. T. Rocha, T. Huffman, A. Swan, J. E. Smith, C. W. Woodall, and T. Krug.
(2018) Delineating managed land for reporting national greenhouse gas emissions and removals to the United
Nations framework convention on climate change. Carbon Balance and Management 13:9.
WFEIS (2023) Wildland Fire Emissions Inventory System, home page, https://wfeis.mtri.org/.
WFIGS (2023) WFIGS, The Wildland Fire Interagency Geospatial Service Interagency Fire Perimeters. National
Interagency Fire Center. https://data-nifc.opendata.arcgis.com/datasets/nifc::wfigs-interagency-fire-
perimeters/explore?location=0.000000%2C0.000000%2C1.82.
Forest Land Remaining Forest Land: N20 Emissions from Soils
Albaugh, T.J., Allen, H.L., Fox, T.R. (2007) Historical Patterns of Forest Fertilization in the Southeastern United
States from 1969 to 2004. Southern Journal of Applied Forestry, 31,129-137(9).
Binkley, D. (2004) Email correspondence regarding the 95 percent confidence interval for area estimates of
southern pine plantations receiving N fertilizer (±20%) and the rate applied for areas receiving N fertilizer (100 to
200 pounds/acre). Dan Binkley, Department of Forest, Rangeland, and Watershed Stewardship, Colorado State
University and Stephen Del Grosso, Natural Resource Ecology Laboratory, Colorado State University. September
19, 2004.
Binkley, D., R. Carter, and H.L Allen (1995) Nitrogen Fertilization Practices in Forestry. In: Nitrogen Fertilization in
the Environment, P.E. Bacon (ed.), Marcel Decker, Inc., New York.
Briggs, D. (2007) Management Practices on Pacific Northwest West-Side Industrial Forest Lands, 1991-2005: With
Projections to 2010. Stand Management Cooperative, SMC Working Paper Number 6, College of Forest Resources,
University of Washington, Seattle.
Fox, T.R., H. L. Allen, T.J. Albaugh, R. Rubilar, and C.A. Carlson (2007) Tree Nutrition and Forest Fertilization of Pine
Plantations in the Southern United States. Southern Journal of Applied Forestry, 31, 5-11.
IPCC (2006) 2006 IPCC Guidelines for National Greenhouse Gas Inventories. The National Greenhouse Gas
Inventories Programme, The Intergovernmental Panel on Climate Change. [H.S. Eggleston, L. Buendia, K. Miwa, T.
Ngara, and K. Tanabe (eds.)]. Hayama, Kanagawa, Japan.
USDA Forest Service (2001) U.S. Forest Facts and Historical Trends. FS-696. U.S. Department of Agriculture Forest
Service, Washington, D.C. Available online at: http://www.fia.fs.fed.us/library/ForestFactslVletric.pdf.
Forest Land Remaining Forest Land: Drained Organic Soils
IPCC (2014) 2013 Supplement to the 2006 IPCC Guidelines for National Greenhouse Gas Inventories: Wetlands,
Hiraishi, T., Krug, T., Tanabe, K., Srivastava, N., Baasansuren, J., Fukuda, M. and Troxler, T.G. (eds.). Published: IPCC,
Switzerland.
IPCC (2013) Climate Change 2013: The Physical Science Basis. Contribution of Working Group I to the Fifth
Assessment Report of the Intergovernmental Panel on Climate Change [Stocker, T.F., D. Qin, G.-K. Plattner, M.
Tignor, S.K. Allen, J. Boschung, A. Nauels, Y. Xia, V. Bex and P.M. Midgley (eds.)]. Cambridge University Press,
Cambridge, United Kingdom and New York, NY, USA, 1535 pp.
References and Abbreviations 10-79
-------
IPCC (2006) 2006IPCC Guidelines for National Greenhouse Gas Inventories. The National Greenhouse Gas
Inventories Programme, The Intergovernmental Panel on Climate Change, H.S. Eggleston, L. Buendia, K. Miwa, T.
Ngara, and K. Tanabe (eds.). Hayama, Kanagawa, Japan.
STATSG02 (2016) Soil Survey Staff, Natural Resources Conservation Service, United States Department of
Agriculture. U.S. General Soil Map (STATSG02). Available online at https://sdmdataaccess.sc.egov.usda.gov.
Accessed 10 November 2016.
USDA Forest Service (2023) Forest Inventory and Analysis National Program: FIA Data Mart. U.S. Department of
Agriculture Forest Service. Washington, DC; 2015. Available online at
https://apps.fs.usda.gov/fia/datamart/datamart.html. Accessed 30 March 2022.
USDA Forest Service (2022) Forest Inventory and Analysis National Program: FIA Data Mart. U.S. Department of
Agriculture Forest Service. Washington, DC; 2015. Available online at
https://apps.fs.usda.gov/fia/datamart/datamart.html. Accessed 30 March 2022.
Land Converted to Forest Land
Birdsey, R. (1996) "Carbon Storage for Major Forest Types and Regions in the Conterminous United States." In R.N.
Sampson and D. Hair, (eds.). Forest and Global Change, Volume 2: Forest Management Opportunities for
Mitigating Carbon Emissions. American Forests. Washington, D.C., 1-26 and 261-379 (appendices 262 and 263).
Brockwell, Peter J., and Richard A. Davis (2016) Introduction to time series and forecasting. Springer.
Domke, G.M., Perry, C.H., Walters, B.F., Woodall, C.W., and Smith, J.E. (2016) A framework for estimating litter
carbon stocks in forests of the United States. Science of the Total Environment 557-558: 469-478.
Domke, G.M., Woodall, C.W., Walters, B.F., Smith, J.E. (2013) From models to measurements: comparing down
dead wood carbon stock estimates in the U.S. forest inventory. PLoS ONE 8(3): e59949.
Harmon, M.E., C.W. Woodall, B. Fasth, J. Sexton, M. Yatkov. (2011) Differences between standing and downed
dead tree wood density reduction factors: A comparison across decay classes and tree species. Res. Paper. NRS-15.
Newtown Square, PA: U.S. Department of Agriculture, Forest Service, Northern Research Station. 40 p.
IPCC (2006) 2006 IPCC Guidelines for National Greenhouse Gas Inventories. The National Greenhouse Gas
Inventories Programme, The Intergovernmental Panel on Climate Change. [H.S. Eggleston, L. Buendia, K. Miwa, T.
Ngara, and K. Tanabe (eds.)]. Hayama, Kanagawa, Japan.
Jenkins, J.C., D.C. Chojnacky, L.S. Heath, and R.A. Birdsey (2003) "National-scale biomass estimators for United
States tree species." Forest Science 49(l):12-35.0gle, S.M., M.D. Eve, F.J. Breidt, and K. Paustian (2003)
"Uncertainty in estimating land use and management impacts on soil organic carbon storage for U.S.
agroecosystems between 1982 and 1997." Global Change Biology 9:1521-1542.
Ogle, S.M., F.J. Breidt, and K. Paustian. (2006) "Bias and variance in model results due to spatial scaling of
measurements for parameterization in regional assessments." Global Change Biology 12:516-523.
Smith, J.E.; Heath, L.S.; Skog, K.E.; Birdsey, R.A. (2006) Methods for calculating forest ecosystem and harvested
carbon with standard estimates for forest types of the United States. Gen. Tech. Rep. NE-343. Newtown Square,
PA: U.S. Department of Agriculture, Forest Service, Northeastern Research Station. 216 p.
USDA Forest Service (2023a) Forest Inventory and Analysis National Program: FIA Data Mart. U.S. Department of
Agriculture Forest Service. Washington, D.C. Available online at:
https://apps.fs.usda.gov/fia/datamart/datamart.html. Accessed on 29 September 2023.
USDA Forest Service (2023b) Forest Inventory and Analysis National Program, FIA library: Field Guides, Methods
and Procedures. U.S. Department of Agriculture Forest Service. Washington, D.C. Available online at:
https://www.fia.fs.usda.gov/library/field-guides-methods-proc/index.php. Accessed on 29 September 2023.
10-80 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
USDA-NRCS (2018) Summary Report: 2015 National Resources Inventory, Natural Resources Conservation Service,
Washington, D.C., and Center for Survey Statistics and Methodology, Iowa State University, Ames, Iowa.
https://www.nrcs.usda.gov/lnternet/FSE DOCUMENTS/nrcseprd 1422028.pdf.
USDA-NRCS (1997) "National Soil Survey Laboratory Characterization Data," Digital Data, Natural Resources
Conservation Service, U.S. Department of Agriculture. Lincoln, NE.
Westfall, J.A., Coulston, J.W., Gray, A.N., Shaw, J.D., Radtke, P.J., Walker, D.M., Weiskittel, A.R., MacFarlane, D.W.,
Affleck, D.L.R., Zhao, D., Temesgen, H., Poudel, K.P., Frank, J.M., Prisley, S.P., Wang, Y., Sanchez Meador, A.J., Auty,
D., and Domke, G.M. 2023. A national-scale tree volume, biomass, and carbon modeling system for the United
States. Gen. Tech. Rep. WO-104.
Woodall, C.W., LS. Heath, G.M. Domke, and M.C. Nichols (2011a) Methods and equations for estimating
aboveground volume, biomass, and carbon for trees in the U.S. forest inventory, 2010. Gen. Tech. Rep. NRS-88.
Newtown Square, PA: U.S. Department of Agriculture, Forest Service, Northern Research Station. 30 p.
Woodall, C.W., and V.J. Monleon (2008) Sampling protocol, estimation, and analysis procedures for the down
woody materials indicator of the FIA program. Gen. Tech. Rep. NRS-22. Newtown Square, PA: U.S. Department of
Agriculture, Forest Service, Northern Research Station. 68 p.
Woodall, C.W., Walters, B.F., Coulston, J.W., D'Amato, A.W., Domke, G.M., Russell, M.B., Sowers, P.A. (2015b)
Monitoring network confirms land use change is a substantial component of the forest carbon sink in the eastern
United States. Scientific Reports. 5: 17028.
Woodall, C.W., Walters, B.F., Oswalt, S.N., Domke, G.M., Toney, C., Gray, A.N. (2013) Biomass and carbon
attributes of downed woody materials in forests of the United States. Forest Ecology and Management 305: 48-59.
Yang, L, Jin, S., Danielson, P., Homer, C., Gass, L., Bender, S. M., Case, A., Costello, C., Dewitz, J., Fry, J., Funk, M.,
Granneman, B., Liknes, G. C., Rigge, M. & Xian, G. (2018) A new generation of the United States National Land
Cover Database: Requirements, research priorities, design, and implementation strategies. ISPRS Journal of
Photogrammetry and Remote Sensing 146:108-123.
Land Converted to Forest Land
Birdsey, R. (1996) "Carbon Storage for Major Forest Types and Regions in the Conterminous United States." In R.N.
Sampson and D. Hair, (eds.). Forest and Global Change, Volume 2: Forest Management Opportunities for
Mitigating Carbon Emissions. American Forests. Washington, D.C., 1-26 and 261-379 (appendices 262 and 263).
Brockwell, Peter J., and Richard A. Davis (2016) Introduction to time series and forecasting. Springer.
Domke, G.M., Perry, C.H., Walters, B.F., Woodall, C.W., and Smith, J.E. (2016) A framework for estimating litter
carbon stocks in forests of the United States. Science of the Total Environment 557-558: 469-478.
Domke, G.M., Woodall, C.W., Walters, B.F., Smith, J.E. (2013) From models to measurements: comparing down
dead wood carbon stock estimates in the U.S. forest inventory. PLoS ONE 8(3): e59949.
Harmon, M.E., C.W. Woodall, B. Fasth, J. Sexton, M. Yatkov. (2011) Differences between standing and downed
dead tree wood density reduction factors: A comparison across decay classes and tree species. Res. Paper. NRS-15.
Newtown Square, PA: U.S. Department of Agriculture, Forest Service, Northern Research Station. 40 p.
IPCC (2006) 2006IPCC Guidelines for National Greenhouse Gas Inventories. The National Greenhouse Gas
Inventories Programme, The Intergovernmental Panel on Climate Change. [H.S. Eggleston, L. Buendia, K. Miwa, T.
Ngara, and K. Tanabe (eds.)]. Hayama, Kanagawa, Japan.
Jenkins, J.C., D.C. Chojnacky, L.S. Heath, and R.A. Birdsey (2003) "National-scale biomass estimators for United
States tree species." Forest Science 49(l):12-35.0gle, S.M., M.D. Eve, F.J. Breidt, and K. Paustian (2003)
"Uncertainty in estimating land use and management impacts on soil organic carbon storage for U.S.
agroecosystems between 1982 and 1997." Global Change Biology 9:1521-1542.
References and Abbreviations 10-81
-------
Ogle, S.M., F.J. Breidt, and K. Paustian. (2006) "Bias and variance in model results due to spatial scaling of
measurements for parameterization in regional assessments." Global Change Biology 12:516-523.
Smith, J.E.; Heath, L.S.; Skog, K.E.; Birdsey, R.A. (2006) Methods for calculating forest ecosystem and harvested
carbon with standard estimates for forest types of the United States. Gen. Tech. Rep. NE-343. Newtown Square,
PA: U.S. Department of Agriculture, Forest Service, Northeastern Research Station. 216 p.
USDA Forest Service (2023b) Forest Inventory and Analysis National Program: FIA Data Mart. U.S. Department of
Agriculture Forest Service. Washington, D.C. Available online at:
https://apps.fs.usda.gov/fia/datamart/datamart.html. Accessed on 29 September 2023.
USDA Forest Service (2023c) Forest Inventory and Analysis National Program, FIA library: Field Guides, Methods
and Procedures. U.S. Department of Agriculture Forest Service. Washington, D.C. Available online at:
https://www.fia.fs.usda.gov/library/field-guides-methods-proc/index.php. Accessed on 29 September 2023.
USDA-NRCS (2020) Summary Report: 2017 National Resources Inventory. Natural Resources Conservation Service,
Washington, DC, and Center for Survey Statistics and Methodology, Iowa State University, Ames, Iowa.
https://www.nrcs.usda.gov/wps/portal/nrcs/main/national/technical/nra/nri/results/.
USDA-NRCS (2018) Summary Report: 2015 National Resources Inventory, Natural Resources Conservation Service,
Washington, D.C., and Center for Survey Statistics and Methodology, Iowa State University, Ames, Iowa.
https://www.nrcs.usda.gov/lnternet/FSE DOCUMENTS/nrcseprd 1422028.pdf.
USDA-NRCS (1997) "National Soil Survey Laboratory Characterization Data," Digital Data, Natural Resources
Conservation Service, U.S. Department of Agriculture. Lincoln, NE.
Westfall, J.A., Coulston, J.W., Gray, A.N., Shaw, J.D., Radtke, P.J., Walker, D.M., Weiskittel, A.R., MacFarlane, D.W.,
Affleck, D.L.R., Zhao, D., Temesgen, H., Poudel, K.P., Frank, J.M., Prisley, S.P., Wang, Y., Sanchez Meador, A.J., Auty,
D., and Domke, G.M. 2023. A national-scale tree volume, biomass, and carbon modeling system for the United
States. Gen. Tech. Rep. WO-104.
Woodall, C.W., LS. Heath, G.M. Domke, and M.C. Nichols (2011a) Methods and equations for estimating
aboveground volume, biomass, and carbon for trees in the U.S. forest inventory, 2010. Gen. Tech. Rep. NRS-88.
Newtown Square, PA: U.S. Department of Agriculture, Forest Service, Northern Research Station. 30 p.
Woodall, C.W., and V.J. Monleon (2008) Sampling protocol, estimation, and analysis procedures for the down
woody materials indicator of the FIA program. Gen. Tech. Rep. NRS-22. Newtown Square, PA: U.S. Department of
Agriculture, Forest Service, Northern Research Station. 68 p.
Woodall, C.W., Walters, B.F., Coulston, J.W., D'Amato, A.W., Domke, G.M., Russell, M.B., Sowers, P.A. (2015b)
Monitoring network confirms land use change is a substantial component of the forest carbon sink in the eastern
United States. Scientific Reports. 5: 17028.
Woodall, C.W., Walters, B.F., Oswalt, S.N., Domke, G.M., Toney, C., Gray, A.N. (2013) Biomass and carbon
attributes of downed woody materials in forests of the United States. Forest Ecology and Management 305: 48-59.
Yang, L, Jin, S., Danielson, P., Homer, C., Gass, L., Bender, S. M., Case, A., Costello, C., Dewitz, J., Fry, J., Funk, M.,
Granneman, B., Liknes, G. C., Rigge, M. & Xian, G. (2018) A new generation of the United States National Land
Cover Database: Requirements, research priorities, design, and implementation strategies. ISPRS Journal of
Photogrammetry and Remote Sensing 146:108-123.
Cropland Remaining Cropland
Armentano, T. V., and E.S. Menges (1986) Patterns of change in the carbon balance of organic soil-wetlands of the
temperate zone. Journal of Ecology 74: 755-774.
Brady, N.C. and R.R. Weil (1999) The Nature and Properties of Soils. Prentice Hall. Upper Saddle River, NJ, 881.
10-82 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
Brockwell, Peter J., and Richard A. Davis (2016) Introduction to time series and forecasting. Springer.
Conant, R. T., K. Paustian, and E.T. Elliott (2001) "Grassland management and conversion into grassland: effects on
soil carbon." Ecological Applications 11: 343-355.
CTIC (2004) National Crop Residue Management Survey: 1989-2004. Conservation Technology Information Center,
Purdue University, Available online at: http://www.ctic.purdue.edu/CRM/.
Daly, C., R.P. Neilson, and D.L Phillips (1994) "A Statistical-Topographic Model for Mapping Climatological
Precipitation Over Mountainous Terrain." Journal of Applied Meteorology 33:140-158.
Del Grosso, S.J., W.J. Parton, A.R. Mosier, M.D. Hartman, J. Brenner, D.S. Ojima, and D.S. Schimel (2001) "Simulated
Interaction of Carbon Dynamics and Nitrogen Trace Gas Fluxes Using the DAYCENT Model." In Modeling Carbon
and Nitrogen Dynamics for Soil Management, Schaffer, M., L Ma, S. Hansen, (eds.). CRC Press, Boca Raton, Florida,
pp. 303-332.
Del Grosso, S.J., S.M. Ogle, W.J. Parton (2011) Soil organic matter cycling and greenhouse gas accounting
methodologies, Chapter 1, pp 3-13 DOI: 10.1021/bk-2011-1072.ch001. In: Understanding Greenhouse Gas
Emissions from Agricultural Management, L. Guo, A. Gunasekara, L. McConnell (eds.). American Chemical Society,
Washington, D.C.
Edmonds, L, R. L. Kellogg, B. Kintzer, L. Knight, C. Lander, J. Lemunyon, D. Meyer, D.C. Moffitt, and J. Schaefer
(2003) "Costs associated with development and implementation of Comprehensive Nutrient Management Plans."
Part I—Nutrient management, land treatment, manure and wastewater handling and storage, and recordkeeping.
Natural Resources Conservation Service, U.S. Department of Agriculture. Available online at:
http://www.nrcs.usda.gov/technical/land/pubs/cnmpl.html.
Euliss, N., and R. Gleason (2002) Personal communication regarding wetland restoration factor estimates and
restoration activity data. Ned Euliss and Robert Gleason of the U.S. Geological Survey, Jamestown, ND, to Stephen
Ogle of the National Resource Ecology Laboratory, Fort Collins, CO. August 2002.
Friedman, J.H. (2001) "Greedy function approximation: A gradient boosting machine." Ann. Statist. 29 (5) 1189 -
1232.
Fry, J., Xian, G., Jin, S., Dewitz, J., Homer, C., Yang, L, Barnes, C., Herold, N., and Wickham, J. (2011) Completion of
the 2006 National Land Cover Database for the Conterminous United States, PE&RS, Vol. 77(9):858-864.
Griscom, B. W., Adams, J., Ellis, P. W., Houghton, R. A., Lomax, G., Miteva, D. A., Schlesinger, W. H., Shoch, D.,
Siikamaki, J. V., Smith, P., Woodbury, P., Zganjar, C., Blackman, A., Campari, J., Conant, R. T., Delgado, C., Elias, P.,
Gopalakrishna, T., Hamsik, M. R., Herrero, M., Kiesecker, J., Landis, E., Laestadius, L., Leavitt, S. M., Minnemeyer, S.,
Polasky, S., Potapov, P., Putz, F. E., Sanderman, J., Silvius, M., Wollenberg, E. & Fargione, J. (2017) "Natural climate
solutions." Proceedings of the National Academy of Sciences of the United States of America 114(44): 11645-11650.
Hagen, S. C., G. Delgado, P. Ingraham, I. Cooke, R. Emery, J. P. Fisk, L. Melendy, T. Olson, S. Patti, N. Rubin, B. Ziniti,
H. Chen, W. Salas, P. Elias, and D. Gustafson. 2020. Mapping Conservation Management Practices and Outcomes in
the Corn Belt Using the Operational Tillage Information System (OpTIS) and the Denitrification-Decomposition
(DNDC) Model. Land 9:408
Homer, C., Dewitz, J., Fry, J., Coan, M., Hossain, N., Larson, C., Herold, N., McKerrow, A., VanDriel, J.N., and
Wickham, J. (2007) Completion of the 2001 National Land Cover Database for the Conterminous United States.
Photogrammetric Engineering and Remote Sensing, Vol. 73, No. 4, pp 337-341.
Homer, C.G., Dewitz, J.A., Yang, L., Jin, S., Danielson, P., Xian, G., Coulston, J., Herold, N.D., Wickham, J.D., and
Megown, K. (2015) Completion of the 2011 National Land Cover Database for the conterminous United States-
Representing a decade of land cover change information. Photogrammetric Engineering and Remote Sensing, v.
81, no. 5, p. 345-354.
References and Abbreviations 10-83
-------
IPCC (2006) 2006IPCC Guidelines for National Greenhouse Gas Inventories. The National Greenhouse Gas
Inventories Programme, The Intergovernmental Panel on Climate Change. [H.S. Eggleston, L. Buendia, K. Miwa, T.
Ngara, and K. Tanabe (eds.)]. Hayama, Kanagawa, Japan.
IPCC (2003) Good Practice Guidance for Land Use, Land-Use Change, and Forestry. The Intergovernmental Panel on
Climate Change, National Greenhouse Gas Inventories Programme, J. Penman, et al., eds. August 13, 2004.
Available online at: http://www.ipcc-nggip.iges.or.ip/public/gpglulucf/gpglulucf.htm.
Lai, R., Kimble, J. M., Follett, R. F. & Cole, C. V. (1998) The potential of U.S. cropland to sequester carbon and
mitigate the greenhouse effect. Chelsea, Ml: Sleeping Bear Press, Inc.
Little, R. (1988) "Missing-data adjustments in large surveys." Journal of Business and Economic Statistics 6: 287-
296.
McGill, W.B., and C.V. Cole (1981) Comparative aspects of cycling of organic C, N, S and P through soil organic
matter. Geoderma 26:267-286.
Metherell, A.K., L.A. Harding, C.V. Cole, and W.J. Parton (1993) "CENTURY Soil Organic Matter Model
Environment." Agroecosystem version 4.0. Technical documentation, GPSRTech. Report No. 4, USDA/ARS, Ft.
Collins, CO.
Mesinger, F., G. DiMego, E. Kalnay, K. Mitchell, P. C. Shafran, W. Ebisuzaki, D. Jovic, J. Woollen, E. Rogers, E. H.
Berbery, M. B. Ek, Y. Fan, R. Grumbine, W. Higgins, H. Li, Y. Lin, G. Manikin, D. Parrish, and W. Shi (2006) North
American regional reanalysis. Bulletin of the American Meteorological Society 87:343-360.
Nelson, Mark D.; Riitters, Kurt H.; Coulston, John W.; Domke, Grant M.; Greenfield, Eric J.; Langner, Linda L.;
Nowak, David J.; O'Dea, Claire B.; Oswalt, Sonja N.; Reeves, Matthew C.; Wear, David N. 2020. Defining the United
States land base: a technical document supporting the USDA Forest Service 2020 RPA assessment. Gen. Tech. Rep.
NRS-191. Madison, Wl: U.S. Department of Agriculture, Forest Service, Northern Research Station. 70 p.
https://doi.org/10.2737/NRS-GTR-191.
NRCS (1999) Soil Taxonomy: A basic system of soil classification for making and interpreting soil surveys, 2nd
Edition. Agricultural Handbook Number 436, Natural Resources Conservation Service, U.S. Department of
Agriculture, Washington, D.C.
NRCS (1997) "National Soil Survey Laboratory Characterization Data," Digital Data, Natural Resources Conservation
Service, U.S. Department of Agriculture. Lincoln, NE.
NRCS (1981) Land Resource Regions and Major Land Resource Areas of the United States, USDA Agriculture
Handbook 296, United States Department of Agriculture, Natural Resources Conservation Service, National Soil
Survey Cente., Lincoln, NE, pp. 156.
Ogle, S.M., Breidt, F.J., Del Grosso, S., Gurung, R., Marx, E., Spencer, S., Williams, S., Manning, D. (2023)
"Counterfactual scenarios reveal historical impact of cropland management on soil organic carbon stocks in the
United States." Scientific Reports 13(1):14564.
Ogle, S. M., Alsaker, C., Baldock, J., Bernoux, M., Breidt, F. J., McConkey, B., Regina, K. & Vazquez-Amabile, G. G.
(2019) "Climate and Soil Characteristics Determine Where No-Till Management Can Store Carbon in Soils and
Mitigate Greenhouse Gas Emissions." Scientific Reports 9(1): 11665.
Ogle, S.M., F.J. Breidt, M. Easter, S. Williams, K. Killian, and K. Paustian (2010) "Scale and uncertainty in modeled
soil organic carbon stock changes for U.S. croplands using a process-based model." Global Change Biology 16:810-
820.
Ogle, S.M., F.J. Breidt, M. Easter, S. Williams and K. Paustian (2007) "Empirically-Based Uncertainty Associated with
Modeling Carbon Sequestration Rates in Soils." Ecological Modeling 205:453-463.
Ogle, S.M., F.J. Breidt, and K. Paustian (2006) "Bias and variance in model results due to spatial scaling of
measurements for parameterization in regional assessments." Global Change Biology 12:516-523.
10-84 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
Ogle, S. M., et al. (2005) "Agricultural management impacts on soil organic carbon storage under moist and dry
climatic conditions of temperate and tropical regions." Biogeochemistry 72: 87-121.
Ogle, S.M., M.D. Eve, F.J. Breidt, and K. Paustian (2003) "Uncertainty in estimating land use and management
impacts on soil organic carbon storage for U.S. agroecosystems between 1982 and 1997." Global Change Biology
9:1521-1542.
Parton, W.J., M.D. Hartman, D.S. Ojima, and D.S. Schimel (1998) "DAYCENT: Its Land Surface Submodel: Description
and Testing". Glob. Planet. Chang. 19: 35-48.
Parton, W.J., D.S. Ojima, C.V. Cole, and D.S. Schimel (1994) "A General Model for Soil Organic Matter Dynamics:
Sensitivity to litter chemistry, texture and management," in Quantitative Modeling of Soil Forming Processes.
Special Publication 39, Soil Science Society of America, Madison, Wl, 147-167.
Parton, W.J., D.S. Schimel, C.V. Cole, D.S. Ojima (1987) "Analysis of factors controlling soil organic matter levels in
Great Plains grasslands." Soil Science Society of America Journal 51:1173-1179.
Parton, W.J., J.W.B. Stewart, C.V. Cole. (1988) "Dynamics of C, N, P, and S in grassland soils: a model."
Biogeochemistry 5:109-131.
Paustian, K., et al. (1997a) "Agricultural soils as a sink to mitigate CO2 emissions." Soil Use and Management 13:
230-244.
Paustian, K., et al. (1997b) Management controls on soil carbon. In Soil organic matter in temperate
agroecosystems: long-term experiments in North America (Paul E.A., K. Paustian, and C.V. Cole, eds.). Boca Raton,
CRC Press, pp. 15-49.
Potter, C. S., J.T. Randerson, C.B. Fields, P.A. Matson, P.M. Vitousek, H.A. Mooney, and S.A. Klooster (1993)
"Terrestrial ecosystem production: a process model based on global satellite and surface data." Global
Biogeochemical Cycles 7:811-841.
Potter, C., S. Klooster, A. Huete, and V. Genovese (2007) Terrestrial carbon sinks for the United States predicted
from MODIS satellite data and ecosystem modeling. Earth Interactions 11, Article No. 13, DOI 10.1175/EI228.1.
PRISM Climate Group (2022) PRISM Climate Data, Oregon State University, http://prism.oregonstate.edu,
downloaded January 2022.
Pukelsheim, F. (1994) 'The 3-Sigma-Rule." American Statistician 48:88-91
Soil Survey Staff (2020) Gridded Soil Survey Geographic (gSSURGO) Database for the Conterminous United States.
United States Department of Agriculture, Natural Resources Conservation Service, Accessed February 2020
(FY2020 official release), Available online at https://gdg.sc.egov.usda.gov/.
Spencer, S., S.M. Ogle, F.J. Breidt, J. Goebel, and K. Paustian. (2011) "Designing a national soil carbon monitoring
network to support climate change policy: a case example for US agricultural lands." Greenhouse Gas Management
& Measurement 1: 167-178.
Towery, D. (2001) Personal Communication. Dan Towery regarding adjustments to the CTIC (1998) tillage data to
reflect long-term trends, Conservation Technology Information Center, West Lafayette, IN, and Marlen Eve,
National Resource Ecology Laboratory, Fort Collins, CO. February 2001.
USDA-ERS (2020) Agricultural Resource Management Survey (ARMS) Farm Financial and Crop Production Practices:
Tailored Reports. Available online at: https://www.ers.usda.gov/data-products/arms-farm-financial-and-crop-
production-practices/.
USDA-ERS (2018) Agricultural Resource Management Survey (ARMS) Farm Financial and Crop Production Practices:
Tailored Reports. Available online at: https://www.ers.usda.gov/data-products/arms-farm-financial-and-crop-
production-practices/.
References and Abbreviations 10-85
-------
USDA-ERS (1997) Cropping Practices Survey Data —1995. Economic Research Service, United States Department of
Agriculture. Available online at: http://www.ers.usda.gov/data/archive/93018/.
USDA Forest Service. (2022) Forest Inventory and Analysis National Program: FIA Data Mart. U.S. Department of
Agriculture Forest Service. Washington, D.C. Available online at:
https://apps.fs.usda.gov/fia/datamart/datamart.html. Accessed on 07 October 2022.
USDA-FSA (2015) Conservation Reserve Program Monthly Summary-September 2015. U.S. Department of
Agriculture, Farm Service Agency, Washington, D.C. Available online at: https://www.fsa.usda.gov/Assets/USDA-
FSA-Public/usdafiles/Conservation/PDF/sep2015summary.pdf.
USDA-NASS (2022) Quick Stats. National Agricultural Statistics Service, United States Department of Agriculture,
Washington, D.C., Accessed October 2022, http://quickstats.nass.usda.gov/.
USDA-NASS (2021) Published crop data layer. Available at https://nassgeodata.gmu.edu/CropScape/, Accessed July
2021, USDA-NASS, Washington, DC.
USDA-NASS (2017) 2017 Census of Agriculture. USDA National Agricultural Statistics Service, Complete data
available at http://www.nass.usda.gov/AgCensus.
USDA-NASS (2012) 2012 Census of Agriculture. USDA National Agricultural Statistics Service, Complete data
available at http://www.nass.usda.gov/AgCensus.
USDA-NASS (2004) Agricultural Chemical Usage: 2003 Field Crops Summary. Report AgChl(04)a. National
Agricultural Statistics Service, U.S. Department of Agriculture, Washington, D.C. Available online at:
http://usda.mannlib.cornell.edu/reports/nassr/other/pcu-bb/agcs0504.pdfH.
USDA-NASS (1999) Agricultural Chemical Usage: 1998 Field Crops Summary. Report AgCHl(99). National
Agricultural Statistics Service, U.S. Department of Agriculture, Washington, DC. Available online at:
http://usda.mannlib.cornell.edu/reports/nassr/other/pcu-bb/agch0599.pdf.
USDA-NASS (1992) Agricultural Chemical Usage: 1991 Field Crops Summary. Report AgChl(92). National
Agricultural Statistics Service, U.S. Department of Agriculture, Washington, D.C. Available online at:
http://usda.mannlib.cornell.edu/reports/nassr/other/pcu-bb/agch0392.txtH.
USDA-NRCS (2020) Summary Report: 2017 National Resources Inventory. Natural Resources Conservation Service,
Washington, DC, and Center for Survey Statistics and Methodology, Iowa State University, Ames, Iowa,
https://www.nrcs.usda.gov/sites/default/files/2022-10/2017NRlSummary Final.pdf.
USDA-NRCS (2018) CEAP Cropland Farmer Surveys. USDA Natural Resources Conservation Service.
https://www.nrcs.usda.gov/wps/portal/nrcs/detail/national/technical/nra/ceap/na/?cid=nrcsl43 014163.
USDA-NRCS (2012) Assessment of the Effects of Conservation Practices on Cultivated Cropland in the Upper
Mississippi River Basin. U.S. Department of Agriculture, Natural Resources Conservation Service,
https://www.nrcs.usda.gov/lnternet/FSE DOCUMENTS/stelprdb 1042093.pdf.
Yang, L, Jin, S., Danielson, P., Homer, C., Gass, L., Bender, S. M., Case, A., Costello, C., Dewitz, J., Fry, J., Funk, M.,
Granneman, B., Liknes, G. C., Rigge, M. & Xian, G. (2018) "A new generation of the United States National Land
Cover Database: Requirements, research priorities, design, and implementation strategies." ISPRS Journal of
Photogrammetry and Remote Sensing 146:108-123.
Land Converted to Cropland
Sampson and D. Hair, (eds.). Forest and Global Change, Volume 2: Forest Management Opportunities for
Mitigating Carbon Emissions. American Forests. Washington, D.C., 1-26 and 261-379 (appendices 262 and 263).
Brockwell, Peter J., and Richard A. Davis (2016) Introduction to time series and forecasting. Springer.
10-86 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
Del Grosso, S.J., W.J. Parton, A.R. Mosier, M.D. Hartman, J. Brenner, D.S. Ojima, and D.S. Schimel (2001) "Simulated
Interaction of Carbon Dynamics and Nitrogen Trace Gas Fluxes Using the DAYCENT Model." In Modeling Carbon
and Nitrogen Dynamics for Soil Management, Schaffer, M., L. Ma, S. Hansen, (eds.). CRC Press, Boca Raton, Florida,
pp. 303-332.
Del Grosso, S.J., S.M. Ogle, W.J. Parton (2011) "Soil organic matter cycling and greenhouse gas accounting
methodologies." Chapter 1, pp 3-13 DOI: 10.1021/bk-2011-1072.ch001. In: Understanding Greenhouse Gas
Emissions from Agricultural Management (L. Guo, A. Gunasekara, L McConnell. Eds.), American Chemical Society,
Washington, D.C.
Del Grosso, S.J., W.J. Parton, A.R. Mosier, M.D. Hartman, J. Brenner, D.S. Ojima, and D.S. Schimel (2001) "Simulated
Interaction of Carbon Dynamics and Nitrogen Trace Gas Fluxes Using the DAYCENT Model." In Schaffer, M., L. Ma,
S. Hansen, (eds.); Modeling Carbon and Nitrogen Dynamics for Soil Management. CRC Press. Boca Raton, Florida.
303-332.
Domke, G.M., J.E. Smith, and C.W. Woodall. (2011) "Accounting for density reduction and structural loss in
standing dead trees: Implications for forest biomass and carbon stock estimates in the United States". Carbon
Balance and Management 6:14.
Domke, G.M., et al. (2013) "From models to measurements: comparing down dead wood carbon stock estimates in
the U.S. forest inventory." PLoS ONE 8(3): e59949.
Domke, G.M., Perry, C.H., Walters, B.F., Woodall, C.W., and Smith, J.E. (2016) "A framework for estimating litter
carbon stocks in forests of the United States." Science of the Total Environment 557-558: 469-478.
Fry, J., Xian, G., Jin, S., Dewitz, J., Homer, C., Yang, L., Barnes, C., Herold, N., and Wickham, J. (2011) "Completion of the
2006 National Land Cover Database for the Conterminous United States." PE&RS, Vol. 77(9):858-864.
Harmon, M.E., C.W. Woodall, B. Fasth, J. Sexton, M. Yatkov. (2011) Differences between standing and downed
dead tree wood density reduction factors: A comparison across decay classes and tree species. Res. Paper. NRS-15.
Newtown Square, PA: U.S. Department of Agriculture, Forest Service, Northern Research Station. 40 p.
Homer, C., Dewitz, J., Fry, J., Coan, M., Hossain, N., Larson, C., Herold, N., McKerrow, A., VanDriel, J.N., and Wickham,
J. (2007) "Completion of the 2001 National Land Cover Database for the Conterminous United States."
Photogrammetric Engineering and Remote Sensing, Vol. 73, No. 4, pp 337-341.
Homer, C.G., Dewitz, J.A., Yang, L., Jin, S., Danielson, P., Xian, G., Coulston, J., Herold, N.D., Wickham, J.D., and
Megown, K. (2015) "Completion of the 2011 National Land Cover Database for the conterminous United States-
Representing a decade of land cover change information." Photogrammetric Engineering and Remote Sensing 81:
345-354.
Houghton, R.A., et al. (1983) "Changes in the carbon content of terrestrial biota and soils between 1860 and 1980:
a net release of CO2 to the atmosphere." Ecological Monographs 53: 235-262.
Houghton, R. A. and Nassikas, A. A. (2017) "Global and regional fluxes of carbon from land use and land cover
change 1850-2015." Global Biogeochemical Cycles 31(3): 456-472.
IPCC (2006) 2006IPCC Guidelines for National Greenhouse Gas Inventories. The National Greenhouse Gas
Inventories Programme, The Intergovernmental Panel on Climate Change. [H.S. Eggleston, L. Buendia, K. Miwa, T.
Ngara, and K. Tanabe (eds.)]. Hayama, Kanagawa, Japan.
Jenkins, J.C., D.C. Chojnacky, L.S. Heath, and R.A. Birdsey (2003) "National-scale biomass estimators for United
States tree species." Forest Science 49(l):12-35.
Metherell, A.K., L.A. Harding, C.V. Cole, and W.J. Parton (1993) CENTURY Soil Organic Matter Model Environment.
Agroecosystem version 4.0. Technical documentation, GPSRTech. Report No. 4, USDA/ARS, Ft. Collins, CO.Ogle,
S.M., Breidt, F.J., Del Grosso, S., Gurung, R., Marx, E., Spencer, S., Williams, S., Manning, D. (2023) "Counterfactual
scenarios reveal historical impact of cropland management on soil organic carbon stocks in the United States."
Scientific Reports 13(1):14564.
References and Abbreviations 10-87
-------
Ogle, S.M., F.J. Breidt, M. Easter, S. Williams, K. Killian, and K. Paustian (2010) "Scale and uncertainty in modeled
soil organic carbon stock changes for U.S. croplands using a process-based model." Global Change Biology 16:810-
820.
Ogle, S.M., M.D. Eve, F.J. Breidt, and K. Paustian (2003) "Uncertainty in estimating land use and management
impacts on soil organic carbon storage for U.S. agroecosystems between 1982 and 1997." Global Change Biology
9:1521-1542.
Ogle, S.M., F.J. Breidt, and K. Paustian (2006) "Bias and variance in model results due to spatial scaling of
measurements for parameterization in regional assessments." Global Change Biology 12:516-523.
Parton, W.J., M.D. Hartman, D.S. Ojima, and D.S. Schimel (1998) "DAYCENT: Its Land Surface Submodel: Description
and Testing". Glob. Planet. Chang. 19: 35-48.
Parton, W.J., D.S. Ojima, C.V. Cole, and D.S. Schimel (1994) "A General Model for Soil Organic Matter Dynamics:
Sensitivity to litter chemistry, texture and management," in Quantitative Modeling of Soil Forming Processes.
Special Publication 39, Soil Science Society of America, Madison, Wl, 147-167.
Parton, W.J., D.S. Schimel, C.V. Cole, D.S. Ojima (1987) "Analysis of factors controlling soil organic matter levels in
Great Plains grasslands." Soil Science Society of America Journal 51:1173-1179.
Parton, W.J., J.W.B. Stewart, C.V. Cole. (1988) "Dynamics of C, N, P, and S in grassland soils: a model."
Biogeochemistry 5:109-131.
PRISM Climate Group (2022) PRISM Climate Data, Oregon State University, http://prism.oregonstate.edu,
downloaded January 2022.
Smith, J.E.; Heath, L.S.; Skog, K.E.; Birdsey, R.A. (2006) Methods for calculating forest ecosystem and harvested
carbon with standard estimates for forest types of the United States. Gen. Tech. Rep. NE-343. Newtown Square,
PA: U.S. Department of Agriculture, Forest Service, Northeastern Research Station. 216 p.
Tubiello, F. N., et al. (2015) "The Contribution of Agriculture, Forestry and other Land Use activities to Global
Warming, 1990-2012." Global Change Biology 21:2655-2660.
USDA Forest Service (2023) Forest Inventory and Analysis National Program: FIA Data Mart v2.0.1. U.S. Department
of Agriculture Forest Service. Washington, D.C. Available online at:
https://apps.fs.usda.gov/fia/datamart/datamart.html Accessed on 13 September 2023.
USDA-NASS (2021) Published crop data layer. Available at https://nassgeodata.gmu.edu/CropScape/, Accessed July
2021, USDA-NASS, Washington, DC.
USDA-NRCS (2020) Summary Report: 2017 National Resources Inventory. Natural Resources Conservation Service,
Washington, DC, and Center for Survey Statistics and Methodology, Iowa State University, Ames, Iowa.
https://www.nrcs.usda.gov/wps/portal/nrcs/main/national/technical/nra/nri/results/.
USDA-NRCS (2018) Summary Report: 2015 National Resources Inventory. Natural Resources Conservation Service,
Washington, D.C., and Center for Survey Statistics and Methodology, Iowa State University, Ames, Iowa.
https://www.nrcs.usda.gov/lnternet/FSE DOCUMENTS/nrcseprd 1422028.pdf.
Westfall, J.A., Coulston, J.W., Gray, A.N., Shaw, J.D., Radtke, P.J., Walker, D.M., Weiskittel, A.R., MacFarlane, D.W.,
Affleck, D.L.R., Zhao, D., Temesgen, H., Poudel, K.P., Frank, J.M., Prisley, S.P., Wang, Y., Sanchez Meador, A.J., Auty,
D., and Domke, G.M. (2023) A national-scale tree volume, biomass, and carbon modeling system for the United
States. Gen. Tech. Rep. WO-XX
Woodall, C.W., and V.J. Monleon (2008) Sampling protocol, estimation, and analysis procedures for the down
woody materials indicator of the FIA program. Gen. Tech. Rep. NRS-22. Newtown Square, PA: U.S. Department of
Agriculture, Forest Service, Northern Research Station. 68 p.
10-88 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
Woodall, C.W., LS. Heath, G.M. Domke, and M.C. Nichols (2011) Methods and equations for estimating
aboveground volume, biomass, and carbon for trees in the U.S. forest inventory, 2010. Gen. Tech. Rep. NRS-88.
Newtown Square, PA: U.S. Department of Agriculture, Forest Service, Northern Research Station. 30 p.
Yang, L, Jin, S., Danielson, P., Homer, C., Gass, L., Bender, S. M., Case, A., Costello, C., Dewitz, J., Fry, J., Funk, M.,
Granneman, B., Liknes, G. C., Rigge, M. & Xian, G. (2018) "A new generation of the United States National Land
Cover Database: Requirements, research priorities, design, and implementation strategies." ISPRS Journal of
Photogrammetry and Remote Sensing 146:108-123.
Grassland Remaining Grassland: Soil Carbon Stock Changes
Brockwell, Peter J., and Richard A. Davis (2016) Introduction to time series and forecasting. Springer.
Del Grosso, S.J., S.M. Ogle, W.J. Parton (2011) Soil organic matter cycling and greenhouse gas accounting
methodologies, Chapter 1, pp 3-13 DOI: 10.1021/bk-2011-1072.ch001. In: Understanding Greenhouse Gas
Emissions from Agricultural Management (L Guo, A. Gunasekara, L. McConnell. Eds.), American Chemical Society,
Washington, D.C.
Del Grosso, S.J., W.J. Parton, A.R. Mosier, M.D. Hartman, J. Brenner, D.S. Ojima, and D.S. Schimel (2001) "Simulated
Interaction of Carbon Dynamics and Nitrogen Trace Gas Fluxes Using the DAYCENT Model." In Modeling Carbon
and Nitrogen Dynamics for Soil Management, Schaffer, M., L Ma, S. Hansen, (eds.). CRC Press, Boca Raton, Florida,
pp. 303-332.
Edmonds, L, R. L. Kellogg, B. Kintzer, L. Knight, C. Lander, J. Lemunyon, D. Meyer, D.C. Moffitt, and J. Schaefer
(2003) "Costs associated with development and implementation of Comprehensive Nutrient Management Plans."
Part I—Nutrient management, land treatment, manure and wastewater handling and storage, and recordkeeping.
Natural Resources Conservation Service, U.S. Department of Agriculture. Available online at:
http://www.nrcs.usda.gov/technical/land/pubs/cnmpl.html.
EPA (1999) Biosolids Generation, Use and Disposal in the United States. Office of Solid Waste, U.S. Environmental
Protection Agency. Available online at: http://biosolids.policy.net/relatives/18941.PDF.
Fry, J., Xian, G., Jin, S., Dewitz, J., Homer, C., Yang, L, Barnes, C., Herold, N., and Wickham, J. (2011) Completion of
the 2006 National Land Cover Database for the Conterminous United States, PE&RS, Vol. 77(9):858-864.
Homer, C., Dewitz, J., Fry, J., Coan, M., Hossain, N., Larson, C., Herold, N., McKerrow, A., VanDriel, J.N., and
Wickham, J. (2007) Completion of the 2001 National Land Cover Database for the Conterminous United States.
Photogrammetric Engineering and Remote Sensing, Vol. 73, No. 4, pp 337-341.
Homer, C.G., Dewitz, J.A., Yang, L., Jin, S., Danielson, P., Xian, G., Coulston, J., Herold, N.D., Wickham, J.D., and
Megown, K. (2015) Completion of the 2011 National Land Cover Database for the conterminous United States-
Representing a decade of land cover change information. Photogrammetric Engineering and Remote Sensing, v.
81, no. 5, p. 345-354.
IPCC (2006) 2006IPCC Guidelines for National Greenhouse Gas Inventories. The National Greenhouse Gas
Inventories Programme, The Intergovernmental Panel on Climate Change. [H.S. Eggleston, L. Buendia, K. Miwa, T.
Ngara, and K. Tanabe (eds.)]. Hayama, Kanagawa, Japan.
Kellogg, R.L, C.H. Lander, D.C. Moffitt, and N. Gollehon (2000) Manure Nutrients Relative to the Capacity of
Cropland and Pastureland to Assimilate Nutrients: Spatial and Temporal Trends for the United States. U.S.
Department of Agriculture, Washington, D.C. Publication number nps00-0579.
Metherell, A.K., L.A. Harding, C.V. Cole, and W.J. Parton (1993) "CENTURY Soil Organic Matter Model
Environment." Agroecosystem version 4.0. Technical documentation, GPSRTech. Report No. 4, USDA/ARS, Ft.
Collins, CO.NEBRA (2007) A National Biosolids Regulation, Quality, End Use & Disposal Survey. North East Biosolids
and Residuals Association. July 21, 2007.
References and Abbreviations 10-89
-------
Nusser, S.M. and J.J. Goebel (1997) The national resources inventory: a long-term multi-resource monitoring
programme. Environmental and Ecological Statistics 4:181-204.
Ogle, S.M., F.J. Breidt, M. Easter, S. Williams, K. Killian, and K. Paustian (2010) "Scale and uncertainty in modeled
soil organic carbon stock changes for U.S. croplands using a process-based model." Global Change Biology 16:810-
820.
Ogle, S.M., M.D. Eve, F.J. Breidt, and K. Paustian (2003) "Uncertainty in estimating land use and management
impacts on soil organic carbon storage for U.S. agroecosystems between 1982 and 1997." Global Change Biology
9:1521-1542.
Parton, W.J., D.S. Ojima, C.V. Cole, and D.S. Schimel (1994) "A General Model for Soil Organic Matter Dynamics:
Sensitivity to litter chemistry, texture and management," in Quantitative Modeling of Soil Forming Processes.
Special Publication 39, Soil Science Society of America, Madison, Wl, 147-167.
Parton, W.J., D.S. Schimel, C.V. Cole, D.S. Ojima (1987) "Analysis of factors controlling soil organic matter levels in
Great Plains grasslands." Soil Science Society of America Journal 51:1173-1179.
Parton, W.J., J.W.B. Stewart, C.V. Cole. (1988) "Dynamics of C, N, P, and S in grassland soils: a model."
Biogeochemistry 5:109-131.
Parton, W.J., M.D. Hartman, D.S. Ojima, and D.S. Schimel (1998) "DAYCENT: Its Land Surface Submodel: Description
and Testing". Glob. Planet. Chang. 19: 35-48.PRISM Climate Group, Oregon State University,
http://prism.oregonstate.edu, created 24 July 2015.
PRISM Climate Group (2022) PRISM Climate Data, Oregon State University, http://prism.oregonstate.edu,
downloaded January 2022.
United States Bureau of Land Management (BLM) (2014) Rangeland Inventory, Monitoring, and Evaluation
Reports. Bureau of Land Management. U.S. Department of the Interior. Available online at:
http://www.blm.gov/wo/st/en/prog/more/rangeland management/rangeland inventory.html.
USDA-NASS (2021) Published crop data layer. Available at https://nassgeodata.gmu.edu/CropScape/, Accessed July
2021, USDA-NASS, Washington, DC.
USDA-NRCS (2020) Summary Report: 2017 National Resources Inventory. Natural Resources Conservation Service,
Washington, DC, and Center for Survey Statistics and Methodology, Iowa State University, Ames, Iowa.
https://www.nrcs.usda.gov/wps/portal/nrcs/main/national/technical/nra/nri/results/.
USDA Forest Service (2023) Forest Inventory and Analysis National Program: FIA Data Mart v2.0.1. U.S. Department
of Agriculture Forest Service. Washington, D.C. Available online at:
https://apps.fs.usda.gov/fia/datamart/datamart.html Accessed on 13 September 2023.
Yang, L, Jin, S., Danielson, P., Homer, C., Gass, L., Bender, S. M., Case, A., Costello, C., Dewitz, J., Fry, J., Funk, M.,
Granneman, B., Liknes, G. C., Rigge, M. & Xian, G. (2018) "A new generation of the United States National Land
Cover Database: Requirements, research priorities, design, and implementation strategies." ISPRS Journal of
Photogrammetry and Remote Sensing 146:108-123.
Grassland Remaining Grassland: Non-C02 Emissions from
Grassland Fires
Anderson, R.C. Evolution and origin of the Central Grassland of North America: climate, fire and mammalian
grazers. Journal of the Torrey Botanical Society 133: 626-647.
Andreae, M.O. and P. Merlet (2001) Emission of trace gases and aerosols from biomass burning. Global
Biogeochemical Cycles 15:955-966.
10-90 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
Brockwell, Peter J., and Richard A. Davis (2016) Introduction to time series and forecasting. Springer.
Chapin, F.S., S.F. Trainor, O. Huntington, A.L Lovecraft, E. Zavaleta, D.C. Natcher, A.D. McGuire, J.L Nelson, L Ray,
M. Calef, N. Fresco, H. Huntington, T.S. Rupp, L. DeWilde, and R.L. Naylor (2008) Increasing wildfires in Alaska's
Boreal Forest: Pathways to potential solutions of a wicked problem. Bioscience 58:531-540.
Daubenmire, R. (1968) Ecology of fire in grasslands. Advances in Ecological Research 5:209-266.
Fry, J., Xian, G., Jin, S., Dewitz, J., Homer, C., Yang, L, Barnes, C., Herold, N., and Wickham, J. (2011) Completion of
the 2006 National Land Cover Database for the Conterminous United States, PE&RS, Vol. 77(9):858-864.
Homer, C., Dewitz, J., Fry, J., Coan, M., Hossain, N., Larson, C., Herold, N., McKerrow, A., VanDriel, J.N., and Wickham,
J. (2007) Completion of the 2001 National Land Cover Database for the Conterminous United States.
Photogrammetric Engineering and Remote Sensing, Vol. 73, No. 4, pp 337-341.
Homer, C.G., Dewitz, J.A., Yang, L., Jin, S., Danielson, P., Xian, G., Coulston, J., Herold, N.D., Wickham, J.D., and
Megown, K. (2015) Completion of the 2011 National Land Cover Database for the conterminous United States-
Representing a decade of land cover change information. Photogrammetric Engineering and Remote Sensing, v.
81, no. 5, p. 345-354.
IPCC (2013) Climate Change 2013: The Physical Science Basis. Contribution of Working Group I to the Fifth
Assessment Report of the Intergovernmental Panel on Climate Change [Stocker, T.F., D. Qin, G.-K. Plattner, M.
Tignor, S.K. Allen, J. Boschung, A. Nauels, Y. Xia, V. Bex and P.M. Midgley (eds.)]. Cambridge University Press,
Cambridge, United Kingdom and New York, NY, USA, 1535 pp.
IPCC (2006) 2006 IPCC Guidelines for National Greenhouse Gas Inventories. The National Greenhouse Gas
Inventories Programme, The Intergovernmental Panel on Climate Change. [H.S. Eggleston, L. Buendia, K. Miwa, T.
Ngara, and K. Tanabe (eds.)]. Hayama, Kanagawa, Japan.
MTBS (2023) Burned Areas Boundaries Dataset. (2023, August - last revised). MTBS Project (USDA Forest
Service/U.S. Geological Survey). Available online: https://mtbs.gov/direct-download Accessed: August 9, 2023.
Ogle, S.M., S. Spencer, M. Hartman, L. Buendia, L. Stevens, D. du Toit, J. Witi (2016) "Developing national baseline
GHG emissions and analyzing mitigation potentials for agriculture and forestry using an advanced national GHG
inventory software system." In Advances in Agricultural Systems Modeling 6, Synthesis and Modeling of
Greenhouse Gas Emissions and Carbon Storage in Agricultural and Forestry Systems to Guide Mitigation and
Adaptation, S. Del Grosso, LR. Ahuja and W.J. Parton (eds.), American Society of Agriculture, Crop Society of
America and Soil Science Society of America, pp. 129-148.
Nusser, S.M. and J.J. Goebel (1997) The national resources inventory: a long-term multi-resource monitoring
programme. Environmental and Ecological Statistics 4:181-204.
Picotte, J.J., K. Bhattarai, D. Howard, J. Lecker, J. Epting, B. Quayle, N. Benson, and K. Nelson (2020) "Changes to
the Monitoring Trends in Burn Severity program mapping production procedures and data products." Fire Ecology.
16:16. https://doi.org/10.1186/s42408-020-00076-v.
USDA-NRCS (2015) Summary Report: 2012 National Resources Inventory, Natural Resources Conservation Service,
Washington, D.C., and Center for Survey Statistics and Methodology, Iowa State University, Ames, Iowa. Available
online at: http://www.nrcs.usda.gov/lnternet/FSE DOCUMENTS/nrcseprd396218.pdf.
land Converted to Grassland
Asner, G.P., Archer, S., Hughes, R.F., Ansley, R.J. and Wessman, C.A. (2003) "Net changes in regional woody
vegetation cover and carbon storage in Texas drylands, 1937-1999." Global Change Biology 9(3): 316-335.
Birdsey, R. (1996) "Carbon Storage for Major Forest Types and Regions in the Conterminous United States." In R.N.
Sampson and D. Hair, (eds.). Forest and Global Change, Volume 2: Forest Management Opportunities for
Mitigating Carbon Emissions. American Forests. Washington, D.C., 1-26 and 261-379 (appendices 262 and 263).
References and Abbreviations 10-91
-------
Breshears, D.D., Knapp, A.K., Law, D.J., Smith, M.D., Twidwell, D. and Wonkka, C.L, 2016. Rangeland Responses to
Predicted Increases in Drought Extremity. Rangelands, 38(4), pp. 191-196.
Brockwell, Peter J., and Richard A. Davis (2016) Introduction to time series and forecasting. Springer.
Del Grosso, S.J., S.M. Ogle, W.J. Parton. (2011) Soil organic matter cycling and greenhouse gas accounting
methodologies, Chapter 1, pp 3-13 DOI: 10.1021/bk-2011-1072.ch001. In: Understanding Greenhouse Gas
Emissions from Agricultural Management (L Guo, A. Gunasekara, L McConnell. Eds.), American Chemical Society,
Washington, D.C.
Del Grosso, S.J., W.J. Parton, A.R. Mosier, M.D. Hartman, J. Brenner, D.S. Ojima, and D.S. Schimel (2001) "Simulated
Interaction of Carbon Dynamics and Nitrogen Trace Gas Fluxes Using the DAYCENT Model." In Modeling Carbon
and Nitrogen Dynamics for Soil Management (Schaffer, M., L. Ma, S. Hansen, (eds.). CRC Press, Boca Raton, Florida,
pp. 303-332.
Domke, G.M., J.E. Smith, and C.W. Woodall. (2011) Accounting for density reduction and structural loss in standing
dead trees: Implications for forest biomass and carbon stock estimates in the United States. Carbon Balance and
Management. 6:14.
Domke, G.M., et al. (2013) From models to measurements: comparing down dead wood carbon stock estimates in
the U.S. forest inventory. PLoS ONE 8(3): e59949.
Domke, G.M., Perry, C.H., Walters, B.F., Woodall, C.W., and Smith, J.E. (2016) A framework for estimating litter
carbon stocks in forests of the United States. Science of the Total Environment 557-558: 469-478.
Domke, G.M., Walters, B.F., Smith, J.E., Woodall, C.W. (2022) Chapter 6: FIA Carbon Attributes. In Westfall, J.A.;
Coulston, J.W.; Moisen, G.G.; Andersen, H.-E., eds. 2022. Sampling and estimation documentation for the
Enhanced Forest Inventory and Analysis Program: 2022. Gen. Tech. Rep. NRS-GTR-207, Madison, Wl: U.S.
Department of Agriculture, Forest Service, Northern Research Station. 129 p. https://doi.org/10.2737/NRS-GTR-
207.
Epstein, H.E., Gill, R.A., Paruelo, J.M., Lauenroth, W.K., Jia, G.J. and Burke, I.C. (2002) The relative abundance of
three plant functional types in temperate grasslands and shrublands of North and South America: effects of
projected climate change. Journal of Biogeography, 29(7), pp.875-888.
Fry, J., Xian, G., Jin, S., Dewitz, J., Homer, C., Yang, L, Barnes, C., Herold, N., and Wickham, J. (2011) Completion of
the 2006 National Land Cover Database for the Conterminous United States, PE&RS, Vol. 77(9):858-864.
Harmon, M.E., C.W. Woodall, B. Fasth, J. Sexton, M. Yatkov. (2011) Differences between standing and downed
dead tree wood density reduction factors: A comparison across decay classes and tree species. Res. Paper. NRS-15.
Newtown Square, PA: U.S. Department of Agriculture, Forest Service, Northern Research Station. 40 p.
Homer, C., Dewitz, J., Fry, J., Coan, M., Hossain, N., Larson, C., Herold, N., McKerrow, A., VanDriel, J.N., and Wickham,
J. (2007) Completion of the 2001 National Land Cover Database for the Conterminous United States.
Photogrammetric Engineering and Remote Sensing, Vol. 73, No. 4, pp 337-341.
Homer, C.G., Dewitz, J.A., Yang, L, Jin, S., Danielson, P., Xian, G., Coulston, J., Herold, N.D., Wickham, J.D., and
Megown, K. (2015) Completion of the 2011 National Land Cover Database for the conterminous United States-
Representing a decade of land cover change information. Photogrammetric Engineering and Remote Sensing, v. 81,
no. 5, p. 345-354.
Houghton, R.A., et al. (1983) "Changes in the carbon content of terrestrial biota and soils between 1860 and 1980:
a net release of CO2 to the atmosphere." Ecological Monographs 53: 235-262.
Houghton, R. A. and Nassikas, A. A. (2017) "Global and regional fluxes of carbon from land use and land cover
change 1850-2015." Global Biogeochemical Cycles 31(3): 456-472.
Huang, C.Y., Asner, G.P., Martin, R.E., Barger, N.N. and Neff, J.C. (2009) "Multiscale analysis of tree cover and
aboveground carbon stocks in pinyon-juniper woodlands." Ecological Applications 19(3): 668-681.
10-92 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
IPCC (2006) 2006IPCC Guidelines for National Greenhouse Gas Inventories. The National Greenhouse Gas
Inventories Programme, The Intergovernmental Panel on Climate Change, [H.S. Eggleston, L. Buendia, K. Miwa, T
Ngara, and K. Tanabe (eds.)]. Hayama, Kanagawa, Japan.
Jenkins, J.C., D.C. Chojnacky, L.S. Heath, and R.A. Birdsey (2003) "National-scale biomass estimators for United
States tree species." Forest Science 49(l):12-35.
Jurena, P.N. and Archer, S., (2003) Woody plant establishment and spatial heterogeneity in grasslands. Ecology,
84(4), pp.907-919.
Lenihan, J.M., Drapek, R., Bachelet, D. and Neilson, R.P., (2003) Climate change effects on vegetation distribution,
carbon, and fire in California. Ecological Applications, 13(6), pp.1667-1681.
Metherell, A.K., L.A. Harding, C.V. Cole, and W.J. Parton (1993) "CENTURY Soil Organic Matter Model
Environment." Agroecosystem version 4.0. Technical documentation, GPSRTech. Report No. 4, USDA/ARS, Ft.
Collins, CO.
Ogle, S.M., F.J. Breidt, M. Easter, S. Williams, K. Killian, and K. Paustian (2010) "Scale and uncertainty in modeled
soil organic carbon stock changes for U.S. croplands using a process-based model." Global Change Biology 16:810-
820.
Ogle, S.M., M.D. Eve, F.J. Breidt, and K. Paustian (2003) "Uncertainty in estimating land use and management
impacts on soil organic carbon storage for U.S. agroecosystems between 1982 and 1997." Global Change Biology
9:1521-1542.
Parton, W.J., D.S. Ojima, C.V. Cole, and D.S. Schimel (1994) "A General Model for Soil Organic Matter Dynamics:
Sensitivity to litter chemistry, texture and management," in Quantitative Modeling of Soil Forming Processes.
Special Publication 39, Soil Science Society of America, Madison, Wl, 147-167.
Parton, W.J., D.S. Schimel, C.V. Cole, D.S. Ojima (1987) "Analysis of factors controlling soil organic matter levels in
Great Plains grasslands." Soil Science Society of America Journal 51:1173-1179.
Parton, W.J., J.W.B. Stewart, C.V. Cole (1988) "Dynamics of C, N, P, and S in grassland soils: a model."
Biogeochemistry 5:109-131.
Parton, W.J., M.D. Hartman, D.S. Ojima, and D.S. Schimel (1998) "DAYCENT: Its Land Surface Submodel: Description
and Testing". Glob. Planet. Chang. 19: 35-48.
PRISM Climate Group (2022) PRISM Climate Data, Oregon State University, http://prism.oregonstate.edu,
downloaded January 2022.
Scholes, R.J. and Archer, S.R. (1997) Tree-grass interactions in savannas 1. Annual review of Ecology and
Systematics, 28(1), pp.517-544.
Sims, P.L, Singh, J.S. and Lauenroth, W.K. (1978) The structure and function often western North American
grasslands: I. Abiotic and vegetational characteristics. The Journal of Ecology, pp.251-285.
Smith, J.E.; Heath, L.S.; Skog, K.E.; Birdsey, R.A. (2006) Methods for calculating forest ecosystem and harvested
carbon with standard estimates for forest types of the United States. Gen. Tech. Rep. NE-343. Newtown Square,
PA: U.S. Department of Agriculture, Forest Service, Northeastern Research Station. 216 p.
Tarhouni, M., et al. (2016) Measurement of the aboveground biomass of some rangeland species using a digital
non-destructive technique. Botany Letters 163(3):281-287.
Tubiello, F. N., et al. (2015) "The Contribution of Agriculture, Forestry and other Land Use activities to Global
Warming, 1990-2012." Global Change Biology 21:2655-2660.
United States Bureau of Land Management (BLM) (2014) Rangeland Inventory, Monitoring, and Evaluation
Reports. Bureau of Land Management. U.S. Department of the Interior. Available online at:
http://www.blm.gov/wo/st/en/prog/more/rangeland management/rangeland inventory.html.
References and Abbreviations 10-93
-------
USDA Forest Service (2023) Forest Inventory and Analysis National Program: FIA Data Mart v2.0.1. U.S. Department
of Agriculture Forest Service. Washington, D.C. Available online at:
https://apps.fs.usda.gov/fia/datamart/datamart.html Accessed on 13 September 2023.
USDA-NASS (2021) Published crop data layer. Available at https://nassgeodata.gmu.edu/CropScape/, Accessed July
2021, USDA-NASS, Washington, DC.
USDA-NRCS (2020) Summary Report: 2017 National Resources Inventory. Natural Resources Conservation Service,
Washington, DC, and Center for Survey Statistics and Methodology, Iowa State University, Ames, Iowa.
https://www.nrcs.usda.gov/wps/portal/nrcs/main/national/technical/nra/nri/results/.
USDA-NRCS (2018) Summary Report: 2015 National Resources Inventory. Natural Resources Conservation Service,
Washington, D.C., and Center for Survey Statistics and Methodology, Iowa State University, Ames, Iowa.
https://www.nrcs.usda.gov/lnternet/FSE DOCUMENTS/nrcseprd 1422028.pdf.
Woodall, C.W., and V.J. Monleon (2008) Sampling protocol, estimation, and analysis procedures for the down
woody materials indicator of the FIA program. Gen. Tech. Rep. NRS-22. Newtown Square, PA: U.S. Department of
Agriculture, Forest Service, Northern Research Station. 68 p.
Woodall, C.W., LS. Heath, G.M. Domke, and M.C. Nichols. (2011) Methods and equations for estimating
aboveground volume, biomass, and carbon for trees in the U.S. forest inventory, 2010. Gen. Tech. Rep. NRS-88.
Newtown Square, PA: U.S. Department of Agriculture, Forest Service, Northern Research Station. 30 p.
Yang, L, Jin, S., Danielson, P., Homer, C., Gass, L., Bender, S. M., Case, A., Costello, C., Dewitz, J., Fry, J., Funk, M.,
Granneman, B., Liknes, G. C., Rigge, M. & Xian, G. (2018) "A new generation of the United States National Land
Cover Database: Requirements, research priorities, design, and implementation strategies." ISPRS Journal of
Photogrammetry and Remote Sensing 146:108-123.
Wetlands Remaining Wetlands: C02, CH4, and N20 Emissions
from Peatlands Remaining Peatlands
Apodaca, L. (2011) Email correspondence. Lori Apodaca, Peat Commodity Specialist, USGS and Emily Rowan, ICF
International. November.
Apodaca, L. (2008) E-mail correspondence. Lori Apodaca, Peat Commodity Specialist, USGS and Emily Rowan, ICF
International. October and November.
Cleary, J., N. Roulet and T.R. Moore (2005) "Greenhouse gas emissions from Canadian peat extraction, 1990-2000:
A life-cycle analysis." Ambio 34:456-461.
Division of Geological & Geophysical Surveys (DGGS), Alaska Department of Natural Resources (1997-2015)
Alaska's Mineral Industry Report (1997-2014). Alaska Department of Natural Resources, Fairbanks, AK. Available
online at http://www.dggs.dnr.state.ak.us/pubs/pubs?reqtype=minerals.
IPCC (2014) Climate Change 2014: Synthesis Report. Contribution of Working Groups I, II and III to the Fifth
Assessment Report of the Intergovernmental Panel on Climate Change. R.K. Pachauri and LA. Meyer (eds.). IPCC,
Geneva, Switzerland.
IPCC (2013) 2013 Supplement to the 2006 IPCC Guidelines for National Greenhouse Gas Inventories: Wetlands.
Hiraishi, T., Krug, T., Tanabe, K., Srivastava, N., Baasansuren, J., Fukuda, M. and Troxler, T.G. (eds.). Published: IPCC,
Switzerland.
IPCC (2007) Climate Change 2007: Synthesis Report. Contribution of Working Groups I, II and III to the Fourth
Assessment Report (AR4) of the IPCC. The Intergovernmental Panel on Climate Change, R.K. Pachauri, A. Resinger
(eds.). Geneva, Switzerland.
10-94 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
IPCC (2006) 2006IPCC Guidelines for National Greenhouse Gas Inventories. The National Greenhouse Gas
Inventories Programme, The Intergovernmental Panel on Climate Change. H.S. Eggleston, L. Buendia, K. Miwa, T.
Ngara, and K. Tanabe (eds.). Hayama, Kanagawa, Japan.
Szumigala, D.J. (2011) Phone conversation. Dr. David Szumigala, Division of Geological and Geophysical Surveys,
Alaska Department of Natural Resources and Emily Rowan, ICF International. January 18, 2011.
Szumigala, D.J. (2008) Phone conversation. Dr. David Szumigala, Division of Geological and Geophysical Surveys,
Alaska Department of Natural Resources and Emily Rowan, ICF International. October 17, 2008.
USGS (1991-2018) Minerals Yearbook: Peat (1994-2018). United States Geological Survey, Reston, VA. Available
online at http://iTiinerals.usgs.gov/iTiinerals/pubs/coiTHTiodity/peat/index.htrol.
USGS (2023a) Minerals Yearbook: Peat (2019) Tables-only release. United States Geological Survey, Reston, VA.
Available online at https://www.usgs.gov/centers/nmic/peat statistics-arid-information.
USGS (2023b) Minerals Yearbook: Peat (2020) Tables-only release. United States Geological Survey, Reston, VA.
Available online at https://www.usgs.gov/centers/nmic/peat statistics-arid-information.
USGS (2023c) Mineral Commodity Summaries: Peat (1996-2023). United States Geological Survey, Reston, VA.
Available online at https://www.usgs.gov/centers/nmic/peat-statistics-and-information.
USGS (2023d) Email correspondence. Amanda Brioche, Mineral Commodity Specialist, USGS. August 2023.
Wetlands Remaining Coastal Wetlands: Emissions and
Removals from Coastal Wetlands Remaining Coastal
Wetlands
Abbott, K. M., Elsey-Quirk, T., and DeLaune, R. D. (2019) Factors influencing blue carbon accumulation across a 32-
year chronosequence of created coastal marshes. Ecosphere, 10(8): e02828.
Allen, J. R., Cornwell, J. C., and Baldwin, A. H. (2021) Contributions of organic and mineral matter to vertical
accretion in tidal wetlands across a Chesapeake Bay subestuary. Journal of Marine Science and Engineering 9(7):
751.
Arias-Ortiz, A., Oikawa, P. Y., Carlin, J., Masque, P., Shahan, J., Kanneg, S.,... and Baldocchi, D. D. (2021) Tidal and
nontidal marsh restoration: a trade-off between carbon sequestration, methane emissions, and soil
accretion. Journal of Geophysical Research: Biogeosciences, 126(12): e2021JG006573.
Arriola, J. M., and Cable, J. E. (2017) Variations in carbon burial and sediment accretion along a tidal creek in a
Florida salt marsh. Limnology and Oceanography 62(S1): S15-S28.
Baustian, M. M., Stagg, C. L., Perry, C. L., Moss, L. C., and Carruthers, T. J. (2021) Long-term carbon sinks in marsh
soils of coastal Louisiana are at risk to wetland loss. Journal of Geophysical Research: Biogeosciences 126(3):
e2020JG005832.
Bianchi, T. S., Allison, M. A., Zhao, J., Li, X., Comeaux, R. S., Feagin, R. A., & Kulawardhana, R. W. (2013) Historical
reconstruction of mangrove expansion in the Gulf of Mexico: linking climate change with carbon sequestration in
coastal wetlands. Estuarine, Coastal and Shelf Science 119: 7-16.
Boyd, B. (2012) Comparison of sediment accumulation and accretion in impounded and unimpounded marshes of
the Delaware Estuary. Doctoral dissertation, University of Delaware.
Boyd, B. M. and Sommerfield, C. K. (2016) Marsh accretion and sediment accumulation in a managed tidal wetland
complex of Delaware Bay. Ecological Engineering, 92: 37-46.
Boyd, B. M., Sommerfield, C. K., and Elsey-Quirk, T. (2017) Hydrogeomorphic influences on salt marsh sediment
References and Abbreviations 10-95
-------
accumulation and accretion in two estuaries of the US Mid-Atlantic coast. Marine Geology, 383: 132-145.
Breithaupt, J. L., Smoak, J. M., Smith III, T. J., and Sanders, C. J. (2014) Temporal variability of carbon and nutrient
burial, sediment accretion, and mass accumulation over the past century in a carbonate platform mangrove forest
of the Florida Everglades. Journal of Geophysical Research: Biogeosciences, 119(10): 2032-2048.
Byrd, K. B., Ballanti, L. R., Thomas, N. M., Nguyen, D. K., Holmquist, J. R., Simard, M., Windham-Myers, L., Schile, L.
M., Parker, V. T.,... and Castaneda-Moya, E. (2017) Biomass/Remote Sensing dataset: 30m resolution tidal marsh
biomass samples and remote sensing data for six regions in the conterminous United States: U.S. Geological Survey
data release, https://doi.org/10.5066/F77943K8.
Byrd, K. B., Ballanti, L., Thomas, N., Nguyen, D., Holmquist, J.R., Simard, M., and Windham-Myers, L. (2018) A
remote sensing-based model of tidal marsh aboveground carbon stocks for the conterminous United States. ISPRS
Journal of Photogrammetry and Remote Sensing 139: 255-271.
Byrd, K. B., Ballanti, L., Thomas, N., Nguyen, D., Holmquist, J.R., Simard, M., and Windham-Myers, L. (2020)
Corrigendum to "A remote sensing-based model of tidal marsh aboveground carbon stocks for the conterminous
United States". ISPRS Journal of Photogrammetry and Remote Sensing 166: 63-67.
Callaway, J. C., Borgnis, E. L., Turner, R. E. & Milan, C. S. (2012a) Carbon sequestration and sediment accretion in
San Francisco Bay tidal wetlands. Estuaries and Coasts 35(5): 1163-1181.
Callaway, J. C., Borgnis, E. L., Turner, R. E., Milan, C. S., Goodfriend, W., & Richmond, S. (2012b) "Wetland Sediment
Accumulation at Corte Madera Marsh and Muzzi Marsh". San Francisco Bay Conservation and Development
Commission.
Church, T. M., Sommerfield, C. K., Velinsky, D. J., Point, D., Benoit, C., Amouroux, D. & Donard, O. F. X. (2006)
Marsh sediments as records of sedimentation, eutrophication and metal pollution in the urban Delaware Estuary.
Marine Chemistry 102(1-2): 72-95.
Couvillion, B. R., Barras, J. A., Steyer, G. D., Sleavin, W., Fischer, M., Beck, H., & Heckman, D. (2011) Land area
change in coastal Louisiana (1932 to 2010) (pp. 1-12). U.S. Department of the Interior, U.S. Geological Survey.
Couvillion, B. R., Fischer, M. R., Beck, H. J. and Sleavin, W. J. (2016) Spatial Configuration Trends in Coastal
Louisiana from 1986 to 2010. Wetlands 1-13.
Craft, C. B., & Richardson, C. J. (1998) Recent and long-term organic soil accretion and nutrient accumulation in the
Everglades. Soil Science Society of America Journal 62(3): 834-843.
Crooks, S., Findsen, J., Igusky, K., Orr, M. K. and Brew, D. (2009) Greenhouse Gas Mitigation Typology Issues Paper:
Tidal Wetlands Restoration. Report by PWA and SAIC to the California Climate Action Reserve.
Crooks, S., Rybczyk, J., O'Connell, K., Devier, D. L, Poppe, K., Emmett-Mattox, S. (2014) Coastal Blue Carbon
Opportunity Assessment for the Snohomish Estuary: The Climate Benefits of Estuary Restoration. Report by
Environmental Science Associates, Western Washington University, EarthCorps, and Restore America's Estuaries.
DeLaune, R. D., & White, J. R. (2012) Will coastal wetlands continue to sequester carbon in response to an increase
in global sea level?: A case study of the rapidly subsiding Mississippi river deltaic plain. Climatic Change, 110(1),
297-314.
Drexler, J. Z., de Fontaine, C. S., and Brown, T. A. (2009) Peat accretion histories during the past 6,000 years in
marshes of the Sacramento-San Joaquin Delta, CA, USA. Estuaries and Coasts 32: 871-892.
Drexler, J. Z., Krauss, K. W., Sasser, M. C., Fuller, C. C., Swarzenski, C. M., Powell, A.,... and Orlando, J. (2013) A
long-term comparison of carbon sequestration rates in impounded and naturally tidal freshwater marshes along
the lower Waccamaw River, South Carolina. Wetlands 33: 965-974.
Drexler, J. Z., Woo, I., Fuller, C. C., and Nakai, G. (2019) Carbon accumulation and vertical accretion in a restored
versus historic salt marsh in southern Puget Sound, Washington, United States. Restoration Ecology 27(5): 1117-
1127.
10-96 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
Ensign, S. H., Noe, G. B., Hupp, C. R., and Skalak, K. J. (2015) Head-of-tide bottleneck of particulate material
transport from watersheds to estuaries. Geophysical Research Letters 42(24): 10-671.
Gerlach, M. J., Engelhart, S. E., Kemp, A. C., Moyer, R. P., Smoak, J. M., Bernhardt, C. E., and Cahill, N. (2017)
Reconstructing Common Era relative sea-level change on the Gulf Coast of Florida. Marine Geology 390: 254-269.
Giblin, A., Forbrich, I., & LTER, P. I. E. (2018) PIE LTER high marsh sediment chemistry and activity measurements,
Nelson Island Creek marsh, Rowley, MA.
Holmquist, J. R., Windham-Myers, L, Bliss, N., Crooks, S., Morris, J. T., Megonigal, J. P. & Woodrey, M. (2018)
Accuracy and Precision of Tidal Wetland Soil Carbon Mapping in the Conterminous United States. Scientific reports
8(1): 9478.
Hu, Z., Lee, J. W., Chandran, K., Kim, S. and Khanal, S. K. (2012) N2O Emissions from Aquaculture: A Review.
Environmental Science & Technology 46(12): 6470-6480.
Hussein, A. H., Rabenhorst, M. C. & Tucker, M. L. (2004) Modeling of carbon sequestration in coastal marsh soils.
Soil Science Society of America Journal 68(5): 1786-1795.
IPCC (2014) 2013 Supplement to the 2006IPCC Guidelines for National Greenhouse Gas Inventories: Wetlands.
Hiraishi, T., Krug, T., Tanabe, K., Srivastava, N., Baasansuren, J., Fukuda, M. and Troxler, T.G. (eds.). Published: IPCC,
Switzerland.
IPCC (2013) Climate Change 2013: The Physical Science Basis. Contribution of Working Group I to the Fifth
Assessment Report of the Intergovernmental Panel on Climate Change. Stocker, T., Qin, D., Plattner, G.-K., Tignor,
M. Allen, S.K., Boschung, J., Nauels, A., Xia, Y., Bex, V. and Midgley, P.M. (eds.). Cambridge University Press,
Cambridge, United Kingdom and New York, NY, USA.
IPCC (2006) IPCC Guidelines for National Greenhouse Gas Inventories. Prepared by the National Greenhouse Gas
Inventories Programme, Eggleston H.S., Buendia L, Miwa K., NgaraT. and Tanabe K. (eds). IGES, Japan.
IPCC (2003) Good Practice Guidance for Land Use, Land-Use Change and Forestry. LUCF Sector Good Practice
Guidance, Chapter 3. Penman, J., Gytarsky, M., Hiraishi, T., Krug, T., Kruger, D., Pipatti, R., Buendia, L, Miwa, K.,
Ngara, T., Tanabe, K. and Wagner, F. (eds). Institute of Global Environmental Strategies (IGES), on behalf of the
Intergovernmental Panel on Climate Change (IPCC): Hayama, Japan.
IPCC (2000) Good Practice Guidance and Uncertainty Management in National Greenhouse Gas Inventories.
Quantifying Uncertainties in Practice, Chapter 6. Penman, J., Kruger, D., Galbally, I., Hiraishi, T., Nyenzi, B.,
Emmanuel, S., Buendia, L., Hoppaus, R., Martinsen, T., Meijer, J., Miwa, K. and Tanabe, K. (eds). Institute of Global
Environmental Strategies (IGES), on behalf of the Intergovernmental Panel on Climate Change (IPCC): Hayama,
Japan.
Kearney, M. S. & Stevenson, J. C. (1991) Island land loss and marsh vertical accretion rate evidence for historical
sea-level changes in Chesapeake Bay. Journal of Coastal Research 7(2): 403-415.
Kemp, A. C., Sommerfield, C. K., Vane, C. H., Horton, B. P., Chenery, S., Anisfeld, S., and Nikitina, D. (2012) Use of
lead isotopes for developing chronologies in recent salt-marsh sediments. Quaternary Geochronology 12: 40-49.
Koster, D., Lichter, J., Lea, P. D., & Nurse, A. (2007) Historical eutrophication in a river-estuary complex in mid-
coast Maine. Ecological Applications 17(3): 765-778.
Krauss, K. W., Noe, G. B., Duberstein, J. A., Conner, W. H., Stagg, C. L, and Jones, M. C. (2018) Carbon budget
assessment of tidal freshwater forested wetland and oligohaline marsh ecosystems along the Waccamaw and
Savannah Rivers, USA (2005-2016). US Geological Survey Data Release. https://doi.Org/10.5066/F7TIVI7930.
Lagomasino, D., Corbett, D. R., and Walsh, J. P. (2013) Influence of wind-driven inundation and coastal
geomorphology on sedimentation in two microtidal marshes, Pamlico River Estuary, NC. Estuaries and Coasts 36:
1165-1180.
Lu, M & Megonigal, J. P. (2017) Final Report for RAE Baseline Assessment Project. Memo to Silvestrum Climate
References and Abbreviations 10-97
-------
Associates by Smithsonian Environmental Research Center, Maryland.
Lynch, J. C. (1989) Sedimentation and nutrient accumulation in mangrove ecosystems of the Gulf of Mexico. M.S.
thesis, Univ. of Southwestern Louisiana, Lafayette, LA.
Luk, S. Y., Todd-Brown, K., Eagle, M., McNichol, A. P., Sanderman, J., Gosselin, K., and Spivak, A. C. (2021) Soil
organic carbon development and turnover in natural and disturbed salt marsh environments. Geophysical
Research Letters 48(2): e2020GL090287.
Marchio, D. A., Savarese, M., Bovard, B., & Mitsch, W. J. (2016) Carbon sequestration and sedimentation in
mangrove swamps influenced by hydrogeomorphic conditions and urbanization in Southwest Florida. Forests 7:
116-135.
McCombs, J. W., Herold, N. D., Burkhalter, S. G. and Robinson C. J. (2016) Accuracy Assessment of NOAA Coastal
Change Analysis Program 2006-2010 Land Cover and Land Cover Change Data. Photogrammetric Engineering &
Remote Sensing. 82:711-718.
McTigue, N., Davis, J., Rodriguez, A. B., McKee, B., Atencio, A., and Currin, C. (2019) Sea level rise explains changing
carbon accumulation rates in a salt marsh over the past two millennia. JGR Biogeosciences.
Merrill, J. Z. (1999) Tidal Freshwater Marshes as Nutrient Sinks: particulate Nutrient Burial and Denitrification.
Ph.D. Dissertation, University of Maryland, College Park, MD, 342 pp.
Miller, C. B., Rodriguez, A. B., Bost, M. C., McKee, B. A., and McTigue, N. D. (2022) Carbon accumulation rates are
highest at young and expanding salt marsh edges. Communications Earth & Environment 3(1): 173.
National Marine Fisheries Service (2022). Fisheries of the United States, 2020. U.S. Department of Commerce,
NOAA Current Fishery Statistics No. 2020. Available at: https://www.fisheries.noaa.gov/ national/sustainable-
fisheries/fisheries-united-states.
National Oceanic and Atmospheric Administration, Office for Coastal Management (2020) Coastal Change Analysis
Program (C-CAP) Regional Land Cover. Charleston, SC: NOAA Office for Coastal Management. Accessed October
2020 at www.coast.noaa.gov/htdata/rasterl/landcover/bulkdownload/30m lc/.
Noe, G. B., Hupp, C. R., Bernhardt, C. E., & Krauss, K. W. (2016) Contemporary deposition and long-term
accumulation of sediment and nutrients by tidal freshwater forested wetlands impacted by sea level rise. Estuaries
and Coasts 39(4): 1006-1019.
Orson, R. A., Simpson, R. L, & Good, R. E. (1990) Rates of sediment accumulation in a tidal freshwater marsh.
Journal of Sedimentary Research 60(6): 859-869.
Orson, R., Warren, R. & Niering, W. (1998) Interpreting sea level rise and rates of vertical marsh accretion in a
southern New England tidal salt marsh. Estuarine, Coastal and Shelf Science 47(4): 419-429.
Peck, E. K., Wheatcroft, R. A., and Brophy, L. S. (2020) Controls on sediment accretion and blue carbon burial in
tidal saline wetlands: insights from the Oregon Coast, USA. Journal of Geophysical Research: Biogeosciences
125(2): e2019JG005464.
Poppe, K. L., and Rybczyk, J. M. (2021) Tidal marsh restoration enhances sediment accretion and carbon
accumulation in the Stillaguamish River estuary, Washington. PloS one 16(9): e0257244.
Roman, C., Peck, J., Allen, J., King, J. & Appleby, P. (1997) Accretion of a New England (USA) salt marsh in response
to inlet migration, storms, and sea-level rise. Estuarine, Coastal and Shelf Science 45(6): 717-727.
Smith, K. E., Flocks, J. G., Steyer, G. D., and Piazza, S. C. (2015) Wetland Paleoecological Study of Southwest Coastal
Louisiana: Sediment Cores and Diatom Calibration Dataset. US Department of the Interior, US Geological Survey.
Thorn, R. M. (1992) Accretion rates of low intertidal salt marshes in the Pacific Northwest. Wetlands 12: 147-156.
Vaughn, D. R., Bianchi, T. S., Shields, M. R., Kenney, W. F., and Osborne, T. Z. (2020) Increased organic carbon burial
in northern Florida mangrove-salt marsh transition zones. Global Biogeochemical Cycles 34(5): e2019GB006334.
10-98 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
Villa, J. A. & Mitsch W. J. (2015) Carbon sequestration in different wetland plant communities of Southwest Florida.
International Journal for Biodiversity Science, Ecosystems Services and Management 11: 17-28
Watson, E. B., and Byrne, R. (2013) Late Holocene Marsh Expansion in Southern San Francisco Bay, California.
Estuaries and Coasts 36: 643-653.
Weis, D. A., Callaway, J. C., and Gersberg, R. M. (2001) Vertical accretion rates and heavy metal chronologies in
wetland sediments of the Tijuana Estuary. Estuaries 24: 840-850.
Weston, N. B., Neubauer, S. C., Velinsky, D. J., & Vile, M. A. (2014) Net ecosystem carbon exchange and the
greenhouse gas balance of tidal marshes along an estuarine salinity gradient. Biogeochemistry 120:163-189.
Weston, N. B., Rodriguez, E., Donnelly, B., Solohin, E., Jezycki, K., Demberger, S.,... and Craft, C. B. (2023) Recent
acceleration of wetland accretion and carbon accumulation along the US East Coast. Earth's Future 11(3):
e2022EF003037.
Wetlands Remaining Wetlands: Flooded Land Remaining
Flooded Land
Abril, G., Gu'erin, F., Richard, S., Delmas, R., Galy-Lacaux, C., Gosse, P., et al. (2005) Carbon dioxide and methane
emissions and the carbon budget of a 10-year old tropical reservoir (Petit Saut, French Guiana). Global
Biogeochem. Cycles 19 (GB4007), 1-16. https://doi.org/10.1029/2005GB002457.
Barros, N., Cole, J.J., Tranvik, L.J., Prairie, Y.T., Bastviken, D., Huszar, V.L.M., et al. (2011) Carbon emission from
hydroelectric reservoirs linked to reservoir age and latitude. Nat. Geosci. 4 (9), 593-596.
https://doi.org/10.1038/ngeol211.
Davis, D. W. (1973) Louisiana Canals and Their Influence on Wetland Development. Louisiana State University and
Agricultural & Mechanical College. LSU Historical Dissertations and Theses. 2386., Louisiana State University.
IPCC (2019) 2019 Refinement to the 2006 Guidelines for National Greenhouse Gas Inventories. Wetlands, Chapter
7. Lovelock, C. E., Evans, C., Barros, N., Prairie, Y. T., Aim, J., Bastviken, D., Beaulieu, J. J., Garneau, M., Harby, A.,
Harrison, J. A., Pare, David, Raadal, Hanne Lerche, Sherman, B., Zhang, Chengyi, Ogle, S. M.IPCC (2013) 2013
Supplement to the 2006 IPCC Guidelines for National Greenhouse Gas Inventories: Wetlands. Hiraishi, T., Krug, T.,
Tanabe, K., Srivastava, N., Baasansuren, J., Fukuda, M. and Troxler, T.G. (eds). In: IPCC, Switzerland.
IPCC (2006) IPCC Guidelines for National Greenhouse Gas Inventories. Prepared by the National Greenhouse Gas
Inventories Programme, Eggleston H.S., Buendia L, Miwa K., NgaraT. and Tanabe K. (eds). IGES, Japan.
IPCC (2003) Good Practice Guidance for Land Use, Land-Use Change and Forestry. LUCF Sector Good Practice
Guidance, Chapter 3. Penman, J., Gytarsky, M., Hiraishi, T., Krug, T., Kruger, D., Pipatti, R., Buendia, L, Miwa, K.,
Ngara, T., Tanabe, K. and Wagner, F. (eds). Institute of Global Environmental Strategies (IGES), on behalf of the
Intergovernmental Panel on Climate Change (IPCC): Hayama, Japan.
Lehner B, Reidy Liermann C, Revenga C, Vorosmarty C, Fekete B, Crouzet P, Doll P, et al. (2011b) Global Reservoir
and Dam Database, Version 1 (GRanDvl): Dams, Revision 01. In: Palisades, NY: NASA Socioeconomic Data and
Applications Center (SEDAC).
Prairie, Y. T., et al. (2017) The GHG ReservoirTool (G-res) User guide. UNESCO/IHA research project on the GHG
status of freshwater reservoirs. Joint publication of the UNESCO Chair in Global Environmental Change and the
International Hydropower Association: 38.
Teodoru, C.R., Bastien, J., Bonneville, M.C., Del Giorgio, P.a., Demarty, M., Garneau, M., et al., 2012. The net
carbon footprint of a newly created boreal hydroelectric reservoir. Global Biogeochem. Cycles 26 (GB2016), 1-14.
https://doi.org/10.1029/2011GB004187.
References and Abbreviations 10-99
-------
Land Converted to Wetlands: Emissions and Removals from
land Converted to Vegetated Coastal Wetlands
Abbott, K. M., Elsey-Quirk, T., and DeLaune, R. D. (2019) Factors influencing blue carbon accumulation across a 32-
year chronosequence of created coastal marshes. Ecosphere, 10(8): e02828.
Allen, J. R., Cornwell, J. C., and Baldwin, A. H. (2021) Contributions of organic and mineral matter to vertical
accretion in tidal wetlands across a Chesapeake Bay subestuary. Journal of Marine Science and Engineering 9(7):
751.
Arias-Ortiz, A., Oikawa, P. Y., Carlin, J., Masque, P., Shahan, J., Kanneg, S.,... and Baldocchi, D. D. (2021) Tidal and
nontidal marsh restoration: a trade-off between carbon sequestration, methane emissions, and soil
accretion. Journal of Geophysical Research: Biogeosciences, 126(12): e2021JG006573.
Arriola, J. M., and Cable, J. E. (2017) Variations in carbon burial and sediment accretion along a tidal creek in a
Florida salt marsh. Limnology and Oceanography 62(S1): S15-S28.
Baustian, M. M., Stagg, C. L., Perry, C. L., Moss, L. C., and Carruthers, T. J. (2021) Long-term carbon sinks in marsh
soils of coastal Louisiana are at risk to wetland loss. Journal of Geophysical Research: Biogeosciences 126(3):
e2020JG005832.
Bianchi, T. S., Allison, M. A., Zhao, J., Li, X., Comeaux, R. S., Feagin, R. A., & Kulawardhana, R. W. (2013) Historical
reconstruction of mangrove expansion in the Gulf of Mexico: linking climate change with carbon sequestration in
coastal wetlands. Estuarine, Coastal and Shelf Science 119: 7-16.
Boyd, B. (2012) Comparison of sediment accumulation and accretion in impounded and unimpounded marshes of
the Delaware Estuary. Doctoral dissertation, University of Delaware.
Boyd, B. M. and Sommerfield, C. K. (2016) Marsh accretion and sediment accumulation in a managed tidal wetland
complex of Delaware Bay. Ecological Engineering, 92: 37-46.
Boyd, B. M., Sommerfield, C. K., and Elsey-Quirk, T. (2017) Hydrogeomorphic influences on salt marsh sediment
accumulation and accretion in two estuaries of the US Mid-Atlantic coast. Marine Geology, 383: 132-145.
Breithaupt, J. L., Smoak, J. M., Smith III, T. J., and Sanders, C. J. (2014) Temporal variability of carbon and nutrient
burial, sediment accretion, and mass accumulation over the past century in a carbonate platform mangrove forest
of the Florida Everglades. Journal of Geophysical Research: Biogeosciences, 119(10): 2032-2048.
Byrd, K. B., Ballanti, L. R., Thomas, N. M., Nguyen, D. K., Holmquist, J. R., Simard, M., Windham-Myers, L., Schile, L.
M., Parker, V. T.,... and Castaneda-Moya, E. (2017) Biomass/Remote Sensing dataset: 30m resolution tidal marsh
biomass samples and remote sensing data for six regions in the conterminous United States: U.S. Geological Survey
data release, https://doi.org/10.5066/F77943K8.
Byrd, K. B., Ballanti, L., Thomas, N., Nguyen, D., Holmquist, J.R., Simard, M., and Windham-Myers, L. (2018) A
remote sensing-based model of tidal marsh aboveground carbon stocks for the conterminous United States. ISPRS
Journal of Photogrammetry and Remote Sensing 139: 255-271.
Byrd, K. B., Ballanti, L., Thomas, N., Nguyen, D., Holmquist, J.R., Simard, M., and Windham-Myers, L. (2020)
Corrigendum to "A remote sensing-based model of tidal marsh aboveground carbon stocks for the conterminous
United States". ISPRS Journal of Photogrammetry and Remote Sensing 166: 63-67.
Callaway, J. C., Borgnis, E. L., Turner, R. E. & Milan, C. S. (2012a) Carbon sequestration and sediment accretion in
San Francisco Bay tidal wetlands. Estuaries and Coasts 35(5): 1163-1181.
Callaway, J. C., Borgnis, E. L., Turner, R. E., Milan, C. S., Goodfriend, W., & Richmond, S. (2012b) "Wetland Sediment
Accumulation at Corte Madera Marsh and Muzzi Marsh". San Francisco Bay Conservation and Development
Commission.
10-100 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
Church, T. M., Sommerfield, C. K., Velinsky, D. J., Point, D., Benoit, C., Amouroux, D. & Donard, O. F. X. (2006)
Marsh sediments as records of sedimentation, eutrophication and metal pollution in the urban Delaware Estuary.
Marine Chemistry 102(1-2): 72-95.
Craft, C. B., & Richardson, C. J. (1998) Recent and long-term organic soil accretion and nutrient accumulation in the
Everglades. Soil Science Society of America Journal 62(3): 834-843.
Crooks, S., Rybczyk, J., O'Connell, K., Devier, D.L., Poppe, K., Emmett-Mattox, S. (2014) Coastal Blue Carbon
Opportunity Assessment for the Snohomish Estuary: The Climate Benefits of Estuary Restoration. Report by
Environmental Science Associates, Western Washington University, EarthCorps, and Restore America's Estuaries.
Drexler, J. Z., de Fontaine, C. S., and Brown, T. A. (2009) Peat accretion histories during the past 6,000 years in
marshes of the Sacramento-San Joaquin Delta, CA, USA. Estuaries and Coasts 32: 871-892.
Drexler, J. Z., Krauss, K. W., Sasser, M. C., Fuller, C. C., Swarzenski, C. M., Powell, A.,... and Orlando, J. (2013) A
long-term comparison of carbon sequestration rates in impounded and naturally tidal freshwater marshes along
the lower Waccamaw River, South Carolina. Wetlands 33: 965-974.
Drexler, J. Z., Woo, I., Fuller, C. C., and Nakai, G. (2019) Carbon accumulation and vertical accretion in a restored
versus historic salt marsh in southern Puget Sound, Washington, United States. Restoration Ecology 27(5): 1117-
1127.
Ensign, S. H., Noe, G. B., Hupp, C. R., and Skalak, K. J. (2015) Head-of-tide bottleneck of particulate material
transport from watersheds to estuaries. Geophysical Research Letters 42(24): 10-671.
Gerlach, M. J., Engelhart, S. E., Kemp, A. C., Moyer, R. P., Smoak, J. M., Bernhardt, C. E., and Cahill, N. (2017)
Reconstructing Common Era relative sea-level change on the Gulf Coast of Florida. Marine Geology 390: 254-269.
Giblin, A., Forbrich, I., & LTER, P. I. E. (2018) PIE LTER high marsh sediment chemistry and activity measurements,
Nelson Island Creek marsh, Rowley, MA.
Hussein, A. H., Rabenhorst, M. C. & Tucker, M. L. (2004) Modeling of carbon sequestration in coastal marsh soils.
Soil Science Society of America Journal 68(5): 1786-1795.
IPCC (2019) Refinement to the 2006IPCC Guidelines for National Greenhouse Gas Inventories, Volume 4:
Agriculture, Forestry, and Other Land Use. Calvo Buendia, E., Tanabe K., Kranjc, A., Baasansuren, J., Fukuda, M.,
Ngarize, S., Osako, A., Pyrozhenko, Y., Shermanau, P., & Federici, S. (eds). IPCC, Switzerland.
IPCC (2006) 2006 IPCC Guidelines for National Greenhouse Gas Inventories. Prepared by the National Greenhouse
Gas Inventories Programme, H.S.Eggleston, L. Buendia, K. Miwa, T. Ngara & K. Tanabe (eds). IGES, Japan.
IPCC (2014) 2013 Supplement to the 2006 IPCC Guidelines for National Greenhouse Gas Inventories: Wetlands.
Hiraishi, T., Krug, T., Tanabe, K., Srivastava, N., Baasansuren, J., Fukuda, M. and Troxler, T.G. (eds.). Published: IPCC,
Switzerland.
IPCC (2013) Climate Change 2013: The Physical Science Basis. Contribution of Working Group I to the Fifth
Assessment Report of the Intergovernmental Panel on Climate Change. Stocker, T., Qin, D., Plattner, G.-K., Tignor,
M. Allen, S.K., Boschung, J., Nauels, A., Xia, Y., Bex, V. and Midgley, P.M. (eds.). Cambridge University Press,
Cambridge, United Kingdom and New York, NY, USA.
IPCC (2003) Good Practice Guidance for Land Use, Land-Use Change and Forestry. LUCF Sector Good Practice
Guidance, Chapter 3. Penman, J., Gytarsky, M., Hiraishi, T., Krug, T., Kruger, D., Pipatti, R., Buendia, L, Miwa, K.,
Ngara, T., Tanabe, K. & F. Wagner (eds). Institute of Global Environmental Strategies (IGES), on behalf of the
Intergovernmental Panel on Climate Change (IPCC): Hayama, Japan.
IPCC (2000) Good Practice Guidance and Uncertainty Management in National Greenhouse Gas Inventories.
Quantifying Uncertainties in Practice, Chapter 6. Penman, J and Kruger, D and Galbally, I and Hiraishi, T and Nyenzi,
B and Emmanuel, S and Buendia, L and Hoppaus, R and Martinsen, T and Meijer, J and Miwa, K and Tanabe, K
(eds). Institute of Global Environmental Strategies (IGES), on behalf of the Intergovernmental Panel on Climate
References and Abbreviations 10-101
-------
Change (IPCC): Hayama, Japan.
Kearney, M. S. & Stevenson, J. C. (1991) Island land loss and marsh vertical accretion rate evidence for historical
sea-level changes in Chesapeake Bay. Journal of Coastal Research 7(2): 403-415.
Kemp, A. C., Sommerfield, C. K., Vane, C. H., Horton, B. P., Chenery, S., Anisfeld, S., and Nikitina, D. (2012) Use of
lead isotopes for developing chronologies in recent salt-marsh sediments. Quaternary Geochronology 12: 40-49.
Koster, D., Lichter, J., Lea, P. D., & Nurse, A. (2007) Historical eutrophication in a river-estuary complex in mid-
coast Maine. Ecological Applications 17(3): 765-778.
Krauss, K. W., Noe, G. B., Duberstein, J. A., Conner, W. H., Stagg, C. L, and Jones, M. C. (2018) Carbon budget
assessment of tidal freshwater forested wetland and oligohaline marsh ecosystems along the Waccamaw and
Savannah Rivers, USA (2005-2016). US Geological Survey Data Release. https://doi.Org/10.5066/F7TIVI7930.
Lagomasino, D., Corbett, D. R., and Walsh, J. P. (2013) Influence of wind-driven inundation and coastal
geomorphology on sedimentation in two microtidal marshes, Pamlico River Estuary, NC. Estuaries and Coasts 36:
1165-1180.
Lu, M & Megonigal, J.P. (2017) Final Report for RAE Baseline Assessment Project. Memo to Silvestrum Climate
Associates by Smithsonian Environmental Research Center, Maryland.
Lynch, J. C. (1989) Sedimentation and nutrient accumulation in mangrove ecosystems of the Gulf of Mexico, M.S.
thesis, Univ. of Southwestern Louisiana, Lafayette, La.
Luk, S. Y., Todd-Brown, K., Eagle, M., McNichol, A. P., Sanderman, J., Gosselin, K., and Spivak, A. C. (2021) Soil
organic carbon development and turnover in natural and disturbed salt marsh environments. Geophysical
Research Letters 48(2): e2020GL090287.
Marchio, D.A., Savarese, M., Bovard, B., & Mitsch, W.J. (2016) Carbon sequestration and sedimentation in
mangrove swamps influenced by hydrogeomorphic conditions and urbanization in Southwest Florida. Forests 7:
116-135.
McCombs, J.W., Herold, N.D., Burkhalter, S.G. and Robinson C.J., (2016) Accuracy Assessment of NOAA Coastal
Change Analysis Program 2006-2010 Land Cover and Land Cover Change Data. Photogrammetric Engineering &
Remote Sensing. 82:711-718.
McTigue, N., Davis, J., Rodriguez, A. B., McKee, B., Atencio, A., and Currin, C. (2019) Sea level rise explains changing
carbon accumulation rates in a salt marsh over the past two millennia. JGR Biogeosciences.
Merrill, J. Z. (1999) Tidal Freshwater Marshes as Nutrient Sinks: particulate Nutrient Burial and Denitrification.
Ph.D. Dissertation, University of Maryland, College Park, MD, 342pp.
Miller, C. B., Rodriguez, A. B., Bost, M. C., McKee, B. A., and McTigue, N. D. (2022) Carbon accumulation rates are
highest at young and expanding salt marsh edges. Communications Earth & Environment 3(1): 173.
National Oceanic and Atmospheric Administration, Office for Coastal Management (2020) Coastal Change Analysis
Program (C-CAP) Regional Land Cover. Charleston, SC: NOAA Office for Coastal Management. Accessed October
2020 at www.coast.noaa.gov/htdata/rasterl/landcover/bulkdownload/30m lc/.
Noe, G. B., Hupp, C. R., Bernhardt, C. E., & Krauss, K. W. (2016) Contemporary deposition and long-term
accumulation of sediment and nutrients by tidal freshwater forested wetlands impacted by sea level rise. Estuaries
and Coasts 39(4): 1006-1019.
Orson, R. A., Simpson, R. L., & Good, R. E. (1990) Rates of sediment accumulation in a tidal freshwater marsh.
Journal of Sedimentary Research 60(6): 859-869.
Orson, R., Warren, R. & Niering, W. (1998) Interpreting sea level rise and rates of vertical marsh accretion in a
southern New England tidal salt marsh. Estuarine, Coastal and Shelf Science 47(4): 419-429.
Peck, E. K., Wheatcroft, R. A., and Brophy, L. S. (2020) Controls on sediment accretion and blue carbon burial in
10-102 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
tidal saline wetlands: insights from the Oregon Coast, USA. Journal of Geophysical Research: Biogeosciences
125(2): e2019JG005464.
Poppe, K. L., and Rybczyk, J. M. (2021) Tidal marsh restoration enhances sediment accretion and carbon
accumulation in the Stillaguamish River estuary, Washington. PloS one 16(9): e0257244.
Roman, C., Peck, J., Allen, J., King, J. & Appleby, P. (1997) Accretion of a New England (USA) salt marsh in response
to inlet migration, storms, and sea-level rise. Estuarine, Coastal and Shelf Science 45(6): 717-727.
Smith, K. E., Flocks, J. G., Steyer, G. D., and Piazza, S. C. (2015) Wetland Paleoecological Study of Southwest Coastal
Louisiana: Sediment Cores and Diatom Calibration Dataset. US Department of the Interior, US Geological Survey.
Thorn, R. M. (1992 Accretion rates of low intertidal salt marshes in the Pacific Northwest. Wetlands 12:147-156.
Vaughn, D. R., Bianchi, T. S., Shields, M. R., Kenney, W. F., and Osborne, T. Z. (2020) Increased organic carbon burial
in northern Florida mangrove-salt marsh transition zones. Global Biogeochemical Cycles 34(5): e2019GB006334.
Villa, J. A. & Mitsch W. J. (2015) "Carbon sequestration in different wetland plant communities of Southwest
Florida". International Journal for Biodiversity Science, Ecosystems Services and Management 11:17-28.
Watson, E. B., and Byrne, R. (2013) Late Holocene Marsh Expansion in Southern San Francisco Bay, California.
Estuaries and Coasts 36: 643-653.
Weis, D. A., Callaway, J. C., and Gersberg, R. M. (2001) Vertical accretion rates and heavy metal chronologies in
wetland sediments of the Tijuana Estuary. Estuaries 24: 840-850.
Weston, N. B., Neubauer, S. C., Velinsky, D. J., & Vile, M. A. (2014) Net ecosystem carbon exchange and the
greenhouse gas balance of tidal marshes along an estuarine salinity gradient. Biogeochemistry 120:163-189.
Weston, N. B., Rodriguez, E., Donnelly, B., Solohin, E., Jezycki, K., Demberger, S.,... and Craft, C. B. (2023) Recent
acceleration of wetland accretion and carbon accumulation along the US East Coast. Earth's Future 11(3):
e2022EF003037.
Land Converted to Wetlands: Land Converted to Flooded Land
Abril, G., Gu'erin, F., Richard, S., Delmas, R., Galy-Lacaux, C., Gosse, P., et al. (2005) Carbon dioxide and methane
emissions and the carbon budget of a 10-year old tropical reservoir (Petit Saut, French Guiana). Global
Biogeochem. Cycles 19 (GB4007), 1-16. https://doi.org/10.1029/2005GB002457.
Barros, N., Cole, J.J., Tranvik, L.J., Prairie, Y.T., Bastviken, D., Huszar, V.L.M., et al. (2011). Carbon emission from
hydroelectric reservoirs linked to reservoir age and latitude. Nat. Geosci. 4 (9), 593-596.
https://doi.org/10.1038/ngeol211.
IPCC (2019) 2019 Refinement to the 2006 Guidelines for National Greenhouse Gas Inventories. Wetlands, Chapter
7. Lovelock, C. E., Evans, C., Barros, N., Prairie, Y. T., Aim, J., Bastviken, D., Beaulieu, J. J., Garneau, M., Harby, A.,
Harrison, J. A., Pare, David, Raadal, Hanne Lerche, Sherman, B., Zhang, Chengyi, Ogle, S. M.
IPCC (2013) 2013 Supplement to the 2006 IPCC Guidelines for National Greenhouse Gas Inventories: Wetlands.
Hiraishi, T., Krug, T., Tanabe, K., Srivastava, N., Baasansuren, J., Fukuda, M. and Troxler, T.G. (eds). In: IPCC,
Switzerland.
IPCC (2006) 2006 IPCC Guidelines for National Greenhouse Gas Inventories. Prepared by the National Greenhouse
Gas Inventories Programme, H.S.Eggleston, L. Buendia, K. Miwa, T. Ngara & K. Tanabe (eds). IGES, Japan.
IPCC (2003) Good Practice Guidance for Land Use, Land-Use Change and Forestry. LUCF Sector Good Practice
Guidance, Chapter 3. Penman, J., Gytarsky, M., Hiraishi, T., Krug, T., Kruger, D., Pipatti, R., Buendia, L, Miwa, K.,
Ngara, T., Tanabe, K. and Wagner, F. (eds). Institute of Global Environmental Strategies (IGES), on behalf of the
Intergovernmental Panel on Climate Change (IPCC): Hayama, Japan.
References and Abbreviations 10-103
-------
Lehner B, Reidy Liermann C, Revenga C, Vorosmarty C, Fekete B, Crouzet P, Doll P, et al. (2011b) Global Reservoir
and Dam Database, Version 1 (GRanDvl): Dams, Revision 01. In: Palisades, NY: NASA Socioeconomic Data and
Applications Center (SEDAC).
Prairie, Y. T., et al. (2017) The GHG Reservoir Tool (G-res) User guide. UNESCO/IHA research project on the GHG
status of freshwater reservoirs. Joint publication of the UNESCO Chair in Global Environmental Change and the
International Hydropower Association: 38.
Teodoru, C.R., Bastien, J., Bonneville, M.C., Del Giorgio, P.a., Demarty, M., Garneau, M., et al. (2012). The net
carbon footprint of a newly created boreal hydroelectric reservoir. Global Biogeochem. Cycles 26 (GB2016), 1-14.
https://doi.org/10.1029/2011GB00418?.
Settlements Remaining Settlements: Soil Carbon Stock
Changes
AAPFCO (2016 through 2022) Commercial Fertilizers: 2013-2017. Association of American Plant Food Control
Officials. University of Missouri. Columbia, MO.
Armentano, T. V., and E.S. Menges (1986) Patterns of change in the carbon balance of organic soil-wetlands of the
temperate zone. Journal of Ecology 74: 755-774.
Brady, N.C. and R.R. Weil (1999) The Nature and Properties of Soils. Prentice Hall. Upper Saddle River, NJ, 881.
Brockwell, Peter J., and Richard A. Davis (2016) Introduction to time series and forecasting. Springer.
Fry, J., Xian, G., Jin, S., Dewitz, J., Homer, C., Yang, L, Barnes, C., Herold, N., and J. Wickham. (2011) Completion of
the 2006 National Land Cover Database for the Conterminous United States, PE&RS 77(9):858-864.
Homer, C., J. Dewitz, J. Fry, M. Coan, N. Hossain, C. Larson, N. Herold, A. McKerrow, J.N. VanDriel and J. Wickham.
(2007) Completion of the 2001 National Land Cover Database for the Conterminous United States.
Photogrammetric Engineering and Remote Sensing 73(4): 337-341.
Homer, C.G., Dewitz, J.A., Yang, L., Jin, S., Danielson, P., Xian, G., Coulston, J., Herold, N.D., Wickham, J.D., and
Megown, K. (2015) Completion of the 2011 National Land Cover Database for the conterminous United States-
Representing a decade of land cover change information. Photogrammetric Engineering and Remote Sensing
81(5):345-354.
IPCC (2006) 2006IPCC Guidelines for National Greenhouse Gas Inventories. The National Greenhouse Gas
Inventories Programme, The Intergovernmental Panel on Climate Change. H.S. Eggleston, L. Buendia, K. Miwa, T.
Ngara, and K. Tanabe (eds.). Hayama, Kanagawa, Japan.
Nelson, Mark D.; Riitters, Kurt H.; Coulston, John W.; Domke, Grant M.; Greenfield, Eric J.; Langner, Linda L.;
Nowak, David J.; O'Dea, Claire B.; Oswalt, Sonja N.; Reeves, Matthew C.; Wear, David N. 2020. Defining the United
States land base: a technical document supporting the USDA Forest Service 2020 RPA assessment. Gen. Tech. Rep.
NRS-191. Madison, Wl: U.S. Department of Agriculture, Forest Service, Northern Research Station. 70 p.
https://doi.org/10.2737/NRS-GTR-191.
NRCS (1999) Soil Taxonomy: A basic system of soil classification for making and interpreting soil surveys, 2nd
Edition. Agricultural Handbook Number 436, Natural Resources Conservation Service, U.S. Department of
Agriculture, Washington, D.C.
Nusser, S.M. and J.J. Goebel (1997) The national resources inventory: a long-term multi-resource monitoring
programme. Environmental and Ecological Statistics 4:181-204.
Ogle, S.M., M.D. Eve, F.J. Breidt, and K. Paustian (2003) Uncertainty in estimating land use and management
impacts on soil organic carbon storage for U.S. agroecosystems between 1982 and 1997. Global Change Biology
9:1521-1542.
10-104 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
Ogle, S.M., F.J. Breidt, M. Easter, S. Williams, K. Killian, and K. Paustian (2010) Scale and uncertainty in modeled soil
organic carbon stock changes for U.S. croplands using a process-based model. Global Change Biology 16:810-822.
Sarndal C-E, Swensson B, Wretman, J (1992). Model Assisted Survey Sampling. Springer, New York.
Soil Survey Staff (2020) Gridded Soil Survey Geographic (gSSURGO) Database for the Conterminous United States.
United States Department of Agriculture, Natural Resources Conservation Service, Accessed February 2020
(FY2020 official release), Available online at https://gdg.sc.egov.usda.gov/.
USDA-NRCS (2020) Summary Report: 2017 National Resources Inventory. Natural Resources Conservation Service,
Washington, DC, and Center for Survey Statistics and Methodology, Iowa State University, Ames, Iowa.
https://www.nrcs.usda.gov/wps/portal/nrcs/main/national/technical/nra/nri/results/.
Yang, L, Jin, S., Danielson, P., Homer, C., Gass, L., Bender, S. M., Case, A., Costello, C., Dewitz, J., Fry, J., Funk, M.,
Granneman, B., Liknes, G. C., Rigge, M. & Xian, G. (2018) A new generation of the United States National Land
Cover Database: Requirements, research priorities, design, and implementation strategies. ISPRS Journal of
Photogrammetry and Remote Sensing 146: 108-123.
Settlements Remaining Settlements: Changes in Carbon Stocks
in Settlement Trees
deVries R.E. (1987) A Preliminary Investigation of the Growth and Longevity of Trees in Central Park. M.S. thesis,
Rutgers University, New Brunswick, NJ.
Fleming, L.E. (1988) Growth Estimation of Street Trees in Central New Jersey. M.S. thesis, Rutgers University, New
Brunswick, NJ.
Frelich, L.E. (1992) Predicting Dimensional Relationships for Twin Cities Shade Trees. University of Minnesota,
Department of Forest Resources, St. Paul, MN, p. 33.
IPCC (2006) 2006IPCC Guidelines for National Greenhouse Gas Inventories. The National Greenhouse Gas
Inventories Programme, The Intergovernmental Panel on Climate Change. H.S. Eggleston, L. Buendia, K. Miwa, T.
Ngara, and K. Tanabe (eds.). Hayama, Kanagawa, Japan.
MRLC (2013) National Land Cover Database 2001 (NLCD2001). Available online at:
http://www.mrlc.gov/nlcd2001.php. Accessed August 2013.
Nowak, D.J. (1986) Silvics of an Urban Tree Species: Norway maple (Acer platanoides L.). M.S. thesis, College of
Environmental Science and Forestry, State University of New York, Syracuse, NY.
Nowak, D.J. (1994) Atmospheric carbon dioxide reduction by Chicago's urban forest. In: Chicago's Urban Forest
Ecosystem: Results of the Chicago Urban Forest Climate Project. E.G. McPherson, D.J. Nowak, and R.A. Rowntree
(eds.). General Technical Report NE-186. U.S. Department of Agriculture Forest Service, Radnor, PA. pp. 83-94.
Nowak, D.J. (2012) Contrasting natural regeneration and tree planting in 14 North American cities. Urban Forestry
and Urban Greening. 11: 374-382.
Nowak, D.J. and D.E. Crane (2002) Carbon storage and sequestration by urban trees in the United States.
Environmental Pollution 116(3):381-389.
Nowak, D.J. and E. Greenfield (2010) Evaluating the National Land Cover Database tree canopy and impervious
cover estimates across the conterminous United States: A comparison with photo-interpreted estimates.
Environmental Management. 46: 378-390.
Nowak, D.J. and E.J. Greenfield (2018a) U.S. urban forest statistics, values and projections. Journal of Forestry.
116(2):164-177
References and Abbreviations 10-105
-------
Nowak, D.J. and E.J. Greenfield (2018b) Declining urban and community tree cover in the United States. Urban
Forestry and Urban Greening. 32:32-55.
Nowak, D.J., D.E. Crane, J.C. Stevens, and M. Ibarra (2002) Brooklyn's Urban Forest. General Technical Report NE-
290. U.S. Department of Agriculture Forest Service, Newtown Square, PA.
Nowak, D.J., R.E. Hoehn, D.E. Crane, J.C. Stevens, J.T. Walton, and J. Bond (2008) Aground-based method of
assessing urban forest structure and ecosystem services. Arboric. Urb. For. 34(6): 347-358.
Nowak, D.J., E.J. Greenfield, R.E. Hoehn, and E. Lapoint (2013) Carbon storage and sequestration by trees in urban
and community areas of the United States." Environmental Pollution 178: 229-236.
Nowak, D.J. A.R. Bodine, R.E. Hoehn, C.B. Edgar, D.R. Hartel, T.W. Lister, T.J. Brandeis (2016) Austin's Urban Forest,
2014. USDA Forest Service, Northern Research Station Resources Bulletin. NRS-100. Newtown Square, PA. 55 p.
Nowak, D.J. A.R. Bodine, R.E. Hoehn, C.B. Edgar, G. Riley, D.R. Hartel, K.J. Dooley, S.M. Stanton, M.A. Hatfield, T.J.
Brandeis, T.W. Lister (2017) Houston's Urban Forest, 2015. USDA Forest Service, Southern Research Station
Resources Bulletin. SRS-211. Newtown Square, PA. 91 p.
Smith, W.B. and S.R. Shifley (1984) Diameter Growth, Survival, and Volume Estimates for Trees in Indiana and
Illinois. Research Paper NC-257. North Central Forest Experiment Station, U.S. Department of Agriculture Forest
Service, St. Paul, MN.
U.S. Department of Interior (2018) National Land Cover Database 2011 (NLCD2011). Accessed online August 16,
2018. Available online at: https://www.mrlc.gov/nlcdll leg.php.
Settlements Remaining Settlements: N20 Emissions from Soils
AAPFCO (2016 through 2022) Commercial Fertilizers: 2013-2017. Association of American Plant Food Control
Officials. University of Missouri. Columbia, MO.
Brakebill, J.W. and Gronberg, J.M. (2017) County-Level Estimates of Nitrogen and Phosphorus from Commercial
Fertilizer for the Conterminous United States, 1987-2012. U.S. Geological Survey,
https://doi.org/10.5066/F7H41PKX.
Brockwell, Peter J., and Richard A. Davis (2016) Introduction to time series and forecasting. Springer.
IPCC (2013) Climate Change 2013: The Physical Science Basis. Contribution of Working Group I to the Fifth
Assessment Report of the Intergovernmental Panel on Climate Change [Stocker, T.F., D. Qin, G.-K. Plattner, M.
Tignor, S.K. Allen, J. Boschung, A. Nauels, Y. Xia, V. Bex and P.M. Midgley (eds.)]. Cambridge University Press,
Cambridge, United Kingdom and New York, NY, USA, 1535 pp.
IPCC (2007) Climate Change 2007: The Physical Science Basis. Contribution of Working Group I to the Fourth
Assessment Report of the Intergovernmental Panel on Climate Change. [S. Solomon, D. Qin, M. Manning, Z. Chen,
M. Marquis, K.B. Averyt, M. Tignor and H.L Miller (eds.)]. Cambridge University Press. Cambridge, United Kingdom
996 pp.
IPCC (2006) 2006 IPCC Guidelines for National Greenhouse Gas Inventories. The National Greenhouse Gas
Inventories Programme, The Intergovernmental Panel on Climate Change. H.S. Eggleston, L. Buendia, K. Miwa, T.
Ngara, and K. Tanabe (eds.). Hayama, Kanagawa, Japan.
Nelson, Mark D.; Riitters, Kurt H.; Coulston, John W.; Domke, Grant M.; Greenfield, Eric J.; Langner, Linda L.;
Nowak, David J.; O'Dea, Claire B.; Oswalt, Sonja N.; Reeves, Matthew C.; Wear, David N. 2020. Defining the United
States land base: a technical document supporting the USDA Forest Service 2020 RPA assessment. Gen. Tech. Rep.
NRS-191. Madison, Wl: U.S. Department of Agriculture, Forest Service, Northern Research Station. 70 p.
https://doi.org/10.2737/NRS-GTR-191.
10-106 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
Ogle, S.M., F.J. Breidt, M. Easter, S. Williams, K. Killian, and K. Paustian (2010) Scale and uncertainty in modeled soil
organic carbon stock changes for U.S. croplands using a process-based model. Global Change Biology 16:810-822.
Sarndal C-E, Swensson B, Wretman, J (1992). Model Assisted Survey Sampling. Springer, New York.
Soil Survey Staff (2020) Gridded Soil Survey Geographic (gSSURGO) Database for the Conterminous United States.
United States Department of Agriculture, Natural Resources Conservation Service, Accessed February 2020
(FY2020 official release), Available online at https://gdg.sc.egov.usda.gov/.
USDA-NRCS (2020) Summary Report: 2017 National Resources Inventory. Natural Resources Conservation Service,
Washington, DC, and Center for Survey Statistics and Methodology, Iowa State University, Ames, Iowa.
https://www.nrcs.usda.gov/wps/portal/nrcs/main/national/technical/nra/nri/results/
Yang, L, et al. (2018). "A new generation of the United States National Land Cover Database: Requirements,
research priorities, design, and implementation strategies." ISPRS Journal of Photogrammetry and Remote Sensing
146: 108-123.
Settlements Remaining Settlements: Changes in Yard
Trimmings and Food Scrap Carbon Stocks in Landfills
Barlaz, M.A. (2008) "Re: Corrections to Previously Published Carbon Storage Factors." Memorandum to Randall
Freed, ICF International. February 28, 2008.
Barlaz, M.A. (2005) "Decomposition of Leaves in Simulated Landfill." Letter report to Randall Freed, ICF Consulting.
June 29, 2005.
Barlaz, M.A. (1998) "Carbon Storage during Biodegradation of Municipal Solid Waste Components in Laboratory-
Scale Landfills." Global Biogeochemical Cycles 12:373-380.
De la Cruz, F.B. and M.A. Barlaz (2010) "Estimation of Waste Component Specific Landfill Decay Rates Using
Laboratory-Scale Decomposition Data" Environmental Science & Technology 44:4722-4728.
Eleazer, W.E., W.S. Odle, Y. Wang, and M.A. Barlaz (1997) "Biodegradability of Municipal Solid Waste Components
in Laboratory-Scale Landfills." Environmental Science & Technology 31:911-917.
EPA (2020) Advancing Sustainable Materials Management: Facts and Figures 2018. U.S. Environmental Protection
Agency, Office of Solid Waste and Emergency Response, Washington, D.C. Available online at
https://www.epa.gov/smm/advancing-sustainable-materials-management-facts-and-figures-report.
EPA (2019) Advancing Sustainable Materials Management: Facts and Figures. U.S. Environmental Protection
Agency, Office of Solid Waste and Emergency Response, Washington, D.C. Available online at
https://www.epa.gov/smm/advancing-sustainable-materials-management-facts-and-figures-report.
EPA (2016) Municipal Solid Waste Generation, Recycling, and Disposal in the United States: Facts and Figures. U.S.
Environmental Protection Agency, Office of Solid Waste and Emergency Response, Washington, D.C. Available
online at https://archive.epa.gov/epawaste/nonhaz/municipal/web/html/msw99.html.
EPA (1995) Compilation of Air Pollutant Emission Factors. U.S. Environmental Protection Agency, Office of Air
Quality Planning and Standards, Research Triangle Park, NC. AP-42 Fifth Edition. Available online at
http://www3.epa.gov/ttnchiel/ap42/.
EPA (1991) Characterization of Municipal Solid Waste in the United States: 1990 Update. U.S. Environmental
Protection Agency, Office of Solid Waste and Emergency Response, Washington, D.C. EPA/530-SW-90-042.
IPCC (2006) 2006IPCC Guidelines for National Greenhouse Gas Inventories. The National Greenhouse Gas
Inventories Programme, The Intergovernmental Panel on Climate Change. H.S. Eggleston, L. Buendia, K. Miwa, T.
Ngara, and K. Tanabe (eds.). Hayama, Kanagawa, Japan.
References and Abbreviations 10-107
-------
IPCC (2003) Good Practice Guidance for Land Use, Land-Use Change, and Forestry. The Intergovernmental Panel on
Climate Change, National Greenhouse Gas Inventories Programme, J. Penman et al. (eds.). Available online at
http://www.ipcc-nggip.iges.or.ip/public/gpglulucf/gpglulucf.htm.
Oshins, C. and D. Block (2000) "Feedstock Composition at Composting Sites." Biocycle 41(9):31—34.
Tchobanoglous, G., H. Theisen, and S.A. Vigil (1993) Integrated Solid Waste Management, 1st edition. McGraw-Hill,
NY. Cited by Barlaz (1998) "Carbon Storage during Biodegradation of Municipal Solid Waste Components in
Laboratory-Scale Landfills." Global Biogeochemical Cycles 12:373-380.
Land Converted to Settlements
Birdsey, R. (1996) "Carbon Storage for Major Forest Types and Regions in the Conterminous United States." In R.N.
Sampson and D. Hair, (eds.). Forest and Global Change, Volume 2: Forest Management Opportunities for
Mitigating Carbon Emissions. American Forests. Washington, D.C., 1-26 and 261-379 (appendices 262 and 263).
Brockwell, Peter J., and Richard A. Davis (2016) Introduction to time series and forecasting. Springer. Domke, G.M.,
Perry, C.H., Walters, B.F., Woodall, C.W., and Smith, J.E. (2016) A framework for estimating litter carbon stocks in
forests of the United States. Science of the Total Environment 557-558: 469-478.
Domke, G.M., J.E. Smith, and C.W. Woodall. (2011) Accounting for density reduction and structural loss in standing
dead trees: Implications for forest biomass and carbon stock estimates in the United States. Carbon Balance and
Management. 6:14.
Domke, G.M., Woodall, C.W., Walters, B.F., Smith, J.E. (2013) From models to measurements: comparing down
dead wood carbon stock estimates in the U.S. forest inventory. PLoS ONE 8(3): e59949.
Domke, G.M., Perry, C.H., Walters, B.F., Woodall, C.W., and Smith, J.E. (2016) A framework for estimating litter
carbon stocks in forests of the United States. Science of the Total Environment 557-558: 469-478.
Fry, J., Xian, G., Jin, S., Dewitz, J., Homer, C., Yang, L, Barnes, C., Herold, N., and Wickham, J. (2011) Completion of
the 2006 National Land Cover Database for the Conterminous United States, PE&RS, Vol. 77(9):858-864.
Harmon, M.E., C.W. Woodall, B. Fasth, J. Sexton, M. Yatkov. (2011) Differences between standing and downed
dead tree wood density reduction factors: A comparison across decay classes and tree species. Res. Paper. NRS-15.
Newtown Square, PA: U.S. Department of Agriculture, Forest Service, Northern Research Station. 40 p.
Homer, C., Dewitz, J., Fry, J., Coan, M., Hossain, N., Larson, C., Herold, N., McKerrow, A., VanDriel, J.N., and
Wickham, J. (2007) Completion of the 2001 National Land Cover Database for the Conterminous United States.
Photogrammetric Engineering and Remote Sensing, Vol. 73, No. 4, pp 337-341.
Homer, C.G., Dewitz, J.A., Yang, L., Jin, S., Danielson, P., Xian, G., Coulston, J., Herold, N.D., Wickham, J.D., and
Megown, K. (2015) Completion of the 2011 National Land Cover Database for the conterminous United States-
Representing a decade of land cover change information. Photogrammetric Engineering and Remote Sensing, v.
81, no. 5, p. 345-354.
IPCC (2006) 2006 IPCC Guidelines for National Greenhouse Gas Inventories. The National Greenhouse Gas
Inventories Programme, The Intergovernmental Panel on Climate Change, H.S. Eggleston, L. Buendia, K. Miwa, T
Ngara, and K. Tanabe (eds.). Hayama, Kanagawa, Japan.
Jenkins, J.C., D.C. Chojnacky, L.S. Heath, and R.A. Birdsey (2003) "National-scale biomass estimators for United
States tree species." Forest Science 49(l):12-35.
Ogle, S.M., M.D. Eve, F.J. Breidt, and K. Paustian (2003) "Uncertainty in estimating land use and management
impacts on soil organic carbon storage for U.S. agroecosystems between 1982 and 1997." Global Change Biology
9:1521-1542.
10-108 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
Ogle, S.M., F.J. Breidt, and K. Paustian (2006) "Bias and variance in model results due to spatial scaling of
measurements for parameterization in regional assessments." Global Change Biology 12:516-523.
Schimel, D.S. (1995) "Terrestrial ecosystems and the carbon cycle." Global Change Biology 1: 77-91.
Smith, J.E.; Heath, L.S.; Skog, K.E.; Birdsey, R.A. (2006) Methods for calculating forest ecosystem and harvested
carbon with standard estimates for forest types of the United States. Gen. Tech. Rep. NE-343. Newtown Square,
PA: U.S. Department of Agriculture, Forest Service, Northeastern Research Station. 216 p.
Tubiello, F. N., et al. (2015) "The Contribution of Agriculture, Forestry and other Land Use activities to Global
Warming, 1990-2012." Global Change Biology 21:2655-2660.
USDA Forest Service. (2022) Forest Inventory and Analysis National Program: FIA Data Mart. U.S. Department of
Agriculture Forest Service. Washington, D.C. Available online at:
https://apps.fs.usda.gov/fia/datamart/datamart.html. Accessed on 07 October 2022.
USDA-NRCS (2020) Summary Report: 2017 National Resources Inventory. Natural Resources Conservation Service,
Washington, DC, and Center for Survey Statistics and Methodology, Iowa State University, Ames, Iowa.
https://www.nrcs.usda.gov/wps/portal/nrcs/main/national/technical/nra/nri/results/.
USDA-NRCS (1997) "National Soil Survey Laboratory Characterization Data," Digital Data, Natural Resources
Conservation Service, U.S. Department of Agriculture. Lincoln, NE.
Westfall, J.A., Coulston, J.W., Gray, A.N., Shaw, J.D., Radtke, P.J., Walker, D.M., Weiskittel, A.R., MacFarlane, D.W.,
Affleck, D.L.R., Zhao, D., Temesgen, H., Poudel, K.P., Frank, J.M., Prisley, S.P., Wang, Y., Sanchez Meador, A.J., Auty,
D., and Domke, G.M. 2023. A national-scale tree volume, biomass, and carbon modeling system for the United
States. Gen. Tech. Rep. WO-XX.
Woodall, C.W., LS. Heath, G.M. Domke, and M.C. Nichols. (2011) Methods and equations for estimating
aboveground volume, biomass, and carbon for trees in the U.S. forest inventory, 2010. Gen. Tech. Rep. NRS-88.
Newtown Square, PA: U.S. Department of Agriculture, Forest Service, Northern Research Station. 30 p.
Woodall, C.W., and V.J. Monleon (2008) Sampling protocol, estimation, and analysis procedures for the down
woody materials indicator of the FIA program. Gen. Tech. Rep. NRS-22. Newtown Square, PA: U.S. Department of
Agriculture, Forest Service, Northern Research Station. 68 p.
Yang, L, Jin, S., Danielson, P., Homer, C., Gass, L., Bender, S. M., Case, A., Costello, C., Dewitz, J., Fry, J., Funk, M.,
Granneman, B., Liknes, G. C., Rigge, M. & Xian, G. (2018) A new generation of the United States National Land
Cover Database: Requirements, research priorities, design, and implementation strategies. ISPRS Journal of
Photogrammetry and Remote Sensing 146:108-123.
Waste
Landfills
40 CFR Part 60, Subpart WWW (2005) Standards of Performance for Municipal Solid Waste Landfills, 60.750-
60.759, Code of Federal Regulations, Title 40. Available online at: https://www.ecfr.gov/current/title-40/chapter-
l/subchapter-C/part-60/subpart-WWW.
40 CFR Part 258, Subtitle D of RCRA (2012) Criteria for Municipal Solid Waste Landfills, 258.1—258.75, Code of
Federal Regulations, Title 40. Available online at: https://www.ecfr.gov/cgi-bin/text-idx?node=pt40.25.258..
BioCycle (2010) "The State of Garbage in America" By L. Arsova, R. Van Haaren, N. Goldstein, S. Kaufman, and N.
Themelis. BioCycle. December 2010. Available online at: https://www.biocycle.net/2010/10/26/the-state-of-
garbage-in america-4/.
References and Abbreviations 10-109
-------
Bronstein, K., Coburn, J., and R. Schmeltz (2012) "Understanding the EPA's Inventory of U.S. Greenhouse Gas
Emissions and Sinks and Mandatory GHG Reporting Program for Landfills: Methodologies, Uncertainties,
Improvements and Deferrals." Prepared for the U.S. EPA International Emissions Inventory Conference, August
2012, Tampa, Florida. Available online at:
https://www3.epa.gov/ttnchiel/conference/ei20/session3/kbronstein.pdf.
Czepiel, P., B. Mosher, P. Crill, and R. Harriss (1996) "Quantifying the Effect of Oxidation on Landfill Methane
Emissions." Journal of Geophysical Research, 101(D11):16721-16730. Dou, Z.; Ferguson, J. D.; Galligan, D. T.; Kelly,
A. M.; Finn, S. T.; Giegengack, R. (2016) "Assessing U.S. food wastage and opportunities for reduction." Global Food
Security Volume 8, March 2016, Pages 19-26. https://doi.Org/10.1016/i.gfs.2016.02.001.
EIA (2007) Voluntary Greenhouse Gas Reports for EIA Form 1605B (Reporting Year 2006). Available online at:
https://www. eia.gov/environ ment/pdfpages/0608sl2009)index.php.
EPA (2023a) Greenhouse Gas Reporting Program (GHGRP). 2022 Amazon S3 Data. Subpart HH: Municipal Solid
Waste Landfills and Subpart TT: Industrial Waste Landfills. Accessed on August 8, 2023.
EPA (2023b) Landfill Methane Outreach Program (LMOP). 2023 Landfill and Project Level Data. July 2023. Available
online at: https://www.epa.gov/lmop/landfill-gas-energy-proiect-data.
EPA (2020a) Wasted Food Measurement Methodology Scoping Memo. July 2020. Available online at
https://www.epa.gov/sites/production/files/202Q-
06/documents/food measurement methodology scoping memo-6-18-20.pdf.
EPA (2020b) Advancing Sustainable Materials Management: Facts and Figures 2018. December 2020. Available
online at: https://www.epa.gov/sites/production/files/202Q-ll/documents/2018 tables and figures fnl 508.pdf.
EPA (2020c) Advancing Sustainable Materials Management: Facts and Figures 2016 and 2017. November 2019.
Available online at: https://www.epa.gov/sites/default/files/2021-
01/documents/2018 tables and figures dec 2020 fnl 508.pdf.
EPA (2018) Advancing Sustainable Materials Management: Facts and Figures 2015. July 2018. Available online at:
https://www.epa.gov/sites/production/files/2018-
07/documents/smm 2015 tables ; ~es 07252018 fnl 508 O.pdf.
EPA (2016a) Industrial and Construction and Demolition Landfills. Available online at:
https://www.epa.gov/landfills/industrial-and-construction-and-demolition-cd-landfills.
EPA (2016b) Advancing Sustainable Materials Management: Facts and Figures 2014. December 2016. Available
online at: https://www.epa.gov/sites/production/files/2016-ll/documents/2014 smm tablesfigures 508.pdf.
EPA (2014) Advancing Sustainable Materials Management: Facts and Figures 2014. February 2014. Available online
at: https://www.epa.gov/sites/production/files/2015-09/documents/2012 msw dat tbls.pdf.
EPA (2008) Compilation of Air Pollution Emission Factors, Publication AP-42, Draft Section 2.4 Municipal Solid
Waste Landfills. October 2008.
EPA (1993) Anthropogenic Methane Emissions in the United States, Estimates for 1990: Report to Congress, U.S.
Environmental Protection Agency, Office of Air and Radiation. Washington, D.C. EPA/430-R-93-003. April 1993.
EPA (1988) National Survey of Solid Waste (Municipal) Landfill Facilities, U.S. Environmental Protection Agency.
Washington, D.C. EPA/530-SW-88-011. September 1988.
EREF (The Environmental Research & Education Foundation) (2016) Municipal Solid Waste Management in the
United States: 2010 & 2013.
ERG (2023) Production Data Supplied by ERG for 1990-2022 for Pulp and Paper, Fruits and Vegetables, and Meat.
September 7, 2021.
10-110 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
Food Waste Reduction Alliance (FWRA) (2016) Analysis of U.S. Food Waste Among Food Manufacturers, Retailers,
and Restaurants. A joint project by the Food Marketing Institute, the Grocery Manufacturers Association, and the
National Restaurant Association. Available online at: https://foodwastealliance.org/wp-
content/uploads/2020/05/FWRA-Food-Waste-Survev-2016-Report Final.pdf.
IPCC (2019) 2019 Refinement to the 2006IPCC Guidelines for National Greenhouse Gas Inventories. Calvo Buendia,
E., Tanabe, K., Kranjc, A., Baasansuren, J., Fukuda, M., Ngarize, S., Osako, A., Pyrozhenko, Y., Shermanau, P. and
Federici, S. (eds). Published: IPCC, Switzerland.
IPCC (2006) 2006 IPCC Guidelines for National Greenhouse Gas Inventories. The National Greenhouse Gas
Inventories Programme, The Intergovernmental Panel on Climate Change. [H.S. Eggleston, L. Buendia, K. Miwa, T.
Ngara, and K. Tanabe (eds.)]. Hayama, Kanagawa, Japan.
Mancinelli, R. and C. McKay (1985) "Methane-Oxidizing Bacteria in Sanitary Landfills." Proc. First Symposium on
Biotechnical Advances in Processing Municipal Wastes for Fuels and Chemicals, Minneapolis, MN, 437-450. August.
RTI (2018) Methodological changes to the scale-up factor used to estimate emissions from municipal solid waste
landfills in the Inventory. Memorandum prepared by K. Bronstein and M. McGrath for R. Schmeltz (EPA). March 22,
2018.
RTI (2017) Methodological changes to the methane emissions from municipal solid waste landfills as reflected in
the public review draft of the 1990-2015 Inventory. Memorandum prepared by K. Bronstein and M. McGrath for R.
Schmeltz (EPA). March 31, 2017.
RTI (2011) Updated Research on Methane Oxidation in Landfills. Memorandum prepared by K. Weitz (RTI) for R.
Schmeltz (EPA). January 14, 2011.
Waste Business Journal (WBJ) (2021) Directory of Waste Processing & Disposal Sites 2021.
WBJ (2016) Directory of Waste Processing & Disposal Sites 2016.
WBJ (2010) Directory of Waste Processing & Disposal Sites 2010.
Wastewater Treatment and Discharge
AF&PA (2022) "AF&PA Members Achieve Progress on Water Stewardship Goal for 2020." American Forest & Paper
Association. Available online at: https://www.afandpa.org/statistics-resources/afpa-members-achieve-progress-
water-stewardship-goal-2020. Accessed July 2022.
AF&PA (2020) "2020 AF&PA Sustainability Report: Advancing the sustainability of an essential industry." American
Forest & Paper Association. Available online at: https://www.afandpa.org/sites/default/files/2021-07/2020 AF-
PA-Sustainability-Report.pdf. Accessed June 2021.
AF&PA (2018) "2018 AF&PA Sustainability Report: Advancing U.S. Paper and Wood Products Industry Sustainability
Performance." American Forest & Paper Association. Available online at: http://sustainability.afandpa.org/wp-
content/uploads/2018/06/2018SustainabilityReport PAGES.pdf. Accessed July 2019.
AF&PA (2016) "2016 AF&PA Sustainability Report: Advancing U.S. Paper and Wood Products Industry Sustainability
Performance." American Forest & Paper Association.
AF&PA (2014) "2014 AF&PA Sustainability Report." American Forest & Paper Association.
Beecher et al. (2007) "A National Biosolids Regulation, Quality, End Use & Disposal Survey, Preliminary Report."
Northeast Biosolids and Residuals Association, April 14, 2007. Available online at:
https://staticl.squarespace.eom/static/54806478e4b0dc44el698e88/t/5480c7a2e4b0787f2c73ad81/1417725858
575/NtlBiosldsRpt-AppD-FINAL.pdf. Accessed August 2021.
Beer Institute (2011) Brewers Almanac. Available online at: http://www.beerinstitute.org/multimedia/brewers-
almanac.
References and Abbreviations 10-111
-------
Benyahia, F., M. Abdulkarim, A. Embaby, and M. Rao. (2006) Refinery Wastewater Treatment: A true Technological
Challenge. Presented at the Seventh Annual U.A.E. University Research Conference.
BIER (2021) 2021 Benchmarking Study Trends & Observations. Available online at
https://www.bieroundtable.com/publication/2021-water-and-energy-use-benchmarking-studv/. Accessed
September 2023.
Brewers Association (2021) Statistics: Number of Breweries. Available online at:
https://www.brewersassociation.org/statistics-and-data/national-beer-stats/. Accessed August 2021.
Brewers Association (2016a) 2015 Sustainability Benchmarking Report. Available online at:
https://www.brewersassociation.org/best-practices/sustainabilitv/sustainability-benchmarking-tools. Accessed
March 2018.
Brewers Association (2016b) Wastewater Management Guidance Manual. Available online at:
https://www.brewersassociation.org/educational-publications/wastewater-management-guidance-manual.
Accessed September 2017.
Cabrera (2017) "Pulp Mill Wastewater: Characteristics and Treatment." Biological Wastewater Treatment and
Resource Recovery. InTech. pp. 119-139.
CAST (1995) Council for Agricultural Science and Technology. Waste Management and Utilization in Food
Production and Processing. U.S.A. October 1995. ISBN 1-887383-02-6. Available online at: https://www.cast-
science.org/publication/waste-management arid-utilization in-food-production arid-processing/.
Climate Action Reserve (CAR) (2011) Landfill Project Protocol V4.0, June 2011. Available online at:
http://www.climateactionreserve.org/how/protocols/us-landfill/.
Cooper (2018) Email correspondence. Geoff Cooper, Renewable Fuels Association to Kara Edquist, ERG. "Wet Mill
vs. Dry Mill Ethanol Production." May 18, 2018.
DOE (2013) U.S. Department of Energy Bioenergy Technologies Office. Biofuels Basics. Available online at:
http://energy.gov/eere/bioenergv/biofuels-basics. Accessed September 2013.
Donovan (1996) Siting an Ethanol Plant in the Northeast. C.T. Donovan Associates, Inc. Report presented to
Northeast Regional Biomass Program (NRBP). (April). Available online at:
https://citeseerx.ist.psu.edu/viewdoc/download?doi=10.1.1.614.856&rep=repl&type=pdf. Accessed October
2006.
EIA (2023) Energy Information Administration. U.S. Refinery and Blender Net Production of Crude Oil and
Petroleum Products (Thousand Barrels). Available online at:
https://www.eia.gov/dnav/pet/pet pnp refp dc nus mbbl ro.htro. Accessed August 2023.
EPA (2019) Preliminary Effluent Guidelines Program Plan 14. EPA-821-R-19-005. Office of Water, U.S.
Environmental Protection Agency. Washington, DC. October 2019. Available online at:
https://www.epa.gov/sites/production/files/2019-10/documents/prelim-eg-pl * ' ; > ^ 319.pdf. Accessed July
2020.
EPA (2013) U.S. Environmental Protection Agency. Report on the Performance of Secondary Treatment
Technology. EPA-821-R-13-001. Office of Water, U.S. Environmental Protection Agency. Washington, D.C. March
2013. Available online at: https://www.epa.gov/sites/production/files/2015-
11/documents/npdes secondary treatment report march2013.pdf.
EPA (2012) U.S. Environmental Protection Agency. Clean Watersheds Needs Survey 2012 - Report to Congress. U.S.
Environmental Protection Agency, Office of Wastewater Management. Washington, D.C. Available online at:
https://www.epa.gOv/cwns/clean-watersheds-needs-survev-cwns-2012-report-and-data#access. Accessed
February 2016.
10-112 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
EPA (2010) U.S. Environmental Protection Agency. Nutrient Control Design Manual. U.S. Environmental Protection
Agency, Office of Research and Development. Washington, D.C. EPA600-R-10-100. Available online at:
https://www.epa.gov/sites/default/files/2019-08/documents/nutrient control design roanual.pdf. Accessed
December 2023.
EPA (2008) U.S. Environmental Protection Agency. Clean Watersheds Needs Survey 2008 - Report to Congress. U.S.
Environmental Protection Agency, Office of Wastewater Management. Washington, D.C. Available online at:
https://www.epa.gov/cwns/clean-watersheds-needs-survev-cwns-2008-report-and-data. Accessed December
2015.
EPA (2004) U.S. Environmental Protection Agency. Clean Watersheds Needs Survey 2004 - Report to Congress. U.S.
Environmental Protection Agency, Office of Wastewater Management. Washington, D.C. Available online at:
https://www.epa.gov/cwns/clean-watersheds-needs-syrvey-cwns-report-congress-2004.
EPA (2000) U.S. Environmental Protection Agency. Clean Watersheds Needs Survey 2000 - Report to Congress.
Office of Wastewater Management, U.S. Environmental Protection Agency. Washington, D.C. Available online at:
https://www.epa.gov/cwns/clean-watersheds-needs-survev-cwns-2000-report-and-data. Accessed July 2007.
EPA (1999) U.S. Environmental Protection Agency. Biosolids Generation, Use and Disposal in the United States.
Office of Solid Waste and Emergency Response, U.S. Environmental Protection Agency. Washington, D.C. EPA530-
R-99-009. September 1999.
EPA (1998) U.S. Environmental Protection Agency. "AP-42 Compilation of Air Pollutant Emission Factors." Chapter
2.4, Table 2.4-3, page 2.4-13. Available online at: https://www.epa.gov/sites/default/files/2020-
10/documents/c02s04.pdf.
EPA (1997a) U.S. Environmental Protection Agency. Estimates of Global Greenhouse Gas Emissions from Industrial
and Domestic Wastewater Treatment. EPA-600/R-97-091. Office of Policy, Planning, and Evaluation, U.S.
Environmental Protection Agency. Washington, D.C. September 1997.
EPA (1997b) U.S. Environmental Protection Agency. Supplemental Technical Development Document for Effluent
Guidelines and Standards (Subparts B & E). EPA-821-R-97-011. Office of Water, U.S. Environmental Protection
Agency. Washington, D.C. October 1997.
EPA (1996) U.S. Environmental Protection Agency. 1996 Clean Water Needs Survey Report to Congress.
Assessment of Needs for Publicly Owned Wastewater Treatment Facilities, Correction of Combined Sewer
Overflows, and Management of Storm Water and Nonpoint Source Pollution in the United States. Office of
Wastewater Management, U.S. Environmental Protection Agency. Washington, D.C.
EPA (1993a) U.S. Environmental Protection Agency, "Anthropogenic Methane Emissions in the U.S.: Estimates for
1990, Report to Congress." Office of Air and Radiation, Washington, DC. April 1993.
EPA (1993b) U.S. Environmental Protection Agency. Development Document for the Proposed Effluent Limitations
Guidelines and Standards for the Pulp, Paper and Paperboard Point Source Category. EPA-821-R-93-019. Office of
Water, U.S. Environmental Protection Agency. Washington, D.C. October 1993.
EPA (1993c) Standards for the Use and Disposal of Sewage Sludge. 40 CFR Part 503.
EPA (1992) U.S. Environmental Protection Agency. Clean Watersheds Needs Survey 1992 - Report to Congress.
Office of Wastewater Management, U.S. Environmental Protection Agency. Washington, D.C.
EPA (1982) U.S. Environmental Protection Agency. Development Document for Effluent Limitations Guidelines and
standards for the Petroleum Refining. United States Environmental Protection Agency, Office of Water. EPA-440/1-
82-014. Washington D.C. October 1982.
EPA (1975) U.S. Environmental Protection Agency. Development Document for Interim Final and Proposed Effluent
Limitations Guidelines and New Source Performance Standards for the Fruits, Vegetables, and Specialties Segment
References and Abbreviations 10-113
-------
of the Canned and Preserved Fruits and Vegetables Point Source Category. United States Environmental Protection
Agency, Office of Water. EPA-440/1-75-046. Washington D.C. October 1975.
EPA (1974) U.S. Environmental Protection Agency. Development Document for Effluent Limitations Guidelines and
New Source Performance Standards for the Apple, Citrus, and Potato Processing Segment of the Canned and
Preserved Fruits and Vegetables Point Source Category. Office of Water, U.S. Environmental Protection Agency,
Washington, D.C. EPA-440/l-74-027-a. March 1974.
ERG (2023) Memorandum: Improvements to the 1990-2021 Wastewater Treatment and Discharge Greenhouse
Gas Inventory. February 2023.
ERG (2021a) Revised Memorandum: Improvements to the 1990-2019 Wastewater Treatment and Discharge
Greenhouse Gas Inventory. March 2021.
ERG (2021b) Draft Memorandum: Improvements to the 1990-2020 Wastewater Treatment and Discharge
Greenhouse Gas Inventory. July 2021.
ERG (2021c) Draft Memorandum: Expert Judgement Documentation for the Wastewater Treatment and Discharge
Greenhouse Gas Inventory Uncertainties, August 2021.
ERG (2019) Memorandum: Recommended Improvements to the 1990-2018 Wastewater Greenhouse Gas
Inventory. August 2019.
ERG (2018a) Memorandum: Updates to Domestic Wastewater BOD Generation per Capita. August 2018.
ERG (2018b) Memorandum: Inclusion of Wastewater Treatment Emissions from Breweries. July 2018.
ERG (2016) Revised Memorandum: Recommended Improvements to the 1990-2015 Wastewater Greenhouse Gas
Inventory. November 2016.
ERG (2013a) Memorandum: Revisions to Pulp and Paper Wastewater Inventory. October 2013.
ERG (2013b) Memorandum: Revisions to the Petroleum Refinery Wastewater Inventory. October 2013.
ERG (2008a) Memorandum: Planned Revisions of the Industrial Wastewater Inventory Emission Estimates for the
1990-2007 Inventory. 10 August 2008.
ERG (2008b) Memorandum: Estimation of Onsite Treatment at Industrial Facilities and Review of Wastewater
Characterization Data. 15 April 2008.
ERG (2006a) Memorandum: Recommended Improvements to EPA's Wastewater Inventory for Industrial
Wastewater. Prepared for Melissa Weitz, EPA. 11 August 2006.
ERG (2006b) Memorandum: Assessment of Greenhouse Gas Emissions from Wastewater Treatment of U.S. Ethanol
Production Wastewaters. Prepared for Melissa Weitz, EPA. 10 October 2006.
FAO (2023a) FAOSTAT-Forestry Database. Available online at: http://www.fao.Org/faostat/en/#data/FQ. Accessed
September 2023.
FAO (2023b) "Pulp and Paper Capacities Report." United States. From 1998 - 2003, 2000 - 2005, 2001 - 2006,
2002 - 2007, 2003 - 2008, 2010 - 2015, 2011 - 2016, 2012 - 2017, 2013 - 2018, 2014 - 2019, 2015 - 2020, 2016 -
2021, 2017 - 2022, 2018 - 2023, 2019 - 2024, 2020-2025, 2021-2026 reports. Available online at:
http://www.fao.org/forestry/statistics/30571/en/. Accessed September 2023.
FAO (2022) FAOSTAT-Food Balance Sheets. Available online at: http://www.fao.Org/faostat/en/#data/FBS.
Accessed September 2023.
Foley et al. (2015) N2O and CHa Emission from Wastewater Collection and Treatment Systems: State of the Science
Report and Technical Report. GWRC Report Series. IWA Publishing, London, UK.
Great Lakes-Upper Mississippi River Board of State and Provincial Public Health and Environmental Managers.
(2004) Recommended Standards for Wastewater Facilities (Ten-State Standards).
10-114 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
Guisasola et al. (2008) Methane formation in sewer systems. Water Research 42(6-7): 1421-1430.
Instituto de Estadisticas de Puerto Rico (2021) Population of Puerto Rico from 1990-1999 from "Estimados anuales
poblacionales de los municipios desde 1950." Accessed February 2021. Available online at:
https://censo.estadisticas.pr/EstimadosPoblacionales.
IPCC (2022) Emission factor database: Emission Factor Detail (ID:625621). The Intergovernmental Panel on Climate
Change. Available online at: https://www.ipcc-nggip.iges.or.ip/EFDB/ef detail.php.
IPCC (2019) 2019 Refinement to the 2006 IPCC Guidelines for National Greenhouse Gas Inventories. The National
Greenhouse Gas Inventories Programme, The Intergovernmental Panel on Climate Change. [CalvoBuendia, E.,
Tanabe, K., Kranjc, A., Baasansuren, J., Fukuda, M., Ngarize S., Osako, A., Pyrozhenko, Y., Shermanau, P. and
Federici, S. (eds)]. Switzerland.
IPCC (2014) 2013 Supplement to the 2006 IPCC Guidelines for National Greenhouse Gas Inventories: Wetlands.
[Hiraishi, T., Krug, T., Tanabe, K., Srivastava, N., Baasansuren, J., Fukuda, M. and Troxler, T.G. (eds.)]. Published:
IPCC, Switzerland.
IPCC (2013) Climate Change 2013: The Physical Science Basis. Contribution of Working Group I to the Fifth
Assessment Report of the Intergovernmental Panel on Climate Change [Stocker, T.F., D. Qin, G.-K. Plattner, M.
Tignor, S.K. Allen, J. Boschung, A. Nauels, Y. Xia, V. Bex and P.M. Midgley (eds.)]. Cambridge University Press,
Cambridge, United Kingdom and New York, NY, USA, 1535 pp.
IPCC (2007) Climate Change 2007: The Physical Science Basis. Contribution of Working Group I to the Fourth
Assessment Report of the Intergovernmental Panel on Climate Change. [S. Solomon, D. Qin, M. Manning, Z. Chen,
M. Marquis, K.B. Averyt, M. Tignor and H.L Miller (eds.)]. Cambridge University Press. Cambridge, United Kingdom
996 pp.
IPCC (2006) 2006 IPCC Guidelines for National Greenhouse Gas Inventories. The National Greenhouse Gas
Inventories Programme, The Intergovernmental Panel on Climate Change. [H.S. Eggleston, L. Buendia, K. Miwa, T.
Ngara, and K. Tanabe (eds.)]. Hayama, Kanagawa, Japan.
Kenari et. al (2010) An Investigation on the Nitrogen Content of a Petroleum Refinery Wastewater and its Removal
by Biological Treatment. Journal of Environmental Health, Sciences, and Engineering. 7(1): 391-394.
Leverenz, H.L., G. Tchobanoglous, and J.L. Darby (2010) "Evaluation of Greenhouse Gas Emissions from Septic
Systems." Water Environment Research Foundation. Alexandria, VA.
Lewis, A. (2019) Email correspondence. Ann Lewis, RFA to Kara Edquist, ERG. "Wet Mill vs Dry Mill Ethanol
Production." August 20, 2019.
Malmberg, B. (2018) Draft Pulp and Paper Information for Revision of EPA Inventory of U.S. Greenhouse Gas
Emissions and Sinks, Waste Chapter. National Council for Air and Stream Improvement, Inc. Prepared for Rachel
Schmeltz, EPA. June 13, 2018.
Merrick (1998) Wastewater Treatment Options for the Biomass-to-Ethanol Process. Report presented to National
Renewable Energy Laboratory (NREL). Merrick & Company. Subcontract No. AXE-8-18020-01. October 22,1998.
Metcalf & Eddy, Inc. (2014) Wastewater Engineering: Treatment and Resource Recovery, 5th ed. McGraw Hill
Publishing.
Metcalf & Eddy, Inc. (2003) Wastewater Engineering: Treatment, Disposal and Reuse, 4th ed. McGraw Hill
Publishing.
NEBRA (2022) "U.S. National Biosolids Data." Northeast Biosolids and Residuals Associations. Available online at:
https://staticl.squarespace.eom/static/601837dlc67bcc4elbll862f/t/62f4f5fbae32804dd9f51ef6/166022092535
6/National BiosolidsDataSummary NBDP 20220811.pdf
Nemerow, N.L and A. Dasgupta (1991) Industrial and Hazardous Waste Treatment. Van Nostrand Reinhold. NY.
ISBN 0-442-31934-7.
References and Abbreviations 10-115
-------
NRBP (2001) Northeast Regional Biomass Program. An Ethanol Production Guidebook for Northeast States.
Washington, D.C. (May 3).
Rendleman, C.M. and Shapouri, H. (2007) New Technologies in Ethanol Production. USDA Agricultural Economic
Report Number 842.
RFA (2023a) Renewable Fuels Association. Annual U.S. Fuel Ethanol Production. Available online at:
https://ethanolrfa.org/statistics/annual-ethanol-production. Accessed August 2023.
RFA (2023b) Renewable Fuels Association. Monthly Grain Use for U.S. Ethanol Production Report. Available online
at: https://ethanolrfa.org/statistics/feedstock-use-co-product-outpyt. Accessed August 2023.
Ruocco (2006a) Email correspondence. Dr. Joe Ruocco, Phoenix Bio-Systems to Sarah Holman, ERG. "Capacity of
Bio-Methanators (Dry Milling)." October 6, 2006.
Ruocco (2006b) Email correspondence. Dr. Joe Ruocco, Phoenix Bio-Systems to Sarah Holman, ERG. "Capacity of
Bio-Methanators (Wet Milling)." October 16, 2006.
Short et al. (2017) Dissolved Methane in the Influent of Three Australian Wastewater Treatment Plants Fed by
Gravity Sewers. Sci Total Environ 599-600: 85-93.
Short et al. (2014) Municipal Gravity Sewers: an Unrecognised Source of Nitrous Oxide. Sci Total Environ 468-469:
211-218.
Stier, J. (2018) Personal communications between John Stier, Brewers Association Sustainability Mentor and Amie
Aguiar, ERG. Multiple dates.
Sullivan (SCS Engineers) (2010) The Importance of Landfill Gas Capture and Utilization in the U.S. Presented to
SWICS, April 6, 2010. Available online at: https://www.scsengineers.com/scs-white-papers/the-importance-of-
landfill-gas-capture-and-utilization-in-the-u-s/.
Sullivan (SCS Engineers) (2007) Current MSW Industry Position and State of the Practice on Methane Destruction
Efficiency in Flares, Turbines, and Engines. Presented to Solid Waste Industry for Climate Solutions (SWICS). July
2007. Available online at: https://www.scsengineers.com/wp-
content/uploads/2015/03/Sullivan LFG Destruction Efficiency White Paper.pdf.
TTB (2022) Alcohol and Tobacco Tax and Trade Bureau. Beer Statistics. Available online at:
https://www.ttb.gov/beer/beer-stats.shtml. Accessed July 2021.
UNFCCC (2012) CDM Methodological tool, Project emissions from flaring (Version 02.0.0). EB 68 Report. Annex 15.
Available online at: http://cdm.unfccc.int/methodologies/PAmethodologies/tools/am-tool-06-vl.pdf/history view.
U.S. Census Bureau (2023) International Database. Available online at: https://www.census.gov/data-
tools/demo/idb/#/trends?YR ANIM=2020&dashPages=DASH&FIPS SINGLE=US&COUNTRY YEAR=2022&menu=tre
ndsViz&TREND RANGE=1990,2022&TREND STEP=5&TREND ADD YRS=&FIPS=AQ,GQ,CQ,RQ,VQ&measures=POP
&CCODE=AS,GU,MP,PR,US,VI&CCODE SINGLE=US&COUNTRY YR ANIM=2022. Accessed September 2023.
U.S. Census Bureau (2022) Annual Estimates of the Resident Population for the United States, Regions, States,
District of Columbia, and Puerto Rico: April 1, 2020 to July 1, 2021. Available online at:
https://www.census.gov/data/tables/time-series/demo/popest/2020s-national-total.html. Accessed July 2023.
U.S. Census Bureau (2021a) "American Housing Survey." Table 1A-4: Selected Equipment and Plumbing-All
Housing Units. From 1989,1991,1993,1995,1997,1999, 2001, 2003, 2005, 2007, and 2009 reports. Table C-04-
AO Plumbing, Water, and Sewage Disposal-All Occupied Units. From 2011, 2013, 2015, 2017, 2019 and 2021
reports. Available online at http://www.census.gov/programs-surveys/ahs/data.html. Accessed August 2023.
U.S. Census Bureau (2021b) Annual Estimates of the Resident Population for the United States, Regions, States,
and Puerto Rico: April 1, 2010 to July 1,2020. Available online at: https://www.census.gov/data/tables/time-
series/demo/popest/2010s-state-total.html.
10-116 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
U.S. Census Bureau (2013) "American Housing Survey." Table 1A-4: Selected Equipment and Plumbing-All Housing
Units. From 1989,1991, 1993, 1995, 1997, 1999, 2001, 2003, 2005, 2007, and 2009 reports. Table C-04-AO
Plumbing, Water, and Sewage Disposal-All Occupied Units. From 2011, and 2013 reports. Available online at
http://www.census.gov/programs-surveys/ahs/data.html. Accessed May 2020.
U.S. Census Bureau, Population Division (2011) Table 1. Intercensal Estimates of the Resident Population for the
United States, Regions, States, and Puerto Rico: April 1, 2000 to July 1, 2010 (ST-EST00INT-01), Release Date:
September 2011. Available online at: https://www2.census.gov/programs-surveys/popest/datasets/2000-
2010/intercensal/state/st-est00int- alldata.csv.
U.S. Census Bureau, Population Division (2002) Table CO-EST2001-12-00 - Time Series of Intercensal State
Population Estimates: April 1,1990 to April 1, 2000.Available online at: https://www2.census.gov/programs-
surveys/popest/tables/1990-2000/intercensal/st- co/co-est2001-12-00.pdf.
USDA (U.S. Department of Agriculture) (2023a) Livestock Slaughter 2022 Summary. Available online at:
https://downloads.usda.library.cornell.edu/usda-esmis/files/r207tp32d/8p58qs65g/g445dv089/lsan0423.pdf
USDA (U.S. Department of Agriculture) (2023b) Poultry Slaughter 2022 Summary. Available online at:
https://downloads.usda.library.cornell.edu/usda-esmis/files/pgl5bd88s/m613p944x/ht24xx05i/pslaan23.pdf
USDA (U.S. Department of Agriculture) (2023c) Vegetables 2022 Summary. Available online at:
https://downloads.usda.library.cornell.edu/usda-esmis/files/02870v86p/hq37xl21v/4b29ck28c/vegean23.pdf
USDA (U.S. Department of Agriculture) (2023d) Noncitrus Fruits and Nuts 2022 Summary. Available online at:
https://downloads.usda.library.cornell.edu/usda-esmis/files/zs25x846c/zk51wx21m/k356bk214/ncit0523.pdf
USDA (U.S. Department of Agriculture) (2022a) Potato Annual 2021 Summary. Available online at:
https://downloads.usda.library.cornell.edu/usda-esmis/files/fx719m44h/gbl9gf71k/37721m72q/pots0922.pdf
USDA (U.S. Department of Agriculture) (2022b) Citrus Fruits 2022 Summary. Available online at:
https://downloads.usda.library.cornell.edu/usda-esmis/files/i9602060k/pn89ff24k/zp38xm24q/cfrt0922.pdf
USDA (2015) U.S. Department of Agriculture. Economic Research Service. Nutrient Availability (food energy,
nutrients, and dietary components). Washington D.C. Available online at: https://www.ers.usda.gov/data-
products/food-availability-per-capita-data-system/food-availability-per-capita-data-system. Accessed September
2023.
U.S. Poultry (2006) Email correspondence. John Starkey, USPOULTRY to D. Bartram, ERG. 30 August 2006.
White and Johnson (2003) White, P.J. and Johnson, L.A. Editors. Corn: Chemistry and Technology. 2nd ed. AACC
Monograph Series. American Association of Cereal Chemists. St. Paul, MN.
World Bank (1999) Pollution Prevention and Abatement Handbook 1998, Toward Cleaner Production. The
International Bank for Reconstruction and Development/The WORLDBANK. 1818 H Street, N.W. Washington, DC.
20433, USA. ISBN 0-8213-3638-X.
Composting
BioCycle (2023) BioCycle Nationwide Survey: Full-Scale Food Waste Composting Infrastructure in the U.S. Prepared
by N. Goldstein, P. Luu, and S. Motta. Available online at: https://www.biocycle.net/us-food-waste-composting-
infrastructure/.
BioCycle (2018a) Organic Waste Bans and Recycling Laws to Tackle Food Waste. Prepared by E. Broad Lieb, K.
Sandson, L. Macaluso, and C. Mansell. Available online at: https://www.biocycle.net/2018/09/ll/organic-waste-
bans-recycling-laws-tackle-food-waste/.
References and Abbreviations 10-117
-------
BioCycle (2018b) State Food Waste Recycling Data Collection, Reporting Analysis. Prepared by Nora Goldstein.
Available online at: http://compostcolab.wpengine.com/wp-content/uploads/2018/ll/State-Food-Waste-
Recycling-Data-Collection-Reporting-Analysis.pdf.
BioCycle (2017) The State of Organics Recycling in the U.S. Prepared by Nora Goldstein. Available online at
http://www.biocycle.net/17 10 06 1/0001/BioCycle StateOfOrganicsUS.pdf.
BioCycle (2010) The State of Garbage in America. Prepared by Rob van Haaren, Nickolas Themelis and Nora
Goldstein. Available online at http://www.biocvcle.net/images/art/1010/bcl01016 s.pdf.
Cornell Composting (1996) Monitoring Compost Moisture. Cornell Waste Management Institute. Available online
at: http://compost.css.cornell.edu/monitor/monitormoisture.html.
Cornell Waste Management Institute (2007) The Science of Composting. Available online at
http://cwmi.css.cornell.edu/chapterl.pdf.
EPA (2020) Advancing Sustainable Materials Management: 2018 Tables and Figures. Office of Solid Waste and
Emergency Response, U.S. Environmental Protection Agency, Washington, D.C. Available online at:
https://www.epa.gov/sites/default/files/2021-01/documents/2018 tables and figures dec 2020 fnl 508.pdf.
EPA (2018) Advancing Sustainable Materials Management: 2015 Tables and Figures. Office of Solid Waste and
Emergency Response, U.S. Environmental Protection Agency, Washington, D.C. Available online at
https://www.epa.gov/sites/production/files/2018-
07/documents/smm 2015 tables and figures 07252018 fnl 508 O.pdf.
EPA (2016) Advancing Sustainable Materials Management: Facts and Figures 2014. Office of Solid Waste and
Emergency Response, U.S. Environmental Protection Agency, Washington, D.C. Available online at
https://www.epa.gov/sites/production/files/2016-ll/documents/2014 smm tablesfigures 508.pdf.
EPA (2014) Municipal Solid Waste in the United States: 2012 Facts and Figures. Office of Solid Waste and
Emergency Response, U.S. Environmental Protection Agency, Washington, D.C. Available online at
https://www.epa.gov/sites/default/files/2015-09/documents/2012 msw fs.pdf.
Harvard Law School and Center for EcoTechnology (CET) (2019) Bans and Beyond: Designing and Implementing
Organic Waste Bans and Mandatory Organics Recycling Laws. Prepared by Katie Sandson and Emily Broad Leib,
Harvard Law School Food Law and Policy Clinic, with input from Lorenzo Macaluso and Coryanne Mansell, Center
for EcoTechnology (CET). Available online at https://wastedfood.cetonline.org/wp-
content/uploads/2019/07/Harvard-Law-School-FLPC-Center-for-EcoTechnologv-CET-Organic-Waste-Bans-
Toolkit.pdf.
IPCC (2006) 2006IPCC Guidelines for National Greenhouse Gas Inventories. Volume 5: Waste, Chapter 4: Biological
Treatment of Solid Waste, Table 4.1. The National Greenhouse Gas Inventories Programme, The
Intergovernmental Panel on Climate Change, H.S. Eggleston, L. Buendia, K. Miwa, T. Ngara, and K. Tanabe (eds.).
Hayama, Kanagawa, Japan. Available online at https://www.ipcc-
nggip.iges.or.ip/public/2006gl/pdf/5 Volume5/V5 4 Ch4 Bio Treat.pdf.
Kijanka (2020) Email correspondence. Kenin Kijanka, EPA Region 2 to Rachel Schmeltz, EPA HQ. "Puerto Rico
Composting Operations." November 13, 2020.
University of Maine (2016) Compost Report Interpretation Guide. Soil Testing Lab. Available online at:
https://umaine.edu/soiltestinglab/wp-content/uploads/sites/227/2016/07/Compost-Report-lnterpretation-
Guide.pdf.
U.S. Census Bureau, Population Division (2023) Table 1. Annual Estimates of the Resident Population for the United
States, Regions, States, the District of Columbia, and Puerto Rico: April 1, 2020 to July 1, 2022 (NST-EST2022-POP).
Available online at https://www.census.gov/data/tables/time-series/demo/popest/2020s-state-total.html.
10-118 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
U.S. Census Bureau, Population Division (2022) Table 1. Annual Estimates of the Resident Population for the United
States, Regions, States, the District of Columbia, and Puerto Rico: April 1, 2020 to July 1, 2021 (NST-EST2021-POP).
Available online at https://www.census.gov/data/datasets/time-series/demo/popest/2020s-national-total.html.
U.S. Census Bureau (2021) Table 1. Annual Estimates of the Resident Population for the United States, Regions,
States, the District of Columbia, and Puerto Rico: April 1, 2010 to July 1, 2019; April 1, 2020; and July 1, 2020 (NST-
EST2020). Available online at https://www.census.gov/programs-surveys/popest/technical-
documentation/research/evaluation-estimates/2020-evaluation-estimates/2010s-totals-national.html.
U.S. Composting Council (2022) State and City Organics Bans, as of June 2021. Accessed on September 29, 2022.
Available at https://www.compostingcouncil.org/page/organicsbans.
U.S. Composting Council (2010) Yard Trimmings Bans: Impact and Support. Prepared by Stuart Buckner, Executive
Director, U.S., Composting Council. Available online at
https://cdn.ymaws.com/www.compostingcouncil.org/resource/resmgr/images/advocacy/Yard-Trimmings-Ban-
lmpacts-a.pdf.
Anaerobic Digestion at Biogas Facilities
Bronstein, Kate (2021) Expert Judgement Uncertainty of quantity of materials digested. RTI International, Solid
Waste Management GHG Expert.
EPA (2023) Anaerobic Digestion Facilities Processing Food Waste in the United States (2019): Survey Results. April
2023 EPA 530-R-23-003. April 2023. Available online at https://www.epa.gov/system/files/documents/2023-
04/Anaerobic Digestion Facilities Processing Food Waste in the United States 2019 20230404 508.pdf.
EPA (2021) Anaerobic Digestion Facilities Processing Food Waste in the United States (2017 & 2018): Survey
Results. January 2021 EPA/903/S-21/001. Available online at https://www.epa.gov/sites/default/files/2021-
02/documents/2021 final ad report feb 2 with links.pdf.
EPA (2020) Types of Anaerobic Digesters: Common Ways to Describe Digesters. Available online at
https://www.epa.gov/anaerobic-digestion/types-anaerobic-digesters.
EPA (2019) Anaerobic Digestion Facilities Processing Food Waste in the United States in 2016: Survey Results.
September 2019 EPA/903/S-19/001. Available online at https://www.epa.gov/sites/production/files/2018-
08/documents/ad data report final 508 compliant no password.pdf.
EPA (2018) Anaerobic Digestion Facilities Processing Food Waste in the United States in 2015: Survey Results. May
2018 EPA/903/S-18/001. Available online at https://www.epa.gov/sites/production/files/2019-
09/documents/ad data report vlO - 508 comp vl.pdf.
EPA (2016) Frequently Asked Questions About Anaerobic Digestion. Available online at
https://www.epa.goV/anaerobic-digestion/frequent-questions-about-anaerobic-digestion#codigestion.
EPA (1993) Anthropogenic Methane Emissions in the U.S.: Estimates for 1990, Report to Congress. Office of Air and
Radiation, Washington, DC. April 1993.
IPCC (2006) 2006IPCC Guidelines for National Greenhouse Gas Inventories. Volume 5: Waste, Chapter 4: Biological
Treatment of Solid Waste, Table 4.1. The National Greenhouse Gas Inventories Programme, The Intergovernmental
Panel on Climate Change, H.S. Eggleston, L. Buendia, K. Miwa, T. Ngara, and K. Tanabe (eds.). Hayama, Kanagawa,
Japan. Available online at https://www.ipcc-
nggip.iges.or.ip/public/2006gl/pdf/5 Volume5/V5 4 Ch4 Bio Treat.pdf.
Water Environment Federation (WEF) (2012) What Every Operator Should Know about Anaerobic Digestion.
Available online at https://www.wef.org/globalassets/assets-wef/direct-download-library/public/operator-
essentials/wet-operator-essentials—anaerobic-digestion—dec.12.pdf.
References and Abbreviations 10-119
-------
Waste Incineration
RTI (2009) Hospital/Medical/Infectious Waste Incinerators: Summary of Requirements for Revised or New Section
lll(d)/129 State Plans Following Amendments to the Emission Guidelines. Available online at
https://nepis.epa.gov/Exe/Zy PDF. cgi/P1003ZW6.PDF?Dockey=P1003ZW6. PDF.
Waste Sources of Precursor Greenhouse Gas Emissions
EPA (2023a) "Criteria pollutants National Tier 1 for 1970 - 2023." National Emissions Inventory (NEI) Air Pollutant
Emissions Trends Data. Office of Air Quality Planning and Standards, March 2024. Available online at:
https://www.epa.gov/air-emissions-inventories/air-pollutant-emissions-trends-data.
EPA (2023b) "2020 National Emissions Inventory Technical Support Document: Introduction." Office of Air Quality
Planning and Standards, March 2023. Available online at: https://www.epa.gov/system/files/documents/2023-
01/NE12020 TSD Sectionl lntroduction.pdf.
Recalculations and Improvements
IPCC (2019) 2019 Refinement to the 2006IPCC Guidelines for National Greenhouse Gas Inventories. The National
Greenhouse Gas Inventories Programme, The Intergovernmental Panel on Climate Change. Calvo Buendia, E.,
Tanabe, K., Kranjc, A., Baasansuren, J., Fukuda, M., Ngarize, S., Osako, A., Pyrozhenko, Y., Shermanau, P. and
Federici, S. (eds). Published: IPCC, Switzerland. IPCC (2013) Climate Change 2013: The Physical Science Basis.
Contribution of Working Group I to the Fifth Assessment Report of the Intergovernmental Panel on Climate Change
[Stocker, T.F., D. Qin, G.-K. Plattner, M. Tignor, S.K. Allen, J. Boschung, A. Nauels, Y. Xia, V. Bex and P.M. Midgley
(eds.)]. Cambridge University Press, Cambridge, United Kingdom and New York, NY, USA, 1535 pp.
IPCC (2007) Climate Change 2007: The Physical Science Basis. Contribution of Working Group I to the Fourth
Assessment Report of the Intergovernmental Panel on Climate Change. [S. Solomon, D. Qin, M. Manning, Z. Chen,
M. Marquis, K.B. Averyt, M. Tignor and H.L Miller (eds.)]. Cambridge University Press. Cambridge, United Kingdom
996 pp.
IPCC (2006) 2006 IPCC Guidelines for National Greenhouse Gas Inventories. The National Greenhouse Gas
Inventories Programme, The Intergovernmental Panel on Climate Change. [H.S. Eggleston, L. Buendia, K. Miwa, T.
Ngara, and K. Tanabe (eds.)]. Hayama, Kanagawa, Japan.
10-120 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
Abbreviations
ABS Acrylonitrile butadiene styrene
AC Air conditioner
ACC American Chemistry Council
AEDT FAA Aviation Environmental Design Tool
AEO Annual Energy Outlook
AER All-electric range
AF&PA American Forest and Paper Association
AFEAS Alternative Fluorocarbon Environmental
Acceptability Study
AFOLU Agriculture, Forestry, and Other Land Use
AFV Alternative fuel vehicle
AGA American Gas Association
AGR Acid gas removal
AHEF Atmospheric and Health Effects Framework
AHRI Air-Conditioning, Heating, and Refrigeration
Institute
AIM Act American Innovation and Manufacturing Act
AISI American Iron and Steel Institute
ALU Agriculture and Land Use
ANGA American Natural Gas Alliance
ANL Argonne National Laboratory
APC American Plastics Council
API American Petroleum Institute
APTA American Public Transportation Association
AR4 IPCC Fourth Assessment Report
AR5 IPCC Fifth Assessment Report
AR6 IPCC Sixth Assessment Report
ARI Advanced Resources International
ARMA Autoregressive moving-average
ARMS Agricultural Resource Management Surveys
ASAE American Society of Agricultural Engineers
ASLRRA American Short-line and Regional Railroad
Association
ASR Annual Statistical Report
ASTM American Society for Testing and Materials
AZR American Zinc Recycling
BCEF Biomass conversion and expansion factors
BEA Bureau of Economic Analysis, U.S. Department
of Commerce
BIER Beverage Industry Environmental Roundtable
BLM Bureau of Land Management
BoC Bureau of Census
BOD Biological oxygen demand
BOD5 Biochemical oxygen demand over a 5-day
period
BOEM Bureau of Ocean Energy Management
BOEMRE Bureau of Ocean Energy Management,
Regulation and Enforcement
BOF Basic oxygen furnace
BRS Biennial Reporting System
BSEE Bureau of Safety and Environmental
Enforcement
BTS Bureau of Transportation Statistics, U.S.
Department of Transportation
Btu British thermal unit
C Carbon
C&D Construction and demolition waste
C&EN Chemical and Engineering News
CAAA Clean Air Act Amendments of 1990
CAFOS Concentrated Animal Feeding Operations
CAGR Compound Annual Growth Rate
CaO Calcium oxide
CAPP Canadian Association of Petroleum Producers
CARB California Air Resources Board
CBI Confidential business information
C-CAP Coastal Change Analysis Program
CDAT Chemical Data Access Tool
CEAP USDA-NRCS Conservation Effects Assessment
Program
CEFM Cattle Enteric Fermentation Model
CEMS Continuous emission monitoring system
CFC Chlorofluorocarbon
CFR Code of Federal Regulations
CGA Compressed Gas Association
CH4 Methane
CHAPA California Health and Productivity Audit
CHP Combined heat and power
CI Confidence interval
CIGRE International Council on Large Electric Systems
CKD Cement kiln dust
CLE Crown Light Exposure
CMA Chemical Manufacturer's Association
CMM Coal mine methane
CMOP Coalbed Methane Outreach Program
CMR Chemical Market Reporter
CNG Compressed natural gas
CO Carbon monoxide
C02 Carbon dioxide
COD Chemical oxygen demand
COGCC Colorado Oil and Gas Conservation Commission
CONUS Continental United States
CRF Common Reporting Format
CRM Component ratio method
CRP Conservation Reserve Program
CSRA Carbon Sequestration Rural Appraisals
CTIC Conservation Technology Information Center
CVD Chemical vapor deposition
CWNS Clean Watershed Needs Survey
d.b.h Diameter breast height
DE Digestible energy
References and Abbreviations 10-121
-------
DESC
Defense Energy Support Center-DoD's Defense
FTA
Federal Transit Authority
Logistics Agency
FTP
Federal Test Procedure
DFAMS
Defense Fuels Automated Management System
g
Gram
DGGS
Division of Geological & Geophysical Surveys
G&B
Gathering and boosting
DHS
Department of Homeland Security
GaAs
Gallium arsenide
DLA
DoD's Defense Logistics Agency
GCV
Gross calorific value
DM
Dry matter
GDP
Gross domestic product
DOC
Degradable organic carbon
GEI
Gulfwide Emissions Inventory
DOC
U.S. Department of Commerce
GHG
Greenhouse gas
DoD
U.S. Department of Defense
GHGRP
EPA's Greenhouse Gas Reporting Program
DOE
U.S. Department of Energy
GIS
Geographic Information Systems
DOI
U.S. Department of the Interior
GJ
Gigajoule
DOM
Dead organic matter
GOADS
Gulf Offshore Activity Data System
DOT
U.S. Department of Transportation
GOM
Gulf of Mexico
DRE
Destruction or removal efficiencies
GPG
Good Practice Guidance
DRI
Direct Reduced Iron
GRI
Gas Research Institute
EAF
Electric arc furnace
GSAM
Gas Systems Analysis Model
EDB
Aircraft Engine Emissions Databank
GTI
Gas Technology Institute
EDF
Environmental Defense Fund
GWP
Global warming potential
EER
Energy economy ratio
ha
Hectare
EF
Emission factor
HBFC
Hydrobromofluorocarbon
EFMA
European Fertilizer Manufacturers Association
HC
Hydrocarbon
EJ
Exajoule
HCFC
Hydrochlorofluorocarbon
EGR
Exhaust gas recirculation
HCFO
Hydrochlorofluoroolefin
EGU
Electric generating unit
HDDV
Heavy duty diesel vehicle
EIA
Energy Information Administration, U.S.
HDGV
Heavy duty gas vehicle
Department of Energy
HDPE
High density polyethylene
El IP
Emissions Inventory Improvement Program
HF
Hydraulically fractured
EOR
Enhanced oil recovery
HFC
Hydrofluorocarbon
EPA
U.S. Environmental Protection Agency
HFO
Hydrofluoroolefin
EPRI
Electric Power Research Institute
HFE
Hydrofluoroether
EREF
Environment Research & Education Foundation
HHV
Higher Heating Value
ERS
Economic Research Service
HMA
Hot Mix Asphalt
ETMS
Enhanced Traffic Management System
HMIWI
Hospital/medical/infectious waste incinerator
EV
Electric vehicle
HTF
Heat Transfer Fluid
EVI
Enhanced Vegetation Index
HTS
Harmonized Tariff Schedule
FAA
Federal Aviation Administration
HVAE
High Voltage Anode Effects
FAO
Food and Agricultural Organization
HWP
Harvested wood product
FAOSTAT
Food and Agricultural Organization database
IBF
International bunker fuels
FAS
Fuels Automated System
IC
Integrated Circuit
FCCC
Framework Convention on Climate Change
ICAO
International Civil Aviation Organization
FEB
Fiber Economics Bureau
ICBA
International Carbon Black Association
FEMA
Federal Emergency Management Agency
ICE
Internal combustion engine
FERC
Federal Energy Regulatory Commission
ICR
Information Collection Request
FGD
Flue gas desulfurization
IEA
International Energy Agency
FHWA
Federal Highway Administration
IFO
Intermediate Fuel Oil
FIA
Forest Inventory and Analysis
IGES
Institute of Global Environmental Strategies
FIADB
Forest Inventory and Analysis Database
IISRP
International Institute of Synthetic Rubber
FIPR
Florida Institute of Phosphate Research
Products
FOD
First order decay
ILENR
Illinois Department of Energy and Natural
FOEN
Federal Office for the Environment
Resources
FOKS
Fuel Oil and Kerosene Sales
IMO
International Maritime Organization
FQSV
First-quarter of silicon volume
IPAA
Independent Petroleum Association of Americ;
FSA
Farm Service Agency
IPCC
Intergovernmental Panel on Climate Change
10-122 Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2022
-------
IPPU Industrial Processes and Product Use MOVES
ISO International Organization for Standardization
ITC U.S. International Trade Commission MPG
ITRS International Technology Roadmap for MRLC
Semiconductors
JWR Jim Walters Resources MRV
KCA Key category analysis MSHA
kg Kilogram MSW
kt Kiloton MT
kWh Kilowatt hour MTBE
LDPE Low density polyethylene MTBS
LDT Light-duty truck MVAC
LDV Light-duty vehicle MY
LEV Low emission vehicles N20
LFG Landfill gas NA
LFGTE Landfill gas-to-energy NACWA
LHV Lower Heating Value NAFTA
LKD Lime kiln dust NAHMS
LLDPE Linear low density polyethylene NAICS
LMOP EPA's Landfill Methane Outreach Program NAPAP
LNG Liquefied natural gas
LPG Liquefied petroleum gas(es) NARR
LTO Landing and take-off NAS
LULUCF Land Use, Land-Use Change, and Forestry
LVAE Low Voltage Anode Effects NASA
M&R Metering and regulating NASF
MARPOL International Convention for the Prevention of NASS
Pollution from Ships NC
MC Motorcycle NCASI
MCF Methane conversion factor
MCL Maximum Contaminant Levels NCV
MCFD Thousand cubic feet per day ND
MDI Metered dose inhalers NE
MDP Management and design practices NEH
MECS EIA Manufacturer's Energy Consumption Survey NEI
MEMS Micro-electromechanical systems NEMA
MER Monthly Energy Review NEMS
MGO Marine gas oil NESHAP
MgO Magnesium oxide
MJ Megajoule NEU
MLRA Major Land Resource Area NEV
mm Millimeter NF3
MMBtu Million British thermal units NFI
MMCF Million cubic feet NGL
MMCFD Million cubic feet per day NGO
MMS Minerals Management Service NID
MMT Million metric tons NIR
MMTCE Million metric tons carbon equivalent NLA
MMT C02 Million metric tons carbon dioxide equivalent NLCD
Eq. NMOC
MODIS Moderate Resolution Imaging NMVOC
Spectroradiometer NMOG
MoU Memorandum of Understanding NO
N02
U.S. EPA's Motor Vehicle Emission Simulator
model
Miles per gallon
Multi-Resolution Land Characteristics
Consortium
Monitoring, reporting, and verification
Mine Safety and Health Administration
Municipal solid waste
Metric ton
Methyl Tertiary Butyl Ether
Monitoring Trends in Burn Severity
Motor vehicle air conditioning
Model year
Nitrous oxide
Not applicable; Not available
National Association of Clean Water Agencies
North American Free Trade Agreement
National Animal Health Monitoring System
North American Industry Classification System
National Acid Precipitation and Assessment
Program
North American Regional Reanalysis Product
National Academies of Sciences, Engineering,
and Medicine
National Aeronautics and Space Administration
National Association of State Foresters
USDA's National Agriculture Statistics Service
No change
National Council of Air and Stream
Improvement
Net calorific value
No data
Not estimated
National Engineering Handbook
National Emissions Inventory
National Electrical Manufacturers Association
National Energy Modeling System
National Emission Standards for Hazardous Air
Pollutants
Non-Energy Use
Neighborhood Electric Vehicle
Nitrogen trifluoride
National forest inventory
Natural gas liquids
Non-Governmental Organization
National inventory of Dams
National Inventory Report
National Lime Association
National Land Cover Dataset
Non-methane organic compounds
Non-methane volatile organic compound
Non-methane organic gas
Not occurring
Nitrogen dioxide
References and Abbreviations 10-123
-------
N0X
Nitrogen oxides
ppm
Parts per million
NOAA
National Oceanic and Atmospheric
ppmv
Parts per million (106) by volume
Administration
pptv
Parts per trillion (1012) by volume
NOF
Not on feed
PRCI
Pipeline Research Council International
NPDES
National Pollutant Discharge Elimination System
PRP
Pasture/Range/Paddock
NPP
Net primary productivity
PS
Polystyrene
NPRA
National Petroleum and Refiners Association
PSU
Primary Sample Unit
NRBP
Northeast Regional Biomass Program
PU
Polyurethane
NRC
National Research Council
PVC
Polyvinyl chloride
NRCS
Natural Resources Conservation Service
PV
Photovoltaic
NREL
National Renewable Energy Laboratory
QA/QC
Quality Assurance and Quality Control
NRI
National Resources Inventory
QBtu
Quadrillion Btu
NSCEP
National Service Center for Environmental
R&D
Research and Development
Publications
RECs
Reduced Emissions Completions
NSCR
Non-selective catalytic reduction
RCRA
Resource Conservation and Recovery Act
NSPS
New source performance standards
RFA
Renewable Fuels Association
NWS
National Weather Service
RFS
Renewable Fuel Standard
OAG
Official Airline Guide
RMA
Rubber Manufacturers' Association
OAP
EPA Office of Atmospheric Programs
RPA
Resources Planning Act
OAQPS
EPA Office of Air Quality Planning and Standards
RTO
Regression-through-the-origin
ODP
Ozone depleting potential
SAE
Society of Automotive Engineers
ODS
Ozone depleting substances
SAGE
System for assessing Aviation's Global Emissions
OECD
Organization of Economic Co-operation and
SAIC
Science Applications International Corporation
Development
SAN
Styrene Acrylonitrile
OEM
Original equipment manufacturers
SAR
IPCC Second Assessment Report
OGJ
Oil & Gas Journal
SCR
Selective catalytic reduction
OGOR
Oil and Gas Operations Reports
SCSE
South central and southeastern coastal
OH
Hydroxyl radical
SDR
Steel dust recycling
OPEC
Organization of Petroleum-Exporting Countries
SEC
Securities and Exchange Commission
OMS
EPA Office of Mobile Sources
SEMI
Semiconductor Equipment and Materials
ORNL
Oak Ridge National Laboratory
Industry
OSHA
Occupational Safety and Health Administration
sf6
Sulfur hexafluoride
OTA
Office of Technology Assessment
SIA
Semiconductor Industry Association
OTAQ
EPA Office of Transportation and Air Quality
SiC
Silicon carbide
OVS
Offset verification statement
SICAS
Semiconductor International Capacity Statistics
PADUS
Protected Areas Database of the United States
SNAP
Significant New Alternative Policy Program
PAH
Polycyclic aromatic hydrocarbons
SNG
Synthetic natural gas
PCA
Portland Cement Association
S02
Sulfur dioxide
PCC
Precipitate calcium carbonate
SOC
Soil Organic Carbon
PDF
Probability Density Function
SOG
State of Garbage survey
PECVD
Plasma enhanced chemical vapor deposition
SOHIO
Standard Oil Company of Ohio
PET
Polyethylene terephthalate
SSURGO
Soil Survey Geographic Database
PET
Potential evapotranspiration
STMC
Scrap Tire Management Council
PEVM
PFC Emissions Vintage Model
SULEV
Super Ultra Low Emissions Vehicle
PFC
Perfluorocarbon
SWANA
Solid Waste Association of North America
PFPE
Perfluoropolyether
SWDS
Solid waste disposal sites
PHEV
Plug-in hybrid vehicles
SWICS
Solid Waste Industry for Climate Solutions
PHMSA
Pipeline and Hazardous Materials Safety
TA
Treated anaerobically (wastewater)
Administration
TAM
Typical animal mass
PI
Productivity index
TAME
Tertiary amyl methyl ether
PLS
Pregnant liquor solution
TAR
IPCC Third Assessment Report
PM
Particulate matter
TBtu
Trillion Btu
POTW
Publicly Owned Treatment Works
TDN
Total digestible nutrients
ppbv
Parts per billion (109) by volume
TEDB
Transportation Energy Data Book
10-124
Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-
-2022
-------
TFI The Fertilizer Institute
TIGER Topological^ Integrated Geographic Encoding
and Referencing survey
TJ Terajoule
TLEV Traditional low emissions vehicle
TMLA Total Manufactured Layer Area
TOW Total organics in wastewater
TPO Timber Product Output
TRI Toxic Release Inventory
TSDF Hazardous waste treatment, storage, and
disposal facility
TTB Tax and Trade Bureau
TVA Tennessee Valley Authority
UAN Urea ammonium nitrate
UDI Utility Data Institute
UFORE U.S. Forest Service's Urban Forest Effects model
UG Underground (coal mining)
U.S. United States
U.S. ITC United States International Trade Commission
UEP United Egg Producers
ULEV Ultra low emission vehicle
UNEP United Nations Environmental Programme
UNFCCC United Nations Framework Convention on
Climate Change
USAA U.S. Aluminum Association
USAF United States Air Force
USDA United States Department of Agriculture
USFS United States Forest Service
USGS United States Geological Survey
USITC U.S. International Trade Commission
VAIP EPA's Voluntary Aluminum Industrial
Partnership
VAM Ventilation air methane
VKT Vehicle kilometers traveled
VMT Vehicle miles traveled
VOCs Volatile organic compounds
VS Volatile solids
WBJ Waste Business Journal
WEF Water Environment Federation
WERF Water Environment Research Federation
WFF World Fab Forecast (previously WFW, World
Fab Watch)
WGC World Gas Conference
WIP Waste-in-place
WMO World Meteorological Organization
WMS Waste management systems
WRRF Water resource recovery facilities
WTE Waste-to-energy
WW Wastewater
WWTP Wastewater treatment plant
ZEVs Zero emissions vehicles
References and Abbreviations 10-125
------- |