EPA 430-P-18-001 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2016 February 6,2018 U.S. Environmental Protection Agency 1200 Pennsylvania Ave., N.W. Washington, DC 20460 U.S.A. ------- 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 HOW TO OBTAIN COPIES You can electronically download this document on the U.S. EPA's homepage at . All data tables of this document for the full time series 1990 through 2016, inclusive, will be made available for the final report published on April 15, 2018 at the internet site mentioned above. FOR FURTHER INFORMATION Contact Ms. Mausami Desai, Environmental Protection Agency, (202) 343-9381, desai.mausami@epa.gov, or Mr. Vincent Camobreco, Environmental Protection Agency, (202) 564-9043, camobreco.vincent@epa.gov. For more information regarding climate change and greenhouse gas emissions, see the EPA web site at . ------- 1 Acknowledgments 2 The Environmental Protection Agency would like to acknowledge the many individual and organizational 3 contributors to this document, without whose efforts this report would not be complete. Although the complete list 4 of researchers, government employees, and consultants who have provided technical and editorial support is too 5 long to list here, EPA's Office of Atmospheric Programs would like to thank some key contributors and reviewers 6 whose work has significantly improved this year's report. 7 Work on emissions from fuel combustion was led by Vincent Camobreco. Sarah Roberts and Justine Geidosch 8 directed the work on mobile combustion and transportation. Work on fugitive methane emissions from the Energy 9 sector was directed by Melissa Weitz, Chris Sherry, and Cate Hight. Calculations for the Waste sector were led by 10 Rachel Schmeltz. Tom Wirth directed work on the Agriculture and the Land Use, Land-Use Change, and Forestry 11 chapters, with support from John Steller. Work on Industrial Processes and Product Use (IPPU) CO2, CH4, and N20 12 emissions was directed by John Steller. Work on emissions of HFCs, PFCs, SF6, and NF3 from the IPPU sector was 13 directed by Deborah Ottinger and Dave Godwin. Cross-cutting work was directed by Mausami Desai. 14 Within the EPA, other Offices also contributed data, analysis, and technical review for this report. The Office of 15 Transportation and Air Quality and the Office of Air Quality Planning and Standards provided analysis and review 16 for several of the source categories addressed in this report. The Office of Solid Waste and the Office of Research 17 and Development also contributed analysis and research. 18 The Energy Information Administration and the Department of Energy contributed invaluable data and analysis on 19 numerous energy-related topics. Other government agencies have contributed data as well, including the U.S. 20 Geological Survey, the Federal Highway Administration, the Department of Transportation, the Bureau of 21 Transportation Statistics, the Department of Commerce, the National Agricultural Statistics Service, the Federal 22 Aviation Administration, and the Department of Defense. 23 We thank the U.S. Department of Agriculture's Forest Service (Grant Domke, Brian Walters, Jim Smith, Mike 24 Nichols, and John Coulston) for compiling the inventories for carbon dioxide (CO2), methane (CH4), and nitrous 25 oxide (N20) fluxes associated with forest land. 26 We thank the Department of Agriculture's Agricultural Research Service (Stephen Del Grosso) and the Natural 27 Resource Ecology Laboratory at Colorado State University (Stephen Ogle, Keith Paustian, Bill Parton, F. Jay Breidt, 28 Shannon Spencer, Kendrick Killian, Ram Gurung, Ernie Marx, Stephen Williams, Cody Alsaker, Amy Swan, and 29 Chris Dorich) for compiling the inventories for CH4 emissions, N20 emissions, and CO2 fluxes associated with soils 30 in croplands, grasslands, and settlements. 31 We thank Silvestrum Climate Associates (Stephen Crooks, Lisa Schile Beers, Christine May), National Oceanic and 32 Atmospheric Administration (Nate Herold, Ariana Sutton-Grier, Meredith Muth), the Smithsonian Environmental 33 Research Center (J. Patrick Megonigal, Blanca Bernal, James Holmquist, Meng Lu) and Florida International 34 University (Tiffany Troxler) and members of the U.S. Coastal Wetland Carbon Working Group for compiling 35 inventories of land use change, soil carbon stocks and stock change, CH4 emissions, and N20 emissions from 36 aquaculture in coastal wetlands. 37 We would also like to thank Marian Martin Van Pelt, Leslie Chinery, Alexander Lataille, Sabrina Andrews and the 38 full Inventory team at ICF including Diana Pape, Robert Lanza, Lauren Marti, Mollie Averyt, Larry O'Rourke, 39 Deborah Harris, Tommy Hendrickson, Rebecca Ferenchiak, Kasey Knoell, Cory Jemison, Emily Kent, Rani Murali, ------- 1 Drew Stilson, Cara Blumenthal, Louise Huttinger, Helena Caswell, Charlotte Cherry, Katie O'Malley, Howard 2 Marano, and Neha Vaingankar for synthesizing this report and preparing many of the individual analyses. 3 We thank Eastern Research Group for their significant analytical support. Deborah Bartram, Kara Edquist, and 4 Amie Aguiar support the development of emissions estimates for wastewater. Cortney Itle, Amie Aguiar, Kara 5 Edquist, Amber Allen, and Spencer Sauter support the inventories for Manure Management, Enteric Fermentation, 6 Wetlands Remaining Wetlands, and Landfilled Yard Trimmings and Food Scraps (included in Settlements 7 Remaining Settlements). Casey Pickering, Brandon Long, Gopi Manne, and Aylin Sertkaya develop estimates for 8 Natural Gas and Petroleum Systems. Brian Guzzone supports the Coal Mining sector. 9 Finally, we thank the following teams for their significant analytical support: RTI International (Kate Bronstein, 10 Meaghan McGrath, Michael Laney, Carson Moss, David Randall, Gabrielle Raymond, Jason Goldsmith, Karen 11 Schaffner, Melissa Icenhour); Raven Ridge Resources, and Ruby Canyon Engineering Inc. (Michael Cote, Samantha 12 Phillips, and Phillip Cunningham). 13 ------- 1 Preface 2 The United States Environmental Protection Agency (EPA) prepares the official U.S. Inventory of Greenhouse Gas 3 Emissions and Sinks to comply with existing commitments under the United Nations Framework Convention on 4 Climate Change (UNFCCC). Under decision 3/CP.5 of the UNFCCC Conference of the Parties, national inventories 5 for UNFCCC Annex I parties should be provided to the UNFCCC Secretariat each year by April 15. 6 In an effort to engage the public and researchers across the country, the EPA has instituted an annual public review 7 and comment process for this document. The availability of the draft document is announced via Federal Register 8 Notice and is posted on the EPA web site. Copies are also emailed upon request. The public comment period is 9 generally limited to 30 days; however, comments received after the closure of the public comment period are 10 accepted and considered for the next edition of this annual report. Public review of this report is occurring in 11 February 2018, and comments received will be posted to the EPA web site. ------- 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 Table of Contents TABLE OF CONTENTS VI LIST OF TABLES, FIGURES, AND BOXES IX EXECUTIVE SUMMARY ES-1 ES. 1 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-18 ES.4 Other Information ES-23 1. INTRODUCTION 1-1 1.1 Background Information 1-3 1.2 National Inventory Arrangements 1-10 1.3 Methodology and Data Sources 1-15 1.4 Key Categories 1-16 1.5 Quality Assurance and Quality Control (QA/QC) 1-19 1.6 Uncertainty Analysis of Emission Estimates 1-21 1.7 Completeness 1-23 1.8 Organization of Report 1-23 2. TRENDS IN GREENHOUSE GAS EMISSIONS 2-1 2.1 Recent Trends in U.S. Greenhouse Gas Emissions and Sinks 2-1 2.2 Emissions by Economic Sector 2-23 2.3 Indirect Greenhouse Gas Emissions (CO, NOx, NMVOCs, and SO2) 2-34 3. ENERGY 3-1 3.1 Fossil Fuel Combustion (CRF Source Category 1A) 3-5 3.2 Carbon Emitted from Non-Energy Uses of Fossil Fuels (CRF Source Category 1A) 3-45 3.3 Incineration of Waste (CRF Source Category lAla) - TO BE UPDATED FOR FINAL INVENTORY REPORT 3-51 3.4 Coal Mining (CRF Source Category lBla) 3-56 3.5 Abandoned Underground Coal Mines (CRF Source Category lBla) 3-61 3.6 Petroleum Systems (CRF Source Category lB2a) 3-65 3.7 Natural Gas Systems (CRF Source Category lB2b) 3-77 vi DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2016 ------- 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 3.8 Abandoned Oil and Gas Wells (CRF Source Categories lB2a and lB2b) 3-93 3.9 Energy Sources of Indirect Greenhouse Gas Emissions 3-96 3.10 International Bunker Fuels (CRF Source Category 1: Memo Items) 3-97 3.11 WoodBiomass andBiofuels Consumption (CRF Source Category 1A) 3-101 4. INDUSTRIAL PROCESSES AND PRODUCT USE 4-1 4.1 Cement Production (CRF Source Category 2A1) 4-8 4.2 Lime Production (CRF Source Category 2A2) - TO BE UPDATED FOR FINAL INVENTORY REPORT. 4-12 4.3 Glass Production (CRF Source Category 2A3) - TO BE UPDATED FOR FINAL INVENTORY REPORT 4-17 4.4 Other Process Uses of Carbonates (CRF Source Category 2A4) - TO BE UPDATED FOR FINAL INVENTORY REPORT 4-20 4.5 Ammonia Production (CRF Source Category 2B1) 4-24 4.6 Urea Consumption for Non-Agricultural Purposes 4-28 4.7 Nitric Acid Production (CRF Source Category 2B2) 4-31 4.8 Adipic Acid Production (CRF Source Category 2B3) 4-34 4.9 Caprolactam, Glyoxal and Glyoxylic Acid Production (CRF Source Category 2B4) 4-38 4.10 Silicon Carbide Production and Consumption (CRF Source Category 2B5) 4-41 4.11 Titanium Dioxide Production (CRF Source Category 2B6) 4-44 4.12 Soda Ash Production (CRF Source Category 2B7) 4-47 4.13 Petrochemical Production (CRF Source Category 2B8) 4-50 4.14 HCFC-22 Production (CRF Source Category 2B9a) 4-56 4.15 Carbon Dioxide Consumption (CRF Source Category 2B10) 4-59 4.16 Phosphoric Acid Production (CRF Source Category 2B10) 4-62 4.17 Iron and Steel Production (CRF Source Category 2C1) and Metallurgical Coke Production 4-66 4.18 Ferroalloy Production (CRF Source Category 2C2) 4-76 4.19 Aluminum Production (CRF Source Category 2C3) 4-79 4.20 Magnesium Production and Processing (CRF Source Category 2C4) 4-84 4.21 Lead Production (CRF Source Category 2C5) 4-89 4.22 Zinc Production (CRF Source Category 2C6) 4-92 4.23 Semiconductor Manufacture (CRF Source Category 2E1) 4-97 4.24 Substitution of Ozone Depleting Substances (CRF Source Category 2F) 4-109 4.25 Electrical Transmission and Distribution (CRF Source Category 2G1) 4-117 4.26 Nitrous Oxide from Product Uses (CRF Source Category 2G3) 4-124 4.27 Industrial Processes and Product Use Sources of Indirect Greenhouse Gases 4-127 5. AGRICULTURE 5-1 5.1 Enteric Fermentation (CRF Source Category 3A) 5-3 5.2 Manure Management (CRF Source Category 3B) 5-9 vii ------- 1 5.3 Rice Cultivation (CRF Source Category 3C) 5-17 2 5.4 Agricultural Soil Management (CRF Source Category 3D) 5-23 3 5.5 Liming (CRF Source Category 3G) 5-42 4 5.6 Urea Fertilization (CRF Source Category 3H) 5-45 5 5.7 Field Burning of Agricultural Residues (CRF Source Category 3F) 5-47 6 6. LAND USE, LAND-USE CHANGE, AND FORESTRY 6-1 7 6.1 Representation of the U.S. Land Base 6-8 8 6.2 Forest Land Remaining Forest Land (CRF Category 4 Al) 6-22 9 6.3 Land Converted to Forest Land (CRF Category 4A2) 6-42 10 6.4 Cropland Remaining Cropland (CRF Category 4B1) 6-48 11 6.5 Land Converted to Cropland (CRF Category 4B2) 6-57 12 6.6 Grassland Remaining Grassland (CRF Category 4C1) 6-63 13 6.7 Land Converted to Grassland (CRF Category 4C2) 6-72 14 6.8 Wetlands Remaining Wetlands (CRF Category 4D1) 6-79 15 6.9 Land Converted to Wetlands (CRF Category 4D2) 6-96 16 6.10 Settlements Remaining Settlements (CRF Category 4E1) 6-99 17 6.11 Land Converted to Settlements (CRF Category 4E2) 6-116 18 6.12 Other Land Remaining Other Land (CRF Category 4F1) 6-121 19 6.13 Land Converted to Other Land (CRF Category 4F2) 6-122 20 7. WASTE 7-1 21 7.1 Landfills (CRF Source Category 5A1) 7-3 22 7.2 Wastewater Treatment (CRF Source Category 5D) 7-19 23 7.3 Composting (CRF Source Category 5B1) 7-33 24 7.4 Waste Incineration (CRF Source Category 5C1) - TO BE UPDATED FOR FINAL INVENTORY 25 REPORT 7-35 26 7.5 Waste Sources of Indirect Greenhouse Gases 7-36 27 8. OTHER 8-1 28 9. RECALCULATIONS AND IMPROVEMENTS 9-1 29 10. REFERENCES 10-1 30 31 32 viii DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2016 ------- 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 List of Tables, Figures, and Boxes 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-6 Table ES-3: CO2 Emissions from Fossil Fuel Combustion by End-Use Sector (MMT CO2 Eq.) ES-11 Table ES-4: Recent Trends in U.S. Greenhouse Gas Emissions and Sinks by Chapter/IPCC Sector (MMT CO2 Eq.) ES-18 Table ES-5: U.S. Greenhouse Gas Emissions and Removals (Net Flux) from Land Use, Land-Use Change, and Forestry (MMT CO Eq.) ES-22 Table ES-6: U.S. Greenhouse Gas Emissions Allocated to Economic Sectors (MMT CO2 Eq.) ES-24 Table ES-7: U.S. Greenhouse Gas Emissions by Economic Sector with Electricity-Related Emissions Distributed (MMTCO2 Eq.) ES-25 Table ES-8: Recent Trends in Various U.S. Data (Index 1990 = 100) ES-26 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 1-10 Table 1-4: Key Categories for the United States (1990-2016) 1-16 Table 1-5: Estimated Overall Inventory Quantitative Uncertainty (MMT CO2 Eq. and Percent) - TO BE UPDATED FOR FINAL INVENTORY REPORT 1-22 Table 1-6: IPCC Sector Descriptions 1-24 Table 1-7: List of Annexes 1-24 Table 2-1: Recent Trends in U.S. Greenhouse Gas Emissions and Sinks (MMT CO2 Eq.) 2-3 Table 2-2: Recent Trends in U.S. Greenhouse Gas Emissions and Sinks (kt) 2-5 Table 2-3: Recent Trends in U.S. Greenhouse Gas Emissions and Sinks by Chapter/IPCC Sector (MMT CO2 Eq.).... 2-7 Table 2-4: Emissions from Energy (MMT CO2 Eq.) 2-10 Table 2-5: CO2 Emissions from Fossil Fuel Combustion by End-Use Sector (MMT CO2 Eq.) 2-12 Table 2-6: Emissions from Industrial Processes and Product Use (MMT CO2 Eq.) 2-16 Table 2-7: Emissions from Agriculture (MMT CO2 Eq.) 2-18 Table 2-8: U.S. Greenhouse Gas Emissions and Removals (Net Flux) from Land Use, Land-Use Change, and Forestry (MMT CO2 Eq.) 2-20 U.S. Greenhouse Gas Emissions Allocated to Economic Sectors (MMT CO2 Eq. and Percent of Total in 2-24 Table 2-9: Emissions from Waste (MMT CO2 Eq.) 2-22 Table 2-10 2016) Table 2-11: Electric Power-Related Greenhouse Gas Emissions (MMT CO2 Eq.) 2-26 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 2016 2-27 ix ------- 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 Table 2-13: Transportation-Related Greenhouse Gas Emissions (MMT CO2 Eq.) 2-30 Table 2-14: Recent Trends in Various U.S. Data (Index 1990 = 100) 2-33 Table 2-15: Emissions of NO,. CO, NMVOCs, and SO -(kt) 2-35 Table 3-1: CO2, CH4, and N20 Emissions from Energy (MMT CO2 Eq.) 3-2 Table 3-2: CO2, CH4, and N20 Emissions from Energy (kt) 3-3 Table 3-3: CO2, CH4, and N20 Emissions from Fossil Fuel Combustion (MMT CO2 Eq.) 3-5 Table 3-4: CO2, CH4, and N20 Emissions from Fossil Fuel Combustion (kt) 3-5 Table 3-5: CO2 Emissions from Fossil Fuel Combustion by Fuel Type and Sector (MMT CO2 Eq.) 3-6 Table 3-6: Annual Change in CO2 Emissions and Total 2016 Emissions fromFossil Fuel Combustion for Selected Fuels and Sectors (MMT CO2 Eq. and Percent) 3-7 Table 3-7: CO2, CH4, andN20 Emissions fromFossil Fuel Combustion by Sector (MMT CO2 Eq.) 3-11 Table 3-8: CO2, CH4, andN20 Emissions fromFossil Fuel Combustion by End-Use Sector (MMT CO2 Eq.).... 3-12 Table 3-9: CO2 Emissions from Stationary Fossil Fuel Combustion (MMT CO2 Eq.) 3-13 Table 3-10: CH4 Emissions from Stationary Combustion (MMT CO2 Eq.) 3-13 Table 3-11: N20 Emissions from Stationary Combustion (MMT CO2 Eq.) 3-14 Table 3-12: CO2 Emissions from Fossil Fuel Combustion in Transportation End-Use Sector (MMT CO2 Eq.)... 3-24 Table 3-13: CH4 Emissions from Mobile Combustion (MMT CO2 Eq.) 3-26 Table 3-14: N2O Emissions from Mobile Combustion (MMT CO2 Eq.) 3-27 Table 3-15: Carbon Intensity from Direct Fossil Fuel Combustion by Sector (MMT CO2 Eq./QBtu) 3-32 Table 3-16: Approach 2 Quantitative Uncertainty Estimates for CO2 Emissions from Energy-Related Fossil Fuel Combustion by Fuel Type and Sector (MMT CO2 Eq. and Percent) 3-35 Table 3-17: Approach 2 Quantitative Uncertainty Estimates for CH4 and N20 Emissions from Energy-Related Stationary Combustion, Including Biomass (MMT CO2 Eq. and Percent) 3-39 Table 3-18: Approach 2 Quantitative Uncertainty Estimates for CH4 and N20 Emissions from Mobile Sources (MMT CO2 Eq. and Percent) 3-42 Table 3-19: CO2 Emissions from Non-Energy Use Fossil Fuel Consumption (MMT CO2 Eq. and Percent) 3-45 Table 3-20: Adjusted Consumption of Fossil Fuels for Non-Energy Uses (TBtu) 3-46 Table 3-21: 2016 Adjusted Non-Energy Use Fossil Fuel Consumption, Storage, and Emissions 3-47 Table 3-22: Approach 2 Quantitative Uncertainty Estimates for CO2 Emissions from Non-Energy Uses of Fossil Fuels (MMT CO2 Eq. and Percent) 3-48 Table 3-23: Approach 2 Quantitative Uncertainty Estimates for Storage Factors of Non-Energy Uses of Fossil Fuels (Percent) 3-49 Table 3-24: CO2, CH4, andN20 Emissions from the Incineration of Waste (MMT CO2 Eq.) 3-52 Table 3-25: CO2, CH4, andN20 Emissions from the Incineration of Waste (kt) 3-52 Table 3-26: Municipal Solid Waste Generation (Metric Tons) and Percent Combusted (BioCycle dataset) 3-54 Table 3-27: Approach 2 Quantitative Uncertainty Estimates for CO2 andN20 from the Incineration of Waste (MMT CO2 Eq. and Percent) 3-55 Table 3-28: Coal Production (kt) 3-57 Table 3-29: CH4 Emissions from Coal Mining (MMT CO2 Eq.) 3-57 x DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2016 ------- 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 Table 3-30: CH4 Emissions from Coal Mining (kt) 3-57 Table 3-31: Approach 2 Quantitative Uncertainty Estimates for CH4 Emissions from Coal Mining (MMT CO2 Eq. and Percent) 3-61 Table 3-32: CH4 Emissions from Abandoned Coal Mines (MMT CO2 Eq.) 3-62 Table 3-33: CH4 Emissions from Abandoned Coal Mines (kt) 3-62 Table 3-34: Number of Gassy Abandoned Mines Present in U.S. Basins in 2016, grouped by Class according to Post-Abandonment State 3-63 Table 3-35: Approach 2 Quantitative Uncertainty Estimates for CH4 Emissions from Abandoned Underground Coal Mines (MMT CO2 Eq. and Percent) 3-65 Table 3-36: CH4 Emissions from Petroleum Systems (MMT CO2 Eq.) 3-66 Table 3-37: CH4 Emissions from Petroleum Systems (kt) 3-66 Table 3-38: CO2 Emissions from Petroleum Systems (MMT CO2) 3-67 Table 3-39: CO2 Emissions from Petroleum Systems (kt) 3-67 Table 3-40: Approach 2 Quantitative Uncertainty Estimates for CH4 Emissions from Petroleum Systems (MMT CO2 Eq. and Percent) 3-69 Table 3-41: Oil Well Testing National CH4 Emissions (Metric Tons CH4) 3-71 Table 3-42: Oil Well Testing National CO2 Emissions (Metric Tons CO2) 3-71 Table 3-43: National Tank CO2 Emissions by Category and National Emissions (kt CO2) 3-72 Table 3-44: Associated Gas Venting and Flaring National CO2 Emissions (kt CO2) 3-73 Table 3-45: Associated Gas Venting and Flaring National CH4 Emissions (Metric Tons CH4) 3-73 Table 3-46: Miscellaneous Production Flaring National CO2 Emissions (kt CO2) 3-73 Table 3-47: Miscellaneous Production Flaring National CH4 Emissions (Metric Tons CH4) 3-74 Table 3-48: Producing Oil Well Count Data 3-74 Table 3-49: Quantity of CO2 Captured and Extracted for EOR Operations (MMT CO2) 3-77 Table 3-50: Quantity of CO2 Captured and Extracted for EOR Operations (kt) 3-77 Table 3-51: CH4 Emissions from Natural Gas Systems (MMT CO2 Eq.)a 3-79 Table 3-52: CH4 Emissions from Natural Gas Systems (kt)a 3-79 Table 3-53: Calculated Potential CH4 and Captured/Combusted CH4 from Natural Gas Systems (MMT CO2 Eq.) 3-80 Table 3-54: Non-combustion CO2 Emissions from Natural Gas Systems (MMT) 3-80 Table 3-55: Non-combustion CO2 Emissions from Natural Gas Systems (kt) 3-80 Table 3-56: Approach 2 Quantitative Uncertainty Estimates for CH4 and Non-energy CO2 Emissions from Natural Gas Systems (MMT CO2 Eq. and Percent) 3-83 Table 3-57: Gas Well Testing National CH4 Emissions (Metric Tons CH4) 3-85 Table 3-58: Gas Well Testing National CO2 Emissions (Metric Tons CO2) 3-85 Table 3-59: Non-HF Gas Well Completions National CH4 Emissions (Metric Tons CH4) 3-86 Table 3-60: Non-HF Gas Well Completions National CO2 Emissions (Metric Tons CO2) 3-86 Table 3-61: HF Gas Well Completions National CO2 Emissions (kt CO2) 3-86 Table 3-62: Non-HF Gas Well Workovers National CH4 Emissions (Metric Tons CH4) 3-87 xi ------- 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 Table 3-63: Non-HF Gas Well Workovers National CO2 Emissions (kt CO2) 3-87 Table 3-64: Producing Gas Well Count Data 3-87 Table 3-65: Miscellaneous Production Flaring National CO2 Emissions (kt CO2) 3-88 Table 3-66: National Condensate Tank Emissions by Category and National Emissions (kt CO2) 3-88 Table 3-67: Processing CO2 Updates, National Emissions (kt CO2) 3-89 Table 3-68: Transmission and Storage CH4 Updates to Flaring, National Emissions (MT CH4) 3-89 Table 3-69: Transmission and Storage CO2 Updates, National Emissions (kt CO2) 3-90 Table 3-70: CH4 Emissions from Abandoned Oil and Gas Wells (MMT CO2 Eq.) 3-93 Table 3-71: CH4 Emissions from Abandoned Oil and Gas Wells (kt) 3-93 Table 3-72: Abandoned Oil Wells Activity Data and Methane Emissions (Metric Tons CH4) 3-94 Table 3-73: Abandoned Gas Wells Activity Data and Methane Emissions (Metric Tons CH4) 3-94 Table 3-74: Approach 2 Quantitative Uncertainty Estimates for CH4 Emissions from Petroleum Systems (MMT CO2 Eq. and Percent) 3-95 Table 3-75: NOx, CO, and NMVOC Emissions from Energy-Related Activities (kt) 3-96 Table 3-76: CO2, CH4, and N2O Emissions from International Bunker Fuels (MMT CO2 Eq.) 3-98 Table 3-77: CO2, CH4, and N20 Emissions from International Bunker Fuels (kt) 3-98 Table 3-78: Aviation Jet Fuel Consumption for International Transport (Million Gallons) 3-100 Table 3-79: Marine Fuel Consumption for International Transport (Million Gallons) 3-100 Table 3-80: CO2 Emissions from Wood Consumption by End-Use Sector (MMT CO2 Eq.) 3-102 Table 3-81: CO2 Emissions from Wood Consumption by End-Use Sector (kt) 3-102 Table 3-82: CO2 Emissions from Ethanol Consumption (MMT CO2 Eq.) 3-102 Table 3-83: CO2 Emissions from Ethanol Consumption (kt) 3-102 Table 3-84: CO2 Emissions from Biodiesel Consumption (MMT CO2 Eq.) 3-103 Table 3-85: CO2 Emissions from Biodiesel Consumption (kt) 3-103 Table 3-86: Woody Biomass Consumption by Sector (Trillion Btu) 3-103 Table 3-87: Ethanol Consumption by Sector (TrillionBtu) 3-104 Table 3-88: Biodiesel Consumption by Sector (TrillionBtu) 3-104 Table 4-1: Emissions from Industrial Processes and Product Use (MMT CO2 Eq.) 4-3 Table 4-2: Emissions from Industrial Processes and Product Use (kt) 4-4 Table 4-3: CO2 Emissions from Cement Production (MMT CO2 Eq. and kt) 4-9 Table 4-4: Clinker Production (kt) 4-10 Table 4-5: Approach 2 Quantitative Uncertainty Estimates for CO2 Emissions from Cement Production (MMT CO2 Eq. and Percent) - TO BE UPDATED FOR FINAL INVENTORY REPORT 4-11 Table 4-6: CO2 Emissions from Lime Production (MMT CO2 Eq. and kt) 4-12 Table 4-7: Potential, Recovered, and Net CO2 Emissions from Lime Production (kt) 4-13 Table 4-8: High-Calcium- and Dolomitic-Quicklime, High-Calcium- and Dolomitic-Hydrated, and Dead-Burned- Dolomite Lime Production (kt) 4-14 Table 4-9: Adjusted Lime Production (kt) 4-14 xii DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2016 ------- 1 Table 4-10: Approach 2 Quantitative Uncertainty Estimates for CO2 Emissions from Lime Production (MMT CO2 2 Eq. and Percent) 4-15 3 Table 4-11: CO2 Emissions from Glass Production (MMT CO2 Eq. and kt) 4-17 4 Table 4-12: Limestone, Dolomite, and Soda Ash Consumption Used in Glass Production (kt) 4-19 5 Table 4-13: Approach 2 Quantitative Uncertainty Estimates for CO2 Emissions from Glass Production (MMT CO2 6 Eq. and Percent) 4-19 7 Table 4-14: CO2 Emissions from Other Process Uses of Carbonates (MMT CO2 Eq.) 4-21 8 Table 4-15: CO2 Emissions from Other Process Uses of Carbonates (kt) 4-21 9 Table 4-16: Limestone and Dolomite Consumption (kt) 4-22 10 Table 4-17: Approach 2 Quantitative Uncertainty Estimates for CO2 Emissions from Other Process Uses of 11 Carbonates (MMT CO2 Eq. and Percent) 4-23 12 Table 4-18: CO2 Emissions from Ammonia Production (MMT CO2 Eq.) 4-25 13 Table 4-19: CO2 Emissions from Ammonia Production (kt) 4-25 14 Table 4-20: Ammonia Production and Urea Production (kt) 4-26 15 Table 4-21: Approach 2 Quantitative Uncertainty Estimates for CO2 Emissions from Ammonia Production (MMT 16 C02 Eq. and Percent) - TO BE UPDATED FOR FINAL INVENTORY REPORT 4-27 17 Table 4-22: CO2 Emissions from Urea Consumption for Non-Agricultural Purposes (MMT CO2 Eq.) 4-28 18 Table 4-23: CO2 Emissions from Urea Consumption for Non-Agricultural Purposes (kt) 4-29 19 Table 4-24: Urea Production, Urea Applied as Fertilizer, Urea Imports, and Urea Exports (kt) 4-30 20 Table 4-25: Approach 2 Quantitative Uncertainty Estimates for CO2 Emissions from Urea Consumption for Non- 21 Agricultural Purposes (MMT C02 Eq. and Percent) - TO BE UPDATED FOR FINAL INVENTORY REPORT 22 4-30 23 Table 4-26: N2O Emissions from Nitric Acid Production (MMT CO2 Eq. and kt N2O) 4-31 24 Table 4-27: Nitric Acid Production (kt) 4-33 25 Table 4-28: Approach 2 Quantitative Uncertainty Estimates for N2O Emissions from Nitric Acid Production (MMT 26 C02 Eq. and Percent) - TO BE UPDATED FOR FINAL INVENTORY REPORT 4-34 27 Table 4-29: N2O Emissions from Adipic Acid Production (MMT CO2 Eq. and kt N2O) 4-35 28 Table 4-30: Adipic Acid Production (kt) 4-37 29 Table 4-31: Approach 2 Quantitative Uncertainty Estimates for N20 Emissions from Adipic Acid Production 30 (MMT C02 Eq. and Percent) - TO BE UPDATED FOR FINAL INVENTORY REPORT 4-37 31 Table 4-32: N20 Emissions from Caprolactam Production (MMT CO2 Eq. and kt N20) 4-39 32 Table 4-33: Caprolactam Production (kt) 4-40 33 Table 4-34: Approach 2 Quantitative Uncertainty Estimates for N20 Emissions from Caprolactam, Glyoxal and 34 Glyoxylic Acid Production (MMT C02 Eq. and Percent) - TO BE UPDATED FOR FINAL INVENTORY 35 REPORT 4-40 36 Table 4-35: CO2 and CH4 Emissions from Silicon Carbide Production and Consumption (MMT CO2 Eq.) 4-42 37 Table 4-36: CO2 and CH4 Emissions from Silicon Carbide Production and Consumption (kt) 4-42 38 Table 4-37: Production and Consumption of Silicon Carbide (Metric Tons) 4-43 39 Table 4-38: Approach 2 Quantitative Uncertainty Estimates for CH4 and CO2 Emissions from Silicon Carbide 40 Production and Consumption (MMT CO2 Eq. and Percent) - TO BE UPDATED FOR FINAL INVENTORY 41 REPORT 4-44 xiii ------- 1 Table 4-39: CO2 Emissions from Titanium Dioxide (MMT CO2 Eq. and kt) 4-45 2 Table 4-40: Titanium Dioxide Production (kt) 4-46 3 Table 4-41: Approach 2 Quantitative Uncertainty Estimates for CO2 Emissions from Titanium Dioxide Production 4 (MMT C02 Eq. and Percent) - TO BE UPDATED FOR FINAL INVENTORY REPORT 4-46 5 Table 4-42: CO2 Emissions from Soda Ash Production (MMT CO2 Eq. and kt CO2) 4-48 6 Table 4-43: Soda Ash Production (kt) 4-49 7 Table 4-44: Approach 2 Quantitative Uncertainty Estimates for CO2 Emissions from Soda Ash Production (MMT 8 C02 Eq. and Percent) - TO BE UPDATED FOR FINAL INVENTORY REPORT 4-49 9 Table 4-45: CO2 and CH4 Emissions from Petrochemical Production (MMT CO2 Eq.) 4-52 10 Table 4-46: CO2 and CH4 Emissions from Petrochemical Production (kt) 4-52 11 Table 4-47: Production of Selected Petrochemicals (kt) 4-54 12 Table 4-48: Approach 2 Quantitative Uncertainty Estimates for CH4 Emissions from Petrochemical Production and 13 CO2 Emissions from Carbon Black Production (MMT CO2 Eq. and Percent) - TO BE UPDATED FOR FINAL 14 INVENTORY REPORT 4-55 15 Table 4-49: HFC-23 Emissions from HCFC-22 Production (MMT C02 Eq. and kt HFC-23) 4-57 16 Table 4-50: HCFC-22 Production (kt) 4-58 17 Table 4-51: Approach 2 Quantitative Uncertainty Estimates for HFC-23 Emissions from HCFC-22 Production 18 (MMT CO2 Eq. and Percent) 4-58 19 Table 4-52: CO2 Emissions from CO2 Consumption (MMT CO2 Eq. and kt) 4-59 20 Table 4-53: CO2 Production (kt CO2) and the Percent Used for Non-EOR Applications 4-61 21 Table 4-54: Approach 2 Quantitative Uncertainty Estimates for CO2 Emissions from CO2 Consumption (MMT CO2 22 Eq. and Percent) - TO BE UPDATED FOR FINAL INVENTORY REPORT 4-62 23 Table 4-55: CO2 Emissions from Phosphoric Acid Production (MMT CO2 Eq. and kt) 4-63 24 Table 4-56: Phosphate Rock Domestic Consumption, Exports, and Imports (kt) 4-64 25 Table 4-57: Chemical Composition of Phosphate Rock (Percent by Weight) 4-64 26 Table 4-58: Approach 2 Quantitative Uncertainty Estimates for CO2 Emissions from Phosphoric Acid Production 27 (MMT C02 Eq. and Percent) - TO BE UPDATED FOR FINAL INVENTORY REPORT 4-65 28 Table 4-59: CO2 Emissions from Metallurgical Coke Production (MMT CO2 Eq.) 4-67 29 Table 4-60: CO2 Emissions from Metallurgical Coke Production (kt) 4-67 30 Table 4-61: CO2 Emissions from Iron and Steel Production (MMT CO2 Eq.) 4-67 31 Table 4-62: CO2 Emissions from Iron and Steel Production (kt) 4-68 32 Table 4-63: CH4 Emissions from Iron and Steel Production (MMT CO2 Eq.) 4-68 33 Table 4-64: CH4 Emissions from Iron and Steel Production (kt) 4-68 34 Table 4-65: Material Carbon Contents for Metallurgical Coke Production 4-69 35 Table 4-66: Production and Consumption Data for the Calculation of CO2 Emissions from Metallurgical Coke 36 Production (Thousand Metric Tons) 4-70 37 Table 4-67: Production and Consumption Data for the Calculation of CO2 Emissions from Metallurgical Coke 38 Production (Million ft3) 4-70 39 Table 4-68: Material Carbon Contents for Iron and Steel Production 4-71 40 Table 4-69: CH4 Emission Factors for Sinter and Pig Iron Production 4-71 xiv DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2016 ------- 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 Table 4-70: CO2 Emission Factors for Sinter Production, Direct Reduced Iron Production and Pellet Production 4-72 Table 4-71: Production and Consumption Data for the Calculation of CO2 and CH4 Emissions from Iron and Steel Production (Thousand Metric Tons) 4-73 Table 4-72: Production and Consumption Data for the Calculation of CO2 Emissions from Iron and Steel Production (Million ft3 unless otherwise specified) 4-73 Table 4-73: Approach 2 Quantitative Uncertainty Estimates for CO2 and CH4 Emissions from Iron and Steel Production and Metallurgical Coke Production (MMT CO2 Eq. and Percent) - TO BE UPDATED FOR FINAL INVENTORY REPORT 4-74 Table 4-74: CO2 and CH4 Emissions from Ferroalloy Production (MMT CO2 Eq.) 4-76 Table 4-75: CO2 and CH4 Emissions from Ferroalloy Production (kt) 4-76 Table 4-76: Production of Ferroalloys (Metric Tons) 4-78 Table 4-77: Approach 2 Quantitative Uncertainty Estimates for CO2 Emissions from Ferroalloy Production (MMT C02 Eq. and Percent) - TO BE UPDATED FOR FINAL INVENTORY REPORT 4-79 Table 4-78: CO2 Emissions from Aluminum Production (MMT CO2 Eq. and kt) 4-80 Table 4-79: PFC Emissions from Aluminum Production (MMT CO2 Eq.) 4-80 Table 4-80: PFC Emissions from Aluminum Production (kt) 4-81 Table 4-81: Production of Primary Aluminum (kt) 4-83 Table 4-82: Approach 2 Quantitative Uncertainty Estimates for CO2 and PFC Emissions from Aluminum Production (MMT CO2 Eq. and Percent) 4-84 Table 4-83: SF6, HFC-134a, FK 5-1-12 and CO2 Emissions from Magnesium Production and Processing (MMT C02 Eq.) 4-85 Table 4-84: SF6, HFC-134a, FK 5-1-12 and CO2 Emissions from Magnesium Production and Processing (kt)... 4-85 Table 4-85: SF6 Emission Factors (kg SF6 per metric ton of magnesium) 4-87 Table 4-86: Approach 2 Quantitative Uncertainty Estimates for SF6, HFC-134a and CO2 Emissions from Magnesium Production and Processing (MMT CO2 Eq. and Percent) 4-88 Table 4-87: CO2 Emissions from Lead Production (MMT CO2 Eq. and kt) 4-90 Table 4-88: Lead Production (Metric Tons) 4-91 Table 4-89: Approach 2 Quantitative Uncertainty Estimates for CO2 Emissions from Lead Production (MMT CO2 Eq. and Percent) - TO BE UPDATED FOR FINAL INVENTORY REPORT 4-91 Table 4-90: Zinc Production (Metric Tons) 4-93 Table 4-91: CO2 Emissions from Zinc Production (MMT CO2 Eq. and kt) 4-94 Table 4-92: Approach 2 Quantitative Uncertainty Estimates for CO2 Emissions from Zinc Production (MMT CO2 Eq. and Percent) - TO BE UPDATED FOR FINAL INVENTORY REPORT 4-96 Table 4-93: PFC, HFC, SF6, NF3, and N20 Emissions from Semiconductor Manufacture (MMT CO2 Eq.) 4-98 Table 4-94: PFC, HFC, SF6, NF3, and N20 Emissions from Semiconductor Manufacture (kt) 4-99 Table 4-95: F-HTF Emissions Based on GHGRP Reporting (MMT CO2 Eq.) 4-99 Table 4-96: F-HTF Compounds with Largest Emissions Based on GHGRP Reporting (tons of gas) 4-99 Table-4-97: Approach 2 Quantitative Uncertainty Estimates for HFC, PFC, SF6, NF3 and N20 Emissions from Semiconductor Manufacture (MMT CO2 Eq. and Percent) 4-108 Table 4-98: Emissions of HFCs and PFCs from ODS Substitutes (MMT CO2 Eq.) 4-109 Table 4-99: Emissions of HFCs and PFCs from ODS Substitution (Metric Tons) 4-110 xv ------- 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 Table 4-100: Emissions of HFCs and PFCs from ODS Substitutes (MMT CO2 Eq.) by Sector 4-110 Table 4-101: Approach 2 Quantitative Uncertainty Estimates for HFC and PFC Emissions from ODS Substitutes (MMT CO2 Eq. and Percent) 4-113 Table 4-102: U.S. HFC Consumption (MMT CO Eq.) 4-114 Table 4-103: Averaged U.S. HFC Demand (MMT CO Eq.) 4-116 Table 4-104: SF6 Emissions from Electric Power Systems and Electrical Equipment Manufacturers (MMT CO2 Eq.) 4-118 Table 4-105: SF6 Emissions from Electric Power Systems and Electrical Equipment Manufacturers (kt) 4-118 Table 4-106: Transmission Mile Coverage (kg) and Regression Coefficients (Percent) 4-121 Table 4-107: Approach 2 Quantitative Uncertainty Estimates for SF6 Emissions from Electrical Transmission and Distribution (MMT CO2 Eq. and Percent) 4-123 Table 4-108: N2O Production (kt) 4-125 Table 4-109: N20 Emissions from N20 Product Usage (MMT CO2 Eq. and kt) 4-125 Table 4-110: Approach 2 Quantitative Uncertainty Estimates for N20 Emissions from N20 Product Usage (MMT CO2 Eq. and Percent) 4-127 Table 4-111: NOx, CO, and NMVOC Emissions from Industrial Processes and Product Use (kt) 4-128 Table 5-1: Emissions from Agriculture (MMT CO2 Eq.) 5-2 Table 5-2: Emissions from Agriculture (kt) 5-2 Table 5-3: CH4 Emissions from Enteric Fermentation (MMT CO2 Eq.) 5-4 Table 5-4: CH4 Emissions from Enteric Fermentation (kt) 5-4 Table 5-5: Cattle Sub-Population Categories for 2016 Population Estimates 5-6 Table 5-6: Approach 2 Quantitative Uncertainty Estimates for CH4 Emissions from Enteric Fermentation (MMT CO2 Eq. and Percent) 5-8 Table 5-7: CH4 and N2O Emissions from Manure Management (MMT CO2 Eq.) 5-11 Table 5-8: CH4 and N20 Emissions from Manure Management (kt) 5-11 Table 5-9: Approach 2 Quantitative Uncertainty Estimates for CH4 and N20 (Direct and Indirect) Emissions from Manure Management (MMT CO2 Eq. and Percent) 5-15 Table 5-10: IPCC (2006) Implied Emission Factor Default Values Compared with Calculated Values for CH4 from Manure Management (kg/head/year) 5-16 Table 5-11: CH4 Emissions from Rice Cultivation (MMT CO2 Eq.) 5-18 Table 5-12: CH4 Emissions from Rice Cultivation (kt) 5-18 Table 5-13: Rice Area Harvested (1,000 Hectares) 5-20 Table 5-14: Average Ratooned Area as Percent of Primary Growth Area (Percent) 5-20 Table 5-15: Approach 2 Quantitative Uncertainty Estimates for CH4 Emissions from Rice Cultivation (MMT CO2 Eq. and Percent) 5-22 Table 5-16: N20 Emissions from Agricultural Soils (MMT CO2 Eq.) 5-25 Table 5-17: N20 Emissions from Agricultural Soils (kt) 5-25 Table 5-18: Direct N20 Emissions from Agricultural Soils by Land Use Type and N Input Type (MMT CO2 Eq.).... 5-25 Table 5-19: Indirect N20 Emissions from Agricultural Soils (MMT CO2 Eq.) 5-26 xvi DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2016 ------- 1 Table 5-20: Quantitative Uncertainty Estimates of N20 Emissions from Agricultural Soil Management in 2016 2 (MMT CO2 Eq. and Percent) 5-40 3 Table 5-21: Emissions from Liming (MMT CO2 Eq.) 5-42 4 Table 5-22: Emissions from Liming (MMT C) 5-42 5 Table 5-23: Applied Minerals (MMT) 5-44 6 Table 5-24: Approach 2 Quantitative Uncertainty Estimates for CO2 Emissions from Liming (MMT CO2 Eq. and 7 Percent) 5-44 8 Table 5-25: CO2 Emissions from Urea Fertilization (MMT CO2 Eq.) 5-45 9 Table 5-26: CO2 Emissions from Urea Fertilization (MMT C) 5-45 10 Table 5-27: Applied Urea (MMT) 5-46 11 Table 5-28: Quantitative Uncertainty Estimates for CO2 Emissions from Urea Fertilization (MMT CO2 Eq. and 12 Percent) 5-46 13 Table 5-29: CH4 and N20 Emissions from Field Burning of Agricultural Residues (MMT CO2 Eq.) 5-47 14 Table 5-30: CH4, N20, CO, and NOx Emissions from Field Burning of Agricultural Residues (kt) 5-48 15 Table 5-31: Agricultural Crop Production (kt of Product) 5-50 16 Table 5-32: U.S. Average Percent Crop Area Burned by Crop (Percent) 5-50 17 Table 5-33: Key Assumptions for Estimating Emissions from Field Burning of Agricultural Residues 5-51 18 Table 5-34: Greenhouse Gas Emission Ratios and Conversion Factors 5-51 19 Table 5-35: Approach 2 Quantitative Uncertainty Estimates for CH4 and N20 Emissions from Field Burning of 20 Agricultural Residues (MMT CO2 Eq. and Percent) 5-51 21 Table 6-1: Net CO2 Flux from Land Use, Land-Use Change, and Forestry (MMT CO2 Eq.) 6-2 22 Table 6-2: Emissions from Land Use, Land-Use Change, and Forestry by Gas (MMT CO2 Eq.) 6-3 23 Table 6-3: Emissions and Removals (Net Flux) from Land Use, Land-Use Change, and Forestry (MMT CO2 Eq.).... 24 6-4 25 Table 6-4: Emissions and Removals from Land Use, Land-Use Change, and Forestry (MMT CO2 Eq.) 6-5 26 Table 6-5: Emissions and Removals from Land Use, Land-Use Change, and Forestry (kt) 6-6 27 Table 6-6: Managed and Unmanaged Land Area by Land-Use Categories for All 50 States (Thousands of Hectares) 28 6-9 29 Table 6-7: Land Use and Land-Use Change for the U.S. Managed Land Base for All 50 States (Thousands of 30 Hectares) 6-10 31 Table 6-8: Data Sources Used to Determine Land Use and Land Area for the Conterminous United States, Hawaii, 32 and Alaska 6-16 33 Table 6-9: Total Land Area (Hectares) by Land-Use Category for U.S. Territories 6-22 34 Table 6-10: Net CO2 Flux from Forest Pools in Forest Land Remaining Forest Land and Harvested Wood Pools 35 (MMT C02 Eq.) 6-26 36 Table 6-11: Net C Flux from Forest Pools in Forest Land Remaining Forest Land and Harvested Wood Pools 37 (MMT C) 6-26 38 Table 6-12: Forest Area (1,000 ha) and C Stocks in Forest Land Remaining Forest Land and Harvested Wood 39 Pools (MMT C) 6-27 40 Table 6-13: Estimates of CO2 (MMT per Year) Emissions from Forest Fires in the Conterminous 48 States and 41 Alaska3 6-29 xvii ------- 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 Table 6-14: Quantitative Uncertainty Estimates for Net CO2 Flux from Forest Land Remaining Forest Land: Changes in Forest C Stocks (MMT CO2 Eq. and Percent) 6-32 Table 6-15: Mean C Stocks, CO2 and CH4 Fluxes in Alaska between 2000 and 2009 6-35 Table 6-16: Non-CCh Emissions from Forest Fires (MMT CO2 Eq.)a 6-35 Table 6-17: Non-C02 Emissions from Forest Fires (kt)a 6-36 Table 6-18: Quantitative Uncertainty Estimates of Non-CCh Emissions from Forest Fires (MMT CO2 Eq. and Percent)3 6-36 Table 6-19: N2O Fluxes from Soils in Forest Land Remaining Forest Land and Land Converted to Forest Land (MMT C02 Eq. and kt N O) 6-37 Table 6-20: 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) 6-39 Table 6-21: Estimated CO2 and Non-CCh Emissions on Drained Organic Forest Soils3 (MMT CO2 Eq.) 6-40 Table 6-22: Estimated C (MMT C) and Non-C02 (kt) Emissions on Drained Organic Forest Soils3 6-40 Table 6-23: States identified as having Drained Organic Soils, Area of Forest on Drained Organic Soils, and Sampling Error 6-41 Table 6-24: Quantitative Uncertainty Estimates for Annual CO2 and Non-C02 Emissions on Drained Organic Forest Soils (MMT CO2 Eq. and Percent)3 6-42 Table 6-25: Net CO2 Flux from Forest C Pools in Land Converted to Forest Land by Land Use Change Category (MMT C02 Eq.) 6-43 Table 6-26: Net C Flux from Forest C Pools in Land Converted to Forest Land by Land Use Change Category (MMT C) 6-44 Table 6-27: Quantitative Uncertainty Estimates for Forest C Pool Stock Changes (MMT CO2 Eq. per Year) in 2016 fxomLand Converted to Forest Landby Land Use Change 6-46 Table 6-28: Net CO2 Flux from Soil C Stock Changes in Cropland Remaining Cropland (MMT CO2 Eq.) 6-49 Table 6-29: Net CO2 Flux from Soil C Stock Changes in Cropland Remaining Cropland (MMT C) 6-49 Table 6-30: Approach 2 Quantitative Uncertainty Estimates for Soil C Stock Changes occurring within Cropland Remaining Cropland (MMT CO2 Eq. and Percent) 6-55 Table 6-31: Net CO2 Flux from Soil, Dead Organic Matter and Biomass C Stock Changes in Land Converted to Cropland by Land Use Change Category (MMT CO2 Eq.) 6-58 Table 6-32: Net CO2 Flux from Soil, Dead Organic Matter and Biomass C Stock Changes in Land Converted to Cropland (MMT C) 6-58 Table 6-33: Approach 2 Quantitative Uncertainty Estimates for Soil, Dead Organic Matter and Biomass C Stock Changes occurring within Land Converted to Cropland (MMT CO2 Eq. and Percent) 6-62 Table 6-34: Net CO2 Flux from Soil C Stock Changes in Grassland Remaining Grassland (MMT CO2 Eq.) 6-64 Table 6-35: Net CO2 Flux from Soil C Stock Changes in Grassland Remaining Grassland (MMT C) 6-64 Table 6-36: Approach 2 Quantitative Uncertainty Estimates for C Stock Changes Occurring Within Grassland Remaining Grassland (MMT CO2 Eq. and Percent) 6-68 Table 6-37: CH4 and N20 Emissions from Biomass Burning in Grassland (MMT CO2 Eq.) 6-69 Table 6-38: CH4, N20, CO, and NOx Emissions from Biomass Burning in Grassland (kt) 6-70 Table 6-39: Thousands of Grassland Hectares Burned Annually 6-70 Table 6-40: Uncertainty Estimates for Non-C02 Greenhouse Gas Emissions from Biomass Burning in Grassland (MMT CO2 Eq. and Percent) 6-71 xviii DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2016 ------- 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 Table 6-41: Net CO2 Flux from Soil, Dead Organic Matter and Biomass C Stock Changes for Land Converted to Grassland (MMT CO2 Eq.) 6-73 Table 6-42: Net CO2 Flux from Soil, Dead Organic Matter and Biomass C Stock Changes for Land Converted to Grassland (MMT C) 6-74 Table 6-43: Approach 2 Quantitative Uncertainty Estimates for Soil, Dead Organic Matter and Biomass C Stock Changes occurring within Land Converted to Grassland (MMT CO2 Eq. and Percent) 6-77 Table 6-44: Emissions from PeatlandsRemaining Peatlands (MMT CO2 Eq.) 6-80 Table 6-45: Emissions from Peatlands Remaining Peatlands (kt) 6-80 Table 6-46: Peat Production of Lower 48 States (kt) 6-82 Table 6-47: Peat Production of Alaska (Thousand Cubic Meters) 6-82 Table 6-48: Approach 2 Quantitative Uncertainty Estimates for CO2, CH4, and N20 Emissions from Peatlands Remaining Peatlands (MMT CO2 Eq. and Percent) 6-83 Table 6-49: Net CO2 Flux from Soil C Stock Changes in Vegetated Coastal Wetlands Remaining Vegetated Coastal Wetlands (MMT CO2 Eq.) 6-86 Table 6-50: Net CO2 Flux from Soil C Stock Changes in Vegetated Coastal Wetlands Remaining Vegetated Coastal Wetlands (MMT C) 6-86 Table 6-51: Net CH4 Flux from Vegetated Coastal Wetlands Remaining Vegetated Coastal Wetlands (MMT CO2 Eq.) 6-86 Table 6-52: Net CH4 Flux from Vegetated Coastal Wetlands Remaining Vegetated Coastal Wetlands (kt CH4). 6-86 Table 6-53: Approach 1 Quantitative Uncertainty Estimates for Emissions from C Stock Changes occurring within Vegetated Coastal Wetlands Remaining Vegetated Coastal Wetlands (MMT CO2 Eq. and Percent) 6-88 Table 6-54: Approach 1 Quantitative Uncertainty Estimates for CH4 Emissions occurring within Vegetated Coastal Wetlands Remaining Vegetated Coastal Wetlands (MMT CO2 Eq. and Percent) 6-88 Table 6-55: Net CO2 Flux from Soil C Stock Changes in Vegetated Coastal Wetlands Converted to Unvegetated Open Water Coastal Wetlands (MMT CO2 Eq.) 6-89 Table 6-56: Net CO2 Flux from Soil C Stock Changes in Vegetated Coastal Wetlands Converted to Unvegetated Open Water Coastal Wetlands (MMT C) 6-89 Table 6-57: Approach 1 Quantitative Uncertainty Estimates for Net CO2 Flux Occurring within Vegetated Coastal Wetlands Converted to Unvegetated Open Water Coastal Wetlands (MMT CO2 Eq. and Percent) 6-91 Table 6-58: Net CO2 Flux from Soil C Stock Changes from Unvegetated Open Water Coastal Wetlands Converted to Vegetated Coastal Wetlands (MMT CO2 Eq.) 6-92 Table 6-59: Net CO2 Flux from Soil C Stock Changes from Unvegetated Open Water Coastal Wetlands Converted to Vegetated Coastal Wetlands (MMT C) 6-92 Table 6-60: Approach 1 Quantitative Uncertainty Estimates for C Stock Changes Occurring within Unvegetated Open Water Coastal Wetlands Converted to Vegetated Coastal Wetlands (MMT CO2 Eq. and Percent) 6-94 Table 6-61: Net N20 Emissions from Aquaculture in Coastal Wetlands (MMT CO2 Eq.) 6-95 Table 6-62: Net N20 Emissions from Aquaculture in Coastal Wetlands (kt N20) 6-95 Table 6-63: Approach 1 Quantitative Uncertainty Estimates for N20 Emissions for Aquaculture Production in Coastal Wetlands (MMT CO2 Eq. and Percent) 6-96 Table 6-64: Net CO2 Flux from Soil C Stock Changes in Land Converted to Vegetated Coastal Wetlands (MMT C02 Eq.) 6-97 Table 6-65: Net CO2 Flux from Soil C Stock Changes in Land Converted to Vegetated Coastal Wetlands (MMT C) 6-97 xix ------- 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 Table 6-66: Net CH4 Flux in Land Converted to Vegetated Coastal Wetlands (MMT CO2 Eq.) 6-97 Table 6-67: Net CH4 Flux from Soil C Stock Changes in Land Converted to Vegetated Coastal Wetlands (kt CH4)... 6-97 Table 6-68: Approach 1 Quantitative Uncertainty Estimates for Net CO2 Flux Changes occurring within Land Converted to Vegetated Coastal Wetlands (MMT CO2 Eq. and Percent) 6-98 Table 6-69: Approach 1 Quantitative Uncertainty Estimates for CH4 Emissions occurring w ithin Land Converted to Vegetated Coastal Wetlands (MMT CO2 Eq. and Percent) 6-99 Table 6-70: Net CO2 Flux from Soil C Stock Changes in Settlements Remaining Settlements (MMT CO2 Eq.) 6-100 Table 6-71: Net CO2 Flux from Soil C Stock Changes in Settlements Remaining Settlements (MMT C) 6-100 Table 6-72: Thousands of Hectares of Drained Organic Soils in Settlements Remaining Settlements 6-101 Table 6-73: Uncertainty Estimates for CO2 Emissions from Drained Organic Soils in Settlements Remaining Settlements (MMT CO2 Eq. and Percent) 6-101 Table 6-74: Net C Flux from Urban Trees (MMT CO Eq. and MMT C) 6-103 Table 6-75: Annual C Sequestration (Metric Tons C7Year), Tree Cover (Percent), and Annual C Sequestration per Area of Tree Cover (kg C/m2-yr) for 50 states plus the District of Columbia (2016) 6-105 Table 6-76: Approach 2 Quantitative Uncertainty Estimates for Net C Flux from Changes in C Stocks in Urban Trees (MMT CO2 Eq. and Percent) 6-106 Table 6-77: N2O Emissions from Soils in Settlements Remaining Settlements (MMT CO2 Eq. and kt N2O) 6-108 Table 6-78: Quantitative Uncertainty Estimates of N2O Emissions from Soils in Settlements Remaining Settlements (MMT CO2 Eq. and Percent) 6-110 Table 6-79: Net Changes in Yard Trimmings and Food Scrap Carbon Stocks in Landfills (MMT CO2 Eq.) 6-112 Table 6-80: Net Changes in Yard Trimmings and Food Scrap Carbon Stocks in Landfills (MMT C) 6-112 Table 6-81: Moisture Contents, C Storage Factors (Proportions of Initial C Sequestered), Initial C Contents, and Decay Rates for Yard Trimmings and Food Scraps in Landfills 6-114 Table 6-82: C Stocks in Yard Trimmings and Food Scraps in Landfills (MMT C) 6-115 Table 6-83: Approach 2 Quantitative Uncertainty Estimates for CO2 Flux from Yard Trimmings and Food Scraps in Landfills (MMT CO2 Eq. and Percent) 6-115 Table 6-84: Net CO2 Flux from Soil, Dead Organic Matter and Biomass C Stock Changes for Land Converted to Settlements (MMT CO2 Eq.) 6-117 Table 6-85: Net CO2 Flux from Soil, Dead Organic Matter and Biomass C Stock Changes for Land Converted to Settlements (MMT C) 6-117 Table 6-86: Approach 2 Quantitative Uncertainty Estimates for Soil, Dead Organic Matter and Biomass C Stock Changes occurring within Land Converted to Settlements (MMT CO2 Eq. and Percent) 6-120 Table 7-1: Emissions from Waste (MMT CO2 Eq.) 7-1 Table 7-2: Emissions from Waste (kt) 7-2 Table 7-3: CH4 Emissions from Landfills (MMT CO2 Eq.) 7-4 Table 7-4: CH4 Emissions from Landfills (kt) 7-5 Table 7-5: Approach 2 Quantitative Uncertainty Estimates for CH4 Emissions from Landfills (MMT CO2 Eq. and Percent) 7-13 Table 7-6: Materials Discarded3 in the Municipal Waste Stream by Waste Type from 1990 to 2014 (Percent)b .. 7-17 Table 7-7: CH4 and N20 Emissions from Domestic and Industrial Wastewater Treatment (MMT CO2 Eq.) 7-20 xx DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2016 ------- 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 Table 7-8: CH4 and N20 Emissions from Domestic and Industrial Wastewater Treatment (kt) 7-20 Table 7-9: U.S. Population (Millions) and Domestic Wastewater BOD5 Produced (kt) 7-22 Table 7-10: Domestic Wastewater CH4 Emissions from Septic and Centralized Systems (2016, MMT CO2 Eq. and Percent) 7-23 Table 7-11: Industrial Wastewater CH4 Emissions by Sector (2016, MMT CO2 Eq. and Percent) 7-24 Table 7-12: U.S. Pulp and Paper, Meat, Poultry, Vegetables, Fruits and Juices, Ethanol, and Petroleum Refining Production (MMT) 7-24 Table 7-13: Variables Used to Calculate Percent Wastewater Treated Anaerobically by Industry (Percent) 7-25 Table 7-14: Wastewater Flow (m3/ton) and BOD Production (g/L) for U.S. Vegetables, Fruits, and Juices Production 7-27 Table 7-15: U.S. Population (Millions), Population Served by Biological Denitrification (Millions), Fraction of Population Served by Wastewater Treatment (percent), Available Protein (kg/person-year), Protein Consumed (kg/person-year), and Nitrogen Removed with Sludge (kt-N/year) 7-30 Table 7-16: Approach 2 Quantitative Uncertainty Estimates for CH4 Emissions from Wastewater Treatment (MMT CO2 Eq. and Percent) 7-31 Table 7-17: CH4 and N20 Emissions from Composting (MMT CO2 Eq.) 7-34 Table 7-18: CH4 and N20 Emissions from Composting (kt) 7-34 Table 7-19: U.S. Waste Composted (kt) 7-34 Table 7-20: Approach 1 Quantitative Uncertainty Estimates for Emissions from Composting (MMT CO2 Eq. and Percent) 7-35 Table 7-21: Emissions of NOx, CO, and NMVOC from Waste (kt) 7-36 Table 9-1: Revisions to U.S. Greenhouse Gas Emissions (MMT CO2 Eq.) 9-4 Table 9-2: Revisions to U.S. Greenhouse Gas Emissions and Removals (Net Flux) from Land Use, Land-Use Change, and Forestry (MMT CO2 Eq.) 9-6 Figure ES-1: Gross U.S. Greenhouse Gas Emissions by Gas (MMT CO2 Eq.) ES-4 Figure ES-2: Annual Percent Change in Gross U.S. Greenhouse Gas Emissions Relative to the Previous Year ..ES-5 Figure ES-3: Cumulative Change in Annual Gross U.S. Greenhouse Gas Emissions Relative to 1990 (1990=0, MM I CO- Eq.) ES-5 Figure ES-4: 2016 U.S. Greenhouse Gas Emissions by Gas (Percentages based on MMT CO2 Eq.) ES-9 Figure ES-5: 2016 Sources of CO2 Emissions (MMT CO2 Eq.) ES-10 Figure ES-6: 2016 CO2 Emissions from Fossil Fuel Combustion by Sector and Fuel Type (MMT C02Eq.) ES-11 Figure ES-7: 2016 End-Use Sector Emissions of CO2 from Fossil Fuel Combustion (MMT CO2 Eq.) ES-11 Figure ES-8: Electric Power Generation (Billion kWh) and Emissions (MMT CO2 Eq.) ES-13 Figure ES-9: 2016 Sources of CH4 Emissions (MMT CO2 Eq.) ES-15 Figure ES-10: 2016 Sources of N20 Emissions (MMT CO2 Eq.) ES-16 Figure ES-11: 2016 Sources of HFCs, PFCs, SF6, and NF3 Emissions (MMT CO2 Eq.) ES-17 Figure ES-12: U.S. Greenhouse Gas Emissions and Sinks by Chapter/IPCC Sector (MMT CO2 Eq.) ES-18 Figure ES-13: 2016 U.S. Energy Consumption by Energy Source (Percent) ES-20 xxi ------- 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 Figure ES-14: U.S. Greenhouse Gas Emissions Allocated to Economic Sectors (MMT CO2 Eq.) ES-24 Figure ES-15: U.S. Greenhouse Gas Emissions with Electricity-Related Emissions Distributed to Economic Sectors (MMTCO2 Eq.) ES-26 Figure ES-16: U.S. Greenhouse Gas Emissions Per Capita and Per Dollar of Gross Domestic Product (GDP) ES-27 Figure ES-17: 2016 Key Categories (MMT CO2 Eq.) ES-28 Figure 1-1: National Inventory Arrangements Diagram Inventory Process 1-12 Figure 1-2: U.S. QA/QC Plan Summary 1-21 Figure 2-1: Gross U.S. Greenhouse Gas Emissions by Gas (MMT CO2 Eq.) 2-1 Figure 2-2: Annual Percent Change in Gross U.S. Greenhouse Gas Emissions Relative to the Previous Year 2-2 Figure 2-3: Cumulative Change in Annual Gross U.S. Greenhouse Gas Emissions Relative to 1990 (1990=0, MMT C02 Eq.) 2-2 Figure 2-4: U.S. Greenhouse Gas Emissions and Sinks by Chapter/IPCC Sector (MMT CO2 Eq.) 2-7 Figure 2-5: 2016 Energy Chapter Greenhouse Gas Sources (MMT CO2 Eq.) 2-9 Figure 2-6: 2016 U.S. Fossil Carbon Flows (MMT CO2 Eq.) 2-10 Figure 2-7: 2016 CO2 Emissions from Fossil Fuel Combustion by Sector and Fuel Type (MMT CO2 Eq.) 2-13 Figure 2-8: 2016 End-Use Sector Emissions of CO2 from Fossil Fuel Combustion (MMT CO2 Eq.) 2-13 Figure 2-9: Electric Power Generation (Billion kWh) and Emissions (MMT CO2 Eq.) 2-14 Figure 2-10: 2016 Industrial Processes and Product Use Chapter Greenhouse Gas Sources (MMT CO2 Eq.) 2-15 Figure 2-11: 2016 Agriculture Chapter Greenhouse Gas Sources (MMT CO2 Eq.) 2-18 Figure 2-12: 2016 LULUCF Chapter Greenhouse Gas Sources and Sinks (MMT CO2 Eq.) 2-20 Figure 2-13: 2016 Waste Chapter Greenhouse Gas Sources (MMT CO2 Eq.) 2-22 Figure 2-14: U.S. Greenhouse Gas Emissions Allocated to Economic Sectors (MMT CO2 Eq.) 2-24 Figure 2-15: U.S. Greenhouse Gas Emissions with Electricity-Related Emissions Distributed to Economic Sectors (MMT C02 Eq.) 2-27 Figure 2-16: U.S. Greenhouse Gas Emissions Per Capita and Per Dollar of Gross Domestic Product 2-34 Figure 3-1: 2016 Energy Chapter Greenhouse Gas Sources (MMT CO2 Eq.) 3-1 Figure 3-2: 2016 U.S. Fossil Carbon Flows (MMT CO2 Eq.) 3-2 Figure 3-3: 2016 U.S. Energy Consumption by Energy Source (Percent) 3-8 Figure 3-4: U.S. Energy Consumption (Quadrillion Btu) 3-8 Figure 3-5: 2016 CO2 Emissions from Fossil Fuel Combustion by Sector and Fuel Type (MMT CO2 Eq.) 3-9 Figure 3-6: Annual Deviations from Normal Heating Degree Days for the United States (1950-2016, Index Normal = 100) 3-10 Figure 3-7: Annual Deviations from Normal Cooling Degree Days for the United States (1950-2016, Index Normal = 100) 3-10 Figure 3-8: Fuels Used in Electric Power Generation (TBtu) and Total Electric Power Sector CO2 Emissions.... 3-16 Figure 3-9: Electric Power Retail Sales by End-Use Sector (Billion kWh) 3-16 Figure 3-10: Industrial Production Indices (Index 2012=100) 3-18 xxii DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2016 ------- 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 Figure 3-11: Fuels Used in Residential and Commercial Sectors (TBtu), Heating Degree Days, and Total Sector CO2 Emissions 3-19 Figure 3-12: Fuels Used in Transportation Sector (TBtu), Onroad VMT, and Total Sector CO2 Emissions 3-21 Figure 3-13: Sales-Weighted Fuel Economy of New Passenger Cars and Light-Duty Trucks, 1990-2016 (miles/gallon) 3-23 Figure 3-14: Sales of New Passenger Cars and Light-Duty Trucks, 1990-2016 (Percent) 3-23 Figure 3-15: Mobile Source CH4 and N20 Emissions (MMT CO2 Eq.) 3-26 Figure 3-16: U.S. Energy Consumption and Energy-Related CO2 Emissions Per Capita and Per Dollar GDP 3-33 Figure 4-1: 2016 Industrial Processes and Product Use Chapter Greenhouse Gas Sources (MMT CO2 Eq.) 4-2 Figure 4-2: U.S. HFC Consumption (MMT CO2 Eq.) 4-115 Figure 5-1: 2016 Agriculture Chapter Greenhouse Gas Emission Sources (MMT CO2 Eq.) 5-1 Figure 5-2: Annual CH4 Emissions from Rice Cultivation, 2012 (MMT CO2 Eq./Year)* 5-19 Figure 5-3: Sources and Pathways of N that Result in N20 Emissions from Agricultural Soil Management 5-24 Figure 5-4: Crops, 2012 Annual Direct N2O Emissions Estimated Using the Tier 3 DAYCENT Model (MMT CO2 Eq./year)* 5-27 Figure 5-5: Grasslands, 2012 Annual Direct N2O Emissions Estimated Using the Tier 3 DAYCENT Model (MMT CO2 Eq./year)* 5-28 Figure 5-6: Crops, 2012 Annual Indirect N2O Emissions from Volatilization Using the Tier 3 DAYCENT Model (MMT C02 Eq./year)* 5-29 Figure 5-7: Grasslands, 2012 Annual Indirect N2O Emissions from Volatilization Using the Tier 3 DAYCENT Model (MMT CO Eq./year)* 5-30 Figure 5-8: Crops, 2012 Annual Indirect N2O Emissions from Leaching and Runoff Using the Tier 3 DAYCENT Model (MMT CO Eq./year)* 5-31 Figure 5-9: Grasslands, 2012 Annual Indirect N20 Emissions from Leaching and Runoff Using the Tier 3 DAYCENT Model (MMT CO Eq./year)* 5-32 Figure 5-10: Comparison of Measured Emissions at Field Sites and Modeled Emissions Using the DAYCENT Simulation Model and IPCC Tier 1 Approach (kg N20 per ha per year) 5-41 Figure 6-1: 2016 LULUCF Chapter Greenhouse Gas Sources and Sinks (MMT CO2 Eq.) 6-4 Figure 6-2: Percent of Total Land Area for Each State in the General Land-Use Categories for 2015a 6-12 Figure 6-3: Changes in Forest Area by Region for Forest Land Remaining Forest Land in the conterminous United States and coastal Alaska (1990-2016, Million Hectares) 6-25 Figure 6-4: Estimated Net Annual Changes in C Stocks for All C Pools in Forest Land Remaining Forest Land in the Conterminous U.S. and Coastal Alaska (1990-2016, MMT C per Year) 6-28 Figure 6-5: Total Net Annual Soil C Stock Changes for Mineral Soils under Agricultural Management within States, 2012, Cropland Remaining Cropland * 6-50 Figure 6-6: Total Net Annual Soil C Stock Changes for Organic Soils under Agricultural Management within States, 2012, Cropland Remaining Cropland* 6-51 Figure 6-7: Total Net Annual Soil C Stock Changes for Mineral Soils under Agricultural Management within States, 2012, Grassland Remaining Grassland * 6-65 Figure 6-8: Total Net Annual Soil C Stock Changes for Organic Soils under Agricultural Management within States, 2012, Grassland Remaining Grassland* 6-65 Figure 7-1: 2016 Waste Chapter Greenhouse Gas Sources (MMT CO2 Eq.) 7-1 xxiii ------- 1 Figure 7-2: Management of Municipal Solid Waste in the United States, 2014 7-16 2 Figure 7-3: MSW Management Trends from 1990 to 2014 7-16 3 Figure 7-4: Percent of Degradable Materials Diverted from Landfills from 1990 to 2014 (Percent) 7-18 4 Boxes 5 Box ES-1: Methodological Approach for Estimating and Reporting U.S. Emissions and Removals ES-1 6 BoxES-2: EPA's Greenhouse Gas Reporting Program ES-2 7 Box ES-3: Improvements and Recalculations Relative to the Previous Inventory ES-5 8 Box ES-4: Use of Ambient Measurements Systems for Validation of Emission Inventories ES-14 9 Box ES-5: Recent Trends in Various U.S. Greenhouse Gas Emissions-Related Data ES-26 10 Box ES-6: Recalculations of Inventory Estimates ES-29 11 Box 1-1: Methodological Approach for Estimating and Reporting U.S. Emissions and Removals 1-2 12 Box 1-2: The IPCC Fifth Assessment Report and Global Warming Potentials 1-9 13 Box 1-3: IPCC Reference Approach 1-15 14 Box 2-1: Methodology for Aggregating Emissions by Economic Sector 2-32 15 Box 2-2: Recent Trends in Various U.S. Greenhouse Gas Emissions-Related Data 2-33 16 Box 2-3: Sources and Effects of Sulfur Dioxide 2-36 17 Box 3-1: Methodological Approach for Estimating and Reporting U.S. Emissions and Removals 3-4 18 Box 3-2: Energy Data from EPA's Greenhouse Gas Reporting Program 3-4 19 Box 3-3: Weather and Non-Fossil Energy Effects on CO2 from Fossil Fuel Combustion Trends 3-9 20 Box 3-4: Uses of Greenhouse Gas Reporting Program Data and Improvements in Reporting Emissions from 21 Industrial Sector Fossil Fuel Combustion 3-31 22 Box 3-5: Carbon Intensity of U.S. Energy Consumption 3-32 23 Box 3-6: Reporting of Lubricants, Waxes, and Asphalt and Road Oil Product Use in Energy Sector 3-51 24 Box 3-7: Carbon Dioxide Transport, Injection, and Geological Storage 3-76 25 Box 4-1: Methodological Approach for Estimating and Reporting U.S. Emissions and Removals 4-6 26 Box 4-2: Industrial Processes Data from EPA's Greenhouse Gas Reporting Program 4-7 27 Box 5-1: Methodological Approach for Estimating and Reporting U.S. Emissions and Removals 5-2 28 Box 5-2: Biennial Inventory Compilation 5-3 29 Box 5-3: Surrogate Data Method 5-21 30 Box 5-4: Tier 1 vs. Tier 3 Approach for Estimating N2O Emissions 5-33 31 Box 5-5: Surrogate Data Method 5-34 32 Box 5-6: Comparison of the Tier 2 U.S. Inventory Approach and IPCC (2006) Default Approach 5-43 33 Box 5-7: Comparison of Tier 2 U.S. Inventory Approach and IPCC (2006) Default Approach 5-49 34 Box 6-1: Methodological Approach for Estimating and Reporting U.S. Emissions and Removals 6-7 35 Box 6-2: Biennial Inventory Compilation 6-8 36 Box 6-3: Preliminary Estimates of Land Use in U.S. Territories 6-21 xxiv DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2016 ------- 1 2 3 4 5 6 7 8 9 10 11 12 13 14 Box 6-4: CO2 Emissions from Forest Fires 6-28 Box 6-5: Preliminary Estimates of Historical Carbon Stock Change and Methane Emissions from Managed Land in Alaska (Represents Mean for Years 2000 to 2009) 6-34 Box 6-6: Surrogate Data Method 6-52 Box 6-7: Tier 3 Approach for Soil C Stocks Compared to Tier 1 or 2 Approaches 6-53 Box 6-8: Grassland Woody Biomass Analysis 6-69 Box 7-1: Methodological Approach for Estimating and Reporting U.S. Emissions and Removals 7-2 Box 7-2: Waste Data from EPA's Greenhouse Gas Reporting Program 7-2 Box 7-3: Nationwide Municipal Solid Waste Data Sources 7-15 Box 7-4: Overview of the Waste Sector 7-16 Box 7-5: Description of a Modern, Managed Landfill 7-18 xxv ------- 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 Executive Summary An emissions inventory that identifies and quantifies a country's primary anthropogenic1 sources and sinks of greenhouse gases is essential for addressing climate change. This inventory adheres to both (1) a comprehensive and detailed set of methodologies for estimating sources and sinks of anthropogenic greenhouse gases, and (2) a common and consistent format that enables Parties to the United Nations Framework Convention on Climate Change (UNFCCC) to compare the relative contribution of different emission sources and greenhouse gases to climate change. In 1992, the United States signed and ratified the UNFCCC. As stated in Article 2 of the UNFCCC, "The ultimate objective of this Convention and any related legal instruments that the Conference of the Parties may adopt is to achieve, in accordance with the relevant provisions of the Convention, stabilization of greenhouse gas concentrations in the atmosphere at a level that would prevent dangerous anthropogenic interference with the climate system. Such a level should be achieved within a time-frame sufficient to allow ecosystems to adapt naturally to climate change, to ensure that food production is not threatened and to enable economic development to proceed in a sustainable manner."2 Parties to the Convention, by ratifying, "shall develop, periodically update, publish and make available... national inventories of anthropogenic emissions by sources and removals by sinks of all greenhouse gases not controlled by the Montreal Protocol, using comparable methodologies.. ."3 The United States views this report as an opportunity to fulfill these commitments. This chapter summarizes the latest information on U.S. anthropogenic greenhouse gas emission trends from 1990 through 2016. To ensure that the U.S. emissions inventory is comparable to those of other UNFCCC Parties, the estimates presented here were calculated using methodologies consistent with those recommended in the 2006 Intergovernmental Panel on Climate Change (IPCC) Guidelines for National Greenhouse Gas Inventories (IPCC 2006). The structure of this report is consistent with the UNFCCC guidelines for inventory reporting, as discussed in BoxES-1.4 Box ES-1: Methodological Approach for Estimating and Reporting U.S. Emissions and Removals In following the United Nations Framework Convention on Climate Change (UNFCCC) requirement under Article 4.1 to develop and submit national greenhouse gas emission inventories, 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 1 Hie term "anthropogenic," in this context, refers to greenhouse gas emissions and removals that are a direct result of human activities or are the result of natural processes that have been affected by human activities (IPCC 2006). 2 Article 2 of the Framework Convention on Climate Change published by the UNEPAVMO Information Unit on Climate Change. See . 3 Article 4(1 )(a) of the United Nations Framework Convention on Climate Change (also identified in Article 12). Subsequent decisions by the Conference of the Parties elaborated the role of Annex I Parties in preparing national inventories. See . 4 See . Executive Summary ES-1 ------- 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 for National Greenhouse Gas Inventories (2006IPCC Guidelines). Additionally, the calculated emissions and removals in a given year for the United States are presented in a common manner in line with the UNFCCC reporting guidelines for the reporting of inventories under this international agreement. The use of consistent methods to calculate emissions and removals by all nations providing their inventories to the UNFCCC ensures that these reports are comparable. The presentation of emissions and removals provided in this Inventory does not preclude alternative examinations, but rather this Inventory presents emissions and removals in a common format consistent with how countries are to report Inventories under the UNFCCC. The report itself, and this chapter, follows this standardized format, and provides an explanation of the application of methods used to calculate emissions and removals. Box ES-2: EPA's Greenhouse Gas Reporting Program On October 30, 2009, the U.S. Enviromnental Protection Agency (EPA) published 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). The rule applies to direct greenhouse gas emitters, fossil fuel suppliers, industrial gas suppliers, and facilities that inject carbon dioxide (CO2) underground for sequestration or other reasons.5 Annual reporting is at the facility level, except for certain suppliers of fossil fuels and industrial greenhouse gases. EPA's GHGRP dataset and the data presented in this Inventory report 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. 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 categories to improve the national estimates presented in this Inventory consistent with IPCC guidance.6 ES.l Background Information Greenhouse gases absorb infrared radiation, thereby trapping heat and making the planet warmer. The most important greenhouse gases directly emitted by humans include carbon dioxide (CO2), methane (CH4), nitrous oxide (N2O), and several other fluorine-containing halogenated substances. Although CO2, CH4, and N20 occur naturally in the atmosphere, human activities have changed their atmospheric concentrations. From the pre-industrial era (i.e., ending about 1750) to 2016, concentrations of these greenhouse gases have increased globally by 44, 163, and 22 percent, respectively (IPCC 2013; NOAA/ESRL 2017a, 2017b, 2017c). This annual report estimates the total national greenhouse gas emissions and removals associated with human activities across the United States. Global Warming Potentials Gases in the atmosphere can contribute to climate change both directly and indirectly. Direct effects occur when the gas itself absorbs radiation. Indirect radiative forcing occurs when chemical transformations of the substance produce other greenhouse gases, when a gas influences the atmospheric lifetimes of other gases, and/or when a gas 5 See and . 6 See ES-2 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2016 ------- 1 affects atmospheric processes that alter the radiative balance of the earth (e.g., affect cloud formation or albedo).7 2 The IPCC developed the Global Warming Potential (GWP) concept to compare the ability of each greenhouse gas to 3 trap heat in the atmosphere relative to another gas. 4 The GWP of a greenhouse gas is defined as the ratio of the accumulated radiative forcing within a specific time 5 horizon caused by emitting 1 kilogram of the gas, relative to that of the reference gas CO2 (IPCC 2014). The 6 reference gas used is CO2, and therefore GWP-weighted emissions are measured in million metric tons of CO2 7 equivalent (MMT CO2 Eq.).8 9 All gases in this Executive Summary are presented in units of MMT CO2 Eq. 8 Emissions by gas in unweighted mass kilotons are provided in the Trends chapter of this report. 9 UNFCCC reporting guidelines for national inventories require the use of GWP values from the IPCC Fourth 10 Assessment Report (AR4) (IPCC 2007).10 All estimates are provided throughout the report in both CO2 equivalents 11 and unweighted units. A comparison of emission values using the AR4 GWP values versus the SAR (IPCC 1996), 12 and the IPCC Fifth Assessment Report (AR5) (IPCC 2013) GWP values can be found in Chapter 1 and, in more 13 detail, in Annex 6.1 of this report. The GWP values used in this report are listed below in Table ES-1. 14 Table ES-1: Global Warming Potentials (100-Year Time Horizon) Used in this Report Gas GWP CO2 1 CH4a 25 N2O 298 HFC-23 14,800 HFC-32 675 HFC-125 3,500 HFC-134a 1,430 HFC-143a 4,470 HFC-152a 124 HFC-227ea 3,220 HFC-236fa 9,810 HFC-4310mee 1,640 CF4 7,390 c2f6 12,200 C4F10 8,860 C6Fl4 9,300 SFe 22,800 NF3 17,200 a The CH4 GWP includes the direct effects and those indirect effects due to the production of tropospheric ozone and stratospheric water vapor. The indirect effect due to production of CO2 is not included. Source: IPCC (2007) 15 7 Albedo is a measure of the Earth's reflectivity, and is defined as the fraction of the total solar radiation incident on a body that is reflected by it. 8 Carbon comprises 12/44 of carbon dioxide by weight. 9 One million metric ton is equal to 1012 grams or one teragram. 10 See . Executive Summary ES-3 ------- 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 ES.2 Recent Trends in U.S. Greenhouse Gas Emissions and Sinks In 2016, total gross U.S. greenhouse gas emissions were 6,546.2 million metric tons (MMT) of CO2 Eq. Total U.S. emissions have increased by 2.8 percent from 1990 to 2016, and emissions decreased from 2015 to 2016 by 2.0 percent (131.1 MMT CO2 Eq.). The decrease in total greenhouse gas emissions between 2015 and 2016 was driven in large part by a decrease in CO2 emissions from fossil fuel combustion. The decrease in CO2 emissions from fossil fuel combustion was a result of multiple factors, including: (1) substitution from coal to natural gas and other sources in the electric power sector; and (2) warmer winter conditions in 2016 resulting in a decreased demand for heating fuel in the residential and commercial sectors. Relative to 1990, the baseline fortius Inventory, gross emissions in 2016 are higher by 2.8 percent, down from a high of 15.6 percent above 1990 levels in 2007. Overall, net emissions in 2016 were 11.6 percent below 2005 levels as shown in Table ES-2. Figure ES-1 through Figure ES-3 illustrate the overall trends in total U.S. emissions by gas, annual changes, and absolute change since 1990, and Table ES-2 provides a detailed summary of gross U.S. greenhouse gas emissions and sinks for 1990 through 2016. Note, unless otherwise stated, all tables and figures provide total emissions without LULUCF. Figure ES-1: Gross U.S. Greenhouse Gas Emissions by Gas (MMT CO2 Eq.) 9,000 8,000 7,000 6,000 & LU 0 5,000 u 1- 1 4,000 3,000 2,000 1,000 0 I HFCs, PFCs, SF, and NF. Subtotal Nitrous Oxide I Methane I Carbon Dioxide ES-4 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2016 ------- 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 Figure ES-2: Annual Percent Change in Gross U.S. Greenhouse Gas Emissions Relative to the Previous Year Figure ES-3: Cumulative Change in Annual Gross U.S. Greenhouse Gas Emissions Relative to 1990 (1990=0, MMT COz Eq.) 1,200- 1,100- Box ES-3: Improvements and Recalculations Relative to the Previous Inventory Each year, some emission and sink estimates in the Inventory are recalculated with improved methods and/or data. These improvements are also implemented consistently across the previous Inventory's time series (i.e., 1990 to 2015) to ensure that the trend is accurate (see also Box ES-6 on the recalculations approach). Below are categories with recalculations resulting in an average change over the time series of greater than 10 MMT CO2 Eq. For more information on specific improvements, please see the Recalculations and Improvements chapter (Chapter 9) and the Energy chapter (Chapter 3). • Fossil Fuel Combustion (CO2). Average increase of 14.0 MMT CO2 Eq. relative to the previous Inventory, resulting primarily from incorporation of updated energy consumption statistics from EIA. • Petroleum Systems (CO2). Average increase of 13.8 MMT CO2 Eq. relative to the previous Inventory, resulting primarily from reallocation of CO2 from flaring to petroleum systems from natural gas systems. Executive Summary ES-5 ------- 1 • Petroleum Systems (CH4). Average decrease of 10.9 MMT CO2 Eq. relative to the previous Inventory, 2 resulting primarily from recalculation of associated gas venting and flaring emissions using a basin-level 3 approach. 4 • Natural Gas Systems (CO2). Average decrease of 10.3 MMT CO2 Eq. relative to the previous Inventory, 5 resulting primarily from reallocation of CO2 from flaring to petroleum systems from natural gas systems. 6 Other improvements of note include recalculations of CH4 estimates from Municipal Solid Waste (MSW) Landfills 7 (See Section 7.1 of the Waste chapter). 8 9 Table ES-2: Recent Trends in U.S. Greenhouse Gas Emissions and Sinks (MMT CO2 Eq.) Gas/Source 1990 2005 2012 2013 2014 2015 2016 CO2 5,136.8 6,150.8 5,383.7 5,541.7 5,590.5 5,449.5 5,333.4 Fossil Fuel Combustion 4,755.8 5,759.1 5,029.8 5,162.3 5,206.1 5,059.3 4,'976.7 Electric Power 1,820.8 2,400.9 2,022.2 2,038.1 2,038.0 1,900.7 1,808.8 Transportation 1,467.2 1,855.8 1,661.9 1,677.6 1,717.1 1,735.5 1,794.9 Industrial 874.5 867.8 818.4 848.7 830.8 819.3 807.6 Residential 338.3 357.8 282.5 329.7 345.3 316.8 296.2 Commercial 227.4 227.0 201.3 225.7 233.6 245.6 227.9 U.S. Territories 27.6 49.7 43.5 42.5 41.4 41.4 41.4 Non-Energy Use of Fuels 119.6 141.7 113.3 133.2 127.8 135.1 121.0 Iron and Steel Production & Metallurgical Coke Production 101.5 68.0 55.4 53.3 58.2 47.7 42.2 Cement Production 33.5 46.2 35.3 36.4 39.4 39.9 39.4 Petrochemical Production 21.2 26.8 26.5 26.4 26.5 28.1 27.4 Natural Gas Systems 29.7 22.5 24.4 26.0 27.0 26.3 26.7 Petroleum Systems 9.4 17.0 25.6 29.7 32.9 38.0 25.5 Lime Production 11.7 14.6 13.8 14.0 14.2 13.3 13.3 Other Process Uses of Carbonates 4.9 6.3 8.0 10.4 11.8 11.2 11.2 Ammonia Production 13.0 9.2 9.4 10.0 9.6 10.6 11.2 Incineration of Waste 8.0 12.5 10.4 10.4 10.6 10.7 10.7 Urea Fertilization 2.4 3.5 4.3 4.4 4.5 4.9 5.1 Carbon Dioxide Consumption 1.5 1.4 4.0 4.2 4.5 4.5 4.5 Urea Consumption for Non- Agricultural Purposes 3.8 3.7 4.4 4.1 1.5 4.2 4.0 Liming 4.7 4.3 6.0 3.9 3.6 3.8 3.9 Ferroalloy Production 2.2 1.4 1.9 1.8 1.9 2.0 1.8 Soda Ash Production 1.4 1.7 1.7 1.7 1.7 1.7 1.7 Titanium Dioxide Production 1.2 1.8 1.5 1.7 1.7 1.6 1.6 Aluminum Production 6.8 4.1 3.4 3.3 2.8 2.8 1.3 Glass Production 1.5 1.9 1.2 1.3 1.3 1.3 1.3 Phosphoric Acid Production 1.5 1.3 1.1 1.1 1.0 1.0 1.0 Zinc Production 0.6 1.0 1.5 1.4 1.0 0.9 0.9 Lead Production 0.5 0.6 0.5 0.5 0.5 0.5 0.5 Silicon Carbide Production and Consumption 0.4 0.2 0.2 0.2 0.2 0.2 0.2 Magnesium Production and Processing + + + + + + + Wood Biomass, Ethanol, and Biodiesel Consumption" 219.4 230.7 276.2 299.8 308.3 294.5 291.1 International Bunker Fuelsb 103.5 113.1 105.8 99.8 103.4 110.9 114.4 CH4c 778.1 679.3 661.3 659.6 665.3 664.0 655.8 Enteric Fermentation 164.2 168.9 166.7 165.5 164.2 166.5 170.1 Natural Gas Systems 193.7 160.0 156.8 159.6 164.2 164.4 162.1 ES-6 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2016 ------- Landfills 179.6 132.7 117.0 113.3 112.7 111.7 107.7 Manure Management 37.2 56.3 65.6 63.3 62.9 66.3 67.7 Coal Mining 96.5 64.1 66.5 64.6 64.6 61.2 53.8 Petroleum Systems 42.3 34.7 35.4 38.8 41.0 39.4 39.3 Wastewater Treatment 15.7 15.8 15.1 14.9 15.0 15.1 14.8 Rice Cultivation 16.0 16.7 11.3 11.5 12.7 12.3 13.7 Stationary Combustion 8.6 7.9 7.3 8.7 8.8 7.8 7.2 Abandoned Oil and Gas Wells 6.5 6.9 7.0 7.0 7.1 7.2 7.1 Abandoned Underground Coal Mines 7.2 6.6 6.2 6.2 6.3 6.4 6.7 Mobile Combustion 9.8 6.6 4.0 3.7 3.4 3.1 3.0 Composting 0.4 1.9 1.9 2.0 2.1 2.1 2.1 Field Burning of Agricultural Residues 0.2 0.2 0.3 0.3 0.3 0.3 0.3 Petrochemical Production 0.2 0.1 0.1 0.1 0.1 0.2 0.2 Ferroalloy Production + + + + + + + Silicon Carbide Production and Consumption + + + + + + + Iron and Steel Production & Metallurgical Coke Production + + + + + + + Incineration of Waste + + + + + + + International Bunker Fuelsb 0.2 0.1 0.1 0.1 0.1 0.1 0.1 N2Oc 354.6 357.4 335.2 362.6 360.5 378.9 368.8 Agricultural Soil Management 250.5 253.5 247.9 276.6 274.0 295.0 283.6 Stationary Combustion 11.1 17.5 16.8 18.6 18.9 18.0 18.4 Manure Management 14.0 16.5 17.5 17.5 17.5 17.7 18.1 Mobile Combustion 41.5 38.4 23.8 22.0 20.2 18.8 17.8 Nitric Acid Production 12.1 11.3 10.5 10.7 10.9 11.6 10.2 Adipic Acid Production 15.2 7.1 5.5 3.9 5.4 4.3 7.0 Wastewater Treatment 3.4 4.4 4.6 4.7 4.8 4.8 5.0 NjO from Product Uses 4.2 4.2 4.2 4.2 4.2 4.2 4.2 Caprolactam, Glyoxal, and Glyoxylic Acid Production 1.7 2.1 2.0 2.0 2.0 2.0 2.0 Composting 0.3 1.7 1.7 1.8 1.9 1.9 1.9 Incineration of Waste 0.5 0.4 0.3 0.3 0.3 0.3 0.3 Semiconductor Manufacture + 0.1 0.2 0.2 0.2 0.2 0.2 Field Burning of Agricultural Residues 0.1 0.1 0.1 0.1 0.1 0.1 0.1 International Bunker Fuelsb 0.9 1.0 0.9 0.9 0.9 0.9 1.0 HFCs 46.6 120.0 156.0 159.1 166.8 173.3 177.1 Substitution of Ozone Depleting Substances'1 0.3 99.8 150.3 154.8 161.4 168.6 173.9 HCFC-22 Production 46.1 20.0 5.5 4.1 5.0 4.3 2.8 Semiconductor Manufacture 0.2 0.2 0.2 0.2 0.3 0.3 0.3 Magnesium Production and Processing 0.0 0.0 + 0.1 0.1 0.1 0.1 PFCs 24.3 6.7 5.9 5.8 5.6 5.1 4.3 Semiconductor Manufacture 2.8 3.3 3.0 2.8 3.1 3.1 3.0 Aluminum Production 21.5 3.4 2.9 3.0 2.5 2.0 1.4 Substitution of Ozone Depleting Substances 0.0 + + + + + + SF« 28.8 11.7 6.6 6.3 6.3 5.9 6.2 Electrical Transmission and Distribution 23.1 8.3 4.6 4.5 4.6 4.2 4.3 Magnesium Production and Processing 5.2 2.7 1.6 1.5 1.0 0.9 1.0 Semiconductor Manufacture 0.5 0.7 0.3 0.4 0.7 0.7 0.8 NF3 + 0.5 0.6 0.6 0.5 0.6 0.6 Semiconductor Manufacture + 0.5 0.6 0.6 0.5 0.6 0.6 Executive Summary ES-7 ------- 1 2 3 4 5 6 7 8 9 10 11 12 Total Emissions 6,369.2 7,326.4 6,549.4 6,735.6 6,795.6 6,677.3 6,546.2 LULUCF Emissions0 10.6 23.0 26.1 19.2 19.6 38.2 38.1 LULUCF CH4 Emissions 6.7 13.3 15.0 10.9 11.2 22.4 22.4 LULUCF N2O Emissions 3.9 H 11.1 8.3 8.4 15.8 15.7 LULUCF Carbon Stock Change6 (830.2) (754.2) (779.5) (755.0) (760.0) (733.4) (754.9) LULUCF Sector Net Total' (819.6) (731.1) (753.5) (735.8) (740.4) (695.2) (716.8) Net Emissions (Sources and Sinks) 5,549.6 6,5^5.3 5,795.9 5,999.9 6,055.2 5,982.1 5,829.3 Notes: Total emissions presented without LULUCF. Net emissions presented with LULUCF. + Does not exceed 0.05 MMT CO2 Eq. a Emissions from Wood 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 Land Use, Land-Use Change, and Forestry. b Emissions from International Bunker Fuels are not included in totals. 0 LULUCF emissions of CH4 andN20 are reported separately from gross emissions totals. LULUCF emissions include the CH4 and N2O 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 N2O emissions from Forest Soils and Settlement Soils. Refer to Table ES-5 for a breakout of emissions and removals for Land Use, Land-Use Change, and Forestry by gas and source category. d Small amounts of PFC emissions also result from this source. 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. Refer to Table ES-5 for a breakout of emissions and removals for Land Use, Land-Use Change, and Forestry by gas and source category. f The LULUCF Sector Net Total is the net sum of all CH4 and N2O emissions to the atmosphere plus net carbon stock changes. Notes: Totals may not sum due to independent rounding. Parentheses indicate negative values or sequestration. Figure ES-4 illustrates the relative contribution of the direct greenhouse gases to total U.S. emissions in 2016, weighted by global warming potential. The primary greenhouse gas emitted by human activities in the United States was CO2, representing approximately 81.5 percent of total greenhouse gas emissions. The largest source of CO2, and of overall greenhouse gas emissions, was fossil fuel combustion. Methane emissions, which have decreased by 15.7 percent since 1990, resulted primarily from enteric fermentation associated with domestic livestock, natural gas systems, and decomposition of wastes in landfills. Agricultural soil management, stationary fuel combustion, manure management, and mobile source fuel combustion were the major sources of N20 emissions. Ozone depleting substance substitute emissions and emissions of HFC-23 during the production of HCFC-22 were the primary contributors to aggregate hydrofluorocarbon (HFC) emissions. Perfluorocarbon (PFC) emissions resulted from semiconductor manufacturing and as a byproduct of primary aluminum production, electrical transmission and distribution systems accounted for most sulfur hexafluoride (SF6) emissions, and semiconductor manufacturing is the only source of nitrogen trifluoride (NF3) emissions. ES-8 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2016 ------- 1 Figure ES-4: 2016 U.S. Greenhouse Gas Emissions by Gas (Percentages based on MMT CO2 2 Eq.) HFCs, PFCs, SF, and NF, Subtotal 2.9% N20 5.6% CH. 10.0% CO, 81.5% 3 4 5 Overall, from 1990 to 2016, total emissions of CO2 increased by 196.5 MMT CO2 Eq. (3.8 percent), while total 6 emissions of CH4 decreased by 122.3 MMT CO: Eq. (15.7 percent), and N;0 emissions increased by 14.2 MMT 7 CO2 Eq. (4.0 percent). During the same period, aggregate weighted emissions of HFCs, PFCs, SF6 and NF;, rose by 8 88.6 MMT CO2 Eq. (88.8 percent). From 1990 to 2016, HFCs increased by 130.5 MMT CO2 Eq. (280.3 percent), 9 PFCs decreased by 19.9 MMT CO2 Eq. (82.1 percent), SF6 decreased by 22.6 MMT CO2 Eq. (78.5 percent), and 10 NF3 increased by 0.5 MMT CO2 Eq. (1,110.2 percent). Despite being emitted in smaller quantities relative to the 11 other principal greenhouse gases, emissions of HFCs, PFCs, SF6 and NF;, are significant because many of these 12 gases have extremely high global warming potentials and, in the cases of PFCs and SF6, long atmospheric lifetimes. 13 Conversely, U.S. greenhouse gas emissions were partly offset by carbon (C) sequestration in forests, trees in urban 14 areas, agricultural soils, landfilled yard trimmings and food scraps, and coastal wetlands, which, in aggregate, offset 15 11.5 percent of total emissions in 2016. The following sections describe each gas's contribution to total U.S. 16 greenhouse gas emissions in more detail. 17 Carbon Dioxide Emissions 18 The global carbon cycle is made up of large carbon flows and reservoirs. Billions of tons of carbon in the form of 19 CO2 are absorbed by oceans and living biomass (i.e., sinks) and are emitted to the atmosphere annually through 20 natural processes (i.e., sources). When in equilibrium, carbon fluxes among these various reservoirs are roughly 21 balanced.11 22 Since the Industrial Revolution (i.e., about 1750), global atmospheric concentrations of CO2 have risen 23 approximately 44 percent (IPCC 2013; NOAA/ESRL 2017a), principally due to the combustion of fossil fuels. 24 Globally, approximately 32,294 MMT of CO2 were added to the atmosphere through the combustion of fossil fuels 25 in 2015, of which the United States accounted for approximately 15 percent.12 26 Within the United States, fossil fuel combustion accounted for 93.3 percent of CO2 emissions in 2016. There are 24 27 additional sources of CO2 emissions included in the Inventory (see Figure ES-5). Although not illustrated in the 28 Figure ES-5, changes in land use and forestry practices can also lead to net CO2 emissions (e.g., through conversion 29 of forest land to agricultural or urban use) or to a net sink for CO2 (e.g., through net additions to forest biomass). 11 Hie term "flux" is used to describe the net emissions of greenhouse gases accounting for both the emissions of CO2 to and the removals of CO2 from the atmosphere. Removal of CO2 from the atmosphere is also referred to as "carbon sequestration." 12 Global CO2 emissions from fossil fuel combustion were taken from International Energy Agency CO: Emissions from Fossil Fuels Combustion -Highlights. IEA (2017). See . The publication has not yet been updated to include 2016 data. Executive Summary ES-9 ------- 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 Figure ES-5: 2016 Sources of CO2 Emissions (MMT CO2 Eq.) Fossil Fuel Combustion Non-Energy Use of Fuels Iron and Steel Prod. & Metallurgical Coke Prod. Cement Production Petrochemical Production Natural Gas Systems Petroleum Systems Lime Production Other Process Uses of Carbonates Ammonia Production Incineration of Waste Urea Fertilization Carbon Dioxide Consumption Urea Consumption for Non-Agricultural Purposes Liming Ferroalloy Production Soda Ash Production Titanium Dioxide Production Aluminum Production Glass Production Phosphoric Acid Production Zinc Production Lead Production Silicon Carbide Production and Consumption Magnesium Production and Processing | 4,977 < .05 < .05 < .05 CO2 as a Portion of all Emissions 0 25 50 75 100 MMT CO= Eq. 125 150 As the largest source of U.S. greenhouse gas emissions, CO2 from fossil fuel combustion has accounted for approximately 77 percent of GWP-weighted emissions since 1990. The fundamental factors influencing emissions levels include: (1) changes in demand for energy; and (2) a general decline in the carbon intensity of fuels combusted for energy in recent years by most sectors of the economy. Between 1990 and 2016, CO2 emissions from fossil fuel combustion increased from 4,755.8 MMT CO2 Eq. to 4,976.7 MMT CO2 Eq., a 4.6 percent total increase over the twenty-seven-year period. Conversely, CO2 emissions from fossil fuel combustion decreased by 782.3 MMT CO2 Eq. from 2005 levels, a decrease of approximately 13.6 percent between 2005 and 2016. From 2015 to 2016, these emissions decreased by 82.6 MMT CO2 Eq. (1.6 percent). Historically, changes in emissions from fossil fuel combustion have been the dominant factor affecting U.S. emission trends. Changes in CO2 emissions from fossil fuel combustion are influenced by many long-term and short-term factors. Long-term factors include population and economic trends, technological changes, shifting energy fuel choices, and various policies at the national, state, and local level. In the short term, the overall consumption and mix of fossil fuels in the United States fluctuates primarily 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. The five major fuel consuming economic sectors contributing to CO2 emissions from fossil fuel combustion are electric power, transportation, industrial, residential, and commercial. Carbon dioxide emissions are produced by the electric power sector as fossil fuel is consumed to provide electricity to one of the other four sectors, or "end-use" sectors. For the discussion below, electric power emissions have been distributed to each end-use sector on the basis of each sector's share of aggregate electricity use. This method of distributing emissions assumes that each end-use sector uses electricity that is generated from the national average mix of fuels according to their carbon intensity. ES-10 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2016 ------- 1 2 3 4 5 6 7 8 9 10 11 12 13 14 Emissions from electric power are also addressed separately after the end-use sectors have been discussed. Note that emissions from U.S. Territories are calculated separately due to a lack of specific consumption data for the individual end-use sectors. Figure ES-6, Figure ES-7, and Table ES-3 summarize CO2 emissions from fossil fuel combustion by end-use sector. Figure ES-6: 2016 CO2 Emissions from Fossil Fuel Combustion by Sector and Fuel Type (MMT COz Eq.) 2,500 2,000 £ 1,500 0 U I- 1 1,000 500 Relative Contribution by Fuel Type 41 228 I Petroleum I Coal I Natural Gas 296 U.S. Territories Commercial Residential 1,795 1,809 Industrial Transportation Electric Power Note on Figure ES-6: Fossil Fuel Combustion includes electric power, which also includes emissions of less than 0.5 MMT CO2 Eq. from geothermal-based generation. Figure ES-7: 2016 End-Use Sector Emissions of CO2 from Fossil Fuel Combustion (MMT CO2 Eq.) 2,000 1,500 8 1,000- b s: z 500- I Direct Fossil Fuel Combustion Indirect Fossil Fuel Combustion 1,798 1,314 U.S. Territories Commercial Residential Industrial Transportation Table ES-3: CO2 Emissions from Fossil Fuel Combustion by End-Use Sector (MMT CO2 Eq.) End-Use Sector 1990 2005 2012 2013 2014 2015 2016 Transportation 1,470.2 1,860.5 1,665.8 1,681.6 1,721.2 1,739.2 1,798.4 Combustion 1,467.2 1,855.8 1,661.9 1,677.6 1,717.1 1,735.5 1,794.9 Electricity 3.0 4.7 3.9 4.0 4.1 3.7 3.5 Industrial 1,561.3 1,604.4 1,411.2 1,443.4 1,424.0 1,368.8 1,313.8 Combustion 874.5 867.8 818.4 848.7 830.8 819.3 807.6 Electricity 686.7 736.6 592.8 594.7 593.2 549.6 506.2 Executive Summary ES-11 ------- 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 Residential 931.4 1,214.1 1,007.8 1,064.6 1,080.0 1,001.1 957.0 Combustion 338.3 357.8 282.5 329.7 345.3 316.8 296.2 Electricity 593.0 856.3 725.3 734.9 734.7 684.3 660.7 Commercial 765.3 1,030.3 901.6 930.2 939.6 908.8 866.2 Combustion 227.4 227.0 201.3 225.7 233.6 245.6 227.9 Electricity 538.0 803.3 700.3 704.5 706.0 663.1 638.3 U.S. Territories3 27.6 49.7 43.5 42.5 41.4 41.4 41.4 Total 4,755.8 5,759.1 5,029.8 5,162.3 5,206.1 5,059.3 4,976.7 Electric Power 1,820.8 2,400.9 2,022.2 2,038.1 2,038.0 1,900.7 1,808.8 aFuel 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. 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. Transportation End-Use Sector. When electricity-related emissions are distributed to economic end-use sectors, transportation activities accounted for 36.1 percent of U.S. CO2 emissions from fossil fuel combustion in 2016. The largest sources of transportation CO2 emissions in 2016 were passenger cars (42.2 percent), medium- and heavy- duty trucks (23.3 percent), light-duty trucks, which include sport utility vehicles, pickup trucks, and minivans (17.0 percent), commercial aircraft (6.6 percent), rail (2.2 percent), other aircraft (2.8 percent), pipelines (2.2 percent), and ships and boats (2.3 percent). Annex 3.2 presents the total emissions from all transportation and mobile sources, including CO2, CH4, N20, and HFCs. In terms of the overall trend, from 1990 to 2016, total transportation CO2 emissions increased due, in large part, to increased demand for travel. The number of VMT by light-duty motor vehicles (i.e., passenger cars and light-duty trucks) increased 43 percent from 1990 to 2016,13 as a result of a confluence of factors including population growth, economic growth, urban sprawl, and low fuel prices during the beginning of this period. 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 diesel fuel for freight trucks and jet fuel for aircraft, accounted for the remainder. Industrial End-Use Sector. Industrial CO2 emissions, resulting both directly from the combustion of fossil fuels and indirectly from the generation of electricity that is used by industry, accounted for 26 percent of CO2 from fossil fuel combustion in 2016. Approximately 61 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. In contrast to the other end-use sectors, emissions from industry have declined 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. Residential and Commercial End-Use Sectors. The residential and commercial end-use sectors accounted for 19 and 17 percent, respectively, of CO2 emissions from fossil fuel combustion in 2016. Both sectors relied heavily on electricity for meeting energy demands, with 69 and 74 percent, respectively, of their emissions attributable to electricity use for lighting, heating, cooling, and operating appliances. The remaining emissions were due to the consumption of natural gas and petroleum for heating and cooking. Emissions from the residential and commercial end-use sectors have increased by 3 percent and 13 percent since 1990, respectively. Electric Power. The United States relies on electricity to meet a significant portion of its energy demands. Electricity generators used 33 percent of U.S. energy from fossil fuels and emitted 36 percent of the CO2 from fossil fuel combustion in 2016. The type of energy source used to generate electricity is the main factor influencing 13 VMT estimates are based on data from FHWA Highway Statistics Table VM-1 (FHWA 1996 through 2017). Table VM-1 data for 2016 has not been published yet, therefore 2016 mileage data is estimated using the 1.7 percent increase in FHWA Traffic Volume Trends from 2015 to 2016. In 2011, FHWA changed its methods for estimating VMT by vehicle class, which led to a shift in VMT and emissions among on-road vehicle classes in the 2007 to 2016 time period. In absence of these method changes, light-duty VMT growth between 1990 and 2016 would likely have been even higher. ES-12 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2016 ------- 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 emissions. For example, some electricity is generated through non-fossil fuel options such as nuclear, hydroelectric, wind, solar, or geothennal energy. Including all electricity generation modes, electric power sector generators relied on coal for approximately 30 percent of their total energy requirements in 2016. In addition, the coal used by electricity generators accounted for 93 percent of all coal consumed for energy in the United States in 2016.14 Recently, a decrease in the carbon intensity of the mix of fuels consumed to generate electricity has occurred due to a decrease in coal consumption, increased natural gas consumption, and increased reliance on non-fossil generation sources. Including all electricity generation modes, electric power sector generators used natural gas for approximately 34 percent of their total energy requirements in 2016. Across the time series, changes in electricity demand and the carbon intensity of fuels used for electric power have a significant impact on CO2 emissions. While emissions from the electric power sector have decreased by approximately 0.2 percent since 1990, the carbon intensity of the electric power sector, in terms of CO2 Eq. per QBtu, input has significantly decreased—by 12 percent—during that same timeframe. This trend away from a direct relationship between electric power and the resulting emissions is shown in Figure ES-8. Figure ES-8: Electric Power Generation (Billion kWh) and Emissions (MMT CO2 Eq.) 5,000 ¦ Petroleum-based Generation (Billion kWh) ¦ Nuclear-based Generation (Billion kWh) 4 50O-I I Renewable-based Generation (Billion kWh) ¦ Natural Gas-based Generation (Billion kWh) ¦ Coal-based Generation (Billion kWh) 3,500 3,000 4,000- Total Emissions (MMT COi Eq.) [Right Axis] 2,500 3,500- 3,000- 2,000 2,500- 1,500 .2 2,000- 1,500- 1,000 1,000- Other significant CO2 trends included the following: • Carbon dioxide emissions from non-energy use of fossil fuels increased by 1.5 MMT CO2 Eq. (1.2 percent) from 1990 through 2016. Emissions from non-energy uses of fossil fuels were 121.0 MMT CO2 Eq. in 2016, which constituted 2.3 percent of total national CO2 emissions, approximately the same proportion as in 1990. 14 See Table 6.2 Coal Consumption by Sector of EIA 2016. Executive Summary ES-13 ------- 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 • Carbon dioxide emissions from iron and steel production and metallurgical coke production have decreased by 59.3 MMT CO2 Eq. (58.4 percent) from 1990 through 2016, due to restructuring of the industry, technological improvements, and increased scrap steel utilization. • Total C stock change (i.e., net CO2 removals) in the LULUCF sector decreased by approximately 9.1 percent between 1990 and 2016. This decrease was primarily due to a decrease in the rate of net C accumulation in forest C stocks and Cropland Remaining Cropland, as well as an increase in emissions from Land Converted to Settlements. Box ES-4: Use of Ambient Measurements Systems for Validation of Emission Inventories In following the UNFCCC requirement under Article 4.1 to develop and submit national greenhouse gas emission inventories, the emissions and sinks presented in this report are organized by source and sink categories and calculated using internationally-accepted methods provided by the IPCC.15 Several recent studies have measured emissions at the national or regional level with 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 greenhouse gas emissions could assist in improving the understanding of inventory estimates. In working with the research community on ambient measurement and remote sensing techniques to improve national greenhouse gas inventories, EPA follows guidance from the IPCC on the use of measurements and modeling to validate emission inventories. 16An area of particular interest in EPA's outreach efforts is how ambient measurement data can be used in a manner consistent with this Inventory report's transparency on its calculation methodologies, and the ability of these techniques to attribute emissions and removals from remote sensing to anthropogenic sources, as defined by the IPCC for this report, versus natural sources and sinks. In an effort to improve the ability to compare the national-level greenhouse gas inventory with measurement results that may be at other scales, a team at Harvard University along with EPA and other coauthors developed a gridded inventory of U.S. anthropogenic methane emissions with 0.1° x 0.1° spatial resolution, monthly temporal resolution and detailed scale-dependent error characterization. The Inventory is designed to be consistent with the 1990 to 2014 U.S. EPA Inventory of U.S. Greenhouse Gas Emissions and Sinks estimates for the year 2012, which presents national totals for different source types.17 Methane Emissions Methane (CH4) is 25 times as effective as CO2 at trapping heat in the atmosphere (IPCC 2007). Over the last two hundred and fifty years, the concentration of CH4 in the atmosphere increased by 163 percent (IPCC 2013; NOAA/ESRL 2017b). Anthropogenic sources of CH4 include natural gas and petroleum systems, agricultural activities, LULUCF, landfills, coal mining, wastewater treatment, stationary and mobile combustion and certain industrial processes (see Figure ES-9). 15 See . 16 See . 17 See . ES-14 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2016 ------- 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 Figure ES-9: 2016 Sources of ChU Emissions (MMT CO2 Eq.) Enteric Fermentation Natural Gas Systems Landfills Manure Management Coal Mining Petroleum Systems Wastewater Treatment Rice Cultivation Stationary Combustion Abandoned Oil and Gas Wells Abandoned Underground Coal Mines Mobile Combustion Composting Field Burning of Agricultural Residues Petrochemical Production Ferroalloy Production Silicon Carbide Production and Consumption Iron and Steel Production & Metallurgical Coke Production Incineration of Waste 170 ¦ ¦ ¦ I I < .05 < .05 < .05 < .05 < .05 < .05 CH. as a Portion Emissions of all 10.0% 25 50 75 100 MMT CO: Eq. 125 150 175 Note: LULUCF emissions are reported separately from gross emissions totals and are not included in Figure ES-9. Refer to Table ES-5 for a breakout of LULUCF emissions by gas. Significant trends for the largest sources of U.S. CH4 emissions include the following: • Enteric fermentation is the largest anthropogenic source of CH4 emissions in the United States. In 2016, enteric fermentation CH4 emissions were 170.1 MMT CO2 Eq. (25.9 percent of total CH4 emissions), which represents an increase of 6.0 MMT CO2 Eq. (3.6 percent) since 1990. This increase in emissions from 1990 to 2016 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 2016 with 162.1 MMT CO2 Eq. of CH4 emitted into the atmosphere. Those emissions have decreased by 31.6 MMT CO2 Eq. (16.3 percent) since 1990. The decrease in CH4 emissions is largely due to the decrease in emissions from transmission, storage, and distribution. The decrease in transmission and storage emissions is largely due to reduced compressor station emissions (including emissions from compressors and fugitives). The decrease in distribution emissions is largely attributed to increased use of plastic piping, which has lower emissions than other pipe materials, and station upgrades at metering and regulating (M&R) stations. • Landfills are the third largest anthropogenic source of CH4 emissions in the United States (107.7 MMT CO2 Eq.), accounting for 16.4 percent of total CH4 emissions in 2016. From 1990 to 2016, CH4 emissions from landfills decreased by 71.9 MMT CO2 Eq. (40.0 percent), with small increases occurring in some interim years. 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,18 which has more than offset the 18 Carbon dioxide emissions from landfills are not included specifically in summing waste sector totals. Net carbon fluxes from changes in biogenic carbon reservoirs and disposed wood products are accounted for in the estimates for LULUCF. Executive Summary ES-15 ------- 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 additional CH4 emissions that would have resulted from an increase in the amount of municipal solid waste landfilled. Nitrous Oxide Emissions Nitrous oxide (N20) 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 N20 emissions are much lower than CO2 emissions, N20 is nearly 300 times more powerful than CO2 at trapping heat in the atmosphere (IPCC 2007). Since 1750, the global atmospheric concentration of N20 lias risen by approximately 22 percent (IPCC 2013; NOAA/ESRL 2017c). The main anthropogenic activities producing N2O in the United States are agricultural soil management, stationary fuel combustion, fuel combustion in motor vehicles, manure management, and nitric acid production (see Figure ES-10). Figure ES-10: 2016 Sources of N2O Emissions (MMT CO2 Eq.) Agricultural Soil Management Stationary Combustion Manure Management Mobile Combustion Nitric Acid Production Adipic Acid Production Wastewater Treatment NiO from Product Uses Caprolactam, Glyoxal, and Glyoxylic Acid Production Composting Incineration of Waste Semiconductor Manufacture Field Burning of Agricultural Residues Note: LULUCF emissions are reported separately from gross emissions totals and are not included in Figure ES-10. Refer to Table ES-5 for a breakout of LULUCF emissions by gas. Significant trends forthe largest sources of U.S. emissions of N20 include the following: • Agricultural soils accounted for approximately 76.9 percent of N20 emissions and 4.3 percent of total emissions in the United States in 2016. Estimated emissions from this source in 2016 were 283.6 MMT CO2 Eq. Annual N20 emissions from agricultural soils fluctuated between 1990 and 2016, although overall emissions were 13.2 percent higher in 2016 than in 1990. Year-to-year fluctuations are largely a reflection of annual variation in weather patterns, synthetic fertilizer use, and crop production. • Nitrous oxide emissions from stationary combustion increased 7.2 MMT CO2 Eq. (64.9 percent) from 1990 through 2016. Nitrous oxide emissions from this source increased primarily as a result of an increase in the number of coal fluidized bed boilers in the electric power sector. • In 2016, total N20 emissions from manure management were estimated to be 18.1 MMT CO2 Eq.; emissions were 14.0 MMT CO2 Eq. in 1990. These values include both direct and indirect N20 emissions from manure management. Nitrous oxide emissions have remained fairly steady since 1990. Small changes in N2O emissions from individual animal groups exhibit the same trends as the animal group populations, with the overall net effect that N20 emissions showed a 29.6 percent increase from 1990 to 2016 and a 2.4 percent increase from 2015 through 2016. 284 NjO as a Portion of all Emissions 5.6% < 0.5 < 0.5 < 0.5 10 15 MMT CO; Eq. 20 25 ES-16 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2016 ------- 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 HFC, PFC, SF6, and NF3 Emissions Hydrofluorocarbons (HFCs) and perfluorocarbons (PFCs) are families of 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 and PFCs do not deplete the stratospheric ozone layer, and are therefore acceptable alternatives under the Montreal Protocol on Substances that Deplete the Ozone Layer. These compounds, however, along with SF6 and NF3, are potent greenhouse gases. In addition to having high global warming potentials, SF6 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 lias evaluated (IPCC 2013). Other emissive sources of these gases include HCFC-22 production electrical transmission and distribution systems, semiconductor manufacturing, aluminum production and magnesium production and processing (see Figure ES-11). Figure ES-11: 2016 Sources of HFCs, PFCs, SFe, and NF3 Emissions (MMT CO2 Eq.) Substitution of Ozone Depleting Substances I Semiconductor Manufacture Electrical Transmission and Distribution HCFC-22 Production Aluminum Production Magnesium Production and Processing 174 HFCs, PFCs, SFi, and NFs as a Portion of all Emissions 2.9% 10 MMT CO: Eq. 20 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 the substitution of ODS (e.g., chlorofluorocarbons [CFCs]) have been consistently increasing, from small amounts in 1990 to 173.9 MMT CO2 Eq. in 2016. This increase was in large part the result of efforts to phase out CFCs and other ODS in the United States. In the short term, this trend is expected to continue, and will likely continue over the next decade as hydrochlorofluorocarbons (HCFCs), which are interim substitutes in many applications, are themselves phased out under the provisions of the Copenhagen Amendments to the Montreal Protocol. • GWP-weighted PFC, HFC, SF6, and NF3 emissions from semiconductor manufacturing have increased by 32.8 percent from 1990 to 2016, due to competing factors of industrial growth and the adoption of emission reduction technologies. Within that time span, emissions peaked at 9.0 MMT CO2 Eq. in 1999, the initial year of EPA's PFC Reduction/Climate Partnership for the Semiconductor Industry, but have since declined to 4.7 MMT CO2 Eq. in 2016 (a 47.6 percent decrease relative to 1999). • Sulfur hexafluoride emissions from electric power transmission and distribution systems decreased by 81.3 percent (18.8 MMT CO2 Eq.) from 1990 to 2016. There are two potential causes for this decrease: (1) a sharp increase in the price of SF6 during the 1990s and (2) a growing awareness of the enviromnental impact of SF6 emissions through programs such as EPA's SF6 Emission Reduction Partnership for Electric Power Systems. Executive Summary ES-17 ------- 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 ES.3 Overview of Sector Emissions and Trends In accordance with the UNFCCC decision to set the 2006IPCC Guidelines for National Greenhouse Gas Inventories (IPCC 2006) as the standard for Annex I countries at the Nineteenth Conference of the Parties (UNFCCC 2014), Figure ES-12 and Table ES-4 aggregate emissions and sinks by the sectors defined by those guidelines. Ov er the twenty-seven-year period of 1990 to 2016, total emissions from the Energy, Industrial Processes and Product Use. and Agriculture sectors grew by 136.2 MMT CO2 Eq. (2.6 percent), 35.2 MMT CO2 Eq. (10.3 percent), and 73.4 MMT CO; Eq. (15.0 percent), respectively. Emissions from the Waste sector decreased by 67.9 MMT CO2 Eq. (34.1 percent). Over the same period, total C sequestration in the Land Use, Land-Use Change, and Forestry (LULUCF) sector increased by 75.3 MMT CO2 (9.1 percent decrease in total C sequestration), and emissions from the LULUCF sector increased by 27.4 MMT CO2 Eq. (258 percent). Figure ES-12: U.S. Greenhouse Gas Emissions and Sinks by Chapter/IPCC Sector (MMT CO2 Eq.) -7 ™ -1 Industrial Processes and Product Use 7,500- Waste \ 7,000- \ ¦ V - M "*'(emissions) 6,500- ^Agriculture '*" 6,000- 5,500- I 5,000- I cr 4,500- I UJ q 4,000- I u 3,500- I 3,000- Energy s 2,500- I 2,000- I 1,500- I 1,000- I 500- I OH I -500 Lanci Use' Land-Use Change and Forestry (LULUCF) (removals) o*HrMcor^covor^coo^OTHr\iroTrinvo c>c^(^(^crt(^o>o^c^(^ooooooooooTHHHHHHH <7>a>cr>c^cr»cr»c^cn<^c^ooooooooooooooooo ¦r-* ¦»-< -»H -r-t r-i r-i *-4 *-4 •*-« T-i f\i CM fM fM CM Cvl fM CM C\1 Csl fNJ Csl CM fNJ OJ fNl Csl Table ES-4: Recent Trends in U.S. Greenhouse Gas Emissions and Sinks by Chapter/IPCC Sector (MMT CO2 Eq.) Chapter/IPCC Sector 1990 2005 2012 2013 2014 2015 2016 Energy 5,340.2 6,295.7 5,527.6 5,691.1 5,739.1 5,596.0 5,476.4 Fossil Fuel Combustion 4,755.8 5,759.1 5,029.8 5,162.3 5,206.1 5,059.3 4,976.7 Natural Gas Systems 223.4 182.5 181.2 185.6 191.2 190.8 188.8 Non-Energy Use of Fuels 119.6 141.7 113.3 133.2 127.8 135.1 121.0 Petroleum Systems 51.7 51.7 61.0 68.5 73.9 77.4 64.8 Coal Mining 96.5 64.1 66.5 64.6 64.6 61.2 53.8 Stationary Combustion 19.8 25.4 24.1 27.2 27.7 25.8 25.6 Mobile Combustion 51.3 45.0 27.8 25.7 23.6 21.9 20.8 Incineration of Waste 8.4 12.9 10.7 10.7 10.9 11.0 11.0 Abandoned Oil and Gas Wells 6.5 6.9 7.0 7.0 7.1 7.2 7.1 Abandoned Underground Coal Mines 7.2 6.6 6.2 6.2 6.3 6.4 6.7 Industrial Processes and Product Use 340.5 354.2 361.6 364.7 380.2 378.8 375.7 Substitution of Ozone Depleting Substances 0.3 99.8 150.4 154.8 161.4 168.6 173.9 Iron and Steel Production & Metallurgical Coke Production 101.5 68.1 55.5 53.4 58.2 47.7 42.2 ES-18 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2016 ------- Cement Production 33.5 46.2 35.3 36.4 39.4 39.9 39.4 Petrochemical Production 21.4 26.9 26.6 26.5 26.6 28.2 27.6 Lime Production 11.7 14.6 13.8 14.0 14.2 13.3 13.3 Other Process Uses of Carbonates 4.9 6.3 8.0 10.4 11.8 11.2 11.2 Ammonia Production 13.0 9.2 9.4 10.0 9.6 10.6 11.2 Nitric Acid Production 12.1 11.3 10.5 10.7 10.9 11.6 10.2 Adipic Acid Production 15.2 7.1 5.5 3.9 5.4 4.3 7.0 Semiconductor Manufacture 3.6 4.7 4.4 4.0 4.9 5.0 5.0 Carbon Dioxide Consumption 1.5 1.4 4.0 4.2 4.5 4.5 4.5 Electrical Transmission and Distribution 23.1 8.3 4.6 4.5 4.6 4.2 4.3 N2O from Product Uses 4.2 4.2 4.2 4.2 4.2 4.2 4.2 Urea Consumption for Non- Agricultural Purposes 3.8 3.7 4.4 4.1 1.5 4.2 4.0 HCFC-22 Production 46.1 20.0 5.5 4.1 5.0 4.3 2.8 Aluminum Production 28.3 7.6 6.4 6.2 5.4 4.8 2.7 Caprolactam, Glyoxal, and Glyoxylic Acid Production 1.7 2.1 2.0 2.0 2.0 2.0 2.0 Ferroalloy Production 2.2 1.4 1.9 1.8 1.9 2.0 1.8 Soda Ash Production 1.4 1.7 1.7 1.7 1.7 1.7 1.7 Titanium Dioxide Production 1.2 1.8 1.5 1.7 1.7 1.6 1.6 Glass Production 1.5 1.9 1.2 1.3 1.3 1.3 1.3 Magnesium Production and Processing 5.2 2.7 1.7 1.5 1.1 1.0 1.1 Phosphoric Acid Production 1.5 1.3 1.1 1.1 1.0 1.0 1.0 Zinc Production 0.6 1.0 1.5 1.4 1.0 0.9 0.9 Lead Production 0.5 0.6 0.5 0.5 0.5 0.5 0.5 Silicon Carbide Production and Consumption 0.4 0.2 0.2 0.2 0.2 0.2 0.2 Agriculture 489.2 520.0 519.8 543.1 539.8 566.9 562.6 Agricultural Soil Management 250.5 253.5 247.9 276.6 274.0 295.0 283.6 Enteric Fermentation 164.2 168.9 166.7 165.5 164.2 166.5 170.1 Manure Management 51.1 72.9 83.2 80.8 80.4 84.0 85.9 Rice Cultivation 16.0 16.7 11.3 11.5 12.7 12.3 13.7 Urea Fertilization 2.4 3.5 4.3 4.4 4.5 4.9 5.1 Liming 4.7 4.3 6.0 3.9 3.6 3.8 3.9 Field Burning of Agricultural Residues 0.3 0.3 0.4 0.4 0.4 0.4 0.4 Waste 199.3 156.4 140.4 136.7 136.5 135.6 131.5 Landfills 179.6 132.7 117.0 113.3 112.7 111.7 107.7 Wastewater Treatment 19.1 20.2 19.7 19.6 19.8 20.0 19.8 Composting 0.7 3.5 3.7 3.9 4.0 4.0 4.0 Total Emissions3 6,369.2 7,326.4 6,549.4 6,735.6 6,795.6 6,677.3 6,546.2 Land Use, Land-Use Change, and Forestry (819.6) (731.1) (753.5) (735.8) (740.4) (695.2) (716.8) Forest land (784.3) (730.0) (723.3) (733.3) (731.7) (709.9) (714.2) Cropland 2.4 (0.7) 1.3 11.9 11.2 16.8 13.8 Grassland 13.8 25.3 0.8 18.5 14.7 33.6 21.0 Wetlands (4.0) (5.3) (4.1) (4.1) (4.1) (4.1) (4.2) Settlements (47.6) (20.5) (28.3) (28.8) (30.5) (31.5) (33.3) Net Emission (Sources and Sinks)b 5,549.6 6,595.3 5,795.9 5,999.9 6,055.2 5,982.1 5,829.3 Notes: Total emissions presented without LULUCF. Net emissions presented with LULUCF. a Total emissions without LULUCF. b Total emissions with LULUCF. Notes: Totals may not sum due to independent rounding. Parentheses indicate negative values or sequestration. 1 Energy 2 The Energy chapter contains emissions of all greenhouse gases resulting from stationary and mobile energy 3 activities including fuel combustion and fugitive fuel emissions, and the use of fossil fuels for non-energy purposes. Executive Summary ES-19 ------- 1 Energy-related activities, primarily fossil fuel combustion, accounted for the vast majority of U.S. CO2 emissions for 2 the period of 1990 through 2016. 3 In 2016, approximately 81 percent of the energy used in the United States (on a Btu basis) was produced through the 4 combustion of fossil fuels. The remaining 19 percent came from other energy sources such as hydropower, bio mass, 5 nuclear, wind, and solar energy (see Figure ES-13). 6 Energy-related activities are also responsible for CH4 and N20 emissions (43 percent and 10 percent of total U.S. 7 emissions of each gas, respectively). Overall, emission sources in the Energy chapter account for a combined 83.7 8 percent of total U.S. greenhouse gas emissions in 2016. 9 Figure ES-13: 2016 U.S. Energy Consumption by Energy Source (Percent) 12 The Industrial Processes and Product Use (IPPU) chapter includes greenhouse gas emissions occurring from 13 industrial processes and from the use of greenhouse gases in products. 14 In many cases, greenhouse gas emissions are produced as the byproducts of many non-energy-related industrial 15 activities. For example, industrial processes can chemically transform raw materials, which often release waste gases 16 such as CO2, CH4, N2O, and fluorinated gases (e.g., HFC-23). These processes include iron and steel production and 17 metallurgical coke production, cement production, lime production, other process uses of carbonates (e.g., flux 18 stone, flue gas desulfurization, and glass manufacturing), ammonia production and urea consumption, petrochemical 19 productioa aluminum production. HCFC-22 production, soda ash production and use, titanium dioxide production. 20 ferroalloy production, glass productioa zinc production, phosphoric acid production, lead productioa silicon 21 carbide production and consumption, nitric acid productioa adipic acid production, and caprolactam production. 22 Industrial manufacturing processes and use by end-consumers also release HFCs, PFCs, SFg, and NF3 and other 23 fluorinated compounds. In addition to the use of HFCs and some PFCs as ODS substitutes, HFCs, PFCs, SF(, NF3, 24 and other fluorinated compounds are employed and emitted by a number of other industrial sources in the United 25 States. These industries include semiconductor manufacture, electric power transmission and distributioa and 26 magnesium metal production and processing. In additioa !vO is used in and emitted by semiconductor 27 manufacturing and anesthetic and aerosol applications, and CO2 is consumed and emitted through various end-use 28 applications. Overall, emission sources in the Industrial Process and Product Use chapter account for 5.7 percent of 29 U.S. greenhouse gas emissions in 2016. Nuclear Electric Power 8.6% Renewable En 10.4% Petroleum 36.9% 10 11 Industrial Processes and Product Use ES-20 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2016 ------- 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 Agriculture The Agriculture chapter contains anthropogenic emissions from agricultural activities (except fuel combustion, which is addressed in the Energy chapter, and some agricultural CO2 fluxes, which are addressed in the Land Use, Land-Use Change, and Forestry chapter). 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. In 2016, agricultural activities were responsible for emissions of 562.6 MMT CO2 Eq., or 8.6 percent of total U.S. greenhouse gas emissions. Methane, N20, and CO2 were the primary greenhouse gases emitted by agricultural activities. Methane emissions from enteric fermentation and manure management represented approximately 25.9 percent and 10.3 percent of total CH4 emissions from anthropogenic activities, respectively, in 2016. Agricultural soil management activities, such as application of synthetic and organic fertilizers, deposition of livestock manure, and growing N-fixing plants, were the largest source of U.S. N20 emissions in 2016, accounting for 76.9 percent. 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 2-11 and Table 2-7 illustrate agricultural greenhouse gas emissions by source. Land Use, Land-Use Change, and Forestry The Land Use, Land-Use Change, and Forestry (LULUCF) chapter contains emissions of CH4 and N20, and emissions and removals of CO2 from managed lands in the United States. Overall, managed land is a net sink for CO2 (C sequestration) in the United States. The primary drivers of fluxes on managed lands include, for example, forest management practices, tree planting in urban areas, the management of agricultural soils, landfilling of yard trimmings and food scraps, and activities that cause changes in C stocks in coastal wetlands. The main drivers for forest C sequestration include forest growth and increasing forest area, as well as a net accumulation of C stocks in harvested wood pools. The net sequestration in Settlements Remaining Settlements, which occurs predominantly from urban forests and landfilled yard trimmings and food scraps, is a result of net tree growth and increased urban forest size, as well as long-term accumulation of yard trimmings and food scraps carbon in landfills. The LULUCF sector in 2016 resulted in a net increase in C stocks (i.e., net CO2 removals) of 754.9 MMT CO2 Eq. (Table ES-5).19 This represents an offset of 11.5 percent of total (i.e., gross) greenhouse gas emissions in 2016. Emissions of CH4 and N2O from LULUCF activities in 2016 are 38.1 MMT CO2 Eq. and represent 0.6 percent of total greenhouse gas emissions.20 Between 1990 and 2016, total C sequestration in the LULUCF sector decreased by 9.1 percent, primarily due to a decrease in the rate of net C accumulation in forests and Cropland Remaining Cropland, as well as an increase in CO2 emissions from Land Converted to Settlements. Forest fires were the largest source of CH4 emissions from LULUCF in 2016, totaling 18.5 MMT CO2 Eq. (740 kt of CH4). Coastal Wetlands Remaining Coastal Wetlands resulted in CH4 emissions of 3.6 MMT CO2 Eq. (143 kt of CH4). Grassland fires resulted in CH4 emissions of 0.3 MMT CO2 Eq. (11 kt of CH4). Peatlands Remaining Peatlands, Land Converted to Wetlands, and Drained Organic Soils resulted in CH4 emissions of less than 0.05 MMT CO2 Eq. each. Forest fires were also the largest source of N20 emissions from LULUCF in 2016, totaling 12.2 MMT CO2 Eq. (41 kt of N20). Nitrous oxide emissions from fertilizer application to settlement soils in 2016 totaled to 2.5 MMT CO2 Eq. (8 kt of N20). Additionally, the application of synthetic fertilizers to forest soils in 2016 resulted in N20 emissions of 0.5 MMT CO2 Eq. (2 kt of N20). Grassland fires resulted in N20 emissions of 0.3 MMT CO2 Eq. (1 kt 19 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. 20 LULUCF emissions include the CH4 and N2O 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 N2O emissions from Forest Soils and Settlement Soils. Executive Summary ES-21 ------- 1 of N2O). Coastal Wetlands Remaining Coastal Wetlands and Drained Organic Soils resulted in N2O emissions of 2 0.1 MMT CO2 Eq. each (less than 0.5 kt of N2O). Peatlands Remaining Peatlands resulted U1N2O emissions of less 3 than 0.05 MMT C02 Eq. 4 Carbon dioxide removals from C stock changes are presented in Table ES-5 along with CH4 and N20 emissions for 5 LULUCF source categories. 6 Table ES-5: U.S. Greenhouse Gas Emissions and Removals (Net Flux) from Land Use, Land- 7 Use Change, and Forestry (MMT CO2 Eq.) Gas/Land-Use Category 1990 2005 2012 2013 2014 2015 2016 Carbon Stock Change3 (830.2) (754.2) (779.5) (755.0) (760.0) (733.4) (754.9) Forest Land Remaining Forest Land (697.7) (664.6) (666.9) (670.9) (669.3) (666.2) (670.5) Land Converted to Forest Land (92.0) (81.6) (74.9) (74.9) (75.0) (75.0) (75.0) Cropland Remaining Cropland (40.9) (26.5) (21.4) (11.4) (12.0) (6.3) (9.9) Land Converted to Cropland 43.3 25.9 22.7 23.3 23.2 23.2 23.8 Grassland Remaining Grassland (4.2) 5.5 (20.8) (3.7) (7.5) 9.6 (1.6) Land Converted to Grassland 17.9 19.2 20.4 21.9 21.5 23.3 22.0 Wetlands Remaining Wetlands (7.6) (8.9) (7.7) (7.8) (7.8) (7.8) (7.9) Land Converted to Wetlands (+) (+) (+) (+) (+) (+) (+) Settlements Remaining Settlements (86.2) (91.4) (99.2) (99.8) (101.2) (102.2) (103.7) Land Converted to Settlements 37.2 68.4 68.3 68.3 68.2 68.1 68.0 CH4 6.7 13.3 15.0 10.9 11.2 22.4 22.4 Forest Land Remaining Forest Land: Forest Fires 3.2 9.4 10.8 7.2 7.2 18.5 18.5 Wetlands Remaining Wetlands: Coastal Wetlands Remaining Coastal Wetlands 3.4 3.5 3.5 3.6 3.6 3.6 3.6 Grassland Remaining Grassland: Grassland Fires 0.1 0.3 0.6 0.2 0.4 0.3 0.3 Forest Land Remaining Forest Land: Drained Organic Soils + + + + + + + Land Converted to Wetlands: Land Converted to Coastal Wetlands + + + + + + + Wetlands Remaining Wetlands: Peatlands Remaining Peatlands + + + + + + + N2O 3.9 9.7 11.1 8.3 8.4 15.8 15.7 Forest Land Remaining Forest Land: Forest Fires 2.1 6.2 7.1 4.8 4.7 12.2 12.2 Settlements Remaining Settlements: Settlement Soilsb 1.4 2.5 2.7 2.6 2.6 2.5 2.5 Forest Land Remaining Forest Land: Forest Soilsc 0.1 0.5 0.5 0.5 0.5 0.5 0.5 Grassland Remaining Grassland: Grassland Fires 0.1 0.3 0.6 0.2 0.4 0.3 0.3 Wetlands Remaining Wetlands: Coastal Wetlands Remaining Coastal Wetlands 0.1 0.2 0.1 0.1 0.1 0.1 0.1 Forest Land Remaining Forest Land: Drained Organic Soils 0.1 0.1 0.1 0.1 0.1 0.1 0.1 Wetlands Remaining Wetlands: Peatlands Remaining Peatlands + + + + + + + LULUCF Emissions'1 10.6 23.0 26.1 19.2 19.6 38.2 38.1 LULUCF Carbon Stock Change3 (830.2) (754.2) (779.5) (755.0) (760.0) (733.4) (754.9) LULUCF Sector Net Totale (819.6) (731.1) (753.5) (735.8) (740.4) (695.2) (716.8) + Absolute value does not exceed 0.05 MMT CO2 Eq. a 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. b Estimates include emissions from N fertilizer additions on both Settlements Remaining Settlements and Land Converted to Settlements. c Estimates include emissions from N fertilizer additions on both Forest Land Remaining Forest Land and Land Converted to Forest Land. ES-22 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2016 ------- d LULUCF emissions include the CH4 and N2O 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 N2O emissions from Forest Soils and Settlement Soils. e The LULUCF Sector Net Total is the net sum of all CH4 and N2O emissions to the atmosphere plus net carbon stock changes. Notes: Totals may not sum due to independent rounding. Parentheses indicate net sequestration. 1 Waste 2 The Waste chapter contains emissions from waste management activities (except incineration of waste, which is 3 addressed in the Energy chapter). Landfills were the largest source of anthropogenic greenhouse gas emissions in the 4 Waste chapter, accounting for 81.9 percent of this chapter's emissions, and 16.4 percent of total U.S. CH4 5 emissions.21 Additionally, wastewater treatment accounts for 15.1 percent of Waste emissions, 2.3 percent of U.S. 6 CH4 emissions, and 1.3 percent of U.S. N20 emissions. Emissions of CH4 and N20 from composting are also 7 accounted for in this chapter, generating emissions of 2.1 MMT CO2 Eq. and 1.9 MMT CO2 Eq., respectively. 8 Overall, emission sources accounted for in the Waste chapter generated 2.0 percent of total U.S. greenhouse gas 9 emissions in 2016. 10 ES.4 Other Information 11 Emissions by Economic Sector 12 Throughout the Inventory of U.S. Greenhouse Gas Emissions and Sinks report, emission estimates are grouped into 13 five sectors (i.e., chapters) defined by the IPCC: Energy; Industrial Processes and Product Use; Agriculture; 14 LULUCF; and Waste. While it is important to use this characterization for consistency with UNFCCC reporting 15 guidelines and to promote comparability across countries, it is also useful to characterize emissions according to 16 commonly used economic sector categories: residential, commercial, industry, transportation, electric power, 17 agriculture, and U.S. Territories. 18 Figure ES-14 shows the trend in emissions by economic sector from 1990 to 2016, and Table ES-6 summarizes 19 emissions from each of these economic sectors. 21 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. Executive Summary ES-23 ------- 1 2 3 4 5 6 7 8 9 10 11 12 13 14 Figure ES-14: U.S. Greenhouse Gas Emissions Allocated to Economic Sectors (MMT CO2 Eq.) 2,500- Electric Power Industry 2,000- Transportation iS" 1/500- Industry 1,000- Agriculture Commercial (Red) 500- Residential (Blue) UD cr ------- 1 The commercial and residential sectors accounted for 6.3 percent and 5.4 percent of emissions, respectively, and 2 U.S. Territories accounted for 0.7 percent of emissions; emissions from these sectors primarily consisted of CO2 3 emissions from fossil fuel combustion. CO2 was also emitted and sequestered by a variety of activities related to 4 forest management practices, tree planting in urban areas, the management of agricultural soils, landfilling of yard 5 trimmings, and changes in C stocks in coastal wetlands. 6 Electricity is ultimately used in the economic sectors described above. Table ES-7 presents greenhouse gas 7 emissions from economic sectors with emissions related to electric power distributed into end-use categories (i.e., 8 emissions from electric power are allocated to the economic sectors in which the electricity is used). To distribute 9 electricity emissions among end-use sectors, emissions from the source categories assigned to electric power were 10 allocated to the residential, commercial, industry, transportation, and agriculture economic sectors according to retail 11 sales of electricity (EIA 2017 and Duffield 2006). These source categories include CO2 from fossil fuel combustion 12 and the use of limestone and dolomite for flue gas desulfurization, CO2 and N20 from incineration of waste, CH4 13 and N20 from stationary sources, and SF6 from electrical transmission and distribution systems. 14 When emissions from electricity use are distributed among these sectors, industrial activities and transportation 15 account for the largest shares of U.S. greenhouse gas emissions (28.9 percent and 28.5 percent, respectively) in 16 2016. The residential and commercial sectors contributed the next largest shares of total U.S. greenhouse gas 17 emissions in 2016. Emissions from these sectors increase substantially when emissions from electricity are included, 18 due to their relatively large share of electricity use (e.g., lighting, appliances). In all sectors except agriculture, CO2 19 accounts for at least 81 percent of greenhouse gas emissions, primarily from the combustion of fossil fuels. 20 Figure ES-15 shows the trend in these emissions by sector from 1990 to 2016. 21 Table ES-7: U.S. Greenhouse Gas Emissions by Economic Sector with Electricity-Related 22 Emissions Distributed (MMT CO2 Eq.) Implied Sectors I'm 2005 2012 2013 2014 2015 2016 Industry 2,321.9 2,225.4 1,977.8 2,040.5 2,033.7 1,985.1 1,888.8 Transportation 1,528.2 1,977.3 1,752.8 1,761.9 1,797.2 1,812.5 1,867.4 Commercial 978.2 1,218.7 1,099.9 1,128.2 1,139.8 1,108.9 1,063.1 Residential 951.2 1,240.4 1,055.7 1,120.6 1,142.2 1,067.3 1,028.4 Agriculture 606.5 614.8 636.4 636.1 656.8 651.9 U.S. Territories 58.1 48.5 48.1 46.6 46.6 46.6 Total Emissions 6,3(t').2 7,326.4 6,549.4 6,735.6 6,795.6 6,677.3 6,546.2 LULUCF Sector Net Total3 (8I'U.) (731.1) (753.5) (735.8) (740.4) (695.2) (716.8) Net Emissions (Sources and Sinks) 5,549.6 6,595.3 5.795.9 5,999.9 6,055.2 5,982.1 5,829.3 a The LULUCF Sector Net Total is the net sum of all CH4 and N2O emissions to the atmosphere plus 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. 23 Executive Summary ES-25 ------- 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 Figure ES-15: U.S. Greenhouse Gas Emissions with Electricity-Related Emissions Distributed to Economic Sectors (MMT CO2 Eq.) 2,500 Industry 2,000 Transportation lS" ~ 1,500 O u H s: Commercial (Red) 1,000 Residential (Blue) Agriculture 500 o o © s © s © c - CO © CTv © © r-J in - o ro in 1—1 CTv Cv -*—1 © © m o = t < o 0 o © o o © Box ES-5: Recent Trends in Various U.S. Greenhouse Gas Emissions-Related Data Total emissions can be compared to other economic and social indices to highlight changes over time. These comparisons include: (1) emissions per unit of aggregate energy use, because energy-related activities are the largest sources of emissions; (2) emissions per unit of fossil fuel consumption because almost all energy-related emissions involve the combustion of fossil fuels; (3) emissions per unit of electricity use, because the electric power industry—utilities and non-utilities combined—was the second largest source of U.S. greenhouse gas emissions in 2016; (4) emissions per unit of total gross domestic product as a measure of national economic activity; and (5) emissions per capita. Table ES-8 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 grown at an average annual rate of 0.1 percent since 1990. This rate is slightly slower than that for total energy use and fossil fuel consumption and much slower than that for electricity use, overall gross domestic product (GDP), and national population (see Figure ES-16). These trends vary relative to 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 1.0 percent since 2005. Total energy use and fossil fuel consumption have also decreased at slower rates than emissions since 2005, while electricity use, GDP, and national population continued to increase. Table ES-8: Recent Trends in Various U.S. Data (Index 1990 = 100) Variable 1990 2005 2012 2013 2014 2015 2016 Avg. Annual Change since 1990 Avg. Annual Change since 2005a Greenhouse Gas Emissions'5 100 115 103 106 107 105 103 0.1% -1.0% Energy Usec 100 118 112 115 117 115 116 0.6% -0.2% Fossil Fuel Consumption0 100 119 107 110 111 110 109 0.4% -0.7% Electricity Usec 100 134 135 136 138 137 136 1.2% 0.1% GDPd 100 159 171 174 179 184 187 2.4% 1.5% Population6 100 118 125 126 127 128 129 1.0% 0.8% a Average annual growth rate ES-26 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2016 ------- 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 b GWP-weighted values c Energy content-weighted values (EIA 2017) d Gross Domestic Product in chained 2009 dollars (BEA 2017) e U.S. Census Bureau (2017) Figure ES-16: U.S. Greenhouse Gas Emissions Per Capita and Per Dollar of Gross Domestic Product (GDP) 200 190- Real GDP 180- 170- 160- — 15°" § 140- II 8 13°" Ol CS 120- 85 ~o 100— Population 110- Emissions per capita 90- 80- 70- Emissions per GDP 60- hv co on o § o o o o vO ro fM ro 8 S o o c c o Source: BEA (2017), U.S. Census Bureau (2017), and emission estimates in this report. Key Categories The 2006IPCC Guidelines (IPCC 2006) defines a key category as a "[category] that is prioritized within the national inventory system because its estimate lias a significant influence on a country's total inventory of greenhouse gases in terms of the absolute level, the trend, or the uncertainty in emissions and removals."23 By definition key categories are sources or sinks that have the greatest contribution to the absolute overall level of national emissions in any of the years covered by the time series. In addition when an entire time series of emission estimates is prepared, a thorough investigation of key categories must also account for the influence of trends of individual source and sink categories. Finally, a qualitative evaluation of key categories should be performed, in order to capture any key categories that were not identified in either of the quantitative analyses. Figure ES-17 presents 2016 emission estimates for the key categories as defined by a level analysis including the LULUCF sector (i.e., the absolute value of the contribution of each source or sink category to the total inventory level). The UNFCCC reporting guidelines request that key category analyses be reported at an appropriate level of disaggregation, which may lead to source and sink category names which differ from those used elsewhere in the Inventory report. For more information regarding key categories, including a complete list of categories accounting for the influence of trends of individual source and sink categories, see Section 1.4 - Key Categories and Annex 1. 23 See Chapter 4 "Methodological Choice and Identification of Key Categories" in IPCC (2006). See . Executive Summary ES-27 ------- 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 Figure ES-17: 2016 Key Categories (MMT CO2 Eq.) COz Emissions from Mobile Combustion: Road COz Emissions from Stationary Combustion - Coal - Electricity Generation Net COz Emissions in Forest Land Remaining Forest Land" COz Emissions from Stationary Combustion - Gas - Electricity Generation CO* Emissions from Stationary Combustion - Gas - Industrial COz Emissions from Stationary Combustion - Oil - Industrial COz Emissions from Stationary Combustion - Gas - Residential Direct NzO Emissions from Agricultural Soil Management Emissions from Substitutes for Ozone Depleting Substances COz Emissions from Stationary Combustion - Gas - Commercial CH4 Emissions from Enteric Fermentation COz Emissions from Mobile Combustion: Aviation CH» Emissions from Natural Gas Systems CO: Emissions from Non-Energy Use of Fuels CH» Emissions from Landfills Net COz Emissions in Settlements Remaining Settlements3 COz Emissions from Mobile Combustion: Other Net COz Emissions in Land Converted to Forest Land3 Net COz Emissions in Land Converted to Settlements3 CH< Emissions from Manure Management COz Emissions from Stationary Combustion - Coal - Industrial COz Emissions from Stationary Combustion - Oil - Residential COz Emissions from Stationaiy Combustion - Oil - Commercial Fugitive Emissions from Coal Mining Indirect NzO Emissions from Applied Nitrogen COz Emissions from Iron and Steel Production & Metallurgical Coke Production COz Emissions from Mobile Combustion: Marine COz Emissions from Cement Production CH< Emissions from Petroleum Systems COz Emissions from Stationary Combustion - Oil - U.S. Territories COz Emissions from Petrochemical Production COz Emissions from Natural Gas Systems COz Emissions from Petroleum Systems Net COz Emissions in Land Converted to Cropland3 Net COz Emissions in Land Converted to Grassland3 CH« Emissions from Forest Firesb NzO Emissions from Forest Fires11 Net COz Emissions in Cropland Remaining Cropland3 Net COz Emissions in Grassland Remaining Grassland3 Key Categories as a Portion of All Emissions 0 200 400 600 800 1,000 1,200 1,400 MMT COz Eq. a Hie absolute values of net CO2 emissions from LULUCF are presented in this figure but reported separately from gross emissions totals. Refer to Table ES-5 for a breakout of emissions and removals for LULUCF by gas and source category. b N011-CO2 emissions from Forest Fires are presented in this figure but reported separately from gross emissions totals. Refer to Table ES-5 for a breakout of emissions and removals for LULUCF by gas and source category. Note: For a complete discussion of the key category analysis, see Annex 1. Blue bars indicate either an Approach 1, or Approach 1 and Approach 2 level assessment key category. Gray bars indicate solely an Approach 2 level assessment key category. Quality Assurance and Quality Control (QA/QC) The United States seeks to continually improve the quality , transparency , and credibility of the Inventory of U.S. Greenhouse Gas Emissions and Sinks. To assist in these efforts, the United States implemented a systematic approach to QA/QC. The procedures followed for the Inventory have been formalized in accordance with the Quality Assurance/Quality Control and Uncertainty Management Plan (QA/QC Management Plan) for the Inventory, and the UNFCCC reporting guidelines. The QA process includes expert and public reviews for both the Inventory estimates and the Inventory report. Uncertainty Analysis of Emission Estimates Uncertainty estimates are an essential element of a complete inventory of greenhouse gas emissions and removals, because they help to prioritize future work and improve overall quality. Some of the current estimates, such as those for CO2 emissions from energy-related activities, are considered to have low uncertainties. This is because the amount of CO2 emitted from energy-related 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 is believed to be relatively small. For some other categories of emissions, however, a lack of data or an incomplete understanding of how emissions are generated increases the uncertainty or systematic ES-28 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2016 ------- 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 error associated with the estimates presented. Recognizing the benefit of conducting an uncertainty analysis, the UNFCCC reporting guidelines follow the recommendations of the 2006IPCC Guidelines (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 provided in accordance with UNFCCC reporting guidelines, a qualitative discussion of uncertainty is presented for all source and sink categories. Within the discussion of each emission source, specific factors affecting the uncertainty surrounding the estimates are discussed. Box ES-6: Recalculations of Inventory Estimates Each year, emission and sink estimates are recalculated and revised for all years in the Inventory of U.S. Greenhouse Gas Emissions and Sinks, as attempts are made to improve both the analyses themselves, through the use of better methods or data, and the overall usefulness of the report. In this effort, the United States follows the 2006 IPCC Guidelines (IPCC 2006), which states, "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 lias increased; new inventory methods become available; and for correction of errors." 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. In each Inventory report, the results of all methodology changes and historical data updates are presented in the Recalculations and Improvements chapter of this report; detailed descriptions of each recalculation are contained within each source's description contained in the report, if applicable. In general, when methodological changes have been implemented, the entire time series (in the case of the most recent Inventory report, 1990 through 2014) has been recalculated to reflect the change, per the 2006 IPCC Guidelines (IPCC 2006). Changes in historical data are generally the result of changes in statistical data supplied by other agencies. References for the data are provided for additional information. Significant changes made since the previous report are also noted in Box ES-3. Executive Summary ES-29 ------- 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 i. Introduction This report presents estimates by the United States government of U.S. anthropogenic greenhouse gas emissions and sinks for the years 1990 through 2016. A summary of these estimates is provided in Table 2-1 and Table 2-2 by gas and source category in the Trends in Greenhouse Gas Emissions chapter. The emission estimates in these tables are presented on both a full molecular mass basis and on a Global Warming Potential (GWP) weighted basis1 in order to show the relative contribution of each gas to global average radiative forcing. This report also discusses the methods and data used to calculate these emission estimates. In 1992, the United States signed and ratified the United Nations Framework Convention on Climate Change (UNFCCC). As stated in Article 2 of the UNFCCC, "The ultimate objective of this Convention and any related legal instruments that the Conference of the Parties may adopt is to achieve, in accordance with the relevant provisions of the Convention, stabilization of greenhouse gas concentrations in the atmosphere at a level that would prevent dangerous anthropogenic interference with the climate system. Such a level should be achieved within a time-frame sufficient to allow ecosystems to adapt naturally to climate change, to ensure that food production is not threatened and to enable economic development to proceed in a sustainable manner."2'3 Parties to the Convention, by ratifying, "shall develop, periodically update, publish and make available... national inventories of anthropogenic emissions by sources and removals by sinks of all greenhouse gases not controlled by the Montreal Protocol, using comparable methodologies.. ."4 The United States views this report as an opportunity to fulfill these commitments under the UNFCCC. 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 2014). Under Working Group 1 of the IPCC, nearly 140 scientists and national experts from more than thirty countries collaborated in the creation of the Revised 1996 IPCC Guidelines for National Greenhouse Gas Inventories (IPCC/UNEP/OECD/IEA 1997) to ensure that the emission inventories submitted to the UNFCCC are consistent and comparable between nations. The IPCC Good Practice Guidance and Uncertainty Management in National Greenhouse Gas Inventories and the IPCC Good Practice Guidance for Land Use, Land-Use Change, and Forestry further expanded upon the methodologies in the Revised 1996 IPCC Guidelines. In 2006, the IPCC accepted the 2006 Guidelines for National Greenhouse Gas Inventories at its Twenty-Fifth Session (Mauritius, April 2006). The 2006 IPCC Guidelines built 1 More information provided in "Global Warming Potentials" section of this chapter on the use of IPCC Fourth Assessment Report (AR4) GWP values. 2 The term "anthropogenic," in this context, refers to greenhouse gas emissions and removals that are a direct result of human activities or are the result of natural processes that have been affected by human activities (IPCC 2006). 3 Article 2 of the Framework Convention on Climate Change published by the UNEP/WMO Information Unit on Climate Change (UNEP/WMO 2000). See . 4 Article 4(1)(a) of the United Nations Framework Convention on Climate Change (also identified in Article 12). Subsequent decisions by the Conference of the Parties elaborated the role of Annex I Parties in preparing national inventories. See . Introduction 1-1 ------- 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 upon the previous bodies of work and include new sources and gases ".. .as well as updates to the previously published methods whenever scientific and technical knowledge have improved since the previous guidelines were issued." The UNFCCC adopted the 2006IPCC Guidelines as the standard methodological approach for Annex I countries at the Nineteenth Conference of the Parties (Warsaw, November 11-23, 2013). This report presents information in accordance with these guidelines. Overall, this Inventory of anthropogenic greenhouse gas emissions and sinks provides a common and consistent mechanism through which Parties to the UNFCCC can estimate emissions and compare the relative contribution of individual sources, gases, and nations to climate change. The Inventory provides a national estimate of sources and sinks for the United States, including all states and U.S. Territories.5 The structure of this report is consistent with the current UNFCCC Guidelines on Annual Inventories (UNFCCC 2014) for Parties included in Annex I of the Convention. Box 1-1: Methodological Approach for Estimating and Reporting U.S. Emissions and Removals In following the UNFCCC requirement under Article 4.1 to develop and submit national greenhouse gas emissions inventories, the gross emissions total presented in this report for the United States excludes emissions and removals from LULUCF. The net emissions total presented in this report for the United States includes emissions and removals from LULUCF. All emissions and removals are calculated using internationally-accepted methods consistent with the IPCC Guidelines.6 Additionally, the calculated emissions and removals in a given year for the United States are presented in a common manner in line with the UNFCCC reporting guidelines for the reporting of inventories under this international agreement.7 The use of consistent methods to calculate emissions and removals by all nations providing their inventories to the UNFCCC ensures that these reports are comparable. The report itself follows this standardized format, and provides an explanation of the IPCC methods used to calculate emissions and removals. On October 30, 2009, the U.S. Enviromnental Protection Agency (EPA) published a rule for the mandatory reporting of greenhouse gases from large greenhouse gas emissions sources in the United States. Implementation of 40 CFR Part 98 is referred to as the EPA's GHGRP. 40 CFR Part 98 applies to direct greenhouse gas emitters, fossil fuel suppliers, industrial gas suppliers, and facilities that inject CO2 underground for sequestration or other reasons.8 Reporting is at the facility level, except for certain suppliers of fossil fuels and industrial greenhouse gases. The GHGRP dataset and the data presented in this Inventory are complementary. The EPA's 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 emissions 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. The GHGRP will continue to enhance QA/QC procedures and assessment of uncertainties. EPA continues to analyze the data on an annual basis to improve the national estimates presented in this Inventory and uses that data for a number of categories consistent with IPCC guidance.9 EPA has integrated GHGRP information for several categories10 this year and also identifies other categories11 where EPA plans to integrate 5 U.S. Territories include American Samoa, Guam, Puerto Rico, U.S. Virgin Islands, Wake Island, and other U.S. Pacific Islands. 6 See . 7 See . 8 See . 9 See . 10 Energy Sector (Coal Mining, Stationary Combustion [Industrial Combustion Disaggregation], and Oil and Gas Systems); Industrial Processes and Product Use (Adipic Acid Production, Aluminum Production, Carbon Dioxide Consumption, Electrical Transmission and Distribution, HCFC-22 Production, Lime Production, Magnesium Production and Processing, ODS Substitutes, Nitric Acid Production, Petrochemical Production, Semiconductor Manufacture); and Waste (Landfills). 11 Industrial Process and Product Use (Ammonia Production, Cement Production, and Other Fluorinated Gas Production) 1-2 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2016 ------- 1 additional GHGRP data in the next edition of this report (see those categories Planned Improvement sections for 2 details). 3 4 1.1 Background Information 5 Sc 6 For over the past 200 years, the burning of fossil fuels such as coal and oil, deforestation, land-use changes, and 7 other activities have caused the concentrations of heat-trapping "greenhouse gases" to increase significantly in our 8 atmosphere (NOAA 2017). These gases in the atmosphere absorb some of the energy being radiated from the 9 surface of the Earth that would otherwise be lost to space, essentially acting like a blanket that makes the Earth's 10 surface warmer than it would be otherwise. 11 Greenhouse gases are necessary to life as we know it. Without greenhouse gases to create the natural heat-trapping 12 properties of the atmosphere, the planet's surface would be about 60 degrees Fahrenheit cooler than present 13 (USGCRP 2017). Carbon dioxide is also necessary for plant growth. With emissions from biological and geological 14 sources, there is a natural level of greenhouse gases that is maintained in the atmosphere. Human emissions of 15 greenhouse gases and subsequent changes in atmospheric concentrations alters the balance of energy transfers 16 between space and the earth system (IPCC 2013). A gauge of these changes is called radiative forcing, which is a 17 measure of a substance's total net effect on the global energy balance for which a positive number represents a 18 warming effect and a negative number represents a cooling effect (IPCC 2013). IPCC concluded in its most recent 19 scientific assessment report that it is extremely likely that human influences have been the dominant cause of 20 warming since the mid-20lh century (IPCC 2013). 21 As concentrations of greenhouse gases continue to increase in from man-made sources, the Earth's temperature is 22 climbing above past levels. The Earth's average land and ocean surface temperature has increased by about 1.2 to 23 1.9 degrees Fahrenheit since 1880. The last three decades have each been the warmest decade successively at the 24 Earth's surface since 1850 (IPCC 2013). Other aspects of the climate are also changing, such as rainfall patterns, 25 snow and ice cover, and sea level. If greenhouse gas concentrations continue to increase, climate models predict that 26 the average temperature at the Earth's surface is likely to increase from 0.5 to 8.6 degrees Fahrenheit above 1986 27 through 2005 levels by the end of this century, depending on future emissions and the responsiveness of the climate 28 system (IPCC 2013). 29 For further information on greenhouse gases, radiative forcing, and implications for climate change, see the recent 30 scientific assessment reports from the IPCC,12 the U.S. Global Change Research Program (USGCRP),13 and the 31 National Academies of Sciences, Engineering, and Medicine (NAS).14 32 Greenhouse Gases 33 Although the Earth's atmosphere consists mainly of oxygen and nitrogen, neither plays a significant role in 34 enhancing the greenhouse effect because both are essentially transparent to terrestrial radiation. The greenhouse 35 effect is primarily a function of the concentration of water vapor, carbon dioxide (CO2), methane (CH4), nitrous 36 oxide (N20), and other trace gases in the atmosphere that absorb the terrestrial radiation leaving the surface of the 37 Earth (IPCC 2013). 38 Naturally occurring greenhouse gases include water vapor, CO2, CH4, N20, and ozone (O3). Several classes of 39 halogenated substances that contain fluorine, chlorine, or bromine are also greenhouse gases, but they are, for the 12 See . 13 See < https://science2017.globalchange.gov/>. 14 See . Introduction 1-3 ------- 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 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 national greenhouse gas inventories.15 Some other fluorine-containing halogenated substances—hydrofluorocarbons (HFCs), perfluorocarbons (PFCs), sulfur hexafluoride (SF6), 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, including CH4) 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, absorb sunlight) and can play a role in affecting cloud formation and lifetime, as well as the radiative forcing of clouds and precipitation patterns. Comparatively, however, while the understanding of aerosols has increased in recent years, they still account for the largest contribution to uncertainty estimates in global energy budgets (IPCC 2013). Carbon dioxide, CH4, and N20 are continuously emitted to and removed from the atmosphere by natural processes on Earth. Anthropogenic activities, 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 CO2 CH4 N2O SF« CF4 Pre-industrial atmospheric concentration Atmospheric concentration Rate of concentration change Atmospheric lifetime (years) 280 ppm 404 ppma 2.3 ppm/yrf See footnote11 0.700 ppm 1.843 ppmb 7 ppb/yrf,B 12.4' 0.270 ppm 0.329 ppmc 0.8 ppb/yr® 121' Oppt 8.9 pptd 0.27 ppt/yr® 3,200 40 ppt 79 ppte 0.7 ppt/yr® 50,000 a The atmospheric CO2 concentration is the 2016 annual average at the Mauna Loa, HI station (NOAA/ESRL 2017a). The concentration in 2017 at Mauna Loa was 407 ppm. The global atmospheric CO2 concentration, computed using an average of sampling sites across the world, was 403 ppm in 2016. b The values presented are global 2016 annual average mole fractions (NOAA/ESRL 2017b). c The values presented are global 2016 annual average mole fractions (NOAA/ESRL 2017c). dThe values presented are global 2016 annual average mole fractions (NOAA/ESRL 2017d). e The 2011 CF4 global mean atmospheric concentration is from the Advanced Global Atmospheric Gases Experiment (IPCC 2013). f 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 7 ppb/year. B The rate of concentration change for CO2 and CH4 is the average rate of change between 2007 and 2016 (NOAA/ESRL 2017a). The rate of concentration change for N2O, SF6, and CF4 is the average rate of change between 2005 and 2011 (IPCC 2013). h For a given amount of carbon dioxide emitted, some fraction of the atmospheric increase in concentration is quickly absorbed by 15 Emissions estimates of CFCs, HCFCs, halons and other ozone-depleting substances are included in this document for informational purposes. 1-4 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2016 ------- 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 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. 1 This lifetime has been defined as an "adjustment time" that takes into account the indirect effect of the gas on its own residence time. Source: Pre-industrial atmospheric concentrations, atmospheric lifetime, and rate of concentration changes for CH4, N2O, SF6, and CF4 are from IPCC (2013). The rate of concentration change for CO2 is an average of the rates from 2011 through 2016 and has fluctuated between 1.9 to 3.0 ppm per year over this period (NOAA/ESRL 2017a). 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 GWPs, which are assigned to individual gases as a measure of their relative average global radiative forcing effect. Water Vapor (H20). 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, this has been determined to have a negligible effect on climate (IPCC 2013). 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 increase in other greenhouse gases, the total warming is greater than would happen in the absence of water vapor. Aircraft emissions of water vapor 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 2013). Carbon Dioxide (C02). 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 404 ppmv in 2016 a 44 percent increase (IPCC 2013; NOAA/ESRL 2017a).1617 The IPCC definitively states that "the increase of CO2 ... is caused by anthropogenic emissions from the use of fossil fuel as a source of energy and from land use and land use changes, in particular agriculture" (IPCC 2013). 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 Fifth Assessment Report, the IPCC stated "it is extremely likely that more than half of the observed increase in global average surface temperature from 1951 to 2010 was caused by the anthropogenic increase in greenhouse gas concentrations and other anthropogenic forcings together," of which CC^is the most important (IPCC 2013). Methane (CH4). 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. 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 CH4 have increased by about 163 percent since 1750, from a pre-industrial value of about 700 ppb to 1,843 ppb in 201618 although the rate of increase decreased to near zero in the early 2000s, and has recently increased again to about 5 ppb/year. The IPCC has estimated that slightly more than half of the current CH4 flux to the atmosphere is anthropogenic, from human activities such as agriculture, fossil fuel use, and waste disposal (IPCC 2007). 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 16 The pre-industrial period is considered as the time preceding the year 1750 (IPCC 2013). 17 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 2013). 18 This value is the global 2016 annual average mole fraction (NOAA/ESRL 2017b). Introduction 1-5 ------- 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 layer, a soil sink, and stratospheric reactions. Increasing emissions of CH4 reduce the concentration of OH, a feedback that increases the atmospheric lifetime of CH4 (IPCC 2013). 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. Nitrous Oxide (N20). Anthropogenic sources of N20 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 22 percent since 1750, from a pre-industrial value of about 270 ppb to 329 ppb in 2016,19 a concentration that has not been exceeded during the last 800 thousand years. Nitrous oxide is primarily removed from the atmosphere by the photolytic action of sunlight in the stratosphere (IPCC 2013). Ozone (03). Ozone is present in both the upper stratosphere,20 where it shields the Earth from harmful levels of ultraviolet radiation, and at lower concentrations in the troposphere,21 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 2013). The depletion of stratospheric ozone and its radiative forcing was expected to reach a maximum in about 2000 before starting to recover. The past increase in tropospheric ozone, which is also a greenhouse gas, is estimated to provide the fourth largest increase in direct radiative forcing since the pre-industrial era, behind CO2, black carbon, and CH4. Tropospheric ozone is produced from complex chemical reactions of volatile organic compounds (including 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 2013). Halocarbons, Sulfur Hexafluoride, and Nitrogen Trifluoride. 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 was placed on the production and importation of HCFCs by non-Article 522 countries, including the U.S., 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. Hydrofluorocarbons, PFCs, SF6, 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 also emitted as a byproduct of the HCFC-22 (chlorodifluoromethane) manufacturing process. 19 This value is the global 2016 annual average (NOAA/ESRL 2017c). 20 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. 21 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. 22 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. 1-6 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2016 ------- 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 Currently, they have a small aggregate radiative forcing impact, but it is anticipated that without further controls their contribution to overall radiative forcing will increase (IPCC 2013). An amendment to the Montreal Protocol was adopted in 2016 which includes obligations for Parties to phase down the production and consumption 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 2013). 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.23 Additionally, NOx emissions are also likely to decrease CH4 concentrations, thus having a negative radiative forcing effect (IPCC 2013). 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. 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 carbonaceous24 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 2013). 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 2013). 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, "despite the large uncertainty ranges on aerosol forcing, there is high confidence that aerosols have offset a substantial portion of GHG forcing" (IPCC 20 13).25 Although because they remain in the atmosphere for 23 NOx emissions injected higher in the stratosphere, primarily from fuel combustion emissions from high altitude supersonic aircraft, can lead to stratospheric ozone depletion. 24 Carbonaceous aerosols are aerosols that are comprised mainly of organic substances and forms of black carbon (or soot) (IPCC 2013). 25 The IPCC (2013) defines high confidence as an indication of strong scientific evidence and agreement in this statement. Introduction 1-7 ------- 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 only days to weeks, their concentrations respond rapidly to changes in emissions.26 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 2013). Black carbon also influences cloud development, but the direction and magnitude of this forcing is an area of active research. A global warming potential is a quantified measure of the globally averaged relative radiative forcing impacts of a particular greenhouse gas (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 2014). 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 million metric tons of CO2 equivalent (MMT CO2 Eq.).27 The relationship between kilotons (kt) of a gas and MMT CO2 Eq. can be expressed as follows: MMT CO2 Eq. = Million metric tons of CO2 equivalent kt = kilotons (equivalent to a thousand metric tons) GWP = Global warming potential MMT = Million metric tons 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 ±35 percent. Parties to the UNFCCC have also agreed to use GWPs based upon a 100-year time horizon, although other time horizon values are available. ... the global warming potential values used by Parties included in Annex I to the Convention (Annex I Parties) to calculate the carbon dioxide equivalence of anthropogenic emissions by sources and removals by sinks of greenhouse gases shall be those listed in the column entitled "Global warming potential for given time horizon " in table 2.14 of the errata to the contribution of Working Group I to the Fourth Assessment Report of the Intergovernmental Panel on Climate Change, based on the effects of greenhouse gases over a 100-year time horizon...28 Greenhouse gases with relatively long atmospheric lifetimes (e.g., CO2, CH4, N20, 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. 26 Volcanic activity can inject significant quantities of aerosol producing sulfur dioxide and other sulfur compounds into the stratosphere, which can result in a longer negative forcing effect (i.e., a few years) (IPCC 2013). 27 Carbon comprises 12/44ths of carbon dioxide by weight. 28 Framework Convention on Climate Change; Available online at: ; 31 January 2014; Report of the Conference of the Parties at its nineteenth session; held in Warsaw from 11 to 23 November 2013; Addendum; Part two: Action taken by the Conference of the Parties at its nineteenth session; Decision 24/CP. 19; Revision of the UNFCCC reporting guidelines on annual inventories for Parties included in Annex I to the Convention; p. 2. (UNFCCC 2014). Global Warming Potentials ( MMT \ Eq. = (kt of gas) x (GWP) x oqq J MMT CO- where, 1-8 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2016 ------- 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 Table 1-2: Global Warming Potentials and Atmospheric Lifetimes (Years) Used in this Report Gas Atmospheric Lifetime GWP1 CO2 See footnoteb 1 CH4c 12 25 N2O 114 298 HFC-23 270 14,800 HFC-32 4.9 675 HFC-125 29 3,500 HFC-134a 14 1,430 HFC-143a 52 4,470 HFC-152a 1.4 124 HFC-227ea 34.2 3,220 HFC-236fa 240 9,810 HFC-4310mee 15.9 1,640 CF4 50,000 7,390 C2F6 10,000 12,200 O J? O 2,600 8,860 CdFl4 3,200 9,300 SFo 3,200 22,800 NF3 740 17,200 a 100-year time horizon. b For a given amount of carbon dioxide emitted, some fraction of the atmospheric increase in concentration is quickly absorbed by the oceans and terrestrial vegetation, some fraction of the atmospheric increase will only slowly decrease over a number of years, and a small portion of the increase will remain for many centuries or more. 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 CO2 is not included. Source: (IPCC 2007) Box 1-2: The IPCC Fifth Assessment Report and Global Warming Potentials In 2014, the IPCC published its Fifth Assessment Report (AR5), which updated its comprehensive scientific assessment of climate change. Within the AR5 report, the GWP values of gases were revised relative to previous IPCC reports, namely the IPCC Second Assessment Report (SAR) (IPCC 1996), the IPCC Third Assessment Report (TAR) (IPCC 2001), and the IPCC Fourth Assessment Report (AR4) (IPCC 2007). Although the AR4 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 AR5, the IPCC applied an improved calculation of CO2 radiative forcing and an improved CO2 response function in presenting updated GWP values. Additionally, the atmospheric lifetimes of some gases have been recalculated, and updated background concentrations were used. In addition, the values for radiative forcing and lifetimes have been recalculated for a variety of halocarbons, and the indirect effects of methane on ozone have been adjusted to match more recent science. Table 1-3 presents the new GWP values, relative to those presented in the AR4 and using the 100-year time horizon common to UNFCCC reporting. For consistency with international reporting standards under the UNFCCC, official emission estimates are reported by the United States using AR4 GWP values, as required by the 2013 revision to the UNFCCC reporting guidelines for national inventories.29 All estimates provided throughout this report are also presented in unweighted units. For 29 See . Introduction 1-9 ------- 1 informational purposes, emission estimates that use GWPs from other IPCC Assessment Reports are presented in 2 detail in Annex 6.1 of this report. 3 Table 1-3: Comparison of 100-Year GWP values AR5 with Gas SAR AR4 AR5 feedbacks6 Comparison to AR4 SAR AR5 AR5 with feedbacksb CO2 1 1 1 1 NC NC NC CH4c 21 25 28 34 (4) 3 9 N2O 310 298 265 298 12 (33) 0 HFC-23 11,700 14,800 12,400 13,856 (3,100) (2,400) (944) HFC-32 650 675 677 817 (25) 2 142 HFC-125 2,800 3,500 3,170 3,691 (700) (330) 191 HFC-134a 1,300 1,430 1,300 1,549 (130) (130) 119 HFC-143a 3,800 4,470 4,800 5,508 (670) 330 1,038 HFC-152a 140 124 138 167 16 14 43 HFC-227ea 2,900 3,220 3,350 3,860 (320) 130 640 HFC-236fa 6,300 9,810 8,060 8,998 (3,510) (1,750) (812) HFC-4310mee 1,300 1,640 1,650 1,952 (340) 10 312 CF4 6,500 7,390 6,630 7,349 (890) (760) (41) C2F6 9,200 12,200 11,100 12,340 (3,000) (1,100) 140 C4F10 7,000 8,860 9,200 10,213 (1,860) 340 1,353 C6Fl4 7,400 9,300 7,910 8,780 (1,900) (1,390) (520) SF« 23,900 22,800 23,500 26,087 1,100 700 3,287 NF3 NA 17,200 16,100 17,885 NA (1,100) 685 NA (Not Applicable) NC (No Change) a The GWPs presented here are the ones most consistent with the methodology used in the AR4 report. b The GWP values presented here from the AR5 report include climate-carbon feedbacks for the non- CO2 gases in order to be consistent with the approach used in calculating the CO2 lifetime. Additionally, the AR5 reported separate values for fossil versus biogenic methane in order to account for the CO2 oxidation product. 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 CO2 is only included in the value from AR5 that includes climate-carbon feedbacks. Note: Parentheses indicate negative values. Source: (IPCC 2013, IPCC 2007, IPCC 2001, IPCC 1996). 4 5 1.2 National Inventory Arrangements 6 The U.S. Environmental Protection Agency (EPA), in cooperation with other U.S. government agencies, prepares 7 the Inventory of U.S. Greenhouse Gas Emissions and Sinks. A wide range of agencies and individuals are involved 8 in supplying data to, planning methodological approaches and improvements, reviewing, or preparing portions of the 9 U.S. Inventory—including federal and state government authorities, research and academic institutions, industry 10 associations, and private consultants. 11 Within EPA, the Office of Atmospheric Programs (OAP) is the lead office responsible for the emission calculations 12 provided in the Inventory, as well as the completion of the National Inventory Report and the Common Reporting 13 Format (CRF) tables. EPA's Office of Transportation and Air Quality (OTAQ) is also involved in calculating 14 emissions for the Inventory. While the U.S. Department of State officially submits the annual Inventory to the 1-10 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2016 ------- 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 UNFCCC, EPA's OAP serves as the National Inventory Focal Point for technical questions and comments on the U.S. Inventory. The staff of EPA coordinate the annual methodological choice, activity data collection, and emission calculations at the individual source category level. EPA, the inventory coordinator, compiles the entire Inventory into the proper reporting format for submission to the UNFCCC, and is responsible for the collection and consistency of cross-cutting issues in the Inventory. Several other government agencies contribute to the collection and analysis of the underlying activity data used in the Inventory calculations. Formal and informal relationships exist between EPA and other U.S. agencies that 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 military fuel consumption and bunker fuels. Informal relationships also exist with other U.S. agencies to provide activity data for use in EPA's emission calculations. These 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. Finally, EPA as the National Inventory Focal Point, in coordination with the U.S. Department of State, officially submits the Inventory to the UNFCCC each April. Figure 1-1 diagrams the National Inventory Arrangements. Introduction 1-11 ------- Figure 1-1: National Inventory Arrangements Diagram Inventory Process United States National Inventory Arrangements United Nations Framework Convention on Climate Change Inventory Submission Inventory Compilation U.S. Environmental Protection Agency Inventory Compiler Emission Calculations U.S. Environmental Protection Agency Other U.S. Government Agencies USDA Forest Service, USDA Agricultural Research Service, NOAA, DOD, FAA Data Collection Energy • U.S. Department of Energy and its National Laboratories • Energy Information Administration • U.S. Department of Transportation • Bureau of Transportation Statistics • Federal Highway Administration • Federal Aviation Administration • U.S. Department of Defense - Defense Logistics Agency • U.S. Department of Commerce - Bureau of the Census • U.S. Department of Homeland Security • U.S. Department of Labor's Mine Safety and Health Administration • EPA Office of Transportation and Air Quality MOVES Model • EPA Greenhouse Gas Reporting Program (GHGRP) and Acid Rain Program • American Association of Railroads • American Public Transportation Association • U.S. Department of Labor - Mine Safety and Health Administration • Data from research studies, trade publications, and —= industry associations A / Agriculture/LULUCF • U.S. Department of Agriculture (USDA) National Agricultural Statistics Service • USDA Natural Resources Conservation Service • USDA Economic Research Service • USDA Farm Service Agency • USDA Animal Plant Health Inspection Service • Conservation Technology Information Service • U.S. Geological Survey ¦ USDA Forest Service • National Oceanic and Atmospheric Administration (NOAA) • U.S. Department of the Interior Bureau of Land Management • EPA Office of Solid Waste • U.S. Census Bureau • Alaska Department of Natural Resources • American Society of Agricultural Engineers • Association of American Plant Food Control Officials • Tennessee Valley Authority • Data from research studies, trade publications, LggflH and industry associations ~\ / Industrial Processes and Product Use • U.S. Geological Survey National Minerals Information Center • EPA GHGRP • U.S. Department of Commerce • American Iron and Steel Institute (AISI) • American Chemistry Council (ACC) • U.S. Aluminum Association • Air-Conditioning, Heating, and Refrigeration Institute • Data from research studies, trade publications, and industry associations Waste • EPA GHGRP • EPA Office of Land and Emergency Management • Data from research studies, trade publications, and industry associations 1-12 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2016 ------- 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 This section describes EPA's approach to preparing the annual U.S. Inventory, which consists of a National Inventory Report (NIR) and Common Reporting Format (CRF) tables. The inventory coordinator at EPA, with support from the data/document manager is responsible for compiling all emission estimates and ensuring consistency and quality throughout the NIR and CRF tables. Emission calculations for individual sources are the responsibility of individual source leads, who are most familiar with each source category and the unique characteristics of its emissions profile. The individual source leads determine the most appropriate methodology and collect the best activity data to use in the emission calculations, based upon their expertise in the source category, as well as coordinating with researchers and contractors familiar with the sources. A multi-stage process for collecting information from the individual source leads and producing the Inventory is undertaken annually to compile all information and data. Methodology Development, Data Collection, and Emissions and Sink Estimation Source leads at EPA collect input data and, as necessary, evaluate or develop the estimation methodology for the individual source categories. Because EPA has been preparing the Inventory for many years, for most source 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 category are being developed for the first time, or if the methodology is changing for an existing source category (e.g., the United States is implementing a higher Tiered approach for that source category), then the source category lead will develop a new methodology, gather the most appropriate activity data and emission factors (or in some cases direct emission measurements) for the entire time series, and conduct a special source-specific review process involving relevant experts from industry, government, and universities (see Box ES-6 on approach to recalculations). Once the methodology is in place and the data are collected, the individual source leads calculate emissions and sink estimates. The source leads then update or create the relevant text and accompanying annexes for the Inventory. Source leads are also responsible for completing the relevant sectoral background tables of the CRF, conducting quality assurance and quality control (QA/QC) checks, and uncertainty analyses. The treatment of confidential business information (CBI) in the Inventory is based on EPA internal guidelines, as well as regulations30 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 Process and Product Use (IPPU) sectors, EPA has used aggregated facility-level data from the Greenhous Gas Reporting Program (GHGRP) to develop, inform, and/or quality-assure U.S. emissions estimates. In 2014, the 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.31 In the Inventory, EPA is publishing only data values that meet the GHGRP aggregation criteria.32 Specific uses of aggregated facility-level data are described in the respective methodological sections within those chapters. In addition, EPA also 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 confidential business information. 30 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 . 31 Federal Register Notice on "Greenhouse Gas Reporting Program: Publication of Aggregated Greenhouse Gas Data." See pp, 79 and 110 of notice at . 32 U.S. EPA Greenhouse Gas Reporting Program. Developments on Publication of Aggregated Greenhouse Gas Data, November 25, 2014. See . Introduction 1-13 ------- 1 Summary Data Compilation and Storage 2 The inventory coordinator at EPA with support from the data/document manager collect the source and sink 3 categories' descriptive text and Annexes, and also aggregates the emission estimates into a summary spreadsheet 4 that links the individual source category spreadsheets together. This summary sheet contains all of the essential data 5 in one central location, in formats commonly used in the Inventory document. In addition to the data from each 6 source category, national trend and related data are also gathered in the summary sheet for use in the Executive 7 Summary, Introduction, and Recent Trends sections of the Inventory report. Electronic copies of each year's 8 summary spreadsheet, which contains all the emission and sink estimates for the United States, are kept on a central 9 server at EPA under the jurisdiction of the inventory coordinator. 10 National Inventory Report Preparation 11 The NIR is compiled from the sections developed by each individual source or sink lead. In addition, the inventory 12 coordinator prepares a brief overview of each chapter that summarizes the emissions from all sources discussed in 13 the chapters. The inventory coordinator then carries out a key category analysis for the Inventory, consistent with the 14 2006IPCC Guidelines for National Greenhouse Gas Inventories, and in accordance with the reporting requirements 15 of the UNFCCC. Also at this time, the Introduction, Executive Summary, and Recent Trends sections are drafted, to 16 reflect the trends for the most recent year of the current Inventory. The analysis of trends necessitates gathering 17 supplemental data, including weather and temperature conditions, economic activity and gross domestic product, 18 population, atmospheric conditions, and the annual consumption of electricity, energy, and fossil fuels. Changes in 19 these data are used to explain the trends observed in greenhouse gas emissions in the United States. Furthermore, 20 specific factors that affect individual sectors are researched and discussed. Many of the factors that affect emissions 21 are included in the Inventory document as separate analyses or side discussions in boxes within the text. Text boxes 22 are also created to examine the data aggregated in different ways than in the remainder of the document, such as a 23 focus on transportation activities or emissions from electricity generation. The document is prepared to match the 24 specification of the UNFCCC reporting guidelines for National Inventory Reports. 25 Common Reporting Format Table Compilation 26 The CRF tables are compiled from individual tables completed by each individual source or sink lead, which contain 27 source emissions and activity data. The inventory coordinator integrates the source data into the UNFCCC's "CRF 28 Reporter" for the United States, assuring consistency across all sectoral tables. The summary reports for emissions, 29 methods, and emission factors used, the overview tables for completeness and quality of estimates, the recalculation 30 tables, the notation key completion tables, and the emission trends tables are then completed by the inventory 31 coordinator. Internal automated quality checks on the CRF Reporter, as well as reviews by the source leads, are 32 completed for the entire time series of CRF tables before submission. 33 Ci\h\€ jn4 34 QA/QC and uncertainty analyses are supervised by the QA/QC and uncertainty coordinators, who have general 35 oversight over the implementation of the QA/QC plan and the overall uncertainty analysis for the Inventory (see 36 sections on QA/QC and Uncertainty, below). These coordinators work closely with the source leads to ensure that a 37 consistent QA/QC plan and uncertainty analysis is implemented across all inventory sources. The inventory QA/QC 38 plan, detailed in a following section, is consistent with the quality assurance procedures outlined by EPA and IPCC. 39 The QA/QC and uncertainty findings also inform overall improvement planning, and specific improvements are 40 noted in the Planned Improvements sections of respective categories. QA processes are outlined below. 41 Expert, Public, and UNFCCC Review Periods 42 During the 30-day Expert Review period, a first draft of the document is sent to a select list of technical experts 43 outside of EPA who are not directly involved in preparing estimates. The purpose of the Expert Review is to provide 44 an objective review, encourage feedback on the methodological and data sources used in the current Inventory, 45 especially for sources which have experienced any changes since the previous Inventory. 1-14 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2016 ------- 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 Once comments are received and addressed, a second draft of the document is released for public review by publishing a notice in the U.S. Federal Register and posting the document on the EPA Web site. The Public Review period allows for a 30-day comment period and is open to the entire U.S. public. Comments may require further discussion with experts and/or additional research and specific Inventory improvements requiring further analysis as a result of comments are noted in categories Planned Improvement sections. See those sections for specific details. EPA publishes comments received with publication of the report on its website. Following completion and submission of the report to the UNFCCC, the report also undergoes review by an independent international team of experts for adherence to UNFCCC reporting guidelines and IPCC Guidance.33 Feedback from these review processes all contribute to improving inventory quality over time. Final Submittal to UNFCCC and Document Printing After the final revisions to incorporate any comments from the Expert Review and Public Review periods, EPA prepares the final National Inventory Report and the accompanying Common Reporting Format Reporter database. EPA as the National Inventory focal point with the U.S. Department of State sends the official submission of the U.S. Inventory to the UNFCCC. The document is then formatted and posted online, available for the public.34 1.3 Methodology and Data Sources Emissions of greenhouse gases from various source and sink categories have been estimated using methodologies that are consistent with the 2006 IPCC Guidelines for National Greenhouse Gas Inventories (IPCC 2006). To a great extent, this report makes use of published official economic and physical statistics for activity data and emission factors. Depending on the emission source category, activity data can include fuel consumption or deliveries, vehicle-miles traveled, raw material processed, 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 data sources. In addition to official statistics, the report utilizes findings from academic studies, trade association surveys and statistical reports, along with expert judgement, consistent with 2006 IPCC Guidelines. The IPCC methodologies provided in the 2006 IPCC Guidelines represent foundational methodologies for a variety of source categories, and many of these methodologies continue to be improved and refined as new research and data become available. This report uses the IPCC methodologies when applicable, and supplements them with other available country-specific methodologies and data where possible. Choices made regarding the methodologies and data sources used are provided in conjunction with the discussion of each source category in the main body of the report. Complete documentation is provided in the annexes on the detailed methodologies and data sources utilized in the calculation of each source category. Box 1-3: IPCC Reference Approach The 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. 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. 33 See . 34 See . Introduction 1-15 ------- 1 and therefore the carbon in them is oxidized and released into the atmosphere. Accounting for actual consumption of 2 fuels at the sectoral or sub-national level is not required. 3 4 1.4 Key Categories 5 The 2006IPCC Guidelines (IPCC 2006) defines a key category as a "[category] that is prioritized within the 6 national inventory system because its estimate lias a significant influence on a country's total inventory of 7 greenhouse gases in terms of the absolute level, the trend, or the uncertainty in emissions and removals."35 By 8 definition key categories include those categories that have the greatest contribution to the absolute level of national 9 emissions. In addition when an entire time series of emission and removal estimates is prepared, a thorough 10 investigation of key categories must also account for the influence of trends and uncertainties of individual source 11 and sink categories. This analysis can identify source and sink categories that diverge from the overall trend in 12 national emissions. Finally, a qualitative evaluation of key categories is performed to capture any categories that 13 were not identified in any of the quantitative analyses. 14 Approach 1, as defined in the 2006 IPCC Guidelines (IPCC 2006), was implemented to identify the key categories 15 for the United States. This analysis was performed twice; one analysis included sources and sinks from the Land 16 Use, Land-Use Change, and Forestry (LULUCF) sector, the other analysis did not include the LULUCF categories. 17 Following Approach 1, Approach 2, as defined in the 2006 IPCC Guidelines (IPCC 2006), was then implemented to 18 identify any additional key categories not already identified in Approach 1 assessment. This analysis, which includes 19 each source category's uncertainty assessments (or proxies) in its calculations, was also performed twice to include 20 or exclude LULUCF categories. 21 In addition to conducting Approach 1 and 2 level and trend assessments, a qualitative assessment of the source 22 categories, as described in the 2006 IPCC Guidelines (IPCC 2006), was conducted to capture any key categories that 23 were not identified by either quantitative method. For this inventory, no additional categories were identified using 24 criteria recommend by IPCC, but EPA continues to update its qualitative assessment on an annual basis. 25 Table 1-4: Key Categories for the United States (1990-2016) CRF Source Categories Gas Approach 1 Approach 2 Quala 2016 Emissions (MMT CO2 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 Energy CO2 Emissions from Mobile Combustion: Road CO2 . . 1,504.0 CO2 Emissions from Stationary Combustion - Coal - Electricity Generation CO2 . . 1,241.3 CO2 Emissions from Stationary Combustion - Gas - Electricity Generation CO2 . . 545.9 CO2 Emissions from Stationary Combustion - Gas - Industrial CO2 . . 478.8 35 See Chapter 4 Volume 1, "Methodological Choice and Identification of Key Categories" in IPCC (2006). See . 1-16 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2016 ------- CO2 Emissions from Stationary Combustion - Oil - Industrial CO2 . . 269.7 CO2 Emissions from Stationary Combustion - Gas - Residential CO2 . • 238.3 CO2 Emissions from Stationary Combustion - Gas - Commercial CO2 . . 170.3 CO2 Emissions from Mobile Combustion: Aviation CO2 . . 169.6 CO2 Emissions from Non-Energy Use of Fuels CO2 • 121.0 CO2 Emissions from Mobile Combustion: Other CO2 80.1 CO2 Emissions from Stationary Combustion - Coal - Industrial CO2 . . 59.0 CO2 Emissions from Stationary Combustion - Oil - Residential CO2 . • 58.0 CO2 Emissions from Stationary Combustion - Oil - Commercial CO2 . • 55.3 CO2 Emissions from Mobile Combustion: Marine CO2 41.1 CO2 Emissions from Stationary Combustion - Oil - U.S. Territories CO2 . 34.3 CO2 Emissions from Natural Gas Systems CO2 26.7 CO2 Emissions from Petroleum Systems CO2 . . 25.5 CO2 Emissions from Stationary Combustion - Oil - Electricity Generation CO2 . . 21.2 CO2 Emissions from Stationary Combustion - Gas - U.S. Territories CO2 • 3.0 CO2 Emissions from Stationary Combustion - Coal - Commercial CO2 • 2.3 CH4 Emissions from Natural Gas Systems CH4 . . 162.1 Fugitive Emissions from Coal Mining ch4 . . 53.8 CH4 Emissions from Petroleum Systems ch4 • • 39.3 N011-CO2 Emissions from Stationary Combustion - Residential ch4 • 3.4 N011-CO2 Emissions from Stationary Combustion - Electricity Generation N2O • • 14.9 Introduction 1-17 ------- N2O Emissions from Mobile Combustion: Road N2O . . 13.1 Non- CO2 Emissions from Stationary Combustion - Industrial N2O • 2.4 International Bunker Fuelsb Several • 115.5 Industrial Processes and Product Use CO2 Emissions from Iron and Steel Production & Metallurgical Coke Production CO2 . . 42.2 CO2 Emissions from Cement Production CO2 • 39.4 CO2 Emissions from Petrochemical Production CO2 • 27.4 CO2 Emissions from Other Process Uses of Carbonates CO2 11.2 N2O Emissions from Adipic Acid Production N2O 7.0 Emissions from Substitutes for Ozone Depleting Substances HiGWP . . 173.9 SFo Emissions from Electrical Transmission and Distribution HiGWP • 4.3 HFC-23 Emissions from HCFC-22 Production HiGWP . • 2.8 PFC Emissions from Aluminum Production HiGWP • 1.4 Agriculture CO2 Emissions from Liming CO2 • 3.9 CH4 Emissions from Enteric Fermentation CH4 • • 170.1 CH4 Emissions from Manure Management ch4 . . 67.7 CH4 Emissions from Rice Cultivation ch4 • 13.7 Direct N2O Emissions from Agricultural Soil Management N2O . . 237.6 Indirect N2O Emissions from Applied Nitrogen N2O . . 45.9 Waste CH4 Emissions from Landfills CH4 . . 107.7 Land Use, Land Use Change, and Forestry Net CO2 Emissions from Land Converted to Settlements CO2 • • 68.0 Net CO2 Emissions from Land Converted to Cropland CO2 • • 23.8 1-18 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2016 ------- Net CO2 Emissions from Land Converted to Grassland CO2 • 22.0 Net CO2 Emissions from Grassland Remaining Grassland CO2 • (1.6) Net CO2 Emissions from Cropland Remaining Cropland CO2 • • (9.9) Net CO2 Emissions from Land Converted to Forest Land CO2 • (75.0) Net CO2 Emissions from Settlements Remaining Settlements CO2 • • (103.7) Net CO2 Emissions from Forest Land Remaining Forest Land CO2 • • (670.5) CH4 Emissions from Forest Fires CH4 • 18.5 N2O Emissions from Forest Fires N2O • 12.2 Subtotal Without LULUCF 6,390.8 Total Emissions Without LULUCF 6,546.2 Percent of Total Without LULUCF 98% Subtotal With LULUCF 5,651.5 Total Emissions With LULUCF 5,829.3 Percent of Total With LULUCF 97% a Qualitative criteria. b Emissions from this source not included in totals. Note: Parentheses indicate negative values (or sequestration). 1 2 1.5 Quality Assurance and Quality Control 3 (QA/QC) 4 As part of efforts to achieve its stated goals for inventory quality, transparency, and credibility, the United States has 5 developed a quality assurance and quality control plan designed to check, document and improve the quality of its 6 inventory over time. QA/QC activities on the Inventory are undertaken within the framework of the U.S. Quality 1 Assurance/Quality Control and Uncertainty Management Plan (QA/QC plan) for the U.S. Greenhouse Gas 8 Inventory: Procedures Manual for OA/OC and Uncertainty Analysis. 9 Key attributes of the QA/QC plan are summarized in Figure 1-2. These attributes include: 10 • Procedures and Forms: detailed and specific systems that serve to standardize the process of documenting 11 and archiving information as well as to guide the implementation of QA/QC and the analysis of 12 uncertainty 13 • Implementation of Procedures: application of QA/QC procedures throughout the whole inventory 14 development process from initial data collection through preparation of the emission estimates, to 15 publication of the Inventory 16 • Quality Assurance: expert and public reviews for both the Inventory estimates and the Inventory report Introduction 1-19 ------- 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 (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 2006IPCC Guidelines (IPCC 2006) • Quality Control, consideration of secondary data and category-specific checks (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 • Multi-Year Implementation: a schedule for coordinating the application of QA/QC procedures across multiple years, especially for category-specific QC, prioritizing key categories • 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. In addition, based on the national QA/QC plan for the Inventory, source-specific QA/QC plans have been developed for a number of sources. These plans follow the procedures outlined in the national QA/QC plan, tailoring the procedures to the specific text and spreadsheets of the individual sources. For each greenhouse gas emissions source or sink included in this Inventory, a minimum of general or Tier 1 QA/QC analysis has been undertaken. Where QA/QC activities for a particular source go beyond the minimum Tier 1 level, and include category-specific checks (Tier 2) further explanation is provided within the respective source 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 category sections in each chapter. 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 data and methods. The QA/QC plan guides the process of ensuring inventory quality by describing data and methodology checks, developing processes governing peer review and public comments, and developing guidance on conducting an analysis of the uncertainty surrounding the emission estimates. The QA/QC procedures also include feedback loops and provide for corrective actions that are designed to improve the inventory estimates over time. 1-20 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2016 ------- 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 Figure 1-2: U.S. QA/QC Plan Summary _>* 03 C < > CI Obtain data in electronic format (if possible) Review spreadsheet construction Avoid hardwiring • Use data validation Protect cells Develop automatic checkers for: Outliers, negative values, or missing data • Variabletypes match values • Time series consistency Maintain tracking tab for status of gathering efforts Check input data for transcription errors Inspect automatic checkers Identify spreadsheet modifications that could provide additional QA/QC checks Contact reports for non- electroniccommunications Provide cell references for primary data elements Obtain copies of all data sources ~stand location of any working/external spreadsheets Document assumptions Check citations in spreadsheet and text for accuracy and style Check reference docket for new citations Review documentation for any data/ methodology changes Clearly label parameters, units, and conversion factors Review spreadsheet integrity ¦ Equations • Units ¦ Inputs and output Develop automated checkers for: ¦ Input ranges ¦ Calculations • Emission aggregation Reproduce calculations Reviewtime series consistency Review changes in data/consistency with IPCC methodology Common starting versions for each inventory year Utilize unalterable summarytab foreach source spreadsheetfor linkingtoa master summary spreadsheet Follow strict version control procedures Document QA/QC procedures Data Gathering Data Documentation Calculating Emissions Cross-Cutting Coordination 1.6 Uncertainty Analysis of Emission Estimates Uncertainty estimates are an essential element of a complete and transparent emissions inventory. Uncertainty information is not intended to dispute the validity of the Inventory estimates, but to help prioritize efforts to improve the accuracy of future Inventories and guide future decisions on methodological choice. While the U.S. Inventory calculates its emission estimates with the highest possible accuracy, uncertainties are associated to a varying degree with the development of emission estimates for any inventory. Some of the current estimates, such as those for carbon dioxide (CO2) emissions from energy-related activities, are considered to have minimal uncertainty associated with them. For some other limited categories of emissions, however, a lack of data or an incomplete understanding of how emissions are generated increases the uncertainty or systematic error associated with the estimates presented. The UNFCCC reporting guidelines follow the recommendation in the 2006 IPCC Guidelines (IPCC 2006) and require that countries provide single point estimates for each gas and emission or removal source category. Within the discussion of each emission source, specific factors affecting the uncertainty associated with the estimates are discussed. Additional research in the following areas could help reduce uncertainty in the U.S. Inventory: • Incorporating excluded emission sources. Quantitative estimates for some of the sources and sinks of greenhouse gas emissions are not available at this time. In particular, emissions from some land-use activities (e.g., emissions and removals from interior Alaska) and industrial processes are not included in the inventory either because data are incomplete or because methodologies do not exist for estimating Introduction 1-21 ------- 1 emissions from these source categories. See Annex 5 of this report for a discussion of the sources of 2 greenhouse gas emissions and sinks excluded from this report. 3 • Improving the accuracy of emission factors. Further research is needed in some cases to improve the 4 accuracy of emission factors used to calculate emissions from a variety of sources. For example, the 5 accuracy of current emission factors applied to CH4 and N20 emissions from stationary and mobile 6 combustion is highly uncertain. 7 • Collecting detailed activity data. Although methodologies exist for estimating emissions for some sources, 8 problems arise in obtaining activity data at a level of detail where more technology or process-specific 9 emission factors can be applied. 10 The overall uncertainty estimate for total U.S. greenhouse gas emissions was developed using the IPCC Approach 2 11 uncertainty estimation methodology. Estimates of quantitative uncertainty for the total U.S. greenhouse gas 12 emissions are shown below, in Table 1-5. 13 The IPCC provides good practice guidance on two approaches—Approach 1 and Approach 2—to estimating 14 uncertainty for individual source categories. Approach 2 uncertainty analysis, employing the Monte Carlo Stochastic 15 Simulation technique, was applied wherever data and resources permitted; further explanation is provided within the 16 respective source category text and in Annex 7. Consistent with the 2006 IPCC Guidelines (IPCC 2006), over a 17 multi-year timeframe, the United States expects to continue to improve the uncertainty estimates presented in this 18 report. 19 Table 1-5: Estimated Overall Inventory Quantitative Uncertainty (MMT CO2 Eq. and Percent) 20 - TO BE UPDATED FOR FINAL INVENTORY REPORT 2015 Emission Uncertainty Range Relative to Emission Standard Estimate3 Estimateb Mean0 Deviation0 Gas (MMT CO2 Eq.) (MMT CO2 Eq.) (%) (MMT CO2 Eq.) Lower Upper Lower Upper Boundd Boundd Bound Bound CO2 5,411.0 5,305.4 5,652.4 -2% 4% 5,474.3 90.2 CH4e 655.7 599.9 779.2 -9% 19% 681.8 45.3 N2Oe 334.8 302.5 424.6 -10% 27% 357.0 30.7 PFC,HFC,SF6,andNF3e 184.7 183.1 204.4 -1% 11% 193.4 5.5 Total 6,586.2 6,505.0 6,919.9 -1% 5% 6,706.6 106.0 LULUCF Emissions' 19.7 14.6 38.2 -26% 94% 23.3 6.3 LULUCF Total Net Flux® (778.7) (993.1) (620.7) -20% 28% (808.4) 94.7 LULUCF Sector Total" (758.9) (969.7) (597.9) -21% 28% (785.1) 94.8 Net Emissions (Sources and Sinks) 5,827.3 5,643.8 6,207.4 -3% 7% 5,921.5 142.8 Notes: Total emissions (excluding emissions for which uncertainty was not quantified) is presented without LULUCF. Net emissions is presented with LULUCF. a Emission estimates reported in this table correspond to emissions from only those source categories for which quantitative uncertainty was performed this year. Thus the totals reported in this table exclude approximately 0.4 MMT CO2 Eq. of emissions for which quantitative uncertainty was not assessed. Hence, these emission estimates do not match the filial total U.S. greenhouse gas emission estimates presented in this Inventory. b The lower and upper bounds for emission estimates correspond to a 95 percent confidence interval, with the lower bound corresponding to 2.5th percentile and the upper bound corresponding to 97.5th percentile. c Mean value indicates the arithmetic average of the simulated emission estimates; standard deviation indicates the extent of deviation of the simulated values from the mean. d The lower and upper bound emission estimates for the sub-source categories do not sum to total emissions because the low and high estimates for total emissions were calculated separately through simulations. e The overall uncertainty estimates did not take into account the uncertainty in the GWP values for CH4, N2O and high GWP gases used in the Inventory emission calculations for 2015. f LULUCF emissions include the CH4 and N2O emissions reported for N011-CO2 Emissions from Forest Fires, Emissions from Drained Organic Soils, N2O Fluxes from Forest Soils, N011-CO2 Emissions from Grassland Fires, N2O Fluxes from Settlement 1-22 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2016 ------- Soils. Coastal Wetlands Remaining Coastal Wetlands, Peatlands Remaining Peatlands, and CI 11 Emissions from Land Converted to Coastal Wetlands.. •" Net C( h llu\ is the net C stock change from the follow ing categories: I'oresl I.and Remaining I'oresl I,and, hind ('onverlcd lo I'oresl Land. ( ropland Remaining ( ropland. Land ( (inverted lo ( ropland, (irassland Remaining (irassland. I,and ( (inverted 10 ('wasshmd. Changes in ()rganic Soils Carbon Stocks. Changes in Urban Tree Carbon Slocks. Changes in Yard Trimmings and Food Scrap Carbon Stocks in 1 .andlills, I.and ('(inverted lo Settlements. II ellands Remaining Wetlands, and {.and ('(inverted lo 11 ellands. 11 The H JI.UCF Sector Total is the net sum of all emissions (i.e., sources) of greenhouse gases to the atmosphere plus removals of C(): (i.e.. sinks or negative emissions) from the atmosphere. Notes: Totals may not sum due to independent rounding. Parentheses indicate net sequestration. 1 Emissions calculated for the U.S. Inventory reflect current best estimates; in some cases, however, estimates are 2 based on approximate methodologies, assumptions, and incomplete data. As new information becomes available in 3 the future, the United States will continue to improve and revise its emission estimates. See Annex 7 of this report 4 for further details on the U. S. process for estimating uncertainty associated with the emission estimates and for a 5 more detailed discussion of the limitations of the current analysis and plans for improvement. Annex 7 also includes 6 details on the uncertainty analysis performed for selected source categories. 7 1.7 Completeness 8 This report, along with its accompanying CRF tables, serves as a thorough assessment of the anthropogenic sources 9 and sinks of greenhouse gas emissions for the United States for the time series 1990 through 2015. This report is 10 intended to be comprehensive and includes the vast majority of emissions and removals identified as anthropogenic, 11 consistent with IPCC and UNFCCC guidelines. In general, sources not accounted for in this Inventory are excluded 12 because they are not occurring in the U.S., or because data are unavailable to develop an estimate and/or the sources 13 were determined to be insignificant36 in terms of overall national emissions per UNFCCC reporting guidelines. 14 The United States is continually working to improve upon the understanding of such sources and seeking to find the 15 data required to estimate related emissions. As such improvements are implemented, new emission sources are 16 quantified and included in the Inventory. For a list of sources not included and more information, see Annex 5 and 17 the respective source category sections in each chapter of this report. is 1.8 Organization of Report 19 In accordance with the revision of the UNFCCC reporting guidelines agreed to at the nineteenth Conference of the 20 Parties (UNFCCC 2014), this Inventory of U.S. Greenhouse Gas Emissions and Sinks is segregated into five sector- 21 specific chapters consistent with the UN Common Reporting Framework (CRF), listed below in Table 1-6. In 22 addition, chapters on Trends in Greenhouse Gas Emissions and Other information to be considered as part of the 23 U.S. Inventory submission are included. 36 See paragraph 32 of Decision 24/CP. 19, the UNFCCC reporting guidelines on annual inventories for Parties included in Annex 1 to the Convention. Paragraph notes that. .An emission should only be considered insignificant if the likely level of emissions is below 0.05 per cent of the national total GHG emissions, and does not exceed 500 kt CO2 Eq. The total national aggregate of estimated emissions for all gases and categories considered insignificant shall remain below 0.1 percent of the national total GHG emissions." Introduction 1-23 ------- 1 Table 1-6: IPCC Sector Descriptions Chapter/IPCC Sector Activities Included Energy Emissions of all greenhouse gases resulting from stationary and mobile energy activities including fuel combustion and fugitive fuel emissions, and non- energy use of fossil fuels. Industrial Processes and Emissions resulting from industrial processes and product use of greenhouse Product Use gases. 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 N2O from land use, Change, and Forestry land-use change and forestry. Waste Emissions from waste management activities. 2 Within each chapter, emissions are identified by the anthropogenic activity that is the source or sink of the 3 greenhouse gas emissions being estimated (e.g., coal mining). Overall, the following organizational structure is 4 consistently applied throughout this report: 5 Chapter/IPCC Sector: Overview of emission trends for each IPCC defined sector. 6 CRF Source or category: Description of source pathway and emission/removal trends based on IPCC 7 methodologies, consistent with UNFCCC reporting guidelines. 8 Methodology: Description of analytical methods (e.g. 2006 IPCC Guidelines) employed to produce emission 9 estimates and identification of data references, primarily for activity data and emission factors. 10 Uncertainty and Time Series Consistency: A discussion and quantification of the uncertainty in emission 11 estimates and a discussion of time-series consistency. 12 QA/QC and Verification: A discussion on steps taken to QA/QC and verily the emission estimates, consistent with 13 the U.S. QA/QC plan, and any key findings. 14 Recalculations: A discussion of any data or methodological changes that necessitate a recalculation of previous 15 years' emission estimates, and the impact of the recalculation on the emission estimates, if applicable. 16 Planned Improvements: A discussion on any category-specific planned improvements, if applicable. 17 Special attention is given to CO2 from fossil fuel combustion relative to other sources because of its share of 18 emissions and its dominant influence on emission trends. For example, each energy consuming end-use sector (i.e., 19 residential, commercial, industrial, and transportation), as well as the electricity generation sector, is described 20 individually. Additional information for certain source categories and other topics is also provided in several 21 Annexes listed in Table 1-7. 22 Table 1-7: List of Annexes ANNEX 1 Key Category Analysis ANNEX 2 Methodology and Data for Estimating CO2 Emissions from Fossil Fuel Combustion 2.1. Methodology for Estimating Emissions of CO2 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, N2O, and Indirect Greenhouse Gases from Stationary Combustion 3.2. Methodology for Estimating Emissions of CH4, N2O, and Indirect Greenhouse Gases from Mobile Combustion and Methodology for and Supplemental Information on Transportation-Related Greenhouse Gas Emissions 3.3. Methodology for Estimating Emissions from Commercial Aircraft Jet Fuel Consumption 3.4. Methodology for Estimating CH4 Emissions from Coal Mining 3.5. Methodology for Estimating CH4 and CO2 Emissions from Petroleum Systems 3.6. Methodology for Estimating CH4 Emissions from Natural Gas Systems 3.7. Methodology for Estimating CO2 and N2O Emissions from Incineration of Waste 3.8. Methodology for Estimating Emissions from International Bunker Fuels used by the U.S. Military 3.9. Methodology for Estimating HFC and PFC Emissions from Substitution of Ozone Depleting Substances 1-24 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2016 ------- 3.10. Methodology for Estimating CH4 Emissions from Enteric Fermentation 3.11. Methodology for Estimating CH4 and N2O Emissions from Manure Management 3.12. Methodology for Estimating N2O 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 CO2 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. Sulfur Dioxide Emissions 6.4. Complete List of Source Categories 6.5. Constants, Units, and Conversions 6.6. Abbreviations 6.7. Chemical Formulas ANNEX 7 Uncertainty 7.1. Overview 7.2. Methodology and Results 7.3. 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 F actors 8.4 . Responses During the Review Process Introduction 1-25 ------- 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 2. Trends in Greenhouse Gas Emissions 2.1 Recent Trends in U.S. Greenhouse Gas Emissions and Sinks In 2016, total gross U.S. greenhouse gas emissions were 6,546.2 MMT, or million metric tons, carbon dioxide (CO2) Eq. Total U.S. emissions have increased by 2.8 percent from 1990 to 2016, and emissions decreased from 2015 to 2016 by 2.0 percent (131.1 MMT CO2 Eq.). The decrease in total greenhouse gas emissions between 2015 and 2016 was driven in large part by a decrease in CO2 emissions from fossil fuel combustion. The decrease in CO2 emissions from fossil fuel combustion was a result of multiple factors, including: (1) substitution from coal to natural gas and other sources in the electric power sector; and (2) warmer winter conditions in 2016 resulting in a decreased demand for heating fuel in the residential and commercial sectors. Since 1990, U.S. emissions have increased at an average annual rate of 0.1 percent. Figure 2-1 through Figure 2-3 illustrate the overall trend in total U.S. emissions by gas, annual changes, and absolute changes since 1990. Overall, net emissions in 2016 were 11.6 percent below 2005 levels as shown in Table 2-1. Figure 2-1: Gross U.S. Greenhouse Gas Emissions by Gas (MMT CO2 Eq.) 9,000 8,000 7,000 6,000 o 5,000 U 4,000 3,000 2,000 1,000 0 I HFCs, PFCs, SFf Nitrous Oxide I Methane I Carbon Dioxide and NF3 Subtotal Trends 2-1 ------- 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 Figure 2-2: Annual Percent Change in Gross U.S. Greenhouse Gas Emissions Relative to the Previous Year 4% 2% 0% -2% -4% -6% -8% 1,8% 1.7% H M 1A% 1.3% 1 0.8% Mjj 0 6% 2.2% 0.5% 0-6% 1.9% 0.1% 1 3.5% 12.8% I 0.9% 1 II || "l| -2.8% I ^HfNro5'U">i£>fsvCOcrvO'-irNjro^u->ioi-vCOvo On Ov On On On On On O O O O u O u O o o iH OnOnOnOnOnOnOnOnOnCjOOOOCsOOOOOOOOOOO Figure 2-3: Cumulative Change in Annual Gross U.S. Greenhouse Gas Emissions Relative to 1990 (1990=0, MMT COz Eq.) 1,200- Overall, from 1990 to 2016, total emissions of CO2 increased by 196.5 MMT CO2 Eq. (3.8 percent), while total emissions of methane (CH4) decreased by 122.3 MMT CO: Eq. (15.7 percent), and total emissions of nitrous oxide (N2O) increased by 14.2 MMT CO2 Eq. (4.0 percent). During the same period, aggregate weighted emissions of hydrofluorocarbons (HFCs), perfluorocarbons (PFCs), sulfur hexafluoride (SF6), and nitrogen trifluoride (NF3) rose by 88.6 MMT CO2 Eq. (88.8 percent). 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 wanning potentials (GWPs), and, in the cases of PFCs, SF6, and NF3, long atmospheric lifetimes. Conversely, U.S. greenhouse gas emissions were partly offset by carbon (C) sequestration in managed forests, trees in urban areas, agricultural soils, landfilled yard trimmings, and coastal wetlands. These were estimated to offset 11.5 percent of total emissions in 2016. Table 2-1 summarizes emissions and sinks from all U.S. anthropogenic sources in weighted units of MMT CO2 Eq., while unweighted gas emissions and sinks in kilotons (kt) are provided in Table 2-2. 2-2 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2016 ------- 1 Table 2-1: Recent Trends in U.S. Greenhouse Gas Emissions and Sinks (MMT CO2 Eq.) Gas/Source 1990 2005 2012 2013 2014 2015 2016 CO2 5,136.8 6,150.8 5,383.7 5,541.7 5,590.5 5,449.5 5,333.4 Fossil Fuel Combustion 4,755.8 5,759.1 5,029.8 5,162.3 5,206.1 5,059.3 4,976.7 Electric Power 1,820.8 2,400.9 2,022.2 2,038.1 2,038.0 1,900.7 1,808.8 Treimportation 1,467.2 1,855.8 1,661.9 1,677.6 1,717.1 1,735.5 1,794.9 Industrial 874.5 867.8 818.4 848.7 830.8 819.3 807.6 Residential 338.3 357.8 282.5 329.7 345.3 316.8 296.2 Commercial 227.4 227.0 201.3 225.7 233.6 245.6 227.9 U.S. Territories 27.6 49.7 43.5 42.5 41.4 41.4 41.4 Non-Energy Use of Fuels 119.6 141.7 113.3 133.2 127.8 135.1 121.0 Iron and Steel Production & Metallurgical Coke Production 101.5 68.0 55.4 53.3 58.2 47.7 42.2 Cement Production 33.5 46.2 35.3 36.4 39.4 39.9 39.4 Petrochemical Production 21.2 26.8 26.5 26.4 26.5 28.1 27.4 Natural Gas Systems 29.7 22.5 24.4 26.0 27.0 26.3 26.7 Petroleum Systems 9.4 17.0 25.6 29.7 32.9 38.0 25.5 Lime Production 11.7 14.6 13.8 14.0 14.2 13.3 13.3 Other Process Uses of Carbonates 4.9 6.3 8.0 10.4 11.8 11.2 11.2 Ammonia Production 13.0 9.2 9.4 10.0 9.6 10.6 11.2 Incineration of Waste 8.0 12.5 10.4 10.4 10.6 10.7 10.7 Urea Fertilization 2.4 3.5 4.3 4.4 4.5 4.9 5.1 Carbon Dioxide Consumption 1.5 1.4 4.0 4.2 4.5 4.5 4.5 Urea Consumption for Non- Agricultural Purposes 3.8 3.7 4.4 4.1 1.5 4.2 4.0 Liming 4.7 4.3 6.0 3.9 3.6 3.8 3.9 Ferroalloy Production 2.2 1.4 1.9 1.8 1.9 2.0 1.8 Soda Ash Production 1.4 1.7 1.7 1.7 1.7 1.7 1.7 Titanium Dioxide Production 1.2 1.8 1.5 1.7 1.7 1.6 1.6 Aluminum Production 6.8 4.1 3.4 3.3 2.8 2.8 1.3 Glass Production 1.5 1.9 1.2 1.3 1.3 1.3 1.3 Phosphoric Acid Production 1.5 1.3 1.1 1.1 1.0 1.0 1.0 Zinc Production 0.6 1.0 1.5 1.4 1.0 0.9 0.9 Lead Production 0.5 0.6 0.5 0.5 0.5 0.5 0.5 Silicon Carbide Production and Consumption 0.4 0.2 0.2 0.2 0.2 0.2 0.2 Magnesium Production and Processing + + + + + + + Wood Biomass, Ethanol, and Biodiesel Consumption" 219.4 230.7 276.2 299.8 308.3 294.5 291.1 International Bunker Fuelsb 103.5 113.1 105.8 99.8 103.4 110.9 114.4 CH4c 778.1 679.3 661.3 659.6 665.3 664.0 655.8 Enteric Fermentation 164.2 168.9 166.7 165.5 164.2 166.5 170.1 Natural Gas Systems 193.7 160.0 156.8 159.6 164.2 164.4 162.1 Landfills 179.6 132.7 117.0 113.3 112.7 111.7 107.7 Manure Management 37.2 56.3 65.6 63.3 62.9 66.3 67.7 Coal Mining 96.5 64.1 66.5 64.6 64.6 61.2 53.8 Petroleum Systems 42.3 34.7 35.4 38.8 41.0 39.4 39.3 Wastewater Treatment 15.7 15.8 15.1 14.9 15.0 15.1 14.8 Rice Cultivation 16.0 16.7 11.3 11.5 12.7 12.3 13.7 Stationary Combustion 8.6 7.9 7.3 8.7 OO OO 7.8 7.2 Abandoned Oil and Gas Wells 6.5 6.9 7.0 7.0 7.1 7.2 7.1 Abandoned Underground Coal Mines 7.2 6.6 6.2 6.2 6.3 6.4 6.7 Mobile Combustion 9.8 6.6 4.0 3.7 3.4 3.1 3.0 Composting 0.4 1.9 1.9 2.0 2.1 2.1 2.1 Field Burning of Agricultural Residues 0.2 0.2 0.3 0.3 0.3 0.3 0.3 Trends 2-3 ------- Petrochemical Production 0.2 0.1 0.1 0.1 0.1 0.2 0.2 Ferroalloy Production + + + + + + + Silicon Carbide Production and Consumption + + + + + + + Iron and Steel Production & Metallurgical Coke Production + + + + + + + Incineration of Waste + + + + + + + International Bunker Fuelsb 0.2 0.1 0.1 0.1 0.1 0.1 0.1 N2Oc 354.6 357.4 335.2 362.6 360.5 378.9 368.8 Agricultural Soil Management 250.5 253.5 247.9 276.6 274.0 295.0 283.6 Stationary Combustion 11.1 17.5 16.8 18.6 18.9 18.0 18.4 Manure Management 14.0 16.5 17.5 17.5 17.5 17.7 18.1 Mobile Combustion 41.5 38.4 23.8 22.0 20.2 18.8 17.8 Nitric Acid Production 12.1 11.3 10.5 10.7 10.9 11.6 10.2 Adipic Acid Production 15.2 7.1 5.5 3.9 5.4 4.3 7.0 Wastewater Treatment 3.4 4.4 4.6 4.7 4.8 4.8 5.0 NjO from Product Uses 4.2 4.2 4.2 4.2 4.2 4.2 4.2 Caprolactam, Glyoxal, and Glyoxylic Acid Production 1.7 2.1 2.0 2.0 2.0 2.0 2.0 Composting 0.3 1.7 1.7 1.8 1.9 1.9 1.9 Incineration of Waste 0.5 0.4 0.3 0.3 0.3 0.3 0.3 Semiconductor Manufacture + 0.1 0.2 0.2 0.2 0.2 0.2 Field Burning of Agricultural Residues 0.1 0.1 0.1 0.1 0.1 0.1 0.1 International Bunker Fuelsb 0.9 1.0 0.9 0.9 0.9 0.9 1.0 HFCs 46.6 120.0 156.0 159.1 166.8 173.3 177.1 Substitution of Ozone Depleting Substances'1 0.3 99.8 150.3 154.8 161.4 168.6 173.9 HCFC-22 Production 46.1 20.0 5.5 4.1 5.0 4.3 2.8 Semiconductor Manufacture 0.2 0.2 0.2 0.2 0.3 0.3 0.3 Magnesium Production and Processing 0.0 0.0 + 0.1 0.1 0.1 0.1 PFCs 24.3 6.7 5.9 5.8 5.6 5.1 4.3 Semiconductor Manufacture 2.8 3.3 3.0 2.8 3.1 3.1 3.0 Aluminum Production 21.5 3.4 2.9 3.0 2.5 2.0 1.4 Substitution of Ozone Depleting Substances 0.0 + + + + + + SF« 28.8 11.7 6.6 6.3 6.3 5.9 6.2 Electrical Transmission and Distribution 23.1 8.3 4.6 4.5 4.6 4.2 4.3 Magnesium Production and Processing 5.2 2.7 1.6 1.5 1.0 0.9 1.0 Semiconductor Manufacture 0.5 0.7 0.3 0.4 0.7 0.7 0.8 NF3 + 0.5 0.6 0.6 0.5 0.6 0.6 Semiconductor Manufacture + 0.5 0.6 0.6 0.5 0.6 0.6 Total Emissions 6,369.2 7,326.4 6,549.4 6,735.6 6,795.6 6,677.3 6,546.2 LULUCF Emissions0 10.6 23.0 26.1 19.2 19.6 38.2 38.1 LULUCF CH4 Emissions 6.7 13.3 15.0 10.9 11.2 22.4 22.4 LULUCF N2O Emissions 3.9 9.7 11.1 8.3 8.4 15.8 15.7 LULUCF Carbon Stock Change' (830.2) (754.2) (779.5) (755.0) (760.0) (733.4) (754.9) LULUCF Sector Net Total' (819.6) (731.1) (753.5) (735.8) (740.4) (695.2) (716.8) Net Emissions (Sources and Sinks) 5,549.6 6,595.3 5,795.9 5,999.9 6,055.2 5,982.1 5,829.3 Notes: Total emissions presented without LULUCF. Net emissions presented with LULUCF. Totals may not sum due to independent rounding. Parentheses indicate negative values or sequestration. + Does not exceed 0.05 MMT CO2 Eq. a Emissions from Wood Biomass, Ethanol, and Biodiesel 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. 2-4 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2016 ------- c LULUCF emissions of CH4 and N2O are reported separately from gross emissions totals. LULUCF emissions include the CH4 and N2O 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 N2O 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 also result from this source. 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. 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 CH4 and N2O emissions to the atmosphere plus net carbon stock changes. 1 Table 2-2: Recent Trends in U.S. Greenhouse Gas Emissions and Sinks (kt) Gas/Source 1W0 2005 2012 2013 2014 2015 2016 CO2 5,136,|SI4 6,150,751 5,383,670 5,541,745 5,590,474 5,449,474 5,333,352 Fossil Fuel Combustion 4,755,819 5,759.056 5,029,830 5,162,315 5,206,135 5,059,288 4,976,737 Electric Power 1,820,SIS 2,400,874 2,022,181 2,038,122 2,038,018 1,900,673 1,808,797 Transportation 1,467,193 1,855.751 1,661,895 1,677,593 1,717,132 1,735,503 1,794,886 Industrial 874,5-IS 867,833 818,402 848,669 830,766 819,273 807,565 Residential 338,347 357,834 282,501 329,742 345,296 316,822 296,238 Commercial 227,358 227,041 201,325 225,722 233,557 245,637 227,871 U.S. Territories 27,555 49,723 43,527 42,467 41,365 41,380 41,380 Non-Energy Use of Fuels 119,588 141.669 113,275 133,176 127,'778 135,106 121,049 Iron and Steel Production & Metallurgical Coke Production 101,487 68.047 55,449 53,348 58,234 47,718 42,219 Cement Production 33,484 46.194 35,270 36,369 39,439 39,907 39,439 Petrochemical Production 21,203 26.794 26,501 26,395 26,496 28,062 27,411 Natural Gas Systems 29,708 22.529 24,398 26,004 27,004 26,329 26,739 Petroleum Systems 9,384 17.004 25,629 29,695 32,895 37,971 25,543 Lime Production 11,700 14.552 13,785 14,028 14,210 13,342 13,342 Other Process Uses of Carbonates 4,907 6.339 8,022 10,414 11,811 11,237 11,237 Ammonia Production 13,047 9,196 9,377 9,962 9,619 10,571 11,234 Incineration of Waste 7,950 12.469 10,392 10,363 10,608 10,676 10,676 Urea Fertilization 2,417 3,504 4,282 4,443 4,538 4,888 5,098 Carbon Dioxide Consumption 1,472 1.375 4,019 4,188 4,471 4,471 4,471 Urea Consumption for Non- Agricultural Purposes 3,784 3.653 4,392 4,074 1,541 4,169 3,959 Liming 4,667 4,349 5,978 3,907 3,609 3,777 3,863 Ferroalloy Production 2,152 1.392 1,903 1,785 1,914 1,960 1,796 Soda Ash Production 1,431 1,655 1,665 1,694 1,685 1,714 1,723 Titanium Dioxide Production 1,195 1.755 1,528 1,715 1,688 1,635 1,608 Aluminum Production 6,831 4,142 3,439 3,255 2,833 2,767 1,334 Glass Production 1,535 1.928 1,248 1,317 1,336 1,299 1,299 Phosphoric Acid Production 1,529 1,342 1,118 1,149 1,038 999 992 Zinc Production 632 1.030 1,486 1,429 956 933 925 Lead Production 516 527 546 459 473 482 Silicon Carbide Production and Consumption 375 219 158 169 173 180 174 Magnesium Production and Processing 1 2 2 2 3 3 Wood Biomass, Ethanol, and Biodiesel Consumption" 219,413 230,700 276,201 299,785 308,346 294,469 291,069 International Bunker Fuelsh 103,463 113,139 105,805 99,763 103,400 110,887 114,394 Trends 2-5 ------- CH4c 31,125 27,170 26,451 26,383 26,613 26,561 26,232 Enteric Fermentation 6,566 6,755 6,670 6,619 6,567 6,661 6,805 Natural Gas Systems 7,748 6,399 6,273 6,385 6,568 6,578 6,483 Landfills 7,182 5,310 4,680 4,531 4,509 4,467 4,306 Manure Management 1,486 2,254 2,625 2,530 2,514 2,651 2,709 Coal Mining 3,860 2,56.5 2,658 2,584 2,583 2,449 2,153 Petroleum Systems 1,693 1,386 1,415 1,553 1,639 1,576 1,571 Wastewater Treatment 627 631 604 596 598 605 593 Rice Cultivation 641 66" 453 462 510 493 549 Stationary Combustion 346 314 292 346 352 313 288 Abandoned Oil and Gas Wells 260 275 279 280 282 286 284 Abandoned Underground Coal Mines 288 264 249 249 253 256 268 Mobile Combustion 393 26o 160 149 136 123 119 Composting 15 7.5 77 81 84 84 85 Field Burning of Agricultural Residues 9 8 11 11 11 11 11 Petrochemical Production 9 3 3 5 7 7 Ferroalloy Production 1 - 1 + 1 1 1 Silicon Carbide Production and Consumption 1 + / + + + + + Iron and Steel Production & Metallurgical Coke Production 1 1 + + + + + Incineration of Waste + + / + + + + + International Bunker Fuelsb 7 .5 4 3 3 3 4 N2Oc 1,190 1,199 1,125 1,217 1,210 1,272 1,238 Agricultural Soil Management 840 851 832 928 920 990 952 Stationary Combustion 37 59 56 62 63 60 62 Manure Management 47 5.5 59 59 59 59 61 Mobile Combustion 139 129 80 74 68 63 60 Nitric Acid Production 41 38 35 36 37 39 34 Adipic Acid Production 51 24 19 13 18 14 23 Wastewater Treatment 11 1.5 16 16 16 16 17 N20 from Product Uses 14 14 14 14 14 14 14 Caprolactam, Glyoxal, and Glyoxylic Acid Production 6 7 7 7 7 7 Composting 1 6 6 6 6 6 6 Incineration of Waste 2 1 1 1 1 1 1 Semiconductor Manufacture + - 1 1 1 1 1 Field Burning of Agricultural Residues + - + + + + + International Bunker Fuelsb 3 3 3 3 3 3 3 HFCs M IV1 M M M M M Substitution of Ozone Depleting Substances'1 M M M M M M M HCFC-22 Production 3 1 + + + + + Semiconductor Manufacture M M M M M M M Magnesium Production and Processing 0 0 + + + + + PFCs M IV1 M M M M M Semiconductor Manufacture M M M M M M M Aluminum Production M M M M M M M Substitution of Ozone Depleting Substances 0 - + + + + + SF« 1 1 + + + + + Electrical Transmission and Distribution 1 - + + + + + 2-6 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2016 ------- 1 2 3 4 5 6 7 8 9 10 11 12 13 14 Magnesium Production and Processing + + + + + + + Semiconductor Manufacture + + + + + + + NF3 + + + + + + + Semiconductor Manufacture + + + + + + + + Does not exceed 0.5 kt. M - Mixture of multiple gases a Emissions from Wood Biomass, Ethanol, and Biodiesel 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. c LULUCF emissions of CUt and N2O are reported separately from gross emissions totals. Refer to Lable 2-8 for a breakout of emissions and removals for LULUCF by gas and source category. d Small amounts of PFC emissions also result from this source. Notes: Lotals may not sum due to independent rounding. Parentheses indicate negative values or sequestration. Emissions of all gases can be summed from each source category into a set of five sectors defined by the Intergovernmental Panel on Climate Change (IPCC). Figure 2-4 and Table 2-3 illustrate that over the twenty-seven- year period of 1990 to 2016, total emissions from the Energy, Industrial Processes and Product Use, and Agriculture sectors grew by 136.2 MMT CO2 Eq. (2.6 percent), 35.2 MMT CO2 Eq. (10.3 percent), and 73.4 MMT CO2 Eq. (15.0 percent), respectively. Emissions from the Waste sector decreased by 67.9 MMT CO2 Eq. (34.1 percent). Over the same period, total C sequestration in the Land Use. Land-Use Change, and Forestry (LULUCF) sector increased by 75.3 MMT CO2 (9.1 percent decrease in total C sequestration), and emissions from the LULUCF sector increased by 27.4 MMT CO2 Eq. (258 percent). Figure 2-4; U.S. Greenhouse Gas Emissions and Sinks by Chapter/IPCC Sector (MMT CO2 Eq.) o u 7,500- 7,000- 6,500- 6,000- 5,500- 5,000- 4,500- 4,000- 3,500- 3,000- 2,500- 2,000- 1,500- 1,000- 500- 0- -500- Waste Industrial Processes and Product Use — Agriculture LULUCF (emissions) Energy Land Use, Land-Use Change and Forestry (LULUCF) (removals) o^rMroo»HrMro^rmvo ^0>O^ON<^0>(TtO^<^C^OOOOOOOOOO»-lrHHHH»-IH cr>cr>cncr>(j>cr»cr*cncr>cr»ooooooooooooooooo H »—I t-i r-t tH i-i y-t t—I fM fN| fM (M CM ------- Petroleum Systems 51.7 51." 61.0 68.5 73.9 77.4 64.8 Coal Mining 96.5 64.1 66.5 64.6 64.6 61.2 53.8 Stationary Combustion 19.8 25.4 24.1 27.2 27.7 25.8 25.6 Mobile Combustion 51.3 45.0 27.8 25.7 23.6 21.9 20.8 Incineration of Waste 8.4 12.9 10.7 10.7 10.9 11.0 11.0 Abandoned Oil and Gas Wells 6.5 6.9 7.0 7.0 7.1 7.2 7.1 Abandoned Underground Coal Mines 7.2 6.6 6.2 6.2 6.3 6.4 6.7 Industrial Processes and Product Use 340.5 354.2 361.6 364.7 380.2 378.8 375.7 Substitution of Ozone Depleting Substances 0.3 99.8 150.4 154.8 161.4 168.6 173.9 Iron and Steel Production & Metallurgical Coke Production 101.5 68.1 55.5 53.4 58.2 47.7 42.2 Cement Production 33.5 46.2 35.3 36.4 39.4 39.9 39.4 Petrochemical Production 21.4 26.9 26.6 26.5 26.6 28.2 27.6 Lime Production 11.7 14.6 13.8 14.0 14.2 13.3 13.3 Other Process Uses of Carbonates 4.9 6.3 8.0 10.4 11.8 11.2 11.2 Ammonia Production 13.0 9.2 9.4 10.0 9.6 10.6 11.2 Nitric Acid Production 12.1 11.3 10.5 10.7 10.9 11.6 10.2 Adipic Acid Production 15.2 7.1 5.5 3.9 5.4 4.3 7.0 Semiconductor Manufacture 3.6 4." 4.4 4.0 4.9 5.0 5.0 Carbon Dioxide Consumption 1.5 1.4 4.0 4.2 4.5 4.5 4.5 Electrical Transmission and Distribution 23.1 8.3 4.6 4.5 4.6 4.2 4.3 NjO from Product Uses 4.2 4.2 4.2 4.2 4.2 4.2 4.2 Urea Consumption for Non- Agricultural Purposes 3.8 3.7 4.4 4.1 1.5 4.2 4.0 HCFC-22 Production 46.1 20.0 5.5 4.1 5.0 4.3 2.8 Aluminum Production 28.3 7.6 6.4 6.2 5.4 4.8 2.7 Caprolactam, Glyoxal, and Glyoxylic Acid Production 1.7 2.1 2.0 2.0 2.0 2.0 2.0 Ferroalloy Production 2.2 1.4 1.9 1.8 1.9 2.0 1.8 Soda Ash Production 1.4 1." 1.7 1.7 1.7 1.7 1.7 Titanium Dioxide Production 1.2 1.8 1.5 1.7 1.7 1.6 1.6 Glass Production 1.5 1.9 1.2 1.3 1.3 1.3 1.3 Magnesium Production and Processing 5.2 2." 1.7 1.5 1.1 1.0 1.1 Phosphoric Acid Production 1.5 1.3 1.1 1.1 1.0 1.0 1.0 Zinc Production 0.6 1.0 1.5 1.4 1.0 0.9 0.9 Lead Production 0.5 0.6 0.5 0.5 0.5 0.5 0.5 Silicon Carbide Production and Consumption 0.4 0.2 0.2 0.2 0.2 0.2 0.2 Agriculture 489.2 520.0 519.8 543.1 539.8 566.9 562.6 Agricultural Soil Management 250.5 253.5 247.9 276.6 274.0 295.0 283.6 Enteric Fermentation 164.2 168.9 166.7 165.5 164.2 166.5 170.1 Manure Management 51.1 72.9 83.2 80.8 80.4 84.0 85.9 Rice Cultivation 16.0 16.7 11.3 11.5 12.7 12.3 13.7 Urea Fertilization 2.4 3.5 4.3 4.4 4.5 4.9 5.1 Liming 4.7 4.3 6.0 3.9 3.6 3.8 3.9 Field Burning of Agricultural Residues 0.3 0.3 0.4 0.4 0.4 0.4 0.4 Waste 199.3 156.4 140.4 136.7 136.5 135.6 131.5 Landfills 179.6 132.7 117.0 113.3 112.7 111.7 107.7 Wastewater Treatment 19.1 20.2 19.7 19.6 19.8 20.0 19.8 Composting 0.7 3.5 3.7 3.9 4.0 4.0 4.0 Total Emissions3 6,369.2 7,326.4 6,549.4 6,735.6 6,795.6 6,677.3 6,546.2 Land Use, Land-Use Change, and Forestry (819.6) (731.1) (753.5) (735.8) (740.4) (695.2) (716.8) Forest land (784.3) (730.0) (723.3) (733.3) (731.7) (709.9) (714.2) 2-8 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2016 ------- 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 Cropland 2.4 (0.7) 1.3 11.9 11.2 16.8 13.8 Grassland 13.8 25.3 0.8 18.5 14.7 33.6 21.0 Wetlands (4.0) (5.3) (4.1) (4.1) (4.1) (4.1) (4.2) Settlements (47.6) (20.5) (28.3) (28.8) (30.5) (31.5) (33.3) Net Emission (Sources and Sinks)b 5,549.6 6,595.3 5,795.9 5,999.9 6,055.2 5,982.1 5,829.3 Notes: Total emissions presented without LULUCF. Net emissions presented with LULUCF. a Total emissions without LULUCF. b Net emissions with LULUCF. Notes: Totals may not sum due to independent rounding. Parentheses indicate negative values or sequestration. Energy Energy-related activities, primarily fossil fuel combustion, accounted for the vast majority of U.S. CO2 emissions for the period of 1990 through 2016. 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 Figure 2-6). In 2016, approximately 81 percent of the energy consumed in the United States (on a Btu basis) was produced through the combustion of fossil fuels. The remaining 19 percent came from other energy sources such as hydropower, biomass, nuclear, wind, and solar energy. A discussion of specific trends related to CO2 as well as other greenhouse gas emissions from energy use is presented in the Energy chapter. Energy-related activities are also responsible for CH4 and N20 emissions (43 percent and 10 percent of total U.S. emissions of each gas, respectively). Table 2-4 presents greenhouse gas emissions from the Energy chapter, by source and gas. Figure 2-5: 2016 Energy Chapter Greenhouse Gas Sources (MMT CO2 Eq.) COi Emissions from Fossil Fuel Combustion Natural Gas Systems Non-Energy Use of Fuels Petroleum Systems Coal Mining Non-COs Emissions from Stationary Combustion Non-CO: Emissions from Mobile Combustion Incineration of Waste Abandoned Oil and Gas Wells Abandoned Underground Coal Mines 0 50 100 150 200 250 300 MMT CO: Eq. 4,977 Energy as a Portion of all Emissions 83.7% Trends 2-9 ------- 1 Figure 2-6: 2016 U.S. Fossil Carbon Flows (MMT CO2 Eq.) 2 4 Table 2-4: Emissions from Energy (MMT CO2 Eq.) Gas/Source 1990 2005 2012 2013 2014 2015 2016 CO2 4,922.4 5,952.7 5,203.5 5,361.6 5,404.4 5,269.4 5,160.7 Fossil Fuel Combustion 4,755.8 5,759.1 5,029.8 5,162.3 5,206.1 5,059.3 4,976.7 Electric Power 1,820.8 2,400.9 2,022.2 2,038.1 2,038.0 1,900.7 1,808.8 Transportation 1,467.2 1,855.8 1,661.9 1,677.6 1,717.1 1,735.5 1,794.9 Industrial 874.5 867.8 818.4 848.7 830.8 819.3 807.6 Residential 338.3 357.8 282.5 329.7 345.3 316.8 296.2 Commercial 227.4 227.0 201.3 225.7 233.6 245.6 227.9 U.S. Territories 27.6 49.7 43.5 42.5 41.4 41.4 41.4 Non-Energy Use of Fuels 119.6 141.7 113.3 133.2 127.8 135.1 121.0 Natural Gas Systems 29.7 22.5 24.4 26.0 27.0 26.3 26.7 Petroleum Systems 9.4 17.0 25.6 29.7 32.9 38.0 25.5 Incineration of Waste 8.0 12.5 10.4 10.4 10.6 10.7 10.7 Biomass- Wood" 215.2 206.9 194.9 211.6 218.9 201.5 190.2 International Bunker Fuelsb 103.5 113.1 105.8 99.8 103.4 110.9 114.4 Biofuels-Ethanol" 4.2 22.9 72.8 74.7 76.1 78.9 81.2 Biofuels-Biodiesel" 0.0 0.9 8.5 13.5 13.3 14.1 19.6 CH4 364.7 286.7 283.1 288.7 295.4 289.5 279.2 Natural Gas Systems 193.7 160.0 156.8 159.6 164.2 164.4 162.1 Coal Mining 96.5 64.1 66.5 64.6 64.6 61.2 53.8 Petroleum Systems 42.3 34.7 35.4 38.8 41.0 39.4 39.3 Stationary Combustion 8.6 7.9 7.3 8.7 00 OO 7.8 7.2 Abandoned Oil and Gas Wells 6.5 6.9 7.0 7.0 7.1 7.2 7.1 Abandoned Underground Coal Mines 7.2 6.6 6.2 6.2 6.3 6.4 6.7 Mobile Combustion 9.8 6.6 4.0 3.7 3.4 3.1 3.0 Incineration of Waste + + + + + + + International Bunker Fuelsb 0.2 0.1 0.1 0.1 0.1 0.1 0.1 N2O 53.1 56.4 40.9 40.9 39.4 37.1 36.5 Stationary Combustion 11.1 17.5 16.8 18.6 18.9 18.0 18.4 Mobile Combustion 41.5 38.4 23.8 22.0 20.2 18.8 17.8 Incineration of Waste 0.5 0.4 0.3 0.3 0.3 0.3 0.3 International Bunker Fuelsb 0.9 1.0 0.9 0.9 0.9 0.9 1.0 Total 5,340.2 6,295.7 5,527.6 5,691.1 5,739.1 5,596.0 5,476.4 + Does not exceed 0.05 MMT CO2 Eq. Coal Emissions 1,316 NEU Exports 146 Combustion Emissions 1,307 NEU Emissions 5 Combustion Emissions 1,477 NEU Emissions Atmospherii Emissions 5,292 Domestic Fossil Fuel Production 4,495 Apparent Consumption 5,390 Petroleum Emissions 2,299 Combustion Emissions 2,193 Petroleum 1,367 - Natural Gas Liquids, Liquefied Refinery Gas, & Other Liquids 295 Petroleum 1,383 „ Fossil Fuel Energy Imports 1,796 Balancing Item (110) NEU U.S. Territories Stock Changes Fossil Fuel Energy Exports Non-Energy Use Carbon Sequestered Note: Totals may not sum due to independent rounding. The "Balancing Item" above accounts for the statistical imbalances and unknowns in the reported data sets combined here. NEU = Non-Energy Use 2-10 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2016 ------- 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 a Emissions from Wood 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. Note: Totals may not sum due to independent rounding. CO2 Emissions from Fossil Fuel Combustion As the largest contributor to U.S. greenhouse gas emissions, CO2 from fossil fuel combustion has accounted for approximately 77 percent of GWP-weighted emissions for the entire time series since 1990. Emissions from this source category grew by 4.6 percent (220.9 MMT CO2 Eq.) from 1990 to 2016 and were responsible for most of the increase in national emissions during this period. Conversely, CO2 emissions from fossil fuel combustion decreased from 2005 levels by 782.3 MMT CO2 Eq., a decrease of approximately 13.6 percent between 2005 and 2016. From 2015 to 2016, these emissions decreased by 1.6 percent (82.6 MMT CO2 Eq.). Historically, changes in emissions from fossil fuel combustion have been the dominant factor affecting U.S. emission trends. Changes in CO2 emissions from fossil fuel combustion are influenced by many long-term and short-term factors, including population and economic growth, energy price fluctuations and market trends, technological changes, energy fuel choices, and seasonal temperatures. On an annual basis, the overall consumption and mix of fossil fuels in the United States fluctuates primarily 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 a number of factors including the relative price of coal and alternative sources, the ability to switch fuels, and longer term trends in coal markets. Likewise, warmer winters will lead to a decrease in heating degree days and result in a decreased demand for heating fuel and electricity for heat in the residential and commercial sector, which leads to a decrease in emissions from reduced fuel consumption. Energy-related CO2 emissions also depend on the type of fuel consumed or energy used and its C intensity. Producing a unit of heat or electricity using natural gas instead of coal, for example, can reduce the CO2 emissions because of the lower C content of natural gas (see Table A-3 9 in Annex 2.1 for more detail on the C Content Coefficient of different fossil fuels). Trends in CO2 emissions from fossil fuel combustion over the past decade have been in large part driven by the electric power sector, which historically has accounted for the majority of emissions from this source (see Figure 2-7). In recent years, the types of fuel consumed to produce electricity have changed. Carbon dioxide emissions from coal consumption for electric power generation decreased by 36.7 percent since 2008, and there has been a shift to the use of less-CCh-intensive natural gas to supply electricity. There has also been a rapid increase in renewable energy capacity additions in the electric power sector in recent years. In 2016, renewable energy sources accounted for 63 percent of capacity additions with natural gas accounting for the majority of the remaining additions. The share of renewable energy capacity additions has grown significantly since 2010, when renewable energy sources accounted for only 28 percent of total capacity additions (EIA 2017d). The decrease in coal-powered electricity generation and increase in renewable energy capacity have contributed to a 4.8 percent decrease in emissions from electric power generation from 2015 to 2016 (see Figure 2-9), and lower CO2 emissions from fossil fuel combustion over the time series (i.e., 1990 through 2016). Total petroleum use is another major driver of CO2 emissions from fossil fuel combustion, particularly in the transportation sector, which represents the second largest source of CO2 emissions from fossil fuel combustion. Emissions from petroleum consumption for transportation have increased by 22.6 percent since 1990, which can be primarily attributed to a 46.8 percent increase in vehicle miles traveled (VMT) over the time series. Fuel economy of light-duty vehicles is another important factor. 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 30 percent of new vehicle sales in 1990 to 48 percent in 2004. Since 2005, average new vehicle fuel economy has increased while the market share of light-duty trucks has decreased. Total transportation sector CO2 emissions have increased by 5.9 percent since 2010. The overall trends in CO2 emissions from fossil fuel combustion in the residential and commercial sectors closely align with heating degree days. Emissions from the residential and commercial sectors decreased by 6.5 percent and 7.2 percent from 2015 to 2016, respectively. This trend can be largely attributed to a 5 percent decrease in heating degree days which led to a decreased demand for heating fuel and electricity for heat in the residential and commercial sectors. In addition, an increase in energy efficiency standards and the use of energy efficient products Trends 2-11 ------- 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 in residential and commercial buildings has resulted in an overall reduction in energy use, contributing to a decrease in emissions in both of these sectors (EIA 2017b). Combined residential and commercial sector emissions have decreased by 6.2 percent since 2010. The increase in transportation sector petroleum CO2 emissions from 2015 to 2016 offset emission reductions from decreased coal use in the electric power sector and decreased demand for heating fuel in the residential and commercial sectors. Although emissions from the transportation sector have increased, emissions from all other sectors and U.S. Territories have decreased in recent years, contributing to a 1.6 percent decrease in total CO2 emissions from fossil fuel combustion from 2015 to 2016 and a 7.3 percent reduction since 2010. Carbon dioxide emissions from fossil fuel combustion are presented in Table 2-5 based on the underlying U.S. energy consumer data collected by the U.S. Energy Information Administration (EIA). 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. (Any additional analysis and refinement of the EIA data is further explained in the Energy chapter of this report.) • Electric Power. EIA's fuel consumption data for the electric power sector are comprised 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. (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. EIA's fuel 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.) • Transportation. EIA'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. • Residential. EIA's fuel consumption data for the residential sector consist of living quarters for private households. • Commercial. EIA'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 and Figure 2-7 summarize CO2 emissions from fossil fuel combustion by end-use sector. Figure 2-8 further describes the total emissions from fossil fuel combustion, separated by end-use sector, including CH4 and N20 in addition to CO2. Table 2-5: CO2 Emissions from Fossil Fuel Combustion by End-Use Sector (MMT CO2 Eq.) End-Use Sector 1990 2005 2012 2013 2014 2015 2016 Transportation 1,470.2 1,860.5 1,665.8 1,681.6 1,721.2 1,739.2 1,798.4 Combustion 1,467.2 1,855.8 1,661.9 1,677.6 1,717.1 1,735.5 1,794.9 Electricity 3.0 4.7 3.9 4.0 4.1 3.7 3.5 Industrial 1,561.3 1,604.4 1,411.2 1,443.4 1,424.0 1,368.8 1,313.8 Combustion 874.5 867.8 818.4 848.7 830.8 819.3 807.6 Electricity 686.7 736.6 592.8 594.7 593.2 549.6 506.2 Residential 931.4 1,214.1 1,007.8 1,064.6 1,080.0 1,001.1 957.0 Combustion 338.3 357.8 282.5 329.7 345.3 316.8 296.2 Electricity 593.0 856.3 725.3 734.9 734.7 684.3 660.7 Commercial 765.3 1,030.3 901.6 930.2 939.6 908.8 866.2 Combustion 227.4 227.0 201.3 225.7 233.6 245.6 227.9 Electricity 538.0 803.3 700.3 704.5 706.0 663.1 638.3 U.S. Territories3 27.6 49.7 43.5 42.5 41.4 41.4 41.4 Total 4,755.8 5,759.1 5,029.8 5,162.3 5,206.1 5,059.3 4,976.7 2-12 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2016 ------- 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 Electric Power 1,820.8 2,400.9 2,022.2 2,038.1 2,038.0 1,900.7 1,808.8 a Fuel consumption by U.S. Territories (i.e., American Samoa, Guam, Puerto Rico, U.S. Virgin Islands, Wake Island, and other U.S. Pacific Islands) is included in this report. 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. Figure 2-7: 2016 CO2 Emissions from Fossil Fuel Combustion by Sector and Fuel Type (MMT COz Eq.) 2,500 2,000 S 1,500 0 u 1- 1 1,000 500 Relative Contribution by Fuel Type I Petroleum Coal I Natural Gas 1,795 1,809 228 41 U.S. Territories Commercial Residential Industrial Transportation Electric Power Note on Figure 2-7: Fossil Fuel Combustion includes electric power, which also includes emissions of less than 0.5 MMT CO2 Eq. from geothermal-based generation. Figure 2-8: 2016 End-Use Sector Emissions of CO2 from Fossil Fuel Combustion (MMT CO2 Eq.) 2,000 1,500- 8 1,000- 500- I Direct Fossil Fuel Combustion Indirect Fossil Fuel Combustion 1,798 U.S. Territories Commercial Residential Industrial Transportation The main driver of emissions in the Energy sector is CO2 from fossil fuel combustion. Electric power is the largest emitter of CO2, and electricity generators used 33 percent of U.S. energy from fossil fuels and emitted 36 percent of the CO2 from fossil fuel combustion in 2016. Changes in electricity demand and the carbon intensity of fuels used for electric power have a significant impact on CO2 emissions. Emissions from the electric power sector have decreased by approximately 0.2 percent since 1990, and the carbon intensity of the electric power sector, in terms of CO2 Eq. per QBtu input has significantly decreased by 12 percent during that same timeframe. This decoupling of electric power and the resulting emissions is shown below in Figure 2-9. Trends 2-13 ------- 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 Figure 2 -9: Electric Power Generation (Billion kWh) and Emissions (MMT CO2 Eq.) I Petroleum-based Generation (Billion kWh) I Nuclear-based Generation (Billion kWh) Renewable-based Generation (Billion kWh) I Natural Gas-based Generation (Billion kWh) Coal-based Generation (Billion kWh) I Total Emissions (MMT CO; Eq.) [Right Axis] 3,500 3,000 Electric power emissions can also be allocated to the end-use sectors that are using that electricity, as presented in Table 2-5. The transportation end-use sector accounted for 1,798.4 MMT CO2 Eq. in 2016 or approximately 36 percent of total CO2 emissions from fossil fuel combustion. The industrial end-use sector accounted for 26 percent of CO2 emissions from fossil fuel combustion. The residential and commercial end-use sectors accounted for 19 and 17 percent, respectively, of CO2 emissions from fossil fuel combustion. Both of these end-use sectors were heavily reliant on electricity for meeting energy needs, with electricity use for lighting, heating, air conditioning, and operating appliances contributing 69 and 74 percent of emissions from the residential and commercial end-use sectors, respectively. Significant trends in emissions from energy source categories over the twenty-seven-year period from 1990 through 2016 included the following: • Total CO2 emissions from fossil fuel combustion increased from 4,755.8 MMT CO2 Eq. in 1990 to 4,976.7 MMT CO2 Eq. in 2016—a 4.6 percent total increase over the twenty-seven-year period. From 2015 to 2016, these emissions decreased by 82.6 MMT CO2 Eq. (1.6 percent). • Methane emissions from natural gas systems and petroleum systems (combined here) decreased from 236.0 MMT CO2 Eq. in 1990 to 201.4 MMT CO2 Eq. in 2016 (34.7 MMT CO2 Eq. or 14.7 percent decrease from 1990 to 2016). Natural gas systems CH4 emissions decreased by 31.6 MMT CO2 Eq. (16.3 percent) since 1990, largely due to a decrease in emissions from distribution, transmission and storage, processing, and exploration. The decrease in distribution emissions is largely attributed to increased use of plastic piping, which lias lower emissions than other pipe materials, and station upgrades at metering and regulating (M&R) stations. The decrease in transmission and storage emissions is largely due to reduced compressor station emissions (including emissions from compressors and leaks). Petroleum systems CH4 emissions decreased by 3.0 MMT CO2 Eq. (or 7.2 percent) since 1990. This decrease is due primarily to decreases in tank emissions and associated gas venting. Carbon dioxide emissions from natural gas and petroleum systems increased by 34% from 1990 to 2016, due to increases in flaring emissions. • Carbon dioxide emissions from non-energy uses of fossil fuels increased by 1.5 MMT CO2 Eq. (1.2 percent) from 1990 through 2016. Emissions from non-energy uses of fossil fuels were 121.0 MMT CO2 Eq. in 2016, which constituted 2.3 percent of total national CO2 emissions, approximately the same proportion as in 1990. 2-14 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2016 ------- 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 • Nitrous oxide emissions from stationary combustion increased by 7.2 MMT CO2 Eq. (64.9 percent) from 1990 through 2016. Nitrous oxide emissions from this source increased primarily as a result of an increase in the number of coal fluidized bed boilers in the electric power sector. • Nitrous oxide emissions from mobile combustion decreased by 23.6 MMT CO2 Eq. (57.0 percent) from 1990 through 2016, primarily as a result of N20 national emission control standards and emission control technologies for on-road vehicles. • Carbon dioxide emissions from incineration of waste (10.7 MMT CO2 Eq. in 2016) increased by 2.7 MMT CO2 Eq. (34.3 percent) from 1990 through 2016, as the volume of plastics and other fossil carbon- containing materials in municipal solid waste grew. Industrial Processes and Product Use The Industrial Processes and Product Use (IPPU) chapter includes greenhouse gas emissions occurring from industrial processes and from the use of greenhouse gases in products. In many cases, greenhouse gas emissions are produced as the byproducts of many non-energy-related industrial activities. For example, industrial processes can chemically transform raw materials, which often release waste gases such as CO2, CH4, N2O, and fluorinated gases (e.g., HFC-23). These processes are shown in Figure 2-10. Industrial manufacturing processes and use by end-consumers also release HFCs, PFCs, SF6, and NF3 and other fluorinated 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 employed and emitted by a number of other industrial sources in the United States. These industries include semiconductor manufacture, electric power transmission and distribution and magnesium metal production and processing. In addition, N20 is used in and emitted by semiconductor manufacturing and anesthetic and aerosol applications. Table 2-6 presents greenhouse gas emissions from industrial processes by source category. Figure 2-10: 2016 Industrial Processes and Product Use Chapter Greenhouse Gas Sources (MMT COz Eq.) Substitution of Ozorie Depleting Substances Iron and Steel Production &. Metallurgical Coke Production Cement Production Petrochemical Production Lime Production Other Process Uses of Carbonates Ammonia Production Nitric Acid Production Adipic Acid Production Semiconductor Manufacture Carbon Dioxide Consumption Electrical Transmission and Distribution NsO from Product Uses Urea Consumption for Non-Agricultural Purposes HCFC-22 Production Aluminum Production Caprolactam, Glyoxal, and Glyoxylic Acid Production Ferroalloy Production Soda Ash Production Titanium Dioxide Production Glass Production Magnesium Production and Processing Phosphoric Acid Production Zinc Production Lead Production Silicon Carbide Production and Consumption 174 ¦ ¦ ¦ I I I I I I I < 0.5 Industrial Processes and Product Use as a Portion of all Emissions 5.7% 0 10 20 30 40 50 60 70 MMT CO, Eg. Trends 2-15 ------- 1 Table 2-6: Emissions from Industrial Processes and Product Use (MMT CO2 Eq.) Gas/Source 1990 2005 2012 2013 2014 2015 2016 CO2 207.3 190.2 169.9 171.8 177.9 171.4 163.6 Iron and Steel Production & Metallurgical Coke Production 101.5 68.0 55.4 53.3 58.2 47.7 42.2 Iron and Steel Production 99.0 66.0 54.9 51.5 56.2 44.9 40.9 Metallurgical Coke Production 2.5 2.0 0.5 1.8 2.0 2.8 1.3 Cement Production 33.5 46.2 35.3 36.4 39.4 39.9 39.4 Petrochemical Production 21.2 26.8 26.5 26.4 26.5 28.1 27.4 Lime Production 11.7 14.6 13.8 14.0 14.2 13.3 13.3 Other Process Uses of Carbonates 4.9 6.3 8.0 10.4 11.8 11.2 11.2 Ammonia Production 13.0 9.2 9.4 10.0 9.6 10.6 11.2 Carbon Dioxide Consumption 1.5 1.4 4.0 4.2 4.5 4.5 4.5 Urea Consumption for Non-Agricultural Purposes 3.8 3.7 4.4 4.1 1.5 4.2 4.0 Ferroalloy Production 2.2 1.4 1.9 1.8 1.9 2.0 1.8 Soda Ash Production 1.4 1.7 1.7 1.7 1.7 1.7 1.7 Titanium Dioxide Production 1.2 1.8 1.5 1.7 1.7 1.6 1.6 Aluminum Production 6.8 4.1 3.4 3.3 2.8 2.8 1.3 Glass Production 1.5 1.9 1.2 1.3 1.3 1.3 1.3 Phosphoric Acid Production 1.5 1.3 1.1 1.1 1.0 1.0 1.0 Zinc Production 0.6 1.0 1.5 1.4 1.0 0.9 0.9 Lead Production 0.5 0.6 0.5 0.5 0.5 0.5 0.5 Silicon Carbide Production and Consumption 0.4 0.2 0.2 0.2 0.2 0.2 0.2 Magnesium Production and Processing + + + + + + + CH4 0.3 0.1 0.1 0.1 0.2 0.2 0.2 Petrochemical Production 0.2 0.1 0.1 0.1 0.1 0.2 0.2 Ferroalloy Production + + + + + + + Silicon Carbide Production and Consumption + + + + + + + Iron and Steel Production & Metallurgical Coke Production + + + + + + + Iron and Steel Production + + + + + + + Metallurgical Coke Production 0.0 0.0 0.0 0.0 0.0 0.0 0.0 N2O 33.3 24.9 22.4 21.0 22.8 22.3 23.7 Nitric Acid Production 12.1 11.3 10.5 10.7 10.9 11.6 10.2 Adipic Acid Production 15.2 7.1 5.5 3.9 5.4 4.3 7.0 N2O from Product Uses 4.2 4.2 4.2 4.2 4.2 4.2 4.2 Caprolactam, Glyoxal, and Glyoxylic Acid 1.7 2.1 2.0 2.0 2.0 2.0 2.0 Semiconductor Manufacturing + 0.1 0.2 0.2 0.2 0.2 0.2 HFCs 46.6 120.0 156.0 159.1 166.8 173.3 177.1 Substitution of Ozone Depleting Substances3 0.3 99.8 150.3 154.8 161.4 168.6 173.9 HCFC-22 Production 46.1 20.0 5.5 4.1 5.0 4.3 2.8 Semiconductor Manufacturing 0.2 0.2 0.2 0.2 0.3 0.3 0.3 Magnesium Production and Processing 0.0 0.0 + 0.1 0.1 0.1 0.1 PFCs 24.3 6.7 5.9 5.8 5.6 5.1 4.3 Semiconductor Manufacturing 2.8 3.3 3.0 2.8 3.1 3.1 3.0 Aluminum Production 21.5 3.4 2.9 3.0 2.5 2.0 1.4 Substitution of Ozone Depleting Substances 0.0 + + + + + + SF« 28.8 11.7 6.6 6.3 6.3 5.9 6.2 Electrical Transmission and Distribution 23.1 8.3 4.6 4.5 4.6 4.2 4.3 Magnesium Production and Processing 5.2 2.7 1.6 1.5 1.0 0.9 1.0 Semiconductor Manufacturing 0.5 0.7 0.3 0.4 0.7 0.7 0.8 NF3 + 0.5 0.6 0.6 0.5 0.6 0.6 Semiconductor Manufacturing + 0.5 0.6 0.6 0.5 0.6 0.6 Total 340.5 354.2 361.6 364.7 380.2 378.8 375.7 + Does not exceed 0.05 MMT CO2 Eq. 2-16 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2016 ------- 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 a Small amounts of PFC emissions also result from this source. Note: Totals may not sum due to independent rounding. Overall, emissions from the IPPU sector increased by 10.3 percent from 1990 to 2016. Significant trends in emissions from IPPU source categories over the twenty-seven-year period from 1990 through 2016 included the following: • Hydrofluorocarbon and perfluorocarbon emissions from ODS substitutes have been increasing from small amounts in 1990 to 173.9 MMT CO2 Eq. in 2016. This increase was in large part the result of efforts to phase out chlorofluorocarbons (CFCs) and other ODSs in the United States. In the short term, this trend is expected to continue, and will likely continue over the next decade as hydrochlorofluorocarbons (HCFCs), which are interim substitutes in many applications, are themselves phased-out under the provisions of the Copenhagen Amendments to the Montreal Protocol. • Combined CO2 and CH4 emissions from iron and steel production and metallurgical coke production decreased by 11.5 percent to 42.2 MMT CO2 Eq. from 2015 to 2016, and have declined overall by 59.3 MMT CO2 Eq. (58.4 percent) from 1990 through 2016, due to restructuring of the industry, technological improvements, and increased scrap steel utilization. • Carbon dioxide emissions from ammonia production (11.2 MMT CO2 Eq. in 2016) decreased by 1.8 MMT CO2 Eq. (13.9 percent) 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. • Nitrous oxide emissions from adipic acid production were 7.0 MMT CO2 Eq. in 2016, and have decreased significantly since 1990 due to both the widespread installation of pollution control measures in the late 1990s and plant idling in the late 2000s. Emissions from adipic acid production have decreased by 53.9 percent since 1990 and by 58.5 percent since a peak in 1995. • PFC emissions from aluminum production decreased by 93.7 percent (20.1 MMT CO2 Eq.) from 1990 to 2016, due to both industry emission reduction efforts and lower domestic aluminum production. Agriculture Agricultural activities contribute directly to emissions of greenhouse gases through a variety of processes, including the following source categories: enteric fermentation in domestic livestock, livestock manure management, rice cultivation, agricultural soil management, liming, urea fertilization, and field burning of agricultural residues. Methane, N20, and CO2 were the primary greenhouse gases emitted by agricultural activities. In 2016, agricultural activities were responsible for emissions of 562.6 MMT CO2 Eq., or 8.6 percent of total U.S. greenhouse gas emissions. Methane emissions from enteric fermentation and manure management represented approximately 25.9 percent and 10.3 percent of total CH4 emissions from anthropogenic activities, respectively, in 2016. Agricultural soil management activities, such as application of synthetic and organic fertilizers, deposition of livestock manure, and growing N-ftxing plants, were the largest source of U.S. N20 emissions in 2016, accounting for 76.9 percent. 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 2-11 and Table 2-7 illustrate agricultural greenhouse gas emissions by source. Trends 2-17 ------- 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 Figure 2-11: 2016 Agriculture Chapter Greenhouse Gas Sources (MMT CO2 Eq.) Agricultural Soil Management Enteric Fermentation Manure Management Rice Cultivation Urea Fertilization Liming Field Burning of Agricultural Residues 284 ¦ Agriculture as a Portion of all Emissions 8.6% < 0.5 0 20 40 60 80 100 120 140 160 180 MMT COi Eg. Table 2-7: Emissions from Agriculture (MMT CO2 Eq.) Gas/Source 1990 2005 2012 2013 2014 2015 2016 CO2 7.1 7.9 10.3 8.4 8.1 8.7 9.0 Urea Fertilization 2.4 3.5 4.3 4.4 4.5 4.9 5.1 Liming 4.7 4.3 6.0 3.9 3.6 3.8 3.9 CH4 217.6 242.1 244.0 240.6 240.1 245.4 251.8 Enteric Fermentation 164.2 168.9 166.7 165.5 164.2 166.5 170.1 Manure Management 37.2 56.3 65.6 63.3 62.9 66.3 67.7 Rice Cultivation 16.0 16.7 11.3 11.5 12.7 12.3 13.7 Field Burning of Agricultural Residues 0.2 0.2 0.3 0.3 0.3 0.3 0.3 N2O 264.5 270.1 265.5 294.2 291.6 312.8 301.8 Agricultural Soil Management 250.5 253.5 247.9 276.6 274.0 295.0 283.6 Manure Management 14.0 16.5 17.5 17.5 17.5 17.7 18.1 Field Burning of Agricultural Residues 0.1 0.1 0.1 0.1 0.1 0.1 0.1 Total 489.2 520.0 519.8 543.1 539.8 566.9 562.6 Note: Totals may not sum due to independent rounding. Some significant trends in U.S. emissions from Agriculture source categories include the following: • Agricultural soils is the largest anthropogenic source of N20 emissions in the United States, accounting for approximately 76.9 percent of N2O emissions in 2016. Estimated emissions from this source in 2016 were 283.6 MMT CO2 Eq. Annual N20 emissions from agricultural soils fluctuated between 1990 and 2016, although overall emissions were 13.2 percent higher in 2016 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 2016, enteric fermentation CH4 emissions were 170.1 MMT CO2 Eq. (25.9 percent of total CH4 emissions), which represents an increase of 6.0 MMT CO2 Eq. (3.6 percent) since 1990. This increase in emissions from 1990 to 2016 in enteric fermentation generally follows the increasing trends in cattle populations. 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 2-18 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2016 ------- 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 as beef cattle populations again decreased. Emissions increased from 2014 to 2016, consistent with an increase in beef cattle population over those same years. • Liming and urea fertilization are the only source of CO2 emissions reported in the Agriculture sector. Estimated emissions from these sources were 3.9 and 5.1 MMT CO2 Eq., respectively. Liming and urea fertilization emissions increased by 2.3 percent and 4.3 percent, respectively, relative to 2015, and decreased by 17.2 percent and increased by 110.9 percent, respectively since 1990. • Overall, emissions from manure management increased 67.8 percent between 1990 and 2016. This encompassed an increase of 82.2 percent for CH4, from 37.2 MMT CO2 Eq. in 1990 to 67.7 MMT CO2 Eq. in 2016; and an increase of 29.6 percent for N2O, from 14.0 MMT CO2 Eq. in 1990 to 18.1 MMT CO2 Eq. in 2016. The majority of the increase observed in CH4 resulted from swine and dairy cattle manure, where emissions increased 63 and 140 percent, respectively, from 1990 to 2016. From 2015 to 2016, there was a 2.2 percent increase in total CH4 emissions from manure management, mainly due to minor shifts in the animal populations and the resultant effects on manure management system allocations. 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 (C) stock fluxes on these lands and cause emissions of CH4 and N20. Overall, managed land is a net sink for CO2 (C sequestration) in the United States. The drivers of fluxes on managed lands include, for example, forest management practices, tree planting in urban areas, the management of agricultural soils, the landfilling of yard trimmings and food scraps, and activities that cause changes in C stocks in coastal wetlands. The main drivers for net forest sequestration include net forest growth, increasing forest area, and a net accumulation of C stocks in harvested wood pools. The net sequestration in Settlements Remaining Settlements, is driven primarily by C stock gains in urban forests through net tree growth and increased urban area, as well as long- term accumulation of C in landfills from additions of yard trimmings and food scraps. The LULUCF sector in 2016 resulted in a net increase in C stocks (i.e., net CO2 removals) of 754.9 MMT CO2 Eq. (Table 2-8).1 This represents an offset of approximately 11.5 percent of total (i.e., gross) greenhouse gas emissions in 2016. Emissions of CHi and N2O from LULUCF activities in 2016 were 38.1 MMT CO2 Eq. and represent 0.6 percent of total greenhouse gas emissions.2 Between 1990 and 2016, total C sequestration in the LULUCF sector decreased by 9.1 percent, primarily due to a decrease in the rate of net C accumulation in forests and Cropland Remaining Cropland, as well as an increase in CO2 emissions from Land Converted to Settlements. Forest fires were the largest source of CHi emissions from LULUCF in 2016, totaling 18.5 MMT CO2 Eq. (740 kt of CH4). Coastal Wetlands Remaining Coastal Wetlands resulted in CH4 emissions of 3.6 MMT CO2 Eq. (143 kt of CH4). Grassland fires resulted in CH4 emissions of 0.3 MMT CO2 Eq. (11 kt of CH4). Peatlands Remaining Peatlands, Land Converted to Wetlands, and Drained Organic Soils resulted in CH4 emissions of less than 0.05 MMT CO2 Eq. each. Forest fires were also the largest source of N20 emissions from LULUCF in 2016, totaling 12.2 MMT CO2 Eq. (41 kt of N20). Nitrous oxide emissions from fertilizer application to settlement soils in 2016 totaled to 2.5 MMT CO2 Eq. (8 kt of N20). Additionally, the application of synthetic fertilizers to forest soils in 2016 resulted in N20 emissions of 0.5 MMT CO2 Eq. (2 kt of N20). Grassland fires resulted in N20 emissions of 0.3 MMT CO2 Eq. (1 kt of N2O). Coastal Wetlands Remaining Coastal Wetlands and Drained Organic Soils resulted in N2O emissions of 0.1 MMT CO2 Eq. each (less than 0.5 kt of N2O). Peatlands Remaining Peatlands resulted U1N2O emissions of less than 0.05 MMT C02 Eq. 1 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. 2 LULUCF emissions include the CH4 and N2O 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 N2O emissions from Forest Soils and Settlement Soils. Trends 2-19 ------- 1 Carbon dioxide removals from C stock changes are presented in Figure 2-12 and Table 2-8 along with CH4 and N20 2 emissions for LULUCF source categories. 3 Figure 2-12: 2016 LULUCF Chapter Greenhouse Gas Sources and Sinks (MMT CO2 Eq.) Forest Land Remaining Forest Land -670.5 | Settlements Remaining Settlements Land Converted to Forest Land Cropland Remaining Cropland Wetlands Remaining Wetlands Grassland Remaining Grassland Land Converted to Wetlands |< 0.5| Non-COi Emissions from Peatlands Remaining Peatlands | |< 0.5| CHi Emissions from Land Converted to Coastal Wetlands I |< 0.5| Non-COi Emissions from Drained Organic Soils | |< 0.5| NiO Emissions from Forest Soils | | < 0.51 Non-COi Emissions from Grassland Fires N:0 Emissions from Settlement Soils Non-COz Emissions from Coastal Wetlands Remaining Coastal Wetlands Land Converted to Grassland Land Converted to Cropland g Carbon Stock Change Non-COi Emissions from Forest Fires | Emissions Land Converted to Settlements -300 -250 -200 -150 -100 -50 0 50 100 MMT CO* Eg. 4 5 Table 2-8: U.S. Greenhouse Gas Emissions and Removals (Net Flux) from Land Use, Land- 6 Use Change, and Forestry (MMT CO2 Eq.) Gas/Land-Use Category 1990 2005 2012 2013 2014 2015 2016 Carbon Stock Change3 (830.2) (754.2) (779.5) (755.0) (760.0) (733.4) (754.9) Forest Land Remaining Forest Land (697.7) (664.6) (666.9) (670.9) (669.3) (666.2) (670.5) Land Converted to Forest Land (92.0) (81.6) (74.9) (74.9) (75.0) (75.0) (75.0) Cropland Remaining Cropland (40.9) (26.5) (21.4) (11.4) (12.0) (6.3) (9.9) Land Converted to Cropland 43.3 25.9 22.7 23.3 23.2 23.2 23.8 Grassland Remaining Grassland (4.2) 5.5 (20.8) (3.7) (7.5) 9.6 (1.6) Land Converted to Grassland 17.9 19.2 20.4 21.9 21.5 23.3 22.0 Wetlands Remaining Wetlands (7.6) (8.9) (7.7) (7.8) (7.8) (7.8) (7.9) Land Converted to Wetlands (+) (+) (+) (+) (+) (+) (+) Settlements Remaining Settlements (86.2) (91.4) (99.2) (99.8) (101.2) (102.2) (103.7) Land Converted to Settlements 37.2 68.4 68.3 68.3 68.2 68.1 68.0 CH4 6.7 13.3 15.0 10.9 11.2 22.4 22.4 Forest Land Remaining Forest Land: Forest Fires 3.2 9.4 10.8 7.2 7.2 18.5 18.5 Wetlands Remaining Wetlands: Coastal Wetlands Remaining Coastal Wetlands 3.4 3.5 3.5 3.6 3.6 3.6 3.6 Grassland Remaining Grassland: Grassland Fires 0.1 0.3 0.6 0.2 0.4 0.3 0.3 Forest Land Remaining Forest Land: Drained Organic Soils + + + + + + + Land Converted to Wetlands: Land Converted to Coastal Wetlands + + + + + + + Wetlands Remaining Wetlands: Peatlands Remaining Peatlands + + + + + + + N2O 3.9 9.7 11.1 8.3 8.4 15.8 15.7 Forest Land Remaining Forest Land: Forest Fires 2.1 6.2 7.1 4.8 4.7 12.2 12.2 Settlements Remaining Settlements: Settlement Soilsb 1.4 2.5 2.7 2.6 2.6 2.5 2.5 2-20 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2016 ------- 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 Forest Land Remaining Forest Land: Forest Soilsc 0.1 0.5 0.5 0.5 0.5 0.5 0.5 Grassland Remaining Grassland: Grassland Fires 0.1 0.3 0.6 0.2 0.4 0.3 0.3 Wetlands Remaining Wetlands: Coastal Wetlands Remaining Coastal Wetlands 0.1 0.2 0.1 0.1 0.1 0.1 0.1 Forest Land Remaining Forest Land: Drained Organic Soils 0.1 0.1 0.1 0.1 0.1 0.1 0.1 Wetlands Remaining Wetlands: Peatlands Remaining Peatlands + + + + + + + LULUCF Emissions'1 10.6 23.0 26.1 19.2 19.6 38.2 38.1 LULUCF Carbon Stock Change3 (830.2) (754.2) (779.5) (755.0) (760.0) (733.4) (754.9) LULUCF Sector Net Total' (819.6) (731.1) (753.5) (735.8) (740.4) (695.2) (716.8) + Absolute value does not exceed 0.05 MMT CO2 Eq. a 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. b Estimates include emissions from N fertilizer additions on both Settlements Remaining Settlements and Land Converted to Settlements. c Estimates include emissions from N fertilizer additions on both Forest Land Remaining Forest Land and Land Converted to Forest Land. d LULUCF emissions include the CH4 and N2O 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 N2O emissions from Forest Soils and Settlement Soils. e Hie LULUCF Sector Net Lotal is the net sum of all CH4 and N2O emissions to the atmosphere plus net carbon stock changes. Notes: Lotals may not sum due to independent rounding. Parentheses indicate net sequestration. Other significant trends from 1990 to 2016 in emissions from LULUCF categories include: • Annual C sequestration by forest land (i.e., annual C stock accumulation in the five C pools and harvested wood products for Forest Land Remaining Forest Land and Land Converted to Forest Land) lias decreased by approximately 5.6 percent since 1990. This is primarily due to decreased C stock gains in Land Converted to Forest Land and the harvested wood products pools within Forest Land Remaining Forest Land. • Annual C sequestration from Settlements Remaining Settlements (which includes organic soils, urban trees, and landfilled yard trimmings and food scraps) has increased by 20.3 percent over the period from 1990 to 2016. This is primarily due to an increase in urbanized land area in the United States. • Annual emissions from Land Converted to Grassland increased by approximately 23.3 percent from 1990 to 2016 due to losses in aboveground biomass, belowground biomass, dead wood, and litter C stocks from Forest Land Converted to Grassland. • Annual emissions from Land Converted to Settlements increased by approximately 82.6 percent from 1990 to 2016 due to losses in aboveground biomass C stocks from Forest Land Converted to Settlements and mineral soils C stocks from Grassland Converted to Settlements. • Nitrous oxide emissions from fertilizer application to settlement soils in 2016 totaled to 2.5 MMT CO2 Eq. (8 kt of N2O). This represents an increase of 74.6 percent since 1990. Additionally, the application of synthetic fertilizers to forest soils in 2016 resulted in N20 emissions of 0.5 MMT CO2 Eq. (2 kt of N20). Nitrous oxide emissions from fertilizer application to forest soils have increased by 455 percent since 1990, but still account for a relatively small portion of overall emissions. Waste Waste management and treatment activities are sources of greenhouse gas emissions (see Figure 2-13). In 2016, landfills were the third-largest source of U.S. anthropogenic CH4 emissions, accounting for 16.4 percent of total Trends 2-21 ------- 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 U.S. CH4 emissions.3 Additionally, wastewater treatment accounts for 15.1 percent of Waste emissions, 2.3 percent of U.S. CH4 emissions, and 1.3 percent of N20 emissions. Emissions of CH4 and N20 from composting grew from 1990 to 2016, and resulted in emissions of 4.0 MMT CO2 Eq. in 2016. A summary of greenhouse gas emissions from the Waste chapter is presented in Table 2-9. Figure 2-13: 2016 Waste Chapter Greenhouse Gas Sources (MMT CO2 Eq.) Landfills Wastewater Treatment Composting 108 Waste as a Portion of all Emissions 2.0% 20 40 60 80 MMT COi Eq. 100 120 Overall, in 2016, waste activities generated emissions of 131.5 MMT CO2 Eq., or 2.0 percent of total U.S. greenhouse gas emissions. Table 2-9: Emissions from Waste (MMT CO2 Eq.) Gas/Source 1990 2005 2012 2013 2014 2015 2016 CH4 195.6 150.4 134.0 130.2 129.8 128.9 124.6 Landfills 179.6 132.7 117.0 113.3 112.7 111.7 107.7 Wastewater Treatment 15.7 15.8 15.1 14.9 15.0 15.1 14.8 Composting 0.4 1.9 1.9 2.0 2.1 2.1 2.1 N2O 3.7 6.1 6.4 6.5 6.7 6.7 6.8 Wastewater Treatment 3.4 4.4 4.6 4.7 4.8 4.8 5.0 Composting 0.3 1.7 1.7 1.8 1.9 1.9 1.9 Total 199.3 156.4 140.4 136.7 136.5 135.6 131.5 Note: Totals may not sum due to independent rounding. Some significant trends in U.S. emissions from waste source categories include the following: • From 1990 to 2016, net CHi emissions from landfills decreased by 71.9 MMT CO2 Eq. (40.0 percent), with small increases occurring in interim years. 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. • Combined CHi and N2O emissions from composting have generally increased since 1990, from 0.7 MMT CO2 Eq. to 4.0 MMT CO2 Eq. in 2016, which represents slightly less than a five-fold increase over the time series. The growth in composting since the 1990s is attributable to primarily two factors: (1) steady growth in population and residential housing, and (2) the enactment of legislation by state and local governments that discouraged the disposal of yard trimmings in landfills. • From 1990 to 2016, CH4 and N20 emissions from wastewater treatment decreased by 0.9 MMT CO2 Eq. (5.5 percent) and increased by 1.6 MMT CO2 Eq. (46.5 percent), respectively. Methane emissions from 3 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. 2-22 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2016 ------- 1 2 3 4 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. Nitrous oxide emissions from wastewater treatment processes gradually increased across the time series as a result of increasing U.S. population and protein consumption. 5 2.2 Emissions by Economic Sector 6 Throughout this report, emission estimates are grouped into five sectors (i.e., chapters) defined by the IPCC and 7 detailed above: Energy; Industrial Processes and Product Use; Agriculture; LULUCF; and Waste. While it is 8 important to use this characterization for consistency with UNFCCC reporting guidelines and to promote 9 comparability across countries, it is also useful to characterize emissions according to commonly used economic 10 sector categories: residential, commercial, industry, transportation, electric power, and agriculture, as well as U.S. 11 Territories. 12 Using this categorization, transportation activities, in aggregate, accounted for the largest portion (28.5 percent) of 13 total U.S. greenhouse gas emissions in 2016. Emissions from electric power, in aggregate, accounted for the second 14 largest portion (28.2 percent). Emissions from industry accounted for about 22 percent of total U.S. greenhouse gas 15 emissions in 2016. Emissions from industry have in general declined over the past decade due to a number of 16 factors, including structural changes in the U.S. economy (i.e., shifts from a manufacturing-based to a service-based 17 economy), fuel switching, and efficiency improvements. 18 The remaining 22 percent of U.S. greenhouse gas emissions were contributed by the residential, agriculture, and 19 commercial sectors, plus emissions from U.S. Territories. The residential sector accounted for 5 percent, and 20 primarily consisted of CO2 emissions from fossil fuel combustion. Activities related to agriculture accounted for 21 roughly 9 percent of U.S. emissions; unlike other economic sectors, agricultural sector emissions were dominated by 22 N20 emissions from agricultural soil management and CH4 emissions from enteric fermentation, rather than CO2 23 from fossil fuel combustion. The commercial sector accounted for roughly 6 percent of emissions, while U.S. 24 Territories accounted for less than 1 percent. Carbon dioxide was also emitted and sequestered (in the form of C) by 25 a variety of activities related to forest management practices, tree planting in urban areas, the management of 26 agricultural soils, landfilling of yard trimmings, and changes in C stocks in coastal wetlands. 27 Table 2-10 presents a detailed breakdown of emissions from each of these economic sectors by source category, as 28 they are defined in this report. Figure 2-14 shows the trend in emissions by sector from 1990 to 2016. Trends 2-23 ------- 1 Figure 2-14: U.S. Greenhouse Gas Emissions Allocated to Economic Sectors (MMT CO2 Eq.) 2,500- Electric Power Industry 2,000- Transportation 1,500- Industry 1,000- Agriculture Commercial (Red) 500- Residential (Blue) o^rMro5,ir>^orvcoo>0'-Jm^u->v£>r^coo%o--JC>^-mio o!flISSo\oIoIoIoIo!o 00000 00 000000000 3 4 5 Table 2-10: U.S. Greenhouse Gas Emissions Allocated to Economic Sectors (MMT CO2 Eq. and 6 Percent of Total in 2016) Sector/Source 1990 2005 2012 2013 2014 2015 2016a Percent3 Transportation 1,525.1 1,972.5 1,748.9 1,757.8 1,793.1 1,808.8 1,863.8 28.5% CO2 from Fossil Fuel Combustion 1,467.2 1,855.8 1,661.9 1,677.6 1,717.1 1,735.5 1,794.9 27.4% Substitution of Ozone Depleting Substances + 67.1 55.1 49.8 47.2 45.1 42.4 0.6% Mobile Combustion 46.1 39.5 23.6 21.6 19.6 18.2 17.1 0.3% Non-Energy Use of Fuels 11.8 10.2 8.3 8.8 9.1 10.0 9.5 0.1% Electric Power Industry 1,861.7 2,439.3 2,055.7 2,074.1 2,075.5 1,936.9 1,845.7 28.2% CO2 from Fossil Fuel Combustion 1,820.8 2,400.9 2,022.2 2,038.1 2,038.0 1,900.7 1,808.8 27.6% Stationary Combustion 6.9 14.0 14.2 15.6 16.0 15.4 16.0 0.2% Incineration of Waste 8.4 12.9 10.7 10.7 10.9 11.0 11.0 0.2% Other Process Uses of Carbonates 2.5 3.2 4.0 5.2 5.9 5.6 5.6 0.1% Electrical Transmission and Distribution 23.1 8.3 4.6 4.5 4.6 4.2 4.3 0.1% Industry 1,654.6 1,515.1 1,418.2 1,477.6 1,474.2 1,466.7 1,412.4 21.6% CO2 from Fossil Fuel Combustion 843.1 820.5 767.3 798.6 780.0 771.8 759.3 11.6% Natural Gas Systems 223.4 182.5 181.2 185.6 191.2 190.8 188.8 2.9% Non-Energy Use of Fuels 102.1 123.4 100.2 119.0 113.5 120.0 106.5 1.6% Petroleum Systems 51.7 51.7 61.0 68.5 73.9 77.4 64.8 1.0% Coal Mining 96.5 64.1 66.5 64.6 64.6 61.2 53.8 0.8% Iron and Steel Production 101.5 68.1 55.5 53.4 58.2 47.7 42.2 0.6% Cement Production 33.5 46.2 35.3 36.4 39.4 39.9 39.4 0.6% Petrochemical Production 21.4 26.9 26.6 26.5 26.6 28.2 27.6 0.4% Substitution of Ozone Depleting Substances + 7.4 18.8 20.4 22.3 24.7 26.9 0.4% Lime Production 11.7 14.6 13.8 14.0 14.2 13.3 13.3 0.2% Ammonia Production 13.0 9.2 9.4 10.0 9.6 10.6 11.2 0.2% Nitric Acid Production 12.1 11.3 10.5 10.7 10.9 11.6 10.2 0.2% Abandoned Oil and Gas Wells 6.5 6.9 7.0 7.0 7.1 7.2 7.1 0.1% 2-24 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2016 ------- Adipic Acid Production 15.2 7.1 5.5 3.9 5.4 4.3 7.0 0.1% Abandoned Underground Coal Mines 7.2 6.6 6.2 6.2 6.3 6.4 6.7 0.1% Other Process Uses of Carbonates 2.5 3.2 4.0 5.2 5.9 5.6 5.6 0.1% Semiconductor Manufacture 3.6 4.7 4.4 4.0 4.9 5.0 5.0 0.1% Carbon Dioxide Consumption 1.5 1.4 4.0 4.2 4.5 4.5 4.5 0.1% N2O from Product Uses 4.2 4.2 4.2 4.2 4.2 4.2 4.2 0.1% Urea Consumption for Non- Agricultural Purposes 3.8 3.7 4.4 4.1 1.5 4.2 4.0 0.1% Stationary Combustion 5.0 4.8 4.0 4.1 4.0 3.9 3.7 0.1% Mobile Combustion 4.4 4.5 3.3 3.3 3.1 3.0 3.0 +% HCFC-22 Production 46.1 20.0 5.5 4.1 5.0 4.3 2.8 +% Aluminum Production 28.3 7.6 6.4 6.2 5.4 4.8 2.7 +% Caprolactam, Glyoxal, and Glyoxylic Acid Production 1.7 2.1 2.0 2.0 2.0 2.0 2.0 +% Ferroalloy Production 2.2 1.4 1.9 1.8 1.9 2.0 1.8 +% Soda Ash Production 1.4 1.7 1.7 1.7 1.7 1.7 1.7 +% Titanium Dioxide Production 1.2 1.8 1.5 1.7 1.7 1.6 1.6 +% Glass Production 1.5 1.9 1.2 1.3 1.3 1.3 1.3 +% Magnesium Production and Processing 5.2 2.7 1.7 1.5 1.1 1.0 1.1 +% Phosphoric Acid Production 1.5 1.3 1.1 1.1 1.0 1.0 1.0 +% Zinc Production 0.6 1.0 1.5 1.4 1.0 0.9 0.9 +% Tead Production 0.5 0.6 0.5 0.5 0.5 0.5 0.5 +% Silicon Carbide Production and Consumption 0.4 0.2 0.2 0.2 0.2 0.2 0.2 +% Agriculture 521.5 568.5 571.8 594.0 591.5 615.1 611.7 9.3% N2O from Agricultural Soil Management 250.5 253.5 247.9 276.6 274.0 295.0 283.6 4.3% Enteric Fermentation 164.2 168.9 166.7 165.5 164.2 166.5 170.1 2.6% Manure Management 51.1 72.9 83.2 80.8 80.4 84.0 85.9 1.3% CO2 from Fossil Fuel Combustion 31.4 47.4 51.1 50.0 50.8 47.5 48.3 0.7% Rice Cultivation 16.0 16.7 11.3 11.5 12.7 12.3 13.7 0.2% Urea Fertilization 2.4 3.5 4.3 4.4 4.5 4.9 5.1 0.1% Timing 4.7 4.3 6.0 3.9 3.6 3.8 3.9 0.1% Mobile Combustion 0.8 1.0 0.9 0.8 0.8 0.7 0.7 +% Field Burning of Agricultural Residues 0.3 0.3 0.4 0.4 0.4 0.4 0.4 +% Stationary Combustion + + + 0.1 0.1 0.1 0.1 +% Commercial 428.2 402.6 388.0 411.3 420.8 433.2 411.7 6.3% CO2 from Fossil Fuel Combustion 227.4 227.0 201.3 225.7 233.6 245.6 227.9 3.5% Tandfills 179.6 132.7 117.0 113.3 112.7 111.7 107.7 1.6% Substitution of Ozone Depleting Substances + 17.6 45.1 47.5 49.3 50.3 50.9 0.8% Wastewater Treatment 15.7 15.8 15.1 14.9 15.0 15.1 14.8 0.2% Human Sewage 3.4 4.4 4.6 4.7 4.8 4.8 5.0 0.1% Composting 0.7 3.5 3.7 3.9 4.0 4.0 4.0 0.1% Stationary Combustion 1.5 1.4 1.2 1.4 1.4 1.6 1.5 +% Residential 344.9 370.4 318.4 372.7 393.9 370.0 354.1 5.4% CO2 from Fossil Fuel Combustion 338.3 357.8 282.5 329.7 345.3 316.8 296.2 4.5% Substitution of Ozone Depleting Substances 0.3 7.7 31.4 37.0 42.6 48.4 53.8 0.8% Stationary Combustion 6.3 4.9 4.5 5.9 6.1 4.7 4.1 0.1% U.S. Territories 33.3 58.1 48.5 48.1 46.6 46.6 46.6 0.7% CO2 from Fossil Fuel Combustion 27.6 49.7 43.5 42.5 41.4 41.4 41.4 0.6% Non-Energy Use of Fuels 5.7 8.1 4.8 5.4 5.1 5.1 5.1 0.1% Stationary Combustion 0.1 0.2 0.2 0.2 0.2 0.2 0.2 +% Total Emissions 6,369.2 7,326.4 6,549.4 6,735.6 6,795.6 6,677.3 6,546.2 100.0% Trends 2-25 ------- 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 LULUCF Sector Net Total" (819.6) (731.1) (753.5) (735.8) (740.4) (695.2) (716.8) (11.0%) Net Emissions (Sources and Sinks) 5,549.6 6,595.3 5,795.9 5,999.9 6,055.2 5,982.1 5,829.3 89.0% Notes: Total emissions presented without LULUCF. Total net emissions presented with LULUCF. + Does not exceed 0.05 MMT CO2 Eq. or 0.05 percent. 3 Percent of total (gross) emissions excluding emissions from LULUCF for 2016. b The LULUCF Sector Net Total is the net sum of all CH4 and N2O emissions to the atmosphere plus net carbon stock changes. Notes: Totals may not sum due to independent rounding. Parentheses indicate negative values or sequestration. Emissions with Electricity Distributed to Economic Sectors It can also be useful to view greenhouse gas emissions from economic sectors with emissions related to electric power distributed into end-use categories (i.e., emissions from electric power are allocated to the economic sectors in which the electricity is used). The generation, transmission, and distribution of electricity, which is the second largest economic sector in the United States, accounted for 28 percent of total U.S. greenhouse gas emissions in 2016. Electric power-related emissions decreased by 1 percent since 1990 and by 4.7 percent from 2015 to 2016, primarily due to decreased CO2 emissions from fossil fuel combustion due to increased natural gas consumption, decreased coal consumption. Overall, between 2015 and 2016, the amount of electricity generated (in kWh) increased by less than 0.1 percent. However, total emissions from the electric power sector decreased by 4.7 percent from 2015 to 2016 due to changes in the consumption of coal and natural gas for electric power, which were driven by changes in their relative prices. Coal consumption decreased by 8.1 percent, while natural gas consumption increased by 3.8 percent. The consumption of petroleum for electric power decreased by 12.9 percent in 2016 relative to 2015. Electricity sales to the residential and commercial end-use sectors increased by 0.2 percent and decreased by 0.1 percent, respectively, from 2015 to 2016. The sales trend in the residential sector can largely be attributed to an increase in the number of households in the United States. The sales trend in the commercial sector can largely be attributed to warmer, less energy-intensive winter conditions compared to 2015. Electricity sales to the industrial sector from 2015 to 2016 decreased by approximately 5.1 percent. Table 2-11 provides a detailed summary of emissions from electric power-related activities. Table 2-11: Electric Power-Related Greenhouse Gas Emissions (MMT CO2 Eq.) Gas/Fuel Type or Source 1990 2005 2012 2013 2014 2015 2016 CO2 1,831.2 2,416.5 2,036.6 2,053.7 2,054.5 1,917.0 1,825.1 Fossil Fuel Combustion 1,820.8 2,400.9 2,022.2 2,038.1 2,038.0 1,900.7 1,808.8 Coal 1,547.6 1,983.8 1,511.2 1,571.3 1,569.1 1,350.5 1,241.3 Natural Gas 175.3 318.8 492.2 444.0 443.2 526.1 545.9 Petroleum 97.5 97.9 18.3 22.4 25.3 23.7 21.2 Geothermal 0.4 0.4 0.4 0.4 0.4 0.4 0.4 Incineration of Waste 8.0 12.5 10.4 10.4 10.6 10.7 10.7 Other Process Uses of Carbonates 2.5 3.2 4.0 5.2 5.9 5.6 5.6 CH4 0.4 0.9 1.1 1.0 1.0 1.1 1.1 Stationary Sources3 0.4 0.9 1.1 1.0 1.0 1.1 1.1 Incineration of Waste + + + + + + + N2O 6.9 13.6 13.4 14.9 15.3 14.6 15.2 Stationary Sources3 6.5 13.2 13.1 14.6 15.0 14.3 14.9 Incineration of Waste 0.5 0.4 0.3 0.3 0.3 0.3 0.3 SF« 23.1 8.3 4.6 4.5 4.6 4.2 4.3 Electrical Transmission and Distribution 23.1 8.3 4.6 4.5 4.6 4.2 4.3 Total 1,861.7 2,439.3 2,055.7 2,074.1 2,075.5 1,936.9 1,845.7 2-26 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2016 ------- 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 + Does not exceed 0.05 MMT CO2 Eq. 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 2017a; Duffield 2006). These source categories include CO2 from Fossil Fuel Combustion, CH4 and N20 from Stationary Combustion. Incineration of Waste, Other Process Uses of Carbonates, and SF6 from Electrical Transmission and Distribution Systems. Note that only 50 percent of the Other Process Uses of Carbonates emissions were associated with electric power and distributed as described; the remainder of Other Process Uses of Carbonates emissions were attributed to the industrial processes economic end-use sector.4 When emissions from electricity use are distributed among these sectors, industrial activities account for the largest share of total U.S. greenhouse gas emissions (28.9 percent), followed closely by emissions from transportation (28.5 percent). Emissions from the residential and commercial sectors also increase substantially when emissions from electricity are included. In all sectors except agriculture, CO2 accounts for more than 81 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 2-15 shows the trend in these emissions by sector from 1990 to 2016. Figure 2-15: U.S. Greenhouse Gas Emissions with Electricity-Related Emissions Distributed to Economic Sectors (MMT CO2 Eq.) 2,500 Industry 2,000 Transportation iff .. 1,500 o u Commercial (Red) 1,000 Residential (Blue) Agriculture 500 Q) Q) 0"» ffi Ov O"* 01 m O rN in = 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 2016 Sector/Gas 1990 2005 2012 2013 2014 2015 2016 Percent3 Industry 2,321.9 2,225.4 1,977.8 2,040.5 2,033.7 1,985.1 1,888.8 28.9% Direct Emissions CO2 1,654.6 1,189.1 1,515.1 1,170.4 1,418.2 1,083.5 1,477.6 1,139.9 1,474.2 1,125.4 1,466.7 1,122.0 1,412.4 1,076.1 21.6% 16.4% 4 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 2-27 ------- ch4 351.9 277.2 275.2 279.6 286.3 281.6 271.9 4.2% n2o 37.3 29.3 26.5 25.2 27.0 26.4 27.7 0.4% HFCs, PFCs, SFs.andNFs 76.3 38.2 33.0 32.9 35.5 36.7 36.9 0.6% Electricity-Related 667.3 710.3 559.6 562.9 559.5 518.4 476.4 7.3% CO2 656.4 703.7 554.4 557.4 553.9 513.1 471.1 7.2% CH4 0.2 0.3 0.3 0.3 0.3 0.3 0.3 +% N2O 2.5 3.9 3.7 4.1 4.1 3.9 3.9 0.1% SFe 8.3 2.4 1.3 1.2 1.2 1.1 1.1 +% Transportation 1,528.2 ; 1,977.3 1,752.8 1,761.9 1,797.2 1,812.5 1,867.4 28.5% Direct Emissions 1,525.1 1,972.5 1,748.9 1,757.8 1,793.1 1,808.8 1,863.8 28.5% CO2 1,479.0 1,865.9 1,670.2 1,686.4 1,726.3 1,745.5 1,804.3 27.6% CH4 5.8 3.1 2.0 1.9 1.7 1.7 1.6 +% N2O 40.2 36.5 21.6 19.7 17.8 16.5 15.5 0.2% HFCsb + 67.1 55.1 49.8 47.2 45.1 42.4 0.6% Electricity-Related 3.1 4.8 3.9 4.1 4.1 3.8 3.6 0.1% CO2 3.1 4.8 3.9 4.0 4.1 3.8 3.6 0.1% CH4 + + + + + + + +% N2O + + + + + + + +% SFe + + + + + + + +% Commercial 978.2 1,218.7 1,099.9 1,128.2 1,139.8 1,108.9 1,063.1 16.2% Direct Emissions 428.2 402.6 388.0 411.3 420.8 433.2 411.7 6.3% CO2 227.4 227.0 201.3 225.7 233.6 245.6 227.9 3.5% CH4 196.7 151.5 135.0 131.3 130.9 130.1 125.8 1.9% N2O 4.1 6.4 6.6 6.8 7.0 7.1 7.2 0.1% HFCs + 17.6 45.1 47.5 49.3 50.3 50.9 0.8% Electricity-Related 550.1 816.1 711.9 716.9 719.0 675.8 651.3 10.0% CO2 541.1 808.5 705.3 709.9 711.7 668.8 644.1 9.8% CH4 0.1 0.3 0.4 0.4 0.4 0.4 0.4 +% N2O 2.0 4.5 4.6 5.2 5.3 5.1 5.4 0.1% SFe 6.8 2.8 1.6 1.5 1.6 1.5 1.5 +% Residential 951.2 1,240.4 1,055.7 1,120.6 1,142.2 1,067.3 1,028.4 15.7% Direct Emissions 344.9 370.4 318.4 372.7 393.9 370.0 354.1 5.4% CO2 338.3 357.8 282.5 329.7 345.3 316.8 296.2 4.5% CH4 5.2 4.1 3.7 5.0 5.1 3.9 3.4 0.1% N2O 1.0 0.9 0.7 1.0 1.0 0.8 0.7 +% HFCs 0.3 7.7 31.4 37.0 42.6 48.4 53.8 0.8% Electricity-Related 606.3 870.0 737.3 747.9 748.2 697.3 674.2 10.3% CO2 596.4 861.9 730.5 740.5 740.7 690.1 666.7 10.2% CH4 0.1 0.3 0.4 0.4 0.4 0.4 0.4 +% N2O 2.3 4.8 4.8 5.4 5.5 5.2 5.5 0.1% SFe 7.5 3.0 1.7 1.6 1.7 1.5 1.6 +% Agriculture 556.3 606.5 614.8 636.4 636.1 656.8 651.9 10.0% Direct Emissions 521.5 568.5 571.8 594.0 591.5 615.1 611.7 9.3% CO2 38.5 55.2 61.3 58.4 58.9 56.1 57.3 0.9% CH4 218.0 242.6 244.3 240.8 240.3 245.5 251.9 3.8% N2O 264.9 270.7 266.2 294.9 292.3 313.5 302.5 4.6% Electricity-Related 34.8 38.0 43.0 42.3 44.6 41.6 40.2 0.6% CO2 34.2 37.7 42.6 41.9 44.1 41.2 39.7 0.6% CH4 + + + + + + + +% N2O 0.1 0.2 0.3 0.3 0.3 0.3 0.3 +% SFe 0.4 0.1 0.1 0.1 0.1 0.1 0.1 +% U.S. Territories 33.3 58.1 48.5 48.1 46.6 46.6 46.6 0.7% Total Emissions 6,369.2 7,326.4 6,549.4 6,735.6 6,795.6 6,677.3 6,546.2 100.0% LULUCF Sector Net Totalc (819.6) (731.1) (753.5) (735.8) (740.4) (695.2) (716.8) (11.0%) Net Emissions (Sources and Sinks) 5,549.6 6,595.3 5,795.9 5,999.9 6,055.2 5,982.1 5,829.3 89.0% Notes: Total emissions presented without LULUCF. Net emissions presented with LULUCF. + Does not exceed 0.05 MMT CO2 Eq. or 0.05 percent. a Percent of total (gross) emissions excluding emissions from LULUCF for year 2016. 2-28 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2016 ------- 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 b Includes primarily HFC-134a. c The LULUCF Sector Net Total is the net sum of all CH4 and N2O emissions to the atmosphere plus 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. Industry The industry end-use sector includes CO2 emissions from fossil fuel combustion from all manufacturing facilities, in aggregate. 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 CH4 emissions from coal mining, byproduct CO2 emissions from cement manufacture, and HFC, PFC, SF6, and NF3 byproduct emissions from semiconductor manufacture, to name a few. Since 1990, industrial sector emissions have declined. 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., from 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 2016. The largest sources of transportation greenhouse gases in 2016 were passenger cars (41.8 percent), freight trucks (22.7 percent), light-duty trucks, which include sport utility vehicles, pickup trucks, and minivans (17.7 percent), commercial aircraft (6.4 percent), other aircraft (2.7 percent), ships and boats (2.4 percent), rail (2.2 percent), and pipelines (2.1 percent). These figures include direct CO2, CH4, and N20 emissions from fossil fuel combustion used in transportation 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. In terms of the overall trend, from 1990 to 2016, total transportation emissions increased due, in large part, to increased demand for travel. The number of vehicle miles traveled (VMT) by light-duty motor vehicles (passenger cars and light-duty trucks) increased 43 percent from 1990 to 2016,5 as a result of a confluence of factors including population growth, economic growth, urban sprawl, and periods of low fuel prices. 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 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 VMT grew only modestly for much of the period. Light-duty VMT grew by less than one percent or declined each year between 2005 and 20146 and has since grown at a faster rate (1.2 percent from 2014 to 2015, and 2.6 percent from 2015 to 2016). Average new vehicle fuel economy has increased almost every year since 2005 while the light-duty truck share decreased to about 33 percent in 2009 and has since varied from year to year between 36 percent and 43 percent. Light-duty truck share is about 43 percent of new vehicles in model year 2016 (EPA 2016a). 5 VMT estimates are based on data from FHWA Highway Statistics Table VM-1 (FHWA 1996 through 2017). Table VM-1 data for 2016 has not been published yet, therefore 2016 mileage data is estimated using the 1.7 percent increase in FHWA Traffic Volume Trends from 2015 to 2016. In 2011, FHWA changed its methods for estimating VMT by vehicle class, which led to a shift in VMT and emissions among on-road vehicle classes in the 2007 to 2016 time period. In absence of these method changes, light-duty VMT growth between 1990 and 2016 would likely have been even higher. 6 In 2007 and 2008 light-duty VMT decreased 3 percent and 2.3 percent, respectively. Note that the decline in light-duty VMT from 2006 to 2007 is due at least in part to a change in FHWA's methods for estimating VMT. In absence of these method changes, light-duty VMT growth between 2006 and 2007 would likely have been higher. See previous footnote. Trends 2-29 ------- 1 Table 2-13 provides a detailed summary of greenhouse gas emissions from transportation-related activities with 2 electricity-related emissions included in the totals. 3 Almost all of the energy consumed for transportation was supplied by petroleum-based products, with more than 4 half being related to gasoline consumption in automobiles and other highway vehicles. Other fuel uses, especially 5 diesel fuel for freight trucks and jet fuel for aircraft, accounted for the remainder. The primary driver of 6 transportation-related emissions was CO2 from fossil fuel combustion, which increased by 22 percent from 1990 to 7 2016.7 This rise in CO2 emissions, combined with an increase in HFCs from close to zero emissions in 1990 to 42.4 8 MMT CO2 Eq. in 2016, led to an increase in overall emissions from transportation activities of 22 percent.8 9 Table 2-13: Transportation-Related Greenhouse Gas Emissions (MMT CO2 Eq.) Gas/Vehicle 1990 2005 2012 2013 2014 2015 2016 Passenger Cars 639.6 693.1 745.9 740.9 756.8 761.1 781.3 CO2 612.3 642.6 711.3 710.9 729.6 735.8 758.4 CH4 3.2 1.3 0.9 0.8 0.7 0.6 0.6 N2O 24.1 17.6 13.1 11.8 10.5 9.7 8.9 HFCs 0.0 31.7 20.6 17.3 16.0 14.9 13.4 Light-Duty Trucks 326.8 539.7 316.2 313.2 334.3 324.9 331.4 CO2 312.3 490.6 281.3 281.5 304.7 297.5 306.3 CH4 1.7 0.8 0.3 0.3 0.3 0.2 0.2 N2O 12.8 15.0 5.3 4.7 4.4 3.8 3.4 HFCs 0.0 33.3 29.3 26.7 25.0 23.4 21.5 Medium- and Heavy-Duty Trucks 230.3 397.8 387.3 394.3 405.6 413.9 424.0 CO2 229.3 395.4 383.6 390.3 401.5 409.5 419.5 CH4 0.3 0.1 0.1 0.1 0.1 0.1 0.1 N2O 0.7 1.2 1.0 0.9 0.9 0.8 0.8 HFCs 0.0 1.1 2.6 2.9 3.2 3.4 3.7 Buses 8.5 12.2 18.0 18.2 19.5 20.0 20.5 CO2 8.4 11.6 17.2 17.5 18.8 19.3 19.8 CH4 + 0.2 0.3 0.2 0.2 0.2 0.2 N2O + + + + + + + HFCs 0.0 0.3 0.4 0.4 0.4 0.4 0.4 Motorcycles 1.7 1.6 4.0 3.8 3.8 3.7 3.8 CO2" 1.7 1.6 3.9 3.7 3.7 3.7 3.8 CH4 + + + + + + + N2O + + + + + + + Commercial Aircraft3 110.9 134.0 114.3 115.4 116.3 120.1 120.1 CO2 109.9 132.7 113.3 114.3 115.2 119.0 119.0 CH4 0.0 0.0 0.0 0.0 0.0 0.0 0.0 N2O 1.0 1.2 1.0 1.1 1.1 1.1 1.1 Other Aircraftb 78.3 59.7 32.1 34.7 35.0 40.4 51.1 CO2 77.5 59.1 31.8 34.4 34.7 40.0 50.6 CH4 0.1 0.1 + + + + + N2O 0.7 0.5 0.3 0.3 0.3 0.4 0.5 Ships and Boats0 45.3 45.8 41.9 41.5 31.0 35.7 44.9 CO2 44.3 44.3 39.3 38.6 28.0 32.3 41.1 CH4 0.5 0.5 0.4 0.4 0.3 0.3 0.3 N2O 0.6 0.6 0.5 0.5 0.3 0.4 0.5 HFCs 0.0 0.5 1.7 2.0 2.3 2.6 2.9 Rail 38.9 51.0 44.1 45.0 46.4 44.3 41.1 CO2 38.5 50.3 43.4 44.2 45.6 43.5 40.3 7 See previous footnote. 8 See previous footnote. 2-30 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2016 ------- 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 ch4 O.I 0.1 0.1 0.1 0.1 0.1 0.1 N20 0.3 0.4 0.3 0.3 0.3 0.3 0.3 HFCs 0.0 0.2 0.3 0.3 0.3 0.4 0.4 Other Emissions from Electric Powerd O.I + + + + + + Pipelines® 36.0 32.4 40.5 46.2 39.4 38.5 39.6 CO2 36.0 32.4 40.5 46.2 39.4 38.5 39.6 Lubricants ll.N 10.2 8.3 8.8 9.1 10.0 9.5 CO2 II.8 10.2 8.3 OO OO 9.1 10.0 9.5 Total Transportation 1,52N.2 1,977.3 1,752.8 1,761.9 1,797.2 1,812.5 1,867.4 International Bunker Fuel/ 104.5 114.2 106.8 100.7 104.4 111.9 115.5 Ethanol CO2s 4.1 22.4 71.5 73.4 74.9 75.9 78.2 Biodiesel CO2s 0.0 0.9 8.5 13.5 13.3 14.1 19.6 + Does not exceed 0.05 MMT CO2 Eq. a Consists of emissions from jet fuel consumed by domestic operations of commercial aircraft (no bunkers). b Consists of emissions from jet fuel and aviation gasoline consumption by general aviation and military aircraft. c Fluctuations in emission estimates are associated with fluctuations in reported fuel consumption, and may reflect issues with data sources. d Other emissions from electric power are a result of waste incineration (as the majority of municipal solid waste is combusted in "trash-to-steam" electric power plants), electrical transmission and distribution, and a portion of Other Process Uses of Carbonates (from pollution control equipment installed in electric power plants). e CO2 estimates reflect natural gas used to power pipelines, but not electricity. While the operation of pipelines produces CH4 and N2O, 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. B Ethanol and biodiesel CO2 estimates are presented for informational purposes only. See Section 3.11 and the estimates in Land Use, Land-Use Change, and Forestry (see Chapter 6), in line with IPCC methodological guidance and UNFCCC reporting obligations, 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. Commercial The commercial sector is heavily reliant on electricity for meeting energy needs, with electricity use for lighting, heating, air conditioning, and operating appliances. The remaining emissions were largely due to the direct consumption of natural gas and petroleum products, primarily for heating and cooking needs. Energy-related emissions from the commercial sector have generally been increasing since 1990, and 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, a reduction in heating degree days, and increases in energy efficiency. Landfills and wastewater treatment are included in the commercial sector, with landfill emissions decreasing since 1990 and wastewater treatment emissions decreasing slightly. Residential The residential sector is heavily reliant on electricity for meeting energy needs, with electricity consumption for lighting, heating, air conditioning, and operating appliances. The remaining emissions were largely due to the direct consumption of natural gas and petroleum products, primarily for heating and cooking needs. Emissions from the residential sector have generally been increasing since 1990, and are often correlated with short-term fluctuations in energy consumption caused by weather conditions, rather than prevailing economic conditions. In the long term, 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 (EIA 2017b). Trends 2-31 ------- 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 Agriculture The agriculture end-use sector includes a variety of processes, including enteric fermentation in domestic livestock, livestock manure management, and agricultural soil management. In 2016, agricultural soil management was the largest source of N20 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 like tractors. Box 2-1: Methodology for Aggregating Emissions by Economic Sector In presenting the Economic Sectors in the annual Inventory of U.S. Greenhouse Gas Emissions and Sinks, the Inventory expands upon the standard IPCC sectors common for UNFCCC reporting. Discussing greenhouse gas emissions relevant to U.S.-specific economic sectors improves communication of the report's findings. The Electric Power economic sector includes CO2 emissions from the combustion of fossil fuels that are included in the EIA electric utility fuel-consuming sector. Stationary combustion emissions of CH4 and N20 are also based on the EIA electric power sector. Additional sources include CO2, CHi and N20 from waste incineration, as the majority of municipal solid waste is combusted in "trash-to-steam" electric power plants. The Electric Power economic sector also includes SF6 from Electrical Transmission and Distribution, 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 N20 from mobile combustion are also apportioned to the Transportation economic sector based on the EIA transportation fuel-consuming sector. Substitution of Ozone Depleting Substances emissions 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 N20 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. Substitution of Ozone Depleting Substances emissions are apportioned based on their specific end-uses within the source category, with most emissions falling within 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 included in supplementary sources of agriculture fuel use, because EIA does not include an agriculture fuel-consuming sector. Agriculture equipment is included in the EIA industrial fuel-consuming sector. Agriculture fuel use estimates are obtained from U.S. Department of Agriculture survey data, in combination with separate EIA fuel sales reports (USDA 2016; EIA 2017c). These supplementary data are subtracted from the industrial fuel use reported by EIA to obtain agriculture fuel use. CO2 emissions from fossil fuel combustion, and CH4 and N20 emissions from stationary and mobile combustion are then apportioned to the Agriculture economic sector based on agricultural fuel use. The other emission sources included in the Agriculture economic sector are intuitive for the agriculture sectors, such as N2O emissions from Agricultural Soils, CH4 from Enteric Fermentation, CH4 and N20 from Manure 2-32 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2016 ------- 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 Management, CH4 from Rice Cultivation. CO2 emissions from Liming and Urea Application, and CH4 and N20 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 N20 are also based on the EIA residential fuel-consuming sector. Substitution of Ozone Depleting Substances are apportioned to the Residential economic sector based on emissions from residential air-conditioning systems. Nitrous oxide 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 N20 from Mobile Combustion are also apportioned to the Commercial economic sector based on the EIA commercial fuel-consuming sector. Substitution of Ozone Depleting Substances emissions are apportioned to the Commercial economic sector based on emissions from commercial refrigeration/air-conditioning systems. Public works sources including direct CH4 from Landfills, CH4 and N20 from Wastewater Treatment, and Composting are also included in the Commercial economic sector. Box 2-2: Recent Trends in Various U.S. Greenhouse Gas Emissions-Related Data Total emissions can be compared to other economic and social indices to highlight changes over time. These comparisons include: (1) emissions per unit of aggregate energy use, because energy-related activities are the largest sources of emissions; (2) emissions per unit of fossil fuel consumption, because almost all energy-related emissions involve the combustion of fossil fuels; (3) emissions per unit of electricity use, because the electric power industry—utilities and non-utilities combined—was the second largest source of U.S. greenhouse gas emissions in 2016; (4) emissions per unit of total gross domestic product as a measure of national economic activity; or (5) emissions per capita. 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 grown at an average annual rate of 0.1 percent since 1990. This rate is slightly slower than that for total energy use and fossil fuel consumption, and much slower than that for electricity use, overall gross domestic product (GDP) and national population (see Table 2-14 and Figure 2-16). These trends vary relative to 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 1.0 percent since 2005. Total energy use and fossil fuel consumption have also decreased at slower rates than emissions since 2005, while electricity use, GDP, and national population continued to increase. Table 2-14: Recent Trends in Various U.S. Data (Index 1990 = 100) Variable 1990 2005 2012 2013 2014 2015 2016 Avg. Annual Avg. Annual Change Change since 1990a since 2005a Greenhouse Gas Emissions'5 100 115 103 106 107 105 103 0.1% -1.0% Energy Usec 100 118 112 115 117 115 116 0.6% -0.2% Fossil Fuel Consumption0 100 119 107 110 111 110 109 0.4% -0.7% Electricity Usec 100 134 135 136 138 137 136 1.2% 0.1% GDPd 100 159 171 174 179 184 187 2.4% 1.5% Population6 100 118 125 126 127 128 129 1.0% 0.8% a Average annual growth rate b GWP-weighted values c Energy-content-weighted values (EIA 2017a) d Gross Domestic Product in chained 2009 dollars (BEA 2017) e U.S. Census Bureau (2017) Trends 2-33 ------- 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 Figure 2-16: U.S. Greenhouse Gas Emissions Per Capita and Per Dollar of Gross Domestic Product 200 190- 180- 170- 160- 150- © 2 140- II o 130- U 120- 110- 100-- 90- 80- 70- 60- 50 5S Real GDP Population Emissions per capita Emissions perGDP Source: BEA (2017), U.S. Census Bureau (2017), and emission estimates in this report. 2.3 Indirect Greenhouse Gas Emissions (CO, NOx, NMVOCs, and S02) The reporting requirements of the UNFCCC9 request that information be provided on indirect greenhouse gases, which include CO, NOx, NMVOCs, and SO2. These gases do not have a direct global warming effect, but indirectly affect terrestrial radiation absorption by influencing the formation and destruction of tropospheric and stratospheric ozone, or, in the case of SO2, by affecting the absorptive characteristics of the atmosphere. Additionally, some of these gases may react with other chemical compounds in the atmosphere to form compounds that are greenhouse gases. Carbon monoxide is produced when carbon-containing fuels are combusted incompletely. Nitrogen oxides (i.e., NO and NO2) are created by lightning, fires, fossil fuel combustion, and in the stratosphere from N20. Non- methane volatile organic compounds—which include hundreds of organic compounds that participate in atmospheric chemical reactions (i.e., propane, butane, xylene, toluene, ethane, and many others)—are emitted primarily from transportation, industrial processes, and non-industrial consumption of organic solvents. In the United States, SO2 is primarily emitted from coal combustion for electric power generation and the metals industry. Sulfur-containing compounds emitted into the atmosphere tend to exert a negative radiative forcing (i.e., cooling) and therefore are discussed separately. One important indirect climate change effect of NMVOCs and NOx is their role as precursors for tropospheric ozone formation. They can also alter the atmospheric lifetimes of other greenhouse gases. Another example of indirect greenhouse gas formation into greenhouse gases is the interaction of CO with the hydroxyl radical—the major 9 See . 2-34 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2016 ------- 1 atmospheric sink for CH4 emissions—to form CO2. Therefore, increased atmospheric concentrations of CO limit the 2 number of hydroxyl molecules (OH) available to destroy CH4. 3 Since 1970, the United States has published estimates of emissions of CO, NOx, NMVOCs, and SO2 (EPA 2016b),10 4 which are regulated under the Clean Air Act. Table 2-15 shows that fuel combustion accounts for the majority of 5 emissions of these indirect greenhouse gases. Industrial processes—such as the manufacture of chemical and allied 6 products, metals processing, and industrial uses of solvents—are also significant sources of CO, NOx, and 7 NMVOCs. 8 Table 2-15: Emissions of NOx, CO, NMVOCs, and SO2 (kt) Gas/Activity 1990 2005 2012 2013 2014 2015 2016 NOx 21,791 17,443 12,038 11,388 10,807 10,252 9,278 Mobile Fossil Fuel Combustion 10,862 10,295 6,871 6,448 6,024 5,417 4,814 Stationary Fossil Fuel Combustion 10,02:' 5,858 3,655 3,504 3,291 3,061 2,692 Oil and Gas Activities 139 321 663 704 745 745 745 Forest Fires 81 239 276 185 185 474 474 Industrial Processes and Product Use 592 572 443 434 424 424 424 Waste Combustion 82 128 82 91 100 100 100 Grassland Fires 5 21 39 13 27 21 19 Agricultural Burning 6 6 7 7 8 8 7 Waste - 2 2 2 2 2 2 CO 132,926 75,569 54,109 48,589 46,875 54,977 52,990 Mobile Fossil Fuel Combustion 119,360 58,615 36,153 34,000 31,848 29,881 27,934 Forest Fires 2,880 8,484 9,804 6,624 6,595 16,752 16,752 Stationary Fossil Fuel Combustion 5,000 4,648 4,027 3,884 3,741 3,741 3,741 Waste Combustion 978 1,403 1,318 1,632 1,947 1,947 1,947 Industrial Processes and Product Use 4,129 1,557 1,246 1,262 1,273 1,273 1,273 Oil and Gas Activities 302 318 666 723 780 780 780 Grassland Fires 84 358 657 217 442 356 324 Agricultural Burning 191 178 232 239 240 239 230 Waste 1 7 6 8 9 9 9 NMVOCs 20,930 13,154 11,464 11,202 10,935 10,647 10,362 Industrial Processes and Product Use 7,638 5,849 3,861 3,793 3,723 3,723 3,723 Mobile Fossil Fuel Combustion 10,932 5,724 4,243 3,924 3,605 3,318 3,032 Oil and Gas Activities 554 510 2,651 2,786 2,921 2,921 2,921 Stationary Fossil Fuel Combustion 912 716 569 539 507 507 507 Waste Combustion 222 241 94 108 121 121 121 Waste 673 114 45 51 57 57 57 Agricultural Burning NA NA NA NA NA NA NA SO2 20,935 13,196 5,876 5,874 4,357 3,448 2,457 Stationary Fossil Fuel Combustion 18,40" 11,541 5,006 5,005 3,640 2,756 1,790 Industrial Processes and Product Use 1,30" 831 604 604 496 496 496 Mobile Fossil Fuel Combustion 390 180 108 108 93 93 93 Oil and Gas Activities 79' 619 142 142 95 70 44 Waste Combustion 38 25 15 15 32 32 32 Waste + 1 + + 1 1 1 Agricultural Burning NA NA NA NA NA NA NA + Does not exceed 0.5 kt. NA (Not Available) Note: Totals may not sum due to independent rounding. Source: (EPA 2016b) except for estimates from Field Burning of Agricultural Residues. 10 NOx and CO emission estimates from Field Burning of Agricultural Residues were estimated separately, and therefore not taken from EPA (2016b). Trends 2-35 ------- 1 Box 2-3: Sources and Effects of Sulfur Dioxide 2 Sulfur dioxide (SO2) emitted into the atmosphere through natural and anthropogenic processes affects the earth's 3 radiative budget through its photochemical transformation into sulfate aerosols that can: 8 The indirect effect of sulfur-derived aerosols on radiative forcing can be considered in two parts. The first indirect 9 effect is the aerosols' tendency to decrease water droplet size and increase water droplet concentration in the 10 atmosphere. The second indirect effect is the tendency of the reduction in cloud droplet size to affect precipitation 11 by increasing cloud lifetime and thickness. Although still highly uncertain, the radiative forcing estimates from both 12 the first and the second indirect effect are believed to be negative, as is the combined radiative forcing of the two 13 (IPCC 2013). 14 Sulfur dioxide is also a major contributor to the formation of regional haze, which can cause significant increases in 15 acute and chronic respiratory diseases. Once SO2 is emitted, it is chemically transformed in the atmosphere and 16 returns to the earth as the primary source of acid rain. Because of these harmful effects, the United States has 17 regulated SO2 emissions in the Clean Air Act. 18 Electric power is the largest anthropogenic source of SO2 emissions in the United States, accounting for 43.8 percent 19 in 2016. Coal combustion contributes nearly all of those emissions (approximately 92 percent). Sulfur dioxide 20 emissions have decreased in recent years, primarily as a result of electric power generators switching from liigh- 21 sulfur to low-sulfur coal and installing flue gas desulfurization equipment. 4 5 6 7 (1) scatter radiation from the sun back to space, thereby reducing the radiation reaching the earth's surface; (2) affect cloud formation; and (3) affect atmospheric chemical composition (e.g., by providing surfaces for heterogeneous chemical reactions). 22 23 2-36 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2016 ------- 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 3. Energy Energy-related activities were the primary sources of U.S. anthropogenic greenhouse gas emissions, accounting for 83.7 percent of total greenhouse gas emissions on a carbon dioxide (CO2) equivalent basis in 2016.1 This included 97, 43, and 10 percent of the nation's CO2, methane (CH4), and nitrous oxide (N20) emissions, respectively. Energy- related CO2 emissions alone constituted 78.8 percent of national emissions from all sources on a CO2 equivalent basis, while the non-C02 emissions from energy-related activities represented a much smaller portion of total national emissions (4.8 percent collectively). Emissions from fossil fuel combustion comprise the vast majority of energy-related emissions, with CO2 being the primary gas emitted (see Figure 3-1). Globally, approximately 32,294 million metric tons (MMT) of CO2 were added to the atmosphere through the combustion of fossil fuels in 2015, of which the United States accounted for approximately 15 percent.2 Due to their relative importance, fossil fuel combustion-related CO2 emissions are considered separately, and in more detail than other energy-related emissions (see Figure 3-2). Fossil fuel combustion also emits CH4 and N20. Stationary combustion of fossil fuels was the second largest source of N20 emissions in the United States and mobile fossil fuel combustion was the fourth largest source. Figure 3-1: 2016 Energy Chapter Greenhouse Gas Sources (MMT CO2 Eq.) CO: Emissions from Fossil Fuel Combustion Natural Gas Systems Non-Energy Use of Fuels Petroleum Systems Coal Mining Non-CO: Emissions from Stationary Combustion Non-CO? Emissions from Mobile Combustion Incineration of Waste Abandoned Oil and Gas Wells Abandoned Underground Coal Mines _ 0 50 100 150 200 250 300 MMT CO* Eq. 4,977 Energy as a Portion of all Emissions 83.7% 1 Estimates are presented in units of million metric tons of carbon dioxide equivalent (MMT CO2 Eq.), which weight each gas by its global wanning potential, or GWP, value. See section on global wanning potentials in the Executive Summary. 2 Global CO2 emissions from fossil fuel combustion were taken from International Energy Agency CO: Emissions from Fossil Fuels Combustion - Highlights IEA (2017). Energy 3-1 ------- 1 Figure 3-2: 2016 U.S. Fossil Carbon Flows (MMT CO2 Eq.) International. Bunkers Industrial Processes Fossil Fuel Energy Exports Coal Emissions 1,316 NEU Exports Combustion Emissions 1,307 Natural Gas Emissions 1,482 Coal 1,377 Combustion Emissions 1,477 NEU Emissions 107 Atmospheric Emissions 5,292 Domestic Fossil Fuel Production 4,495 Apparent Consumption 5,390 Petroleum Emissions 2,299 Natural Gas 1,455 Combustion Emissions 2,193 Petroleum Natural Gas Liquids, Liquefied Refinery Gas, & Other Liquids Non-Energy Use Carbon Sequestered Fossil Fuel Energy Imports Petroleum 1,383 , Balancing Item (110) NEU U.S. Territories Note: Totals may not sum due to independent rounding. The "Balancing Item" above accounts for the statistical imbalances and unknowns in the reported data sets combined here. NEU = Non-Energy Use Fossil Fuel Consumption U.S. Territories Stock Changes (68) Natural Gas 163 Coal 23' NEU Imports Other 226 2 3 Energy-related activities other than fuel combustion such as the production, transmission storage, and distribution 4 of fossil fuels, also emit greenhouse gases. These emissions consist primarily of fugitive CH4 from natural gas 5 systems, petroleum systems, and coal mining. Table 3-1 summarizes emissions from the Energy sector in units of 6 MMT CO2 Eq., while unweighted gas emissions in kilotons (kt) are provided in Table 3-2. Overall, emissions due to 7 energy-related activities were 5,476.4 MMT CO2 Eq. in 2016,3 an increase of 2.6 percent since 1990 and a decrease 8 of 2.1 percent since 2015. 9 Table 3-1: CO2, ChU, and N2O Emissions from Energy (MMT CO2 Eq.) Gas/Source 1990 2005 2012 2013 2014 2015 2016 CO2 4,922.4 5,952.7 5,203.5 5,361.6 5,404.4 5,269.4 5,160.7 Fossil Fuel Combustion 4,755.8 5,759.1 5,029.8 5,162.3 5,206.1 5,059.3 4,916.1 Electric Power 1,820.8 2,400.9 2,022.2 2,038.1 2,038.0 1,900.7 1,808.8 Transportation 1,467.2 1,855.8 1,661.9 1,677.6 1,717.1 1,735.5 1,794.9 Industrial 874.5 867.8 818.4 848.7 830.8 819.3 807.6 Residential 338.3 357.8 282.5 329.7 345.3 316.8 296.2 Commercial 227.4 227.0 201.3 225.7 233.6 245.6 227.9 U.S. Territories 27.6 49.7 43.5 42.5 41.4 41.4 41.4 Non-Energy Use of Fuels 119.6 141.7 113.3 133.2 127.8 135.1 121.0 Natural Gas Systems 29.7 22.5 24.4 26.0 27.0 26.3 26.7 Petroleum Systems 9.4 17.0 25.6 29.7 32.9 38.0 25.5 Incineration of Waste 8.0 12.5 10.4 10.4 10.6 10.7 10.7 Biomass-Wood" 215.2 206.9 194.9 211.6 218.9 201.5 190.2 International Bunker Fuelsb 103.5 113.1 105.8 99.8 103.4 110.9 114.4 Biofuels-Ethanol" 4.2 22.9 72.8 74.7 76.1 78.9 81.2 Biofuels-Biodiesel" 0.0 0.9 8.5 13.5 13.3 14.1 19.6 CH4 364.7 286.7 283.1 288.7 295.4 289.5 279.2 Natural Gas Systems 193.7 160.0 156.8 159.6 164.2 164.4 162.1 3 Following the revised reporting requirements under the UNFCCC, this Inventory report presents CO2 equivalent values based on the IPCC Fourth Assessment Report (AR4) GWP values. See the Introduction chapter for more information. 3-2 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2016 ------- Coal Mining 96.5 64.1 66.5 64.6 64.6 61.2 53.8 Petroleum Systems 42.3 34.7 35.4 38.8 41.0 39.4 39.3 Stationary Combustion 8.6 7.9 7.3 8.7 8.8 7.8 7.2 Abandoned Oil and Gas Wells 6.5 6.9 7.0 7.0 7.1 7.2 7.1 Abandoned Underground Coal Mines 7.2 6.6 6.2 6.2 6.3 6.4 6.7 Mobile Combustion 9.8 6.6 4.0 3.7 3.4 3.1 3.0 Incineration of Waste + + + + + + + International Bunker Fuelsb 0.2 0.1 0.1 0.1 0.1 0.1 0.1 N2O 53.1 56.4 40.9 40.9 39.4 37.1 36.5 Stationary Combustion 11.1 17.5 16.8 18.6 18.9 18.0 18.4 Mobile Combustion 41.5 38.4 23.8 22.0 20.2 18.8 17.8 Incineration of Waste 0.5 0.4 0.3 0.3 0.3 0.3 0.3 International Bunker Fuelsb 0.9 1.0 0.9 0.9 0.9 0.9 1.0 Total 5,340.2 6,295.7 5,527.6 5,691.1 5,739.1 5,596.0 5,476.4 + Does not exceed 0.05 MMT CO2 Eq. a Emissions from Wood Biomass, Ethanol, and Biodiesel 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 may not sum due to independent rounding. 1 Table 3-2: CO2, ChU, and N2O Emissions from Energy (kt) Gas/Source 1990 2005 2012 2013 2014 2015 2016 CO2 4,922,449 5,952,727 5,203,523 5,361,554 5,404,420 5,269,370 5,160,744 Fossil Fuel Combustion 4,755,819 5,759,056 5,029,830 5,162,315 5,206,135 5,059,288 4,976,737 Non-Energy Use of Fuels 119,588 141,669 113,275 133,176 127,778 135,106 121,049 Natural Gas Systems 29,708 22,529 24,398 26,004 27,004 26,329 26,739 Petroleum Systems 9,384 17,004 25,629 29,695 32,895 37,971 25,543 Incineration of Waste 7,950 12,469 10,392 10,363 10,608 10,676 10,676 Biomass-Wood" 215,186 206,901 194,903 211,581 218,922 201,457 190,171 International Bunker Fuelsb 103,463 113,139 105,805 99,763 103,400 110,887 114,394 Biofuels-Ethanol" 4,227 22,943 72,827 74,743 76,075 78,934 81,250 Biofuels-Biodiesel" 0 856 8,470 13,462 13,349 14,077 19,648 cm 14,587 11,467 11,326 11,548 11,814 11,581 11,166 Natural Gas Systems 7,748 6,399 6,273 6,385 6,568 6,578 6,483 Coal Mining 3,860 2,565 2,658 2,584 2,583 2,449 2,153 Petroleum Systems 1,693 1,386 1,415 1,553 1,639 1,576 1,571 Stationary Combustion 346 314 292 346 352 313 288 Abandoned Oil and Gas Wells 260 275 279 280 282 286 284 Abandoned Underground Coal Mines 288 264 249 249 253 256 268 Mobile Combustion 393 263 160 149 136 123 119 Incineration of Waste + + + + + + + International Bunker Fuelsb 7 5 4 3 3 3 4 N2O 178 189 137 137 132 125 122 Stationary Combustion 37 59 56 62 63 60 62 Mobile Combustion 139 129 80 74 68 63 60 Incineration of Waste 2 1 1 1 1 1 1 International Bunker Fuelsb 3 3 3 3 3 3 3 + Does not exceed 0.5 kt. a Emissions from Wood Biomass, Ethanol, and Biodiesel 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 2006 IPCC Guidelines and UNFCCC reporting obligations. Note: Totals may not sum due to independent rounding. Energy 3-3 ------- 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 Box 3-1: Methodological Approach for Estimating and Reporting U.S. Emissions and Removals In following the United Nations Framework Convention on Climate Change (UNFCCC) requirement under Article 4.1 to develop and submit national greenhouse gas emission inventories, 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 manner in line with the UNFCCC reporting guidelines for the reporting of inventories under this international agreement. The use of consistent methods to calculate emissions and removals by all nations providing their inventories to the UNFCCC ensures that these reports are comparable. The presentation of emissions and removals provided in this Inventory do not preclude alternative examinations, but rather, this Inventory presents emissions and removals in a common format consistent with how countries are to report Inventories under the UNFCCC. The report itself, and this chapter, follows this standardized format, and provides an explanation of the application of methods used to calculate emissions and removals. Box 3-2: Energy Data from EPA's Greenhouse Gas Reporting Program On October 30, 2009, the U.S. Enviromnental 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). The rule 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. Data reporting by affected facilities includes the reporting of emissions from fuel combustion at that affected facility. In general, the threshold for reporting is 25,000 metric tons or more of CO2 Eq. per year. EPA's GHGRP dataset and the data presented in this Inventory 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. 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 categories to improve the national estimates presented in this Inventory consistent with IPCC guidelines (see, also. Box 3-4).4 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. Most methodologies used in EPA's GHGRP are consistent with IPCC, though for EPA's GHGRP, facilities collect detailed information specific to their operations according to detailed measurement standards, which may differ with the more aggregated data collected for the Inventory to estimate total national U.S. emissions. It should be noted that the definitions and provisions for reporting fuel types in EPA's GHGRP may differ from those used in the Inventory in meeting the UNFCCC reporting guidelines. In line with the UNFCCC reporting guidelines, the Inventory report is a comprehensive accounting of all emissions from fuel types identified in the IPCC guidelines and provides a separate reporting of emissions from biomass. Further information on the reporting categorizations in EPA's GHGRP and specific data caveats associated with monitoring methods in EPA's GHGRP lias been provided on the GHGRP website.5 EPA presents the data collected by its GHGRP through a data publication tool that allows data to be viewed in several formats including maps, tables, charts and graphs for individual facilities or groups of facilities.6 4 See . 5 See . 6 See . 3-4 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2016 ------- 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 In addition to using GHGRP data to estimate emissions, 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 Format (CRF) tables. The industrial end-use sector activity data collected for the Inventory (EIA 2017) 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, the GHGRP data are used to provide a more detailed breakout of total emissions in the industrial end-use sector within that source category. 3.1 Fossil Fuel Combustion (CRF Source Category 1A) Emissions from the combustion of fossil fuels for energy include the gases CO2, CH4, and N20. 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. Following that is a discussion of emissions of all three gases from fossil fuel combustion presented by sectoral breakdowns. Methodologies for estimating CO2 from fossil fuel combustion also differ from the estimation of CH4 and N20 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, CH4, and N20 emissions from fossil fuel combustion are presented in Table 3-3 and Table 3-4. Table 3-3: CO2, ChU, and N2O Emissions from Fossil Fuel Combustion (MMT CO2 Eq.) Gas loon 2005 2012 2013 2014 2015 2016 CO2 4,755.8 5,759.1 5,029.8 5,162.3 5,206.1 5,059.3 4,976.7 CH4 18.5 14.4 11.3 12.4 12.2 10.9 10.2 N2O 52.6 56.0 40.6 40.6 39.0 36.8 36.2 Total 4,826.') 5,829.5 5,081.7 5,215.3 5,257.4 5,107.0 5,023.1 Note: Totals may not sum due to independent rounding able 3-4: CO2, ChU, and N2O Emissions from Fossil Fuel Combustion (kt) Gas 1990 2005 2012 2013 2014 2015 2016 CO2 4,755,819 5,759,056 5.029,830 5,162,315 5,206,135 5,059,288 4,976,737 CH4 739 578 452 495 488 436 407 N2O 177 188 136 136 131 124 121 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 2016, CO2 emissions from fossil fuel combustion decreased by 1.6 percent relative to the previous year. The decrease in CO2 emissions from fossil fuel combustion was a result of multiple factors, including: (1) substitution from coal to natural gas and other sources in the electric power sector; and (2) warmer winter conditions in 2016 resulting in a decreased demand for heating fuel in the residential and commercial sectors. In 2016, CO2 emissions from fossil fuel combustion were 4,976.7 MMT CO2 Eq., or 4.6 percent above 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 3-5 ------- 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 Table 3-5: CO2 Emissions from Fossil Fuel Combustion by Fuel Type and Sector (MMT CO2 Eq.) Fuel/Sector 1990 2005 2012 2013 2014 2015 2016 Coal 1,718.4 2,112.3 1,592.8 1,653.8 1,652.6 1,423.3 1,306.6 Residential 3.0 0.8 NO NO NO NO 0.0 Commercial 12.0 9.3 4.1 3.9 3.8 2.9 2.3 Industrial 155.3 115.3 74.1 75.7 75.6 65.9 59.0 Transportation NE NE NE NE NE NE NE Electric Power 1,547.6 1,983.8 1,511.2 1,571.3 1,569.1 1,350.5 1,241.3 U.S. Territories 0.6 3.0 3.4 2.8 4.0 4.0 4.0 Natural Gas 1,000.3 1,166.7 1,352.6 1,391.2 1,422.0 1,464.2 1,477.0 Residential 238.0 262.2 224.8 266.2 277.9 253.2 238.3 Commercial 142.1 162.9 156.9 179.1 189.3 175.7 170.3 Industrial 408.9 388.5 434.8 451.9 468.4 466.7 478.8 Transportation 36.0 33.1 41.3 47.0 40.3 39.5 40.6 Electric Power 175.3 318.8 492.2 444.0 443.2 526.1 545.9 U.S. Territories NO 1.3 2.6 3.0 3.0 3.0 3.0 Petroleum 2,036.6 2,479.7 2,084.0 2,116.9 2,131.1 2,171.3 2,192.7 Residential 97.4 94.9 57.7 63.5 67.4 63.6 58.0 Commercial 73.3 54.9 40.4 42.7 40.4 67.0 55.3 Industrial 310.4 364.0 309.6 321.1 286.8 286.7 269.7 Transportation 1,431.2 1,822.7 1,620.6 1,630.6 1,676.9 1,696.0 1,754.2 Electric Power 97.5 97.9 18.3 22.4 25.3 23.7 21.2 U.S. Territories 26.9 45.4 37.5 36.6 34.3 34.3 34.3 Geothermal3 0.4 0.4 0.4 0.4 0.4 0.4 0.4 Total 4,755.8 5,759.1 5,029.8 5,162.3 5,206.1 5,059.3 4,976.7 NE (Not Estimated) NO (Not Occurring) a Although not technically a fossil fuel, geothermal energy-related CO2 emissions are included for reporting purposes. 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. 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, 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 (C) intensity. The amount of C in fuels varies significantly by fuel type. For example, coal contains the highest amount of C per unit of useful energy. Petroleum has roughly 75 percent of the C per unit of energy as coal, and natural gas has only about 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. 8 Based on national aggregate carbon content of all coal, natural gas, and petroleum fuels combusted in the United States. 3-6 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2016 ------- 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 Table 3-6: Annual Change in CO2 Emissions and Total 2016 Emissions from Fossil Fuel Combustion for Selected Fuels and Sectors (MMT CO2 Eq. and Percent) Sector Fuel Type 2012 to 2013 2013 to 2014 2014 to 2015 2015 to 2016 Total 2016 Electric Power Coal 60.1 4.0% -2.2 -0.1% -218.7 -13.9% -109.2 -8.1% 1,241.3 Electric Power Natural Gas -48.3 -9.8% -0.8 -0.2% 82.9 18.7% 19.8 3.8% 545.9 Electric Power Petroleum 4.1 22.3% 2.9 12.8% -1.6 -6.4% -2.5 -10.4% 21.2 Transportation Petroleum 10.0 0.6% 46.3 2.8% 19.2 1.1% 58.2 3.4% 1,754.2 Residential Natural Gas 41.4 18.4% 11.6 4.4% -24.7 -8.9% -14.9 -5.9% 238.3 Commercial Natural Gas 22.3 14.2% 10.2 5.7% -13.6 -7.2% -5.4 -3.1% 170.3 Industrial Coal 1.7 2.3% -0.1 -0.1% -9.8 -12.9% -6.8 -10.4% 59.0 Industrial Natural Gas 17.1 3.9% 16.5 3.7% -1.7 -0.4% 12.2 2.6% 478.8 All Sectors3 All Fuels3 132.5 2.6% 43.8 0.8% -146.8 -2.8% -82.6 -1.6% 4,976.7 a Includes sector and fuel combinations not shown in this table. 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 2.6 percent increase from 2012 to 2013, then a 0.8 percent increase from 2013 to 2014, then a 2.8 percent decrease from 2014 to 2015, and a 1.6 percent decrease from 2015 to 2016. Total electric power generation remained relatively flat over that time period but emission trends generally mirror the trends in the amount of coal used to generate electricity. The consumption of coal used to generate electricity increased by roughly 4 percent from 2012 to 2013, stayed relatively flat from 2013 to 2014, decreased by 14 percent from 2014 to 2015, and decreased by 8 percent from 2015 to 2016. The overall CO2 emission trends from fossil fuel combustion also follow closely changes in heating degree days over that time period. Heating degree days increased by 18 percent from 2012 to 2013, increased by 2 percent from 2013 to 2014, decreased by 10 percent from 2014 to 2015 and decreased by 5 percent from 2015 to 2016. A decrease in heating degree days leads to decreased demand for heating fuel and electricity for heat in the residential and commercial sector, primarily in winter months. The overall CO2 emission trends from fossil fuel combustion also generally follow changes in overall petroleum use and emissions. Carbon dioxide emissions from all petroleum increased by 1.6 percent from 2012 to 2013, increased by 0.7 percent from 2013 to 2014, increased by 1.9 percent from 2014 to 2015, and increased by 1.0 percent from 2015 to 2016. The increase in petroleum CO2 emissions from 2015 to 2016 somewhat offsets emission reductions from other sources like decreased coal use in the electricity sector. In the United States, 81 percent of the energy used in 2016 was produced through the combustion of fossil fuels such as coal, natural gas, and petroleum (see Figure 3-3 and Figure 3-4). The remaining portion was supplied by nuclear electric power (9 percent) and by a variety of renewable energy sources (10 percent), primarily hydroelectric power, wind energy and biofuels (EIA 2017a).9 Specifically, petroleum supplied the largest share of domestic energy demands, accounting for 37 percent of total U.S. energy used in 2016. Natural gas and coal followed in order of energy demand importance, accounting for approximately 29 percent and 15 percent of total U.S. energy used, respectively. Petroleum was consumed primarily in the transportation end-use sector and the vast majority of coal was used in the electric power end-use sector. Natural gas was broadly consumed in all end-use sectors except transportation (see Figure 3-5) (EIA 2017a). 9 Renewable energy, as defined in EIA's energy statistics, includes the following energy sources: hydroelectric power, geothermal energy, biofuels, solar energy, and wind energy. Energy 3-7 ------- 1 Figure 3-3: 2016 U.S. Energy Consumption by Energy Source (Percent) Nuclear Electric Power 8,6% Renewable Energy 10.4% Petroleum 36.9% Coal 14.8% Natural Gas 29.2% Figure 3-4: U.S. Energy Consumption (Quadrillion Btu) 120- 100- CD cy Cl £ 3 <3 >. as 80- 60- 40 20- Total Energy Fossil Fuels Renewable & Nuclear o CM ro LT> UD r** €0 ON o ro in VD fv 00 Ch o V i rsj ro *r U~> a CTv Ch o\ &¦ <3% a\ G\ CT' Q O o o o o CJ CJ a d: ¦H '— •*— ¦*— -•— tH ¦»— o\ c\ o\ & O O © o c o o o o o o c o O o o O 11 * i v 4 V 1 V 1 * i tH i—< fM Psl rsi rsj rsj rsj rsi rsj OJ rsj f*Nj rsj rsi fM 04 fN 3-8 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2016 ------- 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 Figure 3-5: 2016 CO2 Emissions from Fossil Fuel Combustion by Sector and Fuel Type (MMT COz Eq.) 2,500 2,000 S 1,500 o u i- 11,000 500 Relative Contribution by Fuel Type I Petroleum Coal I Natural Gas 228 41 1,795 1,809 U.S. Territories Commercial Residential Industrial Transportation Electric Power Note on Figure 3-5: Fossil Fuel Combustion includes electric power, which also includes emissions of less than 0.5 MMT CO2 Eq. from geothermal-based generation. Fossil fuels are generally combusted for the purpose of producing energy for useful heat and work. During the combustion process, the C stored in the fuels is oxidized and emitted as CO2 and smaller amounts of other gases, including CH4, CO, and NMVOCs.10 These other C-containing non-CO; gases are emitted as a byproduct of incomplete fuel combustion but are, for the most part, eventually oxidized to CO2 in the atmosphere. Therefore, it is assumed all of the C in fossil fuels used to produce energy is eventually converted to atmospheric CO2. Box 3-3: Weather and Non-Fossil Energy Effects on CO2 from Fossil Fuel Combustion Trends In 2016, weather conditions, and a warm first and fourth quarter of the year in particular, caused a significant decrease in demand for heating fuels and is reflected in the decreased residential emissions from 2015 to 2016. The United States in 2016 also experienced a wanner winter overall compared to 2015, as heating degree days decreased (5.1 percent). Wanner winter conditions compared to 2015 resulted in a decrease in the amount of energy required for heating, and heating degree days in the United States were 14.2 percent below nonnal (see Figure 3-6). Cooling degree days increased, by 4.6 percent, and increased demand for air conditioning in the residential and commercial sector, this led in part to an overall residential electricity demand increase of 0.2 percent. Summer conditions were significantly wanner in 2016 compared to 2015, with cooling degree days 28.0 percent above nonnal (see Figure 3-7) (EIA 2017a).11 10 See the sections entitled Stationary Combustion and Mobile Combustion in this chapter for information on non-CCb 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 1981 through 2010. The variation in these normals during this time period was +12 percent and +19 percent for heating and cooling degree days, respectively (99 percent confidence interval). Energy 3-9 ------- 1 Figure 3-6: Annual Deviations from Normal Heating Degree Days for the United States 2 (1950-2016, Index Normal = 100) 10- -20- Normal (4,524 Heating Degree Days) 99% Confidence Note: Ciimatological normal data are highlighted. Statistical confidence interval for "normal" climatology period of 1981 through 2010. orsi*r vDooorsi*Tv£>co o oj t ud co o cm ud co OfMTu^coooj t y) co o r-j t ud Lnir>i/7t/>u->u3vp^0<:0vorvi>»rvr-»rH»aotx>oococoONgNONONONOQOOQ-'-<^-«'-«-r-i on on o>i on on on On on on on on on on on on on on On on on On On on On on oo ooo oooo *-4 -r-i T-) w —< . *-t *-* w ^ tH —• — — tH ^ t-( H N (N Osl CM rM IN fM M M Figure 3-7: Annual Deviations from Normal Cooling Degree Days for the United States (1950-2016, Index Normal = 100) Normal (1,216 cooling degree days) 99% Confidence _20- Note: Ciimatological normal data are highlighted. Statistical confidence interval for "normal" climatology period of 1981 through 2010. ocvj<3-i£»GOOfNjTry>coof"sj unLnLOLOLnuDUDUDU3'UDr*vfN>. . „ _ On On On On On On On On On On On On On On On On On On On On On UDCOOrsJTUDCOOr-JTT O) CO CO CO CO ON ^ ON i£>COOrvJ------- 1 2016, nuclear power represented 20 percent of total electricity production. In recent years, the wind and solar power 2 sectors have been showing strong growth, such that, on the margin, they are becoming relatively important 3 electricity sources. Between 1990 and 2016, renewable energy generation (in kWh) from solar and wind energy 4 have increased from 0.1 percent in 1990 to 7 percent in 2016, which helped drive the decreases in the carbon 5 intensity of the electricity supply in the United States. 6 7 Fossil Fuel Combustion Emissions by Sector 8 In addition to the CO2 emitted from fossil fuel combustion, CH4 and N20 are emitted from stationary and mobile 9 combustion as well. Table 3-7 provides an overview of the CO2, CH4, and N20 emissions from fossil fuel 10 combustion by sector. 11 Table 3-7: CO2, ChU, and N2O Emissions from Fossil Fuel Combustion by Sector (MMT CO2 12 Eq.) End-Use Sector 1990 2005 2012 2013 2014 2015 2016 Electric Power 1,827.7 2,414.9 2,036.3 2,053.8 2,054.0 1,916.1 1,824.8 CO2 1,820.8 2,400.9 2,022.2 2,038.1 2,038.0 1,900.7 1,808.8 CH4 0.4 0.9 1.1 1.0 1.0 1.1 1.1 N2O 6.- 13.2 13.1 14.6 15.0 14.3 14.9 Transportation 1,518.5 1,900.8 1,689.7 1,703.3 1,740.7 1,757.4 1,815.7 CO2 1,467.2 1,855.8 1,661.9 1,677.6 1,717.1 1,735.5 1,794.9 CH4 9.8 6.6 4.0 3.7 3.4 3.1 3.0 N2O 41.5 38.4 23.8 22.0 20.2 18.8 17.8 Industrial 879.fi 872.6 822.5 852.8 834.8 823.2 811.4 CO2 874 - 867.8 818.4 848.7 830.8 819.3 807.6 CH4 1.9 1.8 1.5 1.5 1.5 1.5 1.4 N2O 3.2 3.0 2.6 2.6 2.5 2.5 2.4 Residential 344.fi 362.8 287.0 335.7 351.4 321.6 300.3 CO2 338. ^ 357.8 282.5 329.7 345.3 316.8 296.2 CH4 5.2 4.1 3.7 5.0 5.1 3.9 3.4 N2O 1.0 0.9 0.7 1.0 1.0 0.8 0.7 Commercial 228. S 228.5 202.5 227.1 235.0 247.2 229.4 CO2 227.4 227.0 201.3 225.7 233.6 245.6 227.9 CH4 1.1 1.1 0.9 1.1 1.1 1.2 1.2 N2O 0.4 0.3 0.3 0.3 0.3 0.4 0.3 U.S. Territories3 27.7 49.9 43.7 42.6 41.5 41.5 41.5 Total 4,826.'J 5,829.5 5,081.7 5,215.3 5,257.4 5,107.0 5,023.1 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 consumption by each end-use sector. 13 Other than CO2, gases emitted from stationary combustion include the greenhouse gases CH4 and N20 and the 14 indirect greenhouse gases NOx, CO, and NMVOCs.13 Methane and N20 emissions from stationary combustion 15 sources depend upon fuel characteristics, size and vintage, along with combustion technology, pollution control 16 equipment, ambient environmental conditions, and operation and maintenance practices. Nitrous oxide emissions 17 from stationary combustion are closely related to air-fuel mixes and combustion temperatures, as well as the 18 characteristics of any pollution control equipment that is employed. Methane emissions from stationary combustion 19 are primarily a function of the CH4 content of the fuel and combustion efficiency. 13 Sulfur dioxide (SO2) emissions from stationary combustion are addressed in Annex 6.3. Energy 3-11 ------- 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 Mobile combustion produces greenhouse gases other than CO2, including CH4, N20, and indirect greenhouse gases including NOx, CO, and NMVOCs. As with stationary combustion, N20 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 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 were defined: industrial, transportation, residential, and commercial. In the table 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 N20 from transportation.14 Emissions from U.S. Territories are also calculated separately due to a lack of end-use-specific consumption data.15 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 3-8. Table 3-8: CO2, ChU, and N2O Emissions from Fossil Fuel Combustion by End-Use Sector (MMT COz Eq.) End-Use Sector 1990 2005 2012 2013 2014 2015 2016 Transportation 1,521.5 1,905.5 1,693.6 1,707.3 1,744.7 1,761.1 1,819.2 CO2 1,470.2 1,860.5 1,665.8 1,681.6 1,721.2 1,739.2 1,798.4 CH4 9.8 6.6 4.0 3.7 3.4 3.1 3.0 N2O 41.5 38.4 23.8 22.0 20.2 18.8 17.8 Industrial 1,568.9 1,613.5 1,419.4 1,452.1 1,432.7 1,377.2 1,322.1 CO2 1,561.3 1,604.4 1,411.2 1,443.4 1,424.0 1,368.8 1,313.8 CH4 2.0 2.0 1.8 1.8 1.8 1.8 1.8 N2O 5.6 7.1 6.4 6.8 6.9 6.6 6.5 Residential 939.9 1,224.1 1,017.3 1,076.2 1,091.9 1,011.4 966.9 CO2 931.4 1,214.1 1,007.8 1,064.6 1,080.0 1,001.1 957.0 CH4 5.4 4.4 4.1 5.3 5.4 4.3 3.8 N2O 3.2 5.6 5.5 6.2 6.4 5.9 6.1 Commercial 768.8 1,036.5 907.7 937.0 946.5 915.7 873.3 CO2 765.3 1,030.3 901.6 930.2 939.6 908.8 866.2 CH4 1.2 1.4 1.3 1.4 1.5 1.6 1.6 N2O 2.3 4.8 4.8 5.4 5.5 5.4 5.6 U.S. Territories3 27.7 49.9 43.7 42.6 41.5 41.5 41.5 Total 4,826.9 5,829.5 5,081.7 5,215.3 5,257.4 5,107.0 5,023.1 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. 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 3-9 presents CO2 emissions from fossil 14 Separate calculations were performed for transportation-related CH4 and N2O. The methodology used to calculate these emissions are discussed in the mobile combustion section. 15 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. 3-12 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2016 ------- 1 fuel combustion by stationary sources. The CO2 emitted is closely linked to the type of fuel being combusted in each 2 sector (see Methodology section of CO2 from Fossil Fuel Combustion). Other than CO2, gases emitted from 3 stationary combustion include the greenhouse gases CH4 and N20. Table 3-10 and Table 3-11 present CH4 and N20 4 emissions from the combustion of fuels in stationary sources. The CH4 and N20 emission estimation methodology 5 utilizes facility-specific technology and fuel use data reported to EPA's Acid Rain Program (EPA 2017a) (see 6 Methodology section for CH4 and N20 from Stationary Combustion). Table 3-7 presents the corresponding direct 7 C02, CH4, and N20 emissions from all sources of fuel combustion, without allocating emissions from electricity use 8 to the end-use sectors. 9 Table 3-9: CO2 Emissions from Stationary Fossil Fuel Combustion (MMT CO2 Eq.) Sector/Fuel Type 1990 2005 2012 2013 2014 2015 2016 Electric Power 1,820.8 2,400.9 2,022.2 2,038.1 2,038.0 1,900.7 1,808.8 Coal 1,547.6 1,983.8 1,511.2 1,571.3 1,569.1 1,350.5 1,241.3 Natural Gas 175.3 318.8 492.2 444.0 443.2 526.1 545.9 Fuel Oil 97.5 97.9 18.3 22.4 25.3 23.7 21.2 Geo thermal 0.4 0.4 0.4 0.4 0.4 0.4 0.4 Industrial 874.5 867.8 818.4 848.7 830.8 819.3 807.6 Coal 155.3 115.3 74.1 75.7 75.6 65.9 59.0 Natural Gas 408.9 388.5 434.8 451.9 468.4 466.7 478.8 Fuel Oil 310.4 364.0 309.6 321.1 286.8 286.7 269.7 Commercial 227.4 227.0 201.3 225.7 233.6 245.6 227.9 Coal 12.0 9.3 4.1 3.9 3.8 2.9 2.3 Natural Gas 142.1 162.9 156.9 179.1 189.3 175.7 170.3 Fuel Oil 73.3 54.9 40.4 42.7 40.4 67.0 55.3 Residential 338.3 357.8 282.5 329.7 345.3 316.8 296.2 Coal 3.0 0.8 NO NO NO NO NO Natural Gas 238.0 262.2 224.8 266.2 277.9 253.2 238.3 Fuel Oil 97.4 94.9 57.7 63.5 67.4 63.6 58.0 U.S. Territories 27.6 49.7 43.5 42.5 41.4 41.4 41.4 Coal 0.6 3.0 3.4 2.8 4.0 4.0 4.0 Natural Gas NO 1.3 2.6 3.0 3.0 3.0 3.0 Fuel Oil 26.9 45.4 37.5 36.6 34.3 34.3 34.3 Total 3,288.6 3,903.3 3,367.9 3,484.7 3,489.0 3,323.8 3,181.9 NO (Not Occurring) Note: Totals may not sum due to independent rounding. 10 Table 3-10: ChU Emissions from Stationary Combustion (MMT CO2 Eq.) Sector/Fuel Type 1990 2005 2012 2013 2014 2015 2016 Electric Power 0.4 0.9 1.1 1.0 1.0 1.1 1.1 Coal 0.3 0.4 0.3 0.3 0.3 0.3 0.2 Fuel Oil + + + + + + + Natural gas 0.1 0.5 0.8 0.7 0.7 0.9 0.9 Wood + + + + + + + Industrial 1.9 1.8 1.5 1.5 1.5 1.5 1.4 Coal 0.4 0.3 0.2 0.2 0.2 0.2 0.2 Fuel Oil 0.2 0.3 0.2 0.2 0.2 0.2 0.2 Natural gas 0.2 0.2 0.2 0.2 0.2 0.2 0.2 Wood 1.0 1.0 1.0 0.9 0.9 0.9 0.9 Commercial 1.1 1.1 0.9 1.1 1.1 1.2 1.2 Coal + + + + + + + Fuel Oil 0.3 0.2 0.1 0.2 0.1 0.2 0.2 Natural gas 0.3 0.4 0.4 0.4 0.4 0.4 0.4 Wood 0.5 0.5 0.4 0.5 0.5 0.6 0.6 Energy 3-13 ------- Residential 5.2 4.1 3.7 5.0 5.1 3.9 3.4 Coal 0.2 0.1 NO NO NO NO NO Fuel Oil 0.3 0.3 0.2 0.2 0.2 0.2 0.2 Natural Gas 0.5 0.6 0.5 0.6 0.6 0.6 0.5 Wood 4.1 3.1 3.0 4.1 4.2 3.1 2.7 U.S. Territories + 0.1 0.1 0.1 0.1 0.1 0.1 Coal + + + + + + + Fuel Oil + 0.1 0.1 0.1 0.1 0.1 0.1 Natural Gas NO + + + + + + Wood NO NO NO NO NO NO NO Total 8.6 7.9 7.3 8.7 8.8 7.8 7.2 + Does not exceed 0.05 MMT CO2 Eq. NO (Not Occurring) Note: Totals may not sum due to independent rounding. 1 Table 3-11: N2O Emissions from Stationary Combustion (MMT CO2 Eq.) Sector/Fuel Type 1990 2005 2012 2013 2014 2015 2016 Electric Power 6.5 13.2 13.1 14.6 15.0 14.3 14.9 Coal 6.1 11.1 9.8 11.6 11.9 10.6 11.1 Fuel Oil 0.1 0.1 + + + + + Natural Gas 0.3 1.9 3.2 3.0 3.1 3.6 3.7 Wood + + + + + + + Industrial 3.2 3.0 2.6 2.6 2.5 2.5 2.4 Coal 0.7 0.5 0.4 0.4 0.4 0.3 0.3 Fuel Oil 0.6 0.6 0.5 0.5 0.4 0.4 0.4 Natural Gas 0.2 0.2 0.2 0.2 0.3 0.2 0.3 Wood 1.6 1.6 1.5 1.5 1.5 1.5 1.5 Commercial 0.4 0.3 0.3 0.3 0.3 0.4 0.3 Coal 0.1 + + + + + + Fuel Oil 0.2 0.1 0.1 0.1 0.1 0.2 0.1 Natural Gas 0.1 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 1.0 0.9 0.7 1.0 1.0 0.8 0.7 Coal + + NO NO NO NO NO Fuel Oil 0.2 0.2 0.2 0.2 0.2 0.2 0.2 Natural Gas 0.1 0.1 0.1 0.1 0.1 0.1 0.1 Wood 0.7 0.5 0.5 0.7 0.7 0.5 0.4 U.S. Territories 0.1 0.1 0.1 0.1 0.1 0.1 0.1 Coal + + + + + + + Fuel Oil 0.1 0.1 0.1 0.1 0.1 0.1 0.1 Natural Gas NO + + + + + + Wood NO NO NO NO NO NO NO Total 11.1 17.5 16.8 18.6 18.9 18.0 18.4 + Does not exceed 0.05 MMT CO2 Eq. NO (Not Occurring) Note: Totals may not sum due to independent rounding. 2 Electric Power Sector 3 The process of generating electricity is the single largest source of CO2 emissions in the United States, representing 4 34 percent of total CO2 emissions from all CO2 emissions sources across the United States. Methane and N20 5 accounted for a small portion of total greenhouse gas emissions from electric power, representing 0.1 percent and 6 0.8 percent, respectively. Electric power also accounted for the largest share of CO2 emissions from fossil fuel 7 combustion, approximately 36.3 percent in 2016. Methane and N2O from electric power represented 11.2 and 41.1 8 percent of total CH4 and N20 emissions from fossil fuel combustion in 2016, respectively. 3-14 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2016 ------- 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 For the underlying energy data used in this chapter, the Energy Information Administration (EIA) places electric power generation into three functional categories: the electric power sector, the commercial sector, and the industrial sector. The electric power sector consists of electric utilities and independent power producers whose primary business is the production of electricity. This includes both regulated utilities and non-utilities (e.g., independent power producers, qualifying co-generators, and other small power producers). Electric generation is reported as occurring in other sectors where the producer of the power indicates that its primary business is something other than the production of electricity.16 Emissions from the electric power sector have decreased by 0.2 percent since 1990. The carbon intensity of the electric power sector, in terms of CO2 Eq. per QBtu, input has significantly decreased by 12 percent during that same timeframe with the majority of the emissions and carbon intensity decreases coming in the past decade as shown below in Figure 3-8. This recent decarbonization of the electric power sector is a result of several key drivers. Coal-fired electric power (in kilowatt-hours [kWh]) decreased from almost 54 percent of generation in 1990 to 32 percent in 2016.17 This generation corresponded with an increase in natural gas 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 27-year period to represent 33 percent of electric power sector generation in 2016. In 2016, CO2 emissions from the electric power sector decreased by 4.8 percent relative to 2015. This decrease in CO2 emissions was a result of changes in the types of fuel consumed to produce electricity in the electric power sector in recent years. The shift from coal to less-CCh-intensive natural gas to supply electricity has accelerated in recent years. Consumption of coal for electric power decreased by 8.1 percent from 2015 to 2016, while consumption of natural gas increased by 3.8 percent. There has also been a rapid increase in renewable energy capacity additions in the electric power sector in recent years. In 2016, renewable energy sources accounted for 63 percent of capacity additions, with natural gas accounting for the remaining additions. The share of renewable energy capacity additions has grown significantly since 2010, when renewable energy sources accounted for only 28 percent of total capacity additions (EIA 2017e). The decrease in coal-powered electricity generation and increase in renewable energy capacity contributed to a decrease in emissions from electric power generation over the time series (see Figure 3-8). Decreases in natural gas costs and the associated increase in natural gas generation, particularly between 2005 and 2016, was one of the main driver of the recent fuel switching and decrease in electric power sector carbon intensity. During this time period, the cost of natural gas (in $/MMBtu) decreased by 57 percent while the cost of coal (in$/MMBtu) increased by 83 percent (EIA 2017a). Also, between 1990 and 2016, renewable energy generation (in kWh) from wind and solar energy have increased from 0.1 percent in 1990 to 7 percent in 2016, which also helped drive the decrease in electric power sector carbon intensity. This decrease in carbon intensity occurred even as total electricity retail sales increased 37 percent, from 2,713 billion kWh in 1990 to 3,711 billion kWh in 2016. 16 Utilities primarily generate power for the U.S. electric grid for sale to retail customers. Non-utilities produce electricity for their own use, to sell to large consumers, or to sell on the wholesale electricity market (e.g., to utilities for distribution and resale to customers). 17 Values represent electricity net generation from the electric power sector (EIA 2017a). Energy 3-15 ------- 1 Figure 3-8: Fuels Used in Electric Power Generation (TBtu) and Total Electric Power Sector 2 CO2 Emissions I Petroleum (TBtu) I Nuclear (TBtu) Renewable Energy Sources (TBtu) I Natural Gas (TBtu) Coal (TBtu) Net Generation (Index vs. 1990) [Right Axis] I Sector COs Emissions (Index vs. 1990) [Right Axis] 30,000 25,000 ' 20,000 15,000 10,000 5,000 _ , On On On On On On On On On G* C5 CD • i ~—i ~w—4 < ononononononononononSoooocjcjooooooo coo Electricity was consumed primarily in the residential, commercial, and industrial end-use sectors for lighting, heating, electric motors, appliances, electronics, and air conditioning (see Figure 3-9). Figure 3-9: Electric Power Retail Sales by End-Use Sector (Billion kWh) 160 140 120 100 80 60 40 20 u T3 1,500 1,400- 1,300- 1,200 1,000- 900- 800 Residential Commercia Industrial O rH fN ro IA VD ON On ON On On On On On On On On On On On On oo On ON On gN O -h o o OJ PnJ T id VD N o o o c o o o o N (N fN fN ON O C -H a o r-j fNj 8 The industrial, residential, and commercial end-use sectors, as presented in Table 3-8, were reliant on electricity for 9 meeting energy' needs. The residential and commercial end-use sectors are especially reliant on electricity use for 10 lighting, heating, air conditioning, and operating appliances. In 2016. electricity sales to the residential end-use 11 sector increased by 0.2 percent and sales to the commercial end-use sector decreased by 0.1 percent, respectively. 12 Electricity sales to the industrial sector in 2016 decreased approximately 5.1 percent. Overall, in 2016, the amount of 13 electricity retail sales (in kWh) decreased by 1.2 percent. 3-16 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2016 ------- 1 Industrial Sector 2 Industrial sector CO2, CH4, and N20, emissions accounted for 16, 14, and 7 percent of CO2, CH4, and N20, 3 emissions from fossil fuel combustion, respectively. Carbon dioxide, CH4, and N20 emissions resulted from the 4 direct consumption of fossil fuels for steam and process heat production. 5 The industrial end-use sector, per the underlying energy use data from EIA, includes activities such as 6 manufacturing, construction, mining, and agriculture. The largest of these activities in terms of energy use is 7 manufacturing, of which six industries—Petroleum Refineries, Chemicals, Paper, Primary Metals, Food, and 8 Nonmetallic Mineral Products—represent the vast majority of the energy use (EIA 2017a; EIA 2009b). 9 There are many dynamics that impact emissions from the industrial sector including economic activity, changes in 10 the make-up of the industrial sector, changes in the emissions intensity of industrial processes, and weather impacts 11 on heating of industrial buildings.18 Structural changes within the U.S. economy that lead to shifts in industrial 12 output away from energy-intensive manufacturing products to less energy-intensive products (e.g., from steel to 13 computer equipment) have had a significant effect on industrial emissions. 14 From 2015 to 2016, total industrial production and manufacturing output decreased by 1.2 percent (FRB 2017). 15 Over this period, output increased across production indices for Food, Petroleum Refineries, Chemicals, and 16 Nonmetallic Mineral Products, and decreased slightly for Primary Metals and Paper (see Figure 3-10). Through 17 EPA's Greenhouse Gas Reporting Program (GHGRP), specific industrial sector trends can be discerned from the 18 overall total EIA industrial fuel consumption data used for these calculations. 19 For example, from 2015 to 2016, the underlying EIA data showed decreased consumption of coal, and relatively flat 20 use of natural gas in the industrial sector. The GHGRP data highlights that several industries contributed to these 21 trends, including chemical manufacturing; pulp, paper and print; and food processing, beverages and tobacco.19 18 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. 19 Further details on industrial sector combustion emissions are provided by EPA's GHGRP. See . Energy 3-17 ------- 1 Figure 3-10: Industrial Production Indices (Index 2012=100) 140- 120- Total excluding Computers, Communications Equipment, and Semiconductors 100- Total Industrial 80- Stone, Clay, & Glass Products Chemicals 140- 120- Primary Metals 100J 80- Petroleum Refineries 3 Despite the growth in industrial output (60 percent) and the overall U.S. economy (87 percent) from 1990 to 2016, 4 CO?, emissions from fossil fuel combustion in the industrial sector decreased by 7.7 percent over the same time 5 series. A number of factors are believed to have caused this disparity between growth in industrial output and 6 decrease in industrial emissions, including: (1) more rapid growth in output from less energy-intensive industries 7 relative to traditional manufacturing industries, and (2) energy-intensive industries such as steel are employing new 8 methods, such as electric arc furnaces, that are less carbon intensive than the older methods. In 2016, CO?, CH4. and 9 N2O emissions from fossil fuel combustion and electricity use within the industrial end-use sector totaled 1,322.1 10 MMT CO2 Eq., a 4.0 percent decrease from 2015 emissions. 11 Residential and Commercial Sectors 12 Emissions from the residential and commercial sectors have increased since 1990, and are often correlated with 13 short-term fluctuations in energy use caused by weather conditions, rather than prevailing economic conditions. 14 More significant changes in emissions from the residential and commercial sectors in recent years can be largely 3-18 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2016 ------- 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 attributed to an overall reduction in energy use, a reduction in heating degree days, and increases in energy efficiency (see Figure 3-11). In 2016 the residential and commercial sectors accounted for 6 and 5 percent of CO2 emissions from fossil fuel combustion, 33 and 11 percent of CH4 emissions from fossil fuel combustion, and 2 and 1 percent of N20 emissions from fossil fuel combustion respectively. Emissions from these sectors were largely due to 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 both of these end-use sectors. In 2016, total emissions (CO2, CH4, and N20) from fossil fuel combustion and electricity use within the residential and commercial end-use sectors were 966.9 MMT CO2 Eq. and 873.3 MMT CO2 Eq., respectively. Total CO2, CH4, and N20 emissions from fossil fuel combustion and electricity use within the residential and commercial end-use sectors decreased by 4.4 and 4.6 percent from 2015 to 2016, respectively, and heating degree days decreased by 5 percent over the same time period. A decrease in heating degree days led to a decreased demand for heating fuel and electricity for heat in the residential and commercial sectors. In addition, a shift toward energy efficient products and more stringent energy efficiency standards for household equipment lias also contributed to a decrease in energy demand in households (EIA 2017f), resulting in a decrease in energy-related emissions. In the long term, the residential sector is also affected by population growth, migration trends toward wanner areas, and changes in housing and building attributes (e.g., larger sizes and improved insulation). In 2016, combustion emissions from natural gas consumption represented 80 and 75 percent of the direct fossil fuel CO2 emissions from the residential and commercial sectors, respectively. Natural gas combustion CO2 emissions from the residential and commercial sectors in 2016 decreased by 5.9 percent and 3.1 percent from 2015 levels, respectively. Figure 3-11: Fuels Used in Residential and Commercial Sectors (TBtu), Heating Degree Days, and Total Sector CO2 Emissions Coal (TBtu) | Heating Degree Days (Index vs. 1990) [Right Axis] Renewable Energy Sources (TBtu) ¦ Sector CO: Emissions (Index vs. 1990) [Right Axis] 140 ¦ Petroleum (TBtu) ¦ Natural Gas (TBtu) Electricity Use (TBtu) 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 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, CH4, and N20 emissions are not presented for U.S. Territories in the tables above by sector, though the emissions will include some transportation and mobile combustion sources. Energy 3-19 ------- 1 Transportation Sector and Mobile Combustion 2 This discussion of transportation emissions follows the alternative method of presenting combustion emissions by 3 allocating emissions associated with electricity generation to the transportation end-use sector, as presented in Table 4 3-8. Table 3-7 presents direct CO2, CH4, and N20 emissions from all transportation sources (i.e., excluding 5 emissions allocated to electricity consumption in the transportation end-use sector). 6 The transportation end-use sector and other mobile combustion accounted for 1,819.2 MMT CO2 Eq. in 2016, which 7 represented 36 percent of CO2 emissions, 29 percent of CH4 emissions, and 49 percent of N20 emissions from fossil 8 fuel combustion, respectively.20 Fuel purchased in the United States for international aircraft and marine travel 9 accounted for an additional 115.5 MMT CO2 Eq. in 2016; these emissions are recorded as international bunkers and 10 are not included in U.S. totals according to UNFCCC reporting protocols. 11 Transportation End-Use Sector 12 From 1990 to 2016, transportation emissions from fossil fuel combustion rose by 20 percent due, in large part, to 13 increased demand for travel (see Figure 3-12). The number of vehicle miles traveled (VMT) by light-duty motor 14 vehicles (passenger cars and light-duty trucks) increased 43 percent from 1990 to 2016,21 as a result of a confluence 15 of factors including population growth, economic growth, urban sprawl, and periods of low fuel prices. 16 From 2015 to 2016, CO2 emissions from the transportation end-use sector increased by 3.4 percent. The increase in 17 emissions can largely be attributed to increased VMT and motor gasoline consumption by light duty vehicles, as 18 well as diesel consumption by medium-and heavy-duty vehicles. From 2015 to 2016, there were also increases in 19 residual fuel oil consumption by ships and boats and jet fuel use in general aviation aircraft. 20 Commercial aircraft emissions were similar between 2015 and 2016, but have decreased 15 percent since 2007 21 (FAA 2017).22 Decreases in jet fuel emissions (excluding bunkers) since 2007 are due in part to improved 22 operational efficiency that results in more direct flight routing, improvements in aircraft and engine technologies to 23 reduce fuel burn and emissions, and the accelerated retirement of older, less fuel efficient aircraft. 24 Almost all of the energy consumed for transportation was supplied by petroleum-based products, with more than 25 half being related to gasoline consumption in automobiles and other highway vehicles. Other fuel uses, especially 26 diesel fuel for freight trucks and jet fuel for aircraft, accounted for the remainder. The primary driver of 27 transportation-related emissions was CO2 from fossil fuel combustion. Annex 3.2 presents the total emissions from 28 all transportation and mobile sources, including CO2, N20, CH4, and HFCs. 20 Note that these totals include CO2, CH4 and N2O emissions from some sources in the U.S. Territories (ships and boats, recreational boats, non-transportation mobile sources) and CH4 and N2O emissions from transportation rail electricity. 21 VMT estimates are based on data from FHWA Highway Statistics Table VM-1 (FHWA 1996 through 2017). Table VM-1 data for 2016 has not been published yet, therefore 2016 mileage data is estimated using the 1.7 percent increase in FHWA Traffic Volume Trends from 2015 to 2016. In 2011, FHWA changed its methods for estimating VMT by vehicle class, which led to a shift in VMT and emissions among on-road vehicle classes in the 2007 to 2016 time period. In absence of these method changes, light-duty VMT growth between 1990 and 2016 would likely have been even higher. 22 Commercial aircraft, as modeled in FAA's AEDT (FAA 2017), consists of passenger aircraft, cargo, and other chartered flights. 3-20 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2016 ------- 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 Figure 3-12: Fuels Used in Transportation Sector (TBtu), Onroad VMT, and Total Sector CO2 Emissions ¦ Aviation Gasoline (TBtu) 30,000 ¦ LPG (TBtu) ¦ Residual Fuel (TBtu) I Natural Gas (TBtu) Renewable Energy Sources (TBtu) Jet Fuel Distillate Fuel (TBtu) I Motor Gasoline (TBtu) Onroad VMT (Index vs. 1990) [Right Axis] I Sector CO; Emissions (Index vs. 1990) [Right Axis] 160 140 co eh o h f\i n t o o I - Transportation Fossil Fuel Combustion CO 2 Emissions Domestic transportation CO2 emissions increased by 22 percent (328.2 MMT CO2) between 1990 and 2016, an annualized increase of 0.8 percent. Among domestic transportation sources in 2016, light-duty vehicles (including passenger cars and light-duty trucks) represented 59 percent of CO2 emissions from fossil fuel combustion, medium- and heavy-duty trucks and buses 24 percent, commercial aircraft 7 percent, and other sources 10 percent. See Table 3-12 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 lias increased from 0.7 billion gallons in 1990 to 13.5 billion gallons in 2016, while biodiesel consumption has increased from 0.01 billion gallons in 2001 to 2.1 billion gallons in 2016. For further information, see Section 3.11 on biofuel consumption at the end of this chapter and Table A-96 in Annex 3.2. Carbon dioxide emissions from passenger cars and light-duty trucks totaled 1,064.7 MMT CO2 in 2016. This is an increase of 15 percent (140.2 MMT CO2) from 1990 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 2016). Carbon dioxide emissions from passenger cars and light-duty trucks peaked at 1,150.6 MMT CO2 in 2004, and since then have declined about 7 percent. The decline in new light-duty vehicle fuel economy between 1990 and 2004 (Figure 3-13) reflected 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 VMT grew only 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 . Energy 3-21 ------- 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 modestly for much of the period. Light-duty VMT grew by less than one percent or declined each year between 2005 and 201324 and has since grown at a faster rate (2.6 percent from 2014 to 2015, and 1.7 percent from 2015 to 2016). Average new vehicle fuel economy has increased almost every year since 2005, while the light-duty truck share decreased to about 33 percent in 2009 and has since varied from year to year between 36 and 43 percent. Light-duty truck share is about 38 percent of new vehicles in model year 2016 (EPA 2016a). See also 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 83 percent from 1990 to 2016. This increase was largely due to a substantial growth in medium- and heavy-duty truck VMT, which increased by 98 percent between 1990 and 20 1 6.25 Carbon dioxide from the domestic operation of commercial aircraft increased by 8 percent (9.1 MMT CO2) from 1990 to 2016.26 Across all categories of aviation, excluding international bunkers, CO2 emissions decreased by 9 percent (17.7 MMT CO2) between 1990 and 2016 27 This includes a 65 percent (22.8 MMT CO2) decrease in CO2 emissions from domestic military operations. Transportation sources also produce CH4 and N20; these emissions are included in Table 3-13 and Table 3-14 and in the CH4 and N20 from Mobile Combustion section. Annex 3.2 presents total emissions from all transportation and mobile sources, including CO2, CH4, N20, and HFCs. 24 VMT estimates are based on data from FHWA Highway Statistics Table VM-1 (FHWA 1996 through 2017). Table VM-1 data for 2016 has not been published yet, therefore 2016 mileage data is estimated using the 1.7 percent increase in FHWA Traffic Volume Trends from 2015 to 2016. In 2007 and 2008 light-duty VMT decreased 3.0 percent and 2.3 percent, respectively. Note that the decline in light-duty VMT from 2006 to 2007 is due at least in part to a change in FHWA's methods for estimating VMT. In 2011, FHWA changed its methods for estimating VMT by vehicle class, which led to a shift in VMT and emissions among on-road vehicle classes in the 2007 to 2016 time period. In absence of these method changes, light-duty VMT growth between 2006 and 2007 would likely have been higher. 25 While FHWA data shows consistent growth in medium- and heavy-duty truck VMT over the 1990 to 2016 time period, part of the growth reflects a method change for estimating VMT starting in 2007. This change in methodology in FHWA's VM-1 table resulted in large changes in VMT by vehicle class, thus leading to a shift in VMT and emissions among on-road vehicle classes in the 2007 to 2016 time period. During the time period prior to the method change (1990 to 2006), VMT for medium- and heavy-duty trucks increased by 51 percent. 26 Commercial aircraft, as modeled in FAA's AEDT, consists of passenger aircraft, cargo, and other chartered flights. 27 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. 3-22 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2016 ------- 1 Figure 3-13: Sales-Weighted Fuel Economy of New Passenger Cars and Light-Duty Trucks, 2 1990-2016 (miles/gallon) 30.0. 29.0- 28.0- 27.0- 26,0- c 25.0- o 24.0- 23.0- 22.0- 21.0- 20.0- 19.0- 18.0- 17.0" o-^rMro^-mvDr^oochO-rHfNm'g-cnuDf,^oo\D 4 Source: EPA (2016a) 5 6 Figure 3-14: Sales of New Passenger Cars and Light-Duty Trucks, 1990-2016 (Percent) 100% 75%- % Passenger ° 50%- % Light-Duty Trucks 25%- 0% 8 Source: EPA (2016a) Energy 3-23 ------- 1 2 Table 3-12: CO2 Emissions from Fossil Fuel Combustion in Transportation End-Use Sector 3 (MMT COz Eq.) Fuel/Vehicle Type 1990 2005 2012a 2013a 2014a 2015a 2016a Gasolineb 956.9 1,152.4 1,029.8 1,030.2 1,072.0 1,070.5 1,102.7 Passenger Cars 604.4 638.3 707.2 706.9 725.4 731.3 753.6 Light-Duty Trucks 300.6 464.4 268.2 268.3 290.2 283.2 291.8 Medium- and Heavy-Duty Trucksc 37." 33.9 37.4 38.2 39.5 39.3 40.5 Buses 0.3 0.4 0.8 0.8 0.9 0.9 0.9 Motorcycles ir 1.6 3.9 3.7 3.7 3.7 3.8 Recreational Boats'1 12.3 13.8 12.3 12.2 12.2 12.2 12.1 Distillate Fuel Oil (Diesel)b 262.9 457.5 427.5 433.9 446.3 459.8 466.1 Passenger Cars 7.9 4.2 4.1 4.1 4.1 4.3 4.4 Light-Duty Trucks 11.5 25.8 12.9 12.9 13.8 13.9 14.2 Medium- and Heavy-Duty Trucksc 190.5 360.2 344.4 350.0 360.0 368.6 377.4 Buses 8.0 10.6 15.4 15.5 16.8 17.3 17.7 Rail 35.5 45.5 39.5 40.1 41.5 39.8 36.7 Recreational Boats 2.0 3.2 3.7 3.7 3.8 3.9 4.0 Ships and Non-Recreational Boats6 7.5 8.0 7.5 7.5 6.1 12.0 11.6 International Bunker Fuel/ 11.' 9.4 6.8 5.6 6.1 8.4 8.7 Jet Fuel 184.2 189.3 143.4 147.1 148.4 157.6 168.2 Commercial Aircraft8 109.9 132.7 113.3 114.3 115.2 119.0 119.0 Military Aircraft 35.0 19.4 12.1 11.0 14.0 13.5 12.2 General Aviation Aircraft 39.4 37.3 18.0 21.8 19.2 25.1 37.0 International Bunker Fuel/ 38.0 60.1 64.5 65.7 69.6 71.9 71.9 International Bunker Fuels from Commercial Aviation 30.0 55.6 61.4 62.8 66.3 68.6 68.6 Aviation Gasoline 3.1 2.4 1.7 1.5 1.5 1.5 1.4 General Aviation Aircraft 3.1 2.4 1.7 1.5 1.5 1.5 1.4 Residual Fuel Oil 22.6 19.3 15.8 15.1 5.8 4.2 13.4 Ships and Boats6 22.6 19.3 15.8 15.1 5.8 4.2 13.4 International Bunker Fuel/ 53.' 43.6 34.5 28.5 27.7 30.6 33.8 Natural Gas J 36.0 33.1 41.3 47.0 40.3 39.5 40.6 Passenger Cars + /J + + + + + + Light-Duty Trucks + + + + + + + Medium- and Heavy-Duty Trucks + + + + + + + Buses + :i:; 0.6 0.8 0.8 0.8 0.9 1.0 Pipeline11 36.0 32.4 40.5 46.2 39.4 38.5 39.6 LPGJ 1.4 1.7 2.3 2.7 2.9 2.5 2.5 Passenger Cars + + + + + 0.2 0.4 Light-Duty Trucks 0.2 0.3 0.2 0.3 0.6 0.4 0.2 Medium- and Heavy-Duty Trucks6 1.1 1.3 1.8 2.1 1.9 1.6 1.6 Buses 0.1 0.1 0.3 0.4 0.3 0.3 0.2 Electricity 3.0 4.7 3.9 4.0 4.1 3.7 3.5 Rail 3.0 4.7 3.9 4.0 4.1 3.7 3.5 Totalk 1,470.2 1,860.5 1,665.8 1,681.6 1,721.2 1,739.2 1,798.4 Total (Including Bunkers)' 1,573.7 1,973.6 1,771.6 1,781.4 1,824.6 1,850.1 1,912.8 Biofuels-Ethanol' 4.1 22.4 71.5 73.4 74.9 75.9 78.2 Biofuels-Biodiesel' 0 0.9 8.5 13.5 13.3 14.1 19.6 4 + Does not exceed 0.05 MMT CO2 Eq. 3-24 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2016 ------- 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 a In 2011 FHWA changed its methods for estimating vehicle miles traveled (VMT) and related data. These methodological changes included how vehicles are classified, moving from a system based on body-type to one that is based on wheelbase. These changes were first incorporated for the 1990 through 2010 Inventory and apply to the 2007 through 2016 time period. This resulted in large changes in VMT and fuel consumption data by vehicle class, thus leading to a shift in emissions among on-road vehicle classes. b 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 2017). Table VM-1 data for 2016 has not been published yet, therefore 2016 mileage data is estimated using the 1.7 percent increase in FHWA Traffic Volume Trends from 2015 to 2016. Data from Table VM-1 is used to estimate the share of consumption between each on-road vehicle class. These fuel consumption estimates are combined with estimates of fuel shares by vehicle type from DOE's TEDB Annex Tables A. 1 through A.6 (DOE 1993 through 2016). TEDB data for 2015 and 2016 has not been published yet, therefore 2014 data is used as a proxy. c Includes medium- and heavy-duty trucks over 8,500 lbs. d In 2014, EPA incorporated the NONROAD2008 model into MOVES2014. The current Inventory uses the NONROAD component ofMOVES2014a foryears 1999 through 2016. e Note that large year over year fluctuations in emission estimates partially reflect nature of data collection for these sources. f Official estimates exclude emissions from the combustion of both aviation and marine international bunker fuels; however, estimates including international bunker fuel-related emissions are presented for informational purposes. B Commercial aircraft, as modeled in FAA's AEDT, consists of passenger aircraft, cargo, and other chartered flights. h Pipelines reflect CO2 emissions from natural gas powered pipelines transporting natural gas. 'Ethanol andbiodiesel estimates are presented for informational purposes only. See Section 3.11 of this chapter and the estimates in Land Use, Land-Use Change, and Forestry (see Chapter 6), in line with IPCC methodological guidance and UNFCCC reporting obligations, for more information on ethanol and biodiesel. J Transportation sector natural gas and LPG consumption are based on data fromEIA (2017). Prior to the previous (i.e., 1990 through 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 previous Inventory and apply to the 1990 to 2016 time period. k Includes emissions from rail electricity. 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 CO2 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 andN2O Emissions Mobile combustion includes emissions of CH4 and N20 from all transportation sources identified in the U.S. Inventory with the exception of pipelines and electric locomotives;28 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.). 29 Annex 3.2 includes a summary of all emissions from both transportation 28 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. 29 See the methodology sub-sections of the CO2 from Fossil Fuel Combustion and CH4 and N2O from Mobile Combustion sections of this chapter. Note that N2O and CH4 emissions are reported using different categories than CO2. CO2 emissions are reported by end-use sector (Transportation, Industrial, Commercial, Residential, U.S. Territories), and generally adhere to a top- down approach to estimating emissions. CO2 emissions from non-transportation sources (e.g., lawn and garden equipment, farm equipment, construction equipment) are allocated to their respective end-use sector (i.e., construction equipment CO2 emissions are included in the Industrial end-use sector instead of the Transportation end-use sector). CH4 and N2O emissions are reported using the "Mobile Combustion" category, which includes non-transportation mobile sources. CH4 and N2O 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, CO2 emissions from non- transportation mobile sources are presented separately from their overall end-use sector in Annex 3.2. Energy 3-25 ------- 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 and mobile sources. Table 3-13 and Table 3-14 provide mobile fossil fuel CH4 and N20 emission estimates in MMT C02 Eq.30 Mobile combustion was responsible for a small portion of national CH4 emissions (0.5 percent) but was the fourth largest source of U.S. N20 emissions (4.8 percent). From 1990 to 2016, mobile source CH4 emissions declined by 70 percent, to 3.0 MMT CO: Eq. (119 kt CH4), due largely to control technologies employed in on-road vehicles since the mid-1990s to reduce CO, NOx, NMVOC, and CH4 emissions. Mobile source emissions of N20 decreased by 57 percent, to 17.8 MMT C02 Eq. (60 kt N20). Earlier generation control technologies initially resulted in higher N20 emissions, causing a 30 percent increase in N20 emissions from mobile sources between 1990 and 1997. Improvements in later-generation emission control technologies have reduced N20 output, resulting in a 67 percent decrease in mobile source N20 emissions from 1997 to 2016 (Figure 3-15). Overall, CH4 and N20 emissions were predominantly from gasoline-fueled passenger cars and light-duty trucks. See also Annex 3.2 for data by vehicle mode and information on VMT and the share of new vehicles (in VMT). Figure 3-15: Mobile Source ChU and N2O Emissions (MMT CO2 Eq.) 50- 40- O" LU 8 30- H 2: X 20- 10- Table 3-13: ChU Emissions from Mobile Combustion (MMT CO2 Eq.) Fuel Type/Vehicle Type3 1990 2005 2012 2013 2014 2015 2016 Gasoline On-Roadb 5.2 2.2 1.3 1.1 1.0 0.9 0.8 Passenger Cars 3.2 1.3 0.9 0.8 0.7 0.6 0.6 Light-Duty Trucks 1.7 0.8 0.3 0.3 0.2 0.2 0.2 Medium- and Heavy-Duty Trucks and Buses 0.3 0.1 0.1 0.1 + + + Motorcycles + + + + + + + Diesel On-Roadb + + + + + + + Passenger Cars + + + + + + + Light-Duty Trucks + + + + + + + Medium- and Heavy-Duty Trucks and Buses + + + + + + + Alternative Fuel On-Road + 0.2 0.3 0.3 0.2 0.3 0.3 Non-Roadc 4.6 4.2 2.4 2.3 2.1 1.9 1.8 Ships and Boats 0.5 0.5 0.4 0.4 0.3 0.3 0.3 30 See Annex 3.2 for a complete time series of emission estimates for 1990 through 2016. 3-26 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2016 ------- Rail 0.1 0.1 0.1 0.1 0.1 0.1 0.1 Aircraft 0.1 0.1 + + + + + Agricultural Equipment"1 0.4 0.4 0.2 0.2 0.2 0.1 0.1 Construction/Mining Equipment6 0.4 0.3 0.2 0.3 0.2 0.1 0.1 Otherf 3.1 2.8 1.5 1.4 1.3 1.2 1.2 Total 9.8 6.6 4.0 3.7 3.4 3.1 3.0 + Does not exceed 0.05 MMT CO2 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 (FHWA 1996 through 2017). Table VM-1 data for 2016 has not been published yet, therefore 2016 mileage data is estimated using the 1.7 percent increase in FHWA Traffic Volume Trends from 2015 to 2016. These mileage estimates are combined with estimates of fuel shares by vehicle type from DOE's TEDB Annex Tables A.l through A.6 (DOE 1993 through 2016). TEDB data for 2015 and 2016 has not been published yet, therefore 2014 data is used as a proxy for the Public Review draft. c Rail emissions do not include emissions from electric powered locomotives. Class II, Class III, commuter, and intercity rail diesel consumption data for 2014 to 2016 are not available yet, therefore 2013 data is used as a proxy for the Public Review draft. d Includes equipment, such as tractors and combines, as well as fuel consumption from trucks that are used off- road in agriculture. e Includes equipment, such as cranes, dumpers, and excavators, as well as fuel consumption from trucks that are used off-road in construction. f "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. Notes: In 2011, FHWA changed its methods for estimating vehicle miles traveled (VMT) and related data. These methodological changes included how vehicles are classified, moving from a system based on body-type to one that is based on wheelbase. These changes were first incorporated for the 1990 through 2010 Inventory and apply to the 2007 through 2016 time period. This resulted in large changes in VMT and fuel consumption data by vehicle class, thus leading to a shift in emissions among on-road vehicle classes. Totals may not sum due to independent rounding. 1 Table 3-14: N2O Emissions from Mobile Combustion (MMT CO2 Eq.) Fuel Type/Vehicle Type3 19911 2005 2012 2013 2014 2015 2016 Gasoline On-Roadb 37.5 33.5 19.1 17.2 15.4 14.0 12.7 Passenger Cars 24.1 17.5 13.1 11.8 10.5 9.7 8.9 Light-Duty Trucks 12.8 15.0 5.3 4.7 4.4 3.8 3.4 Medium- and Heavy-Duty Trucks and Buses 0.5 0.9 0.7 0.6 0.5 0.5 0.4 Motorcycles - + + + + + Diesel On-Roadb 0.2 0.3 0.4 0.4 0.4 0.4 0.4 Passenger Cars - - + + + + + Light-Duty Trucks - + + + + + Medium- and Heavy-Duty Trucks and Buses 0.2 0.3 0.4 0.4 0.4 0.4 0.4 Alternative Fuel On-Road + + + + + + + Non-Road 3.S 4.6 4.4 4.5 4.3 4.5 4.7 Ships and Boats 0.6 0.6 0.5 0.5 0.3 0.4 0.5 Rail0 0.3 0.3 0.3 0.3 0.3 0.3 0.3 Aircraft 1." 1.8 1.3 1.4 1.4 1.5 1.6 Agricultural Equipment"1 0.4 0.6 0.7 0.7 0.7 0.7 0.7 Construction/Mining Equipment6 0.5 0.8 0.9 0.9 0.9 0.9 0.9 Otherf 0.4 0.6 0.7 0.7 0.7 0.7 0.7 Total 41.5 38.4 23.8 22.0 20.2 18.8 17.8 Energy 3-27 ------- 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 + Does not exceed 0.05 MMT CO2 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 (FHWA 1996 through 2017). Table VM-1 data for 2016 has not been published yet, therefore 2016 mileage data is estimated using the 1.7 percent increase in FHWA Traffic Volume Trends from 2015 to 2016. These mileage estimates are combined with estimates of fuel shares by vehicle type from DOE's TEDB Annex Tables A. 1 through A.6 (DOE 1993 through 2016). TEDB data for 2015 and 2016 has not been published yet, therefore 2014 data is used as a proxy for the Public Review draft. c Rail emissions do not include emissions from electric powered locomotives. Class II, Class III, commuter, and intercity rail diesel consumption data for 2014 to 2016 are not available yet, therefore 2013 data is used as a proxy for the Public Review draft. d Includes equipment, such as tractors and combines, as well as fuel consumption from trucks that are used off-road in agriculture. e Includes equipment, such as cranes, dumpers, and excavators, as well as fuel consumption from trucks that are used off-road in construction. f "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: In 2011, FHWA changed its methods for estimating vehicle miles traveled (VMT) and related data. These methodological changes included how vehicles are classified, moving from a system based on body type to one that is based on wheelbase. These changes were first incorporated for the 1990 through 2010 Inventory and apply to the 2007 through 2016 time period. This resulted in large changes in VMT and fuel consumption data by vehicle class, thus leading to a shift in emissions among on-road vehicle classes. Totals may not sum due to independent rounding. C02 from Fossil Fuel Combustion Methodology 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) with some exceptions as discussed below.31 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, etc.), primary fuel type (e.g., coal, petroleum, gas), and secondary fuel category (e.g., motor gasoline, distillate fuel oil, etc.). 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 2017a). The EIA does not include territories in its national energy statistics, so fuel consumption data for territories were collected separately fromEIA's International Energy Statistics (EIA 2017b).32 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 31 The IPCC Tier 3B methodology is used for estimating emissions from commercial aircraft. 32 Fuel consumption by U.S. Territories (i.e., American Samoa, Guam, Puerto Rico, U.S. Virgin Islands, Wake Island, and other U.S. Pacific Islands) is included in this report and contributed total emissions of 41.4 MMT CO2 Eq. in 2016. 3-28 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2016 ------- 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 periodic basis (every four years). These consumption data sets help inform the annual surveys to arrive at the national total and sectoral breakdowns for that total.33 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).34 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 2016), Coffeyville (2012), U.S. Census Bureau (2001 through 2011), EIA (2017a, 2017c, 2017d), USAA (2008 through 2017), USGS (1991 through 2015a), (USGS 2016a), USGS (2014 through 2016a), USGS (2014 through 2016b), USGS (1995 through 2013), USGS (1995, 1998, 2000, 2001), USGS (2017), USGS (1991 through 2013), USGS (2016d), USGS (2015b), USGS (2014), USGS (1996 through 2013), USGS (1991 through 2015b), USGS (2015 and 2016), USGS (1991 through 2015c).35 3. Adjust for conversion offuels and exports of CO 2. Fossil fuel consumption estimates are adjusted downward to exclude fuels created from other fossil fuels and exports of CO2.36 Synthetic natural gas is created from industrial coal, and is currently included in EIA statistics for both coal and natural gas. Therefore, synthetic natural gas is subtracted from energy consumption statistics.37 Since October 2000, the Dakota Gasification Plant has been exporting CO2 to Canada by pipeline. Since this CO2 is not emitted to the atmosphere in the United States, the associated fossil fuel burned to create the exported CO2 is subtracted from fossil fuel consumption statistics. 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, additional data for ethanol and biodiesel were collected from EIA (2017a), data for synthetic natural gas were collected from EIA (2017d), and data for CO2 exports were collected from the Eastman Gasification Services Company (2011), Dakota Gasification Company (2006), Fitzpatrick (2002), Erickson (2003), EIA (2008) and DOE (2012). 4. Adjust Sectoral Allocation of Distillate Fuel Oil and Motor Gasoline. EPA had 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 was 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 2017), Benson 33 See IPCC Reference Approach for Estimating CO2 Emissions from Fossil Fuel Combustion in Annex 4 for a comparison of U.S. estimates using top-down and bottom-up approaches. 34 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. 35 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. 36 Energy statistics from EIA (2017a) are already adjusted downward to account for ethanol added to motor gasoline, biodiesel added to diesel fuel, and biogas in natural gas. 37 These adjustments are explained in greater detail in Annex 2.1. Energy 3-29 ------- 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 (2002 through 2004), DOE (1993 through 2016), EIA (2007), EIA (1991 through 2016), EPA (2017b), and FHWA (1996 through 2017).38 5. 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 C contained in the fuel for a period of time. As the emission pathways of C used for non-energy purposes are vastly different than fuel combustion (since the C in these fuels ends up in products instead of being combusted), these emissions are estimated separately in Section 3.2 - Carbon Emitted and Stored in Products from Non-Energy Uses of Fossil Fuels. Therefore, the amount of fuels used for non-energy purposes was subtracted from total fuel consumption. Data on non-fuel consumption was provided by EIA (2017a). 6. Subtract consumption of international bunker fuels. According to the 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 for emissions from consumption of all fossil fuels (i.e., estimation of consumption, and determination of C content).39 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 2017) supplied data on military jet fuel and marine fuel use. Commercial jet fuel use was obtained from FAA (2017); residual and distillate fuel use for civilian marine bunkers was obtained from DOC (1991 through 2017) for 1990 through 2001 and 2007 through 2014, and DHS (2008) for 2003 through 2006. Consumption of these fuels was subtracted from the corresponding fuels in the transportation end-use sector. Estimates of international bunker fuel emissions for the United States are discussed in detail in Section 3.10 - International Bunker Fuels. 7. Determine the total C content of fuels consumed. Total C was estimated by multiplying the amount of fuel consumed by the amount of C in each fuel. This total C estimate defines the maximum amount of C that could potentially be released to the atmosphere if all of the C in each fuel was converted to CO2. The C content coefficients used by the United States were obtained from EIA's Emissions of Greenhouse Gases in the United States 2008 (EIA 2009a), and an EPA analysis of C content coefficients developed for the GHGRP (EPA 2010). A discussion of the methodology used to develop the C content coefficients are presented in Annexes 2.1 and 2.2. 8. Estimate C02 Emissions. Total CO2 emissions are the product of the adjusted energy consumption (from the previous methodology steps 1 through 6), the C content of the fuels consumed, and the fraction of C 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). 9. 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 (2017a) and USAF (1998).40 • For on-road vehicles, annual estimates of combined motor gasoline and diesel fuel consumption by vehicle category were obtained fromFHWA (1996 through 2017); for each vehicle category, the 38 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 2017). 39 See International Bunker Fuels section in this chapter for a more detailed discussion. 40 For a more detailed description of the data sources used for the analysis of the transportation end use sector see the Mobile Combustion (excluding CO2) and International Bunker Fuels sections of the Energy chapter, Annex 3.2, and Annex 3.8, respectively. 3-30 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2016 ------- 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 percent gasoline, diesel, and other (e.g., CNG, LPG) fuel consumption are estimated using data from DOE (1993 through 2016).4142 • For non-road vehicles, activity data were obtained from AAR (2008 through 2017), APTA (2007 through 2016), APTA (2006)' BEA (2016), Benson (2002 through 2004), DOE (1993 through 2016), DLA Energy (2017), DOC (1991 through 2017), DOT (1991 through 2016), EIA (2009a), EIA (2017a), EIA (2017), EIA (1991 through 2016), EPA (2017b),43 and Gaffney (2007). • 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. Box 3-4: 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 lias provided an opportunity to better characterize the industrial sector's energy consumption and emissions in the United States, through a disaggregation of EIA's industrial sector fuel consumption data from select industries. For GHGRP 2010 through 2016 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 the 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.44 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 CRF tables that are submitted to the 41 Data from FHWA's Table VM-1 is used to estimate the share of fuel consumption between each on-road vehicle class. Since VM-1 data for 2016 has not been published yet, fuel consumption shares from 2015 are used as a proxy for the current Inventory. These fuel consumption estimates are combined with estimates of fuel shares by vehicle type from DOE's TEDB Annex Tables A.l through A.6 (DOE 1993 through 2016). TEDB data for 2016 has not been published yet, therefore 2015 data is used as a proxy. In 2011, FHWA changed its methods for estimating data in the VM-1 table. These methodological changes included how vehicles are classified, moving from a system based on body-type to one that is based on wheelbase. These changes were first incorporated for the 1990 through 2010 Inventory and apply to the 2007 through 2015 time period. This resulted in large changes in VMT and fuel consumption data by vehicle class, thus leading to a shift in emissions among on-road vehicle classes. 42 Transportation sector natural gas and LPG consumption are based on data from EIA (2017a). In previous Inventory years, data from DOE 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 current Inventory and apply to the 1990 to 2015 time period. 43 In 2014, EPA incorporated the NONROAD2008 model into MOVES2014. The current Inventory uses the NONROAD component of MOVES2014a for years 1999 through 2016. 44 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 . Energy 3-31 ------- 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 UNFCCC along with this report.45 The efforts in reconciling fuels focus on standard, common fuel types (e.g., natural gas, distillate fuel oil, etc.) where the fuels inEIA's national statistics aligned well with facility-level GHGRP data. For these reasons, the current information presented in the CRF tables 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-level data and national statistics. The current analysis includes the full time series presented in the CRF tables. 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 2016 time series in the CRF tables. 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. Box 3-5: Carbon Intensity of U.S. Energy Consumption The amount of C emitted from the combustion of fossil fuels is dependent upon the C content of the fuel and the fraction of that C that is oxidized. Fossil fuels vary in their average C 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.46 In general, the C content per unit of energy of fossil fuels is the highest for coal products, followed by petroleum, and then natural gas. The overall C 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 3-15 provides a time series of the C 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 C 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 3-15 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 C intensity, which is related to the large percentage of its energy derived from natural gas for heating. The C intensity of the commercial sector lias 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 C intensities over this period. The C intensity of the transportation sector was closely related to the C content of petroleum products (e.g., motor gasoline and jet fuel, both around 70 MMT CO2 Eq./EJ), which were the primary sources of energy. Lastly, the electric power sector had the highest C intensity due to its heavy reliance on coal for generating electricity. Table 3-15: Carbon Intensity from Direct Fossil Fuel Combustion by Sector (MMT CO2 Eq./QBtu) Sector 1990 2005 2012 2013 2014 2015 2016 Residential3 57.4 56.6 55.5 55.3 55.4 55.5 55.4 Commercial3 59.6 57.7 56.3 56.1 55.8 57.2 56.7 Industrial3 64.4 64.5 62.3 62.1 61.6 61.2 60.7 Transportation3 71.1 71.4 71.5 71.4 71.5 71.5 71.5 Electric Powerb 87.3 85.8 79.9 81.3 81.2 78.1 76.9 U.S. Territories0 73.0 73.5 72.2 71.9 72.3 72.3 72.3 All Sectors0 73.0 73.5 70.9 70.9 70.7 69.7 69.2 3 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. 45 See . 46 One exajoule (E.T) is equal to 1018 joules or 0.9478 QBtu. 3-32 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2016 ------- 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 For the time period of 1990 through about 2008. the C 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 C intensity lias 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 in 2016 was approximately 10.5 percent below levels in 1990 (see Figure 3-16). To differentiate these estimates from those of Table 3-15, the C intensity trend shown in Figure 3-16 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 2017). Figure 3-16: U.S. Energy Consumption and Energy-Related CO2 Emissions Per Capita and Per Dollar GDP 110 100 8 90 S 80 rH X -8 £ 70 60 50 CO:/Energy Consumption Energy Consumption/capita CO./capita Energy Consumption/GDP COj/GDP o-«-»fNro^*u^vor^oo^O'^r«jm^-mvor>»ooo>0'^-'fNro^ri/^vo G^O^ONO^G^C^O^O^O^O^OOOO COC OO O »—« 4 0^0>(^a>^O^OGNO^(^COOCOOCOOOOCOOOOO f>4 fNJ OJ Oi CM fN fN CM CM CM CM CM CM CM CM CM CM C intensity estimates were developed using nuclear and renewable energy data from EIA (2017a), EPA (2010), and fossil fuel consumption data as discussed above and presented in Annex 2.1. Uncertainty and Time-Series Consistency 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 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). Although 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 Energy 3-33 ------- 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 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 problems 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 C 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 3.10 - International Bunker Fuels). 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 120 input variables were modeled for CO2 from energy-related Fossil Fuel Combustion (including about 10 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.47 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.48 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).49 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 3-16. Fossil fuel combustion CO2 emissions in 2016 were estimated to be between 4,868.8 and 5,202.9 MMT CO2 Eq. at a 95 percent 47 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. 48 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. 49 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. 3-34 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2016 ------- 1 confidence level. This indicates a range of 2 percent below to 5 percent above the 2016 emission estimate of 4,976.7 2 MMT C02 Eq. 3 Table 3-16: Approach 2 Quantitative Uncertainty Estimates for CO2 Emissions from Energy- 4 Related Fossil Fuel Combustion by Fuel Type and Sector (MMT CO2 Eq. and Percent) 2016 Emission Estimate Uncertainty Range Relative to Emission Estimate3 Fuel/Sector (MMT CO2 Eq.) (MMT CO2 Eq.) (%) Lower Bound Upper Bound Lower Bound Upper Bound Coal" 1,306.6 1,261.2 1,430.4 -3% 9% Residential NE NE NE NE NE Commercial 2.3 2.2 2.7 -5% 15% Industrial 59.0 56.1 68.3 -5% 16% Transportation NE NE NE NE NE Electric Power 1,241.3 1,192.8 1,361.2 -4% 10% U.S. Territories 4.0 3.5 4.8 -12% 19% Natural Gasb 1,477.0 1,460.2 1,545.5 -1% 5% Residential 238.3 231.5 255.0 -3% 7% Commercial 170.3 165.5 182.3 -3% 7% Industrial 478.8 464.4 513.3 -3% 7% Transportation 40.6 39.5 43.5 -3% 7% Electric Power 545.9 530.1 574.0 -3% 5% U.S. Territories 3.0 2.6 3.5 -13% 17% Petroleumb 2,192.7 2,056.6 2,320.3 -6% 6% Residential 58.0 54.8 61.0 -5% 5% Commercial 55.3 52.1 58.1 -6% 5% Industrial 269.7 212.6 322.4 -21% 20% Transportation 1,754.2 1,639.4 1,866.5 -7% 6% Electric Power 21.2 20.0 23.2 -6% 9% U.S. Territories 34.3 31.7 38.2 -8% 11% Total (excluding Geothermal)1 " 4,976.3 4,868.4 5,202.4 -2% 5% Geothermal 0.4 NE NE NE NE Total (including Geothermal)b'c 4,976.7 4,868.8 5,202.9 -2% 5% NE (Not Estimated) a Range of emission estimates predicted by Monte Carlo Stochastic Simulation for a 95 percent confidence interval. b The low and high estimates for total emissions were calculated separately through simulations and, hence, the low and high emission estimates for the sub-source categories do not sum to total emissions. c Geothermal emissions added for reporting purposes, but an uncertainty analysis was not performed for CO2 emissions from geothermal production. 5 Methodological recalculations were applied to the entire time series to ensure time-series consistency from 1990 6 through 2016. Details on the emission trends through time are described in more detail in the Methodology section, 7 above. 8 QA/QC and Verification 9 A source-specific QA/QC plan for CO2 from fossil fuel combustion was developed and implemented consistent with 10 the 2006IPCC Guidelines and the Quality Assurance/Quality Control and Uncertainty Management Plan (QA/QC 11 Management Plan) referenced in this report and described further in Annex 8. This effort included a general (Tier 1) 12 analysis, as well as portions of a category-specific (Tier 2) analysis. The Tier 2 procedures that were implemented 13 involved checks specifically focusing on the activity data and methodology used for estimating CO2 emissions from 14 fossil fuel combustion in the United States. Emission totals for the different sectors and fuels were compared and Energy 3-35 ------- 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 trends were investigated to determine whether any corrective actions were needed. Minor corrective actions were taken. Recalculations Discussion The Energy Information Administration (EIA 2017a) updated energy consumption statistics across the time series relative to the previous Inventory. EIA revised ethylene consumption in the industrial sector for the years 1990 through 2015, which had a significant impact on emissions from Liquefied Petroleum Gas (LPG) across the time series (i.e., 1990 through 2015). EIA revised LPG consumption in the residential sector for the years 2010 through 2015, and in the commercial and transportation sectors for the years 2011 through 2015. EIA also revised 2014 and 2015 distillate fuel consumption in the residential, commercial, industrial, and transportation sectors, and 2015 natural gas consumption in the residential, commercial, transportation, and electric power sectors. Revisions to LPG and distillate fuel consumption resulted in an average annual increase of 14.0 MMT CO2 Eq. (0,6 percent) in CO2 emissions from petroleum. Revisions to natural gas consumption resulted in an average annual increase of less than 0.5 MMT CO2 Eq. (less than 0.05 percent) in CO2 emissions from natural gas. Overall, these changes resulted in an average annual increase of 14.0 MMT CO2 Eq. (0.3 percent) in CO2 emissions from fossil fuel combustion for the period 1990 through 2015, relative to the previous Inventory. In addition, changes were made to the historic allocation of gasoline to on-road and non-road applications. In 2016, the Federal Highway Administration (FHWA) changed its methods for estimating the share of gasoline used in on- road and non-road applications. Among other updates, FHWA included lawn and garden equipment as well as off- road recreational equipment in its estimates of non-road gasoline consumption for the first time. This change created a time-series inconsistency between the data reported for years 2015 and 2016 and previous years. To create a more consistent time series of motor gasoline consumption and emissions data for the current Inventory, the historical time series was modified. Specifically, the lawn, garden, and recreational vehicle gasoline consumption from EPA's NONROAD model is subtracted from the highway motor gasoline consumption from FHWA Table MF-21 when determining the total highway motor gasoline consumption foryears 1990 through 2014. Planned Improvements To reduce uncertainty of CO2 from fossil fuel combustion estimates for U.S. Territories, efforts will be made to improve the quality of the U.S. Territories data, including through work with EIA and other agencies. This improvement is part of an ongoing analysis and efforts to continually improve the CO2 from fossil fuel combustion estimates. In addition, further expert elicitation may be conducted to better quantify the total uncertainty associated with emissions from this source. The availability of facility-level combustion emissions through EPA's GHGRP will continue to be examined 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, though 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 UNFCCC reporting guidelines, some facility-level fuel combustion emissions reported under the GHGRP may also include industrial process emissions.50 In line with 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, efforts will be taken to incorporate updated industrial fuel consumption data from 50 See . 3-36 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2016 ------- 1 EIA's Manufacturing Energy Consumption Survey (MECS), with updated data for 2014. Additional work will look 2 at CO2 emissions from biomass to ensure they are separated in the facility-level reported data, and maintaining 3 consistency with national energy statistics provided by EIA. In implementing improvements and integration of data 4 from EPA's GHGRP, the latest guidance from the IPCC on the use of facility-level data in national inventories will 5 continue to be relied upon.51 6 An ongoing planned improvement is to develop improved estimates of domestic waterborne fuel consumption. The 7 Inventory estimates for residual and distillate fuel used by ships and boats is based in part on data on bunker fuel use 8 from the U.S. Department of Commerce. Domestic fuel consumption is estimated by subtracting fuel sold for 9 international use from the total sold in the United States. It may be possible to more accurately estimate domestic 10 fuel use and emissions by using detailed data on marine ship activity. The feasibility of using domestic marine 11 activity data to improve the estimates will continue to be investigated. 12 EPA received a comment from FHWA that the trend of decreasing electricity use in the transportation sector does 13 not align with increased sales of electric and plug-in hybrid vehicles. Electricity data is allocated between economic 14 sectors based on electricity sales data provided by the industry through EIA reports. The data for electricity used in 15 transportation only includes electricity used for railroads and railways. Electricity used to charge electric vehicles 16 would fall under other sectors like residential and commercial use associated with home and public charging 17 stations. As a planned improvement, EPA will look into the possibility of breaking out electricity used to charge 18 electric vehicles and report that electricity use under the transportation sector. 19 EPA will evaluate and potentially update methods for allocating motor gasoline consumption to the transportation, 20 industrial, and commercial sectors. In 2016, FHWA changed its methods for estimating the share of gasoline used in 21 on-road and non-road applications, creating a time-series inconsistency in the current Inventory between 2015 and 22 previous years.52 EPA will continue to explore approaches to address this inconsistency, including using MOVES 23 on-road fuel consumption output to define the percentage of the FHWA consumption totals (from MF -21) that are 24 attributable to "transportation", and applying that percentage to the EIA total. This would define gasoline 25 consumption from "transportation," such that the remainder would be defined as consumption by the industrial and 26 commercial sectors. 27 CH4 and N20 from Stationary Combustion 28 Methodology 29 Methane and N20 emissions from stationary combustion were estimated by multiplying fossil fuel and wood 30 consumption data by emission factors (by sector and fuel type for industrial, residential, commercial, and U.S. 31 Territories; and by fuel and technology type for the electric power sector). The electric power sector utilizes a Tier 2 32 methodology, whereas all other sectors utilize a Tier 1 methodology. The activity data and emission factors used are 33 described in the following subsections. 34 Industrial, Residential, Commercial, and U.S. Territories 35 National coal, natural gas, fuel oil, and wood consumption data were grouped by sector: industrial, commercial, 36 residential, and U.S. Territories. For the CH4 and N20 estimates, consumption data for each fuel were obtained from 37 EIA's Monthly Energy Review (EIA 2017a). Because the United States does not include territories in its national 38 energy statistics, fuel consumption data for territories were provided separately by EIA's International Energy 51 See . 52 The previous and new FHWA methodologies for estimating non-road gasoline are described in Off-Highway and Public-Use Gasoline Consumption Estimation Models Used in the Federal Highway Administration, Publication Number FHWA-PL-17-012. Energy 3-37 ------- 1 Statistics (EIA 2017b).53 Fuel consumption for the industrial sector was adjusted to subtract out construction and 2 agricultural use, which is reported under mobile sources.54 Construction and agricultural fuel use was obtained from 3 EPA (2017b) and FHWA (1996 through 2016). Estimates for wood biomass consumption for fuel combustion do 4 not include wood wastes, liquors, municipal solid waste, tires, etc., that are reported as biomass by EIA. Tier 1 5 default emission factors for these three end-use sectors were provided by the 2006IPCC Guidelines for National 6 Greenhouse Gas Inventories (IPCC 2006). U.S. Territories' emission factors were estimated using the U.S. emission 7 factors for the primary sector in which each fuel was combusted. 8 Electric Power Sector 9 The electric power sector uses a Tier 2 emission estimation methodology as fuel consumption for the electric power 10 sector by control-technology type was obtained from EPA's Acid Rain Program Dataset (EPA 2017a). These 11 combustion technology- and fuel- use data were available by facility from 1996 to 2016. The Tier 2 emission factors 12 used are based in part on emission factors published by EPA, and EPA's Compilation of Air Pollutant Emission 13 Factors, AP-42 (EPA 1997) for combined cycle natural gas units.55 14 Since there was a difference between the EPA (2017a) and EIA (2017a) total fuel consumption estimates, the 15 remaining consumption from EIA (2017a) was apportioned to each combustion technology type and fuel 16 combination using a ratio of fuel consumption by technology type from 1996 to 2016. 17 Fuel consumption estimates were not available from 1990 to 1995 in the EPA (2017a) dataset, and as a result, 18 consumption was calculated using total electric power production from EIA (2017a) and the ratio of combustion 19 technology and fuel types from EPA (2017a). The consumption estimates from 1990 to 1995 were estimated by 20 applying the 1996 consumption ratio by combustion technology type to the total EIA consumption for each year 21 from 1990 to 1995. Emissions were estimated by multiplying fossil fuel and wood consumption by technology - and 22 fuel-specific Tier 2 country specific emission factors. 23 Lastly, there were significant differences between wood biomass consumption in the electric power sector between 24 the EPA (2017a) and EIA (2017a) datasets. The higher wood biomass consumption from EIA (2017a) in the electric 25 power sector was distributed to the residential, commercial, and industrial sectors according to their percent share of 26 wood biomass energy consumption calculated from EIA (2017a). 27 More detailed information on the methodology for calculating emissions from stationary combustion, including 28 emission factors and activity data, is provided in Annex 3.1. 29 Uncertainty and Time-Series Consistency 30 Methane emission estimates from stationary sources exhibit high uncertainty, primarily due to difficulties in 31 calculating emissions from wood combustion (i.e., fireplaces and wood stoves). The estimates of CH4 and N20 32 emissions presented are based on broad indicators of emissions (i.e., fuel use multiplied by an aggregate emission 33 factor for different sectors), rather than specific emission processes (i.e., by combustion technology and type of 34 emission control). 35 An uncertainty analysis was performed by primary fuel type for each end-use sector, using the IPCC-recommended 36 Approach 2 uncertainty estimation methodology, Monte Carlo Stochastic Simulation technique, with @RISK 37 software. 38 The uncertainty estimation model for this source category was developed by integrating the CH4 and N20 stationary 39 source inventory estimation models with the model for CO2 from fossil fuel combustion to realistically characterize 53 U.S. Territories data also include combustion from mobile activities because data to allocate territories' energy use were unavailable. For this reason, CH4 and N2O emissions from combustion by U.S. Territories are only included in the stationary combustion totals. 54 Though emissions from construction and farm use occur due to both stationary and mobile sources, detailed data was not available to determine the magnitude from each. Currently, these emissions are assumed to be predominantly from mobile sources. 55 Several of the U.S. Tier 2 emission factors were used in IPCC 2006 as Tier 1 emission factors. 3-38 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2016 ------- 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 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 N20 emission factors, based on the SAIC/EIA (2001) report.56 For these variables, the uncertainty ranges were assigned to the input variables based on the data reported in SAIC/EIA (2001).57 However, the CH4 emission factors differ from those used by EIA. These factors and uncertainty ranges are based on IPCC default uncertainty estimates (IPCC 2006). The results of the Approach 2 quantitative uncertainty analysis are summarized in Table 3-17. Stationary combustion CH4 emissions in 2016 (including biomass) were estimated to be between 5.1 and 15.5 MMT CO2 Eq. at a 95 percent confidence level. This indicates a range of 29 percent below to 115 percent above the 2016 emission estimate of 7.2 MMT CO2 Eq.58 Stationary combustion N20 emissions in 2016 (including biomass) were estimated to be between 14.2 and 27.5 MMT CO2 Eq. at a 95 percent confidence level. This indicates a range of 23 percent below to 50 percent above the 2016 emission estimate of 18.4 MMT CO2 Eq. Table 3-17: Approach 2 Quantitative Uncertainty Estimates for ChU and N2O Emissions from Energy-Related Stationary Combustion, Including Biomass (MMT CO2 Eq. and Percent) Source Gas 2016 Emission Estimate Uncertainty Range Relative to Emission Estimate3 (MMT CO2 Eq.) (MMT CO2 Eq.) (%) Lower Upper Lower Upper Bound Bound Bound Bound Stationary Combustion CH4 7.2 5.1 15.5 -29% +115% Stationary Combustion N2O 18.4 14.2 27.5 -23% +50% 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 N20 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 CH4 and N20 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. Methodological recalculations were applied to the entire time series to ensure time-series consistency from 1990 through 2016 as discussed below. Details on the emission trends through time are described in more detail in the Methodology section, above. For more information on the general QA/QC process applied to this source category, consistent with Volume 1, Chapter 6 of the 2006 IPCC Guidelines, see QA/QC and Verification Procedures section in the introduction of the IPPU Chapter. 56 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. 57 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. 58 The low emission estimates reported in this section have been rounded down to the nearest integer values and the high emission estimates have been rounded up to the nearest integer values. Energy 3-39 ------- 1 QA/QC and Verification 2 A source-specific QA/QC plan for stationary combustion was developed and implemented consistent with the 2006 3 IPCC Guidelines and the QA/QC Management Plan referenced in this report and described further in Annex 8. This 4 effort included a general (Tier 1) analysis, as well as portions of a category-specific (Tier 2) analysis. The Tier 2 5 procedures that were implemented involved checks specifically focusing on the activity data and emission factor 6 sources and methodology used for estimating CH4, N20, and the indirect greenhouse gases from stationary 7 combustion in the United States. Emission totals for the different sectors and fuels were compared and trends were 8 investigated. 9 Recalculations Discussion 10 Methane and N20 emissions from stationary sources (excluding CO2) across the entire time series were revised due 11 to revised data from EIA (2017a), EIA (2017b), and EPA (2017a) relative to the previous Inventory. Methane and 12 N20 emission factors for combined cycle natural gas units were updated to be consistent with EPA's Compilation of 13 Air Pollutant Emission Factors, AP-42 (EPA 1997). In addition, the GWPs for CH4 and N2O for the Acid Rain 14 Program Dataset (EPA 2017a) were updated to be consistent with the IPCC Fourth Assessment Report (AIM) 15 values. The historical data changes resulted in an average annual increase 0.4 MMT CO2 Eq. (5.2 percent) in CH4 16 emissions, and an average annual decrease 2.3 MMT CO2 Eq. (12 percent) in N20 emissions from stationary 17 combustion for the 1990 through 2015 period. 18 Planned Improvements 19 Several items are being evaluated to improve the CH4 and N20 emission estimates from stationary combustion and 20 to reduce uncertainty for U.S. Territories. Efforts will be taken to work with EIA and other agencies to improve the 21 quality of the U.S. Territories data. Because these data are not broken out by stationary and mobile uses, further 22 research will be aimed at trying to allocate consumption appropriately. In addition, the uncertainty of bio mass 23 emissions will be further investigated since it was expected that the exclusion of biomass from the estimates would 24 reduce the uncertainty; and in actuality the exclusion of biomass increases the uncertainty. These improvements are 25 not all-inclusive, but are part of an ongoing analysis and efforts to continually improve these stationary combustion 26 estimates from U.S. Territories. 27 Fuel use was adjusted for the industrial sector to subtract out construction and agricultural use, which is reported 28 under mobile sources. Mobile source CH4 and N20 also include emissions from sources that may be captured as part 29 of the commercial sector. Future research will look into the need to adjust commercial sector fuel consumption to 30 account for sources included elsewhere. 31 CH4 and N20 from Mobile Combustion 32 Methodology 33 Estimates of CH4 and N20 emissions from mobile combustion were calculated by multiplying emission factors by 34 measures of activity for each fuel and vehicle type (e.g., light-duty gasoline trucks). Activity data included vehicle 35 miles traveled (VMT) for on-road vehicles and fuel consumption for non-road mobile sources. The activity data and 36 emission factors used are described in the subsections that follow. A complete discussion of the methodology used to 37 estimate CH4 and N20 emissions from mobile combustion and the emission factors used in the calculations is provided 38 in Annex 3.2. 3-40 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2016 ------- 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 On-Road Vehicles Estimates of CH4 and N20 emissions from gasoline and diesel on-road vehicles are based on VMT and emission factors 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 by vehicle and fuel type.59 Emissions factors for N20 from newer on-road gasoline vehicles were calculated based upon a regression analysis done by EPA (Browning 2017). Methane emission factors were calculated based on the ratio of NMOG emission standards for newer vehicles. Older gasoline vehicles on-road emissions factors were developed by ICF (2004). These factors were derived from EPA, California Air Resources Board (CARB) and Environment Canada laboratory test results of different vehicle and control technology types. The EPA, CARB and Environment Canada tests were designed following the Federal Test Procedure (FTP), which 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 segment 2 were used to define running emissions, and subtracted from the total FTP emissions to determine start emissions. These were then recombined based upon the ratio of start to running emissions for each vehicle class from MOBILE6.2, an EPA emission factor model that predicts gram per mile emissions of CO2, CO, HC, NOx, and PM from vehicles under various conditions, to approximate average driving characteristics.60 Diesel on-road vehicle emission factors were developed by ICF (2006b). CH4 and N20 emission factors for AFVs were developed based on the 2016 GREET model. For light-duty trucks, EPA used a curve fit of 1999 through 2011 travel fractions for LDT1 and LDT2 (MOVES Source Type 31 for LDT1 and MOVES Source Type 32 for LDT2). 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 operation only (tank-to- wheels); well-to-tank emissions are calculated elsewhere in the Inventory. Annual VMT data for 1990 through 2015 were obtained from the Federal Highway Administration's (FHWA) Highway Performance Monitoring System database as reported in Highway Statistics (FHWA 1996 through 2016).61 VMT data in the VM-1 table for 2016 has not been published yet; therefore 2016 mileage data is estimated using the 1.7 percent increase in FHWA Traffic Volume Trends from 2015 to 2016. VMT estimates were then allocated from FHWA's vehicle categories to fuel-specific vehicle categories using the calculated shares of vehicle fuel use for each vehicle category by fuel type reported in DOE (1993 through 2017) and information on total motor vehicle fuel consumption by fuel type from FHWA (1996 through 2017). VMT for AFVs were estimated based on Browning (2017). The age distributions of the U.S. vehicle fleet were obtained from EPA (2017b, 2000), and the average annual age-specific vehicle mileage accumulation of U.S. vehicles were obtained from EPA (2017b). 59 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. 60 Additional information regarding the MOBILE model can be found online at . 61 The source of VMT is FHWA Highway Statistics Table VM-1. In 2011, FHWA changed its methods for estimating data in the VM-1 table. These methodological changes included how vehicles are classified, moving from a system based on body-type to one that is based on wheelbase. These changes were first incorporated for the 1990 through 2010 Inventory and apply to the 2007 through 2016 time period. This resulted in large changes in VMT by vehicle class, thus leading to a shift in emissions among on-road vehicle classes. For example, the category "Passenger Cars" has been replaced by "Light-duty Vehicles-Short Wheelbase" and "Other 2 axle-4 Tire Vehicles" has been replaced by "Light-duty Vehicles, Long Wheelbase." This change in vehicle classification has moved some smaller trucks and sport utility vehicles from the light truck category to the passenger vehicle category in the current Inventory. These changes are reflected in a large drop in light-truck emissions between 2006 and 2007. Energy 3-41 ------- 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 Control technology and standards data for on-road vehicles were obtained from EPA's Office of Transportation and Air Quality (EPA 2007a, 2007b, 2000, 1998, and 1997) and Browning (2005). These technologies and standards are defined in Annex 3.2, and were compiled from EPA (1994a, 1994b, 1998, 1999a) and IPCC (2006). Non-Road Mobile Sources To estimate 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 N20 and CH4 per kilogram of fuel consumed).62 Activity data were obtained from AAR (2008 through 2016), APTA (2007 through 2016), APTA (2006), BEA (1991 through 2015), Benson (2002 through 2004), DHS (2008), DLA Energy (2015), DOC (1991 through 2015), DOE (1993 through 2016), DOT (1991 through 2017), EIA (2002, 2007, 2016a), EIA (2007 through 2016), EIA (1991 through 2017), EPA (2017b), Esser (2003 through 2004), FAA (2017), FHWA (1996 through 20 1 7),63 Gaffney (2007), and Whorton (2006 through 2014). Emission factors for non-road modes were taken from IPCC (2006) and Browning (2017). Uncertainty and Time-Series Consistency-TO BE UPDATED FOR FINAL INVENTORY REPORT 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 VvRISK software. The uncertainty analysis was performed on 2015 estimates of CH4 and N20 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 and (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 since 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 3.9 - Uncertainty Analysis of Emission Estimates. However, a much higher level of uncertainty is associated with CH4 and N20 emission factors due to limited emission test data, and because, unlike CO: emissions, the emission pathways of CH4 and N20 are highly complex. Mobile combustion CH4 emissions from all mobile sources in 2015 were estimated to be between 1.6 and 2.5 MMT C02 Eq. at a 95 percent confidence level. This indicates a range of 18 percent below to 27 percent above the corresponding 2015 emission estimate of 2.0 MMT C02 Eq. Also at a 95 percent confidence level, mobile combustion N20 emissions from mobile sources in 2015 were estimated to be between 13.2 and 17.9 MMT C02 Eq., indicating a range of 13 percent below to 19 percent above the corresponding 2015 emission estimate of 15.1 MMT C02 Eq. Table 3-18: Approach 2 Quantitative Uncertainty Estimates for ChU and N2O Emissions from Mobile Sources (MMT CO2 Eq. and Percent) Source Gas 2015 Emission Estimate3 Uncertainty Range Relative to Emission Estimate3 62 Hie consumption of international bunker fuels is not included in these activity data, but is estimated separately under the International Bunker Fuels source category. 63 This Inventory uses FHWA's Agriculture, Construction, and Commercial/Industrial MF-24 fuel volumes along with the MOVES NONROAD model gasoline volumes to estimate non-road mobile source CH4 and N2O emissions for these categories. For agriculture, the MF-24 gasoline volume is used directly because it includes both off-road trucks and equipment. For construction and commercial/industrial gasoline estimates, the 2014 and older MF-24 volumes represented off-road trucks only; therefore, the MOVES NONROAD gasoline volumes for construction and commercial/industrial 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 NONROAD equipment gasoline volumes in the construction and commercial/industrial categories. 3-42 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2016 ------- (MMT CO2 Eq.) (MMT CO2 Eq.) (%) Lower Upper Lower Upper Bound Bound Bound Bound Mobile Sources CH4 2.0 1.6 2.5 -18% +27% Mobile Sources N2O 15.1 13.2 17.9 -13% +19% a Range of emission estimates predicted by Monte Carlo Stochastic Simulation for a 95 percent confidence interval. 1 This uncertainty analysis is a continuation of a multi-year process for developing quantitative uncertainty estimates 2 for this source category using the IPCC Approach 2 uncertainty analysis. As a result, as new information becomes 3 available, uncertainty characterization of input variables may be improved and revised. For additional information 4 regarding uncertainty in emission estimates for CH4 and N20 please refer to the Uncertainty Annex. 5 QA/QC and Verification 6 A source-specific Quality Assurance/Quality Control plan for mobile combustion was developed and implemented. 7 This plan is based on the IPCC-recommended QA/QC Plan. The specific plan used for mobile combustion was 8 updated prior to collection and analysis of this current year of data. This effort included a general (Tier 1) analysis, 9 as well as portions of a category-specific (Tier 2) analysis. The Tier 2 procedures focused on the emission factor and 10 activity data sources, as well as the methodology used for estimating emissions. These procedures included a 11 qualitative assessment of the emission estimates to determine whether they appear consistent with the most recent 12 activity data and emission factors available. A comparison of historical emissions between the current Inventory and 13 the previous Inventory was also conducted to ensure that the changes in estimates were consistent with the changes 14 in activity data and emission factors. 15 Recalculations Discussion 16 Updates were made to the on-road, non-road and alternative fuel CH4 and N20 emissions calculations this year 17 resulting in both increases and decreases to different source categories. Decreases in on-road gasoline emissions 18 were offset by large increases in alternative fuel and non-road emissions. The collective result of all of these changes 19 was a net increase in CH4 and N20 emissions from mobile combustion relative to the previous Inventory. CH4 20 emissions increased by 52.7 percent. N20 emissions increased by 24.5 percent. Each of these changes is described 21 below. 22 New emissions factors for N20 emissions were developed for on-road vehicles based on an EPA regression analysis 23 of the relationship between NOx and N20. New CH4 emission factors were calculated based on the ratio of NMOG 24 emission standards. These new emission factors allowed the inclusion of additional emissions standards, including 25 Federal Tier 3 emission standards and two levels of California emission standards (LEV II and LEV III) to the 26 control technology breakouts. 27 In addition new non-road emissions factors were developed. Previously emission factors were taken from the 1996 28 IPCC Guidelines and represented the IPCC Tier 1 factors. This year new emission factors were calculated using the 29 updated 2006 IPCC Tier 3 guidance and EPA's MOVES2014a model. CH4 emission factors were calculated directly 30 from MOVES. N20 emission factors were calculated using NONROAD activity and emission factors by fuel type 31 from the European Enviromnent Agency. Gasoline engines were broken out by 2- and 4-stroke engine types. 32 Equipment using liquefied petroleum gas (LPG) and compressed natural gas (CNG) were included. 33 New emission factors for alternative fuel vehicles were estimated using GREET 2016. The updated emission factors 34 have been generated for CH4 and N20. For light-duty trucks, EPA used a curve fit of 1999 through 2011 travel 35 fractions for LDT1 and LDT2 (MOVES Source Type 31 for LDT1 and MOVES Source Type 32 for LDT2). For 36 medium duty vehicles, EPA used emission factors for Light Heavy-Duty Vocational Trucks. For heavy-duty 37 vehicles, EPA used emission factors for Long Haul Combination Trucks. For Buses, EPA used emission factors for 38 Transit Buses. The emissions factors developed represent vehicle operation only (tank-to-wheels). 39 In addition changes were made to the historic allocation of gasoline to on-road and non-road applications. In 2016, 40 the Federal Highway Administration (FHWA) changed its methods for estimating the share of gasoline used in on- 41 road and non-road applications. Among other updates, FHWA included lawn and garden equipment as well as off- Energy 3-43 ------- 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 road recreational equipment in its estimates of non-road gasoline consumption for the first time. This change created a time-series inconsistency between the data reported for years 2015 and 2016 and previous years. To create a more consistent time series of motor gasoline consumption and emissions data for the current Inventory, the historical time series was modified. Specifically, the lawn, garden, and recreational vehicle gasoline consumption from EPA's NONROAD model is subtracted from the highway motor gasoline consumption from FHWA Table MF-21 when determining the total highway motor gasoline consumption for years 1990 through 2014. Methodological recalculations were applied to the entire time series to ensure time-series consistency from 1990 through 2016 with one recent notable exception. An update by FHWA to the method for estimating on-road VMT created an inconsistency in on-road CH4 and N20 for the time periods 1990 to 2006 and 2007 to 2016. Details on the emission trends and methodological inconsistencies through time are described in the Methodology section, above. Planned Improvements While the data used for this report represent the most accurate information available, several areas have been identified that could potentially be improved in the near term given available resources. • Evaluate and potentially update EPA's method for estimating motor gasoline consumption for non-road mobile sources to improve accuracy and create a more consistent time series. As discussed in the Methodology section above and in Annex 3.2, CH4 and N20 estimates for gasoline-powered non-road sources in this Inventory are based on a variety of inputs, including FHWA Highway Statistics Table MF- 24. In 2016, FHWA changed its methods for estimating the share of gasoline used in on-road and non-road applications.64 These method changes created a time-series inconsistency in the current Inventory between 2015 and previous years in CH4 and N20 estimates for agricultural, construction, commercial, and industrial non-road mobile sources. In the current Inventory EPA has implemented one approach to address this inconsistency. EPA will test other approaches including using MOVES on-road fuel consumption output to define the percentage of the FHWA consumption totals (from MF-21) that are attributable to "transportation." This percentage would then be applied to the EIA total, thereby defining gasoline consumption from "transportation," such that the remainder would be defined as consumption by the industrial and commercial sectors. • Explore updates to on-road diesel emissions factors for CH4 and N20 to incorporate diesel after treatment technology for light-duty vehicles. • 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. It may be possible to more accurately estimate domestic fuel use and emissions by using detailed data on marine ship activity. The feasibility of using domestic marine activity data to improve the estimates continues to be investigated. Additionally, the feasibility of including data from a broader range of domestic and international sources for domestic bunker fuels, including data from studies such as the Third 1MO GHG Study 2014, continues to be explored. 64 The previous and new FHWA methodologies for estimating non-road gasoline are described in Off-Highway and Public-Use Gasoline Consumption Estimation Models Used in the Federal Highway Administration, Publication Number FHWA-PL-17-012. 3-44 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2016 ------- 1 3.2 Carbon Emitted from Non-Energy Uses of 2 Fossil Fuels (CRF Source Category 1A) 3 In addition to being combusted for energy, fossil fuels are also consumed for non-energy uses (NEU) in the United 4 States. The fuels used for these purposes are diverse, including natural gas, liquefied petroleum gases (LPG), asphalt 5 (a viscous liquid mixture of heavy crude oil distillates), petroleum coke (manufactured from heavy oil), and coal 6 (metallurgical) coke (manufactured from coking coal). The non-energy applications of these fuels are equally 7 diverse, including feedstocks for the manufacture of plastics, rubber, synthetic fibers and other materials; reducing 8 agents for the production of various metals and inorganic products; and non-energy products such as lubricants, 9 waxes, and asphalt (IPCC 2006). Emissions from non-energy uses of fossil fuels are reported in the Energy sector, 10 as opposed to the IPPU sector, to reflect national circumstances in its choice of methodology and to increase 11 transparency of this source category's unique country-specific data sources and methodology (see Box 3-6). 12 Carbon dioxide emissions arise from non-energy uses via several pathways. Emissions may occur during the 13 manufacture of a product, as is the case in producing plastics or rubber from fuel-derived feedstocks. Additionally, 14 emissions may occur during the product's lifetime, such as during solvent use. Overall, throughout the time series 15 and across all uses, about 61 percent of the total C consumed for non-energy purposes was stored in products, and 16 not released to the atmosphere; the remaining 39 percent was emitted. 17 There are several areas in which non-energy uses of fossil fuels are closely related to other parts of this Inventory. 18 For example, some of the NEU products release CO2 at the end of their commercial life when they are combusted 19 after disposal; these emissions are reported separately within the Energy chapter in the Incineration of Waste source 20 category. In addition, there is some overlap between fossil fuels consumed for non-energy uses and the fossil- 21 derived CO2 emissions accounted for in the Industrial Processes and Product Use chapter, especially for fuels used 22 as reducing agents. To avoid double-counting, the "raw" non-energy fuel consumption data reported by EIA are 23 modified to account for these overlaps. There are also net exports of petrochemicals that are not completely 24 accounted for in the EIA data, and the Inventory calculations adjust for the effect of net exports on the mass of C in 25 non-energy applications. 26 As shown in Table 3-19, fossil fuel emissions in 2016 from the non-energy uses of fossil fuels were 121.0 MMT 27 CO2 Eq., which constituted approximately 2 percent of overall fossil fuel emissions. In 2016, the consumption of 28 fuels for non-energy uses (after the adjustments described above) was 4,844.4 TBtu (see Table 3-20). A portion of 29 the C in the 4,844.4 TBtu of fuels was stored (207.7 MMT CO2 Eq.), while the remaining portion was emitted 30 (121.0 MMT C02Eq.). 31 Table 3-19: CO2 Emissions from Non-Energy Use Fossil Fuel Consumption (MMT CO2 Eq. and 32 Percent) Year 1990 2005 2012 2013 2014 2015 2016 Potential Emissions 312.1 377.5 312.6 329.3 323.8 339.6 328.7 C Stored 192.5 235.9 199.4 196.2 196.0 204.5 207.7 Emissions as a % of Potential 38% 38% 36% 40% 39% 40% 37% Emissions 119.6 141.7 113.3 133.2 127.8 135.1 121.0 33 Methodology 34 The first step in estimating C stored in products was to determine the aggregate quantity of fossil fuels consumed for 35 non-energy uses. The C content of these feedstock fuels is equivalent to potential emissions, or the product of 36 consumption and the fuel-specific C content values. Both the non-energy fuel consumption and C content data were 3 7 supplied by the EIA (2013, 2016) (see Annex 2.1). Consumption of natural gas, LPG, pentanes plus, naphthas, other 38 oils, and special naphtha were adjusted to subtract out net exports of these products that are not reflected in the raw 39 data from EIA. Consumption values for industrial coking coal, petroleum coke, other oils, and natural gas in Table 40 3-20 and Table 3-21 have been adjusted to subtract non-energy uses that are included in the source categories of the Energy 3-45 ------- 1 Industrial Processes and Product Use chapter.65-66 Consumption values were also adjusted to subtract net exports of 2 intermediary chemicals. 3 For the remaining non-energy uses, the quantity of C stored was estimated by multiplying the potential emissions by 4 a storage factor. 5 • For several fuel types—petrochemical feedstocks (including natural gas for non-fertilizer uses, LPG, 6 pentanes plus, naphthas, other oils, still gas, special naphtha, and industrial other coal), asphalt and road oil, 7 lubricants, and waxes—U.S. data on C stocks and flows were used to develop C storage factors, calculated 8 as the ratio of (a) the C stored by the fuel's non-energy products to (b) the total C content of the fuel 9 consumed. A lifecycle approach was used in the development of these factors in order to account for losses 10 in the production process and during use. Because losses associated with municipal solid waste 11 management are handled separately in the Energy sector under the Incineration of Waste source category, 12 the storage factors do not account for losses at the disposal end of the life cycle. 13 • For industrial coking coal and distillate fuel oil, storage factors were taken from IPCC (2006), which in turn 14 draws from Marland and Rotty (1984). 15 • For the remaining fuel types (petroleum coke, miscellaneous products, and other petroleum), IPCC does not 16 provide guidance on storage factors, and assumptions were made based on the potential fate of C in the 17 respective NEU products. 18 Table 3-20: Adjusted Consumption of Fossil Fuels for Non-Energy Uses (TBtu) Year 1990 2005 2012 2013 2014 2015 2016 Industry 4,215.!S 5,110.7 4,373.3 4,627.5 4,591.4 4,762.0 4,626.6 Industrial Coking Coal 0.0 80.4 132.5 119.3 48.8 121.8 88.6 Industrial Other Coal 8.2 11.9 10.3 10.3 10.3 10.3 10.3 Natural Gas to Chemical Plants 281.6 260.9 292.7 297.1 305.1 302.2 289.5 Asphalt & Road Oil 1,170.2 1,323.2 826.7 783.3 792.6 831.7 853.4 LPG 1,120.5 1,610.0 1,883.4 2,069.2 2,103.4 2,160.0 2,117.6 Lubricants 186.:- 160.2 130.5 138.1 144.0 156.8 148.9 Pentanes Plus 117.6 95.5 40.3 45.4 43.5 78.4 53.0 Naphtha (<401 °F) 326.} 679.5 432.2 498.8 435.2 417.8 396.6 Other Oil (>401 °F) 662.1 499.4 267.4 209.1 236.2 216.8 203.8 Still Gas 36." 67.7 160.6 166.7 164.5 162.2 166.1 Petroleum Coke 27.2 105.2 0.0 0.0 0.0 0.0 0.0 Special Naphtha 100.9 60.9 14.1 96.6 104.4 97.0 88.7 Distillate Fuel Oil 7.0 11.7 5.8 5.8 5.8 5.8 5.8 Waxes 33.3 31.4 15.3 16.5 14.8 12.4 12.9 Miscellaneous Products 137.8 112.8 161.6 171.2 182.7 188.9 191.3 Transportation 176.0 151.3 123.2 130.4 136.0 148.1 140.6 Lubricants 176.0 151.3 123.2 130.4 136.0 148.1 140.6 U.S. Territories 85/. 123.2 72.0 82.4 77.3 77.3 77.3 Lubricants 0." 4.6 1.0 1.0 1.0 1.0 1.0 Other Petroleum (Misc. Prod.) 84.9 118.6 71.0 81.4 76.2 76.2 76.2 Total 4,477.4 5,385.2 4,568.5 4,840.3 4,804.7 4,987.4 4,844.5 65 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. 66 Some degree of double counting may occur between these estimates of non-energy use of fuels and process emissions from petrochemical production presented in the Industrial Processes and Produce Use sector. Data integration is not feasible at this time as feedstock data from EIA used to estimate non-energy uses of fuels are aggregated by fuel type, rather than disaggregated by both fuel type and particular industries (e.g., petrochemical production) as currently collected through EPA's GHGRP and used for the petrochemical production category. 3-46 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2016 ------- 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 Table 3-21: 2016 Adjusted Non-Energy Use Fossil Fuel Consumption, Storage, and Emissions Adjusted Carbon Non-Energy Content Potential Storage Carbon Carbon Carbon Use3 Coefficient (MMT Carbon Factor Stored Emissions Emissions (MMT Sector/Fuel Type (TBtu) C/QBtu) (MMT C) (MMT C) (MMT C) CO2 Eq.) Industry 4,626.6 NA 85.3 NA 56.2 29.0 106.5 Industrial Coking Coal 88.6 31.00 2.7 0.10 0.3 2.5 9.1 Industrial Other Coal 10.3 25.82 0.3 0.66 0.2 0.1 0.3 Natural Gas to Chemical Plants 289.5 14.47 4.2 0.66 2.8 1.4 5.2 Asphalt & Road Oil 853.4 20.55 17.5 1.00 17.5 0.1 0.3 LPG 2,117.6 17.06 36.1 0.66 23.8 12.3 45.2 Lubricants 148.9 20.20 3.0 0.09 0.3 2.7 10.0 Pentanes Plus 53.0 19.10 1.0 0.66 0.7 0.3 1.3 Naphtha (<401° F) 396.6 18.55 7.4 0.66 4.8 2.5 9.2 Other Oil (>401° F) 203.8 20.17 4.1 0.66 2.7 1.4 5.1 Still Gas 166.1 17.51 2.9 0.66 1.9 1.0 3.6 Petroleum Coke 0.0 27.85 0.0 0.30 0.0 0.0 0.0 Special Naphtha 88.7 19.74 1.8 0.66 1.2 0.6 2.2 Distillate Fuel Oil 5.8 20.17 0.1 0.50 0.1 0.1 0.2 Waxes 12.9 19.80 0.3 0.58 0.1 0.1 0.4 Miscellaneous Products 191.3 20.31 3.9 0.00 0.0 3.9 14.2 Transportation 140.6 NA 2.8 NA 0.3 2.6 9.5 Lubricants 140.6 20.20 2.8 0.09 0.3 2.6 9.5 U.S. Territories 77.3 NA 1.5 NA 0.2 1.4 5.1 Lubricants 1.0 20.20 0.0 0.09 0.0 0.0 0.1 Other Petroleum (Misc. Prod.) 76.2 20.00 1.5 0.10 0.2 1.4 5.0 Total 4,844.5 89.7 56.6 33.0 121.0 + Does not exceed 0.05 TBtu, MMT C, MMT CO2 Eq. NA (Not Applicable) aTo avoid double counting, net exports have been deducted. Note: Totals may not sum due to independent rounding. Lastly, emissions were estimated by subtracting the C stored from the potential emissions (see Table 3-19). 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, 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), National Emissions Inventory (NEI) Air Pollutant Emissions Trends Data (EPA 2016a), Toxics Release Inventory, 1998 (EPA 2000b), Biennial Reporting System (EPA 2000a, 2009), Resource Conservation and Recovery Act Information System (EPA 2013b, 2015b, 2016c), pesticide sales and use estimates (EPA 1998, 1999, 2002, 2004, 2011, 2017), and the Chemical Data Access Tool (EPA 2012); the EIA Manufacturer's Energy Consumption Survey (MECS) (EIA 1994, 1997, 2001, 2005, 2010, 2013, 2017a); the National Petrochemical & Refiners Association (NPRA 2002); the U.S. Census Bureau (1999, 2004, 2009, 2014); Bank of Canada (2012, 2013, 2014, 2016, 2017); Financial Planning Association (2006); INEGI (2006); the United States International Trade Energy 3-47 ------- 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 Commission (1990 through 2016); Gosselin, Smith, and Hodge (1984); EPA's Municipal Solid Waste (MSW) Facts and Figures (EPA 2013a, 2014a, 2016b); the Rubber Manufacturers' Association (RMA 2009, 2011, 2014, 2016); the International Institute of Synthetic Rubber Products (IISRP 2000, 2003); the Fiber Economics Bureau (FEB 2001, 2003, 2005, 2007, 2009, 2010, 2011, 2012, 2013, 2017); the EPA Chemical Data Access Tool (CDAT) (EPA 2014b); the American Chemistry Council (ACC 2003 through 2011, 2013, 2014, 2015a, 2016b, 2017); and the Guide to the Business of Chemistry (ACC 2012, 2015b, 2016a). Specific data sources are listed in full detail in Annex 2.3. Uncertainty and lime-Serfi insistency 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 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, LPG, pentanes plus, 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 3-20 and Table 3-21), 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 3-22 (emissions) and Table 3-23 (storage factors). Carbon emitted from non-energy uses of fossil fuels in 2016 was estimated to be between 93.5 and 166.5 MMT CO2 Eq. at a 95 percent confidence level. This indicates a range of 23 percent below to 38 percent above the 2016 emission estimate of 121.0 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 3-22: Approach 2 Quantitative Uncertainty Estimates for CO2 Emissions from Non- Energy Uses of Fossil Fuels (MMT CO2 Eq. and Percent) Source Gas 2016 Emission Estimate Uncertainty Range Relative to Emission Estimate3 (MMT CO2 Eq.) (MMT CO2 Eq.) (%) Lower Upper Lower Upper Bound Bound Bound Bound Feedstocks CO2 72.3 51.0 121.3 -29% 68% Asphalt CO2 0.3 0.1 0.6 -58% 120% Lubricants CO2 19.5 16.1 22.8 -18% 16% Waxes CO2 0.4 0.3 0.7 -23% 77% Other CO2 28.5 17.3 31.1 -39% 9% Total CO2 121.0 93.5 166.5 -23% 38% 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. 3-48 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2016 ------- 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 Table 3-23: Approach 2 Quantitative Uncertainty Estimates for Storage Factors of Non- Energy Uses of Fossil Fuels (Percent) Source Gas 2016 Storage Factor (%) Uncertainty Range Relative to Emission Estimate3 (%) (%, Relative) Lower Bound Upper Bound Lower Bound Upper Bound Feedstocks CO2 65.9% 50% 70% -19% 5% Asphalt CO2 99.6% 99% 100% -0.5% 0.3% Lubricants CO2 9.2% 4% 17% -58% 91% Waxes CO2 57.8% 48% 67% -17% 17% Other CO2 6.4% 6% 43% -4% 569% 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 3-23, feedstocks and asphalt contribute least to overall storage factor uncertainty on a percentage basis. Although the feedstocks category—the largest use category in terms of total carbon flows—appears to have tight confidence limits, this is to some extent an artifact of the way the uncertainty analysis was structured. As discussed in Annex 2.3, the storage factor for feedstocks is based on an analysis of six fates that result in long-term storage (e.g., plastics production), and eleven that result in emissions (e.g., volatile organic compound emissions). Rather than modeling the total uncertainty around all of these fate processes, the current analysis addresses only the storage fates, and assumes that all C that is not stored is emitted. As the production statistics that drive the storage values are relatively well-characterized, this approach yields a result that is probably biased toward understating uncertainty. 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. Methodological recalculations were applied to the entire time series to ensure time-series consistency from 1990 through 2016 as discussed below. Details on the emission trends through time are described in more detail in the Methodology section, above. QA/QC and Verification A source-specific QA/QC plan for non-energy uses of fossil fuels was developed and implemented. 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 category-specific procedures that were implemented involved checks specifically focusing on the activity data and methodology for estimating the fate of C (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 2015 totals as well as their trends across the time series. Petrochemical input data reported by EIA will continue to be investigated in an attempt to address an input/output discrepancy in the NEU model. Prior to 2001, the C balance inputs exceed outputs, then starting in 2001 through 2009, outputs exceeded inputs. In 2010 through 2016, inputs exceeded outputs. A portion of this discrepancy has been reduced and two strategies have been developed to address the remaining portion (see the Planned Improvements section, below). Energy 3-49 ------- 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 Recalculations Discussion A number of updates to historical production values were included in the most recent Monthly Energy Review; these have been populated throughout the Inventory. Pesticide production data for 2007 through 2015 were updated using EPA's Pesticides Industry Sales and Usage 2008 - 2012 Market Estimates (EPA 2017). This resulted in a slight increase in emissions from pesticides compared to previous estimates for 2007 through 2015. Pesticide production data for 1990 through 2015 were updated by correcting rounding errors and molecular weights and chemical formulas for certain pesticides. The calculated ratio of urea production to melamine production from 2001 to 2015 was updated to approximately 95/5 based on ICIS (2016) and ICIS (2008), rather tlian an even 50/50 split as previously estimated. Planned Improvements There are several future improvements planned: • Analyzing the fuel and feedstock data from EPA's GHGRP subpart X (Petrochemical Production) to better disaggregate CO2 emissions in NEU model and CO2 process emissions from petrochemical production. • More accurate accounting of C in petrochemical feedstocks. EPA has worked with EI A to determine the cause of input/output discrepancies in the C mass balance contained within the NEU model. In the future, two strategies to reduce or eliminate this discrepancy will continue to be pursued. First, accounting of C 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 C input calculation in estimating emissions will be reconsidered. Alternative approaches that rely more substantially on the bottom-up C 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 C. Additional fates may be researched, including the fossil C 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 (miscellaneous products). A better understanding of these trends will be pursued to identify any mischaracterized or misreported fuel consumption for non-energy uses. For example, "miscellaneous products" category includes miscellaneous products that are not reported elsewhere in the EIA data set. The EIA does not have firm data concerning the amounts of various products that are being reported in the "miscellaneous products" category; however, EIA has indicated that recovered sulfur from petroleum and natural gas processing, and potentially also C black feedstock could be reported in this category. Recovered sulfur would not be reported in the NEU calculation or elsewhere in the Inventory. • Updating the average C content of solvents was researched, since the entire time series depends on one year's worth of solvent composition data. The data on C 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 C in solvents. Additional sources of solvents data will be investigated in order to update the C content assumptions. • Updating the average C 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 sources may facilitate updating the average C content for this category. • Revising the methodology for consumption, production, and C 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 3-50 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2016 ------- 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 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 C 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. Box 3-6: 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 Industrial Processes and Product Use (IPPU) sector. 67 In this Inventory, C storage and C 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 (CRF Source Category 1A).68 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. 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., C 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 Section 3.2, Table 3-21). For those inputs, U.S. country-specific data on C stocks and flows are used to develop carbon storage factors, which are calculated as the ratio of the C stored by the fossil fuel non-energy products to the total C content of the fuel consumed, taking into account losses in the production process and during product use.69 The country-specific methodology to reflect national circumstances starts with the aggregate amount of fossil fuels used for non-energy uses and applies a C balance calculation, breaking out the C emissions from non-energy use of lubricants, waxes, and asphalt and road oil. Due to U.S. national circumstances, reporting these C emissions separately under IPPU would involve making artificial adjustments to allocate both the C inputs and C outputs of the non-energy use C balance. These artificial adjustments would also result in the C emissions for lubricants, waxes, and asphalt and road oil being reported under IPPU, while the C storage for lubricants, waxes, and asphalt and road oil would be reported under Energy. To avoid presenting an incomplete C balance and a less transparent approach for the Carbon Emitted from Non-Energy Uses of Fossil Fuels source category calculation the entire calculation of C storage and C emissions is therefore conducted in the Non-Energy Uses of Fossil Fuels category calculation methodology, and both the C storage and C emissions for lubricants, waxes, and asphalt and road oil are reported under the Energy sector. However, 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 IPPU chapter, as they were consumed during non-energy related industrial activity. Emissions from 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. 3.3 Incineration of Waste (CRF Source 67 See Volume 3: Industrial Processes and Product Use, Chapter 5: Non-Energy Products from Fuels and Solvent Use of the 2006 IPCC Guidelines for National Greenhouse Gas Inventories (IPCC 2006). 68 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. 69 Data and calculations for lubricants and waxes and asphalt and road oil are in Annex 2.3 - Methodology and Data for Estimating CO2 Emissions from Fossil Fuel Combustion. Energy 3-51 ------- 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 Category lAla) - TO BE UPDATED FOR FINAL INVENTORY REPORT Incineration 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; Goldstein andMadtes 2001; Kaufman et al. 2004; Simmons et al. 2006; van Haarenet al. 2010). In the context of this section waste includes all municipal solid waste (MSW) as well as scrap tires. In the United States, incineration of MSW tends to occur at waste-to-energy facilities or industrial facilities where useful energy is recovered, and thus emissions from waste incineration 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. Incineration of waste results in conversion of the organic inputs to CO2. According to 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 incineration are calculated by estimating the quantity of waste combusted and the fraction of the waste that is C derived from fossil sources. Most of the organic materials in municipal solid wastes are of biogenic origin (e.g., paper, yard trimmings), and have their net C flows accounted for under the Land Use, Land-Use Change, and Forestry chapter. However, some components—plastics, synthetic rubber, synthetic fibers, and carbon black in scrap tires—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 municipal solid wastes 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 incineration estimate, though waste disposal practices for tires differ from municipal solid waste. Estimates on emissions from hazardous waste incineration can be found in Annex 2.3 and are accounted for as part of the C mass balance for non-energy uses of fossil fuels. Approximately 30.1 million metric tons of MSW were incinerated in the United States in 2014 (EPA 2016). Data for the amount of MSW incinerated in 2015 were not available, so data for 2015 was assumed to be equal to data for 2014. CO2 emissions from incineration of waste rose 34 percent since 1990, to an estimated 10.7 MMT CO2 Eq. (10,676 kt) in 2015, as the volume of scrap tires and other fossil C-containing materials in waste increased (see Table 3-24 and Table 3-25). Waste incineration is also a source of CH4 and N20 emissions (De Soete 1993; IPCC 2006). Methane emissions from the incineration of waste were estimated to be less than 0.05 MMT CO2 Eq. (less than 0.5 kt CH4) in 2015, and have decreased by 32 percent since 1990. Nitrous oxide emissions from the incineration of waste were estimated to be 0.3 MMT CO2 Eq. (1 kt N20) in 2015, and have decreased by 32 percent since 1990. Table 3-24: CO2, ChU, and N2O Emissions from the Incineration of Waste (MMT CO2 Eq.) Gas/Waste Product 1990 2005 2011 2012 2013 2014 2015 CO2 8.0 12.5 10.6 10.4 10.4 10.6 10.7 Plastics 5.6 6.9 5.8 5.7 5.8 5.9 5.9 Synthetic Rubber in Tires 0.3 1.6 1.4 1.3 1.2 1.2 1.2 Carbon Black in Tires 0.4 2.0 1.7 1.5 1.4 1.4 1.5 Synthetic Rubber in MSW 0.9 0.8 0.7 0.7 0.7 0.7 0.7 Synthetic Fibers 0.8 1.2 1.1 1.1 1.3 1.3 1.3 CH4 + + + + + + + N2O 0.5 0.4 0.3 0.3 0.3 0.3 0.3 Total 8.4 12.8 10.9 10.7 10.7 10.9 11.0 + Does not exceed 0.05 MMT CO2 Eq. Table 3-25: CO2, ChU, and N2O Emissions from the Incineration of Waste (kt) Gas/Waste Product 1990 2005 2011 2012 2013 2014 2015 3-52 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2016 ------- 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 C02 7,950 12,469 10,564 10,379 10,398 10,608 10,676 Plastics 5,588 6,919 5,757 5,709 5,815 5,928 5,928 Synthetic Rubber in Tires 308 1,599 1,363 1,261 1,158 1,189 1,220 Carbon Black in Tires 385 1,958 1,663 1,537 1,412 1,449 1,487 Synthetic Rubber in MSW 854 766 712 706 729 729 729 Synthetic Fibers 816 1,227 1,070 1,166 1,284 1,313 1,313 ch4 + + + + + + + n2o 2 1 1 1 1 1 1 + Does not exceed 0.5 kt Methodology Emissions of CO2 from the incineration of waste include CO2 generated by the incineration of plastics, synthetic fibers, and synthetic rubber in MSW, as well as the incineration of synthetic rubber and carbon black in scrap tires. The emission estimates are calculated for all four sources on a mass-basis based on the data available. These emissions were estimated by multiplying the mass of each material incinerated by the C content of the material and the fraction oxidized (98 percent). Plastics incinerated in municipal solid wastes were categorized into seven plastic resin types, each material having a discrete C content. Similarly, synthetic rubber is categorized into three product types, and synthetic fibers were categorized into four product types, each having a discrete C content. Scrap tires contain several types of synthetic rubber, carbon black, and synthetic fibers. Each type of synthetic rubber has a discrete C content, and carbon black is 100 percent C. 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. More detail on the methodology for calculating emissions from each of these waste incineration sources is provided in Annex 3.7. For each of the methods used to calculate CO2 emissions from the incineration of waste, data on the quantity of product combusted and the C content of the product are needed. For plastics, synthetic rubber, and synthetic fibers in MSW, the amount of specific materials discarded as municipal solid waste (i.e., the quantity generated minus the quantity recycled) was taken from Municipal Solid H aste Generation, Recycling, and Disposal in the United States: Facts and Figures (EPA 2000 through 2003, 2005 through 2014), and Advancing Sustainable Materials Management: Facts and Figures: Assessing Trends in Material Generation, Recycling and Disposal in the United States (EPA 2015, 2016) and detailed unpublished backup data for some years not shown in the reports (Schneider 2007). For 2015, the amount of MSW incinerated was assumed to be equal to that in 2014, due to the lack of available data. The proportion of total waste discarded that is incinerated was derived from Shin (2014). Data on total waste incinerated was not available in detail for 2012 through 2015, so these values were assumed to equal to the 2011 value (Shin 2014). For synthetic rubber and carbon black in scrap tires, information was obtained from U.S. Scrap Tire Management Summary for 2005 through 2015 data (RMA 2016). Average C contents for the "Other" plastics category and synthetic rubber in municipal solid wastes were calculated from 1998 and 2002 production statistics: C content for 1990 through 1998 is based on the 1998 value; C content for 1999 through 2001 is the average of 1998 and 2002 values; and C content for 2002 to date is based on the 2002 value. Carbon content for synthetic fibers was calculated from a weighted average of production statistics from 1990 to date. Information about scrap tire composition was taken from the Rubber Manufacturers' Association internet site (RMA 2012a). The mass of incinerated material is multiplied by its C content to calculate the total amount of carbon stored. The assumption that 98 percent of organic C is oxidized (which applies to all waste incineration categories for CO2 emissions) was reported in EPA's life cycle analysis of greenhouse gas emissions and sinks from management of solid waste (EPA 2006). This percentage is multiplied by the carbon stored to estimate the amount of carbon emitted. Incineration of waste, including MSW, also results in emissions of CH4 and N20. These emissions were calculated as a function of the total estimated mass of waste incinerated and emission factors. As noted above, CH4 and N20 emissions are a function of total waste incinerated in each year; for 1990 through 2008, these data were derived from the information published in BioCvcle (van Haaren et al. 2010). Data for 2009 and 2010 were interpolated between 2008 and 2011 values. Data for 2011 were derived from Shin (2014). Data on total waste incinerated was not Energy 3-53 ------- 1 available in the BioCvcle data set for 2012 through 2015, so these values were assumed to equal the 2011 BioCvcle 2 data set value. 3 Table 3-26 provides data on municipal solid waste discarded and percentage combusted for the total waste stream. 4 The emission factors of N20 and CH4 emissions per quantity of municipal solid waste combusted are default 5 emission factors for the default continuously-fed stoker unit MSW incineration technology type and were taken from 6 IPCC (2006). 7 Table 3-26: Municipal Solid Waste Generation (Metric Tons) and Percent Combusted 8 (BioCycle dataset) Incinerated (% of Year Waste Discarded Waste Incinerated Discards) 1990 235,733,657 30,632,057 13.0% 2005 259,559,787 25,973,520 10.0% 2011 273,116,704 20,756,870 7.6% 2012 273,116,704a 20,756,870 7.6% 2013 273,116,704a 20,756,870 7.6% 2014 273,116,704a 20,756,870 7.6% 201 5 273,116,704a 20,756,870 7.6% a Assumed equal to 2011 value. Source: van Haaren et al. (2010) 9 Uncertainty and Time-Series Consistency 10 An Approach 2 Monte Carlo analysis was performed to determine the level of uncertainty surrounding the estimates 11 of CO2 emissions and N20 emissions from the incineration of waste (given the very low emissions for CH4, no 12 uncertainty estimate was derived). IPCC Approach 2 analysis allows the specification of probability density 13 functions for key variables within a computational structure that mirrors the calculation of the Inventory estimate. 14 Uncertainty estimates and distributions for waste generation variables (i.e., plastics, synthetic rubber, and textiles 15 generation) were obtained through a conversation with one of the authors of the Municipal Solid Waste in the 16 United States reports. Statistical analyses or expert judgments of uncertainty were not available directly from the 17 information sources for the other variables; thus, uncertainty estimates for these variables were determined using 18 assumptions based on source category knowledge and the known uncertainty estimates for the waste generation 19 variables. 20 The uncertainties in the waste incineration emission estimates arise from both the assumptions applied to the data 21 and from the quality of the data. Key factors include MSW incineration rate; fraction oxidized; missing data on 22 waste composition; average C content of waste components; assumptions on the synthetic/biogenic C ratio; and 23 combustion conditions affecting N20 emissions. The highest levels of uncertainty surround the variables that are 24 based on assumptions (e.g., percent of clothing and footwear composed of synthetic rubber); the lowest levels of 25 uncertainty surround variables that were determined by quantitative measurements (e.g., combustion efficiency, C 26 content of C black). 27 The results of the Approach 2 quantitative uncertainty analysis are summarized in Table 3-27. Waste incineration 28 CO2 emissions in 2015 were estimated to be between 9.6 and 12.1 MMT CO2 Eq. at a 95 percent confidence level. 29 This indicates a range of 10 percent below to 13 percent above the 2015 emission estimate of 10.7 MMT CO2 Eq. 30 Also at a 95 percent confidence level, waste incineration N20 emissions in 2015 were estimated to be between 0.2 31 and 1.3 MMT CO2 Eq. This indicates a range of 51 percent below to 3 3 0 percent above the 2015 emission estimate 32 of 0.3 MMT C02 Eq. 3-54 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2016 ------- 1 2 Table 3-27: Approach 2 Quantitative Uncertainty Estimates for CO2 and N2O from the Incineration of Waste (MMT CO2 Eq. and Percent) 2015 Emission Estimate Uncertainty Range Relative to Emission Estimate3 Source Gas (MMT CP2 Eq.) (MMT CO2 Eq.) [%) Lower Upper Lower Upper Bound Bound Bound Bound Incineration of Waste CO2 10.7 9.6 12.1 -10% +13% Incineration of Waste N2O (13 02 L3 -51% +330% a Range of emission estimates predicted by Monte Carlo Simulation for a 95 percent confidence interval. 3 Methodological recalculations were applied to the entire time series to ensure time-series consistency from 1990 4 through 2015 as discussed below. Details on the emission trends through time are described in more detail in the 5 Methodology section above. 6 QA/QC and Verification 7 A source-specific Quality Assurance/Quality Control plan was implemented for incineration of waste. This effort 8 included a general (Tier 1) analysis, as well as portions of a category-specific (Tier 2) analysis. The Tier 2 9 procedures that were implemented involved checks specifically focusing on the activity data and specifically 10 focused on the emission factor and activity data sources and methodology used for estimating emissions from 11 incineration of waste. Trends across the time series were analyzed to determine whether any corrective actions were 12 needed. Actions were taken to streamline the activity data throughout the calculations on incineration of waste. 13 Recalculations Discussion 14 For the current Inventory, emission estimates for 2014 have been updated based on Advancing Sustainable 15 Materials Management: 2014 Fact Sheet (EPA 2016). The data used to calculate the percent incineration was not 16 updated in the current Inventory. BioCvcle lias not released a new State of Garbage in America Report since 2010 17 (with 2008 data), which used to be a semi-annual publication which publishes the results of the nation-wide MSW 18 survey. The results of the survey have been published in Shin (2014). This provided updated incineration data for 19 2011, so the generation and incineration data for 2012 through 2015 are assumed equivalent to the 2011 values. The 20 data for 2009 and 2010 were based on interpolations between 2008 and 2011. 21 A transcription error in 2013 plastics production data from EPA's Advancing Sustainable Materials Management: 22 Facts and Figures 2013: Assessing Trends in Material Generation, Recycling and Disposal in the United States 23 (EPA 2015) was identified and corrected. The amount of HDPE discarded in 2013 was misreported and the value 24 has been updated. This update results in updated emission estimate for the CO2 from Plastics for 2013. 25 Previously, the carbon content for synthetic fiber was assumed to be the weighted average of carbon contents of four 26 fiber types (polyester, nylon olefin and acrylic) based on 1999 fiber production data. This methodology lias been 27 updated. A weighted average for the carbon content of synthetic fibers based on production data from 1990 through 28 2015 was developed for each year based on the amount of fiber produced. For each year, the weighted average 29 carbon content was used to develop the amount of carbon emitted. This methodology update affects the synthetic 30 fiber CO2 estimates. 31 Planned Improvements 32 The availability of facility-level waste incineration data through EPA's Greenhouse Gas Reporting Program 33 (GHGRP) will be examined to help better characterize waste incineration operations in the United States. This 34 characterization could include future improvements as to the operations involved in waste incineration for energy, 35 whether in the power generation sector or the industrial sector. Additional examinations will be necessary as, unlike Energy 3-55 ------- 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 the reporting requirements for this chapter under the UNFCCC reporting guidelines,70 some facility-level waste incineration emissions reported under EPA's GHGRP may also include industrial process emissions. In line with UNFCCC reporting guidelines, emissions for waste incineration with energy recovery 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 waste incineration 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 ensuring CO2 emissions from the biomass component of waste are separated in the facility-level reported data, and on maintaining consistency with national waste generation and fate statistics currently used to estimate total, national U.S. greenhouse gas emissions. 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.71 GHGRP data is available for MSW combustors, which contains information on the CO2, CH4, and N20 emissions from MSW combustion plus the fraction of the emissions that are biogenic. To calculate biogenic versus total CO2 emissions, a default biogenic fraction of 0.6 is used. The biogenic fraction will be calculated using the current input data and assumptions to verily the current MSW emission estimates. If GHGRP data would not provide a more accurate estimate of the amount of solid waste combusted, new data sources for the total MSW generated will be explored given that the data previously published semi-annually in BioCvcIe (van Haaren et al. 2010) has ceased to be published, according to the authors. Equivalent data was derived from Shin (2014) for 2011. A new methodology would be developed considering the available data within the annual update of EPA's Ach'ancing Sustainable Materials Management: Facts and Figures 2014: Assessing Trends in Material Generation, Recycling and Disposal in the United States (EPA 2016) and a report from the Enviromnental Research & Education Foundation (2016), MSW Management in the U.S.: 2010 & 2013, that has data for 2010 and 2013. In developing the new methodology, appropriate assumptions would need to be made to ensure that the MSW figures include the same boundaries. Consideration would also be made to be consistent with calculations in other waste categories including landfilling and composting. Additional improvements will be conducted to improve the transparency in the current reporting of waste incineration. Currently, hazardous industrial waste incineration is included within the overall calculations for the Carbon Emitted from Non-Energy Uses of Fossil Fuels source category. Waste incineration activities that do not include energy recovery will be examined. Synthetic fibers within scrap tires are not included in this analysis and will be explored for future Inventories. The carbon content of fibers within scrap tires would be used to calculate the associated incineration emissions. Updated fiber content data from the Fiber Economics Bureau will also be explored. 3.4 Coal Mining (CRF Source Category lBla) Three types of coal mining-related activities release CH4 to the atmosphere: underground mining, surface mining, and post-mining (i.e., coal-handling) activities. While surface mines account for the majority of U.S. coal production underground coal mines contribute the largest share of CH4 emissions (see Table 3-29 and Table 3-30) due to the higher CH4 content of coal in the deeper underground coal seams. In 2016, 251 underground coal mines and 439 surface mines were operating in the United States. In recent years the total number of active coal mines in the United States lias declined. In 2016, the United States was the third largest coal producer in the world (660 MMT), after China (3,242 MMT) and India (708 MMT) (IEA 2017). 70 See . 71 See . 3-56 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2016 ------- 1 Table 3-28: Coal Production (kt) Year Underground Surface Total Number of Mines Production Number of Mines Production Number of Mines Production 1990 1,683 384,244 1,656 546,808 3,339 931,052 2005 586 334,398 789 691.448 1.398 1,025.846 2012 488 310,608 719 610,307 1,207 920,915 2013 395 309,546 637 581,270 1,032 890,815 2014 345 321,783 613 583,974 958 905,757 2015 305 278,342 529 534,127 834 812,469 2016 251 228,403 439 431,485 690 659,888 2 Underground mines liberate CH4 from ventilation systems and from degasification systems. Ventilation systems 3 pump air through the mine workings to dilute noxious gases and ensure worker safety; these systems can exhaust 4 significant amounts of CH4 to the atmosphere in low concentrations. Degasification systems are wells drilled from 5 the surface or boreholes drilled inside the mine that remove large, often highly concentrated volumes of CH4 before, 6 during, or after mining. Some mines recover and use CH4 generated from ventilation and degasification systems, 7 thereby reducing emissions to the atmosphere. 8 Surface coal mines liberate CH4 as the overburden is removed and the coal is exposed to the atmosphere. CH4 9 emissions are normally a function of coal rank (a classification related to the percentage of carbon in the coal) and 10 depth. Surface coal mines typically produce lower-rank coals and remove less than 250 feet of overburden, so their 11 level of emissions is much lower than from underground mines. 12 In addition, CH4 is released during post-mining activities, as the coal is processed, transported, and stored for use. 13 Total CH4 emissions in 2016 were estimated to be 2,153 kt (53.8 MMT CO2 Eq.), a decline of 44 percent since 1990 14 (see Table 3-29 and Table 3-30). Of these total emissions, underground mines accounted for approximately 76 15 percent, surface mines accounted for 13 percent, and post-mining activities accounted for 12 percent. 16 Table 3-29: ChU Emissions from Coal Mining (MMT CO2 Eq.) Activity 1990 2005 2012 2013 2014 2015 2016 Underground (UG) Mining 74.2 42.0 47.3 46.2 46.1 44.9 40.7 Liberated 80.8 59.7 65.8 64.5 63.1 61.2 57.1 Recovered & Used (6.6) (17.7) (18.5) (18.3) (17.0) (16.4) (16.3) Surface Mining 10.8 11.9 10.3 9.7 9.6 8.7 6.8 Post-Mining (UG) 9.2 7.6 6.7 6.6 6.7 5.8 4.8 Post-Mining (Surface) 2.3 2.6 2.2 2.1 2.1 1.9 1.5 Total 96.5 64.1 66.5 64.6 64.6 61.2 53.8 Notes: Totals may not sum due to independent rounding. Parentheses indicate negative values. 17 Table 3-30: ChU Emissions from Coal Mining (kt) Activity 1990 2005 2012 2013 2014 2015 2016 UG Mining 2,968 1,682 1,891 1,849 1,844 1,796 1,629 Liberated 3,234 2,390 2,631 2,580 2,524 2,450 2,282 Recovered & Used (266) (708) (740) (730) (680) (654) (654) Surface Mining 430 475 410 388 386 347 273 Post-Mining (UG) 368 306 268 263 270 231 192 Post-Mining (Surface) 93 103 89 84 84 75 59 Total 3,860 2,565 2,658 2,584 2,583 2,449 2,153 Notes: Totals may not sum due to independent rounding. Parentheses indicate negative values. Energy 3-57 ------- 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 Methodology The methodology for estimating CH4 emissions from coal mining consists of two steps: • Estimate 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 from degasification systems. Some mines recover and use the generated CH4, thereby reducing emissions to the atmosphere. Total CH4 emitted from underground mines equals the CH4 liberated from ventilation systems, plus the CH4 liberated from degasification systems, minus the CH4 recovered and used. Step 1.1: Estimate CH4 Liberatedfrom Ventilation Systems To estimate CH4 liberated from ventilation systems, EPA uses data collected through its Greenhouse Gas Reporting Program (GHGRP)72 (subpart FF, "Underground Coal Mines"), data provided by the U.S. Mine Safety and Health Administration (MSHA), 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 1 6).73 Mines that report to EPA's GHGRP must report quarterly measurements of CH4 emissions from ventilation systems to EPA; they have the option of recording 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.74 Since 2013, ventilation emission estimates have been calculated based on both GHGRP data submitted by underground mines, and on quarterly measurement data obtained directly from MSHA for the remaining mines. The quarterly measurements are used to determine the average daily emissions rate for the reporting year quarter. 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. Step 1.2: Estimate CH4 Liberatedfrom 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. Twenty-five mines used degasification systems in 2016, and the CH4 removed through these systems was reported to EPA's GHGRP under subpart FF (EPA 2017). Based on the weekly measurements reported to EPA's GHGRP, degasification data summaries for each mine were added to estimate the CH4 liberated from degasification systems. Fifteen of the 25 mines with degasification systems had operational CH4 recovery and use projects (see step 1.3 72 In implementing improvements and integrating data from EPA's GHGRP, the EPA followed the latest guidance from the IPCC on the use of facility-level data in national inventories (IPCC 2011). 73 Underground coal mines report to EPA under Subpart FF of the GHGRP. In 2016, 90 underground coal mines reported to the program. 74 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. 3-58 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2016 ------- 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 below), and EPA's GHGRP reports show the remaining ten mines vented CH4from degasification systems to the atmosphere.75 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 to estimate CH4 liberated from degasification systems at 20 of the 25 mines that used degasification systems in 2016. 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.76 EPA's GHGRP does not require gas production from virgin coal seams (coalbed methane) to be reported by coal mines under subpart FF.77 Most pre-mining wells drilled from the surface are considered coalbed methane wells prior to mine-through and associated CH4 emissions are reported under another subpart of the program (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. For five mines with degasification systems that include pre-mining wells that were mined through in 2016, GHGRP information was supplemented with historical data from state gas well production databases (DMME 2017; GSA 2017; WVGES 2017), 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). EPA's GHGRP reports with CH4 liberated from degasification systems are reviewed for errors in reporting. For one of the 25 mines, due to a lack of mine-provided information used in prior years and a GHGRP reporting discrepancy, the CH4 liberated was based on both an estimate from historical mine-provided CH4 recovery and use rates and state gas sales records (DMME 2017). Step 1.3: Estimate CH4 Recovered from Ventilation and Degasification Systems, and Utilized or Destroyed (Emissions Avoided) Fifteen mines had CH4 recovery and use projects in place in 2016. Fourteen of these mines sold the recovered CH4 to a pipeline, including one that also used CH4 to fuel a thermal coal dryer. In addition, one mine used recovered CH4to heat mine ventilation air. EPA's GHGRP data was exclusively used to estimate the CH4 recovered and used from ten of the 15 mines that deployed degasification systems in 2016. Based on weekly measurements, the GHGRP degasification destruction data summaries for each mine were added together to estimate the CH4 recovered and used from degasification systems. All 10 mines with degasification systems used pre-mining wells as part of those systems, but only four of the mines intersected pre-mining wells in 2016. EPA's GHGRP and supplemental data were used to estimate CH4 recovered and used at two of these four mines; supplemental data alone (GSA 2017) was used to estimate CH4 recovered and used at the other two mines. Supplemental information was used for these four 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. The supplemental data came from state gas production databases as well as mine-specific information on the timing of mined-through pre-mining wells. EPA's GHGRP information was not used to estimate CH4 recovered and used at two mines. At one of these mines, a portion of reported CH4 vented was applied to an ongoing mine air heating project. Because of a lack of mine- provided information used in prior years and a GHGRP reporting discrepancy, the 2016 CH4 recovered and used from pre-mining wells at the other mine was based on an estimate from historical mine-provided CH4 recovery and use rates. Emissions recovered and used from the active mine degasification system were estimated based on a state gas production data information system. 75 Several of the mines venting CH4 from degasification systems use a small portion the gas to fuel gob well blowers in remote locations where electricity is not available. However, this CH4use is not considered to be a formal recovery and use project. 76 A well is "mined through" when coal mining development or the working face intersects the borehole or well. 77 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 3-59 ------- 1 In 2016, one mine destroyed a portion of its CH4 emissions from ventilation systems using thermal oxidation 2 technology. The amount of CH4 recovered and destroyed by the project was determined through publicly-available 3 emission reduction project information (ACR 2017). 4 Step 2: Estimate CH4 Emitted from Surface Mines and Post-Mining Activities 5 Mine-specific data are not available for estimating CH4 emissions from surface coal mines or for post-mining 6 activities. For surface mines, basin-specific coal production obtained from the Energy Information Administration's 7 Annual Coal Report (EIA 2017) was multiplied by basin-specific CHi contents (EPA 1996, 2005) and a 150 percent 8 emission factor (to account for CH4from over- and under-burden) to estimate CH4 emissions (King 1994; Saghafi 9 2013). For post-mining activities, basin-specific coal production was multiplied by basin-specific gas contents and a 10 mid-range 32.5 percent emission factor for CH4 desorption during coal transportation and storage (Creedy 1993). 11 Basin-specific in situ gas content data were compiled from AAPG (1984) and USBM (1986). 12 Uncertainty and Time-Series Consistency 13 A quantitative uncertainty analysis was conducted for the coal mining source category using the IPCC- 14 recommended Approach 2 uncertainty estimation methodology. Because emission estimates from underground 15 ventilation systems were based on actual measurement data from EPA's GHGRP or from MSHA, uncertainty is 16 relatively low. A degree of imprecision was introduced because the ventilation air measurements used were not 17 continuous but rather quarterly instantaneous readings that were used to determine the average daily emissions rate 18 for the quarter. Additionally, the measurement equipment used can be expected to have resulted in an average of 10 19 percent overestimation of annual CH4 emissions (Mutmansky & Wang 2000). GHGRP data were used for a 20 significant number of the mines that reported their own measurements to the program beginning in 2013; however, 21 the equipment uncertainty is applied to both GHGRP and MSHA data. 22 Estimates of CH4 recovered by degasification systems are relatively certain for utilized CH4 because of the 23 availability of EPA's GHGRP data and gas sales information. Many of the recovery estimates use data on wells 24 within 100 feet of a mined area. However, uncertainty exists concerning the radius of influence of each well. The 25 number of wells counted, and thus the avoided emissions, may vary if the drainage area is found to be larger or 26 smaller than estimated. 27 EPA's GHGRP requires weekly CH4 monitoring of mines that report degasification systems, and continuous CH4 28 monitoring is required for utilized CH4 on- or off-site. Since 2012, GHGRP data have been used to estimate CH4 29 emissions from vented degasification wells, reducing the uncertainty associated with prior MSHA estimates used for 30 this subsource. Beginning in 2013, GHGRP data were also used for determining CH4 recovery and use at mines 31 without publicly available gas usage or sales records, which has reduced the uncertainty from previous estimation 32 methods that were based on information from coal industry contacts. 33 In 2015 and 2016, a small level of uncertainty was introduced with using estimated rather than measured values of 34 recovered methane from two of the mines with degasification systems. An increased level of uncertainty was applied 35 to these two subsources, but the change had little impact on the overall uncertainty. 36 Surface mining and post-mining emissions are associated with considerably more uncertainty than underground 37 mines, because of the difficulty in developing accurate emission factors from field measurements. However, since 38 underground emissions constitute the majority of total coal mining emissions, the uncertainty associated with 39 underground emissions is the primary factor that determines overall uncertainty. 40 The results of the Approach 2 quantitative uncertainty analysis are summarized in Table 3-31. Coal mining CH4 41 emissions in 2016 were estimated to be between 47.5 and 61.7 MMT CO2 Eq. at a 95 percent confidence level. This 42 indicates a range of 11.8 percent below to 14.6 percent above the 2016 emission estimate of 53.8 MMT CO2 Eq. 3-60 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2016 ------- 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 Table 3-31: Approach 2 Quantitative Uncertainty Estimates for ChU Emissions from Coal Mining (MMT CO2 Eq. and Percent) Source Gas 2016 Emission Estimate (MMT CO2 Eq.) Uncertainty Range Relative to Emission Estimate3 (MMT CO2 Eq.) (%) Lower Upper Bound Bound Lower Upper Bound Bound Coal mining CH4 53.8 47.5 61.7 -11.8% +14.6% a Range of emission estimates predicted by Monte Carlo stochastic simulation for a 95 percent confidence interval. Methodological recalculations were applied to the entire time series to ensure consistency from 1990 through 2016. Details on the emission trends through time are described in more detail in the methodology section. Recalculations Discussion For the current Inventory, revisions were made to the 2014 and 2015 underground liberated and recovered emissions. The EPA's GHGRP data that was used to calculate the emissions liberated and destroyed in 2014 and 2015 from a mine with a ventilation air methane (VAM) project was incorrect. The GHGRP spreadsheet for Subpart FF reporting does not accommodate methane destruction from VAM, and therefore the emissions avoided are reported as degasification. In 2016, the VAM project's verified emission reductions registered with the California Air Resources Board were deducted from the total reported destroyed methane; and the remaining emissions destroyed were applied to the mine's degasification emissions recovered and destroyed total. The revised methodology was used to recalculate and update the emissions avoided in 2014 and 2015. 3.5 Abandoned Underground Coal Mines (CRF 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 or via ground water aquifers. 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. Annual gross abandoned mine CH4 emissions ranged from 7.2 to 10.8 MMT CO2 Eq. from 1990 through 2016, 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 (10.8 MMT CO2 Eq.) due to the large number Energy 3-61 ------- 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 of gassy mine78 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. There were five gassy mine closures in 2016. In 2016, gross abandoned mine emissions increased slightly from 9.0 to 9.5 MMT CO2 Eq. (see Table 3-32 and Table 3-33). Gross emissions are reduced by CH4 recovered and used at 45 mines, resulting in net emissions in 2016 of 6.7 MMT CO2 Eq. Table 3-32: ChU Emissions from Abandoned Coal Mines (MMT CO2 Eq.) Activity 1990 2005 2012 2013 2014 2015 2016 Abandoned Underground Mines 7.2 8.4 8.9 8.8 8.7 9.0 9.5 Recovered & Used + 1.8 2.7 2.6 2.4 2.6 2.8 Total 7.2 6.6 6.2 6.2 6.3 6.4 6.7 + Does not exceed 0.05 MMT CO2 Eq. Note: Totals may not sum due to independent rounding. Table 3-33: ChU Emissions from Abandoned Coal Mines (kt) Activity 1990 2005 2012 2013 2014 2015 2016 Abandoned Underground Mines 288 334 OO 353 350 359 380 Recovered & Used + 70 109 104 97 102 112 Total 288 264 249 249 253 256 268 + Does not exceed 0.5 kt Note: Totals may not sum due to independent rounding. Methodology Estimating CH4 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 and method 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 which 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. In order to estimate CH4 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 CH4 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 CH4 flow rate is determined by the laws of gas flow through porous media, such as Darcy's Law. Permeability and isotherm data were gathered from each coal basin and histograms were generated. The low, mid and high values of each parameter were combined in nine separate flow simulations for each coal basin using a computational fluid dynamics simulation model used in the oil and gas industry, which generated individual decline curves. These decline curves fit a hyperbolic equation commonly used in the oil and 78 A mine is considered a "gassy" mine if it emits more than 100 thousand cubic feet of CH4 per day (100 mcfd). 3-62 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2016 ------- 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 gas industry for forecasting gas well production. A rate-time equation can be generated that can be used to predict future emissions. This equation is expressed as: q = (1 + &A0("1/fc) where, q b Dj t Gas flow rate at time t in million cubic feet per day (mmcfd) Initial gas flow rate at time zero (tQ), mmcfd The hyperbolic exponent, dimensionless Initial decline rate, 1/year Elapsed time from tQ (years) This equation is applied to mines of various initial emission rates that have similar initial pressures, permeability and adsorption isotherms relative to their coal basin (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 CH4 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. There was not enough data to establish basin-specific equations as was done with the vented, non-flooding mines (EPA 2004). Seals have an inhibiting effect on the flow rate of CH4 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 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) account for about 98 percent of all CH4 emissions. This same relationship is assumed for abandoned mines. It was determined that the 531 abandoned mines closed after 1972 produced emissions greater than 100 mcfd when active. Further, the status of 304 of the 531 mines (or 57 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 CH4 flow to the atmosphere). The remaining 43 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 3-34: Number of Gassy Abandoned Mines Present in U.S. Basins in 2016, grouped by Class according to Post-Abandonment State q = q^ where, q q> D t Gas flow rate at time t in mmcfd Initial gas flow rate at time zero (tQ), mmcfd Decline rate, 1/year Elapsed time from tQ (years) Basin Sealed Vented Flooded Total Known Unknown Total Mines Central Appl. Illinois Northern Appl. 40 34 47 26 3 22 52 14 16 118 51 85 147 31 39 265 82 124 Energy 3-63 ------- 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 Warrior Basin Western Basins 0 28 0 4 16 2 16 34 0 10 16 44 Total 149 55 100 304 227 531 Inputs to the decline equation require the average emission rate and the date of abandonment. Generally, this data is available for mines abandoned after 1971; however, such data are largely unknown for mines closed before 1972, which marks the beginning of comprehensive methane emissions data by the Bureau of Mines. 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 CH4 emissions from coal mining came from seventeen counties in seven states. In addition, 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 CH4 emissions rates during the 1970s (EPA 2004) and closure dates were summarized by decade. Emissions from pre-1972 mines represent approximately 17 percent of total abandoned mine methane emissions. Abandoned mine emission estimates are based on all closed mines known to have active mine CH4 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 2016). Coal mine degasification data are not available for years prior to 1990, thus the initial emission rates used reflect ventilation emissions only for pre-1990 closures. CH4 degasification amounts were added to the quantity of CH4 vented to determine the total CH4 liberation rate for all mines that closed between 1992 and 2016. Since the sample of gassy mines 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 to account for all U.S. abandoned mine emissions. From 1993 through 2016, emission totals were reduced by subtracting abandoned mine CH4 emissions avoided. The Inventory totals were not adjusted for abandoned mine reductions from 1990 through 1992 because no data was reported for abandoned coal mining CH4 recovery projects during that time. A quantitative uncertainty analysis was conducted to estimate the uncertainty surrounding the estimates of emissions from abandoned underground coal mines. The uncertainty analysis described below 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 low, mid and high model generated decline curves for each basin were fitted to a hyperbolic decline curve. The decline curve parameters, Di and b, for the low, mid and high decline curves were then used to define a triangular distribution and together with the initial rate value of a mine's emissions and time from abandonment, a probability density function for each mine in the coal basin was generated. These density functions were then summed together using Monte Carlo simulation software to produce the AMM inventory which would be expressed in terms of a 95 percent confidence interval. The results of the Approach 2 quantitative uncertainty analysis are summarized in Table 3-35. Annual abandoned coal mine CH4 emissions in 2016 were estimated to be between 5.5 and 8.2 MMT CO2 Eq. at a 95 percent confidence level. This indicates a range of 18 percent below to 22 percent above the 2016 emission estimate of 6.7 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 after 1972. Emissions from mines closed prior to 1972 have the largest degree of uncertainty because no mine-specific CH4 liberation rates exist. Pre-1972 mines represent 17 percent of the total abandoned mine inventory. Uncertainty and Time-Series Consistency 3-64 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2016 ------- 1 Table 3-35: Approach 2 Quantitative Uncertainty Estimates for ChU Emissions from 2 Abandoned Underground Coal Mines (MMT CO2 Eq. and Percent) Source Gas 2016 Emission Estimate (MMT CO2 Eq.) Uncertainty Range Relative to Emission Estimate3 (MMT CO2 Eq.) (%) Lower Upper Bound Bound Lower Upper Bound Bound Abandoned Underground Coal Mines CH4 6.7 5.5 8.2 -18% +22% a Range of emission estimates predicted by Monte Carlo Simulation for a 95 percent confidence interval. 3 4 Methodological recalculations were applied to the entire time series to ensure time-series consistency from 1990 5 through 2016. Details on the emission trends through time are described in more detail in the Methodology section, 6 above. 7 3.6 Petroleum Systems (CRF Source Category s lB2a) 9 Methane emissions from petroleum systems are primarily associated with onshore and offshore crude oil production, 10 transportation, and refining operations. During these activities, CH4 is released to the atmosphere as leak emissions, 11 vented emissions (including emissions from operational upsets) and emissions from fuel combustion. Leak and 12 vented CO2 emissions from petroleum systems are primarily associated with crude oil production and refining 13 operations but are negligible in transportation operations. Total CH4 emissions from petroleum systems in 2016 14 were 39.3 MMT CO2 Eq. (1,571 kt), a decrease of 7 percent from 1990. Total CO2 emissions from petroleum 15 systems in 2016 were 25.5 MMT CO2 Eq. (25,543 kt), an increase of a factor of 1.7 from 1990. 16 Exploration. Exploration includes well drilling, testing, and completions. Exploration accounts for approximately 5 17 percent of total CH4 emissions from petroleum systems. The predominant sources of emissions from exploration are 18 hydraulically fractured oil well completions and well testing. Other sources include well completions without 19 hydraulic fracturing and well drilling. Since 1990, exploration CH4 emissions have increased 168 percent due to 20 increases in the number of wells completed. Emissions of CH4 from exploration decreased 7 percent from 2015 to 21 2016. Exploration accounts for less than 1 percent of total CO2 emissions from petroleum systems. Emissions of 22 CO2 from exploration in 2016 decreased by 84 percent from 1990, and 85 percent from 2015, due to a decrease in 23 well testing flaring CO2 emissions. 24 Production. Production accounts for approximately 92 percent of total CH4 emissions from petroleum systems. The 25 predominant sources of emissions from production field operations are pneumatic controllers, offshore oil platforms, 26 oil tanks, gas engines, chemical injection pumps, associated gas venting and flaring, and leaks from oil wellheads. 27 Since 1990, CH4 emissions from production have decreased by 12 percent, due to decreases in tank emissions and in 28 associated gas venting. Overall, production segment methane emissions decreased by less than 1 percent from 2015 29 levels, although emissions from tanks increased by 53 percent, emissions from associated gas venting and flaring 30 decreased by 40 percent, and emissions from miscellaneous production flaring decreased by 45 percent in 2016 31 compared to 2015. The change in CH4 emissions from 2015 to 2016 for tanks, associated gas venting and flaring, 32 and miscellaneous production flaring reflects differences in reported GHGRP subpart W emissions levels for 33 reporting year (RY) 2016 as compared to RY2015. Production field operations account for approximately 85 percent 34 of the total CO2 emissions from petroleum systems. The principal sources of CO2 emissions are associated gas 35 flaring, oil tanks with flares, and miscellaneous production flaring. These three sources together account for over 99 36 percent of the CO2 emissions from production. 37 Crude Oil Transportation. Crude oil transportation activities account for less than 1 percent of total CH4 emissions 38 from the oil industry. Emissions from tanks, truck loading, rail loading, and marine vessel loading operations 39 account for 88 percent of CH4 emissions from crude oil transportation. Leak emissions, almost entirely from floating Energy 3-65 ------- 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 roof tanks, account for approximately 12 percent of CH4 emissions from crude oil transportation. Since 1990, CH4 emissions from transportation have increased by 27 percent. However, because emissions from crude oil transportation account for such a small percentage of the total emissions from the petroleum industry, this has had little impact on the overall emissions. Methane emissions from transportation in 2016 decreased by less than 1 percent from 2015 levels. Crude Oil Refining. Crude oil refining processes and systems account for approximately 2 percent of total CH4 emissions from the oil industry. 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 an insignificant amount of CH4 in all refined products. Within refineries, incomplete combustion accounts for 38 percent of the CH4 emissions, while vented and leak emissions account for approximately 52 and 10 percent, respectively. Flaring accounts for 82 percent of combustion CH4 emissions. Refinery system blowdowns for maintenance and process vents are the primary venting contributors (97 percent). Most of the leak CH4 emissions from refineries are from equipment leaks and storage tanks (85 percent). Methane emissions from refining of crude oil have increased by approximately 51 percent since 1990; however, similar to the transportation subcategory, this increase has had little effect on the overall emissions of CH4. From 1990 to 2015, CH4 emissions from crude oil refining fluctuated between 24 and 28 kt; in 2016, emissions increased to 37 kt as process vent emissions increased. Crude oil refining processes and systems account for approximately 15 percent of total CO2 emissions from the oil industry. Almost all (97 percent) of the CO2 from refining is from flaring. Refinery CO2 emissions increased by approximately 13 percent from 1990 to 2016. Table 3-36: ChU Emissions from Petroleum Systems (MMT CO2 Eq.) Activity 1990 2005 2012 2013 2014 2015 2016 Exploration3 0.8 1.0 2.8 3.0 3.3 2.2 2.1 Production (Total) 40.8 32.8 31.7 35.0 36.8 36.3 36.1 Pneumatic controller venting 19.1 16.6 14.3 17.2 17.9 18.0 18.5 Offshore platforms 5.3 4.6 4.7 4.7 4.7 4.7 4.7 Associated gas venting and flaring 3.5 3.0 3.4 3.2 3.6 2.7 1.6 Tanks 6.4 2.0 1.4 1.6 1.9 2.1 3.2 Gas engines 2.1 1.8 2.1 2.2 2.3 2.3 2.2 Chemical injection pumps 1.2 1.7 2.0 2.1 2.1 2.1 2.0 Other Sources 3.0 3.1 3.9 4.1 4.3 4.4 3.9 Crude Oil Transportation 0.2 0.1 0.2 0.2 0.2 0.2 0.2 Refining 0.6 0.7 0.7 0.7 0.7 0.7 0.9 Total 42.3 34.7 35.4 38.8 41.0 39.4 39.3 a Exploration includes well drilling, testing, and completions. Note: Totals may not sum due to independent rounding. Table 3-37: ChU Emissions from Petroleum Systems (kt) Activity 1990 2005 ; 2012 2013 2014 2015 2016 Exploration3 31 39 113 120 131 89 82 Production (Total) 1,631 1,314 1,269 1,399 1,474 1,451 1,443 Pneumatic controller venting 766 663 570 687 716 721 739 Offshore platforms 211 185 188 188 188 188 188 Associated gas venting and flaring 140 120 136 127 145 106 64 Tanks 258 84 57 65 77 82 127 Gas Engines 85 70 83 87 92 93 89 Chemical injection pumps 49 67 80 82 85 85 81 Other Sources 122 125 155 163 172 175 156 Crude Oil Transportation 7 5 6 7 8 8 8 Refining 24 28 27 27 26 28 37 Total 1,693 1,386 1,415 1,553 1,639 1,576 1,571 a Exploration includes well drilling, testing, and completions. Note: Totals may not sum due to independent rounding. 3-66 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2016 ------- 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 Table 3-38: CO2 Emissions from Petroleum Systems (MMT CO2) Activity 1990 2005 2012 2013 2014 2015 2016 Exploration 0.2 0.2 0.2 0.3 0.3 0.3 0.0 Production 5.9 13.1 22.0 25.8 29.2 33.7 21.8 Crude Refining 3.3 3.7 3.4 3.6 3.4 4.0 3.7 Total 9.4 17.0 25.6 29.7 32.9 38.0 25.5 Note: Totals may not sum due to independent rounding. ible 3-39: CO2 Emissions from Petroleum Systems (kt) Activity 1990 2005 2012 2013 2014 2015 2016 Exploration 243 207 247 255 264 262 39 Production 5,859 13,071 21,957 25,835 29,217 33,695 21,794 Crude Refining 3,282 3,726 3,425 3,605 3,414 4,014 3,710 Total 9,384 17,004 25,629 29,695 32,895 37,971 25,543 Note: Totals may not sum due to independent rounding. Methodology See Annex 3.5 for the Ml time series of emissions data, activity data, and emission factors, and additional information on methods and data sources. The estimates of CH4 emissions from petroleum systems are largely based on RY2010 through RY2016 GHGRP data, Drillinglnfo, EPA/GRI 1996, and EPA 1999. Petroleum systems includes emission estimates for activities occurring in petroleum systems from the oil wellhead through crude oil refining, including activities for crude oil production field operations, crude oil transportation activities, and refining operations. Annex 3.5 provides further detail on the emission estimates for these activities, including year-specific emission factor and activity data information. 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). References for emission factors include 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), Drillinglnfo (2017), consensus of industry peer review panels, Bureau of Ocean Energy Management (BOEM) reports and analysis of GHGRP data. The emission factors for pneumatic controllers and chemical injection pumps were developed using GHGRP data for reporting year 2014. The emission factors for tanks, well testing, associated gas venting and flaring, and miscellaneous production flaring were developed using GHGRP data for reporting year 2015 and 2016. Emission factors for hydraulically fractured (HF) oil well completions (controlled and uncontrolled) were developed using Drillinglnfo data analyzed for the 2015 NSPS OOOOa proposal. For offshore oil production, two emission factors were calculated using data collected for all federal offshore platforms; one for oil platforms in shallow water, and one for oil platforms in deep water. For most sources, emission factors were held constant for the period 1990 through 2016, and trends in emissions reflect changes in activity levels. For tanks, well testing, and associated gas venting and flaring, year-specific emission factors were developed for 2015 and 2016 and the 2015 emission factors were applied back to 1990. Basin-specific emission factors for associated gas venting and flaring were developed and applied for the four basins that in any year from 2011 through 2016 contributed at least 10 percent of total emissions (on a CO2 Eq. basis) from associated gas venting and flaring in the GHGRP: Williston, Permian, Gulf Coast, and Anadarko basins. Associated gas venting and flaring data in all other basins were combined, and emission factors and activity factors developed for the other basins as a single group. For more information, see Recalculations Discussion below. For miscellaneous production flaring, year-specific emission factors were developed for 2015 and 2016, an emission factor of 0 was assumed for 1990 through 1992, and linear interpolation was applied to develop emission factors for 1993 through 2014. Emission factors from EPA 1999 are used for all other production and transportation activities. Energy 3-67 ------- 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 References for activity data include Drillinglnfo (2017), 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 the GHGRP (RY2010 through RY2016). 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. For floating roof tanks, the activity data were held constant from 1990 through 2016 based on EPA 1999. 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, the previous year's data were used for domestic barges and tankers as current year were not yet available. For offshore production, the number of platforms in shallow water and the number of platforms in deep water are used as activity data and are taken from BOEM datasets. For the production segment, in general, CO2 emissions for each source are estimated with GHGRP data or by multiplying CO2 emission factors by the corresponding CH4 data, as the CO2 content of gas relates to the CH4 content of gas. Sources with CO2 emissions calculated from GHGRP data are associated gas venting and flaring, tanks, well testing, pneumatic controllers, chemical injection pumps, and miscellaneous production flaring. For these sources, CO2 was calculated using the same methods as used for CH4. Emission factors for offshore oil production (shallow and deep water) were derived using data from BOEM. 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 petroleum refining activities, 2010 to 2016 emissions were directly obtained from EPA's GHGRP. All U.S. refineries have been required to report CH4 and CO2 emissions for all major activities starting with emissions that occurred in 2010. However, GHGRP does have provisions that refineries are not required to report to the GHGRP if their emissions fall below certain thresholds (see Planned Improvements for additional discussion). The reported total of CH4 and CO2 emissions for each activity was used for the 2010 to 2016 emissions. 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), which were then applied to the annual refinery feed to estimate CH4 and CO2 emissions for 1990 to 2009. A complete list of references for emission factors and activity data by emission source is provided in Annex 3.5. Through EPA's stakeholder process on oil and gas in the Inventory, EPA received initial stakeholder feedback on updates under consideration for the Inventory. Stakeholder feedback is noted below in Uncertainty and Time -Series Consistency, Recalculations Discussion, and Planned Improvements. Uncertainty and Time-Series Consistency In recent years, EPA has made significant revisions to the Inventory methodology to use updated activity and emissions data. To update its characterization of uncertainty, EPA has conducted a draft quantitative uncertainty analysis using the IPCC Approach 2 methodology (Monte Carlo Simulation technique). The 95 percent confidence intervals presented here are based on 2015 data from the previous (i.e., 1990 through 2015) Inventory. EPA is still seeking comment on the approach to calculate uncertainty and may update its approach in the final version of the current Inventory. Initial stakeholder feedback on the uncertainty analysis included support for annual updates to the uncertainty assessment, so that the uncertainty ranges will continue to reflect new data as they become available. Stakeholders supported the approach of calculating uncertainty for the top emitters. For more information, please see the Planned Improvements section, and the memorandum Inventory of U.S. Greenhouse Gas Emissions and Sinks 1990-2016: Updates Under Consideration for Natural Gas and Petroleum Systems Uncertainty Estimates (Draft 2018 Uncertainty Memo)?9 79 See 3-68 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2016 ------- 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 To develop the values in Table 3-40 below, EPA applied the uncertainty bounds calculated for the 2015 emission estimates presented in the previous Inventory. To develop the uncertainty bounds, EPA used the IPCC Approach 2 methodology (Monte Carlo Simulation technique). Microsoft Excel's @RISK add-in tool was used to estimate the 95 percent confidence bound around methane emissions from petroleum systems. For the analysis, EPA focused on the five highest methane-emitting sources for the year 2015, which together emitted 79 percent of methane from petroleum systems in 2015, and extrapolated the estimated uncertainty for the remaining 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. 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 provide the 95 percent confidence bound within which actual emissions from this source category are likely to fall for the year 2016, using the recommended IPCC methodology. The results of the Approach 2 uncertainty analysis are summarized in Table 3-40. Petroleum systems CH4 emissions in 2016 were estimated to be between 26.7 and 53.4 MMT CO2 Eq., while CO2 emissions were estimated to be between 17.3 and 34.7 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. Table 3-40: Approach 2 Quantitative Uncertainty Estimates for ChU Emissions from Petroleum Systems (MMT CO2 Eq. and Percent) Source Gas 2016 Emission Estimate (MMT CO2 Eq.)b Uncertainty Range Relative to Emission Estimate3 (MMT CO2 Eq.) (%) Lower Upper Bound Bound Lower Upper Bound Bound Petroleum Systems CH4 39.3 26.7 53.4 -32% +36% Petroleum Systems CO2 25.5 17.3 34.7 -32% +36% a Range of emission estimates estimated by applying the 95 percent confidence intervals obtained from the Monte Carlo Simulation analysis conducted for the year 2015. 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. c An uncertainty analysis for the non-energy CO2 emissions was not performed. The relative uncertainty estimated (expressed as a percent) from the CH4 uncertainty analysis was applied to the point estimate of non-energy CO2 emissions. 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 1990 through 2016, 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 2014 by interpolating activity data or emission factors or both between 1992 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 Discussion above. QA/QC and Verification Discussion 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 Energy 3-69 ------- 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 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. EPA held stakeholder workshops on greenhouse gas data for oil and gas in June and October of 2017, and held webinars in April and August of 2017. In advance of each workshop, EPA released memos detailing updates under consideration and requesting stakeholder feedback. Stakeholder feedback received through these processes is discussed in the Recalculations Discussion and Planned Improvements sections below. 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 (i.e., the independent study should assess data from the Inventory and not another data set, such as EDGAR). In an effort to improve the ability to compare the national-level Inventory with measurement results that may be at other scales, a team at Harvard University along with EPA and other coauthors 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 gridded methane inventory is designed to be consistent with the U.S. EPA Inventory of U.S. Greenhouse Gas Emissions and Sinks (1990-2014) estimates for the year 2012, which presents national totals.82 Recalculations Discussion The EPA received information and data related to the emission estimates through GHGRP reporting, the annual Inventory formal public notice periods, stakeholder feedback on updates under consideration, and new studies. In June and October 2017, the EPA released draft memoranda: Inventory of U.S. Greenhouse Gas Emissions and Sinks 1990-2016: Revisions Under Consideration for CO2 Emissions {Draft 2018 CO 2 Memo),^ and Inventory of U.S. Greenhouse Gas Emissions and Sinks 1990-2016: Additional Revisions Under Consideration {Draft 2018 Other Updates Memo).84 The memos discussed changes under consideration, and requested stakeholder feedback on those changes. The EPA thoroughly evaluated relevant information available, and made updates to exploration and production segment methodologies for the Inventory, including to define an exploration segment separate from production (not a methodological change, but a change in presentation of information), revising activity and CH4 and CO2 emissions data for associated gas venting and flaring, miscellaneous production flaring, and well testing. Production segment CO2 emissions data were also revised for oil tanks, pneumatic controllers, and chemical injection pumps. 80 See 81 See 82 See 83 See < https://www. epa.gov/sites/production/files/2017-10/documents/2018_ghgi_co2_revisions_under_consideration_2017- 10-25_to_post.pdf> 84 See < https://www.epa.gov/sites/production/files/2017-10/documents/2018_ghgi_ng- petro_revisions_under_consideration_2017- 10-26_pdf_to_post.pdf 3-70 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2016 ------- 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 The combined impact of revisions to 2015 petroleum systems CH4 emissions, compared to the previous Inventory, is a decrease from 39.9 to 39.4 MMT CO2 Eq. (0.5 MMT CO2 Eq., or 1 percent). The recalculations resulted in an average decrease in CH4 emission estimates across the 1990 through 2015 time series, compared to the previous Inventory, of 11 MMT CO2 Eq, or 22 percent. The CH4 emissions estimate decrease was primarily due to recalculations related to associated gas venting and flaring which were updated to use a basin-level approach, and has the largest impact on years prior to 2013. The combined impact of revisions to 2015 petroleum systems CO2 emissions, compared to the previous Inventory, is an increase from 3.6 to 38.0 MMT CO2 (34.4 MMT CO2, or by a factor of 9). The recalculations resulted in an average increase in emission estimates across the 1990 through 2015 time series, compared to the previous Inventory, of 13.8 MMT CO2 Eq, or 360 percent. The CO2 emissions estimate increase was primarily due to recalculations related to the reallocation of CO2 from flaring to petroleum systems from natural gas systems. Previously, data were not available to disaggregate flared emissions between natural gas systems and petroleum systems. The largest sources of CO2 from flaring are associated gas flaring, tanks with flares, and miscellaneous production flaring. Exploration Petroleum systems was reorganized for the current Inventory to include an exploration segment to improve conformance with the IPCC guidelines. Exploration activities were previously included under the production segment. The activities included under exploration are hydraulically fractured oil well completions, oil well completions without hydraulic fracturing, well drilling, and well testing. Of these activities, well testing was the only source with a new methodology, which is discussed below. Well Testing EPA developed a new estimate for oil well testing (during non-completion events) using GHGRP data. In previous Inventories, only well testing conducted as part of a completion event was included. CH4 and CO2 emission factors were developed, on a per-event basis, for vented and flared oil well testing events using RY2015 and RY2016 data. EPA developed activity factors (i.e., number of events per oil well) to determine the number of well testing events in a year, also using RY2015 and RY2016 data. GHGRP RY2015 activity and emission factors are applied to all prior years of the time series. Methane emissions from well testing averaged 8.1 kt (or 0.2 MMT CO2 Eq.) over the time series. There was a large decrease in emissions from oil well testing from 2015 to 2016 as observed in reported GHGRP data. Carbon dioxide emissions from well testing averaged 216 kt (0.2 MMT CO2) over the time series. See the Draft 2018 Other Updates Memo for additional discussion. Table 3-41: Oil Well Testing National ChU Emissions (Metric Tons ChU) Source 1'WO 2005 2012 2013 2014 2015 2016 Non-Completion Well Testing - Vented 8.043 6,819 8,022 8,272 8,559 8,567 2,811 Non-Completion Well Testing - Flared %l 815 959 989 1,023 1,024 157 Table 3-42: Oil Well Testing National CO2 Emissions (Metric Tons CO2) Source I'J'JO 2005 2012 2013 2014 2015 2016~ Non-Completion Well Testing- ^ 3Q8 363 ^ ^ ^ ^ Vented Non-Completion Well Testing- 241.362 204,643 240,754 248,234 256,853 257,101 34,481 r lared Production In addition to the memos discussed above, this section references the memorandum, Inventory of U.S. Greenhouse Gas Emissions and Sinks 1990-2015: Revisions for Natural Gas and Petroleum Systems Production Emissions Energy 3-71 ------- 1 (2017 Production Memo).85 The 2017 Production Memo contains further details and documentation of 2 recalculations. 3 CO2 Updates 4 EPA updated CO2 emissions for a number of sources in the Inventory. See the Draft 2018 CO2 Memo for more 5 details. The overall impact was an average increase of 13.8 MMT CO2 (or 364 percent) over the time series in 6 petroleum systems, which is primarily due to the reallocation of flaring CO2 emissions from natural gas systems to 7 petroleum systems, which was not possible in the past because the previous data source aggregated venting and 8 flaring activity data from both petroleum and natural gas systems, but is now possible through use of the GHGRP 9 data. A stakeholder noted that the update uses the best available data for this source. 10 Sources with the largest impacts include tanks with flares, associated gas flaring, and miscellaneous production 11 flaring. These sources are discussed in detail below. Other sources (i.e., pneumatic controllers and chemical 12 injection pumps) had increases or decreases of less than 1 MMT CO2. 13 Tanks 14 EPA developed CO2 emissions estimates for oil tanks using GHGRP data and a throughput-based approach. This 15 approach is identical to the methodology to calculate CH4 emissions; for more information, please see the 2017 16 Production Memo. The overall impact of the change is an increase in calculated CO2 emissions by a factor of nine 17 on average over the time series. 18 Table 3-43: National Tank CO2 Emissions by Category and National Emissions (kt CO2) CO2 Emissions mo 2005 2012 2013 2014 2015 2016 Large Tanks w/ Flares 0 3,407 5,978 6,870 8,054 8,657 7,282 Large Tanks w/ VRU 0 6 11 13 15 16 11 Large Tanks w/o Control 25 " 4 5 6 6 9 Small Tanks w/ Flares 0 4 7 8 9 10 21 Small Tanks w/o Flares 9 6 7 8 8 7 Malfunctioning Dump Valves 20 14 17 20 23 25 22 Total Emissions 53 3,444 6,023 6,922 8,115 8,722 7,351 Previous Estimated Emissions 329 247 366 433 520 520 NA NA (Not Applicable) 19 Associated Gas Venting and Flaring 20 EPA developed a new estimate for CO2 from associated gas venting and flaring. EPA's considerations for this 21 source are documented in the Draft 2018 C02 Memo. As noted above in the Methodology section, EPA used a 22 basin-level and well-based approach to calculate emissions from this source. In the Draft 2018 C02 Memo, a NEMS 23 Region-level and well-based approach was presented, however, stakeholder feedback on the Draft 2018 C02 Memo 24 supported the use of GHGRP data to calculate emissions from this source at a basin-level. EPA evaluated basin- 25 level associated gas venting and flaring data reported to GHGRP from 2011 to 2016 and developed the estimates 26 below with that approach. If a basin contributed at least 10 percent of total annual emissions (on a CO2 Eq. basis) 27 from associated gas venting and flaring in any year, then basin-specific emission factors and activity factors were 28 developed. Four basins met this criteria: Williston, Permian, Gulf Coast, and Anadarko. Associated gas venting and 29 flaring data in all other basins were combined, and emission factors and activity factors developed for the other 30 basins as a single group. For each basin or group, emission factors were calculated for 2015 and 2016; the 2015 31 emission factors were applied to all prior years. Two activity factors were also calculated for each basin or group: 32 the percent of oil wells that either flare or vent associated gas and, for those wells in that category (those that vent or 33 flare associated gas), the fraction that vents and the fraction that flares. The percent of oil wells that flare or vent 34 associated gas was calculated for 2015 and 2016, and the 2015 activity factors applied to all prioryears. The specific 35 fractions that vent and flare associated gas were developed for 2011 through 2016, and the 2011 fractions applied to 85 Available at 3-72 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2016 ------- 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 all prior years. Stakeholders also suggested a production-based approach for the basin-level analysis, and EPA will continue evaluating this approach for the final version of the current Inventory (as noted in the Planned Improvements section). Stakeholders have noted that past (e.g., 1990 through 2010) associated gas venting and flaring likely varied significantly from year to year and from region to region. However, data are not presently available to take variation prior to 2011 into account. Table 3-44: Associated Gas Venting and Flaring National CO2 Emissions (kt CO2) Source Associated Gas Well Venting ^ ^ ^ Emissions Associated Gas Well Flaring . Emissions "' Previous Estimated emissions j from stripper wells 2005 2012 2013 2014 2015 2016 95 89 75 83 41 19 5,102 8,820 11,582 13,510 17,414 10,137 I 1 1 1 1 NA NA (Not Applicable) The CH4 methodology was previously developed for the previous Inventory and used a national-level approach. EPA updated its CH4 calculations for associated gas venting and flaring to be consistent with the basin-level approach to calculate CO2 emissions from this source. Overall, the change decreased calculated CH4 emissions over the time series by around 70 percent for both associated gas venting and associated gas flaring, with the largest decreases occurring early in the time series. Table 3-45: Associated Gas Venting and Flaring National ChU Emissions (Metric Tons ChU) Source Two 2005 2012 2013 2014 2015 2016 Associated Gas Well Venting Emissions Associated Gas Well Flaring Emissions Previous Associated Gas Well Venting Emissions Previous Associated Gas Well Flaring Emissions 119.697 101,157 105,312 87,013 97,477 44,636 31,040 20.769 18,380 31,024 40,203 47,101 61,576 33,015 608,758 511,701 482,816 214,665 89,333 42,518 NA 76,176 64,031 104,513 146,292 149,694 105,706 NA NA (Not Applicable) Miscellaneous Production Flaring The EPA developed new estimates for CO2 and CH4 emissions from miscellaneous production flaring using GHGRP subpart W data. Along with other updates to flaring emissions in both oil and gas production, this replaces the estimate for flaring that was previously reported in the natural gas systems emissions totals. EPA developed emission factors from 2015 and 2016 GHGRP data; the 2015 emission factor is applied to all prior years. The emission factors are on a per-well basis and were applied to all oil wells in each year. Details are provided in the Draft 2018 CO2 Memo. Initial stakeholder feedback on this update suggested use of production-based emission factors as opposed to well-based emission factors. Table 3-46: Miscellaneous Production Flaring National CO2 Emissions (kt CO2) Source 19911 2005 2012 2013 2014 2015 2016 Miscellaneous Production Flaring Previous Estimated emissions from flaring (natural gas and petroleum)" 0 4,349 6,944 7,160 7,409 7,416 4,183 9,093 7,193 12,704 15,684 17,629 17,629 NA a The previous estimated emissions from flaring were reported under Natural Gas Systems and included emissions from multiple sources, including associated gas, and natural gas systems, but is provided for reference. NA (Not Applicable) Energy 3-73 ------- 1 Table 3-47: Miscellaneous Production Flaring National ChU Emissions (Metric Tons ChU) Source 1990 2005 2012 2013 2014 2015 2016 Miscellaneous Production Flaring Previous Estimated emissions from flaring" I) 14,889 23,773 24,511 25,362 25,387 13,844 0 0 0 0 0 0 NA a Prior Inventories did not estimate methane emissions from a source similar to miscellaneous production flaring. NA (Not Applicable) 2 Activity Data Updates 3 Well Counts 4 EPA has used a more recent version of the Drillinglnfo data set to update well counts data in the Inventory. There 5 are not methodological changes to this source in the current Inventory or major changes to the activity data, but 6 because this is a key input, results are highlighted here. 7 Table 3-48: Producing Oil Well Count Data Oil Well Count 1990 2005 2012 2013 2014 2015 2016 Number of Oil Wells Previous Estimate 553,899 572,639 469,632 481,340 552,504 564,348 569,670 580,960 589,450 598,627 590,017 586,896 561,964 NA NA (Not Applicable) 8 In December 2017, EIA released a 2000 through 2016 time series of national oil and gas well counts. EIA total (oil 9 and gas) well counts for 2016 were 1,010,441. EPA's total well counts were 978,845. Over the 2000 to 2016 time 10 series, EPA's well counts were on average 2 percent lower than EIA's. EIA's well counts include side tracks, 11 completions, and recompletions, and therefore are expected to be higher than EPA's which include only producing 12 wells. EPA and EIA use a different threshold for distinguishing between oil versus gas (EIA uses 6 mcf/bbl, while 13 EPA uses 100 mcf/bbl), which results in EIA having a lower fraction of oil wells and a higher fraction of gas wells 14 than EPA. Across the 2000 through 2016 EIA time series, EIA estimates on average 111,420 (or 20 percent) fewer 15 oil wells in each year than EPA. 16 Equipment Counts 17 EPA recalculated activity factors of equipment per well using the latest GHGRP RY2015 data, which included some 18 resubmissions. This resulted in minor changes across the time series. For example, the number of heater/treaters per 19 well decreased by 9 percent over the time series, the number of separators and headers per well decreased by 4 20 percent and 3 percent, respectively, while chemical injection pumps and pneumatic controllers per well increased by 21 4 percent and less than 1 percent, respectively. The impact of the changes in equipment counts per well along with 22 changes in well counts resulted in minor changes in methane emissions across the time series for heater/treaters (-12 23 percent), separators (17 percent), headers (-5 percent), pneumatic controllers (-2 percent), and chemical injection 24 pumps (2 percent). 25 Transportation 26 Recalculations due to updated activity data for quantity of petroleum transported by barge or tanker in the 27 transportation segment have resulted in an average decrease in calculated emissions over the time series from this 28 segment of less than 0.01 percent. 29 Refining 30 Recalculations due to resubmitted GHGRP data-in particular from flaring in the refining segment have resulted in an 31 average increase in calculated CH4 emissions over the time series from this segment of 3 percent and an average 32 increase in calculated CO2 emissions over the time series of 6 percent. 3-74 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2016 ------- 1 Planned improvements 2 Plans for Final 2018 Inventory (1990 through 2016) and Future Inventories 3 Associated Gas Venting and Flaring 4 For the final version of the current Inventory, EPA is considering applying a production-based approach at the 5 basin-level, as opposed to a well-based approach at the basin-level, to calculate emissions for this source based on 6 stakeholder feedback. Preliminary analysis of emissions calculated using a basin-level, production-based approach 7 indicates that the emission estimates will be lower than those calculated with the basin-level, well-based approach. 8 EPA applied a basin-level approach to avoid potential overestimates of venting and flaring activity that might occur 9 by applying data from GHGRP at a national level without basin-specific adjustments. EPA seeks feedback on this 10 approach and how other approaches might avoid over- or underestimating emissions from this source. 11 Uncertainty 12 The uncertainty analysis results presented for this public review Inventory were based on the top five methane- 13 emitting sources for 2015 from the previous Inventory. EPA will re-evaluate the highest emitting sources, based on 14 the final version of the current Inventory, and update the uncertainty analysis to reflect these sources and their 15 methodology, as necessary. EPA will also consider further stakeholder feedback on the Draft 2018 Uncertainty 16 Memo. 17 Miscellaneous Production Flaring 18 Miscellaneous production flaring emission factors are currently applied on a well-basis at the national-level. EPA is 19 considering two additional options for the final version of the current Inventory, based on stakeholder feedback: a 20 production-based approach and developing factors at a basin-level. Each of these options are being considered to 21 avoid over- or underestimating emissions from this source. 22 Refineries 23 The GHGRP includes provisions at 40 CFR 98.2(i) that allows facilities to discontinue complying with the GHGRP 24 if their emissions fall below certain thresholds. EPA is assessing to what extent this provision has affected the 25 subpart Y reported emissions. If certain refineries are not reporting emissions to the GHGRP, options to address this 26 will be considered. 27 Offshore Platforms 28 EPA is considering updates to the offshore platform emissions calculation methodology, as discussed in the Draft 29 2018 Other Updates Memo. The current emission factors were based on data from the 2011 DOI/Bureau of Ocean 30 Energy Management's (BOEM) Gulf Offshore Activity Data System (GOADS), and 2014 GOADS data is 31 available. A different source for platform counts is also being considered. 32 N2() Emissions 33 N20 emissions are currently not included in petroleum systems estimates, but EPA is considering developing a 34 methodology to estimate N2O emissions. The Draft 2018 Other Updates Memo provides discussion on this topic. 35 EPA will consider options such as using GHGRP data for sources that already rely on GHGRP data for CH4 or CO2 36 estimates. GHGRP RY2015 reported N20 flaring emissions specific to petroleum systems were 124 metric tons (or 37 0.04 MMT CO2 Eq. In addition, 36 metric tons N2O (or 0.01 MMT CO2 Eq.) flaring emissions were reported for 38 GHGRP RY2015 for sources that fall within both natural gas and petroleum systems. 39 Well-Related Activity Data Energy 3-75 ------- 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 As described in the Recalculations Discussion EPA lias updated the emission factors for certain well-related emission sources, including well testing. EPA will continue to assess available data, including data from the GHGRP and stakeholder feedback on considerations, to improve activity estimates for these types of sources. Upcoming Data, and Additional Data that Could Inform the Inventory EPA will continue to review data available from the GHGRP, in particular new data on hydraulically fractured oil well completions and workovers and new well-specific information, available in 2017 for the first time (for RY2016). EPA will consider revising its methods to take into account the new GHGRP data. EPA will assess new data received by the Methane Challenge Program 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, such as an upcoming field study by API on pneumatic controllers. 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 stakeholder comments. EPA also continues to seek new data that could be used to assess or update the estimates in the Inventory. For example, stakeholder comments have highlighted areas where additional data that could inform the Inventory are currently limited or unavailable: • Tank malfunction and control efficiency data. See Tanks in Recalculations Discussion. • Activity data and emissions data for production facilities that do not report to GHGRP. • Associated gas venting and flaring data on practices from 1990 through 2010. See Associated Gas Venting and Flaring in Recalculations Discussion. • Refineries emissions data. One stakeholder noted a recent study (Lavoie et al. 2017) that measured three refineries and found higher average emissions than in the Inventory, and the stakeholder suggested that EPA evaluate the study and any additional information available on this source. One stakeholder suggested that the Inventory should be updated with site-level and basin-level data, noting the EPA could first use basin-level data to assess the inventory, and that future research could focus on collecting data in basins with the largest discrepancies. EPA will continue to seek available data on these and other sources as part of the process to update the Inventory. Box 3-7: 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. 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 from naturally-produced CO2 are estimated based 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. The naturally-occurring CO2 used in EOR operations is assumed to be fully sequestered. Additionally, all anthropogenic CO2 emitted from natural gas processing and ammonia plants is assumed to be emitted to the atmosphere, regardless of whether the CO2 is captured or not. These emissions are currently included in the Natural Gas Systems and the Ammonia Production sections of the Inventory report, respectively. 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. 3-76 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2016 ------- 1 In the United States, facilities that produce CO2 for various end-use applications (including capture facilities such as 2 acid gas removal plants and ammonia plants), importers of CO2, exporters of CO2, facilities that conduct geologic 3 sequestration of CO2, and facilities that inject CO2 underground, are required to report greenhouse gas data annually 4 to EPA through its GHGRP. Facilities conducting geologic sequestration of CO2 are required to develop and 5 implement an EPA-approved site-specific monitoring, reporting and verification plan, and to report the amount of 6 CO2 sequestered using a mass balance approach. 7 GHGRP data relevant for this inventory estimate consists of national-level annual quantities of CO2 captured and 8 extracted for EOR applications for 2010 to 2016. However, for 2015 and 2016, data from EPA's GHGRP (Subpart 9 PP) were unavailable for use in the current Inventory report due to data confidentiality reasons. The estimate for 10 2014 was held constant here to estimate 2015 and 2016 emissions. EPA will continue to evaluate the availability of 11 additional GHGRP data and other opportunities for improving the emission estimates. For reporting year 2016, one 12 facility reported data to the GHGRP under subpart RR (Geologic Sequestration of Carbon Dioxide). This facility 13 reported 3.1 MMT of CO2 sequestered in subsurface geological formations and 56 metric tons of CO2 emitted from 14 surface equipment leaks and vents. 15 These estimates indicate that the amount of CO2 captured and extracted from natural and industrial sites for EOR 16 applications in 2016 is 59.3 MMT CO2 Eq. (59,318 kt) (see Table 3-49 and Table 3-50). Site-specific monitoring 17 and reporting data for CO2 injection sites (i.e., EOR operations) were not readily available, therefore, the quantity of 18 CO2 captured and extracted is noted here for information purposes only; CO2 captured and extracted from industrial 19 and commercial processes is assumed to be emitted and included in emissions totals from those processes. 20 Table 3-49: Quantity of CO2 Captured and Extracted for EOR Operations (MMT CO2) Stage 1990 2005 2012 2013 2014 2015 2016 Capture Facilities 4.8 6 9.3 12.2 13.1 13.1 13.1 Extraction Facilities 20.8 28.3 48.9 47.0 46.2 46.2 46.2 Total 25.6 34.7 58.1 59.2 59.3 59.3 59.3 Note: Totals may not sum due to independent rounding. 21 Table 3-50: Quantity of CO2 Captured and Extracted for EOR Operations (kt) Stage 1990 2005 2012 2013 2014 2015 2016 Capture Facilities 4,832 6,475 9,267 12,205 13,093 13,093 13,093 Extraction Facilities 20,811 28,267 48,869 46,984 46,225 46,225 46,225 Total 25,643 34,742 58,136 59,189 59,318 59,318 59,318 Note: Totals may not sum due to independent rounding. 22 23 3.7 Natural Gas Systems (CRF Source Category 24 lB2b) 25 The U.S. natural gas system encompasses hundreds of thousands of wells, hundreds of processing facilities, and 26 over a million miles of transmission and distribution pipelines. Overall, natural gas systems emitted 162.1 MMT 27 CO2 Eq. (6,483 kt) of CH4 in 2016, a 16 percent decrease compared to 1990 emissions, and a 1.4 percent decrease 28 compared to 2015 emissions (see Table 3-51, Table 3-52, and Table 3-53) and 26.7 MMT CO2 Eq. (26,739 kt) of 29 non-combustion CO2 in 2016, a 10 percent decrease compared to 1990 emissions. 30 The 1990 to 2016 trend in CH4 is not consistent across segments. Overall, the 1990 to 2016 decrease in CH4 31 emissions is due primarily to the decrease in emissions from distribution (75 percent decrease), transmission and 32 storage (44 percent decrease), processing (48 percent decrease), and exploration (81 percent decrease) segments. 33 Over the same time period, the production segments saw increased methane emissions of 60 percent (with onshore Energy 3-77 ------- 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 production emissions increasing 30 percent, offshore production emissions increasing 7 percent, and gathering and boosting emissions increasing 103 percent). The 1990 to 2016 decrease in CO2 is due primarily to decreases in acid gas removal emissions in the processing segment, where acid gas removal emissions per plant have decreased over time. Methane and non-combustion 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, bleed and discharge emissions from pneumatic controllers, and fugitive 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. Below is a characterization of the five major stages of the natural gas system. Each of the stages is described and the different factors affecting CH4 and non-combustion CO2 emissions are discussed. Exploration. Exploration includes well drilling, testing, and completions. Emissions from exploration account for less than 1 percent of CH4 emissions and 2 percent of non-combustion CO2 emissions from natural gas systems in 2016. Well completions account for most of the CH4 emissions in 2016, with well testing and drilling also contributing emissions. Flaring emissions account for most of the non-combustion CO2 emissions. Methane emissions from exploration decreased by 81 percent from 1990 to 2016, with the largest decreases coming from hydraulically fractured gas well completions without reduced emissions completions (RECs) or flaring. Carbon dioxide emissions from exploration increased by 74 percent from 1990 to 2016 due to increases in flaring. Production (including gathering and boosting). In the production stage, wells are used to withdraw raw gas from underground formations. Emissions arise from the wells themselves, and well-site gas treatment equipment such as dehydrators and separators. 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). Emissions from production (including gathering and boosting) account for 66 percent of CH4 emissions and 15 percent of non-combustion CO2 emissions from natural gas systems in 2016. Emissions from gathering stations, pneumatic controllers, gas engines, liquids unloading, and offshore platforms account for most of the CH4 emissions in 2016. Flaring emissions account for most of the non-combustion CO2 emissions with the highest emissions coming from miscellaneous flaring, flaring from tanks, offshore flaring, and flaring at workovers. Methane emissions from production increased by 60 percent from 1990 to 2016, due primarily to increases in emissions from gathering and boosting stations (driven by an increase in gas production), increases in emissions from pneumatic controllers (due to an increase in the number of controllers, particularly in the number of intermittent bleed controllers), and gas engines. Carbon dioxide emissions from production increased by a factor of 4.9 from 1990 to 2016 due to increases in flaring. Processing. In this stage, 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. Fugitive CH4 emissions from compressors, including compressor seals, are the primary emission source from this stage. Most of the non-combustion CO2 emissions come from acid gas removal (AGR) units, which are designed to remove CO2 from natural gas. Processing plants account for 7 percent of CH4 emissions and 82 percent of non-combustion CO2 emissions from natural gas systems. Methane emissions from processing decreased by 48 percent from 1990 to 2016 as emissions from compressors (leaks and venting) and equipment leaks decreased. Carbon dioxide emissions from processing decreased by 22 percent from 1990 to 2016, due to a decrease in acid gas removal emissions. 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 CH4 emissions from these compressor stations, and venting from pneumatic controllers account for most of the emissions from this stage. Uncombusted engine exhaust and pipeline venting are also sources of CH4 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). In 2016, emissions from the final months of the Aliso Canyon leak event in Southern California contributed 0.5 MMT CO2 Eq. to transmission and storage emissions, around 2 percent of total emissions for this segment. Compressors and dehydrators are the primary contributors to emissions from storage. Methane emissions from the transmission and storage sector account for 3-78 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2016 ------- 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 approximately 20 percent of emissions from natural gas systems, while CO2 emissions from transmission and storage account for less than 1 percent of the non-combustion CO2 emissions from natural gas systems. CH4 emissions from this source decreased by 44 percent from 1990 to 2016 due to reduced compressor station emissions (including emissions from compressors and leaks). CO2 emissions from transmission and storage have decreased by 14 percent from 1990 to 2016, also due to reduced compressor station emissions. 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 1,284,241 miles of distribution mains in 2016, an increase of over 340,000 miles since 1990 (PHMSA 2017a; PHMSA 2017b). Distribution system emissions, which account for 7 percent of CH4 emissions from natural gas systems and less than 1 percent of non-combustion CO2 emissions, 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 CH4 and CO2 emissions from this stage, as have station upgrades at metering and regulating (M&R) stations. Distribution system CH4 emissions in 2016 were 75 percent lower than 1990 levels (changed from 43.5 MMT CO2 Eq. to 11.0 MMT CO2 Eq.), while distribution CO2emissions in 2016 were 72 percent lower than 1990 levels (CO2 emission from this segment are less than 0.1 MMT CO2 Eq. across the time series). Total CH4 emissions for the five major stages of natural gas systems are shown in MMT CO2 Eq. (Table 3-51) and kt (Table 3-52). Table 3-53 provides additional information on how the estimates in Table 3-49 were calculated. With recent updates to the Inventory, most emissions are calculated using a net emission approach. However, certain sources are still calculated with a potential emission approach. Table 3-53 shows the calculated potential CH4 release (i.e., potential emissions before any controls are applied) from each stage, and the amount of CH4 that is estimated to have been flared, captured, or otherwise controlled, and therefore not emitted to the atmosphere. Subtracting the value for CH4 that is controlled, from the value for calculated potential release of CH4, results in the total net emissions values. More disaggregated information on potential emissions and emissions is available in Annex 3.6. See Methodology for Estimating CH4 and CO2 Emissions from Natural Gas Systems. Table 3-51: ChU Emissions from Natural Gas Systems (MMT CO2 Eq.)a Stage 1990 2005 2012 2013 2014 2015 2016 Exploration6 3.5 9.7 3.5 2.6 2.6 1.0 0.7 Production 66.7 85.9 104.1 104.2 107.2 107.4 106.6 Onshore Production 34.8 47.7 49.6 48.8 47.8 45.6 45.2 Offshore Production 3.5 4.3 3.8 3.8 3.8 3.8 3.8 Gathering and Boosting0 28.4 33.8 50.7 51.6 55.6 58.1 57.7 Processing 21.3 11.6 10.0 10.8 11.0 11.0 11.2 Transmission and Storage 58.6 30.8 27.9 30.8 32.2 34.0 32.7 Distribution 43.5 22.1 11.3 11.2 11.2 11.0 11.0 Total 193.7 160.0 156.8 159.6 164.2 164.4 162.1 a These values represent CH4 emitted to the atmosphere. CH4 that is captured, flared, or otherwise controlled (and not emitted to the atmosphere) has been calculated and removed from emission totals. b Exploration includes well drilling, testing, and completions. c Gathering and boosting includes gathering and boosting stations, gathering pipeline leaks, and gathering and boosting station episodic events. Note: Totals may not sum due to independent rounding. Table 3-52: ChU Emissions from Natural Gas Systems (kt)a Stage 1990 2005 2012 2013 2014 2015 2016 Explorationb 142 387 138 105 104 39 26 Production 2,669 3,435 4,165 4,169 4,288 4,296 4,264 Onshore Production 1,392 1,907 1,985 1,954 1,912 1,822 1,806 Offshore Production 141 173 151 151 151 151 151 Gathering and Boosting0 1,136 1,354 2,029 2,064 2,226 2,324 2,307 Processing 853 463 401 430 441 441 448 Transmission and Storage 2,343 1,230 1,117 1,232 1,287 1,360 1,306 Distribution 1,741 884 451 450 447 441 439 Total 7,748 6,399 6,273 6,385 6,568 6,578 6,483 Energy 3-79 ------- a These values represent CH4 emitted to the atmosphere. CH4 that is captured, flared, or otherwise controlled (and not emitted to the atmosphere) has been calculated and removed from emission totals. b Exploration includes well drilling, testing, and completions. c Gathering and boosting includes gathering and boosting stations, gathering pipeline leaks, and gathering and boosting station episodic events. Note: Totals may not sum due to independent rounding. 1 Table 3-53: Calculated Potential CH4 and Captu red/Com busted Cm from Natural Gas 2 Systems (MMT CO2 Eq.) 1990 2005 2012 2013 2014 2015 2016 Calculated Potential3 193.7 179.8 176.5 178.1 182.6 182.9 180.3 Exploration 3.5 9.7 3.5 2.6 2.6 1.0 0.7 Production 66.7 92.1 113.3 113.3 116.3 116.5 115.5 Processing 21.3 11.6 10.0 10.8 11.0 11.0 11.2 Transmission and Storage 58.6 43.2 37.3 39.1 40.5 42.3 40.9 Distribution 43.5 | 23.3 12.4 12.3 12.2 12.0 12.0 Captured/Combusted NA i 19.8 19.6 18.4 18.4 18.4 18.4 Exploration NA i NA NA NA NA NA NA Production NA 6.2 9.1 9.1 9.1 9.1 8.9 Processing NA NA NA NA NA NA NA Transmission and Storage NA 12.4 9.4 8.3 8.3 8.3 8.3 Distribution NA 1.2 1.1 1.0 1.0 1.0 1.0 Net Emissions 193.7 160.0 156.8 159.6 164.2 164.4 162.1 Exploration 3.5 9.7 3.5 2.6 2.6 1.0 0.7 Production 66.7 85.9 104.1 104.2 107.2 107.4 106.6 Processing 21.3 11.6 10.0 10.8 11.0 11.0 11.2 Transmission and Storage 58.6 30.8 27.9 30.8 32.2 34.0 32.7 Distribution 43.5 22.1 11.3 11.2 11.2 11.0 11.0 a In this context, "potential" means the total emissions calculated before voluntary reductions and regulatory controls are applied. NA (Not Applicable) Note: Totals may not sum due to independent rounding. 3 Table 3-54: Non-combustion CO2 Emissions from Natural Gas Systems (MMT) Stage 1990 2005 2012 2013 2014 2015 2016 Exploration 0.3 1.4 1.6 1.5 1.9 1.1 0.6 Production 0.8 2.1 3.6 3.8 3.9 4.0 4.0 Processing 28.3 18.9 19.1 20.5 21.0 21.0 22.0 Transmission and Storage 0.2 0.1 0.1 0.1 0.1 0.1 0.1 Distribution 0.1 + + + + + + Total 29.7 22.5 24.4 26.0 27.0 26.3 26.7 + Does not exceed 0.1 MMT CO2 Eq. Note: Totals may not sum due to independent rounding. 4 Table 3-55: Non-combustion CO2 Emissions from Natural Gas Systems (kt) Stage 1990 2005 2012 2013 2014 2015 2016 Exploration 328 1,406 1,568 1,517 1,874 1,101 571 Production 825 2,082 3,560 3,822 3,924 4,023 4,002 Processing 28,338 18,875 19,120 20,508 21,044 21,044 22,009 Transmission and Storage 166 140 135 142 148 147 143 Distribution 51 27 15 14 14 14 14 Total 29,708 22,529 24.398 26,004 27,004 26,329 26,739 Note: Totals may not sum due to independent rounding. 3-80 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2016 ------- 1 Methodology 2 See Annex 3.6 for the Ml time series of emissions data, activity data, and emission factors, and additional 3 information on methods and data sources. 4 The methodology for natural gas emission estimates in the Inventory involves the calculation of CH4 and CO2 5 emissions for over 100 emissions sources, and then the summation of emissions for each natural gas segment. 6 The approach for calculating emissions for natural gas systems generally involves the application of emission factors 7 to activity data. For most sources, the approach uses technology-specific emission factors or emission factors that 8 vary over time and take into account changes to technologies and practices, which are used to calculate net 9 emissions directly. For others, the approach uses what are considered "potential methane factors" and reduction data 10 to calculate net emissions. 11 Emission Factors. Key references for emission factors for CH4 and non-combustion-related CO2 emissions from the 12 U.S. natural gas industry include a 1996 study published by the Gas Research Institute (GRI) and EPA (GRI/EPA 13 1996), the EPA's Greenhouse Gas Reporting Program (GHGRP 2017), and others. 14 The EPA/GRI study developed over 80 CH4 emission factors to characterize emissions from the various components 15 within the operating stages of the U.S. natural gas system. The EPA/GRI study was based on a combination of 16 process engineering studies, collection of activity data, and measurements at representative gas facilities conducted 17 in the early 1990s. Year-specific natural gas CH4 compositions are calculated using U.S. Department of Energy's 18 Energy Information Administration (EIA) annual gross production for National Energy Modeling System (NEMS) 19 oil and gas supply module regions in conjunction with data from the Gas Technology Institute (GTI, formerly GRI) 20 Unconventional Natural Gas and Gas Composition Databases (GTI 2001). These year-specific CH4 compositions are 21 applied to emission factors, which therefore may vary from year to year due to slight changes in the CH4 22 composition for each NEMS region. 23 GHGRP Subpart W data were used to develop both CH4 and CO2 emission factors for several sources in the 24 Inventory. In the onshore production segment, GHGRP data were used to develop emission factors used for all time 25 series years for well testing, gas well completions and workovers with and without hydraulic fracturing, pneumatic 26 controllers and chemical injection pumps, condensate tanks, liquids unloading, and miscellaneous flaring. In the 27 processing segment, for recent years of the times series, GHGRP data were used to develop emission factors for 28 fugitives, compressors, flares, dehydrators, and blowdowns/venting. In the transmission and storage segment, for 29 recent years of the times series, GHGRP data were used to develop factors for pneumatic controllers. 30 Other data sources used for CH4 emission factors include Marchese et al. (2015) for gathering stations, Zimmerle et 31 al. (2015) for transmission and storage station fugitives and compressors, and Lamb et al. (2015) for recent years for 32 distribution pipelines and meter/regulator stations. 33 For sources in the exploration, production and processing segments that use emission factors not directly calculated 34 from GHGRP data, data from the 1996 GRI/EPA study and a 2001 GTI publication were used to adapt the CH4 35 emission factors into non-combustion related CO2 emission factors. For sources in the transmission and storage 36 segment that use emission factors not directly calculated from GHGRP data, and for sources in the distribution 37 segment, data from the 1996 GRI/EPA study and a 1993 GTI publication were used to adapt the CH4 emission 38 factors into non-combustion related CO2 emission factors. See Annex 3.6 for more detailed information on the 39 methodology and data used to calculate CH4 and non-combustion CO2 emissions from natural gas systems. 40 Activity Data. Activity data were taken from various published data sets, as detailed in Annex 3.6. Key activity data 41 sources include data sets developed and maintained by EPA's GHGRP; Drillinglnfo, Inc.; U.S. Department of the 42 Interior's Bureau of Ocean Energy Management, Regulation and Enforcement (BOEMRE, previously Minerals and 43 Management Service); Federal Energy Regulatory Commission (FERC); EIA; the Natural Gas STAR Program 44 annual emissions savings data; Oil and Gas Journal; PHMSA; the Wyoming Conservation Commission; and the 45 Alabama State Oil and Gas Board. 46 For a few sources, recent direct activity data are not available. For these sources, either 2015 data were used as a 47 proxy for 2016 data, or a set of industry activity data drivers was developed and used to calculate activity data over 48 the time series. Drivers include statistics on gas production, number of wells, system throughput, miles of various Energy 3-81 ------- 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 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 certain sectors, some sources are calculated using potential emission factors, and the step of deducting CH4 that is not emitted from the total CH4 potential estimates to develop net CH4 emissions is applied. To take into account 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 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 for certain sources in the production, transmission, and distribution segments. In fall of 2015, a well in a California storage field began leaking methane at an initial average rate of around 50 metric tons (MT) of methane (CH4) an hour, and continued leaking until it was permanently sealed in February of 20 1 6.86 An emission estimate from the leak event was included for 2015 and 2016, using the estimate of the leak published by the California Air Resources Board (99,638 MT CH4 for the duration of the leak). The 2015 and 2016 emission estimates of 78,350 MT CH4 and 21,288 MT CH4, respectively, were added to the 2015 and 2016 estimates of fugitive emissions from storage wells. For more information, please see Inventory of U.S. Greenhouse Gas Emissions and Sinks 1990-2015: Update for Storage Segment Emissions,87 Through EPA's stakeholder process on oil and gas in the Inventory, EPA received initial stakeholder feedback on updates under consideration for the Inventory. Stakeholder feedback is noted below in Uncertainty and Time -Series Consistency, Recalculations Discussion, and Planned Improvements. Uncertainty and Time-Series Consistency In recent years, EPA has made significant revisions to the Inventory methodology to use updated activity and emissions data. To update its characterization of uncertainty, EPA has conducted a draft quantitative uncertainty analysis using the IPCC Approach 2 methodology (Monte Carlo Simulation technique). The 95 percent confidence intervals presented here are based on 2015 data from the previous (i.e., 1990 through 2015) Inventory. EPA is still seeking comment on the approach to calculate uncertainty and may update its approach in the current Inventory. Initial stakeholder feedback on the uncertainty analysis included support for annual updates to the uncertainty assessment, so that the uncertainty ranges will continue to reflect new data as they become available. Similarly, a stakeholder cautioned against not updating the uncertainty range to reflect updated data, in particular for transmission and storage, in future Inventories. Stakeholders supported the approach of calculating uncertainty for the top emitters. For more information, please see the Planned Improvements section, and the Draft 2018 Uncertainty Memo88. To develop the values in Table 3-56 below, EPA has applied the uncertainty bounds calculated for the 2015 emission estimates presented in the previous Inventory. To develop the uncertainty bounds, EPA used the IPCC Approach 2 methodology (Monte Carlo Simulation technique). Microsoft Excel's @RISK add-in tool was used to estimate the 95 percent confidence bound around methane emissions from natural gas systems. For the analysis, EPA focused on the 14 highest-emitting sources for the year 2015, which together emitted 77 percent of methane from natural gas systems in 2015, and extrapolated the estimated uncertainty for the remaining sources. The @RISK 86 For more information on the Aliso Canyon event, and the measurements conducted of the leak, please see Ensuring Safe and Reliable Underground Natural Gas Storage, Final Report of the Interagency Task Force on Natural Gas Storage Safety, available at . 87 - 88 See < https://www.epa.gov/sites/production/files/2017- 10/documents/revision_under_consideration_for_ghgi_ng_and_petro_uncertainty_2017-10-25_to_post.pdf> 3-82 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2016 ------- 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 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. 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." 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 2016, using the IPCC methodology. The results of the Approach 2 uncertainty analysis are summarized in Table 3-56. Natural gas systems CH4 emissions in 2016 were estimated to be between 137.9 and 190.4 MMT CO2 Eq. at a 95 percent confidence level. Natural gas systems non-energy CO2 emissions in 2016 were estimated to be between 22.8 and 31.4 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. For example, years where many emission sources are calculated with interpolated data would likely have higher uncertainty than years with predominantly year-specific data. Table 3-56: Approach 2 Quantitative Uncertainty Estimates for CH4 and Non-energy CO2 Emissions from Natural Gas Systems (MMT CO2 Eq. and Percent) Source Gas 2016 Emission Estimate Uncertainty Range Relative to Emission Estimate3 (MMT CO2 Eq.)b (MMT CO2 Eq.) (%) Lower Upper Boundb Boundb Lower Upper Boundb Boundb Natural Gas Systems CH4 162.1 137.9 190.4 -15% +17% Natural Gas Systems0 CO2 26.7 22.8 31.4 -15% +17% a Range of emission estimates estimated by applying the 95 percent confidence intervals obtained from the Monte Carlo Simulation analysis conducted for the year 2015. 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 3-51 and Table 3-52. c An uncertainty analysis for the non-energy CO2 emissions was not performed. The relative uncertainty estimated (expressed as a percent) from the CH4 uncertainty analysis was applied to the point estimate of non-energy CO2 emissions. 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 1990 through 2016, 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. QA/QC and Verification Discussion The natural gas 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 Energy 3-83 ------- 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 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.89 As in previous years, EPA conducted early engagement and communication with stakeholders on updates prior to public review. EPA held stakeholder workshops on greenhouse gas data for oil and gas in June and October of 2017, and held webinars in April and August of 2017. In advance of each workshop, EPA released memos detailing updates under consideration and requesting stakeholder feedback. Stakeholder feedback received through these processes is discussed in the Recalculations Discussion and Planned Improvements sections below. 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 (i.e., the independent study should assess data from the Inventory and not another data set, such as EDGAR.). In an effort to improve the ability to compare the national-level inventory with measurement results that may be at other scales, a team at Harvard University along with EPA and other coauthors developed a gridded inventory of U.S. anthropogenic methane emissions with 0.1° x 0.1° spatial resolution, monthly temporal resolution, and detailed scale-dependent error characterization.90 The gridded methane inventory is designed to be consistent with the 2016 Inventory of U.S. Greenhouse Gas Emissions and Sinks (1990-2014) estimates for the year 2012, which presents national totals.91 Recalculations Discussion The EPA received information and data related to the emission estimates through GHGRP reporting, the annual Inventory formal public notice periods, stakeholder feedback on updates under consideration, and new studies. In June and October 2017, the EPA released draft memoranda, Inventory of U.S. Greenhouse Gas Emissions and Sinks 1990-2016: Revisions Under Consideration for CO2 Emissions {Draft 2018 CO2 Memo),92 Inventory of U.S. Greenhouse Gas Emissions and Sinks 1990-2016: Updates Under Consideration for Natural Gas and Petroleum Systems Uncertainty Estimates {Draft 2018 Uncertainty Memo) ,93 and Inventory of U.S. Greenhouse Gas Emissions and Sinks 1990-2016: Additional Revisions Under Consideration {Draft 2018 Other Updates Memo),94. The memos discussed changes under consideration, and requested stakeholder feedback on those changes. The EPA thoroughly evaluated relevant information available, and made several updates to the Inventory, including to define an exploration segment separate from production (not a methodological change, but a change in presentation of information), calculate activity and emission factors for well testing and non-hydraulically fractured completions from GHGRP data, recalculate production segment major equipment activity factors using updated GHGRP data, and calculate new CO2 emission factors for several sources throughout all segments directly from GHGRP data. 89 See . 90 See . 91 See . 92 See 93 See < https://www.epa.gov/sites/production/files/2017- 10/documents/revision_under_consideration_for_ghgi_ng_and_petro_uncertainty_2017-10-25_to_post.pdf> 94 3-84 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2016 ------- 1 The combined impact of revisions to 2015 natural gas sector CH4 emissions, compared to the previous Inventory, is 2 an increase from 162.4 to 164.4 MMT CO2 Eq. (2.0 MMT CO2 Eq., or 1.2 percent). The recalculations resulted in 3 an average increase in CH4 emission estimates across the 1990 through 2015 time series, compared to the previous 4 Inventory, of 0.1 MMT CO2 Eq, or 0.1 percent. 5 The combined impact of revisions to 2015 natural gas sector CO2 emissions, compared to the previous Inventory, is 6 a decrease from 42.4 to 26.3 MMT CO2 (16.0 MMT CO2, or 38 percent). The recalculations resulted in an average 7 decrease in emission estimates across the 1990 through 2015 time series, compared to the previous Inventory, of 8 10.3 MMT CO2 Eq, or 29 percent. The decreased estimate results primarily from recalculations related to the 9 reallocation of CO2 from flaring to petroleum systems from natural gas systems. Previously, data were not available 10 to disaggregate flared emissions between natural gas and petroleum. 11 Exploration 12 The natural gas system segments were reorganized for the current Inventory and now include a specific exploration 13 segment to improve conformance with the IPCC guidelines. Exploration activities were previously included under 14 the production segment. The activities included under exploration are hydraulically fractured (HF) gas well 15 completions, gas well completions without HF, well drilling, and well testing. EPA developed a new methodology 16 to estimate emissions from well testing (not during completions) using GHGRP data, revised the methodology for 17 non-HF gas well completions to use GHGRP data, and updated the HF gas well completions methodology for CO2 18 emissions. These recalculations are discussed below. 19 Well Testing 20 EPA developed a new estimate for gas well testing (during non-completion events) using GHGRP data. In previous 21 Inventories, only well testing conducted as part of a completion event was included. CH4 and CO2 emission factors 22 were developed, on a per-event basis, for vented and flared gas well testing events using RY2015 and RY2016 data. 23 EPA developed activity factors (i.e., number of events per gas well) to determine the number of well testing events 24 in a year, also using RY2015 and RY2016 data. GHGRP RY2015 activity and emission factors are applied to all 25 prior years of the time series. Methane emissions from well testing averaged 1.5 kt (or less than 0.05 MMT CO2 Eq.) 26 over the time series. There was a large decrease in methane emissions from gas well testing from 2015 to 2016 as 27 observed in reported GHGRP data. Carbon dioxide emission from well testing averaged 3.1 kt (or less than 0.05 28 MMT CO2) over the time series. See the Draft 2018 Other Updates Memo for additional discussion. 29 Table 3-57: Gas Well Testing National ChU Emissions (Metric Tons ChU) Source 1'WO 2005 2012 2013 2014 2015 2016 Non-Completion Well Testing - Vented 949 1,673 2,080 2,054 2,071 2,043 614 Non-Completion Well Testing - Flared 13 23 29 29 29 29 2 30 Table 3-58: Gas Well Testing National CO2 Emissions (Metric Tons CO2) Source 1'WO 2005 2012 2013 2014 2015 2016 Non-Completion Well Testing - Vented Non-Completion Well Testing - Flared 30 1.914 53 3,375 66 4,198 65 4,144 65 4,179 64 4,123 39 323 31 Non-HF Gas Well Completions 32 EPA developed new emission factors for controlled and uncontrolled non-HF gas well completions using GHGRP 33 data, and applied the new factors over all time series years. The emission factor for non-HF gas well completions in 34 the Inventory was previously derived from the GRI 1996 study which defines the factor as covering both gas well Energy 3-85 ------- 1 completions and well flow testing, and based on the assumption that all gas is flared. CH4 and CO2 emission factors 2 were developed, on a per-event basis, for vented and flared gas well non-HF completion events using RY2015 and 3 RY2016 GHGRP data. EPA did not revise the overall counts of non-HF gas well completions. For the split between 4 vented and flared events, EPA used GHGRP data for year 2011 forward, and 2011 data (which show 3 percent of 5 events flared) as a proxy for all earlier years. Methane emissions from non-HF completions averaged 8.6 kt CH4 (or 6 0.2 MMT CO2 Eq.) over the time series. The previous estimate was an average of 0.01 kt CH4 over the time series. 7 Carbon dioxide emission from non-HF completions averaged 7.3 kt (or less than 0.05 MMT CO2) over the time 8 series. The previous estimate was an average of 0.001 kt CH4 over the time series. See the Draft 2018 Other 9 Updates Memo for additional discussion. 10 Table 3-59: Non-HF Gas Well Completions National ChU Emissions (Metric Tons ChU) Source 1990 2005 2012 2013 2014 2015 2016 Non-HF Completions - Vented 5,713 10,074 11,009 5,890 1,404 13,680 8,065 Non-HF Completions - Flared 20 35 2 39 12 36 89 11 Table 3-60: Non-HF Gas Well Completions National CO2 Emissions (Metric Tons CO2) Source 1990 2005 2012 2013 2014 2015 2016 Non-HF Completions - Vented 216 381 101 182 72 172 829 Non-HF Completions - Flared 4,643 8,187 565 6,695 2,683 5,909 16,407 12 CO2 Updates 13 EPA developed new CO2 emission factors for the four control categories of HF gas well completions using the same 14 GHGRP data sets and methodology as established for CH4. EPA did not change the activity data methodology for 15 this source, other than to break out HF completions and workovers as separate line items (where completions are 16 included in Exploration and workovers remain within the Production segment). As noted in Planned Improvements, 17 EPA is considering year-specific GHGRP-based emission factor for this source. See the Draft 2018 CO2 Memo for 18 additional discussion. 19 Table 3-61: HF Gas Well Completions National CO2 Emissions (kt CO2) Source 1990 2005 2012 2013 2014 2015 2016 HF Completions - Non-REC with Venting HF Completions - Non-REC with Flaring HF Completions - REC with Venting HF Completions - REC with Flaring Total Emissions 10 311 26 1,062 8 759 6 587 6 613 + 273 + 114 0 0 2 304 2 794 3 911 2 1,246 3 814 2 437 321 1,394 1,563 1,506 1,867 1,091 553 Previous Estimated Emissions 74 305 99 75 66 66 NA NA (Not Applicable) 20 + Does not exceed 0.5 kt CO2. 21 Production 22 In addition to the memos discussed above, this section references the memorandum, Inventory of U.S. Greenhouse 23 Gas Emissions and Sinks 1990-2015: Revisions for Natural Gas and Petroleum Systems Production Emissions 24 (2017 Production Memo).95 95 See . 3-86 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2016 ------- 1 Non-HF Gas Well Workovers 2 EPA developed new emission factors for controlled and uncontrolled non-HF gas well workovers using GHGRP 3 data, and applied the new factors over all time series years. The emission factor for non-HF gas well workovers in 4 the Inventory was previously derived from the GRI 1996 study. Methane and CO2 emission factors were developed, 5 on a per-event basis, for vented and flared gas well non-HF workover events using RY2015 andRY2016 data. EPA 6 did not revise the overall counts of non-HF gas well workovers. For the split between vented and flared events, EPA 7 used GHGRP data for year 2011 forward, and interpolated to 100 percent vented and 0 percent flared in year 1992 8 (GRI basis). Methane emissions from non-HF workovers averaged 0.6 kt CH4 (or 0.02 MMT CO2 Eq.) over the time 9 series. The previous estimate was an average of 0.4 kt CH4 over the time series. Carbon dioxide emission from non- 10 HF workovers averaged 2.2 kt (or less than 0.05 MMT CO2) over the time series. The previous estimate was an 11 average of 0.03 kt CH4 over the time series. See the Draft 2018 Other Updates Memo for additional discussion. 12 Table 3-62: Non-HF Gas Well Workovers National ChU Emissions (Metric Tons ChU) Source 1990 2005 2012 2013 2014 2015 2016 Non-HF Workovers - Vented Non-HF Workovers - Flared 509 0 631 19 1,486 0 429 6 441 2 525 26 517 1 13 14 Table 3-63: Non-HF Gas Well Workovers National CO2 Emissions (kt CO2) ~~Source 1990 2005 2012 2013 2014 2015 2016 Non-HF Workovers - Vented 30 38 92 24 28 45 25 Non-HF Workovers - Flared 0 3,164 97 942 548 3,192 5,836 15 Activity data updates 16 Well Counts 17 EPA has used a more recent version of the Drillinglnfo data set to update well counts data in the Inventory. There 18 are not methodological changes to this source in the 2018 Inventory or major changes to the activity data, but 19 because this is a key input, results are highlighted here. 20 Table 3-64: Producing Gas Well Count Data Gas Well Count 1990 2005 2012 2013 2014 2015 2016 Number of Gas Wells Previous Estimate 197,626 202,628 348,470 355,234 433,390 438,672 427,828 431,926 431,446 433,941 425,651 421,893 416,881 NA NA (Not Applicable) 21 In December 2017, EIA released a 2000 through 2016 time series of national oil and gas well counts. EIA total (oil 22 and gas) well counts for 2016 were 1,010,441. EPA's total well counts were 978,845. Over the 2000 through 2016 23 time series, EPA's well counts were on average 2 percent lower than EIA's. EIA's well counts include side tracks, 24 completions, and recompletions, and therefore are expected to be higher than EPA's which include only producing 25 wells. EPA and EIA use different thresholds for distinguishing between oil and gas (EIA uses 6 mcf/bbl, while EPA 26 uses 100 mcf/bbl), which results in EIA having a lower fraction of oil wells and a higher fraction of gas wells than 27 EPA. Across the 2000 through 2016 EIA time series, EIA estimates (which include multiple well categories, as 28 noted above) on average 128,335 (or 31 percent) more gas wells in each year than EPA's gas well counts (which 29 include only producing wells). 30 Equipment Counts 31 EPA recalculated activity factors of equipment per well using the latest GHGRP RY2015 data. This resulted in 32 changes across the time series. For example, the number of heaters per well decreased by 20 percent over the time 33 series, the number of chemical injection pumps per well decreased by 4 percent, and the number of dehydrators per 34 well increased by 5 percent. The impact of the changes in equipment counts per well along with changes in well Energy 3-87 ------- 1 counts resulted in changes in methane emissions across the time series for heaters (-21 percent), chemical injection 2 pumps (-6 percent), and dehydrators (+3 percent). 3 CO2 Updates 4 EPA updated CO2 emissions for a number of sources in the production segment. See the Draft 2018 C02 Memo for 5 more details. The overall impact was an average decrease of 9.6 MMT CO2 (or 81 percent) over the time series, 6 which is partially due to the reallocation of CO2 emissions from associated gas and miscellaneous onshore 7 production flaring from Natural Gas Systems to Petroleum Systems, which was not possible in the past because the 8 previous data source aggregated venting and flaring activity data from both petroleum and natural gas systems, but 9 is now possible because through use of the GHGRP data. 10 Sources with the largest impacts include flaring (decrease of 9.9 MMT CO2 on average over the time series), and 11 tanks (increase of 0.5 MMT CO2 over the time series). These sources are discussed in detail below. Other sources 12 recalculated had increases or decreases of less than 0.5 MMT CO2. 13 Miscellaneous Production Flaring 14 The EPA developed new estimates for CO2 and CH4 emissions from miscellaneous production flaring using 15 GHGRP subpart W data. Along with other updates to flaring emissions in both oil and gas production, this replaces 16 the estimate for onshore flaring that was previously reported in the natural gas systems CO2 emissions totals. EPA 17 developed emission factors from 2015 and 2016 GHGRP data; the 2015 emission factor is applied to all prior years. 18 The emission factors are on a per-well basis and were applied to all gas wells in each year. Details are provided in 19 the Draft 2018 C02 Memo. Initial stakeholder feedback on this update suggested use of production-based emission 20 factors as opposed to well-based emission factors. 21 Table 3-65: Miscellaneous Production Flaring National CO2 Emissions (kt CO2) Source 1WO 2005 2012 2013 2014 2015 2016 Miscellaneous Production Flaring Previous Estimated emissions from flaring (natural gas and petroleum)" 0 1,009 1,834 1,810 1,826 1,801 1,445 9,093 7,193 12,704 15,684 17,629 17,629 NA a The previous estimated emissions from ilaring included emissions from multiple sources in the production and processing segments, and also included petroleum systems flaring emissions. NA (Not Applicable) 22 Tanks 23 EPA developed CO2 emissions estimates for condensate tanks using GHGRP data and a throughput-based approach. 24 This approach is identical to the methodology to calculate CH4 emissions; for more information, please see the 2017 25 Production Memo. The overall impact of the change is an increase in calculated CO2 emissions by 0.5 MMT CO2 26 over the time series. 27 Table 3-66: National Condensate Tank Emissions by Category and National Emissions (kt 28 COz) CO2 Emissions 1990 2005 2012 2013 2014 2015 2016 Large Tanks w/ Flares 28" 363 819 985 1,030 1,044 1,398 Large Tanks w/ VRU 0 1 2 3 3 3 3 Large Tanks w/o Control 1 + + 1 1 1 1 Small Tanks w/ Flares 0 9 27 33 35 35 42 Small Tanks w/o Flares 6 4 8 9 10 10 15 Malfunctioning Dump Valves Total Emissions + 294 + / 378 + 857 + 1,030 + 1,078 + 1,093 + 1,460 Previous Estimated Emissions 296 383 870 1,045 1,093 1,108 NA NA (Not Applicable) + Does not exceed 0.5 kt CO2. 3-88 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2016 ------- 1 Processing 2 There were no updates to the CH4 emissions estimation methodology for the processing segment. Updates to activity 3 data resulted in a minor decrease (less than 0.1 MMT CO2 Eq., or 0.5 percent) in CH4 emissions estimates for this 4 segment across the time series. EPA updated CO2 emissions for a number of sources in the processing segment to 5 use emission factors directly calculated from subpart W data. See the Draft 2018 C02 Memo for more details. The 6 overall impact was an average decrease of 1.7 MMT CO2 (or 8 percent) over the time series, which is primarily due 7 to the incorporation of GHGRP data for acid gas removal vents. Acid gas removal CO2 emissions decreased by an 8 average of 4.7 MMT CO2, or 21 percent over the time series. Incorporation of GHGRP data for flaring in 9 processing increased emissions by 3.0 MMT CO2. 10 Table 3-67: Processing CO2 Updates, National Emissions (kt CO2) Source 1990 2005 2012 2013 2014 2015 2016 Acid Gas Removal 28,282 15,320 13,579 14,565 14,946 14,946 16,565 Previous Acid Gas Removal 27,708 21,694 21,404 21,690 23,643 23,643 NA Processing-flaring 0 3,516 5,502 5,902 6,056 6,056 5,404 Previous processing flaring a NA NA NA NA NA NA NA NA (Not Applicable) a The previous estimated emissions from flaring included emissions from multiple sources in the natural gas and petroleum production segments, and natural gas processing segment. The previous estimate was presented as a single emission source in the natural gas systems production segment. 11 Transmission and Storage 12 Changes in the estimates for CH4 from transmission and storage include the addition of flaring emissions and 13 recalculations due to updated data (e.g., GHGRP station counts, the GHGRP split between dry and wet seal 14 centrifugal compressors, and GHGRP pneumatic controller data). Stakeholder feedback (one stakeholder) expressed 15 support for this approach. These changes resulted in an average increase in calculated emissions over the time series 16 from this segment of less than 0.1 MMT CO2 Eq., or 0.1 percent. 17 Additional information on inclusion of the Aliso Canyon emissions can be found in the Methodology section above 18 and in the 2017 Transmission and Storage Memo96 and not in the Recalculation Discussion section as it did not 19 involve recalculation of a previous year of the Inventory. 20 Table 3-68: Transmission and Storage ChU Updates to Flaring, National Emissions (MT ChU) Source 1990 2005 2012 2013 2014 2015 2016 Transmission-flaring * 307 276 281 303 326 326 395 Storage-flaring* 235 223 231 232 232 227 198 Previous flaring (transmission NA NA •• NA NA NA NA NA and storage) NA (Not Applicable) *Estimates are developed from GHGRP data, wherein compressor stations that service underground storage fields might be classified as transmission compression as the primary function. A significant fraction of the transmission station flaring emissions presented in this table likely occurs at stations that service storage facilities; such stations typically require flares, compared to a typical transmission compressor station used solely for mainline compression that does not require liquids separation, dehydration, and flaring. 21 EPA updated CO2 emissions for pneumatic controllers and flares in the transmission and storage segment. See the 22 Draft 2018 CO2 Memo for more details. The overall impact was an average increase of 0.1 MMT CO2 (or by a 23 factor of 3) over the time series. The updated CO2 data for pneumatic controllers increased estimated emissions 24 from pneumatic controllers by less than 0.1 MMT CO2, or 53 percent over the time series. The addition of an 25 estimate for flares increased CO2 emissions from transmission and storage by 2.5 MMT CO2 over the time series. 96 See . Energy 3-89 ------- 1 Table 3-69: Transmission and Storage CO2 Updates, National Emissions (kt CO2) Source mo 2005 2012 2013 2014 2015 2016 Transmission-pneumatic controllers 6.3 4.0 2.9 3.0 0.9 0.8 0.7 Previous transmission pneumatic controllers 6.1 2.1 0.6 0.8 0.8 0.8 NA Storage-pneumatic controllers Previous Storage pneumatic controllers 1.3 1.3 1.2 1.0 O 00 0 1.0 0.9 0.9 0.9 0.6 0.7 1.0 NA Transmission-flaring * 78.8 71.0 72.2 78.0 83.7 83.9 88.4 Storage-flaring* Previous flaring (transmission and storage) 24.5 a :i 23.2 V,l 24.0 NA 24.1 NA 24.1 NA 23.6 NA 15.3 NA NA (Not Applicable) * Estimates are developed from GHGRP data, wherein compressor stations that service underground storage fields might be classified as transmission compression as the primary function. A significant fraction of the transmission station flaring emissions presented in this table likely occurs at stations that service storage facilities; such stations typically require flares, compared to a typical transmission compressor station used solely for mainline compression that does not require liquids separation, dehydration, and flaring. 2 Distribution 3 Although there were no methodological updates to the distribution segment, recalculations due to updated data (e.g., 4 GHGRP M&R station counts) resulted in an average increase in calculated emissions over the time series from this 5 segment of less than 0.01 MMT CO2 Eq. CH4 (or less than 0.1 percent) and less than 0.01 MMT CO2 (or 1.9 6 percent). 7 Planned Improvements 8 Plans for 2018 Inventory (1990 through 2016) and Future Inventories 9 EPA seeks stakeholder feedback on the improvements noted below for the final version of the current Inventory and 10 future Inventories. 11 Uncertainty 12 The uncertainty analysis results presented for this public review Inventory were based on the top methane-emitting 13 sources for 2015 from the previous Inventory. EPA will re-evaluate the highest emitting sources for 2016, based on 14 the final version of the current Inventory, and update the uncertainty analysis to reflect these sources and their 15 methodology, as necessary. EPA will also consider further stakeholder feedback on the Draft 2018 Uncertainty 16 Memo. 17 Miscellaneous Production Flaring 18 Miscellaneous production flaring emission factors are currently applied on a well-basis at the national-level. EPA is 19 considering two additional options for the final version of the current Inventory, based on stakeholder feedback: a 20 production-based approach and developing factors at a basin-level. Each of these options is being considered to 21 avoid over- or underestimating emissions from this source. 22 Gas STAR Reductions 23 As detailed in the Draft 2018 Other Updates Memo, EPA is continuing to evaluate sources that currently use 24 voluntary reduction data in calculating emissions to identify instances where an emission source's calculation 25 methodology could be updated to calculate net emissions or instances where the current methodology could be 26 simplified to acknowledge sources that likely no longer necessitate consideration of Gas STAR reductions. EPA has 27 assessed Gas STAR reduction data and is considering several updates for the final version of the current Inventory. 3-90 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2016 ------- 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 In the Production Segment, a spreadsheet error in the public review draft resulted in a miscalculation of the scaling factor for the "other reductions" value. Correcting this error results in a decrease in the Gas STAR reductions for this segment of around 1 MMT CO2 Eq. (from 9 MMT CO2 Eq. to 8 MMT CO2 Eq.). In the transmission and storage segment, Gas STAR reductions associated with transmission station fugitives were not removed from the Gas STAR data when the emissions were recalculated with a "net approach." Correcting the estimates to remove those reductions results in a decrease in the Gas STAR reductions of 0.1 MMT CO2 Eq. (from 8.3 MMT CO2 Eq. to 8.2 MMT CO2 Eq.). In the distribution segment, Gas STAR reductions associated with mishaps (dig-ins) and pipeline blowdowns might be removed, which would result in a small decrease in the Gas STAR reductions of less than 0.1 MMT CO2 Eq. (from 1.0 MMT CO2 Eq. to 0.99 MMT CO2 Eq.). See Draft 2018 Other Updates Memo for more information. EPA continues to review unassigned Gas STAR reductions in the transmission and storage segment and distribution segment (currently grouped and identified as "other" reductions). Liquids Unloading EPA is considering several updates to liquids unloading for the final version of the current Inventory. Based on stakeholder feedback, EPA is considering developing region-specific liquids unloading emissions and activity factors, rather than national-level. Preliminary analysis of emissions calculated using a basin-level approach indicates that the emission estimates will be lower than those calculated with the national-level approach. Additionally, the emissions and activity data methodology used in the current Inventory rely exclusively on recently collected data (from 2011 or later). The EPA is evaluating the liquids unloading data collected for the 1996 GRI/EPA study to determine if it better represents early time series years. Year-Specific Emission Factors EPA is considering the development of year-specific emission factors, using GHGRP data, for a number of sources with annual emissions currently calculated with data from one year or an average of a several years. For example, for hydraulically fractured gas well completions and workovers, while changes in practices over time are currently reflected in the Inventory due to annual practice-specific activity data, changes in emissions within each practice- specific category are not currently reflected. Preliminary analysis of emissions calculated using a year-specific emission factor approach for hydraulically fractured gas well completions and workovers indicates that the emission estimates will be lower than those calculated with the average emission factor approach. EPA is considering updating emission factors for this and potentially other sources for the final version of the current Inventory. Well-Related Activity Data As described in the Recalculations Discussion, EPA has updated the emission factors for several well-related emission sources, including testing, completions, and workovers. EPA will continue to assess available data that and stakeholder feedback on considerations to improve activity estimates for these types of sources. For example, the current Inventory assumes that 1 percent of HF gas wells and 4.35 percent of non-HF gas wells undergo workovers each year based on historical assumptions; EPA will review available data including from the GHGRP to consider updating the activity data methodology. LNG Segment Emissions The current Inventory estimates emissions from LNG storage stations and LNG import terminals in the transmission and storage segment of natural gas systems. The emission factors are based on the 1996 GRI/EPA study, which developed emission factors using underground natural gas storage and transmission compressor station data; specific emissions data for LNG storage stations and LNG import terminals were not available in the GRI/EPA study. EPA's GHGRP subpart W collects data from LNG storage and LNG import and export facilities that meet a reporting threshold of 25,000 metric tons of CO2 equivalent (MT CO2 Eq.) emissions. EPA is considering approaches and seeking stakeholder feedback on incorporating GHGRP data to improve LNG emissions estimates in the Inventory. Refer to the Draft 2018 Other Updates Memo for additional details. Incorporating GHGRP data would likely decrease emissions from this segment. Energy 3-91 ------- 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 N2() Emissions N20 emissions are currently not included in petroleum systems estimates, but EPA is considering developing a methodology to estimate N2O emissions. The Draft 2018 Other Updates Memo provides discussion on this topic. EPA will consider options such as using GHGRP data directly, for sources that already rely on GHGRP data for CH4 or CO2 estimates. GHGRP RY2015 reported N20 flaring emissions specific to natural gas systems were 26 metric tons (or less than 0.01 MMT CO2 Eq.). In addition, 36 metric tons N2O (or 0.01 MMT CO2 Eq.) flaring emissions were reported for GHGRP RY2015 for sources that fall within both natural gas and petroleum systems. Offshore Platforms EPA is considering updates to the offshore platform emissions calculation methodology, as discussed in the Draft 2018 Other Updates Memo. The current emission factors were based on data from the 2011 DOI/Bureau of Ocean Energy Management's (BOEM) Gulf Offshore Activity Data System (GOADS), and 2014 GOADS data is available. A different source for platform counts is also being considered. Upcoming Data, and Additional Data that Could Inform the Inventory EPA will continue to review data available from its GHGRP, in particular new data on gathering and boosting stations, gathering pipelines, and transmission pipeline blowdowns and new well-specific information, available in 2017 (for reporting year 2016) for the first time. EPA will consider revising its methods to take into account the new GHGRP data. EPA will assess new data received by the Methane Challenge Program 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. Key studies in progress include DOE-funded work on the following sources: vintage and new plastic pipelines (distribution segment), industrial meters (distribution segment), and sources within the gathering and storage segments97; an API field study on pneumatic controllers; and a Pipeline Research Council International (PRCI) project in which researchers are gathering and analyzing subpart W data on transmission compressor stations and underground storage facilities. 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 stakeholder comments. EPA also continues to seek new data that could be used to assess or update the estimates in the Inventory. For example, stakeholder comments have highlighted areas where additional data that could inform the Inventory are currently limited or unavailable: • Tank malfunction and control efficiency data. • Consider updating engine emission factors, including using subpart W data to the extent possible, and considering whether and how to represent differences between rich- and lean-burn engines. • Activity data and emissions data for production facilities that do not report to GHGRP. • Natural gas leaks at point of use estimates. A recent study (Lavoie et al. 2017) measured three natural gas power plants and found them to be large sources of natural gas leak emissions, and the stakeholder suggested that EPA evaluate the study and any additional information available on this source. At least one country, the United Kingdom, includes an emission estimate for residential and commercial customer natural gas use leaks (e.g., domestic heating boiler cycling and pre-ignition losses from domestic and commercial gas appliances) in its national greenhouse gas emissions inventory; the EPA seeks available data to estimate emissions from this source in the U.S. Stakeholder feedback (one stakeholder) supports use of data from Lavoie et al. or use of the U.K. approach to calculate emissions from this source. One stakeholder suggested that the Inventory should be updated with site-level and basin-level data, noting the EPA could first use basin-level data to assess the Inventory, and that future research could focus on collecting data in basins with the largest discrepancies. 97 See . 3-92 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2016 ------- 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 EPA will continue to seek available data on these and other sources as part of the process to update the Inventory. 3.8 Abandoned Oil and Gas Wells (CRF Source Categories lB2a and lB2b) The term "abandoned wells" encompasses various types of 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 well is around 3 million (with around 2.6 million abandoned oil wells and 0.6 million abandoned gas wells). Wells that are plugged have much lower methane emissions than wells that are unplugged (less than 1 kg CH4 per well per year, versus over 100 kg CH4 per well per year). Around 30 percent of the abandoned well population in the U.S. is plugged. This fraction has increased over the time series (from around 19 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 230 kt CH4 in 2016. Emissions increased by 3 percent from 1990, as the total population of abandoned oil wells increased 25 percent. Emissions decreased by 1 percent between 2015 and 2016 as a result of well plugging activities. Abandoned gas wells. Abandoned gas wells emitted 54 kt CH4 in 2016. Emissions increased by 51 percent from 1990, as the total population of abandoned gas wells increased 73 percent. Emissions decreased by 1 percent between 2015 and 2016 as a result of well plugging activities. Table 3-70: ChU Emissions from Abandoned Oil and Gas Wells (MMT CO2 Eq.) Activity 1990 2005 2012 2013 2014 2015 2016 Abandoned Oil Wells 5.6 5.8 5.8 5.8 5.8 5.8 5.8 Abandoned Gas Wells 0.9 I.I 1.2 1.2 1.3 1.4 1.4 Total 6.5 7.0 7.0 7.1 7.2 7.1 Note: Totals may not sum due to independent rounding. able 3-71: ChU Emissions from Abandoned Oil and Gas Wells (kt) Activity 1990 2005 2012 2013 2014 2015 2016 Abandoned Oil Wells 224 ] 233 231 230 230 232 230 Abandoned Gas Wells 36 ! 42 48 50 52 55 54 Total 260 275 279 280 282 286 284 Note: Totals may not sum due to independent rounding. Methodology EPA developed abandoned well emission factors using data from Kang et al. (2016) and Townsend-Small et al. (2016). Plugged and unplugged abandoned well emission factors were developed at the national-level (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 all other abandoned wells. Energy 3-93 ------- 1 2 3 4 5 6 7 8 9 10 11 12 13 The total population of abandoned wells over the time series was estimated using historical data and Drillinglnfo data. The abandoned well population was then split into plugged and unplugged wells by assuming that all abandoned wells were unplugged in 1950, 31 percent of abandoned wells were plugged in 2016 (based on an analysis of Drillinglnfo data), and applying linear interpolation for intermediate years. See the Draft 2018 Abandoned Wells Memo for details.98 Abandoned Oil Wells Table 3-72: Abandoned Oil Wells Activity Data and Methane Emissions (Metric Tons ChU) Source Plugged abandoned oil wells Unplugged abandoned oil wells Total Abandoned Oil Wells Abandoned oil wells in Appalachia Abandoned oil wells outside of Appalachia Methane from plugged abandoned oil wells (MT) Methane from unplugged abandoned oil wells (MT) Total Methane from Abandoned oil wells (MT) 1990 2005 2012 2013 2014 2015 2016 382,446 610,884 719,901 736,830 754,118 776,450 788,396 1,666,399 1,769,214 1,768,266 1,769,425 1,770,862 1,783,308 1,771,362 2,048,846 2,380,098 2,488,167 2,506,255 2,524,980 2,559,758 2,559,758 26% 24% 24% 23% 23% 23% 23% 74% 76% 76% 77% 77% 77% 77% 314 471 539 549 560 574 582 223,780 232,546 230,070 229,885 229,735 231,011 229,464 224,094 233,017 230,609 230,434 230,295 231,585 230,046 Abandoned Gas Wells Table 3-73: Abandoned Gas Wells Activity Data and Methane Emissions (Metric Tons ChU) Source 1990 2005 2012 2013 2014 2015 2016 Plugged abandoned gas wells 59,480 103,379 138,537 146,187 156,144 167,011 169,580 Unplugged abandoned gas wells 259,166 299,402 340,284 351,053 366,666 383,580 381,011 Total Abandoned Gas Wells 318,645 402,781 478,821 497,239 522,810 550,591 550,591 Abandoned gas wells in Appalachia 28% 29% 30% 30% 30% 30% 30% Abandoned gas wells outside S ' / of Appalachia 72% 71% 70% 70% 70% 70% 70% Methane from plugged abandoned gas wells (MT) 53 96 131 139 149 159 162 Methane from unplugged abandoned gas wells (MT) 35,810 42,064 48,176 49,754 52,024 54,483 54,118 Total Methane from Abandoned gas wells (MT) 35,863 42,160 18,307 49,893 52,173 54,643 54,280 Uncertainty and Time-Series Consistency An uncertainty analysis for abandoned well emissions was not performed. To develop the values in Table 3-74 below, EPA has applied the uncertainty bounds calculated for the 2015 emission estimates presented in the previous (i.e., 1990 through 2015) Inventory for Petroleum Systems and Natural Gas Systems. EPA is still seeking comment 98 https://www.epa.gOv/sites/production/files/2017-10/documents/2018_ghgi_draft_revision_-_abandoned_wells_2017-10- 25_to_post.pdf 3-94 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2016 ------- 1 on the approach to calculate uncertainty and may update its approach in the final version of the current Inventory, 2 such as by incorporating uncertainty information from Townsend-Small et al. and Kang et al. For more information, 3 please see the Planned Improvements sections for Petroleum Systems and Natural Gas Systems, and the 2018 4 Uncertainty Memo." 5 Table 3-74: Approach 2 Quantitative Uncertainty Estimates for ChU Emissions from 6 Petroleum Systems (MMT CO2 Eq. and Percent) Source Gas 2016 Emission Estimate Uncertainty Range Relative to Emission Estimate3 (MMT CO2 Eq.)b (MMT CO2 Eq.) (%) Lower Upper Bound Bound Lower Upper Bound Bound Abandoned Oil Wells CH4 5.8 3.9 7.9 -32% +36% Abandoned Gas Wells ch4 1.4 1.2 1.6 -15% +17% a Range of emission estimates estimated by applying the 95 percent confidence intervals obtained from the Monte Carlo Simulation analysis conducted for the year 2015 for natural gas and petroleum systems. 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. 7 To calculate a time series of emissions for abandoned wells, EPA developed annual activity data for 1990 through 8 2016 by summing an estimate of total abandoned wells not included in recent databases, to an annual estimate of 9 abandoned wells in the Drillinglnfo data set. As discussed above, the abandoned well population was split into 10 plugged and unplugged wells by assuming that all abandoned wells were unplugged in 1950, 31 percent of 11 abandoned wells were plugged in 2016 (based on an analysis of Drillinglnfo data), and applying linear interpolation 12 for intermediate years. The same emission factors were applied to the corresponding categories for each year of the 13 time series. 14 QA/QC and Verific cussion 15 The emission estimates in the Inventory are continually being reviewed and assessed to determine whether emission 16 factors and activity factors accurately reflect current industry practices. A QA/QC analysis was performed for data 17 gathering and input, documentation, and calculation. QA/QC checks are consistently conducted to minimize human 18 error in the model calculations. 19 As in previous years, EPA conducted early engagement and communication with stakeholders on updates prior to 20 public review. EPA held stakeholder workshops on greenhouse gas data for oil and gas in June and October of 21 2017, and held webinars in April and August of 2017. In advance of each workshop, EPA released memos detailing 22 updates under consideration and requesting stakeholder feedback. Stakeholder feedback received through these 23 processes is discussed in the Planned Improvements sections below. 24 Planned Improvements 25 Through EPA's stakeholder process on oil and gas in the Inventory, EPA received initial stakeholder feedback on 26 the abandoned wells update to the Inventory. Stakeholders noted varying definitions regarding abandoned well 27 populations and subpopulations and plugging status, and noted varying degrees of plugging, due to state-level 28 programs to plug abandoned wells. A stakeholder noted limited coverage of abandoned wells studies in the U.S., and 29 cautioned that it may be premature to develop national level estimates for this source, while another stakeholder 30 supported the inclusion of this emission sources and noted that the update uses the best available data for this source. 31 EPA will also continue to assess new data and stakeholder feedback on considerations (such as the disaggregation of 32 the well population into Appalachia and other regions) to improve the abandoned well count estimates and emission 99 https://www.epa.gov/sites/production/files/2017- 10/documents/revision_under_consideration_for_ghgi_ng_and_petro_uncertainty_2017-10-25_to_post.pdf Energy 3-95 ------- 1 factors. In addition, the studies used to develop CH4 emission factors did not provide CO2 data. EPA will consider 2 developing a methodology to estimate CO2 emissions from abandoned wells, such as applying a ratio of CO2 to CH4 3 content in gas. 4 EPA will assess updates to the Uncertainty Analysis, such as incorporating data from Townsend-Small et al. and 5 Kang et al. to improve the uncertainty estimates. 6 3.9 Energy Sources of Indirect Greenhouse Gas 7 Emissions 8 In addition to the main greenhouse gases addressed above, many energy-related activities generate emissions of 9 indirect greenhouse gases. Total emissions of nitrogen oxides (NOx), carbon monoxide (CO), and non-CH4 volatile 10 organic compounds (NMVOCs) from energy-related activities from 1990 to 2016 are reported in Table 3-75. 11 Table 3-75: NOx, CO, and NMVOC Emissions from Energy-Related Activities (kt) Gas/Activity 1990 2005 2012 2013 2014 2015 2016 NOx 21,106 16.602 11,271 10,747 10,161 9,323 8,352 Mobile Fossil Fuel Combustion 10,862 10.295 6,871 6,448 6,024 5,417 4,814 Stationary Fossil Fuel Combustion 10,023 5.858 3,655 3,504 3,291 3,061 2,692 Oil and Gas Activities 139 321 663 704 745 745 745 Waste Combustion 82 128 82 91 100 100 100 International Bunker Fuels" 1,956 1,70-1 1,398 1,139 1,139 1,226 1,313 CO 125,640 64,'M5 42,164 40,239 38,315 36,348 34,401 Mobile Fossil Fuel Combustion 119,360 58.615 36,153 34,000 31,848 29,881 27,934 Stationary Fossil Fuel Combustion 5,000 4.648 4,027 3,884 3,741 3,741 3,741 Waste Combustion 978 1.403 1,318 1,632 1,947 1,947 1,947 Oil and Gas Activities 302 318 666 723 780 780 780 International Bunker Fuels" 103 133 133 129 135 141 143 NMVOCs 12,620 7,m 7,558 7,357 7,154 6,867 6,581 Mobile Fossil Fuel Combustion 10,932 5.724 4,243 3,924 3,605 3,318 3,032 Oil and Gas Activities 554 510 2,651 2,786 2,921 2,921 2,921 Stationary Fossil Fuel Combustion 912 716 569 539 507 507 507 Waste Combustion 222 241 94 108 121 121 121 International Bunker Fuels" 57 5-1 46 41 42 47 49 a These values are presented for informational purposes only and are not included in totals. Note: Totals may not sum due to independent rounding. 12 Methodology 13 Emission estimates for 1990 through 2016 were obtained from data published on the National Emission Inventory 14 (NEI) Air Pollutant Emission Trends web site (EPA 2016), and disaggregated based on EPA (2003). Emission 15 estimates for 2012 and 2013 for non-electric generating units (EGU) were updated to the most recent available data 16 in EPA (2016). Emission estimates for 2012 and 2013 for non-mobile sources are recalculated emissions by 17 interpolation from 2016 in EPA (2016). Emissions were calculated either for individual categories or for many 18 categories combined, using basic activity data (e.g., the amount of raw material processed) as an indicator of 19 emissions. National activity data were collected for individual applications from various agencies. 20 Activity data were used in conjunction with emission factors, which together relate the quantity of emissions to the 21 activity. Emission factors are generally available from the EPA's Compilation of Air Pollutant Emission Factors, 22 AP-42 (EPA 1997). The EPA currently derives the overall emission control efficiency of a source category from a 23 variety of information sources, including published reports, the 1985 National Acid Precipitation and Assessment 24 Program emissions inventory, and other EPA databases. 3-96 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2016 ------- 1 Uncertainty and Time-Series Consistency 2 Uncertainties in these estimates are partly due to the accuracy of the emission factors used and accurate estimates of 3 activity data. A quantitative uncertainty analysis was not performed. 4 Methodological recalculations were applied to the entire time series to ensure time-series consistency from 1990 5 through 2016. Details on the emission trends through time are described in more detail in the Methodology section, 6 above. 7 3.10 International Bunker Fuels (CRF Source 8 Category 1: Memo Items) 9 Emissions resulting from the combustion of fuels used for international transport activities, termed international 10 bunker fuels under the UNFCCC, are not included in national emission totals, but are reported separately based upon 11 location of fuel sales. The decision to report emissions from international bunker fuels separately, instead of 12 allocating them to a particular country, was made by the Intergovernmental Negotiating Committee in establishing 13 the Framework Convention on Climate Change.100 These decisions are reflected in the IPCC methodological 14 guidance, including IPCC (2006), in which countries are requested to report emissions from ships or aircraft that 15 depart from their ports with fuel purchased within national boundaries and are engaged in international transport 16 separately from national totals (IPCC 2006).101 17 Two transport modes are addressed under the IPCC definition of international bunker fuels: aviation and marine.102 18 Greenhouse gases emitted from the combustion of international bunker fuels, like other fossil fuels, include CO2, 19 CH4 and N20 for marine transport modes, and CO2 and N20 for aviation transport modes. Emissions from ground 20 transport activities—by road vehicles and trains—even when crossing international borders are allocated to the 21 country where the fuel was loaded into the vehicle and, therefore, are not counted as bunker fuel emissions. 22 The 2006 IPCC Guidelines distinguish between different modes of air traffic. Civil aviation comprises aircraft used 23 for the commercial transport of passengers and freight, military aviation comprises aircraft under the control of 24 national armed forces, and general aviation applies to recreational and small corporate aircraft. The 2006 IPCC 25 Guidelines further define international bunker fuel use from civil aviation as the fuel combusted for civil (e.g., 26 commercial) aviation purposes by aircraft arriving or departing on international flight segments. However, as 27 mentioned above, and in keeping with the 2006 IPCC Guidelines, only the fuel purchased in the United States and 28 used by aircraft taking-off (i.e., departing) from the United States are reported here. The standard fuel used for civil 29 aviation is kerosene-type jet fuel, while the typical fuel used for general aviation is aviation gasoline.103 30 Emissions of CO2 from aircraft are essentially a function of fuel use. Nitrous oxide emissions also depend upon 31 engine characteristics, flight conditions, and flight phase (i.e., take-off, climb, cruise, decent, and landing). Recent 32 data suggest that little or no CH4 is emitted by modern engines (Anderson et al. 2011), and as a result, CH4 33 emissions from this category are considered zero. Injet engines, N20 is primarily produced by the oxidation of 34 atmospheric nitrogen, and the majority of emissions occur during the cruise phase. International marine bunkers 35 comprise emissions from fuels burned by ocean-going ships of all flags that are engaged in international transport. 1°° 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). 101 Note that the definition of international bunker fuels used by the UNFCCC differs from that used by the International Civil Aviation Organization. 102 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). 103 Naphtha-type jet fuel was used in the past by the military in turbojet and turboprop aircraft engines. Energy 3-97 ------- 1 Ocean-going ships are generally classified as cargo and passenger carrying, military (i.e., U.S. Navy), fishing, and 2 miscellaneous support ships (e.g., tugboats). For the purpose of estimating greenhouse gas emissions, international 3 bunker fuels are solely related to cargo and passenger carrying vessels, which is the largest of the four categories, 4 and military vessels. Two main types of fuels are used on sea-going vessels: distillate diesel fuel and residual fuel 5 oil. Carbon dioxide is the primary greenhouse gas emitted from marine shipping. 6 Overall, aggregate greenhouse gas emissions in 2016 from the combustion of international bunker fuels from both 7 aviation and marine activities were 115.5 MMT CO2 Eq., or 10.5 percent above emissions in 1990 (see Table 3-76 8 and Table 3-77). Emissions from international flights and international shipping voyages departing from the United 9 States have increased by 88.9 percent and decreased by 35.1 percent, respectively, since 1990. The majority of these 10 emissions were in the form of CO2; however, small amounts of CH4 (from marine transport modes) and N20 were 11 also emitted. 12 Table 3-76: CO2, ChU, and N2O Emissions from International Bunker Fuels (MMT CO2 Eq.) Gas/Mode 1990 2005 2012 2013 2014 2015 2016 CO2 103.5 113.1 105.8 99.8 103.4 110.9 114.4 Aviation 38.0 60.1 64.5 65.7 69.6 71.9 71.9 Commercial 30.0 55.6 61.4 62.8 66.3 68.6 68.6 Military 8.1 4.5 3.1 2.9 3.3 3.3 3.3 Marine 65.4 53.0 41.3 34.1 33.8 38.9 42.5 CH4 0.2 0.1 0.1 0.1 0.1 0.1 0.1 Aviation8 0.0 0.0 0.0 0.0 0.0 0.0 0.0 Marine 0.2 0.1 0.1 0.1 0.1 0.1 0.1 N2O 0.9 1.0 0.9 0.9 0.9 0.9 1.0 Aviation 0.4 0.6 0.6 0.6 0.7 0.7 0.7 Marine 0.5 0.4 0.3 0.2 0.2 0.3 0.3 Total 104.5 114.2 106.8 100.7 104.4 111.9 115.5 + Does not exceed 0.05 MMT CO2 Eq. a CH4 emissions from aviation are estimated to be zero. Notes: Totals may not sum due to independent rounding. Includes aircraft cruise altitude emissions. 13 Table 3-77: CO2, ChU, and N2O Emissions from International Bunker Fuels (kt) Gas/Mode 1990 2005 2012 2013 2014 2015 2016 CO2 103,463 113,139 105,805 99,763 103,400 110,887 114,394 Aviation 38,034 60,125 64,524 65,664 69,609 71,942 71,859 Marine 65,429 53,014 41,281 34,099 33,791 38,946 42,535 CH4 7 5 4 3 3 3 4 Aviation3 0 0 0 0 0 0 0 Marine 7 5 4 3 3 3 4 N2O 3 3 3 3 3 3 3 Aviation 1 2 2 2 2 2 2 Marine 2 1 1 1 1 1 1 aCH4 emissions from aviation are estimated to be zero. Notes: Totals may not sum due to independent rounding. Includes aircraft cruise altitude emissions. 14 Methodology 15 Emissions of CO2 were estimated by applying C content and fraction oxidized factors to fuel consumption activity 16 data. This approach is analogous to that described under Section 3.1- CO2 from Fossil Fuel Combustion. Carbon 17 content and fraction oxidized factors for jet fuel, distillate fuel oil, and residual fuel oil were taken directly from EIA 18 and are presented in Annex 2.1, Annex 2.2, and Annex 3.8 of this Inventory. Density conversions were taken from 19 Chevron (2000), ASTM (1989), and USAF (1998). Heat content for distillate fuel oil and residual fuel oil were 20 taken from EIA (2017) and USAF (1998), and heat content for jet fuel was taken from EIA (2017). A complete 21 description of the methodology and a listing of the various factors employed can be found in Annex 2.1. See Annex 3-98 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2016 ------- 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 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 N20 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 N20 emissions were obtained from the 2006IPCC 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 N20 (IPCC 2006). For marine vessels consuming either distillate diesel or residual fuel oil the following values (g/MJ), were employed: 0.32 for CH4 and 0.08 for N:0. Activity data for aviation included solely jet fuel consumption statistics, while the marine mode included both distillate diesel and residual fuel oil. 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, 2000 through 2016 as modeled with the Aviation Enviromnental 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 2006 IPCC Guidelines (IPCC 2006). International aviation CO: estimates for 1990 and 2000 through 2016 are obtained from FAA's AEDT model (FAA 2017). The radar-informed method that was used to estimate CO: emissions for commercial aircraft for 1990, and 2000 through 2016 is not possible for 1991 through 1999 because the radar data set is not available for years prior to 2000. FAA developed OAG schedule-informed inventories modeled with AEDT and great circle trajectories for 1990, 2000 and 2010. Because fuel consumption and CO: 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. 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 Enviromnent), 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 2017). 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 3-78. See Annex 3.8 for additional discussion of military data. Activity data on distillate diesel and residual fuel oil consumption by cargo or passenger carrying marine vessels departing from U.S. ports were taken from unpublished data collected by the Foreign Trade Division of the U.S. Department of Commerce's Bureau of the Census (DOC 2017) for 1990 through 2001, 2007 through 2016, 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 (2017). 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 3-79. Energy 3-99 ------- 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 Table 3-78: Aviation Jet Fuel Consumption for International Transport (Million Gallons) Nationality 1990 2005 2012 2013 2014 2015 2016 U.S. and Foreign Carriers 3,222 5,983 6,604 6,748 7,126 7,383 7,383 U.S. Military 862 462 321 294 339 341 333 Total 4,084 6,445 6,925 7,042 7,465 7,725 7,716 Table 3-79: Marine Fuel Consumption for International Transport (Million Gallons) Fuel Type 1990 2005 2012 2013 2014 2015 2016 Residual Fuel Oil 4,781 3,881 3,069 2,537 2,466 2,718 3,011 Distillate Diesel Fuel & Other 617 444 280 235 261 492 534 U.S. Military Naval Fuels 522 471 381 308 331 326 314 Total 5,920 4,796 3,730 3,081 3,058 3,536 3,858 Note: Totals may not sum due to independent rounding. Uncertainty and Time-Series Consistency 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.104 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, and in the activity data on military operations and training that were used to estimate percentages of total fuel use reported as bunker fuel emissions. Total aircraft and ship fuel use estimates were developed from DoD records, which document fuel sold to the Navy and Air Force from the Defense Logistics Agency. These data may slightly over or under estimate actual total fuel use in aircraft and ships because each Service may have procured fuel from, and/or may have sold to, traded with, and/or given fuel to other ships, aircraft, governments, or other entities. There are uncertainties in 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 2001. 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, Department and military Service data, and expert judgments. The magnitude of the potential errors 104 See uncertainty discussions under Carbon Dioxide Emissions from Fossil Fuel Combustion. 3-100 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2016 ------- 1 related to the various uncertainties has not been calculated, but is believed to be small. The uncertainties associated 2 with future military bunker fuel emission estimates could be reduced through additional data collection. 3 Although aggregate fuel consumption data have been used to estimate emissions from aviation, the recommended 4 method for estimating emissions of gases other than CO2 in the 2006IPCC Guidelines (IPCC 2006) is to use data by 5 specific aircraft type, number of individual flights and, ideally, movement data to better differentiate between 6 domestic and international aviation and to facilitate estimating the effects of changes in technologies. The IPCC also 7 recommends that cruise altitude emissions be estimated separately using fuel consumption data, while landing and 8 take-off (LTO) cycle data be used to estimate near-ground level emissions of gases other than CO2.105 9 There is also concern regarding the reliability of the existing DOC (2017) data on marine vessel fuel consumption 10 reported at U.S. customs stations due to the significant degree of inter-annual variation. 11 Methodological recalculations were applied to the entire time series to ensure time-series consistency from 1990 12 through 2016. Details on the emission trends through time are described in more detail in the Methodology section, 13 above. 14 QA/QC and Verification 15 A source-specific QA/QC plan for international bunker fuels was developed and implemented. This effort included a 16 general analysis, as well as portions of a category specific analysis. The category specific procedures that were 17 implemented involved checks specifically focusing on the activity data and emission factor sources and 18 methodology used for estimating CO2, CH4, and N20 from international bunker fuels in the United States. Emission 19 totals for the different sectors and fuels were compared and trends were investigated. No corrective actions were 20 necessary. 21 Planned Improvements 22 The feasibility of including data from a broader range of domestic and international sources for bunker fuels, 23 including data from studies such as the Third IMO GHG Study 2014 (IMO 2014), is being considered. 24 3.11 Wood Biomass and Biofuefs 25 Consumption (CRF Source Category 1A) 26 The combustion of biomass fuels such as wood, charcoal, and wood waste and biomass-based fuels such as ethanol, 27 biogas, and biodiesel generates CO2 in addition to CH4 and N20 already covered in this chapter. In line with the 28 reporting requirements for inventories submitted under the UNFCCC, CO2 emissions from biomass combustion 29 have been estimated separately from fossil fuel CO2 emissions and are not directly included in the energy sector 30 contributions to U.S. totals. In accordance with IPCC methodological guidelines, any such emissions are calculated 31 by accounting for net carbon (C) fluxes from changes in biogenic C reservoirs in wooded or crop lands. For a more 32 complete description of this methodological approach, see the Land Use, Land-Use Change, and Forestry chapter 105 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. The estimates in Mobile Combustion are also likely to include emissions from ocean-going vessels departing from U.S. ports on international voyages. Energy 3-101 ------- 1 (Chapter 6), which accounts for the contribution of any resulting CO2 emissions to U.S. totals within the Land Use, 2 Land-Use Change, and Forestry sector's approach. 3 Therefore, CO2 emissions from wood biomass and biofuel consumption are not included specifically in summing 4 energy sector totals and are instead included in net carbon fluxes from changes in biogenic carbon reservoirs in the 5 estimates for Land Use, Land-Use Change, and Forestry. However, they are presented here for informational 6 purposes and to provide detail on wood biomass and biofuels consumption. 7 In 2016, total CO2 emissions from the burning of woody biomass in the industrial, residential, commercial, and 8 electric power sectors were approximately 190.2 MMT CO2 Eq. (190,171 kt) (see Table 3-80 and Table 3-81). As 9 the largest consumer of woody biomass, the industrial sector was responsible for 63.3 percent of the CO2 emissions 10 from this source. The residential sector was the second largest emitter, constituting 20.2 percent of the total, while 11 the commercial and electric power sectors accounted for the remainder. 12 Table 3-80: CO2 Emissions from Wood Consumption by End-Use Sector (MMT CO2 Eq.) End-Use Sector 1990 2005 2012 2013 2014 2015 2016 Industrial 135.3 136.3 125.7 123.1 124.4 122.6 120.4 Residential 59.8 44.3 43.3 59.8 60.9 45.4 38.4 Commercial 6.8 7.2 6.3 7.2 7.8 8.4 8.5 Electric Power 13.3 19.1 19.6 21.4 25.9 25.1 22.9 Total 215.2 206.9 194.9 211.6 218.9 201.5 190.2 Note: Totals may not sum due to independent rounding. 13 Table 3-81: CO2 Emissions from Wood Consumption by End-Use Sector (kt) End-Use Sector 1990 2005 2012 2013 2014 2015 2016 Industrial 135,348 136,269 125,724 123,149 124,369 122,575 120,417 Residential 59,808 44,340 43,309 59,808 60,884 45,359 38,419 Commercial 6,779 7,218 6,257 7,235 7,760 8,377 8,457 Electric Power 13,252 19,074 19,612 21,389 25,908 25,146 22,878 Total 215,186 206,901 I'M,903 211,581 218,922 201,457 190,171 Note: Totals may not sum due to independent rounding. 14 The transportation sector is responsible for most of the fuel ethanol consumption in the United States. Ethanol used 15 for fuel is currently produced primarily from corn grown in the Midwest, but it can be produced from a variety of 16 biomass feedstocks. Most ethanol for transportation use is blended with gasoline to create a 90 percent gasoline, 10 17 percent by volume ethanol blend known as E-10 or gasohol. 18 In 2016, the United States transportation sector consumed an estimated 1,186.9 trillion Btu of ethanol, and as a 19 result, produced approximately 81.2 MMT CO2 Eq. (81,250 kt) (see Table 3-82 and Table 3-83) of CO2 emissions. 20 Ethanol fuel production and consumption has grown significantly since 1990 due to the favorable economics of 21 blending ethanol into gasoline and federal policies that have encouraged use of renewable fuels. 22 Table 3-82: CO2 Emissions from Ethanol Consumption (MMT CO2 Eq.) End-Use Sector 1990 2005 2012 2013 2014 2015 2016 Transportation3 4.1 22.4 71.5 73.4 74.9 75.9 78.2 Industrial 0.1 0.5 1.1 1.2 1.0 1.2 1.2 Commercial 0.0 0.1 0.2 0.2 0.2 1.8 1.8 Total 4.2 22.9 72.8 74.7 76.1 78.9 81.2 + Does not exceed 0.05 MMT CO2 Eq. a See Annex 3.2, Table A-95 for additional information on transportation consumption of these fuels. Note: Totals may not sum due to independent rounding. 23 Table 3-83: CO2 Emissions from Ethanol Consumption (kt) End-Use Sector 1990 2005 2012 2013 2014 2015 2016 Transportation3 4,136 22,414 4 71,510 73,359 74,857 75,946 78,174 3-102 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2016 ------- 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 Industrial 56 468 1,142 1,202 970 1,203 1,238 Commercial 34 60 175 183 249 1,785 1,838 Total 4,227 22,943 72,827 74,743 76,075 78,934 81,250 a See Annex 3.2, Table A-95 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 2017a). 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 2017b). In 2016, the United States consumed an estimated 266.1 trillion Btu of biodiesel, and as a result, produced approximately 19.6 MMT CO2 Eq. (19,648 kt) (see Table 3-84 and Table 3-85) 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 2017b). There was no measured biodiesel consumption prior to 2001 EIA (2017a). Table 3-84: CO2 Emissions from Biodiesel Consumption (MMT CO2 Eq.) End-Use Sector 1990 2005 2012 2013 2014 2015 2016 Transportation3 0.0 0.9 8.5 13.5 13.3 14.1 19.6 Total 0.0 0.9 8.5 13.5 13.3 14.1 19.6 + Does not exceed 0.05 MMT CO2 Eq. a See Annex 3.2, Table A-95 for additional information on transportation consumption of these fuels. Note: Totals may not sum due to independent rounding. Table 3-85: CO2 Emissions from Biodiesel Consumption (kt) End-Use Sector 1990 2005 2012 2013 2014 2015 2016 Transportation3 0 856 8,470 13,462 13,349 14,077 19,648 Total 0 856 8,470 13,462 13,349 14,077 19,648 a See Annex 3.2, Table A-95 for additional information on transportation consumption of these fuels. Note: Totals may not sum due to independent rounding. Methodology Woody biomass emissions were estimated by applying two EIA gross heat contents (Lindstrom 2006) to U.S. consumption data (EIA 2017a) (see Table 3-86), provided in energy units for the industrial, residential, commercial, and electric generation 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. It was assumed that the woody biomass contains black liquor and other wood wastes, has a moisture content of 12 percent, and is converted into CO2 with 100 percent efficiency. The emissions from ethanol consumption were calculated by applying an emission factor of 18.7 MMT C/QBtu (EPA 2010) to U.S. ethanol consumption estimates that were provided in energy units (EIA 2017a) (see Table 3-87). The emissions from biodiesel consumption were calculated by applying an emission factor of 20.1 MMT C/QBtu (EPA 2010) to U. S. biodiesel consumption estimates that were provided in energy units (EIA 2017a) (see Table 3-88). Table 3-86: Woody Biomass Consumption by Sector (Trillion Btu) End-Use Sector 1990 2005 2012 2013 2014 2015 2016 Industrial 1,441.9 1,451.7 1,339.4 1,312.0 1,325.0 1,305.8 1,282.9 Residential 580.0 430.0 420.0 580.0 590.4 439.9 372.6 Commercial 65.7 70.0 60.7 70.2 75.3 81.2 82.0 Energy 3-103 ------- Electric Power 128.5 185.0 190.2 207.4 251.3 243.9 221.9 Total 2,216.2 2,136.7 2,010.3 2,169.5 2,241.9 2,070.8 1,959.3 Note: Totals may not sum due to independent rounding. 1 Table 3-87: Ethanol Consumption by Sector (Trillion Btu) End-Use Sector 1990 2005 2012 2013 2014 2015 2016 Transportation 60.4 327.4 1.044.6 1,071.6 1,093.5 1,109.4 1,142.0 Industrial 0.8 6.8 16.7 17.6 14.2 17.6 18.1 Commercial 0.5 0.9 2.6 2.7 3.6 26.1 26.8 Total 61.7 335.1 I.I 163.8 1,091.8 1,111.3 1,153.1 1,186.9 Note: Totals may not sum due to independent rounding. 2 Table 3-88: Biodiesel Consumption by Sector (Trillion Btu) End-Use Sector 1990 2005 2012 2013 2014 2015 2016 Transportation 0.0 11.6 1 114.7 182.3 180.8 190.6 266.1 Total 0.0 11.6 114.7 182.3 180.8 190.6 266.1 Note: Totals may not sum due to independent rounding. 3 Uncertainty and Time-Series Consistency 4 It is assumed that the combustion efficiency for woody biomass is 100 percent, which is believed to be an 5 overestimate of the efficiency of wood combustion processes in the United States. Decreasing the combustion 6 efficiency would decrease emission estimates for CO2. Additionally, the heat content applied to the consumption of 7 woody biomass in the residential, commercial, and electric power sectors is unlikely to be a completely accurate 8 representation of the heat content for all the different types of woody biomass consumed within these sectors. 9 Emission estimates from ethanol and biodiesel production are more certain than estimates from woody biomass 10 consumption due to better activity data collection methods and uniform combustion techniques. 11 Methodological recalculations were applied to the entire time series to ensure time-series consistency from 1990 12 through 2016. Details on the emission trends through time are described in more detail in the Methodology section, 13 above. 14 Recalculations Discussion 15 Ethanol and biodiesel values for 1990 through 2015 were not revised relative to the previous Inventory, as there 16 were no historical revisions fromEIA's Monthly Energy Review (EIA 2017a). 17 Planned Improvements 18 Future research will look into the availability of data on woody biomass heat contents and carbon emission factors 19 the see if there are newer, improved data sources available for these factors. 20 The availability of facility-level combustion emissions through EPA's GHGRP will be examined to help better 21 characterize the industrial sector's energy consumption in the United States, and further classify woody biomass 22 consumption by business establishments according to industrial economic activity type. Most methodologies used in 23 EPA's GHGRP are consistent with IPCC, though for EPA's GHGRP, facilities collect detailed information specific 24 to their operations according to detailed measurement standards, which may differ with the more aggregated data 25 collected for the Inventory to estimate total, national U.S. emissions. In addition, and unlike the reporting 26 requirements for this chapter under the UNFCCC reporting guidelines, some facility-level fuel combustion 3-104 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2016 ------- 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 emissions reported under EPA's GHGRP may also include industrial process emissions.106 In line with 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.107 Currently emission estimates from biomass and biomass-based fuels included in this inventory are limited to woody biomass, ethanol, and biodiesel. Other forms of biomass-based fuel consumption include biogas. An effort will be made to examine sources of data for biogas including data from EIA for possible inclusion. EIA (2017a) 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. As per discussion in Section 3.1, an additional planned improvement is to evaluate and potentially update EPA's method for allocating motor gasoline consumption across the Transportation, Industrial and Commercial sectors to improve accuracy and create a more consistent time series. Further research will be conducted to determine if changes also need to be made to ethanol allocation between these sectors to match gasoline's sectoral distribution. 106 See . 107 See . Energy 3-105 ------- 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 4. Industrial Processes and Product Use The 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. 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. 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 (N20), and fluorinated GHGs (e.g., HFC-23). The GHG byproduct generating processes included in this chapter include iron and steel production and metallurgical coke production, cement production, lime production, other process uses of carbonates (e.g., flux stone, flue gas desulfurization, and glass manufacturing), ammonia production and urea consumption, petrochemical production, aluminum production, HCFC-22 production, soda ash production and use, titanium dioxide production, ferroalloy production, glass production, zinc production, phosphoric acid production, lead production, silicon carbide production and consumption, nitric acid production, adipic acid production, and caprolactam production. Greenhouse gases that are used in manufacturing processes or by end-consumers include man-made compounds such as hydrofluorocarbons (HFCs), perfluorocarbons (PFCs), sulfur hexafluoride (SF6), 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 accumulate in the atmosphere as long as emissions continue. 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 is growing rapidly 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. Hydrofluorocarbons, PFCs, SF6, and NF3 are employed and emitted by a number of other industrial sources in the United States, such as semiconductor manufacture, electric power transmission and distribution, 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 semiconductor manufacturing and anesthetic and aerosol applications. In 2016, IPPU generated emissions of 375.7 million metric tons of CO2 equivalent (MMT CO2 Eq.), or 5.7 percent of total U.S. greenhouse gas emissions. Carbon dioxide emissions from all industrial processes were 163.6 MMT CO2 Eq. (163,647 kt CO2) in 2016, or 3.1 percent of total U.S. CO2 emissions. Methane emissions from industrial processes resulted in emissions of approximately 0.2 MMT CO2 Eq. (8 kt CH4) in 2016, which was less than 1 percent of U.S. CH4 emissions. Nitrous oxide emissions from IPPU were 23.7 MMT CO2 Eq. (79 kt N2O) in 2016, or 6.4 percent of total U.S. N20 emissions. In 2016 combined emissions of HFCs, PFCs, SF6, and NF3 totaled 188.2 MMT CO2 Eq. Total emissions from IPPU in 2016 were 10.3 percent more than 1990 emissions. Indirect greenhouse gas emissions also result from IPPU, and are presented in Table 4-111 in kilotons (kt). Industrial Processes and Product Use 4-1 ------- 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 Figure 4-1: 2016 Industrial Processes and (MMT COz Eq.) Product Use Chapter Greenhouse Gas Sources Substitution of Ozone Depleting Substances Iron and Steel Production & Metallurgical Coke Production Cement Production Petrochemical Production Lime Production Other Process Uses of Carbonates Ammonia Production Nitric Acid Production Adipic Acid Production Semiconductor Manufacture Carbon Dioxide Consumption Electrical Transmission and Distribution NiO from Product Uses Urea Consumption for Non-Agricultural Purposes HCFC-22 Production Aluminum Production Caprolactam, Glyoxal, and Glyoxylic Acid Production Ferroalloy Production Soda Ash Production Titanium Dioxide Production Glass Production Magnesium Production and Processing Phosphoric Acid Production Zinc Production Lead Production Silicon Carbide Production and Consumption 174 ¦ ¦ ¦ I I I I I < 0.5 Industrial Processes and Product Use as a Portion of all Emissions 5.7% 10 20 30 40 50 60 70 MMT CO, Eq. The increase in overall IPPU emissions since 1990 reflects a range of emission trends among the emission sources. 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 production (in the case of iron and steel) and to improved practices (in the case of PFC emissions from aluminum production). Similarly, CO2 and CH4 emissions from many chemical production sources have either decreased or not changed significantly since 1990, with the exception of petrochemical production which lias steadily increased. Emissions from mineral sources have either increased (e.g., cement manufacturing) or not changed significantly (e.g., glass and lime manufacturing) since 1990 but largely follow economic cycles. Hydrofluorocarbon emissions from the substitution of ODS have increased drastically since 1990, while the emissions of HFCs, PFCs, SF6, and NF3 from other sources have generally declined. Nitrous oxide emissions from the production of adipic and nitric acid have decreased, while N2O emissions from product uses have remained nearly constant over time. Trends are explained further within each emission source category throughout the chapter. Table 4-1 summarizes emissions for the IPPU chapter in MMT CO2 Eq. using IPCC Fourth Assessment Report (AR4) GWP values, following the requirements of the revised United Nations Framework Convention on Climate Change (UNFCCC) reporting guidelines for national inventories (IPCC 2007).1 Unweighted native 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 Format (CRF) tables, corresponding generally to: mineral products, chemical production, metal production, and emissions from the uses of HFCs, PFCs, SF6, and NF3. 1 See . 4-2 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2016 ------- 1 Table 4-1: Emissions from Industrial Processes and Product Use (MMT CO2 Eq.) Gas/Source 1990 2005 2012 2013 2014 2015 2016 co2 207.3 190.2 169.9 171.8 177.9 171.4 163.6 Iron and Steel Production & Metallurgical Coke Production 101.5 68.0 55.4 53.3 58.2 47.7 42.2 Iron and Steel Production 99.0 66.0 54.9 51.5 56.2 44.9 40.9 Metallurgical Coke Production 2.5 2.0 0.5 1.8 2.0 2.8 1.3 Cement Production 33.5 46.2 35.3 36.4 39.4 39.9 39.4 Petrochemical Production 21.2 26.8 26.5 26.4 26.5 28.1 27.4 Lime Production 11.7 14.6 13.8 14.0 14.2 13.3 13.3 Other Process Uses of Carbonates 4.9 6.3 8.0 10.4 11.8 11.2 11.2 Ammonia Production 13.0 9.2 9.4 10.0 9.6 10.6 11.2 Carbon Dioxide Consumption 1.5 1.4 4.0 4.2 4.5 4.5 4.5 Urea Consumption for Non- Agricultural Purposes 3.8 3.7 4.4 4.1 1.5 4.2 4.0 Ferroalloy Production 2.2 1.4 1.9 1.8 1.9 2.0 1.8 Soda Ash Production 1.4 1.7 1.7 1.7 1.7 1.7 1.7 Titanium Dioxide Production 1.2 1.8 1.5 1.7 1.7 1.6 1.6 Aluminum Production 6.8 4.1 3.4 3.3 2.8 2.8 1.3 Glass Production 1.5 1.9 1.2 1.3 1.3 1.3 1.3 Phosphoric Acid Production 1.5 1.3 1.1 1.1 1.0 1.0 1.0 Zinc Production 0.6 1.0 1.5 1.4 1.0 0.9 0.9 Lead Production 0.5 0.6 0.5 0.5 0.5 0.5 0.5 Silicon Carbide Production and Consumption 0.4 0.2 0.2 0.2 0.2 0.2 0.2 Magnesium Production and Processing + + + + + + + CH4 0.3 0.1 0.1 0.1 0.2 0.2 0.2 Petrochemical Production 0.2 0.1 0.1 0.1 0.1 0.2 0.2 Ferroalloy Production + + + + + + + Silicon Carbide Production and Consumption + + + + + + + Iron and Steel Production & Metallurgical Coke Production + + + + + + + Iron and Steel Production + + + + + + + Metallurgical Coke Production 0.0 0.0 0.0 0.0 0.0 0.0 0.0 N2O 33.3 24.9 22.4 21.0 22.8 22.3 23.7 Nitric Acid Production 12.1 11.3 10.5 10.7 10.9 11.6 10.2 Adipic Acid Production 15.2 7.1 5.5 3.9 5.4 4.3 7.0 N20 from Product Uses 4.2 4.2 4.2 4.2 4.2 4.2 4.2 Caprolactam, Glyoxal, and Glyoxylic Acid Production 1.7 2.1 2.0 2.0 2.0 2.0 2.0 Semiconductor Manufacturing + 0.1 0.2 0.2 0.2 0.2 0.2 HFCs 46.6 120.0 156.0 159.1 166.8 173.3 177.1 Substitution of Ozone Depleting Substances3 0.3 99.8 150.3 154.8 161.4 168.6 173.9 HCFC-22 Production 46.1 20.0 5.5 4.1 5.0 4.3 2.8 Semiconductor Manufacturing 0.2 0.2 0.2 0.2 0.3 0.3 0.3 Magnesium Production and Processing 0.0 0.0 + 0.1 0.1 0.1 0.1 PFCs 24.3 6.7 5.9 5.8 5.6 5.1 4.3 Semiconductor Manufacturing 2.8 3.3 3.0 2.8 3.1 3.1 3.0 Aluminum Production 21.5 3.4 2.9 3.0 2.5 2.0 1.4 Substitution of Ozone Depleting Substances3 0.0 + + + + + + SF« 28.8 11.7 6.6 6.3 6.3 5.9 6.2 Electrical Transmission and Distribution 23.1 8.3 4.6 4.5 4.6 4.2 4.3 Industrial Processes and Product Use 4-3 ------- Magnesium Production and Processing 5.2 2.7 Semiconductor Manufacturing 0.5 0.7 NFs + 0.5 Semiconductor Manufacturing + 0.5 1.6 0.3 0.6 0.6 1.5 0.4 0.6 0.6 1.0 0.7 0.5 0.5 0.9 0.7 0.6 0.6 1.0 0.8 0.6 0.6 Total 340.5 354.2 361.6 364.7 380.2 378.8 375.7 + Does not exceed 0.05 MMT CO2 Eq. a Small amounts of PFC emissions also result from this source. Note: Totals may not sum due to independent rounding. 1 Table 4-2: Emissions from Industrial Processes and Product Use (kt) Gas/Source 1990 2005 2012 2013 2014 2015 2016 CO2 207,281 190,171 169,888 171,841 177,906 171,439 163,647 Iron and Steel Production & Metallurgical Coke Production 101,48" 68,047 55,449 53,348 58,234 47,718 42,219 Iron and Steel Production 98,984 66,003 54,906 51,525 56,220 44,879 40,896 Metallurgical Coke Production 2,503 2,044 543 1,824 2,014 2,839 1,323 Cement Production 33,484 46,194 35,270 36,369 39,439 39,907 39,439 Petrochemical Production 21,20 ^ 26,794 26,501 26,395 26,496 28,062 27,411 Lime Production 11,700 14,552 13,785 14,028 14,210 13,342 13,342 Other Process Uses of Carbonates 4,90" 6,339 8,022 10,414 11,811 11,237 11,237 Ammonia Production 13,04" 9,196 9,377 9,962 9,619 10,571 11,234 Carbon Dioxide Consumption 1,472 1,375 4,019 4,188 4,471 4,471 4,471 Urea Consumption for Non- Agricultural Purposes 3,784 3,653 4,392 4,074 1,541 4,169 3,959 Ferroalloy Production 2,152 1,392 1,903 1,785 1,914 1,960 1,796 Soda Ash Production 1,431 1,655 1,665 1,694 1,685 1,714 1,723 Titanium Dioxide Production 1,195 1,755 1,528 1,715 1,688 1,635 1,608 Aluminum Production 6,831 4,142 3,439 3,255 2,833 2,767 1,334 Glass Production 1,535 1,928 1,248 1,317 1,336 1,299 1,299 Phosphoric Acid Production 1,529 1,342 1,118 1,149 1,038 999 992 Zinc Production 632 1,030 1,486 1,429 956 933 925 Lead Production 516 553 527 546 459 473 482 Silicon Carbide Production and Consumption 375 219 158 169 173 180 174 Magnesium Production and Processing 1 3 2 2 2 3 3 CH4 12 4 4 4 6 9 8 Petrochemical Production 9 3 3 3 5 7 7 Ferroalloy Production 1 + 1 + 1 1 1 Silicon Carbide Production and Consumption 1 + + + + + + Iron and Steel Production & Metallurgical Coke Production 1 1 + + + + + Iron and Steel Production / 1 + + + + + Metallurgical Coke Production 0 0 0 0 0 0 0 N2O 112 84 75 71 77 75 79 Nitric Acid Production 41 38 35 36 37 39 34 Adipic Acid Production 51 24 19 13 18 14 23 NjO from Product Uses 14 14 14 14 14 14 14 Caprolactam, Glyoxal, and Glyoxylic Acid Production 6 7 7 7 7 7 7 Semiconductor Manufacturing + + 1 1 1 1 1 HFCs 1V1 M M M M M M Substitution of Ozone Depleting Substances3 M M M M M M M 4-4 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2016 ------- 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 HCFC-22 Production 1 + + + + + Semiconductor Manufacturing M M M M M M M Magnesium Production and Processing 0 0 ! + + + + + PFCs M M M M M M M Semiconductor Manufacturing M M j M M M M M Aluminum Production M M i M M M M M Substitution of Ozone Depleting Substances3 0 + + + + + + SF« 1 1 + + + + + Electrical Transmission and Distribution 1 + + + + + + Magnesium Production and Processing + + + + + + + Semiconductor Manufacturing + + + + + + + NF3 + + + + + + + Semiconductor Manufacturing - + + + + + + + Does not exceed 0.5 kt. M (Mixture of gases) a Small amounts of PFC emissions also result from this source. Note: Totals may not sum due to independent rounding. The UNFCCC incorporated the 2006IPCC Guidelines for National Greenhouse Gas Inventories (2006IPCC Guidelines) as the standard for Annex I countries at the Nineteenth Conference of the Parties (Warsaw, November 11-23, 2013). This chapter presents emission estimates calculated in accordance with the methodological guidance provided in these guidelines. For additional detail on IPPU sources that are not estimated in this Inventory report, please review Annex 5, Assessment of the Sources and Sinks of Greenhouse Gas Emissions Not Included. These sources are not estimated due to various national circumstances, such as emissions from a source may not be currently occurring in the United States, data are not currently available for those emission sources (e.g., ceramics, non-metallurgical magnesium production), emissions are included elsewhere within the Inventory report, or also that data suggest that emissions are not significant. Information on planned improvements for specific IPPU source categories can be found in the Planned Improvements section of the individual source category. Finally, as mentioned in the Energy chapter of this report (Box 3-6), 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 IPPU, rather than Energy; however, due to national circumstances regarding the allocation of energy statistics and carbon (C) balance data, the United States reports non-energy uses in the Energy chapter of this Inventory. Reporting these non-energy use emissions under IPPU would involve making artificial adjustments to the non-energy use C balance. These artificial adjustments would also result in the C emissions for lubricants, waxes, and asphalt and road oil being reported under IPPU, while the C storage for lubricants, waxes, and asphalt and road oil would be reported under Energy. To avoid presenting an incomplete C balance, double- counting, and adopting a less transparent approach, the entire calculation of C storage and C emissions is therefore conducted in the Non-Energy Uses of Fossil Fuels category calculation methodology and reported under the Energy sector. For more information, see the Methodology section for CO2 from Fossil Fuel Combustion and Section 3.2, Carbon Emitted from Non-Energy Uses of Fossil Fuels. In addition, 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 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 and zinc production) are reported in the IPPU chapter, unless otherwise noted due to specific national circumstances. 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 (CRF Source Category 1 A)) and Annex 2.1, Methodology for Estimating Emissions of CO2 from Fossil Fuel Combustion. Industrial Processes and Product Use 4-5 ------- 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 QA/QC and Verification Procedures For IPPU sources, a detailed QA/QC plan was developed and implemented for specific categories. This plan was based on the overall Quality Assurance/Quality Control and Uncertainty Management Plan for the U.S. Greenhouse Gas Inventory (QA/QC Management Plan), but was tailored to include specific procedures recommended for these sources. Two types of checks were performed using this plan: (1) general (Tier 1) procedures consistent with Volume 1, Chapter 6 of the 2006IPCC 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 to identify significant changes; that, where possible, consistent and reputable data sources are used and specified across sources; that interpolation or extrapolation techniques are consistent across sources; and that common datasets, units, and conversion factors are used where applicable. The IPPU QA/QC plan also checked for transcription errors in data inputs required for emission calculations, including activity data and emission factors; and confirmed that estimates were calculated and reported for all applicable and able portions of the source categories for all years. General or tier 1 QA/QC procedures and calculation-related QC (category-specific, Tier 2) have been performed for all IPPU sources. Consistent with the 2006 IPCC Guidelines, additional category-specific QC procedures were performed for more significant emission categories (such as the comparison of reported consumption with modeled consumption using EPA's Greenhouse Gas Reporting Program (GHGRP) data within Substitution of ODS) or sources where significant methodological and data updates have taken place. The QA/QC implementation did not reveal any significant inaccuracies, and all errors identified were 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, and planned improvement sections. For most IPPU categories, activity data are obtained via aggregation of facility-level data from EPA's GHGRP, national commodity surveys conducted by U.S. Geologic Survey National Minerals Information Center, U.S. Department of Energy (DOE), U.S. Census Bureau, 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. Descriptions of uncertainties and assumptions for activity data and emission factors are included within the uncertainty discussion sections for each IPPU source category. The uncertainty analysis performed to quantify uncertainties associated with the 2016 emission estimates from IPPU continues a multi-year process for developing credible quantitative uncertainty estimates for these source categories using the IPCC Tier 2 approach. As the process continues, the type and the characteristics of the actual probability density functions underlying the input variables are identified and better characterized (resulting in development of more reliable inputs for the model, including accurate characterization of correlation between variables), based primarily on expert judgment. Accordingly, the quantitative uncertainty estimates reported in this section should be considered illustrative and as iterations of ongoing efforts to produce accurate uncertainty estimates. The correlation among data used for estimating emissions for different sources can influence the uncertainty analysis of each individual source. While the uncertainty analysis recognizes very significant connections among sources, a more compiclicnsn c ippio i :h that accounts for all linkages will be identified as the uncertainty analysis moves forward. cal Approach for Estimating and Reporting U.S. Emissions and I In following ilic U nncd Nations Framework Convention on Climate Change (UNFCCC) requirement under Article 4.1 to develop and submit national greenhouse gas emission inventories, 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). Additionally, the calculated emissions and removals in a given year for the United States are presented in a common manner in line with the UNFCCC reporting guidelines for the reporting of inventories under this international agreement. The use of consistent methods to calculate emissions and removals by all nations providing their inventories to the UNFCCC ensures that 4-6 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2016 ------- 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 these reports are comparable. The presentation of emissions and removals provided in this Inventory do not preclude alternative examinations, but rather, this Inventory presents emissions and removals in a common format consistent with how countries are to report Inventories under the UNFCCC. The report itself, and this chapter, follows this standardized format, and provides an explanation of the application of methods used to calculate emissions and removals. Box 4-2: Industrial Processes Data from EPA's Greenhouse Gas Reporting Program On October 30, 2009, the U.S. 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). The rule 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 ("Subparts"). 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 reported for facilities subject to 40 CFR Part 98, though some source categories first reported data for calendar year 2011. EPA's GHGRP dataset and the data presented in this Inventory 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. 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 categories to improve the national estimates presented in this Inventory consistent with IPCC guidelines. While many methodologies used in EPA's GHGRP are consistent with IPCC, it should be noted that the definitions for source categories in EPA's GHGRP may differ from those used in this Inventory in meeting the UNFCCC reporting guidelines (IPCC 2011). In line with the UNFCCC reporting guidelines, the Inventory is a comprehensive accounting of all emissions from source categories identified in the 2006 IPCC Guidelines. Further information on the reporting categorizations in EPA's GHGRP and specific data caveats associated with monitoring methods in EPA's GHGRP lias been provided on the GHGRP website.2 For certain source categories in this Inventory (e.g., nitric acid production, cement production and petrochemical production), EPA has also 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 only publishing data values that meet these aggregation criteria.3 Specific uses of aggregated facility-level data are described in the respective methodological sections. For other source categories in this chapter, as indicated in the respective planned improvements sections, 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. As stated previously in the Introduction chapter, this year EPA has integrated GHGRP information for various Industrial Processes and Product Use categories4 and also identified places where EPA plans to integrate additional GHGRP data in additional categories5 (see those categories' Planned Improvement sections for details). EPA lias paid particular attention to ensuring time-series consistency for major recalculations that have occurred from the incorporation of GHGRP data into these categories, consistent with 2006 IPCC 2 See . 3 U.S. EPA Greenhouse Gas Reporting Program. Developments on Publication of Aggregated Greenhouse Gas Data, November 25, 2014. See . 4 Adipic Acid Production, Aluminum Production, Carbon Dioxide Consumption, Cement Production, Electrical Transmission and Distribution, HCFC-22 Production, Lime Production, Magnesium Production and Processing, Substitution of ODS, Nitric Acid Production, Petrochemical Production, and Semiconductor Manufacture. 5 Ammonia Production, Glass Production and Other fluorinated gas production. Industrial Processes and Product Use 4-7 ------- 1 Guidelines and IPCC Technical Bulletin on Use of Facility-Specific Data in National GHG Inventories.6 EPA 2 verifies annual facility-level reports through a multi-step process to identify potential errors and ensure that data 3 submitted to EPA are accurate, complete, and consistent.7 The GHGRP dataset is a particularly important annual 4 resource and will continue to be important for improving emissions estimates from Industrial Process and Product 5 Use in future Inventory reports. Additionally, EPA's GHGRP has and will continue to enhance QA/QC procedures 6 and assessment of uncertainties within the IPPU categories (see those categories for specific QA/QC details 7 regarding the use of GHGRP data). 8 9 4.1 Cement Production (CRF Source Category 10 2A1) 11 Cement production is an energy- and raw material-intensive process that results in the generation of carbon dioxide 12 (CO2) from both the energy consumed in making the cement and the chemical process itself. Emissions from fuels 13 consumed for energy purposes during the production of cement are accounted for in the Energy chapter. 14 During the cement production process, calcium carbonate (CaCCh) is heated in a cement kiln at a temperature range 15 of about 700 to 1000 degrees Celsius (1,292 to 1,832 degrees Fahrenheit) to form lime (i.e., calcium oxide or CaO) 16 and CO2 in a process known as calcination or calcining. The quantity of CO2 emitted during cement production is 17 directly proportional to the lime content of the clinker. During calcination, each mole of limestone (CaCCh) heated 18 in the clinker kiln forms one mole of lime (CaO) and one mole of CO2: 19 CaC03 + heat -» CaO + C02 20 Next, the lime is combined with silica-containing materials to produce clinker (an intermediate product), with the 21 earlier byproduct CO2 being released to the atmosphere. The clinker is then rapidly cooled to maintain quality, 22 mixed with a small amount of gypsum and potentially other materials (e.g., slag, etc.), and used to make Portland 23 cement.8 24 Carbon dioxide emitted from the chemical process of cement production is the second largest source of industrial 25 CO2 emissions in the United States. Cement is produced in 34 states and Puerto Rico. Texas, California, Missouri, 26 Florida, and Alabama were the leading cement-producing states in 2016 and accounted for almost 50 percent of total 27 U.S. production (USGS 2017). Clinker production in 2016 decreased approximately 1 percent from 2015 levels as 28 cement sales increased significantly in 2016, with much of the increase accounted for by imports. In 2016, U.S. 29 clinker production totaled 75,800 kilotons (EPA 2017). The resulting CO2 emissions were estimated to be 39.4 30 MMT C02 Eq. (39,439 kt) (see Table 4-3). 6 See . 7 See . 8 Approximately three percent of total clinker production is used to produce masonry cement, which is produced using plasticizers (e.g., ground limestone, lime, etc.) and Portland cement (USGS 2011). Carbon dioxide emissions that result from the production of lime used to create masonry cement are included in the Lime Manufacture source category. 4-8 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2016 ------- 1 Table 4-3: CO2 Emissions from Cement Production (MMT CO2 Eq. and kt) Year MMT CO2 Eq. kt 1000 33.5 33,484 2005 46.2 46,104 2012 35.3 35,270 2013 36.4 36,360 2014 30.4 30,430 2015 30.0 30,007 2016 30.4 30,430 2 Greenhouse gas emissions from cement production increased every year from 1991 through 2006 (with the 3 exception of a slight decrease in 1997), but decreased in the following years until 2009. Emissions from cement 4 production were at their lowest levels in 2009 (2009 emissions are approximately 28 percent lower than 2008 5 emissions and 12 percent lower than 1990). Since 2010, emissions have increased by roughly 25 percent. In 2016, 6 emissions from cement production decreased by 1 percent from 2015 levels. 7 Emissions since 1990 have increased by 18 percent. Emissions decreased significantly between 2008 and 2009, due 8 to the economic recession and associated decrease in demand for construction materials. Emissions increased 9 slightly from 2009 levels in 2010, and continued to gradually increase during the 2011 through 2015 time period due 10 to increasing consumption. Emissions in 2016 decreased slightly from 2015 levels. Cement continues to be a critical 11 component of the construction industry; therefore, the availability of public and private construction funding, as well 12 as overall economic conditions, have considerable impact on the level of cement production. 13 Methodology 14 Carbon dioxide emissions were estimated using the Tier 2 methodology from the 2006IPCC Guidelines. The Tier 2 15 methodology was used because detailed and complete data (including weights and composition) for carbonate(s) 16 consumed in clinker production are not available, and thus a rigorous Tier 3 approach is impractical. Tier 2 specifies 17 the use of aggregated plant or national clinker production data and an emission factor, which is the product of the 18 average lime fraction for clinker of 65 percent and a constant reflecting the mass of CO2 released per unit of lime. 19 The U.S. Geological Survey (USGS) mineral commodity expert for cement has confirmed that this is a reasonable 20 assumption for the United States (VanOss 2013a). This calculation yields an emission factor of 0.51 tons of CO2 per 21 ton of clinker produced, which was determined as follows: 22 EFciinker = 0.650 CaO X [(44.01 g/mole CO2) (56.08 g/mole CaO)] = 0.510 tons CCh/ton clinker 23 During clinker production, some of the clinker precursor materials remain in the kiln as non-calcinated, partially 24 calcinated, or fully calcinated cement kiln dust (CKD). The emissions attributable to the calcinated portion of the 25 CKD are not accounted for by the clinker emission factor. The IPCC recommends that these additional CKD CO2 26 emissions should be estimated as two percent of the CO2 emissions calculated from clinker production (when data 27 on CKD generation are not available). Total cement production emissions were calculated by adding the emissions 28 from clinker production to the emissions assigned to CKD (IPCC 2006). 29 Furthermore, small amounts of impurities (i.e., not calcium carbonate) may exist in the raw limestone used to 30 produce clinker. The proportion of these impurities is generally minimal, although a small amount (1 to 2 percent) of 31 magnesium oxide (MgO) may be desirable as a flux. Per the IPCC Tier 2 methodology, a correction for MgO is not 32 used, since the amount of MgO from carbonate is likely very small and the assumption of a 100 percent carbonate 33 source of CaO already yields an overestimation of emissions (IPCC 2006). 34 The 1990 through 2012 activity data for clinker production (see Table 4-4) were obtained from USGS (Van Oss 35 2013a, Van Oss 2013b). Clinker production data for 2013 were also obtained from USGS (USGS 2014). The data 36 were compiled by USGS (to the nearest ton) through questionnaires sent to domestic clinker and cement 37 manufacturing plants, including the facilities in Puerto Rico. During the 1990 through 2015 Inventory report cycle, Industrial Processes and Product Use 4-9 ------- 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 EPA began incorporating clinker production data from EPA's GHGRP to estimate emissions in these respective years. Clinker production values in the current Inventory report utilize GHGRP data for the years 2014, 2015 and 2016 (EPA 2017). More details on how this change compares to USGS reported data can be found in the section on Uncertainty and Time-Series Consistency. Table 4-4: Clinker Production (kt) Year Clinker 1000 64.355 2005 88.783 2012 67,788 2013 60,000 2014 75,800 2015 76,700 201 6 75,800 Notes: Clinker production from 1000 through 2016 includes Puerto Rico. Uncertainty and Time-Series Consistency 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 CaCCb, 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). CKD loss can range from 1.5 to 8 percent depending upon plant specifications. 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 surface area. Because the amount of CO2 reabsorbed is thought to be minimal, it was not estimated. However, see Planned Improvements described below to reassess this assumption by conducting a review to identify recent studies that may provide information or data on reabsorption rates of cement products. Total U.S. clinker production is assumed to have low uncertainty. USGS takes a number of manual steps to review clinker production reported through their voluntary surveys. EPA continues to assess the accuracy of reported clinker production data required by GHGRP Subpart H facilities for current and future Inventory reports. 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, the EPA follows up with facilities to resolve mistakes that may have occurred.9 Facilities are also required to monitor and maintain records of monthly clinker production. 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 GHGRP with existing data from USGS. This was to ensure time-series consistency of the emission estimates presented in the Inventory. For the year 2014, USGS and GHGRP clinker production data showed a difference of approximately 2 percent, while in 2015 and in 2016 that difference decreased to less than 1 percent between the two sets of activity data. This difference resulted in an increase of emissions compared to USGS data (USGS 2016a) by 0.7 MMT CO2 Eq. in 2014 and 0.0 MMT CO2 Eq. in 2015 and in 2016. The results of 1 he \pproach 2 quaulitali\ e uuccriaiiits ; 111; 11\ sis are suniniari/cd 111 Table 4-5 IJased 011 ihc uncertainties associated u 11 li total I S clinker production. I lie ('() emission factor for clinker production, and the emission factor lor additional CO emissions from CKI). ) I(> CO emissions from cement production were estimated to he hem con "— 0 ;nn.l 41 X MMT CO l!q at the l>5 percent confidence le\ el I his confidence le\ el 9 See . 4-10 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2016 ------- 1 indicalcsa laimc <> pcivciil lvlo\\ and (> pcicciil ;ihn\c I lie emission esiimale of 'lM \1\1T('() 2 I t| 3 Table 4-5: Approach 2 Quantitative Uncertainty Estimates for CO2 Emissions from Cement 4 Production (MMT CO2 Eq. and Percent) - TO BE UPDATED FOR FINAL INVENTORY REPORT Souriv (¦;is 21116 remission 1'sii 111:1 lc- (MM'I'CO: i:|KT I.OXUT I |)|KT 1 Sound Bound Bound Bound Cement Prodn ction t ( ) 30.4 37.0 41.8 -6% +6% Range of emission estimates predicted by Monte Carlo Stochastic Simulation for a 95 percent confidence interval. 5 Methodological approaches were applied to the entire time series to ensure time-series consistency from 1990 6 through 2016. Details on the emission trends through time are described in more detail in the Methodology section, 7 above. More information on the consistency in clinker production data and emissions across the time series with the 8 use of GHGRP clinker data for 2014 through 2016 can be found in the Uncertainty and Time-Series Consistency 9 section. 10 For more information on the general QA/QC process applied to this source category, consistent with Volume 1, 11 Chapter 6 of the 2006IPCC Guidelines, see QA/QC and Verification Procedures section in the introduction of the 12 IPPU chapter. 13 Planned Improvements 14 In response to comments from the Portland Cement Association (PCA) and UNFCCC expert technical reviews, EPA 15 is continuing to evaluate and analyze data reported under EPA's GHGRP that would be useful to improve the 16 emission estimates for the Cement Production source category. EPA held a technical meeting with PCA in August 17 2016 to review Inventory methods and available data from the GHGRP data set. Most cement production facilities 18 reporting under EPA's GHGRP use Continuous Emission Monitoring Systems (CEMS) to monitor and report CO2 19 emissions, thus reporting combined process and combustion emissions from kilns. In implementing further 20 improvements and integration of data from EPA's GHGRP, the latest guidance from the IPCC on the use of facility - 21 level data in national inventories will be relied upon, in addition to category specific QC methods recommended by 22 2006 IPCC Guidelines,10 EPA's long-term improvement plan includes continued assessment of the feasibility of 23 using additional GHGRP information, in particular disaggregating aggregated GHGRP emissions consistent with 24 IPCC and UNFCCC guidelines to present both national process and combustion emissions streams. This long-term 25 planned analysis is still in development and has not been updated for this current Inventory. 26 Finally, in response to feedback from PCA during the public review of the draft Inventory in March 2017, EPA 27 plans to meet with PCA to discuss additional long-term improvements to review methods and data used to estimate 28 CO2 emissions from cement production to account for both organic material and magnesium carbonate in the raw 29 material, and to discuss the carbonation that occurs later in the cement product lifecycle. EPA will work with PCA 30 to identify data and studies on the average MgO content of clinker produced in the United States, the average carbon 31 content for organic materials in kiln feed in the United States, and CO2 reabsorption rates via carbonation for various 32 cement products. 10 See . Industrial Processes and Product Use 4-11 ------- 1 2 3 4 I.inic is ;in iniporimil ni;iiiiil';icliircd product wiili iii;in\ iiidusiri;il. chcniic;il. ;uid cu\ iroiinicui;il ;ipphc;ilious l.unc 5 production iu\ol\cs lliree 111:1111 processes' sioue prcp;ir;ilioii. ji 11:11 k>ii. ;iud hulr;ilioii Ciirbou dio\idc (CO ) is 6 ucucriilcd diiriuu ilie c;ilciu;iliou si;iuc. w lieu limestone niosiK :isictl :il liiuli 7 icnipcmiiircs 111 ;i kiln lo produce ji 11111 o\ide (( :K )) ;nid C() I lie CO is m\cu i»IT;is ;i u;is ;md is ik>i'iii:iIl\ 8 eniilled In I he ;iiniosphcrc 9 C.uCO ¦ — CuO + CO. 10 Sonic ii|" 1 lie ('() uc 1 ici": 11ctl diiiinu 1 lie production process. Iriwe\cr. is rcco\ crcd ;il some f;ieililies for use 111 su&ir 11 rcfiiiiim ;md preeipil;iled e;ileiiim c;irbou;ilc (l'C(') prodiielion 11 IEmissions from fuels consumed lor ciicrus 12 purposes durum ilie prodiielion ol" lime :ire ;ieeouiiied lor 1111 lie l!ueru\ eli;ipier 13 I"nr I S. operations. ilie lerm "lime " ;ielu;ill\ refers 10 :i \:irielv of elieniienl eonipouuds These include C;iO. or 14 hmh-c;ilcuini quicklime. c;ilciuni h\dro\ide (( ;i(OI 11 1. or h\dr;iled lime, dolomiiic (|iucklime (|C;i( )»\1u()|i. ;md 15 dolomiiic 11> tlr: 11e (|( ;i(OI I) •\1uO| or |( :i(OI 11 'MuiOl I) 11 16 The currciii lime nuirkel is ;ippro\ini;iicl> disirihuied ;icioss l'i\e cud-use ciilcuories ;is follows nicl;illurmc;il uses. 17 perccui. eu\ iroiinicui;il uses. "51 pcrcciii. chcniic;il ;uid uidiisin;il uses. 22 percciii. coiisirucliou uses. l) pereeui. 18 ;uid relr;ielor\ doloniiie. I pereeui (I S( iS 2d I (ih 1 .The ni;ijor uses :ire 111 si eel ni;ikiim. Hue u:is desiill'uri/;iiioii 19 svsienis ;ii eo;il-l'ired eleeirie power phnils. coiisirucliou. ;iud w;iler ire;iinieul. ;is well ;is uses 111 milium, pulp ;iud 20 p;iper;uid preeipil;iled ciilciuni e;irhou;ile iii;iiiiil';ieluriim l.inie is ;ilso used ;is ;i CO scrubber. ;iud llierc h;is hcen 21 c\pcriniciii;iliou 011 ilie use of lime lo c:ipiure ('() from elcciric power pl;uils 22 l.inie prodiielion 111 ilie I niled Si:iics iucludiim I'ucrio kico w;is reported lo be IX.2~lH\iloloiis 111 2D 15 23 (Coniihers 2(>l~i. \i \c: 1 r-e 1 it! 2<)I5. llierc were opcriiiuii: pruii;ir\ lime phnils 1111 lie I lined Si;iics. iiicludiim 24 I'ucrio kico 1 - Prineip:il lime producing suites ;irc Missouri. \l:ih:i 111:1. Kentucky. Ohio. Tc\:is (I S(iS 2ul<>;n 25 I S. lime production resulted 111 estimated ucl CO emissions of H ' MMICO Lq (I V ^42 kn (see I :ible 4-(> ;md 26 T;ihle 4-~i The ireuds 111 C() emissions from lime prodiielion ;ire dircclK proporiiou;il lo ireuds 111 production. 27 w liieli ;ire described below 28 Table 4-6: CO2 Emissions from Lime Production (MMT CO2 Eq. and kt) ^ i ;i r ,MM"I" CO; i:<|. kl I 990 I I 700 201 | 2012 14.6 14.0 13.8 14,552 13,982 13,785 2013 2014 I4.0 I4.2 I"? 14.028 14.210 1"? ,4? 11 PCC is obtained from the reaction of CO2 with calcium hydroxide. It is used as a filler and/or coating in the paper, food, and plastic industries. 12 In 2015, 74 operating primary lime facilities in the United States reported to the EPA Greenhouse Gas Reporting Program. 4-12 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2016 ------- 1 Table 4-7: Potential, Recovered, and Net CO2 Emissions from Lime Production (kt) U-;ir Puk-nlhil RlinWIVll 1 Nil 1! miss in ns I 990 I 1.959 259 1 1 700 2005 15.074 522 14.552 201 I 14,389 407 13.982 20I2 14.258 473 13,785 2013 14,495 467 14.028 20I4 14.715 S()S 14,210 2015 13.764 422 13.342 " for s ugar rclining ant: [ l'CC production. Note: Totals may nol si mi due to indepen dent rounding. 2 In 2<>15. lime prndiielinn decreased cnnip;ired In 2(>I4 levels (decrease nl'jihmil (> percciili ;il IX.2~9 kilnlnns. nwinu 3 in decreased cniisnnipiinii hv ilie I S iiniil'errniis nicl;illiirme;il indiisiries (prini;irilv enpperi ;md si eel industries 4 (( nnithers 2o I I S( iS 2(> I(<;ii 5 Methodology 6 Ti< c;ilcul;itc ennssiniis. the miinuuis nf limh-c;ilcuuii ;md dnlnniitic lime produced were multiplied In ilieir 7 respective eniissinn liictnrs iisiiiu ilie Tier 2 ;ipprn;ich Irnni llie 2nnf- ll'< '<' < ini,!i. /iih > The emission liictnr is ilie 8 prnduct ill ilie stoichiometric r;itio between ('() ;md (;i(). ;md 1 lie ;iver;me ( ;i() ;md \lu() content for lime The (;i() 9 ;md \lu() content for lime is ;issiimed In he 95 pereenl for both limh-c;ilciuni ;md doloniitic lime (IIH ( 2()()<¦) I lie 10 emission factors were e;ileul;iled ;is follows' 11 I'nr limh-c;ilcuuii lime 12 |( 14.01 ti/niok' CO ¦) -h (.Ki.OH n/niok' CnO)| x (()/).">()() (I;i()/1i 111c) = 0.7-l-.">."> n CO ¦/}• lime 13 I'nrdnlnmilie lime' 14 |(tm.02 o/mnk-CO ) h- t>/mc»k* C:i())| x (().').">()() CnO/limt') = O.H(i7.~> ?¦ CO ¦/}• lime 15 Production w;is ;id|iistcd In remove I he iikiss of chcniic;illv enmhined w;iler found in lmlr;iled lime, delermined 16 ;ieenrdinu In I he innleeiil;ir weiulil nil ins nf II () In (( ;ii( )l 11 ;ind |( ;ii( )l h • \ 1 u< <) 11) 11 < ll'('(' 2()()(ii These liictors 17 sel lhe chcniic;illv enmhined w;iler ennleiil In 24 ' pereenl for hmh-c;ilciuni hvdmtcd lime. ;md 2" 2 pereenl Inr 18 dnlnmilie hulniled lime 19 The 2nnf- ll'< '<' i ini,L/iih> ( Tier 2 nielhnd) ;ilsn reeniiiniends ;iccoiiiilum I'nr ennssiniis from lime kiln dnsi i l.kDi 20 ihrnimh ;ipphe;ilinn of ;i enrreelinn liictor l.kl) is ;i hv product of the lime iii;iiiiif;icluriim prneess tvpic;illv nnl 21 reeve led h;iek In kilns I ,kl) is ;i v erv I'liie-unined ni;ileri;il ;md is cspccmllv useful Inr ;ipplie;ilinns rci|iiiriim v erv 22 siikiII p;iriiele si/e Mnsi enmmnii l.kl) ;ipphe;iliniis include soil reel;im;ilinii mid ;mriciilliirc. ( iirrenllv. d;il;i on 23 ;iiiiiii;iI I .kl) prndiielinn is nnl rendilv ;iv ;nl;ihle In dev elnp ;i enimirv specific enrreelinn l';ielnr I .line eniissinn 24 esiiiiKiles were multiplied hv ;i l';iclnrnf I ()2 In ;iccninil I'nr eniissimis from l.kl) (ll'( ( 2()(K>i See I he Hmnicd 25 I niprnv enienis seclinn ;issnci;iled w nil elTnrls in iniprnv e iiiiceri;niilv ;m;ilv sis ;md eniissinn es|ini;iles ;issnci;iled w nil 26 l.kl) 27 I .line eniissinn esiini;iles were liiriher ;id| listed In ;iccniiiil I'nr I he ;inininil nf ('() c;ipiiircd Inr use in nn-siie 28 prncesses \ll ilie dnniesiic lime l;icililies ;ire rei|inred In repnri iliese d;il;i In I P \ under iIs (il l( ikl' The lnl;il 29 n;ilinii;il-lcv el ;iiimi;il ;imniinl nf ('() c;ipiiircd Inr nn-siie prneess use w;is nhliiined I rnni IP Vs (il l(d{lJ (IP \ 30 2o l(>) h;ised nil repnried liieihlv lev el d;il;i Inr vein's 2o lo ihrnimh 2d 15 I lie ;ininiiiil nf ('() e;ipiiired reenv ered Inr 31 nn-siie prneess use is dedneled Irnni I he lnl;il pnlenli;il ennssiniis n e . Irnni lime prndiielinn ;md l.kl)) Ilie nel lime 32 ennssiniis ;ire presenled in T;ihlc 4-<> ;md T;ihlc 4-~ (il l( ikf d;il;i on ( () reninv;iIs (i e . CO e;ipiiired reenv ered) 33 w;is ;iv ;nl;ihle oulv I'nr 2o In llirniiuh 2o 15 Smee (il l( ikl'd;il;i ;ire nnl ;iv ;nl;ihle I'nr I990 ihrnimh 2<>(>l>. II'CC 34 "splieinu"' leehini|iies were used ;is per I he 2m if' ll'< '<'' :ni------- 1 I.line prndiiclimi d;il;i (In l\pe. hmh-c;ilcinni- ;md dnlnnnlic-t|iiickhnic. hmh-c;ilciiini- ;ind dnlnniilic-hulr;ilcd. ;ind 2 dc;id-hiiriicd dnlnnnici Inr Iihrnimh 2ul5 (sec T;ihlc 4-Si were nhijiincd Irnni llic I S (icnlnmc;il Siir\c> 3 (I S(iS)il S( iS 2(> | I ~i ;imiii;il rcpnris ;md ;irc cnmpilcd In I S( iS in llic iicnrcsi inn Vilnrcil 4 lis d mil lie I line, w Inch is produced I'rnni ( ;i() ;md hulmiihc c;ilci n in silic;iles. is iu(>(>) Since d;il;i Inr 8 llic iiidi\ idinl lime l\pes ihiuh cnlcinni ;md dnlnmilio were ik ;il c;ipii\c lime prndnclinii r;icihlics As nnled ;ihn\c. lime h;is ni;in\ different chcniic;il. 21 iiiclnsiri;il. cn\ irniimeiil;il. ;md cniisiriiclinn ;ipplic;ilinns. In iii;in\ processes. ('() rc;icls w illi llic lime In crc;ilc 22 c;ilcinni c;irhnii;ilc (c.u . w;ilcr snl'lciiiimi (';irhnn din\idc rcnhsnrpiinn r;11cs \;ny Imwe\er. dcpciidiim nil llic 23 ;ipphc;ilinn I'nr cviniplc. I nil percent nf llic lime used In prndncc prccipil;ilcd c;ilcinni c;irhnn;ilc re;icls Willi ('() . 24 w licrcns mnsi oil lie lime used in siccl ninkiim rc;icls w illi i mpiiril ics such ;is silic;i. siill'iir. ;md ;il n mi nil in 4-14 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2016 ------- 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 compounds Oiinulil\ inu I lie nniouul of ('() lh;il is renhsorhed would require ;i dclmlcd nccouuliim of lime use in ilie I lined Sink's ;iud nddilionnl iiiforninliou nhoul ilie nssocinlcd processes w here holli I he lime ;md h\ product ('() nrc "reused" nrc required lo quniiiil\ ilie ;iiiku2i lu nccordnuce Willi ll'CC niclhodolomcnl uuiclelilies. nu\ such emissions nrc cnlculnled In nccouuiiuu lor uel ( Hiincs from chnuucs mi hiouemc C reser\oirs mi wooded or crop Inuds (see I he l.nud I sc. I.nud-l se Clinuue. nud l-'oresir\ chnpien. In i he ense of wnler ireninieul plnuis. lime is used iu i he sofiemim process Some In rue wnler ireninieui plnuis nin> rcco\er ilieir wnsie cnlciuni cnrhounlc nud cnlcuie il iulo c|iii pcrccul flies nlso uoic llinl nddilionnl emissions (nppro\ininlel> 2 perceui i nin> nlso he ueuernled lliroimh prodiicliou of oilier In products wnsies (off-spec lime llinl is noi recvcled. scruhher sluduc) nl lime plnnis i Seeuer 2o| m I'uhlicK n\nilnhleou I.KI) ueuernliou mies. loinl qunuiiiics uoi used iiicemeui prodiicliou. nud i\pes of oilier h\ products wnsies produced nl lime fnciluics is linuied IT \ niiiinled n dinlouue willi \l. \ lo discuss dnln needs io ueuernle n coiuiirs -specific I.KI) I'nclor nud is rc\ lew nm ihe iiilorninlioii pro\ ided In \l. \ \l. \ compiled nnd slinred hisioricnl eniissious iiiforninliou nud qunuiilies for some wnsie products reported In nieniher fncililies nssocinled w illi ueuernliou of loinl cnlciued In products nud I .Kl). ns well ns nielhodolous nud cnlciilnliou worksheets thnl nieniher fncililies complete w lieu reportiim There is iiuceriniuis reunrdum the n\ nilnhilits of dnln ncross ilie lime series needed lo ueuernle n represeuinine couuirv-specific I.KI) I'nclor I iiceriniuis of ihe ncli\ n\ dnln is nlso n fuucliou of ilie relinhiln\ nud conipleleiiess of \ oluuinriK reported plnui-lc\cl production dnln I 'uriher resenrch nud dnln is needed lo impro\ e iiiidersinudiim of nddilionnl cnlciunliou emissions io consider rc\ ismu I lie curreiii nssumpiioiis Mini nrc hnsed on ll'CC guidelines More iiiforninliou cnu he found mi I he Plnuued I nipro\ enieuis section helow The results of the \ppronch 2 qunutiinli\ e uuceriniui\ minis sis nrc sunininri/.ed iu I'nhle 4-In I .line CO emissions for 2o 15 were esiininled lo he helweeu I2l> nud I ^ - \1\1l(() l!q nl I he l>5 perceui confidence lc\ el This confidence lc\cl nidicnlesn rnuue ofnppro\ininlel> ' perceui helow nud ' perceui nho\e ilie emission esiimnle of I ^ ^ \l\11 CO l!q Table 4-10: Approach 2 Quantitative Uncertainty Estimates for CO2 Emissions from Lime Production (MMT CO2 Eq. and Percent) 2015 l.missiim l!siim;iu- I iui-rl;iiiil> K;iii------- 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 I.IHUT lillllll(l I |)|KT lilllllHl I.IHUT lilllllHl I |)|KT liuLMUl I.ime Production C( h I 12.9 13.7 -V'/n + i% Range of emission estimates predicted by Monle Carlo Stochastic Simulation lor a 95 percent confidence interval. \1clhodolomc;il ;ippro;ichcs were ;ipphcd u» I lie enure lime series io ensure cousisteucv in emissions from I'WD lIiixmiuIi 2d 15 I )el:iils on ilie emission trends throimh lime ;ire described in more del;nl in ilie Melhodolouv seeliou. ;iho\ e I'or more iuforni;iiiou on ilie ueiler;iI o \ OC proeess applied lo ilns source c;ilcuor\. eousisieni with N'olunie I. ( kipicr<> of ilie li'< '<'' iititU-linc*. see o \ OC ;md \ erifie;iliou I'roeedures seeliou mi ilie iiiirodiieliou of ilie IH'l ( h;ipier I pd;iled d;il;i from I.is;i ('outliers (I S (ieolomc;il Sur\e\ i (( 'omihers 2u I ~i resulted in I lmli-( ;ileuini Oiiicklinic prodiielion d;il;i ekiuues for 2d 14 ;iud I )olonnlic Oiiicklinic production d;il;i ch;iimcs for 2d I i ;md 2d 14. ;is show u mi Tnhlc 4-X kcco\ ered emissions show u mi T;iblc 4-~ were upd;iled usuiu ;murcu;ilcd Cil l( ¦ l esiini;iie eniissious from prodiielion of l.kl) lu response lo Mils iechuic;il nieeinm. mi .l;iiiu;ir\ ;uid l'ehru;ir> 2d|(i. \|. \ compiled ;uid sh;ired lnsioric;il emissions MiforiiKiiiou reported In nieniher f:icililies on ;iii ;iiiiiu;il h;isis under \oluui;ir\ reporiiuu unti;iii\ es o\er 2uo2 throimh 2d I I ;issoci;iled w ilh ueuer;iliou of tol;il c;ilciued In products ;iud I .Kl) (I .Kl) reporinm oul\ dilfereiili;iled s|;iriiuu mi 2d Id i This emissions iiiform;iliou w;is reported oil ;i \ oluul;ir\ h;isis consisieiil willi \ l. Vs l;icilit> -lc\ el reportum protocol ;ilso receuil\ pro\ ided IP \ needs ;iddiliou;il lime lo rc\ lew the iiiforni;iiiou pro\ ided In \l. \ ;uid pl;nis to work w ilh llieni lo ;iddress needs for I P X's ;iu;il\ sis. ;is there is hniiled iiiforni;iliou ;icross the lime series I)uc lo hniiled resources ;uid need lor ;iddiliou;il o \ of uiforni;iiiou. this pkuiued iniproN enieui is still mi process ;iud h;is not heeu iiicorpomtcd into this curreui lu\eiitors report As ;ui imenm step. IPX h;is updated the c|ii;ilil;ili\ e description of uuccri;iiiil\ lo reflecl the iufornuitiou pro\ ided In \l. X In ;iddilion. I !l* X pkius lo ;ilso re\ lew (il l( ikl' emissions ;iud ;icli\ il\ d;il;i reported lo I PA under Suhp;iri S. mi p;irticiil;ir. rc\ lew of ;mureu;iled ;icli\ il\ d;il;i on lime prodiielion In t\pe IJ:irticuhir ;ilteuliou w ill he m;idc lo ;ilso eusiiriiiu iinie-series coiisisieucs of ihe emissions esiinuites presented mi fiiiure lu\ euior\ reports, eousisieni w ilh IK ( ;iud I \l'('('(' uiiidehues This is required ;is ilie l;icihi\-lexel rcporlumd;it;i from l!P X's (il ICiKP. with the prour;ini's iuiiiiil requirements for rcporiiuu of emissions in c;ilcud;ir\e;ir 2d Id. ;ire not ;i\;ul;ihle for;ill iu\ciiior\ \ c;irs 11 e . I iwu ihrouuh 2dd>Ji ;is required for this 11 in cuiors In i mple nieiiii uu impro\ cine ills ;md iuieur;iiiou of d;it;i from I P.X's (il l( ikl\ ihe kilest miid;iucc from ilie IK ( on ihe use of l;icihl \-lexel d;il;i mi ii;iiiou;il inxeniories will he relied upon 15 See. 4-16 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2016 ------- 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 (il;iss product ion is ;in cuerus ;iud mw -ni;iieri;il iuieusn e process lh;ii results in the ucueniliou of ('() from hoili i lie cuerus consumed in ni;ikiuu ul;iss;uid ilie d;iss process itself Emissions from fuels consumed forciierus purposes durum ilie prodiielion of ul;iss ;ire ;ieeouiiied lor in ilie I ineruv seelor (ihiss producliou eniploss ;i \;iriel> of r;iw m;iieri;ils iii;i ul;iss-h;iieli These include lornieiY Hu\es. si;ihih/ers. ;uid sonielinies color;iuis The nuijor r;n\ ni;iieri;ils 11 e.. rinses ;uid si;ihih/ersi w Inch emu process-rekiled c;irhou dioxide (CO i emissions durum ilie ul;iss nielliuu process ;ue liniesioiie. dolomite. ;uid sod;i ;ish The ni;iiu rornier mi ;ill i\ pes of d;iss is sihc;i (Si() i Oilier imjor rorniers in ul;iss include feldspar ;iud hone ;icid lie. hor;i\) I'Iiincs ;ire ;idded lo lower I he leniper;ilure ;il w Inch I he h;ilch mells Most coniniouK used llu\ iu;ileri;i Is ;ire sod;i ;ish (soduini c;irhon;ile. \;i CO i ;uid pol;ish (pol;issuini c;irhou;ile. K Oi Si;ihih/ers ;ire used lo ni;ike ul;iss more cheniic;ill\ s|;ihle ;uid lo keep I he finished ukiss Irom dissol\ niu ;uid or l;il h uu ;ip;irt. ('onimouK used s|;ibili/um ;meuls in ul;iss producliou ;ire liniesioiie (C;iCO i. doloniile (C;iCO \lu('() i. ;ilunnii;i I \l O i. ni;muesi;i i\1u()i. h;iriiini c;irhou;ile (I5;i( () i. siroiiliuni c;irhou;ile (SrCO i. Ill hiuiu c;irhon;ile 11 .i CO i. ;iud /ircoum i/.ri) i (Oil 2(i(>21 (il;iss milkers ;ilsi> use ;i cerl;nii ;uuouui of recvcled scr;ip ul;iss icullel i. w Inch conies from iii-house reluru of ul;issw;ire broken mi ilie process or oilier ul;iss spilkme or releuiiou such ;is recschuu or cullel broker ser\ ices The mw ni;iieri;ils (pnni;iril> liniesioiie. doloniile ;uid sod;i ;isln release ('() emissions in ;i coniple\ liiuh- leniperniiire cheniic;il re;icliou duriim I lie d;iss niellum process This process is noi direclK conip;ir;ihle lo l lie c;ilciu;iliou process used mi lime ni;iiiuf;icluriim. cenieul ni;iiiiif;icliirum. ;iud process uses of c;irhou;iles (i.e.. liniesioiie doloniile usei. hul h;is ilie s;une uel effect mi lernis of ( () emissions i ll'( ( ' 2uu<>) I lie I S uhiss industry c;iii he di\ ided nilo four ni;iiu c;ileuorics coiil;iiiicrs. fl;ii i\\ mdow i ul;iss. fiher ul;iss. ;iud speci;ill\ uhiss. The ni;i|orii\ of coniniercuil ul;iss produced is coiii;uuer ;iud fl;il ukiss i IT \ 2(>ui>). The I in led Si;iies is one of I lie nuijor uloh;il exporters of uhiss l)oniesiic;ill>. denuuid conies ni;uul> from I lie coiisirucliou. ;iuio. hoillum. ;uid couijiiiier iiidiisiries There ;ire o\er l.5oo companies ih;ii ni;uiuf;iclure ukiss in I lie I mied Si;iies. with I lie kiruest heiuu Corinim. (iii;irdi;iu Iiidiisiries. Owens-Illinois. ;uid lJKi Iiidiisiries lu 2d 15. kilolous of liniesioiie ;uid 2. kilolous of sod:i ;ish were consumed lor ul;iss producliou (I S( iS 2d 15c. Willell 2d I ~) I )olomilc cousiinipiioii d;il;i for ukiss ni;iiiiif;icluriim w;is reporled lo he /ero for 2o 15 I se of liniesioiie ;uid sod;i ;ish mi ul;iss producliou resulted in ;murcu;ilc ('() emissions oi l ' \I\1T CO I !i|. (1,2l>l> kl) (see I;ihle 4-1 I) ()\er;ill. emissions h;i\e decreased 15 perceui from I wo ihrouuh 2ol5 Imissions in 2o 15 decreased ;ippro\ini;iiel\ ' perceui from 2o 14 le\els w Inle. mi uenenil. emissions from ul;iss producliou h;i\e reni;uiied rcl;ili\el> coustjiui o\er I he lime series with some I1iicIii;iiioiis since ll^<> lu ueuenil. lliese fliiclii;iiious were rekiled lo l lie heh;i\ lor of I lie e\pori ni;irkel ;uid I lie I S ecouoim. Speci fic;i 11\. ilie extended dowiiiuni mi resideuii;il ;uid coninierci;il coiisirucliou ;uid ;iuiomoii\c iiidiisiries helweeu 2()(>X ;ind 2<>l<) resulied in reduced coiisumpiiou of ukiss producls. c;iusiim ;i drop in uloh;il demmid for limestone doloniile ;iud sod;i ;ish. ;uid ;i correspoudiim decrease mi emissions I 'uriherniore. ilie ukiss coui;iiuer sector is one of the le;idiuu sod;i ;ish coiisunnim sectors in the I mied Suites Some coninierci;il food ;uid hc\ crime p;ick;me ni;iiiiif;iclurers ;ire sin ft i uii from ul;iss coiit;uuers towards lmhler;uid more cost effecti\e pol\elh\ lene lerephlh;il;ite (HT) h;ised coiit;uuers. pulliim dow uw;ird pressure oil domestic consumption of sod;i ;ish (I S(iS Iiw5 throimh 2o15c). Table 4-11: CO2 Emissions from Glass Production (MMT CO2 Eq. and kt) U-:ir MM l ( (): l'.(|. kl_ 1990 I I ^^5 16 Excerpt from Glass & Glass Product Manufacturing Industry Profile, First Research. Available online at: . Industrial Processes and Product Use 4-17 ------- 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 2005 1.9 1,928 2011 1.3 1.299 2012 1.2 1.248 2013 1.3 1,317 2014 1.3 1.336 2015 1.3 1,299 Nolo: Totals may not sum due to independent rounding C;irhou dio\ide emissions were e;ileul;iled h;ised on i lie yitit- li'( V 1 (iuiiLiiiuw Tier ' method In niullipK niu I lie c | m ; 111111\ of inpiil e;irhon;iles (liniesioiie. dolomite. ;ind sod;i ;isln In I lie e;irhou;ile-h;ised emission f;ielor (in nieli'ie UHis ('() nieli'ie Ion e;irhon;ile) liniesioiie. n4'l)"l. dolomite. "4 ^2. ;ind sod;i ;ish. D.414^2 ( ousu mpi ion d;il;i lor I 'J'Jii lliroimh 2t> I 5 of limesioiie. dolomile. ;md sod;i ;isli used lor ul;iss ni;iiiiif;ieluriim were ohl;nued from llie I S (ieolome;il Sur\e> il S(iSi \/iihr,i/.\ )l-,ir/'uull < rn>lhi! Sinih-. iimn.il lu /'ori (|iw5 lliroimli 2ol5hi. 2 <) 15 prel i mi ikiia d;il;i Irom i lie I S( iS (rushed Sione ( oniniodils l!\peri (Willell 2< > I-1. i lie ' V/.s \liihr,il.s )ciirhi»iL Su./n . i.sh . imin.il AV/"*/-/ i |'W5 lliroimh 2t) 151 < I S(iS l'W5 lliroimh 2t>l5e). I S( iS \liner;il liidiisir\ Siiia e\ s lor Sod;i \sli in .l;iiiu;ir\ 2t)l5(l S( iS 2d 15;n ;ind llie I S I5ure;iu of Mines i liw| ;ind I'N'in. w lneli ;ire reporied lo llie ne;iresi ion Durum ivwo ;iud ll^2.ihel S( iS did noi eondiiel ;i detailed sur\ e\ of limesioiie ;md dolomile eoiisiimpiioii In end-use ('oiisiinipiiou lor Iiwt) w;is esiini;iied h\ ;ippl\ iiiu ilie II pereeuiiiues of iol;il liniesioiie ;iud dolomile use eousiiiuied In llie indix idu;il limesioiie ;iud dolomile uses lo I'wo loi;il use Siniil;irl\. llie I ^>2 eoiisiinipiiou fmures were ;ippro\ini;iied h\ ;ippl> iiiu ;iu ;i\ er;me of llie I w I ;iud I ' percentiles of ioi;il limesioiie ;iud dolomile use eousiiiuied h\ llie iudi\ idu;il limesioiie ;iud dolomile uses lo I lie ly): loini \ddilioii;ill\. e;ich \e;ir llie I S( iS w iililiolds d;it;i oil eei1;nu limesioiie ;iud dok proprietors d;il;i I'orilie purposes of iliis ;iu;il\sis. eniissn e end-uses iluii eoui;iiued w illilield d;il;i were esinu;iied usiuu one of i lie follow iiiu ieeliuic|iies 111 llie \ ;ilue for ;ill I lie w illilield d;il;i pouils lor liniesioiie or dolomile use w;is disirihuied c\ eiiI\ lo ;ill w illilield end-uses, or 121 llie ;i\ crime pereeui of ioi;il limesioiie or diilonnie I'orilie w illilield end-use in i lie preceding ;iud sueeeediuu\e;irs. I here is ;i l;irue t|ii;iuiii\ of liniesioiie ;iud dolomile reporied lo llie I S( iS under I lie e;ileuories "unspecified reporied" ;iud "unspecified esiini;iied " \ poriiou of iliis eoiisiinipiiou is lvlie\ ed lo he liniesioiie or dolomile used for uhiss ni;iuuf;ieluriim I lie i|ii;iulilies lisied under llie "iinspeeified" e;ilei:ories were, llierelore. ;illoe;iled lo uhiss iii;iiiuf;icluriim ;iccordiuu lo llie pereeui limesioiie or dolomite eoiisiinipiiou for uhiss iii;iiiuf;icluriim end use for lh;il >e;ir1 I5;ised oil llie 2t) I 5 reporied d;il;i. llie es|ini;iled disirihuliou of sod;i ;ish eoiisiinipiiou lor uhiss prodiieliou compared lo lol;il donieslie sod;i ;ish eoiisiinipiiou is 4X pereeui (I S( iS I lliroimh 2d 15ei 17 This approach was recommended by USGS. 4-18 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2016 ------- 1 Table 4-12: Limestone, Dolomite, and Soda Ash Consumption Used in Glass Production (kt) .Uli\ il\ mil 2005 2011 2012 2013 2014 2015 Limestone 430 920 614 > > > 693 765 699 Dolomite 59 541 0 0 0 0 0 Soda Ash 3.177 3.050 2.480 2.420 2.440 2.410 2.390 Tuliil 4.511 3.0'M 2,975 3.133 3,175 3,0N'J 2 unceridiruy anu i imeoeries Lonsisiency 3 The uiiccri;iiul\ lc\cls presented in lliis section ;irisc in p;iri due lo \ ;iri;ilious in llie chcniic;il composition oT 4 limestone used in uhiss production In ;iddiliou lo c;ilciuni c;irbou;ilc. limestone ni;i\ coiilmu siii;illcr miiouuls oT 5 ni;mucsi;i. sihc;i. ;ind sulfur. jinioim oilier niiiicr;ils i pol;issiuni c;irbou;ilc. siroiiliuni c;irbou;ilc ;ind bmiiini c;irbou;ilc. 6 ;ind dc;id burned dolomite) SiniikirK. I lie i|ii;ilil\ ol" llie limestone (;ind mix oTc;irbou;ilcsi used lor ul;iss 7 11i;iiiiil;icl11riiiu w ill depend on ilie i\ pe ol' ul;iss benm nimiiiTiiclurcd 8 The esinmies below ;ilso ;iccoiiul Tor uuccri;iiiil\ ;issoci;ilcd w illi ;icli\ its d;il;i I .;iruc lluclimlioiis in reporied 9 consumption exist. rcUccliim >e;ir-io-\e;ireli;iimes in ilie number oTsur\e\ respouders The iiuccri;iiui\ resiiliinu 10 I'roni ;i shilling sur\e\ population is e\;ieerh;iled In llie mips in ilie lime series of reports The ;iccur;ic> of 11 distribution In end use is ;ilso iiuccrimu lve;nise llns \;ilne is reporied In ilie iii;iinir;ieliirer of llie inpiii e;irhon;iies 12 i limesioiie. dolomiie ;md sod;i ;isln mid noi llie end user Tor 2d 15. ihere li;is heen no reporied eoiisuinpiioii of 13 dolomiie lor ul;iss iii;iiiuT;icluriim These d;il;i h;i\ e been reporied in I S(iS b\ dolomiie iii;iinir;ieliirers ;md nol end- 14 users ii.e . ul;iss inmiiiliieliirersi There is ;i hiuli uiiccri;uui\ ;issoei;iled w illi lliis esiiin;iie. ;is dolomiie is ;i nuijor r;iw 15 m;ileri;i I eonsiimed in uhiss prodiielion \ddiliou;ill\. I lie re is siumfiemil inhereiil uuccri;iiiil\ ;issoei;iled w illi 16 esiiiiKiiinu w iihheld d;il;i ponils Tor specific end uses oT IniiesUiiie ;md dolomite The iiiiccri;uul\ oT llie es|iin;iies Tor 17 limestone mid dolomiie used mi ul;iss ni;ikum is cspcci;ill> lnuli l.;istl\. niueli oT llie limestone consumed 11i llie 18 I lined Suites is reporied ;is "oilier uuspeeil'ied uses;" therelore. it is diTTieuli to ;iccur;iicl> ;illoe;ite this unspecified 19 i|ii;iuiii\ lo llie correct cud-uses T'uriher research is needed mio ;ilteru;iie ;md more complete sources ol'd;il;i ou 20 c;irboii;ilc-b;iscd r;iw ni;iteri;il eoiisuinpiioii b\ llie ul;iss mdiistn "This \e;ir. I !l* \ reiuiti;iled dmlouuc w illi the I S(iS 21 Vitioiinl \ 11ner;iIs luTorm;iiiou Center Crushed Stone coniniodits e\peri lo ;issess the current uuccri;iiiii\ rmmes 22 ;issoci;iled willi c| ii;i lit i lies oT c;irbou;ilcs coiisunied Tor ul;iss production compiled ;uid puhhshed ml S( iS reports 23 The results ol" the \ppro;ich 2 i|ii;inlil;ili\ e iuiccri;iiiil\ ;ui;il> sis ;ire sunini;iri/ed in T;iblc 4-1 ' lu 2d 15. ul;iss 24 prodiielion C() emissions were estinuiied lo he helweeu I 2 ;iud I 4 \ 1 \ 1"T ("<) lx| ;il the '->5 percent confidence 25 lc\ el This iudic;iles ;i r;umc ol";ippro\ini;iiel> 4 perceui helow mid 5 pcrccul ;iho\e llie emission esiini;iie ol" I ' 26 \l\TTCO I iq 27 Table 4-13: Approach 2 Quantitative Uncertainty Estimates for CO2 Emissions from Glass 28 Production (MMT CO2 Eq. and Percent) Si ill I'l l" Ci;is 2015 llmissiiin Ksiim.iU- I iui'il;iinl\ Ki-I.iMm- In 1". miss in 11 r!sliiii;ik"' (MMT CO: Kii.) (MM 1 ( (): Kd.) ("..) I.I HUT I |)|KT I.I HUT I |>|KT liiiund 1 {iiiiihI 1 {iiiiihI ISiimihI Class Production I ( ) 1.3 1.2 1.4 -4% +5% Range ol"emission estimates predicted by Monte Carlo Stochastic Simulation for a 95 percent confidence interval. 29 \1clhodolomc;il ;ippro;ichcs were ;ipphed lo llie eulire lime series to ensure coiisisiencv 111 emissions I'roni llwo 30 ihroimh 2d 15 I )el;nls 011 llie emission I rends ihroimh lime ;ire described iu more del;iil 1111 he Melhodolouv secliou. 31 ;ibo\c 32 I'or more luTornuiiiou 011 the ue 1 ler; 11 o \ OC process applied lo llns source c;ileuor\. coiisisieut with Volume I. 33 Ch;ipier (> ol" llie Jwir. ll'< '< 11 ;///,/, 7/'//r.v see o \ n(' mid VcriTic;iliou I'rocedures secliou 111 llie iiiiroducliou ol" ihe 34 ll'l'l Ckipier Industrial Processes and Product Use 4-19 ------- 1 2 I .Milestone ;iud dolomite consumption d:il:i for 2d 14 were rc\ iscd rcl;ili\ c lo I lie prc\ ions 11 in cnu>r\ bused on i lie 3 pivlimiikiia d;il;i obtained dircclK from ilie I S(iS ( rushed Stone ( oniniodils expert. Tisou \\ iNell (\\ illell 4 In llie pie\ kmis In\eiiun"\ n e . Il>l><> throimh 141. prelimiikiia d;il;i were used I'oi'2d|4. which were upd;iled for 5 I lie ciirreiil lii\cutor\ I lie published lime series w;is re\ lewed lo ensure lime-series coiisisiciic> I his upd;ile e;iused 6 ;i dee reuse in 2d 14 emissions of less ih;iu I pereeni compared lo 2d 14 emissions preseuied mi ilie prc\ kmis 11 in ciiioia 7 ne. I wd throimh 2Dl4i 8 riannea improvemenis 9 \s noted mi I lie prc\ kmis re pons, currcul puhl iel\ ;i\;ul;iblc ;icli\ n\ d;il;i shows eoiisunipiioii ol'oiik hniesione ;iud 10 sod;i ;ish lor ul;iss ni;iiiiif;icliirum While hniesione ;iud sod:i ;ish ;ire I lie predominant e;irhou;iles used in ul;iss 11 iii;iiiiif;icliirnm. I lie re ;ire oilier e;irhou;iles lli;il ;ire ;ilso eousunied lor ul;iss ni;iiiuf;icluriim. ;ill IkmiuIi hi siii;i I ler 12 (|ii;iulilies. I :P \ h;is niilKiled re\ lew of ;i\;ul;ihlc ;icli\ its d;il;i oil c;irhou;ile eoiisunipiioii In |\pe mi I he d;iss iudusir\ 13 from IPX's (iieeiihouse (us keportnm I'rournni i( il l( ikl'i reported ;imiii;ill\ siuee 2d Id. ;is well ;is I S(iS 14 piihhe;ilious 15 I P \ h;is Minuted rc\ iew of I Ins ;icli\ it\ d;it;i ;iud ;uiticip;ilcs lo fui;ili/e ;issessmeui lor fiiiure uitcurnliou of d;il;i 16 reported under I P. Vs (il l( ¦ |< I* mi the sprum of 2d I" in inipro\ e I he eonipleleuess of emission esiini;iles ;iud 17 f;ieilil;ile c;ilcuor\ -speeifie n( per Volume I of I lie 2nnf- ll'< '<' < iniiL/iih* for I lie (ikiss Production source c;itcuor\. 18 I P Vs (il l( ikl' h;is; 11 i emission ihreshold for rcporiiui:. so i lie ;isscssniciii w ill consider I lie eonipleleuess of 19 c;irhou;ilc eoiisunipiioii d;il;i for uhiss production mi 1 he I nil eel Slnlcs knrticiikir ;illeuiiou will ;ilso he nindc lo ;ilso 20 eiisuriim lime-series coiisisieucs of I lie emissions csiini;iics preseuied mi fiiiure ln\ euior\ reporis. cousisieui w nil 21 I lJ("(' ;iirI I Nl'('('(' uiudehiies I'liis is required ;is i lie l;icihi\ -lex el report um d;il;i from IP Vs (il l( ¦ k I*. w illi I lie 22 prouriun's uiiiiiil re(|iiirenieuis lor reporiiuu of emissions in c;ilcud;ir xe;ir 2d Id. ;ire uoi ;i\;ul;ihlc fornll iii\euior\ 23 \e;irs 11 e . Iwo ihrouuh 2dd^i ;is required for ilns 11inciiioia In implement inu inipro\enieuis ;uid uitcurniiou of 24 d;il;i from IP Vs (il l( ikl'. I he l;iles| miidnucc from the ll'('(' on the use of f;icilil> -lex el d;il;i in u;iliou;il i nx eulories 25 will he relied upon |s These pliiuued iniproxenienis;ireouuoiim;uid IP \ m;ix ;ilso 111111;11e research mio oilier 26 sources of ;iclix nx d;il;i for c;irhou;ile eoiisunipiioii h\ I he ul;iss industry 27 28 29 30 I .Milestone <( ;i( () i. dolonule (( ;i('() \lu('() i.1'1 ;ind oilier c;nliou;iles such ;is sod;i ;ish. ninmicsitc. ;uid siderile ;ire 31 h;isic ni;ileri;ils used In ;i wide x;inelx of industries, iiicludiim construction. nuriculturc. chcniic;il. mel;illurux. ulnss 32 production. ;iud eiin iroiimeui;il pollution control This section addresses oiils hniesione ;uid dolomite use for 33 iikIiisIri;iI ;ipphc;ilious. c;irbou;itcs such ;is limesioue ;uid dolomite ;ire honied siifficieutIx eiiouuh lo cnlciuc the 34 ni;iteri;il ;iud uciicmtc ( () ;is ;i In product 35 CuCO, — CuO + CO. 36 MyCO-. - Mi]() + CO. 18 See . 19 Limestone and dolomite are collectively referred to as limestone by the industry, and intermediate varieties are seldom distinguished. 4-20 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2016 ------- 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 I A;imples of such ;ipphc;ilious include limestone used ;is ;i l ln\ or purifier in nicl;illurmc;il lurii;iccs. ;is ;i sorhcni in Hue u;is dcsiill'iiri/;iliou (l '(il )i s\ stems lor ulilils ;iud indnsiri;il pkiuls. ;ind ;is ;i r;i\\ ni;ilcri;il lor I lie production of ul;iss. 11inc. ;md ccnicul I Emissions from limestone ;ind dolomite used in oilier process seclors such ;is cement. lime. ul;iss prodiielion. ;md iron ;md slccl. ;ire c\cludcd from llns section ;ind reported under ilieir respccli\c source c;ilcuorics ic u . Section 4 V (ikiss I'roduclioui I Emission from sod;i ;isli consumption is rcporicd under respecli\ c c;ilcuorics (c u . (ihiss \1;iiiiil';icliirum (( kl' Source ( ;ileuor\ 2 \') ;iud Sod;i \sli I'roduclioii ;ind ('ousiinipiioii (( kl' Source C;ilcuoi'\ 215-)) I Emissions from fuels coiisunied forcucrus purposes durum iliese processes ;ire ;iccoiiuied lor in llie I !nerii\ cluipicr. I.iniesioiie is w idel> disirihuied iliroimlioui ilie world in deposiis til'\;ir\ nm si/cs ;iud decrees of purii\ I.;irue deposiis ol' limesione occur m ue;irl\ c\ er\ sinle iu llie I lined Sinles. ;iud siumric;uil t|iinuiiiics ;irc cMrnclcd lor iudiisiri;il ;ipplic;ilious lu 2d 14. ilie Icndiuu liniesioiie prodiicum sinles ;ire I e\;is. Missouri. I'londn. ()lno. ;uid kciilucks. w Inch couirihulc 4^ percent of the lotnl I S ouipiilll S(iS l'W5n lliroimh 2d| 5) Siniilnrk. dolomite deposiis ;ire nlso w idesprend ihroimhoui ihe world I)oloniilc deposiis ;irc found iu the I uiied Sinles. (nundn. \le\ico. I mi lope. Allien. ;uid Urn/il lu llic I lined Sinles. llic lendum dolonnle prodiiciuu s|;iies ;irc Illinois. IVuuss l\ ;ini;i. ;ind New York, w Inch contribute 55 perccul of the lotnl 2d 14 I S ouipiilll S(iS l'W5n lliroimh 2dI5i lu 2dI5. 2^.251 kl ol'liniesioiie ;iud \244 kl ol'doloniiie were coiisunied lor iliese ennssi\e ;ipphc;ilious. c\cludiim ul;iss niniiiifncluriim i Wi licit 2d I "hi I snuc ol' liniesioiie ;iud dolonnle resulied iu nuurcunlc ('() emissions ol' I 1.2 \1\11 ( () I !c| (I 1.2 '(i kl i (see Tnhle 4-14 ;iud Tnhle 4-15) While 2d 15 emissions h;i\c decreased 5 perceui compared lo 2d|4. o\er;ill eniissious h;i\e incre;ised I2'J perceni Ironi I1>1>d ihrouuh 2d|5 Table 4-14: CO2 Emissions from Other Process Uses of Carbonates (MMT CO2 Eq.) W;ir l"lu\ Slum- k;d M;ir acid water treatment. acid neutralization. and sugar Note: Totals may not sum due lo independent round Table 4-15: CO2 Emissions from Other Process Uses of Carbonates (kt) ^ e;ir l"lu\ Slime ix;d Mii^iU'siinii I'l'iiduiiiiiii Oilier Misielhiiieiiiis I SI S 1 Tiil.il 1990 2.592 1,432 64 819 4.907 2005 2.649 2,973 0 718 6,339 201 1 1.467 5.420 0 2.449 9.335 2012 1.077 5.797 0 1.148 8.022 2013 2.307 6.309 0 1.798 10.414 2014 2.91 1 7.1 1 1 0 1.790 1 1.81 1 2015 3,031 7.335 0 871 11.236 ;i "()ther miscellaneous uses" include chemical stone, mine dusting or acid water treatment, acid neutralization, and sugar refining. Note: Totals may not sum due to independent rounding. Industrial Processes and Product Use 4-21 ------- 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 Cnrhou dii<\idc emissions were cnlculnlcd hnsed on the ym/- H'( '<' (iuiiLiiiuw Tier 2 method In niullipK inu I lie 1111;1111il\ of limestone iir dolomite consumed In the emission I'nctor lor limestone or dolomite cnlciiinliou. rcspccti\cl>. Inblc 2 I limestone o4^~l metric inn ('() metric iiincnrbounlc. ;ind dolomite.<) 4 ^2 metric ton ('() metric Iiin cnrhounlc This niclhodolous wns used liir flu\ sioue. Hue uns dcsiilliiri/nliou s\ skins. j;iiiI use ol' limestone ;ind dolonnic llinl produced CO emissions Ai ilie end of 2<>t)|. ilie sole ninmicsiiini production plnui opernliuu mi ilie I lined Sinles llinl produced iiinuuesiuni nielnl usinun dolonuiic process Mini resulted iu llic relense of CO eniissious censed its opernlious 11 S(iS llJ'J5h Miroimh 2<>I2. I S(iS2<)Pi ('oiisiinipliou dnln for Iiwt) lliroimh 2t> 15 of limestone nud doloniiie used for flu\ sioue. flue uns desiilfiiri/nliou s\ sienis. c I ic 111 i en I sioue. mine diisiiuu orncid wnler irenimeui. nciil iieuirnli/nliou. nud suunr reluiiiiu (see Tnhle 4-l(>i were ohinuied from die I S. (icolomcnl Sur\c> (I S(iSi \liihru/.\ < rn>lhJ Siniw.iiiiiihi/ lu/'uri i ll>l>5n Miroimh 2ul5i. preliniiunr\ dnln l'or2n|5 from I S(iS ( rushed Sionc Coniniodils l\peri iWillell 2u|~hi. \mericnii Iron nud Sieel Iusimile Iimesioiie nud doloniiie consumption dnln i MSI 2t> l<>). nud llic I S. IJurenii of Mines i llw| nud ll->l^m. which nre reporied lo llic uenresi ion The production cnpncits dnln for I'Wt) Miroimh 2<>I5 of dolonuiic ninuiiesiuni nielnl nlso enme I'roni the I S(iS i l'W5h Miroimh 2ul2. I S(iS 2(>| i) nud the I S IJurenu of Mines 11lwo throimh IWhi Durum I'wnniul ll>l>2. the I SCSdid not conduct n detniled sur\ e\ of liniestoue nud doloniiie consumption In end-use ('oiisiinipliou for I'J'Jt) wns estininled In nppKiuuihc II pcrcciilnucs of lotnl Iimesioiie nud dolomite use constituted h\ the i ml i\ idunl Iimesioiie nud dolomite uses to I lwo totnl use Sum In rl\. the I ^>2 eoiisiinipiioii I'iuures were nppro.Mninled In nppl> nm nil n\ ernue of the I I nud I W perceutnues of totnl limestone nud dolomite use constituted In the iudi\ idunl Iimesioiie nud dolomite uses to the Il)l>2 totnl \ddiliounll>. encli \enr the I S( iS w iihhokls dnln on certniu Iimesioiie nud dolomite end-uses due lo coufidciiiinhis nureenients reunrdnm conipnm proprielnr\ dntn I 'or llic purposes of this niinls sis. eniissi\ c end-uses Mint coiiiniued w itliliekl dnln were esiimnled usiuu one of the follow nm iechuii|iies. (I) the \ nine for nil the w itliliekl dnln points for limestone or doloniiie use wns distributed c\cnl\ lo nil withheld end-uses. (2) the n\crnuc percent oftoinl liniestoue or doloniiie for the w ilhheld end-use iu llic prcccdum nud succccdiim \ en in. or i ' i llic n\ ernue Irnclioii of lotnl limestone or doloniiie for llic cud-use o\ cr llic entire time period There is n In rue i|iinuiii\ of crushed sioue reporied lo llic I S( iS under llic cnlcuorv "unspecified uses " \ poriiou of lliis eoiisiinipiioii is helie\ ed to he limestone or doloniiie used for ciiiissin c cud uses The i|iiniiiu\ listed for "unspecified uses" wns. therefore, nllocnled lo encli reporied end-use nccordnm lo encli cud-use's I'rnction of lotnl eoiisiinipiioii mi Mint \enr Table 4-16: Limestone and Dolomite Consumption (kt) Aili\ il\ I'WO 2005 2011 2012 2013 2014 2015 l-'lux Stone C) 7^7 7.022 4.396 1 (S()(S 6.345 7 *^90 7.834 Limestone 5.804 3.165 2.531 3.108 4.380 4.243 4 ^90 Dolomite 3.857 1.865 S50 1.965 3.356 3 244 l'Cil) 3.258 6.761 12.326 13.185 14.347 16.171 16.680 Oilier Miscellaneon; ; Uses 1.835 1,632 5.548 2,610 i 973 4 069 1.980 lnhil 11.S30 15,415 22,270 24,------- 1 uncertainty ana i imeoeries Lonsisiency 2 The iiuccn;iiul\ lc\ els presented in this section iiccouul for iiiiccri;iuily ;issoci;ilcd w ilh ;icli\ ily d;il;i l);il;ion 3 linicsiouc ;uid dolomite consumption ;ire collected b\ I S(iS lliroimh \ oluul;ir\ u;iliou;il siiiac> s I S( iS contacts the 4 mimics (i c . producers ii|" \ ;irious |\ pes of crushed stone) for ;iiiiim;iI s;ilcs d;il;i. I );ii;i on other c;irbou;ilc consumption 5 ;ire iml rc;idil> ;i\;iil;iblc The producers report I lie ;iiiiim;iI (|ii;nilil\ sold lo \;irious end-users ;md industrs l\pes. 6 I S( iS estimi;iics ilie historical response r;iie lor ilie erushed stone snr\ e\ lo he ;ippro\im;iiel> pereeiil. ;ind llie resi 7 is esiimined In I S(iS I.;iruc I'luelikiikiiix mi reporied consumption e\isi. rcflccliuu \e;ir-ui-\e;ir ckinucs in ihe 8 number of sur\ c> responders I'lie iiiieei"l;iiiil\ resiilnim from ;i shiftiim siiia e\ population is e\;icerb;iled In llie iz;ips 9 mi ihe lime series of reporis. The ;iccur;ic> of distribution In end use is ;ilso iiiiccrl;iiii hee;iuse lliis \ ;ilue is reporied 10 h\ ihe producer mines ;md noi ihe end user \ddilion;ill\. I lie re is sium fie;i nl inhereiil inieeri;ii ni\ ;issoci;ilcd w ilh 11 esiiiikiiiii'-i w uhheld d;il;i pomis lor speeil ie end uses of hmesioiie ;iud dolomiie I .;istl\. much of ihe hmesioiie 12 eoiisunied iu ihe I lined Scues is reporied ;is "oilier uuspeeified uses." iherefore. il is diffieuli lo ;iccur;itcl> ;illoe;ile 13 llns uuspeeified i|ii;iulil> lo ihe eorreel end-uses This \e;ir. I P \ reuiili;iled dmlouuc w ilh ihe I S( iS \;iiiou;il 14 Minerals luforni;iliou Ceuler ( rushed Sioue eoniniodils e\peri lo ;issess ihe curreiil iiuccri;iiul\ r;umcs ;issoei;iled 15 w ilh ihe hmesioiie ;iud dolomiie consumption d;il;i compiled ;iud published b\ I S( iS I )uriuu llns discussion, ihe 16 e\peri confirnied lluii lil' Vs r;nmc of iiiiccri;uui\ w;is still rc;isou;iblc (W illed 2t)|~;n 17 I iiccri;iint\ mi ihe cs|im;ilcs ;ilso ;irises iu p;ul due lo \ ;iri;ilious mi ihe chemic;il coniposiiKiu of hmesioiie lu 18 ;iddiliou lo c;ilcium c;irhou;ile. Iimesioue ni;i> coul;uu sm;iller ;imouiils of ni;imicsi;i. sihc;i. ;md sulfur. ;inioim oilier 19 niineriils The evict specil ic;ilious for hniesioiie or dolomiie used ;is llu\ sioue \ ;ir\ w ilh ihe p\ ronicl;illuruic;il 20 process ;uid ihe kind of lire processed 21 The results of ihe \ppro;ich 2 i|ii;nilil;ili\ e iiiiccri;iiul> ;111;11\ sis ;ue sunini;ui/ed in T;iblc 4-1 ~ ( ';irboii dio\ide 22 emissions from oilier process uses of c;irhoii;iics iu 2<) 15 were esiini;iied lo he hem ecu ;uid I ' 2 \1\1T ('() I !i|. 23 ;ii I he l>5 pereeiil confidence lex el This indicates ;i nuiue of ;ippro\ini;iiel> I ' perceui below ;ind l(> perceni ;ibo\e 24 ihe emission cs|ini;ilc of I 1.2 \1\1T ( '() I !i| 25 Table 4-17: Approach 2 Quantitative Uncertainty Estimates for CO2 Emissions from Other 26 Process Uses of Carbonates (MMT CO2 Eq. and Percent) 2015 I'.mission Si ill I'l l' (iilS I'.siiniiiU' I nii'i'hiiim Riinm'Kikilivi'In I'missiun I'slimiili'' I .MM'I'CO: Kii.) (MM 1 CO: i«i» 1.1 HUT l |>|KT I.I HUT I |>|KT lii >11 ml lililllld Bound Bound ()ther Process 1 Jses 1 1 2 1 i 2 -1 i% +16% of Carbonates Range of emission estimates predicted by Monte Curio Stochastic Simulation for a 95 percent confidence interval. 27 \1clhodolomc;il ;ippro;ichcs were applied lo 1 lie eulire nine series ui ensure coiisisiencs 111 eniissioiis from I'Wt) 28 ihroimh 2d 15 I )el;nls 0111 lie emission ireuds ihroimh lime ;ire described 111 more dcl;iil 111 ihe Methodology seclion. 29 ;ibo\e 30 for more iuforni;iiiou 011 ihe ucucnil o \ OC process applied lo llns source c;ilcuor\. cousisieui Willi Volume I. 31 ( h;ipler (> of 1 lie 2mif- li'< '<' < inith liih .s. see OA 0(' ;ind Verification I'rocediires seel 10111111 lie introduction of the 32 ll'l'l ( kiptcr 33 Recalculations Discussion 34 I .Milestone ;iud dokiniite coiisunipiiou d;il;i by cud-use for 2<> 14 were updated rcl;iti\ e lo the pre\ ions 11 in cuiory 35 b;ised 011 the prchnnii;iry d;il;i pro\ ided by I S(iS ( rush Stone ( oniniodily expert. l;isou \\ illell lu ihe pre\ ions 36 Iuxeiiuirx lie.. I'J'Ji) lliroimh 2<) 14). prchniiii;iry d;il;i were used for 2<> 14 w Inch were upd;ilcd for ihe currcui 37 lii\ciitory The published lime series w;is rc\ icwed lo ensure linie-senes consistency This updiilc c;iused ;i decrease 38 iuioi;il limestone ;md dolomite consumption for cnnssi\c cud uses 111 2ul4 by ;ipproxini;itcly 2 perceui. rcl;ili\clo 39 the prc\ ious report Industrial Processes and Product Use 4-23 ------- 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 I'eiidiiiu ;i\ ailahle lesnuiees. llns seelimi \x ill iiilemale and present emissmiis Irnm snda ;ish eniisumpiinii Inr niher chemical uses ( nmi-ulass prndiiclimi) ( uiTeiilh . in llns dneiimeiil. I hose esii males are presented almm w nil emissions Irmii snda ;ish prndiiclimi (ll'CC Caleunrs 2I>~) This imprn\email is planned ;md will he implemented iiilii ihe ne\i In\enU'i"\ repnride. Ii u» 2<>I( 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 4-24 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2016 ------- 1 Only the CO2 emitted directly to the atmosphere from the synthetic ammonia production process is accounted for in 2 determining emissions from ammonia production. The CO2 that is captured during the ammonia production process 3 and used to produce urea does not contribute to the CO2 emission estimates for ammonia production presented in 4 this section. Instead, CO2 emissions resulting from the consumption of urea are attributed to the urea consumption or 5 urea application source category (under the assumption that the carbon stored in the urea during its manufacture is 6 released into the environment during its consumption or application). Emissions of CO2 resulting from agricultural 7 applications of urea are accounted for in the Agriculture chapter. Previously, these emission estimates from the 8 agricultural application of urea were accounted for in the Cropland Remaining Cropland section of the Land Use, 9 Land Use Change, and Forestry chapter. Emissions of CO2 resulting from non-agricultural applications of urea (e.g., 10 use as a feedstock in chemical production processes) are accounted for in the Urea Consumption for Non- 11 Agricultural Purposes section of this chapter. 12 Total emissions of CO2 from ammonia production in 2016 were 11.2 MMT CO2 Eq. (11,234 kt), and are 13 summarized in Table 4-18 and Table 4-19. Ammonia production relies on natural gas as both a feedstock and a fuel, 14 and as such, market fluctuations and volatility in natural gas prices affect the production of ammonia. Since 1990, 15 emissions from ammonia production have decreased by 14 percent. Emissions in 2016 have increased by 16 approximately 6 percent from the 2015 levels. 17 Table 4-18: CO2 Emissions from Ammonia Production (MMT CO2 Eq.) Source 1990 2005 2012 2013 2014 2015 2016 Ammonia Production 13.0 9.2 . 9.4 10.0 9.6 10.6 11.2 Total 13.0 9.2 9.4 10.0 9.6 10.6 11.2 18 Table 4-19: CO2 Emissions from Ammonia Production (kt) Source 1990 2005 2012 2013 2014 2015 2016 Ammonia Production 13,047 9,196 9,377 9,962 9,619 10,571 11,234 Total 13,047 9,196 9,377 9,962 9,619 10,571 11,234 19 Methodology 20 For the United States Inventory, carbon dioxide emissions from production of synthetic ammonia from natural gas 21 feedstock are estimated using a country-specific approach modified from the 2006IPCC Guidelines (IPCC 2006) 22 Tier 1 and 2 methods. In the country-specific approach, emissions are not based on total fuel requirement per the 23 2006 IPCC Guidelines due to data disaggregation limitations of energy statistics provided by the Energy 24 Information Administration (EIA). A country-specific emission factor is developed and applied to national ammonia 25 production to estimate emissions. The method uses a CO2 emission factor published by the European Fertilizer 26 Manufacturers Association (EFMA) that is based on natural gas-based ammonia production technologies that are 27 similar to those employed in the United States. This CO2 emission factor of 1.2 metric tons CCVmetric ton NH3 28 (EFMA 2000a) is applied to the percent of total annual domestic ammonia production from natural gas feedstock. 29 Emissions of CO2 from ammonia production are then adjusted to account for the use of some of the CO2 produced 30 from ammonia production as a raw material in the production of urea. The CO2 emissions reported for ammonia 31 production are reduced by a factor of 0.733 multiplied by total annual domestic urea production. This corresponds to 32 a stoichiometric CCVurea factor of 44/60, assuming complete conversion of ammonia (NH3) and CO2 to urea (IPCC 33 2006; EFMA 2000b). 34 All synthetic ammonia production and subsequent urea production are assumed to be from the same process— 35 conventional catalytic reforming of natural gas feedstock, with the exception of ammonia production from 36 petroleum coke feedstock at one plant located in Kansas. Annual ammonia and urea production are shown in Table 37 4-20. The CO2 emission factor for production of ammonia from petroleum coke is based on plant-specific data, 38 wherein all carbon contained in the petroleum coke feedstock that is not used for urea production is assumed to be 39 emitted to the atmosphere as CO2 (Bark 2004). Ammonia and urea are assumed to be manufactured in the same 40 manufacturing complex, as both the raw materials needed for urea production are produced by the ammonia Industrial Processes and Product Use 4-25 ------- 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 production process. The CO2 emission factor of 3.57 metric tons CCh/metric ton NH3 for the petroleum coke feedstock process (Bark 2004) is applied to the percent of total annual domestic ammonia production from petroleum coke feedstock. The emission factor of 1.2 metric ton CCVmetric ton NH3 for production of ammonia from natural gas feedstock was taken from the EFMA Best Available Techniques publication, Production of Ammonia (EFMA 2000a). The EFMA reported an emission factor range of 1.15 to 1.30 metric ton C02/metric ton NH3, with 1.2 metric ton CCVmetric ton NH3 as a typical value (EFMA 2000a). Technologies (e.g., catalytic reforming process, etc.) associated with this factor are found to closely resemble those employed in the United States for use of natural gas as a feedstock. The EFMA reference also indicates that more than 99 percent of the CH4 feedstock to the catalytic reforming process is ultimately converted to CO2. 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 CO2 from Fossil Fuel Combustion. See the Planned Improvements section on improvements of reporting fuel and feedstock CO2 emissions utilizing EPA's GHGRP data to improve consistency with 2006IPCC Guidelines. The total ammonia production data for 2011 through 2015 were obtained from American Chemistry Council (2016). ACC ammonia production data for 2016 was not yet available and so 2015 data were used as a proxy. For years before 2011, ammonia production data (see Table 4-20) were obtained from Coffeyville Resources (Coffeyville 2005, 2006, 2007a, 2007b, 2009, 2010, 2011, and 2012) and 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. Urea-ammonia nitrate production from petroleum coke for years through 2011 was obtained from Coffeyville Resources (Coffeyville 2005, 2006, 2007a, 2007b, 2009, 2010, 2011, and 2012), and from CVR Energy, Inc. Annual Report (CVR 2012, 2014, 2015, 2016, and 2017) for 2012 through 2016. Urea production data for 1990 through 2008 were obtained from the Minerals Yearbook: Nitrogen (USGS 1994 through 2009). Urea production data for 2009 through 2010 were obtained from the U.S. Census Bureau (U.S. Census Bureau 2010 and 2011). The U.S. Census Bureau ceased collection of urea production statistics, and urea production data for 2011 through 2015 were obtained from the Minerals Yearbook: Nitrogen (USGS 2015, 2016, 2017). USGS urea production data for 2016 was not yet published and so 2015 data were used as a proxy. Table 4-20: Ammonia Production and Urea Production (kt) Year Ammonia Production Urea Production 1990 15,425 7,450 2005 10.143 5.270 5,220 5,480 5,230 5,540 5,540 2012 10,305 2013 10,930 2014 10,515 2015 11,505 2016 11,505 Uncertainty and Time-Sern insistency 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. 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 collocated plants from the same natural gas raw 4-26 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2016 ------- 1 material. The uncertainty of the total urea production activity data, based on USGS Minerals Yearbook: Nitrogen 2 data, is a function of the reliability of reported production data and is influenced by the completeness of the survey 3 responses. In addition, due to the fact that 2016 nitrogen data has yet to be published, 2015 is used as a proxy which 4 may result in greater uncertainty. 5 Recovery of CO2 from ammonia production plants for purposes other than urea production (e.g., commercial sale, 6 etc.) has not been considered in estimating the CO2 emissions from ammonia production, as data concerning the 7 disposition of recovered CO2 are not available. Such recovery may or may not affect the overall estimate of CO2 8 emissions depending upon the end use to which the recovered CO2 is applied. Further research is required to 9 determine whether byproduct CO2 is being recovered from other ammonia production plants for application to end 10 uses that are not accounted for elsewhere. 11 The icsnlis nf ilie \ppmach 2 qiianinali\e iiiiceriaiiiis aiials sis are snniniai'i/ed 111 I able 4-2 I ( aihuii dio\ide 12 emissions I'min ammonia pmdnclkiii 111 2t> I<¦ were esiinialed lo be heluceii Id ' and 12 I \1\1TCO l!q al ilie l>5 13 pei'ceni confidence le\el This mdieales a ranue of appiiv\inialel> N peiveni helou and X pereeni aho\ e llie emission 14 esiimale of I I 2 \1\11 (() l!q 15 Table 4-21: Approach 2 Quantitative Uncertainty Estimates for CO2 Emissions from 16 Ammonia Production (MMT CO2 Eq. and Percent) - TO BE UPDATED FOR FINAL INVENTORY 17 REPORT Si ill l\v (¦;is 2016 Emission Ksiim;iu- IMMK O: i:<|.) I iHiTl.iiim Ki-hiliu- In Kmissimi I'slimiik-' (MM 1 ('(): Ku.) I.I HUT I |)|KT 1111111(1 1$111111(1 I.I HUT I |>|KT Bound liiniiid Ammonia Prodnc lion t ( ) 1 1.2 10.3 12.1 -8% +8% Range of emission estimates predicted by Monte Carlo Stochastic Simulation for a 95 percent confidence interval. 18 Methodological approaches were applied to the entire time series to ensure consistency in emissions from 1990 19 through 2016. Details on the emission trends through time are described in more detail in the Methodology section, 20 above. 21 For more information on the general QA/QC process applied to this source category, consistent with Volume 1, 22 Chapter 6 of the 2006IPCC Guidelines, see QA/QC and Verification Procedures section in the introduction of the 23 IPPU Chapter. 24 Recalculations Discussion 25 For the previous version of the Inventory (i.e., 1990 through 2015), 2015 urea production data were not published 26 and so 2014 activity data was used as a proxy. Production estimates for urea production for 2015 were updated 27 relative to the previous Inventory using information obtained from the recent 2015 Minerals Yearbook: Nitrogen 28 (USGS 2017). This update resulted in a slight increase of emissions by approximately 6 percent for 2015 relative to 29 the previous Inventory. 30 Planned Improvements 31 Future improvements involve continuing to evaluate and analyze data reported under EPA's GHGRP to improve the 32 emission estimates for the Ammonia Production source category, in particular new data from updated reporting 33 requirements finalized in October of 2014 (79 FR 63750) and December 2016 (81 FR 89188),22 that include facility- 34 level ammonia production data and feedstock consumption. This data will first be reported by facilities in 2018 and 35 available post-verification to assess in early 2019 for use in future reports (e.g. 2020 Inventory report) if the data 36 meets GHGRP CBI aggregation criteria. Particular attention will be made to ensure time-series consistency of the 37 emission estimates presented in future Inventory reports, along with application of appropriate category-specific QC 38 procedures consistent with IPCC and UNFCCC guidelines. For example, data reported in 2018 will reflect activity 22 See . Industrial Processes and Product Use 4-27 ------- 1 in 2017 and may not be representative of activity in prior years of the time series. This assessment is required as the 2 new facility-level reporting data from EPA's GHGRP associated with new requirements is only applicable starting 3 with reporting of emissions in calendar year 2017, and thus is not available for all inventory years (i.e., 1990 through 4 2016) as required for this Inventory. 5 In implementing improvements and integration of data from EPA's GHGRP, the latest guidance from the IPCC on 6 the use of facility-level data in national inventories will be relied upon.23 Specifically, the planned improvements 7 include assessing the anticipated new data to update the emission factors to include both fuel and feedstock CO2 8 emissions to improve consistency with 2006 IPCC Guidelines, in addition to reflecting CO2 capture and storage 9 practices (beyond use of CO2 for urea production). Methodologies will also be updated if additional ammonia 10 production plants are found to use hydrocarbons other than natural gas for ammonia production. Due to limited 11 resources and ongoing data collection effort, this planned improvement is still in development and so is not 12 incorporated into this Inventory. 13 4.6 Urea Consumption for Non-Agricultural 14 Purposes 15 Urea is produced using ammonia and carbon dioxide (CO2) as raw materials. All urea produced in the United States 16 is assumed to be produced at ammonia production facilities where both ammonia and CO2 are generated. There were 17 31 plants producing ammonia in the United States during 2016, with two additional plants sitting idle for the entire 18 year (USGS 2017b). 19 The chemical reaction that produces urea is: 20 2NH3+ C02 -> NH2COONH4 -> CO(NH2)2 + H20 21 This section accounts for CO2 emissions associated with urea consumed exclusively for non-agricultural purposes. 22 Carbon dioxide emissions associated with urea consumed for fertilizer are accounted for in the Agriculture chapter. 23 Urea is used as a nitrogenous fertilizer for agricultural applications and also in a variety of industrial applications. 24 The industrial applications of urea include its use in adhesives, binders, sealants, resins, fillers, analytical reagents, 25 catalysts, intermediates, solvents, dyestuffs, fragrances, deodorizers, flavoring agents, humectants and dehydrating 26 agents, formulation components, monomers, paint and coating additives, photosensitive agents, and surface 27 treatments agents. In addition, urea is used for abating nitrogen oxide (NOx) emissions from coal-fired power plants 28 and diesel transportation motors. 29 Emissions of CO2 from urea consumed for non-agricultural purposes in 2016 were estimated to be 4.0 MMT CO2 30 Eq. (3,959 kt), and are summarized in Table 4-22 and Table 4-23. Net CO2 emissions from urea consumption for 31 non-agricultural purposes in 2016 have increased by approximately 5 percent from 1990 and decreased by 32 approximately 5 percent from 2015. The significant decrease in emissions during 2014 can be attributed to a 33 decrease in the amount of urea imported by the United States during that year. 34 Table 4-22: CO2 Emissions from Urea Consumption for Non-Agricultural Purposes (MMT CO2 35 Eq.) Source 1990 2005 2012 2013 2014 2015 2016 Urea Consumption 3.8 3.7 4.4 4.1 1.5 4.2 4.0 Total 3.8 3.7 4.4 4.1 1.5 4.2 4.0 23 See . 4-28 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2016 ------- 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 Table 4-23: CO2 Emissions from Urea Consumption for Non-Agricultural Purposes (kt) Source 1990 2005 2012 2013 2014 2015 2016 Urea Consumption 3,784 C 3,653 4,392 4,074 1,541 4,169 3,959 Total 3,784 3,653 4,392 4,074 1,541 4,169 3,959 Methodology Emissions of CO2 resulting from urea consumption for non-agricultural purposes are estimated by multiplying the amount of urea consumed in the United States for non-agricultural purposes 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, and consistent with the 2006IPCC Guidelines. 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) and is reported in Table 4-24, from the total domestic supply of urea. In previous Inventory reports, the quantity of urea fertilizer applied to agricultural lands was obtained directly from the Cropland Remaining Cropland section of the Land Use, Land Use Change, and Forestry chapter. The domestic supply of urea is estimated based on the amount of urea produced plus the sum of net urea imports and 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 producing urea from ammonia and CO2. This corresponds to a stoichiometric CCh/urea factor of 44/60, assuming complete conversion of NH3 and CChto urea (IPCC 2006; EFMA 2000). Urea production data for 1990 through 2008 were obtained from the Minerals Yearbook: Nitrogen (USGS 1994 through 2009). 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, therefore, urea production data from 2011 to 2015 were obtained from the Minerals Yearbook: Nitrogen (USGS 2014 through 2017a). Urea production data for 2016 are not yet publicly available and so 2015 data (ACC 2015) have been used as proxy. Urea import data for 2016 are not yet publicly available and so 2015 data have been used as proxy. Urea import data for 2013 to 2015 were obtained from the Minerals Yearbook: Nitrogen (USGS 2016; USGS 2017a). 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 1997 through 2010 (U.S. Census Bureau 2001 through 2011), The Fertilizer Institute (TFI2002) 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 4-24). Urea export data for 2016 are not yet publicly available and so 2015 data have been used as proxy. Urea export data for 2013 to 2015 were obtained from the Minerals Yearbook: Nitrogen (USGS 2016; USGS 2017a). 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. Industrial Processes and Product Use 4-29 ------- 1 Table 4-24: Urea Production, Urea Applied as Fertilizer, Urea Imports, and Urea Exports (kt) Year Urea Production Urea Applied as Fertilizer Urea Imports Urea Exports 1990 7,450 3,296 1,860 854 2005 5,270 4.779 5.026 536 2012 5,220 5,838 6,944 336 2013 5,480 6,059 6,470 335 2014 5,230 6,188 3,510 451 2015 5,540 6,665 7,190 380 2016 5,540 6,952 7,190 380 2 Uncertainty and Time-Series Consistency 3 There is limited publicly-available data on the quantities of urea produced and consumed for non-agricultural 4 purposes. Therefore, the amount of urea used for non-agricultural purposes is estimated based on a balance that 5 relies on estimates of urea production, urea imports, urea exports, and the amount of urea used as fertilizer. The 6 primary uncertainties associated with this source category are associated with the accuracy of these estimates as well 7 as the fact that each estimate is obtained from a different data source. Because urea production estimates are no 8 longer available from the USGS, there is additional uncertainty associated with urea produced beginning in 2011. 9 There is also uncertainty associated with the assumption that all of the carbon in urea is released into the 10 environment as CO2 during use. 11 The icmiIis of 1 he \ppmadi 2 i|ii;inlil;ili\e iiiiceriaiiiis aiials sis are siininian/ed 111 I able 4-25 ( aihun dio\ide 12 emissions associated u illi urea aiiisiinipiiiin IV»r iioii-auricnliiiral purposes were esiimaled lo he heluceii v5 and 4 4 13 \1\1I 'CO I !c| ;illhc l>5 perceiii confidence le\ el This mdieales a rauue nf apprn\inialcl> 12 pereeui helow and 12 14 pereeiii alxne llie emission esiimale ol'4 0 \l\ITCO l!i|. 15 Table 4-25: Approach 2 Quantitative Uncertainty Estimates for CO2 Emissions from Urea 16 Consumption for Non-Agricultural Purposes (MMT CO2 Eq. and Percent) - TO BE UPDATED 17 FOR FINAL INVENTORY REPORT Si hi I'l l' (¦as 2016 llmission I'isiimaU' (MMT CO: l.(|.) I ni'iTlaiim kan^i' Ki-laliu- In 1". miss ion Ksiimak"1 (MM 1 ( (): l.ii.) ("..) 1 .ON l'l* I ppiT 1 {oiiiid 1 Sound I.I HUT I ppi'l' Bound liound I Jrea Consiimp for Non-Agrii Purposes lion uiltural t ( ) 4.0 3.5 4.4 -12% +12% Range of emission estimates predicted by Monte Carlo Stochastic Simulation for a 95 percent confidence interval. 18 Methodological approaches were applied to the entire time series to ensure consistency in emissions from 1990 19 through 2016 Methodological approaches were applied to the entire time series to ensure consistency above. 20 For more information on the general QA/QC process applied to this source category, consistent with Volume 1, 21 Chapter 6 of the 2006IPCC Guidelines, see QA/QC and Verification Procedures section in the introduction of the 22 IPPU Chapter. 23 Recalculations Discussion 24 The amount of urea consumed for agricultural purposes (used for calculating urea consumption for non-agricultural 25 purposes) in the United States for the years 1990 through 2016 was revised based on updated urea application 26 estimates obtained from the Agriculture chapter (see Table 5-25). These updates resulted in the following changes to 27 the emission estimates relative to the previous Inventory report: a decrease of less than 1 percent in 2012, an 4-30 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2016 ------- 1 increase of 1.5 percent in 2013, an increase of 12 percent in 2014, and an increase of 270 percent in 2015. As stated 2 previously in the Methodology section, in previous Inventory reports the quantity of urea fertilizer applied to 3 agricultural lands was obtained directly from the Cropland Remaining Cropland section of the Land Use, Land-Use 4 Change, and Forestry chapter; urea consumption is reported in the Agriculture chapter for the current report. 5 4.7 Nitric Acid Production (CRF Source 6 Category 2B2) 7 Nitrous oxide (N20) is emitted during the production of nitric acid (HNO3), an inorganic compound used primarily 8 to make synthetic commercial fertilizers. It is also a major component in the production of adipic acid—a feedstock 9 for nylon—and explosives. Virtually all of the nitric acid produced in the United States is manufactured by the high- 10 temperature catalytic oxidation of ammonia (EPA 1998). There are two different nitric acid production methods: 11 weak nitric acid and high-strength nitric acid. The first method utilizes oxidation, condensation, and absorption to 12 produce nitric acid at concentrations between 30 and 70 percent nitric acid. High-strength acid (90 percent or greater 13 nitric acid) can be produced from dehydrating, bleaching, condensing, and absorption of the weak nitric acid. The 14 basic process technology for producing nitric acid has not changed significantly over time. Most U.S. plants were 15 built between 1960 and 2000. As of 2016, there were 35 active weak nitric acid production plants, including one 16 high-strength nitric acid production plant in the United States (EPA 2010; EPA 2017). 17 During this reaction, N20 is formed as a byproduct and is released from reactor vents into the atmosphere. 18 Emissions from fuels consumed for energy purposes during the production of nitric acid are accounted for in the 19 Energy chapter. 20 Nitric acid is made from the reaction of ammonia (NH3) with oxygen (O2) in two stages. The overall reaction is: 21 4NH3 + 802 -> 4HNO:i +4H20 22 Currently, the nitric acid industry controls emissions of NO and NO2 (i.e., NOx). As such, the industry in the United 23 States uses a combination of non-selective catalytic reduction (NSCR) and selective catalytic reduction (SCR) 24 technologies. In the process of destroying NOx, NSCR systems are also very effective at destroying N20. However, 25 NSCR units are generally not preferred in modern plants because of high energy costs and associated high gas 26 temperatures. NSCR systems were installed in nitric plants built between 1971 and 1977 with NSCRs installed at 27 approximately one-third of the weak acid production plants. U.S. facilities are using both tertiary (i.e., NSCR) and 28 secondary controls (i.e., alternate catalysts). 29 Nitrous oxide emissions from this source were estimated to be 10.2 MMT CO2 Eq. (34 kt of N20) in 2016 (see 30 Table 4-26). Emissions from nitric acid production have decreased by 16 percent since 1990, with the trend in the 31 time series closely tracking the changes in production. Emissions have decreased by 30 percent since 1997, the 32 highest year of production in the time series. 33 Table 4-26: N2O Emissions from Nitric Acid Production (MMT CO2 Eq. and kt N2O) Year MMT CO2 Eq. kt N2O 1990 12.1 41 2005 11.3 38 35 36 37 39 34 2012 10.5 2013 10.7 2014 10.9 2015 11.6 2016 10.2 Industrial Processes and Product Use 4-31 ------- 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 Methodology Emissions of N20 were calculated using the estimation methods provided by the 2006IPCC Guidelines and country specific methods from EPA's GHGRP. The 2006IPCC Guidelines Tier 2 method was used to estimate emissions from nitric acid production for 1990 through 2009, and a country-specific approach similar to the IPCC Tier 3 method was used to estimate N20 emissions for 2010 through 2016. 2010 through 2016 Process N20 emissions and nitric acid production data were obtained directly from EPA's GHGRP for 2010 through 2016 by aggregating reported facility-level data (EPA 2017). In the United States, all nitric acid facilities producing weak nitric acid (30 to 70 percent in strength) are required to report annual greenhouse gas emissions data to EPA as per the requirements of its GHGRP. As of 2016, there were 35 facilities that reported to EPA, including the known single high-strength nitric acid production facility in the United States (EPA 2017). All nitric acid (weak acid) facilities are required to calculate process emissions using a site-specific emission factor developed through annual performance testing under typical operating conditions or by directly measuring N20 emissions using monitoring equipment.24 The high-strength nitric acid facility also reports N20 emissions associated with weak acid production and this may capture all relevant emissions, pending additional further EPA research. More details on the calculation, monitoring and QA/QC methods applicable to nitric acid facilities can be found under Subpart V: Nitric Acid Production of the regulation, Part 98.25 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. Based on the results of the verification process, the EPA follows up with facilities to resolve mistakes that may have occurred.26 To calculate emissions from 2010 through 2016, the GHGRP nitric acid production data are utilized to develop weighted country specific emission factors used to calculate emissions estimates. Based on aggregated nitric acid production data by abatement type (i.e., with, without) provided by EPA's GHGRP, the percent of production values and associated emissions of nitric acid with and without abatement technologies are calculated. These percentages are the basis for developing the country specific weighted emission factors which vary from year to year based on the amount of nitric acid production with and without abatement technologies. 1990 through 2009 Using GHGRP data for 2010,27 country-specific N20 emission factors were calculated for nitric acid production with abatement and without abatement (i.e., controlled and uncontrolled emission factors), as previously stated. The following 2010 emission factors were derived for production with abatement and without abatement: 3.3 kg N20/metric ton HNO3 produced at plants using abatement technologies (e.g., tertiary systems such as NSCR systems) and 5.99 kg N20/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 N20 emissions from nitric acid production for years prior to the availability of GHGRP data (i.e., 1990 through 2008 and 2009). A separate weighted factor is included for 2009 due to data availability for that year. At that time, EPA had initiated compilation of a nitric acid database to improve estimation of emissions from this industry and obtained updated information on application of controls via review of permits and outreach with facilities and trade associations. The research indicated recent installation of abatement technologies at additional facilities. 24 Facilities must use standard methods, either EPA Method 320 or ASTM D6348-03 and must follow associated QA/QC procedures consistent during these performance test consistent with category-specific QC of direct emission measurements. 25 See . 26 See . 27 National N2O process emissions, national production, and national share of nitric acid production with abatement and without abatement technology was aggregated from the GHGRP facility-level data for 2010 to 2016 (i.e., percent production with and without abatement). 4-32 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2016 ------- 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 Based on the available data, it was assumed that emission factors for 2010 would be more representative of operating conditions in 1990 through 2009 than more recent years. Initial review of historical data indicates that percent production with and without abatement can change over time and also year over year due to changes in application of facility-level abatement technologies, maintenance of abatement technologies, and also due to plant closures and start-ups (EPA 2012, 2013; Desai 2012; CAR 2013). The installation dates of N20 abatement technologies are not known at most facilities, but it is assumed that facilities reporting abatement technology use have had this technology installed and operational for the duration of the time series considered in this report (especially NSCRs). The country-specific weighted N20 emission factors were used in conjunction with annual production to estimate N20 emissions for 1990 through 2009, using the following equations: Ei — Pi X EFwelg)lt:ecl l where, EFWeighted,i = [(°/oPc,i X EFc) + (%Punc,L X EFunc)\ E = Annual N20 Emissions for year i (kg/yr) Pi = Annual nitric acid production for year i (metric tons HNO3) EF weighted,i = Weighted N20 emission factor for year i (kg N20/metric ton HNO3) %Pc,i = Percent national production of HNO3 with N20 abatement technology (%) EFC = N20 emission factor, with abatement technology (kg N20/metric ton HNO3) %Punc,i = Percent national production of HNO3 without N20 abatement technology (%) EFunc = N20 emission factor, without abatement technology (kg N20/metric ton HNO3) i = year from 1990 through 2009 For 2009: Weighted N20 emission factor = 5.46 kg N20/metric ton HNO3. For 1990 through 2008: Weighted N20 emission factor = 5.66 kg N20/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 4-27). Publicly-available information on plant-level abatement technologies was used to estimate the shares of nitric acid production with and without abatement for 2008 and 2009 (EPA 2012, 2013; Desai 2012; CAR 2013). EPA has previously conducted a review of operating permits to obtain more current information due to the lack of publicly-available data on use of abatement technologies for 1990 through 2007, as stated previously; therefore, the share of national production with and without abatement for 2008 was assumed to be constant for 1990 through 2007. Table 4-27: Nitric Acid Production (kt) Year kt 1990 7,200 2005 6.710 2012 7,460 2013 7,580 2014 7,660 2015 7,210 2016 7,810 Uncertainty and Time-Series Consistency Uncertainty associated with the parameters used to estimate N20 emissions includes the share of U.S. nitric acid production attributable to each emission abatement technology 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 Industrial Processes and Product Use 4-33 ------- 1 (especially prior to 2010) for a variety of facility level variables, including plant specific production levels, plant 2 production technology (e.g., low, high pressure, etc.), and abatement technology type, installation date of abatement 3 technology, and accurate destruction and removal efficiency rates. Production data prior to 2010 were obtained from 4 National Census Bureau, which does not provide uncertainty estimates with their data. Facilities reporting to EPA's 5 GHGRP must measure production using equipment and practices used for accounting purposes. At this time EPA 6 does not estimate uncertainty of the aggregated facility-level information. As noted in the Methodology section, 7 EPA verifies annual facility-level reports through a multi-step process (e.g., combination of electronic checks and 8 manual reviews by staff) to identify potential errors and ensure that data submitted to EPA are accurate, complete, 9 and consistent. The annual production reported by each nitric acid facility under EPA's GHGRP and then 10 aggregated to estimate national N20 emissions is assumed to have low uncertainty. 11 The results olThis \pproach 2 quauiiiali\ e uuceriaiuis aiials sis are sununari/ed in I able 4-2S \iirous o\ide 12 emissions limn uiiiicacid producnm! were esiimaled In he helueeu ') "and Ins \l\l I (() l!q al I lie l>5 perceui 13 confidence le\el This mdicales a rauue of approMinaleK 5 peiveui below lo <> peiveui aho\e llie )I(> emissions 14 esiiniale of Id 2 \1\1T CO l!q. 15 Table 4-28: Approach 2 Quantitative Uncertainty Estimates for N2O Emissions from Nitric 16 Acid Production (MMT CO2 Eq. and Percent) - TO BE UPDATED FOR FINAL INVENTORY 17 REPORT Si ill I'l l" 2016 l.missiim l'.slim;ili' (MM 1 ( (): Kii.) I iiii'i'hiiiih l{:iii|;r Riliiliu- In Kmissimi l;.sliiii;iu-' (MM 1 ( (): l u.) ("..) I.I HUT I |1|KT I.IHUT I |1|KT lilllllld 1$111111(1 1 $111111(1 1$111111(1 Nitric Acid l'n )duction N:() 10.2 9.7 10.8 -5% +6% Range of emission estimates predicted by Monte Carlo Stochastic Simulation lor a l)5 percent confidence interval. 18 Methodological approaches were applied to the entire time series to ensure consistency in emissions from 1990 19 through 2016. To maintain consistency across the time series and with the rounding approaches taken by other data 20 sets, a new rounding approach was performed for the GHGRP Subpart V: Nitric Acid data. This resulted in 21 production data changes across the time series of 2010 to 2016, in which EPA's GHGRP data have been utilized. 22 The results of this update have had an insignificant impact on the emission estimates across the 2010 to 2016 time 23 series. Details on the emission trends through time are described in more detail in the Methodology section, above. 24 For more information on the general QA/QC process applied to this source category, consistent with Volume 1, 25 Chapter 6 of the 2006IPCC Guidelines, see QA/QC and Verification Procedures section in the introduction of the 26 IPPU Chapter. 27 Planned Improvements 28 Pending resources, EPA is considering both near-term and long-term improvement to estimates and associated 29 characterization of uncertainty. In the short-term, with 7 years of EPA's GHGRP data, EPA anticipates completing 30 updates of category-specific QC procedures to potentially also improve both qualitative and quantitative uncertainty 31 estimates. Longer term, in 2020, EPA anticipates having information from EPA's GHGRP facilities on the 32 installation date of any N20 abatement equipment, per recent revisions finalized in December 2016 to EPA's 33 GHGRP. This information will enable more accurate estimation of N20 emissions from nitric acid production over 34 the time series. 35 4.8 Adipic Acid Production (CRF Source 36 Category 2B3) 37 Adipic acid is produced through a two-stage process during which nitrous oxide (N20) is generated in the second 38 stage. Emissions from fuels consumed for energy purposes during the production of adipic acid are accounted for in 4-34 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2016 ------- 1 the Energy chapter. The first stage of manufacturing usually involves the oxidation of cyclohexane to form a 2 cyclohexanone/cyclohexanol mixture. The second stage involves oxidizing this mixture with nitric acid to produce 3 adipic acid. Nitrous oxide is generated as a byproduct of the nitric acid oxidation stage and is emitted in the waste 4 gas stream (Thiemens and Trogler 1991). The second stage is represented by the following chemical reaction: 5 ('CH2)5CO(cyclohexanone) + (CH2)zCHOH (cyclohexanol) + wHN03 6 -» HOOC(CH2)4COOH(adipic acid) + xN20 + yH20 7 Process emissions from the production of adipic acid vary with the types of technologies and level of emission 8 controls employed by a facility. In 1990, two major adipic acid-producing plants had N20 abatement technologies in 9 place and, as of 1998, three major adipic acid production facilities had control systems in place (Reimer et al. 1999). 10 In 2016, catalytic reduction, non-selective catalytic reduction (NSCR) and thermal reduction abatement technologies 11 were applied as N20 abatement measures at adipic acid facilities (EPA 2017). 12 Worldwide, only a few adipic acid plants exist. The United States, Europe, and China are the major producers, with 13 the United States accounting for the largest share of global adipic acid production capacity in recent years. In 2016, 14 the United States had two companies with a total of two adipic acid production facilities (one in Texas and one in 15 Florida) following the ceased operations of a third major production facility at the end of 2015 (EPA 2017). 16 Adipic acid is a white crystalline solid used in the manufacture of synthetic fibers, plastics, coatings, urethane 17 foams, elastomers, and synthetic lubricants. Commercially, it is the most important of the aliphatic dicarboxylic 18 acids, which are used to manufacture polyesters. Eighty-four percent of all adipic acid produced in the United States 19 is used in the production of nylon 6,6; 9 percent is used in the production of polyester polyols; 4 percent is used in 20 the production of plasticizers; and the remaining 4 percent is accounted for by other uses, including unsaturated 21 polyester resins and food applications (ICIS 2007). Food grade adipic acid is used to provide some foods with a 22 "tangy" flavor (Thiemens and Trogler 1991). 23 National adipic acid production has increased by approximately 40 percent over the period of 1990 through 2016, to 24 approximately 1,055,000 metric tons (ACC 2016). Nitrous oxide emissions from adipic acid production were 25 estimated to be 7.0 MMT CO2 Eq. (23 kt N2O) in 2016 (see Table 4-29). Over the period 1990 through 2016, 26 emissions have been reduced by 54 percent due to both the widespread installation of pollution control measures in 27 the late 1990s and plant idling in the late 2000s. Very little information on annual trends in the activity data exist for 28 adipic acid. 29 Table 4-29: N2O Emissions from Adipic Acid Production (MMT CO2 Eq. and kt N2O) Year MMT CO2 Eq. kt N2O 1990 15.2 51 2005 7.1 24 2012 5.5 19 2013 3.9 13 2014 5.4 18 2015 4.3 14 2016 7.0 23 30 Methodology 31 Emissions are estimated using both Tier 2 and Tier 3 methods consistent with the 2006IPCC Guidelines. Due to 32 confidential business information, plant names are not provided in this section. Therefore, the four adipic acid- 33 producing facilities that have operated over the time series will be referred to as Plants 1 through 4. Overall, as noted 34 above, the two currently operating facilities use catalytic reduction, NSCR and thermal reduction abatement 35 technologies. Industrial Processes and Product Use 4-35 ------- 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 2010 through 2016 All emission estimates for 2010 through 2016 were obtained through analysis of GHGRP data (EPA 2014 through 2017), which is consistent with the 2006IPCC Guidelines Tier 3 method. Facility-level greenhouse gas emissions data were obtained from EPA's GHGRP for the years 2010 through 2016 (EPA 2014 through 2017) and aggregated to national N20 emissions. Consistent with IPCC Tier 3 methods, all adipic acid production facilities are required to calculate emissions using a facility-specific emission factor developed through annual performance testing under typical operating conditions or by directly measuring N20 emissions using monitoring equipment.28 More information on the calculation, monitoring and QA/QC methods for process N20 emissions applicable to adipic acid production facilities under Subpart E can be found in the electronic code of federal regulations.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 1990 through 2009 For years prior to EPA's GHGRP reporting, for both Plants 1 and 2, 1990 to 2009 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 confidential business information 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 N20 emissions were estimated using the following Tier 2 equation from the 2006 IPCC Guidelines'. Eaa = Qaa x EFaa X (1 - [DF X UF]) where, Eaa N20 emissions from adipic acid production, metric tons Qaa Quantity of adipic acid produced, metric tons EFaa Emission factor, metric ton N20/metric ton adipic acid produced DF N20 destruction factor UF Abatement system utility factor The adipic acid production is multiplied by an emission factor (i.e., N20 emitted per unit of adipic acid produced), which has been estimated, based on experiments that the reaction stoichiometry for N20 production in the preparation of adipic acid, to be approximately 0.3 metric tons of N20 per metric ton of product (IPCC 2006). The "N20 destruction factor" in the equation represents the percentage of N20 emissions that are destroyed by the installed abatement technology. The "abatement system utility factor" represents the percentage of time that the 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 2016; 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 28 Facilities must use standard methods, either EPA Method 320 or ASTM D6348-03, and must follow associated QA/QC procedures during these performance tests consistent with category-specific QC of direct emission measurements. 29 See . 30 See . 4-36 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2016 ------- 1 same as 1994 data. The 1997 plant capacities were taken from Chemical Market Reporter, "Chemical Profile: 2 Adipic Acid" (CMR 1998). The 1998 plant capacities for all four plants and 1999 plant capacities for three of the 3 plants were obtained from Chemical Week, Product Focus: Adipic Acid/Adiponitrile (CW 1999). Plant capacities for 4 2000 for three of the plants were updated using Chemical Market Reporter, "Chemical Profile: Adipic Acid" (CMR 5 2001). For 2001 through 2003, the plant capacities for three plants were held constant at year 2000 capacities. Plant 6 capacity for 1999 to 2003 for the one remaining plant was kept the same as 1998. 7 National adipic acid production data (see Table 4-30) from 1990 through 2015 were obtained from the American 8 Chemistry Council (ACC 2016). Updated ACC data for 2016 was not currently available and 2015 data were used 9 as a proxy. 10 Table 4-30: Adipic Acid Production (kt) Year kt 1990 755 2005 865 2012 950 2013 980 2014 1,025 2015 1,055 2016 1,055 11 Uncertainty and Time-Series Consistency 12 Uncertainty associated with N20 emission estimates includes the methods used by companies to monitor and 13 estimate emissions. While some information has been obtained through outreach with facilities, limited information 14 is available over the time series on these methods, abatement technology destruction and removal efficiency rates 15 and plant specific production levels. 16 The ivsiilis olTlns \ppronch 2 i|ii;inlil;ili\c iiiiceilami\ ;111;11\ sis are snnininii/cd in Table 4-' I \iirons o\idc 17 emissions from adipie acid production I'm' 2u I(> were esiimaled In he heUxeen <•." and ~ ' \1\1l (() I a| al llic l>5 18 pci'cciil confidence lc\cl. These \nines indicnlc n innue of ;ippi'o\ininlcl> 4 pcrccni heknx In 4 peiceni nho\c 1 he 19 2dl------- 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 4.9 Caprolactam, Glyoxal and Glyoxylic Acid Production (CRF Source Category 2B4) Caprolactam Caprolactam (CV,Hi iNO) is a colorless monomer produced for nylon-6 fibers and plastics, with a substantial proportion of the fiber used in carpet manufacturing. Commercial processes for the manufacture of caprolactam are based on either toluene or benzene. The production of caprolactam can give rise to emissions of nitrous oxide (N20). During the production of caprolactam, emissions of N20 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 (Reimschuessel 1977). Where caprolactam is produced from benzene, the main process, the benzene is hydrogenated to cyclohexane which is then oxidized to produce cyclohexanone (CY,Hh,0). The classical route (Raschig process) and basic reaction equations for production from cyclohexanone are (Reimschuessel 1977; Lowenheim and Moran 1975; IPCC 2006): Oxidation of NH3 to NO/N02 I NH3 reacted with C02/H20 to yield ammonium carbonate (NH4)2C03 I (NH4)2C03 reacted with NO/N02 (from NH3 oxidation) to yield ammonium nitrite (NH4N02) I NH3 reacted with S02/H20 to yield ammonium bisulphite (NH4HS03) I NH4N02 and (NH4HS03) reacted to yield hydroxylamine disulphonate (NOH(S03NH4)2) I (NOH(S03NH4)2) hydrolised to yield hydroxylamine sulphate ({NH2OH)2. H2S04) and ammonium sulphate ((NH4)2S04) I Cylohexanone reaction-. 1 C6H10O + ~(NH2OH)2.H2S04(+NH3 and H2S04) -> C6H10NOH + (NH4)2S04 + H20 I Beckmann rearrangement: C6H10NOH (+H2S04 and S02) -> C6HuNO.H2S04 (+4NH3 and H20) -> C6HuNO + 2(NH4)2S04 In 1999, there were four caprolactam production facilities in the United States. As of 2016, the United States had 3 companies with a total of 3 caprolactam production facilities: AdvanSix in Virginia (AdvanSix 2017), BASF in Texas (BASF 2017), and Fibrant LLC in Georgia (Fibrant 2017) (TechSci n.d. 2017). Nitrous oxide emissions from caprolactam production in the United States were estimated to be 2.0 MMT CO2 Eq. (7 kt N20) in 2016 (see Table 4-32). National caprolactam production has increased by approximately 21 percent over the period of 1990 through 2016, to approximately 760 thousand metric tons (ACC 2015). Very little information on annual trends in the activity data exist for caprolactam. 4-38 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2016 ------- 1 Table 4-32: N2O Emissions from Caprolactam Production (MMT CO2 Eq. and kt N2O) Year MMT CO2 Eq. kt N2O 1990 1.7 6 2005 2.1 " 2012 2.0 7 2013 2.0 7 2014 2.0 7 2015 2.0 7 2016 2.0 7 2 3 Glyoxal 4 Glyoxal is mainly used as a crosslinking agent for acrylic resins, disinfectant, gelatin hardening agent, and textile 5 finishing agent etc. It's produced from oxidation of acetaldehyde with concentrated nitric acid, or from the catalytic 6 oxidation of ethylene glycol, and N20 is emitted in the process of oxidation of acetaldehyde. 7 8 Glyoxal (ethanedial) (C2H2O2) is produced from oxidation of acetaldehyde (ethanal) (C2H4O) with concentrated 9 nitric acid (HNO3). Glyoxal can also be produced from catalytic oxidation of ethylene glycol (ethanediol) 10 (CH2OHCH2OH). Glyoxal is used as a crosslinking agent for vinyl acetate/acrylic resins, disinfectant, gelatin 11 hardening agent, textile finishing agent (permanent-press cotton, rayon fabrics), wet-resistance additive (paper 12 coatings) (Ashford 1994; IPCC 2006). 13 14 Glyoxylic Acid 15 Glyoxylic acid is produced by nitric acid oxidation of glyoxal. Glyoxylic acid is used for the production of synthetic 16 aromas, agrochemicals and pharmaceutical intermediates (Babusiaux 2005). 17 The EPA does not currently estimate the emissions associated with the production of Glyoxal and Glyoxylic Acid 18 due to data availability and a lack of publicly available information on the industry in the United States. 19 Methodology 20 Emissions of N2O were calculated using the estimation methods provided by the 2006 IPCC Guidelines. The 2006 21 IPCC Guidelines Tier 1 method was used to estimate emissions from caprolactam production for 1990 through 22 2016, as shown in this formula: 23 EN20 = EF x CP 24 where, 25 En2o — Annual N2O Emissions (kg) 26 EF = N20 emission factor (default) (kg N20/metric ton caprolactam produced) 27 CP = Caprolactam production (metric tons) 28 During the caprolactam production process, nitrous oxide is generated as a byproduct of the high temperature 29 catalytic oxidation of ammonia (NH3), which is the first reaction in the series of reactions to produce caprolactam. 30 The amount of nitrous oxide emissions can be estimated based on the chemical reaction shown above. Based on this 31 formula, which is consistent with an IPCC Tier 1 approach, approximately 111.1 metric tons of caprolactam are 32 required to generate one metric ton of N20, or an emission factor of 9.0 kg N20 per metric ton of caprolactam (IPCC 33 2006). When applying the Tier 1 method, the 2006IPCC Guidelines state that it is good practice to assume that 34 there is no abatement of N20 emissions and to use the highest default emission factor available in the guidelines. In 35 addition, EPA did not find support for the use of secondary catalysts to reduce N20 emissions, like those employed 36 at nitric acid plants. Thus, the 760 thousand metric tons (kt) of caprolactam produced in 2016 (ACC 2015) resulted 37 in N2O emissions of approximately 2.0 MMT CO2 Eq. (7 kt). Industrial Processes and Product Use 4-39 ------- 1 The activity data for caprolactam production (see Table 4-33) from 1990 to 2015 were obtained from the ACC 2 Guide to the Business of Chemistry report (ACC 2015). For 2016, caprolactam production data was not available 3 and EPA used 2015 production data as proxy. EPA will continue to analyze and assess alternative sources of 4 production data as a quality control measure. 5 Table 4-33: Caprolactam Production (kt) Year kt 1990 626 2005 795 2012 750 2013 750 2014 755 2015 760 2016 760 6 7 Carbon dioxide and methane emissions may also occur from the production of caprolactam but currently the IPCC 8 does not have methodologies for calculating these emissions associated with caprolactam production. 9 Uncertainty and Time-Series Consistency 10 Estimation of emissions of N20 from caprolactam production can be treated as analogous to estimation of 11 emissions of N20 from nitric acid production. Both production processes involve an initial step of NH3 oxidation 12 which is the source of N20 formation and emissions (IPCC 2006). Therefore, uncertainties for the default values in 13 the 2006IPCC Guidelines is an estimate based on default values for nitric acid plants. In general, default emission 14 factors for gaseous substances have higher uncertainties because mass values for gaseous substances are influenced 15 by temperature and pressure variations and gases are more easily lost through process leaks. The default values for 16 caprolactam production have a relatively high level of uncertainty due to the limited information available (IPCC 17 2006). 18 The rcsnlis olThis \pproach 2 i|ii;inlil;ili\ c iiiiccriainis aiials sis are summari/cd in I able 4-^4 \iirons o\ide 19 emissions from (aprolaclam. (il\o\al and (il>i'\\ lie \cid hodiiclion lor 2<> l<> were csiinialcd lo he helween I 2 20 and 2 S \I\1T CO I al llie l>5 pereeni confidence lex el. These \ allies mdicalc a mime of appro\imalel> 4 emission esiiniale of 2 u \I\1T ('() I a| 22 Table 4-34: Approach 2 Quantitative Uncertainty Estimates for N2O Emissions from 23 Caprolactam, Glyoxal and Glyoxylic Acid Production (MMT CO2 Eq. and Percent) - TO BE 24 UPDATED FOR FINAL INVENTORY REPORT Siuinv (¦us 2016 riiiiissiiui I'.sliniiik' (MM 1 CO: l.i|.) I luvriiiiim in I", miss in 11 r'.siiinuk-1 (MM 1 ( (): l.i|.) ("..) I.I HUT I |1|KT I.IIIUT I |1|KT liiillllil Bound liiillllil liiillllil Caprolacta 111 Production \ ( 1 2.0 1.2 2.8 -40% +40% Range of emission estimates predicted by Monte Carlo Stochastic Simulation for a 95 percent confidence interval. 25 Details on the emission trends through time are described in more detail in the Methodology section, above. 26 Caprolactam was not reported as a source category in previous Inventory reports (previously reported as "NE") and 27 EPA has taken measures to ensure emission estimates are consistent with 2006 IPCC Guidelines and good practice, 28 ensuring time-series consistency. 4-40 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2016 ------- 1 For more information on the general QA/QC process applied to this source category, consistent with Volume 1, 2 Chapter 6 of the 2006IPCC Guidelines, see QA/QC and Verification Procedures section in the introduction of the 3 IPPU Chapter. 4 Recalculations Discussion 5 With the addition of caprolactam production emissions reported within the current Inventory report, recalculations 6 have occurred to the IPPU chapter aggregated total emissions estimate. Across the 1990 to 2016 time series, the 7 addition of caprolactam production has added emissions ranging between 1.6 and 2.2 MMT CO2 Eq. 8 Planned Improvements 9 Pending resources, EPA will research other available datasets for caprolactam production and industry trends, 10 including facility-level data. EPA will also research the production process and emissions associated with the 11 production of glyoxal and glyoxylic acid. During the Expert Review comment period for the current Inventory 12 report, EPA sought expert solicitation on data available for these emissions source categories. EPA did not receive 13 information regarding these industries during Expert Review but will continue to research alternative datasets. 14 4.10 Silicon Carbide Production and 15 Consumption (CRF Source Category 2B5) 16 Carbon dioxide (CO2) and methane (CH4) are emitted from the production of silicon carbide (SiC), a material used 17 as an industrial abrasive. Silicon carbide is produced for abrasive, metallurgical, and other non-abrasive applications 18 in the United States. Production for metallurgical and other non-abrasive applications is not available and therefore 19 both CO2 and CH4 estimates are based solely upon production estimates of silicon carbide for abrasive applications. 20 Emissions from fuels consumed for energy purposes during the production of silicon carbide are accounted for in the 21 Energy chapter. 22 Carbon dioxide and CH4 are also emitted during the production of calcium carbide, a chemical used to produce 23 acetylene. Carbon dioxide is implicitly accounted for in the storage factor calculation for the non-energy use of 24 petroleum coke in the Energy chapter. However, data are currently not available to estimate CH4 emissions from this 25 source. See Annex 5 for additional information on sources and sinks not included in this report. 26 To produce SiC, silica sand or quartz (SiCh) is reacted with C in the form of petroleum coke. A portion (about 35 27 percent) of the carbon contained in the petroleum coke is retained in the SiC. The remaining C is emitted as CO2, 28 CH4, or carbon monoxide (CO). The overall reaction is shown below (but in practice it does not proceed according 29 to stoichiometry): 30 Si02 + 3C -> SiC + 2CO (+ 02 -> 2C02) 31 Carbon dioxide is also emitted from the consumption of SiC for metallurgical and other non-abrasive applications. 32 Markets for manufactured abrasives, including SiC, are heavily influenced by activity in the U.S. manufacturing 33 sector, especially in the aerospace, automotive, furniture, housing, and steel manufacturing sectors. The U.S. 34 Geological Survey (USGS) reports that a portion (approximately 50 percent) of SiC is used in metallurgical and 35 other non-abrasive applications, primarily in iron and steel production (USGS 2006a). As a result of the economic 36 downturn in 2008 and 2009, demand for SiC decreased in those years. Low cost imports, particularly from China, 37 combined with high relative operating costs for domestic producers, continue to put downward pressure on the 38 production of SiC in the United States. However, demand for SiC consumption in the United States has recovered 39 somewhat from its low in 2009 (USGS 2012a). Abrasive-grade silicon carbide was manufactured at one facility in 40 2015 in the United States (USGS 2017a). 41 Carbon dioxide emissions from SiC production and consumption in 2016 were 0.2 MMT CO2 Eq. (174 kt CO2) (see 42 Table 4-35 and Table 4-36). Approximately 51 percent of these emissions resulted from SiC production while the Industrial Processes and Product Use 4-41 ------- 1 remainder resulted from SiC consumption. Methane emissions from SiC production in 2016 were 0.01 MMT CO2 2 Eq. (0.4 kt CH4) (see Table 4-35 and Table 4-36). Emissions have not fluctuated greatly in recent years, but 2016 3 emissions are about 52 percent lower than emissions in 1990. 4 Table 4-35: CO2 and ChU Emissions from Silicon Carbide Production and Consumption (MMT 5 COz Eq.) Year 1990 2005 2012 2013 2014 2015 2016 CO2 CH4 0.4 + 0.2 + 0.2 + 0.2 + 0.2 + 0.2 + 0.2 + Total 0.4 0.2 0.2 0.2 0.2 0.2 0.2 + Does not exceed 0.05 MMT CO2 Eq. 6 Table 4-36: CO2 and ChU Emissions from Silicon Carbide Production and Consumption (kt) Year 1990 2005 2012 2013 2014 2015 2016 CO2 CH4 375 1 219 + 158 + 169 + 173 + 180 + 174 + + Does not exceed 0.5 kt. 7 Methodology 8 Emissions of CO2 and CH4 from the production of SiC were calculated31 using the Tier 1 method provided by the 9 2006IPCC Guidelines. Annual estimates of SiC production were multiplied by the appropriate emission factor, as 10 shown below: 11 ESc,C02 = EFsc,co2 X Qsc /I metric ton\ 12 Esc'CH4 = EFsc'CHA X Qsc X { 1000 kg ) 13 where, 14 Esc,co2 15 EF sc,co2 16 QSc 17 ESCjch4 18 EFsc,ch4 19 20 Emission factors were taken from the 2006 IPCC Guidelines: 21 • 2.62 metric tons CCVmetric ton SiC 22 • 11.6 kg CH i/mctric ton SiC 23 Emissions of CO2 from silicon carbide consumption for metallurgical uses were calculated by multiplying the 24 annual utilization of SiC for metallurgical uses (reported annually in the USGS Minerals Yearbook: Silicon) by the 25 carbon content of SiC (31.5 percent), which was determined according to the molecular weight ratio of SiC. 26 Emissions of CO2 from silicon carbide consumption for other non-abrasive uses were calculated by multiplying the 27 annual SiC consumption for non-abrasive uses by the carbon content of SiC (31.5 percent). The annual SiC 28 consumption for non-abrasive uses was calculated by multiplying the annual SiC consumption (production plus net CO2 emissions from production of SiC, metric tons Emission factor for production of SiC, metric ton CCh/metric ton SiC Quantity of SiC produced, metric tons CH4 emissions from production of SiC, metric tons Emission factor for production of SiC, kilogram CH4/metric ton SiC 31 EPA has not integrated aggregated facility-level GHGRP information to inform these estimates. 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. 4-42 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2016 ------- 1 imports) by the percent used in metallurgical and other non-abrasive uses (50 percent) (USGS 2006a) and then 2 subtracting the SiC consumption for metallurgical use. 3 The petroleum coke portion of the total CO2 process emissions from silicon carbide production is adjusted for within 4 the Energy chapter, as these fuels were consumed during non-energy related activities. Additional information on 5 the adjustments made within the Energy sector for Non-Energy Use of Fuels is described in both the Methodology 6 section of CO2 from Fossil Fuel Combustion (3.1 Fossil Fuel Combustion (CRF Source Category 1 A)) and Annex 7 2.1, Methodology for Estimating Emissions of CO2 from Fossil Fuel Combustion. 8 Production data for 1990 through 2013 were obtained from the Minerals Yearbook: Manufactured Abrasives (USGS 9 1991a through 2015). Production data for 2014 and 2015 were obtained from the Minerals Industry Surveys: 10 Abrasives (Manufactured) (USGS 2016). Production data for 2016 were obtained from the Mineral Industry 11 Surveys: Abrasives (Manufactured) (USGS 2017b). Silicon carbide production data obtained through the USGS 12 National Minerals Information Center has been previously been rounded to the nearest 5,000 metric tons to avoid 13 disclosing company proprietary data. Silicon carbide consumption by major end use for 1990 through 2015 were 14 obtained from the Minerals Yearbook: Silicon (USGS 1991b through 2017c) (see Table 4-37). 2016 silicon carbide 15 consumption was not yet published by the USGS; therefore, 2015 data are used as a proxy for 2016. Net imports and 16 exports for the entire time series were obtained from the U.S. International Trade Commission (USITC) database 17 updated from data provided by the U.S. Census Bureau (2005 through 2017). 18 Table 4-37: Production and Consumption of Silicon Carbide (Metric Tons) Year Production Consumption 1000 105.000 172.465 2005 35.000 220.140 2012 35,000 114,265 2013 35,000 134,055 2014 35,000 140,733 2015 35,000 153,475 2016 35,000 142,104 19 Uncertainty and Time-Series Consistency 20 There is uncertainty associated with the emission factors used because they are based on stoichiometry as opposed to 21 monitoring of actual SiC production plants. An alternative would be to calculate emissions based on the quantity of 22 petroleum coke used during the production process rather than on the amount of silicon carbide produced. However, 23 these data were not available. For CH4, there is also uncertainty associated with the hydrogen-containing volatile 24 compounds in the petroleum coke (IPCC 2006). There is also uncertainty associated with the use or destruction of 25 methane generated from the process in addition to uncertainty associated with levels of production, net imports, 26 consumption levels, and the percent of total consumption that is attributed to metallurgical and other non-abrasive 27 uses. 28 The ivsiilis of ilic \pproach 2 (|iianiiiali\ e uiiceriaiiiis ; 111; 11\ sis are suniniaii/ed 111 I able 4-^X. Silicon carhide 29 produclioii and consunipiioii ('() emissions fmm 2d I(> were esiimaled lo he bclweeii l) peiceiii helow and peiceiii 30 alxn e llie emission esiiniale oft) I~ \1\1T( O I !t| al I lie l)5 peiceiii confidence le\ el. Silicon carhide produclion 31 CI I emissions were esiimaled lo he hclweeii l) perceni below and It) perceni aho\e 1 lie emission esiiniale oft) t)l 32 M\11 'CO I !c| al I lie l)5 perceni confidence le\ el Industrial Processes and Product Use 4-43 ------- 1 Table 4-38: Approach 2 Quantitative Uncertainty Estimates for ChU and CO2 Emissions from 2 Silicon Carbide Production and Consumption (MMT CO2 Eq. and Percent) - TO BE UPDATED 3 FOR FINAL INVENTORY REPORT Si HI I'l l' C;is 2016 I'lniissiiiii l.slim;ili' (MMT CO: i:<|.) I iHiTi.iiim R;mm'Ri-hiiiw in Kmissimi l.siim;iii:' (MM 1 C (): i:(|.) ("i.) Li HUT liiiund I ppi'i' Bmind I.I HUT 1 $111111(1 I ppi'i' liiilllHl Silicon Carbide Production and Consumption t ( ) 0.I7 0.15 0.19 -9% +9% Silicon Carbide Production CI 11 + + + -9% + 10% + I )oes not exceed 0.05 MMT C(): ] Cq. 11 Range of emission estimates predicted by Monte Carlo Stochastic Simulation for a 95 percent confidence interval. 4 Methodological approaches were applied to the entire time series to ensure consistency in emissions from 1990 5 through 2016. Details on the emission trends through time are described in more detail in the Methodology section, 6 above. 7 For more information on the general QA/QC process applied to this source category, consistent with Volume 1, 8 Chapter 6 of the 2006IPCC Guidelines, see QA/QC and Verification Procedures section in the introduction of the 9 IPPU Chapter. 10 Recalculations Discussion 11 In the previous Inventory report (i.e., 1990 through 2015), 2015 silicon carbide consumption data by end-use was 12 not available which resulted in the use of 2014 data as proxy. In the current Inventory, advance release data were 13 available for 2015 and the value was updated (USGS 2017c). This recalculation resulted in an insignificant change 14 to the total silicon carbide emissions estimate for the year 2015 compared to the previous Inventory report. 15 4.11 Titanium Dioxide Production (CRF Source is Category 2B6) 17 Titanium dioxide (TiCh) is manufactured using one of two processes: the chloride process and the sulfate process. 18 The chloride process uses petroleum coke and chlorine as raw materials and emits process-related carbon dioxide 19 (CO2). Emissions from fuels consumed for energy purposes during the production of titanium dioxide are accounted 20 for in the Energy chapter. The chloride process is based on the following chemical reactions: 21 2FeTi03 +7Cl2 +3C -> 2TiCl4 +2FeCl3 +3C02 22 2TiCl4 + 202 -> 2Ti02 +4Cl2 23 The sulfate process does not use petroleum coke or other forms of carbon as a raw material and does not emit CO2. 24 The C in the first chemical reaction is provided by petroleum coke, which is oxidized in the presence of the chlorine 25 and FeTiCb (rutile ore) to form CO2. Since 2004, all TiC>2 produced in the United States has been produced using the 26 chloride process, and a special grade of "calcined" petroleum coke is manufactured specifically for this purpose. 27 The principal use of TiC>2 is as a pigment in white paint, lacquers, and varnishes; it is also used as a pigment in the 28 manufacture of plastics, paper, and other products. In 2016, U.S. TiC>2 production totaled 1,200,000 metric tons 29 (USGS 2017). There were a total five plants producing TiC>2 in the United States in 2016. 4-44 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2016 ------- 1 Emissions of CO2 from titanium dioxide production in 2016 were estimated to be 1.6 MMT CO2 Eq. (1,608 kt CO2), 2 which represents an increase of 35 percent since 1990 (see Table 4-39). Compared to 2015, emissions from titanium 3 dioxide production decreased by 2 percent in 2016 due to a 2 percent decrease in production. 4 Table 4-39: CO2 Emissions from Titanium Dioxide (MMT CO2 Eq. and kt) Year MMT CO2 Eq. kt 1990 1.2 1,195 2005 1.8 1.755 2012 1.5 1,528 2013 1.7 1,715 2014 1.7 1,688 2015 1.6 1,635 2016 1.6 1,608 5 Methodology 6 Emissions of CO2 from TiCh production were calculated by multiplying annual national TiCh production by chloride 7 process-specific emission factors using a Tier 1 approach provided in 2006IPCC Guidelines. The Tier 1 equation is 8 as follows: 9 Etd = EFtd X Qtd 10 where, 11 Etd CO2 emissions from Ti02 production, metric tons 12 EFtd = Emission factor (chloride process), metric ton C02/metric ton Ti02 13 Qtd Quantity of Ti02 produced 14 The petroleum coke portion of the total CO2 process emissions from TiC>2 production is adjusted for within the 15 Energy chapter as these fuels were consumed during non-energy related activities. Additional information on the 16 adjustments made within the Energy sector for Non-Energy Use of Fuels is described in both the Methodology 17 section of CO2 from Fossil Fuel Combustion (3.1 Fossil Fuel Combustion (CRF Source Category 1A)) and Annex 18 2.1, Methodology for Estimating Emissions of CO2 from Fossil Fuel Combustion. 19 Data were obtained for the total amount of TiC>2 produced each year. For years prior to 2004, it was assumed that 20 TiC>2 was produced using the chloride process and the sulfate process in the same ratio as the ratio of the total U.S. 21 production capacity for each process. As of 2004, the last remaining sulfate process plant in the United States 22 closed; therefore, 100 percent of post-2004 production uses the chloride process (USGS 2005b). The percentage of 23 production from the chloride process is estimated at 100 percent since 2004. An emission factor of 1.34 metric tons 24 CCVmetric ton TiC>2 was applied to the estimated chloride-process production (IPCC 2006). It was assumed that all 25 TiC>2 produced using the chloride process was produced using petroleum coke, although some TiC>2 may have been 26 produced with graphite or other carbon inputs. 27 The emission factor for the TiC>2 chloride process was taken from the 2006 IPCC Guidelines. Titanium dioxide 28 production data and the percentage of total TiC>2 production capacity that is chloride process for 1990 through 2013 29 (see Table 4-40) were obtained through the Minerals Yearbook: Titanium Annual Report (USGS 1991 through 30 2015). Production data for 2014 through 2016 were obtained from the Minerals Commodity Summary: Titanium and 31 Titanium Dioxide (USGS 2016; USGS 2017).32 Data on the percentage of total TiC>2 production capacity that is 32 chloride process were not available for 1990 through 1993, so data from the 1994 USGS Minerals Yearbook were 32 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 4-45 ------- 1 used for these years. Because a sulfate process plant closed in September 2001, the chloride process percentage for 2 2001 was estimated based on a discussion with Joseph Gambogi (2002). By 2002, only one sulfate process plant 3 remained online in the United States and this plant closed in 2004 (USGS 2005b). 4 Table 4-40: Titanium Dioxide Production (kt) Year kt 1990 979 2005 1,310 2012 1,140 2013 1,280 2014 1,260 2015 1,220 2016 1,200 5 Uncertainty and Time-Series Consistency 6 Each year, the U.S. Geological Survey (USGS) collects titanium industry data for titanium mineral and pigment 7 production operations. If TiCh pigment plants do not respond, production from the operations is estimated on the 8 basis of prior year production levels and industry trends. Variability in response rates varies from 67 to 100 percent 9 of Ti02 pigment plants over the time series. 10 Although some TiCh may be produced using graphite or other carbon inputs, information and data regarding these 11 practices were not available. Titanium dioxide produced using graphite inputs, for example, may generate differing 12 amounts of CO2 per unit of TiCh produced as compared to that generated through the use of petroleum coke in 13 production. While the most accurate method to estimate emissions would be to base calculations on the amount of 14 reducing agent used in each process rather than on the amount of TiCh produced, sufficient data were not available 15 to do so. 16 As of 2004, the last remaining sulfate-process plant in the United States closed. Since annual TiCh production was 17 not reported by USGS by the type of production process used (chloride or sulfate) prior to 2004 and only the 18 percentage of total production capacity by process was reported, the percent of total TiCh production capacity that 19 was attributed to the chloride process was multiplied by total TiCh production to estimate the amount of TiCh 20 produced using the chloride process. Finally, the emission factor was applied uniformly to all chloride-process 21 production, and no data were available to account for differences in production efficiency among chloride-process 22 plants. In calculating the amount of petroleum coke consumed in chloride-process TiCh production, literature data 23 were used for petroleum coke composition. Certain grades of petroleum coke are manufactured specifically for use 24 in the TiCh chloride process; however, this composition information was not available. 25 The results oil lie \pproach 2 t|ii;inlil;ili\ e iiuccriniuis ;in;il\sis are suniniari/cd in Table 4-41 Tiinmum dioxide 26 consumption ( () emissions limn 2u I (> were csiinialed In he between I 4and I X \l\l'l CO Lq al llie lJ5 perccui 27 confidence le\el This indicates a rnime of approximate^ 12 percent below and I ' percent abo\e llie emission 28 esiimale of I <> \1\1T CO \ x\ 29 Table 4-41: Approach 2 Quantitative Uncertainty Estimates for CO2 Emissions from Titanium 30 Dioxide Production (MMT CO2 Eq. and Percent) - TO BE UPDATED FOR FINAL INVENTORY 31 REPORT Si HI I'l l' (¦as 2016 1"missiiin KslimaU- IMMK O: i:c|.) I iHi-iiaiim Kan^c Ri hilhi-1<> l.missiim (MM 1 ( (): l.t|.) ("¦ llsiinialr1 i>) I.I HUT I |)|KT 1.1 HUT lilllllld 1$111111(1 1 $111111(1 I ppi-r liiillllll Titanium Dioxide Prod uclion t ( ) 1.6 1.4 1.8 -12% + 13% Range of emission estimates predicted by Monte Carlo Stochastic Simulation lor a 95 percent confidence interval. 4-46 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2016 ------- 1 Methodological approaches were applied to the entire time series to ensure consistency in emissions from 1990 2 through 2016. Details on the emission trends through time are described in more detail in the Methodology section, 3 above. 4 For more information on the general QA/QC process applied to this source category, consistent with Volume 1, 5 Chapter 6 of the 2006IPCC Guidelines, see QA/QC and Verification Procedures section in the introduction of the 6 IPPU Chapter. 7 Planned Improvements 8 Planned improvements include researching the significance of titanium-slag production in electric furnaces and 9 synthetic-rutile production using the Becher process in the United States. Significant use of these production 10 processes will be included in future Inventory reports. Due to resource constraints, this planned improvement is still 11 in development by EPA and is not included in this report. EPA continues to assess the potential of integrating 12 aggregated facility-level GHGRP information for titanium dioxide production facilities based on criteria to shield 13 underlying CBI from public disclosure. Pending available resources, EPA will also evaluate use of GHGRP data to 14 improve category-specific QC consistent with both Volume 1, Chapter 6 of 2006 IPCC Guidelines and the latest 15 IPCC guidance on the use of facility-level data in national inventories.33 16 4.12 Soda Ash Production (CRF Source n Category 2B7) 18 Carbon dioxide (CO2) is generated as a byproduct of calcining trona ore to produce soda ash, and is eventually 19 emitted into the atmosphere. In addition, CO2 may also be released when soda ash is consumed. Emissions from 20 soda ash consumption in chemical production processes are reported under Section 4.4 Other Process Uses of 21 Carbonates (IPCC Category 2A4) and emissions from fuels consumed for energy purposes during the production 22 and consumption of soda ash are accounted for in the Energy sector. 23 Calcining involves placing crushed trona ore into a kiln to convert sodium bicarbonate into crude sodium carbonate 24 that will later be filtered into pure soda ash. The emission of CO2 during trona-based production is based on the 25 following reaction: 26 2Na2C03 • NaHC03 • 2H20(Trona) -» 3Na2C03(Soda Ash) + 5H20 + C02 27 Soda ash (sodium carbonate, Na2CC>3) is a white crystalline solid that is readily soluble in water and strongly 28 alkaline. Commercial soda ash is used as a raw material in a variety of industrial processes and in many familiar 29 consumer products such as glass, soap and detergents, paper, textiles, and food. Emissions from soda ash used in 30 glass production are reported under Section 4.3, Glass Production (CRF Source Category 2A3). Glass production is 31 its own source category and historical soda ash consumption figures have been adjusted to reflect this change. After 32 glass manufacturing, soda ash is used primarily to manufacture many sodium-based inorganic chemicals, including 33 sodium bicarbonate, sodium chromates, sodium phosphates, and sodium silicates (USGS 2015b). Internationally, 34 two types of soda ash are produced, natural and synthetic. The United States produces only natural soda ash and is 35 second only to China in total soda ash production. Trona is the principal ore from which natural soda ash is made. 36 The United States represents about one-fourth of total world soda ash output. Only two states produce natural soda 37 ash: Wyoming and California. Of these two states, only net emissions of CO2 from Wyoming were calculated due to 38 specifics regarding the production processes employed in the state.34 Based on 2016 reported data, the estimated 33 See . 34 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 CO2 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 Industrial Processes and Product Use 4-47 ------- 1 distribution of soda ash by end-use in 2016 (excluding glass production) was chemical production, 57 percent; soap 2 and detergent manufacturing, 12 percent; distributors, 11 percent; flue gas desulfurization, 8 percent; other uses, 7 3 percent; pulp and paper production, 3 percent, and water treatment, 2 percent (USGS 2017).35 4 U.S. natural soda ash is competitive in world markets because the majority of the world output of soda ash is made 5 synthetically. Although the United States continues to be a major supplier of world soda ash, China, which 6 surpassed the United States in soda ash production in 2003, is the world's leading producer. 7 In 2016, CO2 emissions from the production of soda ash from trona were approximately 1.7 MMT CO2 Eq. (1,723 kt 8 CO2) (see Table 4-42). Total emissions from soda ash production in 2016 increased by approximately 1 percent 9 from emissions in 2015, and have increased by approximately 20 percent from 1990 levels. 10 Emissions have remained relatively constant over the time series with some fluctuations since 1990. In general, 11 these fluctuations were related to the behavior of the export market and the U.S. economy. The U.S. soda ash 12 industry continued a trend of increased production and value in 2016 since experiencing a decline in domestic and 13 export sales caused by adverse global economic conditions in 2009. 14 Table 4-42: CO2 Emissions from Soda Ash Production (MMT CO2 Eq. and kt CO2) Year MMT CO2 Eq. kt CO2 1990 1.4 1,431 2005 1.7 1.655 2012 1.7 1,665 2013 1.7 1,694 2014 1.7 1,685 2015 1.7 1,714 2016 1.7 1,723 Note: Totals may not sum due to independent rounding. 15 Methodology 16 During the production process, trona ore is calcined in a rotary kiln and chemically transformed into a crude soda 17 ash that requires further processing. Carbon dioxide and water are generated as byproducts of the calcination 18 process. Carbon dioxide emissions from the calcination of trona can be estimated based on the chemical reaction 19 shown above. Based on this formula, which is consistent with an IPCC Tier 1 approach, approximately 10.27 metric 20 tons of trona are required to generate one metric ton of CO2, or an emission factor of 0.0974 metric tons CO2 per 21 metric ton trona (IPCC 2006). Thus, the 17.7 million metric tons of trona mined in 2016 for soda ash production 22 (USGS 2017) resulted in CO2 emissions of approximately 1.7 MMT CO2 Eq. (1,723 kt). 23 Once produced, most soda ash is consumed in chemical production, with minor amounts in soap production, pulp 24 and paper, flue gas desulfurization, and water treatment (excluding soda ash consumption for glass manufacturing). 25 As soda ash is consumed for these purposes, additional CO2 is usually emitted. Consistent with the 2006 IPCC 26 Guidelines for National Greenhouse Gas Inventories, emissions from soda ash consumption in chemical production 27 processes are reported under Section 4.4 Other Process Uses of Carbonates (IPCC Category 2A4). sodium carbonate. Although CO2 is generated as a byproduct, the CO2 is recovered and recycled for use in the carbonation stage and is not emitted. A third state, Colorado, produced soda ash until the plant was idled in 2004. The lone producer of sodium bicarbonate no longer mines trona in the state. For a brief time, sodium bicarbonate was produced using soda ash feedstocks mined in Wyoming and shipped to Colorado. Prior to 2004, because the trona was mined in Wyoming, the production numbers given by the USGS included the feedstocks mined in Wyoming and shipped to Colorado. In this way, the sodium bicarbonate production that took place in Colorado was accounted for in the Wyoming numbers. 35 Percentages may not add up to 100 percent due to independent rounding. 4-48 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2016 ------- 1 The activity data for trona production (see Table 4-43) for 1990 to 2016 were obtained from the U.S. Geological 2 Survey (USGS) Minerals Yearbook for Soda Ash (1994 through 2015b) and USGS Mineral Industry Surveys for 3 Soda Ash (USGS 2017). Soda ash production36 data were collected by the USGS from voluntary surveys of the U.S. 4 soda ash industry. EPA will continue to analyze and assess opportunities to use facility-level data from EPA's 5 GHGRP to improve the emission estimates for Soda Ash Production source category consistent with IPCC37 and 6 UNFCCC guidelines. 7 Table 4-43: Soda Ash Production (kt) Year Production3 1990 14,700 2005 17.000 2012 17,100 2013 17,400 2014 17,300 2015 17,600 2016 17,700 a Soda ash produced from trona ore only. 8 Uncertainty and Time-Series Consistency 9 Emission estimates from soda ash production have relatively low associated uncertainty levels in that reliable and 10 accurate data sources are available for the emission factor and activity data for trona-based soda ash production. 11 EPA plans to work with other entities to reassess the uncertainty of these emission factors and activity data based on 12 the most recent information and data. Through EPA's GHGRP, EPA is aware of one facility producing soda ash 13 from a liquid alkaline feedstock process. Soda ash production data was collected by the USGS from voluntary 14 surveys. A survey request was sent to each of the five soda ash producers, all of which responded, representing 100 15 percent of the total production data (USGS 2016). One source of uncertainty is the purity of the trona ore used for 16 manufacturing soda ash. The emission factor used for this estimate assumes the ore is 100 percent pure, and likely 17 overestimates the emissions from soda ash manufacture. The average water-soluble sodium carbonate-bicarbonate 18 content for ore mined in Wyoming ranges from 85.5 to 93.8 percent (USGS 1995). 19 The results nl" I lie \pproaeli 2 qiiaiitiiali\ e iiiiccriaiiits ;iii;iI\ sis are siininiari/ed in Table 4-44 Soda Ash hodiiclion 20 CO emissions I'm- 2<) I (> were estimated In he between I (> and I.XMMTCO liq alllie l>5 pereenl coiilideiice le\ el 21 This indicates a ranue ol appi'o\inialel> ~ pereenl below and <> percent aho\e the emission estimate of I ~ MMT 22 CO l!q 23 Table 4-44: Approach 2 Quantitative Uncertainty Estimates for CO2 Emissions from Soda Ash 24 Production (MMT CO2 Eq. and Percent) - TO BE UPDATED FOR FINAL INVENTORY REPORT Sou I'll' (¦as 2016 rimissiiiii l.sliniiili' I MMT CO: l!i|.) I niirlaiim kan^i'ki'laliu'In llniissiiui 1'siimale-1 (MM 1 ( (): i:t|.) C'i.) I.I HUT I |1|KT I.IHUT I |1|KT Bound Bound Bound Bound Soda Ash Production CO: 1.7 1.6 1.8 36 EPA has assessed feasibility of using emissions information (including activity data) from EPA's GHGRP program; however, 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. 37 See . Industrial Processes and Product Use 4-49 ------- 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 ;i Range of emission estimates predicted by Monte Curio Stochastic Simulation for a 95 percent confidence interval. Methodological approaches were applied to the entire time series to ensure consistency in emissions from 1990 through 2016. 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 QA/QC and Verification Procedures section in the introduction of the IPPU Chapter. Recalculations Discussion Emissions from soda ash consumption in chemical production processes were removed from this section and reported under Section 4.4 Other Process Uses of Carbonates (IPCC Category 2A4) to be consistent with 2006 IPCC Guidelines. This revision resulted in a decrease in emissions associated with the soda ash production source category ranging from approximately 39 percent to 49 percent across the time series of 1990 through 2015 compared to the previous Inventory report. Planned Improvements EPA plans to use GHGRP data for conducting category-specific QC of emission estimates consistent with both Volume 1, Chapter 6 of 2006 IPCC Guidelines and the latest IPCC guidance on the use of facility-level data in national inventories.38 This planned improvement is ongoing and has not been incorporated into this Inventory report. 4.13 Petrochemical Production (CRF Source Category 2B8) The production of some petrochemicals results in the release of small amounts of carbon dioxide (CO2) and methane (CH4) emissions. Petrochemicals are chemicals isolated or derived from petroleum or natural gas. Carbon dioxide emissions from the production of acrylonitrile, carbon black, ethylene, ethylene dichloride, ethylene oxide, and methanol, and CH4 emissions from the production of methanol and acrylonitrile are presented here and reported under CRF Source Category 2B8. 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. 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 in the Energy sector. The allocation and reporting of emissions from feedstocks transferred out of the system for use in energy purposes to the Energy Chapter is consistent with 2006 IPCC Guidelines. 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 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 also produces byproduct CO2, carbon monoxide (CO), and water 38 See . 4-50 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2016 ------- 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 from the direct oxidation of the propylene feedstock, and produces other hydrocarbons from side reactions in the ammoxidation process. 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 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 emissions are released from thermal incinerators used as control devices, process dryers, and equipment leaks. Carbon black is also produced in the United States by the thermal cracking of acetylene-containing feedstocks (i.e., acetylene black process), by the thermal cracking of other hydrocarbons (i.e., thermal black process), and by the open burning of carbon black feedstock (i.e., lamp black process); each of these processes is used at only one U.S. plant (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: C2H6 -» C2H4 + H2 Small amounts of CH4 are also generated from the steam cracking process. In addition, CO2 and CH4 emissions are also generated from combustion units. Ethylene dichloride (C2H4CI2) is used to produce vinyl chloride monomer, which is the precursor to polyvinyl chloride (PVC). Ethylene dichloride was 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 + 3 02 —> 2C02 + 2 H20 (direct oxidation of ethylene during oxychlorination) In addition to the byproduct CO2 produced from the direction oxidation of the ethylene feedstock, CO2 and CH4 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, etc.) 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. Industrial Processes and Product Use 4-51 ------- 1 Emissions of CO2 and CH4 from petrochemical production in 2016 were 27.4 MMT CChEq. (27,411 kt CO2) and 2 0.2 MMT CO2 Eq. (7 kt CH4), respectively (see Table 4-45 and Table 4-46). Since 1990, total CO2 emissions from 3 petrochemical production increased by 29 percent. Methane emissions from petrochemical (methanol and 4 acrylonitrile) production have decreased by approximately 18 percent since 1990, given declining production; 5 however, CH4 emissions have been increasing since 2011 due to a rebound in methanol production. 6 Table 4-45: CO2 and ChU Emissions from Petrochemical Production (MMT CO2 Eq.) Year 1990 2005 2012 2013 2014 2015 2016 CO2 21.2 26.8 26.5 26.4 26.5 28.1 27.4 CH4 0.2 0.1 0.1 0.1 0.1 0.2 0.2 Total 21.4 26.9 26.6 26.5 26.6 28.2 27.6 + Does not exceed 0.05 MMT CO2 Eq. Note: Totals may not sum due to independent rounding. 7 Table 4-46: CO2 and ChU Emissions from Petrochemical Production (kt) Year 1990 2005 2012 2013 2014 2015 2016 CO2 CH4 21,203 9 26,794 3 26,501 3 26,395 3 26,496 5 28,062 7 27,411 7 8 Methodology 9 Emissions of CO2 and CH4 were calculated using the estimation methods provided by the 2006IPCC Guidelines 10 and country-specific methods from EPA's GHGRP. The 2006 IPCC Guidelines Tier 1 method was used to estimate 11 CO2 and CH4 emissions from production of acrylonitrile and methanol,39 and a country-specific approach similar to 12 the IPCC Tier 2 method was used to estimate CO2 emissions from carbon black, ethylene, ethylene oxide, and 13 ethylene dichloride. The Tier 2 method for petrochemicals is a total feedstock C mass balance method used to 14 estimate total CO2 emissions, but is not applicable for estimating CH4 emissions. As noted in the 2006 IPCC 15 Guidelines, the total feedstock C mass balance method (Tier 2) is based on the assumption that all of the C input to 16 the process is converted either into primary and secondary products or into CO2. Further, the guideline states that 17 while the total C mass balance method estimates total C emissions from the process but does not directly provide an 18 estimate of the amount of the total C emissions emitted as CO2, CH4, or non-CH4 volatile organic compounds 19 (NMVOCs). This method accounts for all the C as CO2, including CH4. Note, a subset of facilities reporting under 20 EPA's GHGRP use alternate methods to the C balance approach (e.g., Continuous Emission Monitoring Systems 21 (CEMS) or other engineering approaches) to monitor CO2 emissions and these facilities are required to also report 22 CH4 and N20 emissions from combustion of process off-gas. Preliminary analysis of aggregated annual reports 23 shows that these emissions are less than 500 kt/year. EPA's GHGRP is still reviewing this data across reported years 24 to facilitate update of category-specific QC documentation and EPA plans to address this more completely in future 25 reports. 26 Carbon Black, Ethylene, Ethylene Dichloride and Ethylene Oxide 27 2010 through 2016 28 Carbon dioxide emissions and national production were aggregated directly from EPA's GHGRP dataset for 2010 29 through 2016 (EPA 2017). In 2016, GHGRP data reported CO2 emissions of 3,160,000 metric tons from carbon 30 black production; 19,600,000 metric tons of CChfrom ethylene production; 447,000 metric tons of CChfrom 31 ethylene dichloride production; and 1,100,000 metric tons of CO2 from ethylene oxide production. These emissions 39 EPA has not integrated aggregated facility-level GHGRP information for acrylonitrile and methanol production. The aggregated information associated with production of these petrochemicals did not meet criteria to shield underlying CBI from public disclosure. 4-52 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2016 ------- 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 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, and ethylene oxide. Since 2010, EPA's GHGRP, under Subpart X, requires all domestic producers of petrochemicals to report annual emissions and supplemental emissions information (e.g., production data, etc.) to facilitate verification of reported emissions. Under EPA's GHGRP, 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 facilities40 and assumes that all the carbon input is converted into primary and secondary products, byproducts, 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. 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).41 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.42 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 CO2 from Fossil Fuel Combustion (3.1 Fossil Fuel Combustion (CRF Source Category 1 A)) and Annex 2.1, Methodology for Estimating Emissions of CO2 from Fossil Fuel Combustion. 1990 through 2009 Prior to 2010, for each of these 4 types of petrochemical processes, an average national CO2 emission factor was calculated based on the 2010 through 2016 GHGRP data and applied to production for earlier years in the time series (i.e., 1990 through 2009) to estimate CO2 emissions from carbon black, ethylene, ethylene dichloride, and ethylene oxide production. Carbon dioxide emission factors were derived from EPA's GHGRP data by dividing annual CO2 emissions for petrochemical type "i" with annual production for petrochemical type "i" and then averaging the derived emission factors obtained for each calendar year 2010 through 2016. 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 2010 through 2016 GHGRP data are as follows: • 2.62 metric tons CCVmetric ton carbon black produced • 0.77 metric tons CCVmetric ton ethylene produced • 0.040 metric tons CCVmetric ton ethylene dichloride produced 40 A few facilities producing ethylene dichloride used CO2 CEMS, those CO2 emissions have been included in the aggregated GHGRP emissions presented here. For ethylene production processes, nearly all process emissions are from the combustion of process off-gas. Under EPA's GHGRP, Subpart X, ethylene facilities can report CO2 emissions from burning of process gases using the optional combustion methodology for ethylene production processes, which is requires estimating emissions based on fuel quantity and carbon contents of the fuel. This is consistent with the 2006 IPCC Guidelines (p. 3.57) which recommends including combustion emissions from fuels obtained from feedstocks (e.g., off-gases) in petrochemical production under in the IPPU sector. In 2014, for example, this methodology was used by more than 20 of the 65 reporting facilities. In addition to CO2, these facilities are required to report emissions of CH4 and N2O from combustion of ethylene process off-gas in 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, as applicable. Preliminary analysis of the aggregated reported CH4 and N2O emissions from facilities using CEMS and the optional combustion methodology suggests that these annual emissions are less than 500 kt/yr so not significant enough to prioritize for inclusion in the report at this time. Pending resources and significance, EPA may include these emissions in future reports to enhance completeness. 41 See . 42 See . Industrial Processes and Product Use 4-53 ------- 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 • 0.43 metric tons COVmctric 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 and ethylene dichloride for 1990 through 2009 were obtained from the American Chemistry Council's (ACC's) Guide to the Business of Chemistry (ACC 2002, 2003, 2005 through 2011). Annual production data for ethylene oxide were obtained from ACC's U.S. Chemical Industry Statistical Handbook for 2003 through 2009 (ACC 2014a) and from ACC's Business of Chemistry for 1990 through 2002 (ACC 2014b). As noted above, annual 2010 through 2016 production data for carbon black, ethylene, ethylene dichloride, and ethylene oxide, were obtained fromEPA's GHGRP. Acrylonitrile Carbon dioxide and methane emissions from acrylonitrile production were estimated using the Tier 1 method in the 2006IPCC Guidelines. Annual acrylonitrile production data were used with IPCC default Tier 1 CO2 and CH4 emission factors to estimate emissions for 1990 through 2016. Emission factors used to estimate acrylonitrile production emissions are as follows: • 0.18 kg CH4/metric ton acrylonitrile produced • 1.00 metric tons COVmctric ton acrylonitrile produced Annual acrylonitrile production data for 1990 through 2015 were obtained from ACC's Business of Chemistry (ACC 2016). In this current report, 2016 ACC production data was not yet available and 2015 data was used as proxy. Methanol Carbon dioxide and methane emissions from methanol production were estimated using Tier 1 method in the 2006 IPCC Guidelines. Annual methanol production data were used with IPCC default Tier 1 CO2 and CH4 emission factors to estimate emissions for 1990 through 2016. Emission factors used to estimate methanol production emissions are as follows: • 2.3 kg CH4/metric ton methanol produced • 0.67 metric tons COVmctric ton methanol produced Annual methanol production data for 1990 through 2015 were obtained from the ACC's Business of Chemistry (ACC 2016). As mentioned previously, 2016 ACC production data was not yet available and 2015 data was used as proxy. Table 4-47: Production of Selected Petrochemicals (kt) Chemical 1990 2005 2012 2013 2014 2015 2016 Carbon Black 1,307 1.651 1,280 1,230 1,210 1,220 1,190 Ethylene 16,542 23.975 24,800 25,300 25,500 26,900 26,600 Ethylene Dichloride 6,283 11.260 11,300 11,500 11,300 11,300 11,700 Ethylene Oxide 2,429 3.220 3,110 3,150 3,140 3,240 3,210 Acrylonitrile 1,214 1.325 1,220 1,075 1,095 1,050 1,050 Methanol 3,750 1,225 995 1,235 2,105 3,065 3,065 As noted earlier in the introduction section of the Petrochemical Production chapter, 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 reporting year 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 4-54 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2016 ------- 1 (81 FR 89260).43 Unfortunately, both of these data elements have not passed GHGRP's CBI aggregation criteria and 2 are thus not available for use in the Inventory at this time. As a result, the United States is currently unable to report 3 non-energy fuel use from petrochemical production under the IPPU chapter. Therefore, consistent with 2006IPCC 4 Guidelines, fuel consumption data reported by EIA are modified to account for these overlaps to avoid double- 5 counting. More information on the non-energy use of fossil fuel feedstocks for petrochemical production can be 6 found in Annex 2.3. 7 Uncertainty and Time-Series Consistency 8 The CH4 and CO2 emission factors used for acrylonitrile and methanol production are based on a limited number of 9 studies. Using plant-specific factors instead of default or average factors could increase the accuracy of the emission 10 estimates; however, such data were not available for the current Inventory report. 11 The results of the quantitative uncertainty analysis for the CO2 emissions from carbon black production, ethylene, 12 ethylene dichloride, and ethylene oxide are based on reported GHGRP data. Refer to the Methodology section for 13 more details on how these emissions were calculated and reported to EPA's GHGRP. There is some uncertainty in 14 the applicability of the average emission factors for each petrochemical type across all prior years. While 15 petrochemical production processes in the United States have not changed significantly since 1990, some 16 operational efficiencies have been implemented at facilities over the time series. 17 The results nl" llie \pproach 2 quaulilali\e iiuccriaiulv aiials sis ;iiv summarized in Table 4-4S. IVtrochcniical 18 production ('() emissions I'roni 2<) l<> were esiimuled in he between 2<> <> mid 2X S \ 1 \1T (() l\q. ill ilie l>5 percent 19 confidence le\el This indicates a rauue ol appro\inialcl> 5 percent below lo 5 pereeui aho\c llie emission esiiniale 20 of 2" 4 \1\1T ('() I x|. IVirochcniical production CM emissions IV0111 2t> l<> were est 1 mated lo he between t> <><1 and 21 11.22 \I\IT ('() I a| at llie l>5 pereeui confidence le\ el This indicates a rauuc ol appio\inialel> 5" percent helow to 22 4(> pereeui aho\e the emission esiiniale ol'u 2 \l\1TCO l!q. 23 Table 4-48: Approach 2 Quantitative Uncertainty Estimates for ChU Emissions from 24 Petrochemical Production and CO2 Emissions from Carbon Black Production (MMT CO2 Eq. 25 and Percent) - TO BE UPDATED FOR FINAL INVENTORY REPORT Si ill I'l l' (¦;is 2016 liiiiissiiin r. si i m:i u- (MM"I" CO: i:<|. 1 I iiiirl;iiiil\ Ki-hiliu- In 1". miss inn (MM 1 ( (): i:t|.l r.slilll;lli'' I.I HUT I |1|KT I.IHUT lillllllll Bound 1$111111(1 l |1|KT lillllllll Petrochemical Production Petrochemical Production t ( 1 CI Ii 27.4 0.2 26.0 28.8 -5% 0 Of) 0 22 -^7. Industrial Processes and Product Use 4-55 ------- 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 this category included the QA/QC requirements and verification procedures of EPA's GHGRP. Future QC efforts to validate the use of Tier 1 default EFs and report on the comparison of Tier 1 emissions estimates and GHGRP data are described below in the Planned Improvements section. Planned Improvements Improvements include completing category-specific QC of activity data and EFs, along with further assessment of CH4 and N20 emissions to enhance completeness in reporting of emissions from U.S. petrochemical production, pending resources, significance and time series consistency considerations. As of this current report, timing and resources have not allowed EPA to complete this analysis of activity data and EFs and remains a priority improvement 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. Some degree of double counting may occur between CO2 estimates of non-energy use of fuels in the energy sector and CO2 process emissions from petrochemical production in this sector. As noted previously in the methodology section, data integration is not feasible at this time as 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. As described in the methodology section of this source category, EPA is currently unable to use GHGRP reported data on quantities of fuel consumed as feedstocks by petrochemical producers, only feedstock type, due to the data failing GHGRP CBI aggregation criteria. Incorporating this data into future inventories 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 (CRF Source Category 2B9a) Trifluoromethane (HFC-23 or CHF3) is generated as a byproduct during the manufacture of chlorodifluoromethane (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 is scheduled to be phased out by 2020 under the U.S. Clean Air Act.44 Feedstock production, however, is permitted to continue indefinitely. HCFC-22 is produced by the reaction of chloroform (CHCI3) 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 (CHCI2F), HCFC-22 (CHCIF2), HFC-23 (CHF3), HC1, 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, HC1 and residual HF. The HC1 is recovered as a useful 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. 44 As construed, interpreted, and applied in the terms and conditions of the Montreal Protocol on Substances that Deplete the Ozone Layer. [42 U.S.C. §7671m(b), CAA §614] 4-56 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2016 ------- 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 Two facilities produced HCFC-22 in the United States in 2016. Emissions of HFC-23 from this activity in 2016 were estimated to be 2.8 MMT CO2 Eq. (0.19 kt) (see Table 4-49). This quantity represents a 34 percent decrease from 2015 emissions and a 94 percent decrease from 1990 emissions. The decrease from 2015 emissions and the decrease from 1990 emissions were caused primarily by changes in the HFC-23 emission rate (kg HFC-23 emitted/kg 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 factor was higher than that of the other two plants, ceased production of HCFC-22 in 2013. Table 4-49: HFC-23 Emissions from HCFC-22 Production (MMT COz Eq. and kt HFC-23) Year MMT CChEq. kt HFC-23 1990 46.1 3 2005 20.0 1 2012 5.5 0.4 2013 4.1 0.3 2014 5.0 0.3 2015 4.3 0.3 2016 2.8 0.2 Methodology 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. Emissions for 2010 through 2016 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 to estimate their emissions of HFC-23. Plants using thermal oxidation to abate their HFC- 23 emissions monitor the performance of their oxidizers to verily that the HFC-23 is almost completely destroyed. Plants that release (or historically have released) some of their byproduct HFC-23 periodically measure HFC-23 concentrations in the output stream using gas chromatography. This information is combined with information on quantities of products (e.g., HCFC-22) to estimate HFC-23 emissions. 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 2016 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 (RTI 1997; RTI2008). The 1997 and 2008 reviews enabled U.S. totals to be reviewed, updated, and where necessary, corrected, and also for 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 4-50. Industrial Processes and Product Use 4-57 ------- 1 Table 4-50: HCFC-22 Production (kt) Year kt 1990 139 2005 156 2012 96 2013 C 2014 C 2015 C 2016 C C (CBI) Note: HCFC-22 production in 2012 through 2016 is considered Confidential Business Information (CBI) as there were only two producers of HCFC-22 in those years. 2 Uncertainty and Time-Series Consistency 3 The uncertainty analysis presented in this section was based on a plant-level Monte Carlo Stochastic Simulation for 4 2006. The Monte Carlo analysis used estimates of the uncertainties in the individual variables in each plant's 5 estimating procedure. This analysis was based on the generation of 10,000 random samples of model inputs from the 6 probability density functions for each input. A normal probability density function was assumed for all 7 measurements and biases except the equipment leak estimates for one plant; a log-normal probability density 8 function was used for this plant's equipment leak estimates. The simulation for 2006 yielded a 95-percent 9 confidence interval for U.S. emissions of 6.8 percent below to 9.6 percent above the reported total. 10 The relative errors yielded by the Monte Carlo Stochastic Simulation for 2006 were applied to the U.S. emission 11 estimate for 2016. The resulting estimates of absolute uncertainty are likely to be reasonably accurate because (1) 12 the methods used by the two remaining plants to estimate their emissions are not believed to have changed 13 significantly since 2006, and (2) although the distribution of emissions among the plants has changed between 2006 14 and 2016 (because one plant has closed), the plant that currently accounts for most emissions had a relative 15 uncertainty in its 2006 (as well as 2005) emissions estimate that was similar to the relative uncertainty for total U.S. 16 emissions. Thus, the closure of one plant is not likely to have a large impact on the uncertainty of the national 17 emission estimate. 18 The results of the Approach 2 quantitative uncertainty analysis are summarized in Table 4-51. HFC-23 emissions 19 from HCFC-22 production were estimated to be between 2.6 and 3.1 MMT CO2 Eq. at the 95 percent confidence 20 level. This indicates a range of approximately 7 percent below and 10 percent above the emission estimate of 2.8 21 MMT C02 Eq. 22 Table 4-51: Approach 2 Quantitative Uncertainty Estimates for HFC-23 Emissions from 23 HCFC-22 Production (MMT CO2 Eq. and Percent) Source _ 2016 Emission Estimate (MMT CO2 Eq.) Uncertainty Range Relative to Emission Estimate3 (MMT CO2 Eq.) (%) Lower Upper Bound Bound Lower Upper Bound Bound HCFC-22 Production HFC-23 2.8 2.6 3.1 -7% +10% a Range of emissions reflects a 95 percent confidence interval. 24 QA/QC and Verification 25 Tier 1 and Tier 2 QA/QC activities were conducted consistent with the U.S. QA/QC plan. Source-specific quality 26 control measures for the HCFC-22 Production category included the QA/QC requirements and verification 4-58 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2016 ------- 1 procedures of EPA's GHGRP. Under EPA's GHGRP, HCFC-22 producers are required to (1) measure 2 concentrations of HFC-23 and HCFC-22 in the product stream at least weekly using equipment and methods (e.g., 3 gas chromatography) with an accuracy and precision of 5 percent or better at the concentrations of the process 4 samples, (2) measure mass flows of HFC-23 and HCFC-22 at least weekly using measurement devices (e.g., 5 flowmeters) with an accuracy and precision of 1 percent of full scale or better, (3) calibrate mass measurement 6 devices at the frequency recommended by the manufacturer using traceable standards and suitable methods 7 published by a consensus standards organization, (4) calibrate gas chromatographs at least monthly through analysis 8 of certified standards, and (5) document these calibrations. 9 EPA verifies annual facility-level reports from HCFC-22 producers through a multi-step process (e.g., a 10 combination of electronic checks and manual reviews by staff) to identify potential errors and ensure that data 11 submitted to EPA are accurate, complete, and consistent. Based on the results of the verification process, EPA 12 follows up with facilities to resolve mistakes that may have occurred.45 n 4.15 Carbon Dioxide Consumption (CRF Source 14 Category 2610} 15 Carbon dioxide (CO2) is used for a variety of commercial applications, including food processing, chemical 16 production, carbonated beverage production, and refrigeration, and is also used in petroleum production for 17 enhanced oil recovery (EOR). Carbon dioxide used for EOR is injected underground to enable additional petroleum 18 to be produced. For the purposes of this analysis, CO2 used in commercial applications other than EOR is assumed 19 to be emitted to the atmosphere. Carbon dioxide used in EOR applications is discussed in the Energy chapter under 20 "Carbon Capture and Storage, including Enhanced Oil Recovery" and is not discussed in this section. 21 Carbon dioxide is produced from naturally-occurring CO2 reservoirs, as a byproduct from the energy and industrial 22 production processes (e.g., ammonia production, fossil fuel combustion, ethanol production), and as a byproduct 23 from the production of crude oil and natural gas, which contain naturally occurring CO2 as a component. Only CO2 24 produced from naturally occurring CO2 reservoirs, and as a byproduct from energy and industrial processes, and 25 used in industrial applications other than EOR is included in this analysis. Carbon dioxide captured from biogenic 26 sources (e.g., ethanol production plants) is not included in the Inventory. Carbon dioxide captured from crude oil 27 and gas production is used in EOR applications and is therefore reported in the Energy chapter. 28 Carbon dioxide is produced as a byproduct of crude oil and natural gas production. This CO2 is separated from the 29 crude oil and natural gas using gas processing equipment, and may be emitted directly to the atmosphere, or 30 captured and reinjected into underground formations, used for EOR, or sold for other commercial uses. A further 31 discussion of CO2 used in EOR is described in the Energy chapter in Box 3-7 titled "Carbon Dioxide Transport, 32 Injection, and Geological Storage." 33 In 2016, the amount of CO2 produced and captured for commercial applications and subsequently emitted to the 34 atmosphere was 4.5 MMT C02Eq. (4,471 kt) (see Table 4-52). This is consistent with 2014 and 2015 levels and is 35 an increase of approximately 204 percent since 1990. 36 Table 4-52: CO2 Emissions from CO2 Consumption (MMT CO2 Eq. and kt) Year MMT CO2 Eq. kt 1990 1.5 1,472 2005 1.4 1.37 5 2012 2013 4.0 4.2 4,019 4,188 45 See . Industrial Processes and Product Use 4-59 ------- Year MMT CO2 Eg. kt 2014 2015 2016 4.5 4.5 4.5 4,471 4,471 4,471 1 Methodology 2 Carbon dioxide emission estimates for 1990 through 2016 were based on the quantity of CO2 extracted and 3 transferred for industrial applications (i.e., non-EOR end-uses). Some of the CO2 produced by these facilities is used 4 for EOR and some is used in other commercial applications (e.g., chemical manufacturing, food production). It is 5 assumed that 100 percent of the CO2 production used in commercial applications other than EOR is eventually 6 released into the atmosphere. 8 For 2010 through 2014, data from EPA's GHGRP (Subpart PP) were aggregated from facility-level reports to 9 develop a national-level estimate for use in the Inventory (EPA 2016). Facilities report CO2 extracted or produced 10 from natural reservoirs and industrial sites, and CO2 captured from energy and industrial processes and transferred to 11 various end-use applications to EPA's GHGRP. This analysis includes only reported CO2 transferred to food and 12 beverage end-uses. EPA is continuing to analyze and assess integration of CO2 transferred to other end-uses to 13 enhance the completeness of estimates under this source category. Other end-uses include industrial applications, 14 such as metal fabrication. EPA is analyzing the information reported to ensure that other end-use data excludes non- 15 emissive applications and publication will not reveal confidential business information (CBI). Reporters subject to 16 EPA's GHGRP Subpart PP are also required to report the quantity of CO2 that is imported and/or exported. 17 Currently, these data are not publicly available through the GHGRP due to data confidentiality reasons and hence 18 are excluded from this analysis. 19 Facilities subject to Subpart PP of EPA's GHGRP are required to measure CO2 extracted or produced. More details 20 on the calculation and monitoring methods applicable to extraction and production facilities can be found under 21 Subpart PP: Suppliers of Carbon Dioxide of the regulation, Part 98.46 The number of facilities that reported data to 22 EPA's GHGRP Subpart PP (Suppliers of Carbon Dioxide) for 2010 through 2014 is much higher (ranging from 44 23 to 48) than the number of facilities included in the Inventory for the 1990 to 2009 time period prior to the 24 availability of GHGRP data (4 facilities). The difference is largely due to the fact the 1990 to 2009 data includes 25 only CO2 transferred to end-use applications from naturally occurring CO2 reservoirs and excludes industrial sites. 26 Starting in 2015, data from EPA's GHGRP (Subpart PP) was unavailable for use in the current Inventory report due 27 to data confidentiality reasons. As a result, the emissions estimates for 2015 and 2016 have been held constant from 28 2014 levels to avoid disclosure of proprietary information. EPA will continue to evaluate options for utilizing 29 GHGRP data to update these values in future inventories. 31 For 1990 through 2009, data from EPA's GHGRP are not available. For this time period, CO2 production data from 32 four naturally-occurring CO2 reservoirs were used to estimate annual CO2 emissions. These facilities were Jackson 33 Dome in Mississippi, Brave and West Bravo Domes in New Mexico, and McCallum Dome in Colorado. The 34 facilities in Mississippi and New Mexico produced CO2 for use in both EOR and in other commercial applications 35 (e.g., chemical manufacturing, food production). The fourth facility in Colorado (McCallum Dome) produced CO2 36 for commercial applications only (New Mexico Bureau of Geology and Mineral Resources 2006). 37 Carbon dioxide production data and the percentage of production that was used for non-EOR applications for the 38 Jackson Dome, Mississippi facility were obtained from Advanced Resources International (ARI 2006, 2007) for 46 See . 7 2010 through 2016 30 1990 through 2009 4-60 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2016 ------- 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 1990 to 2000, and from the Annual Reports of Denbury Resources (Denbury Resources 2002 through 2010) for 2001 to 2009 (see Table 4-53). 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 4-53: CO2 Production (kt CO2) and the Percent Used for Non-EOR Applications Year Jackson Dome, Bravo Dome, West Bravo McCallum Total CO2 % MS NM Dome, NM CO2 Dome, CO Production Non- CO2 Production CO2 Production Production CO2 Production from Extraction EOR3 (kt) (% Non- (kt) (% Non- (kt) (% Non- (kt) (% Non- and Capture EOR) EOR) EOR) EOR) Facilities (kt) 1990 1,344(100%) 63 (1%) + 65 (100%) NA NA 2005 1.254 (27%) 58(1%) 63 (100%) NA NA 2012 NA NA NA NA 66,326 6% 2013 NA NA NA NA 68,435 6% 2014 NA NA NA NA 72,000 6% 2015 NA NA NA NA 72,000 6% 2016 NA NA NA NA 72,000 6% + Does not exceed 0.5 percent. NA (Not available) - For 2010 through 2014, the publicly available GHGRP data were aggregated at the national level. For 2015 and 2016, values were held constant with those from 2014. Facility-level data are not publicly available from EPA's GHGRP. a Includes only food & beverage applications. Uncertainty and Time-Series Consistency 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 a threshold for reporting under GHGRP applicable to those reporting under Subpart PP, 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, the EPA follows up with facilities to resolve mistakes that may have occurred.47 The resiilis nl" 1 lie \pprnaeh 2 qiianliiali\ e iiiiceriaiiiis ; 111; 11\ sis are suiiimari/ed in Table 4-54 ( ailxin dioside anMimpikni CO emissions fur 2<>l<> wereesimialed in he helueen 4. ^ and 4 " \l\ITCO l!q al llie l>5 pereeni 47 See . Industrial Processes and Product Use 4-61 ------- 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 ainl'idcMcc le\el. This indicates ;i mime iifapprii\iiiialcl> 5 pcrcciil hckm in 5 percent :ihi»\e I lie emission esiiniale ol'4.5 \1\1T CO \ x\ Table 4-54: Approach 2 Quantitative Uncertainty Estimates for CO2 Emissions from CO2 Consumption (MMT CO2 Eq. and Percent) - TO BE UPDATED FOR FINAL INVENTORY REPORT Sounv 21116 rimissiiin l!siiin;iU' (MMT CO: l'.(|.) I iuvrl;iinl\ K;iii|KT I.I HUT lilllllld 1 $111111(1 1$111111(1 I |)|KT lion 11(1 C(): Consumption C'(): 4.5 4.3 4.7 -5% Range of emission estimates predicted by Monte Carlo Stochastic Simulation for a l)5 percent confidence interval. Methodological approaches were applied to the entire time series to ensure consistency in emissions from 1990 through 2016. 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 QA/QC and Verification Procedures section in the introduction of the IPPU chapter. Planned Improvements EPA will continue to evaluate the potential to include additional GHGRP data on other emissive end-uses to improve 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, the latest guidance from the IPCC on the use of facility-level data in national inventories will be relied upon.48 These improvements, in addition to updating the time series when new data is available, are still in process and will be incorporated into future Inventory reports. 4.16 Phosphoric Acid Production (CRF Source 0) 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. Phosphate rock is mined in Florida and North Carolina, which account for more than 75 percent of total domestic output, as well as in Idaho and Utah and is used primarily as a raw material for wet-process phosphoric acid production (USGS 2017). The composition of natural phosphate rock varies depending upon 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 also may contain organic carbon. The calcium carbonate component of the phosphate rock is integral to the phosphate rock chemistry. Phosphate rock can also contain organic carbon that is physically incorporated into the mined rock but is not an integral component of the phosphate rock chemistry. 48 See . 4-62 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2016 ------- 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 The phosphoric acid production process involves chemical reaction of the calcium phosphate (Ca3(P04)2) component of the phosphate rock with sulfuric acid (H2SO4) and recirculated phosphoric acid (H3PO4) (EFMA 2000). However, the generation of CO2 is due to the associated limestone-sulfuric acid reaction, as shown below: CaCO3 + //2SO4 + H20 —* CaS04 • 2H20 + C02 Total U.S. phosphate rock production sold or used in 2016 was 26.5 million metric tons (USGS 2017). Total imports of phosphate rock to the United States in 2016 were approximately 1.6 million metric tons (USGS 2017). Between 2012 and 2015, most of the imported phosphate rock (58 percent) came from Peru, with the remaining 42 percent being from Morocco (USGS 2017). All phosphate rock mining companies are vertically integrated with fertilizer plants that produce phosphoric acid located near the mines. Some additional phosphoric acid production facilities are located in Texas, Louisiana, and Mississippi that used imported phosphate rock. Over the 1990 to 2016 period, domestic production has decreased by nearly 47 percent. Total CO2 emissions from phosphoric acid production were 1.0 MMT CO2 Eq. (992 kt CO2) in 2016 (see Table 4-55). Domestic consumption of phosphate rock in 2016 was estimated to have gone unchanged over 2015 levels (USGS 2017). Table 4-55: CO2 Emissions from Phosphoric Acid Production (MMT CO2 Eq. and kt) Year MMT CO2 Eq. kt 1990 1.5 1,529 2005 1.3 1.342 2012 1.1 1,118 2013 1.1 1,149 2014 1.0 1,038 2015 1.0 999 2016 1.0 992 Methodology 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: 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 is based on the assumption that all of the inorganic C (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 C content of the phosphate rock is converted to CO2 and that all of the organic C content remains in the phosphoric acid product. The United States uses a country-specific methodology to calculate emissions from production of phosphoric acid from phosphate rock. 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 4-56). For the years 1990 through 1992, and 2005 through 2016, 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 average share of U.S. production in those states Industrial Processes and Product Use 4-63 ------- 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 from 1993 to 2004 data. For the years 2005 through 2016, the same approximation method is used, but the share of U.S. production in those states data were obtained from the USGS commodity specialist for phosphate rock (USGS 2012). Data for domestic sales or consumption of phosphate rock, exports of phosphate rock (primarily from Florida and North Carolina), and imports of phosphate rock for consumption for 1990 through 2016 were obtained from USGS Minerals Yearbook: Phosphate Rock (USGS 1994 through 2015b), and from USGS Minerals Commodity Summaries: Phosphate Rock (USGS 2016, 2017). From 2004 through 2016, the USGS reported no exports of phosphate rock from U.S. producers (USGS 2005 through 2015b). 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 (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 carbon. Calcined phosphate rock mined in North Carolina and Idaho contains approximately 0.41 percent and 0.27 percent inorganic C, respectively (see Table 4-57). 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 (80 percent of domestic production) and carbonate content data for phosphate rock mined in Morocco are used to calculate CO2 emissions from consumption of imported phosphate rock. The CO2 emissions calculation is based on the assumption 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 4-56: Phosphate Rock Domestic Consumption, Exports, and Imports (kt) Location/Year 1990 2005 2012 2013 2014 2015 2016 U.S. Domestic Consumption 49,800 35,200 27,300 28,800 26,700 26,200 26,500 FLandNC 42,494 28,160 21,840 23,040 21,360 20,960 21,200 ID and UT 7,306 7,040 5,460 5,760 5,340 5,240 5,300 Exports—FL and NC 6,240 0 0 0 0 0 0 Imports 451 2,630 3,570 3,170 2,390 1,960 1,600 Total U.S. Consumption 44,011 37,830 30,870 31,970 29,090 28,160 28,100 Table 4-57: Chemical Composition of Phosphate Rock (Percent by Weight) Central North North Carolina Idaho Composition Florida Florida (calcined) (calcined) Morocco Total Carbon (as C) 1.60 1.76 0.76 0.60 1.56 Inorganic Carbon (as C) 1.00 0.93 0.41 0.27 1.46 Organic Carbon (as C) 0.60 0.83 0.35 0.00 0.10 Inorganic Carbon (as CO2) 3.67 3.43 1.50 1.00 5.00 Source: FIPR (2003a). Uncertainty and Time-Series Consistency 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 2016. For previous years in the time series, USGS provided the data disaggregated regionally; however, beginning in 2006, only total U.S. phosphate rock production was reported. Regional production for 2016 was estimated based on regional production data from previous years and multiplied by regionally-specific emission factors. There is uncertainty associated with the degree to which the estimated 2016 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 calculation are reported 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. 4-64 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2016 ------- 1 An additional source of uncertainty in the calculation of CO2 emissions from phosphoric acid production is the 2 carbonate composition of phosphate rock; the composition of phosphate rock varies depending upon where the 3 material is mined, and may also vary over time. The Inventory relies on one study (FIPR 2003a) of chemical 4 composition of the phosphate rock; limited data are available beyond this study. Another source of uncertainty is the 5 disposition of the organic carbon content of the phosphate rock. A representative of the Florida Institute of 6 Phosphate Research (FIPR) indicated that in the phosphoric acid production process, the organic C content of the 7 mined phosphate rock generally remains in the phosphoric acid product, which is what produces the color of the 8 phosphoric acid product (FIPR 2003b). Organic carbon is therefore not included in the calculation of CO2 emissions 9 from phosphoric acid production. 10 A third source of uncertainty is the assumption that all domestically-produced phosphate rock is used in phosphoric 11 acid production and used without first being calcined. Calcination of the phosphate rock would result in conversion 12 of some of the organic C in the phosphate rock into CO2. However, according to air permit information available to 13 the public, at least one facility has calcining units permitted for operation (NCDENR 2013). 14 Finally, USGS indicated that approximately 7 percent of domestically-produced phosphate rock is used to 15 manufacture elemental phosphorus and other phosphorus-based chemicals, rather than phosphoric acid (USGS 16 2006). According to USGS, there is only one domestic producer of elemental phosphorus, in Idaho, and no data 17 were available concerning the annual production of this single producer. Elemental phosphorus is produced by 18 reducing phosphate rock with coal coke, and it is therefore assumed that 100 percent of the carbonate content of the 19 phosphate rock will be converted to CO2 in the elemental phosphorus production process. The calculation for CO2 20 emissions is based on the assumption that phosphate rock consumption, for purposes other than phosphoric acid 21 production, results in CO2 emissions from 100 percent of the inorganic carbon content in phosphate rock, but none 22 from the organic carbon content. 23 The ivsiilis of ihe \pproach 2 (|ii;inlil;ili\e iiiiceriaiiils aiials sis are summari/ed 111 I ahle 4-5X 2d l(> phosphoric acid 24 produclioii (() emissions were csiimaled lo he helweeii 0 X and I 2\l\l l ( () lx| al 1 lie l>5 percciil confidence 25 le\ el This mdicales a mime of appi'o\inialel\ 11> percenl below and 2<> percciil aho\ e I he emission esiimale of I 0 26 \1\1T'CO \x\ 27 Table 4-58: Approach 2 Quantitative Uncertainty Estimates for CO2 Emissions from 28 Phosphoric Acid Production (MMT CO2 Eq. and Percent) - TO BE UPDATED FOR FINAL 29 INVENTORY REPORT Shu ni- (¦iis 20H> rimissiiin Kslim;iU- 1 MMT CO: Kil l I iHiTl;iinl> ki'liiliu'In I'.iiiissiiui Kslimuk"' (MM I' CO: Km.) ("..) I.I HUT I |)|KT I.I HUT I |)|KT liiillllll Bound liiillllil liiillllil Phosphoric Acid l'roi luction t () 1.0 0.8 1.2 -19% +20% Range of emission estimates predicted by Monte Carlo Stochastic Simulation lor a 95 percenl confidence interval. 30 Methodological approaches were applied to the entire time series to ensure consistency in emissions from 1990 31 through 2016. Details on the emission trends through time are described in more detail in the Methodology section, 32 above. 33 For more information on the general QA/QC process applied to this source category, consistent with Volume 1, 34 Chapter 6 of the 2006IPCC Guidelines, see QA/QC and Verification Procedures section in the introduction of the 35 IPPU Chapter. 36 Planned Improvements 37 EPA continues to evaluate potential improvements to the Inventory estimates for this source category, which include 38 direct integration of EPA's GHGRP data for 2010 through 2016 and the use of reported GHGRP data to update the 39 inorganic C content of phosphate rock for prior years. Confidentiality of CBI continues to be assessed, in addition to 40 the applicability of GHGRP data for the averaged inorganic C content data (by region) from 2010 through 2016 to 41 inform estimates in prior years in the required time series (i.e., 1990 through 2009). In implementing improvements 42 and integration of data from EPA's GHGRP, the latest guidance from the IPCC on the use of facility-level data in Industrial Processes and Product Use 4-65 ------- 1 national inventories will be relied upon.49 This planned improvement is still in development by EPA and have not 2 been implemented into the current Inventory report. 3 4.17 Iron and Steel Production (CRF Source 4 Category 2C1) and Metallurgical Coke 5 Production 6 Iron and steel production is a multi-step process that generates process-related emissions of carbon dioxide (CO2) 7 and methane (CH4) as raw materials are refined into iron and then transformed into crude steel. Emissions from 8 conventional fuels (e.g., natural gas, fuel oil) consumed for energy purposes during the production of iron and steel 9 are accounted for in the Energy chapter. 10 Iron and steel production includes six distinct production processes: coke production, sinter production, direct 11 reduced iron (DRI) production, pig iron50 production, electric arc furnace (EAF) steel production, and basic oxygen 12 furnace (BOF) steel production. The number of production processes at a particular plant is dependent upon the 13 specific plant configuration. Most process CO2 generated from the iron and steel industry is a result of the 14 production of crude iron. 15 In addition to the production processes mentioned above, CO2 is also generated at iron and steel mills through the 16 consumption of process byproducts (e.g., blast furnace gas, coke oven gas) used for various purposes including 17 heating, annealing, and electricity generation. Process byproducts sold for use as synthetic natural gas are deducted 18 and reported in the Energy chapter. In general, CO2 emissions are generated in these production processes through 19 the reduction and consumption of various carbon-containing inputs (e.g., ore, scrap, flux, coke byproducts). In 20 addition, fugitive CH4 emissions can also be generated from these processes, as well as from sinter, direct iron and 21 pellet production. 22 Currently, there are approximately 11 integrated iron and steel steelmaking facilities that utilize BOFs to refine and 23 produce steel from iron. These facilities have 21 active blast furnaces between them as of 2015. More than 100 24 steelmaking facilities utilize EAFs to produce steel primarily from recycled ferrous scrap (USGS 2017). The trend in 25 the United States for integrated facilities has been a shift towards fewer BOFs and more EAFs. EAFs use scrap steel 26 as their main input and use significantly less energy than BOFs. In addition, there are 16 cokemaking facilities, of 27 which 6 facilities are co-located with integrated iron and steel facilities (ACCCI2016). In the United States, raw 28 steel is produced in 37 states, but six states - Alabama, Arkansas, Indiana, Kentucky, Mississippi, and Tennessee - 29 count for roughly 50 percent of total production (AISI2017). 30 Total annual production of crude steel in the United States was fairly constant between 2000 and 2008 ranged from a 31 low of 99,320,000 tons to a high of 109,880,000 tons (2001 and 2004, respectively). Due to the decrease in demand 32 caused by the global economic downturn (particularly from the automotive industry), crude steel production in the 33 United States sharply decreased to 65,459,000 tons in 2009. In 2010, crude steel production rebounded to 34 88,731,000 tons as economic conditions improved and then continued to increase to 95,237,000 tons in 2011 and 35 97,769,000 tons in 2012; crude steel production slightly decreased to 95,766,000 tons in 2013 and then slightly 36 increased to 97,195,000 tons in 2014 (AISI 2017); crude steel production decreased to 86,912,000 tons in 2015 and 37 decreased again slightly in 2016 to 86,504,000 tons, a decrease of roughly 11 percent from 2014 levels. The United 38 States was the fourth largest producer of raw steel in the world, behind China, Japan and India, accounting for 39 approximately 4.8 percent of world production in 2016 (AISI 2017). 49 See . 50 pjg jj.on js 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 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. 4-66 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2016 ------- 1 The majority of CO2 emissions from the iron and steel production process come from the use of coke in the 2 production of pig iron and from the consumption of other process byproducts, with lesser amounts emitted from the 3 use of flux and from the removal of carbon from pig iron used to produce steel. 4 According to the 2006IPCC Guidelines, the production of metallurgical coke from coking coal is considered to be 5 an energy use of fossil fuel and the use of coke in iron and steel production is considered to be an industrial process 6 source. Therefore, the 2006 IPCC Guidelines suggest that emissions from the production of metallurgical coke 7 should be reported separately in the Energy sector, while emissions from coke consumption in iron and steel 8 production should be reported in the Industrial Processes and Product Use sector. However, the approaches and 9 emission estimates for both metallurgical coke production and iron and steel production are presented here because 10 much of the relevant activity data is used to estimate emissions from both metallurgical coke production and iron 11 and steel production. For example, some byproducts (e.g., coke oven gas) of the metallurgical coke production 12 process are consumed during iron and steel production, and some byproducts of the iron and steel production 13 process (e.g., blast furnace gas) are consumed during metallurgical coke production. Emissions associated with the 14 consumption of these byproducts are attributed at the point of consumption. Emissions associated with the use of 15 conventional fuels (e.g., natural gas, fuel oil) for electricity generation, heating and annealing, or other 16 miscellaneous purposes downstream of the iron and steelmaking furnaces are reported in the Energy chapter. 17 Metallurgical Coke Production 18 Emissions of CO2 from metallurgical coke production in 2016 were 1.3 MMT CO2 Eq. (1,323 kt CO2) (see Table 19 4-59 and Table 4-60). Emissions decreased significantly in 2016 by 54 percent from 2015 levels and have decreased 20 by 47 percent (1.2 MMT CO2 Eq.) since 1990. Coke production in 2016 was 43 percent lower than in 2000 and 57 21 percent below 1990. 22 Table 4-59: CO2 Emissions from Metallurgical Coke Production (MMT CO2 Eq.) Gas 1990 2005 2012 2013 2014 2015 2016 CO2 2.5 : 2.0 0.5 1.8 2.0 2.8 1.3 Total 2.5 2.0 0.5 1.8 2.0 2.8 1.3 23 Table 4-60: CO2 Emissions from Metallurgical Coke Production (kt) Gas 1990 2005 2012 2013 2014 2015 M16 CO2 2,503 2,044 543 1,824 2,014 2,839 1,323 Total 2,503 2,044 543 1,824 2,014 2,839 1,323 25 26 Iron and Steel Production 27 Emissions of CO2 and CH4 from iron and steel production in 2016 were 40.9 MMT CO2 Eq. (40,896 kt) and 0.0074 28 MMT CO2 Eq. (0.3 kt CH4), respectively (see Table 4-61 through Table 4-64), totaling approximately 40.9 MMT 29 CO2 Eq. Emissions decreased in 2016 from 2015 and have decreased overall since 1990 due to restructuring of the 30 industry, technological improvements, and increased scrap steel utilization. Carbon dioxide emission estimates 31 include emissions from the consumption of carbonaceous materials in the blast furnace, EAF, and BOF, as well as 32 blast furnace gas and coke oven gas consumption for other activities at the steel mill. 33 In 2016, domestic production of pig iron decreased by 12 percent from 2015 levels. Overall, domestic pig iron 34 production has declined since the 1990s. Pig iron production in 2016 was 53 percent lower than in 2000 and 55 35 percent below 1990. Carbon dioxide emissions from iron production have decreased by 78 percent since 1990. 36 Carbon dioxide emissions from steel production have decreased by 14 percent (1.1 MMT CO2 Eq.) since 1990, 37 while overall CO2 emissions from iron and steel production have declined by 59 percent (58.1 MMT CO2 Eq.) from 38 1990 to 2016. 39 Table 4-61: CO2 Emissions from Iron and Steel Production (MMT CO2 Eq.) Source/Activity Data 1990 2005 2012 2013 2014 2015 2016" Sinter Production 2.4 1.7 1.2 1.1 1.1 1.0 0.9 Industrial Processes and Product Use 4-67 ------- Iron Production 45.6 17.5 10.9 11.9 18.6 11.7 9.9 Pellet Production 1.8 1.5 1.2 1.2 1.1 1.0 0.9 Steel Production 7.9 9.4 9.9 8.6 7.5 6.9 6.8 Other Activities3 41.2 35.9 31.7 28.7 27.9 24.3 22.4 Total 99.0 66.0 54.9 51.5 56.2 44.9 40.9 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. 1 Table 4-62: CO2 Emissions from Iron and Steel Production (kt) Source/Activity Data 1990 2005 2012 2013 2014 2015 2016 Sinter Production 2,448 1,663 1,159 1,117 1,104 1,016 877 Iron Production 45,592 17,545 10,918 11,935 18,629 11,696 9,853 Pellet Production 1,817 1,503 1,219 1,146 1,126 964 869 Steel Production 7,933 9,356 9,860 8,617 7,450 6,924 6,850 Other Activitiesa 41,193 35,934 31,750 28,709 27,911 24,280 22,448 Total 98,984 66,003 54,906 51,525 56,220 44,879 40,896 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. 2 Table 4-63: Cm Emissions from Iron and Steel Production (MMT CO2 Eq.) Source/Activity Data 1990 2005 2012 2013 2014 2015 2016 Sinter Production + +; + + + + + Total + + + + + + + + Does not exceed 0.05 MMT CO2 Eq. 3 Table 4-64: Cm Emissions from Iron and Steel Production (kt) Source/Activity Data 1990 2005 2012 2013 2014 2015 2016 Sinter Production 0.9 0.6 5 0.4 0.4 0.4 0.3 0.3 Total 0.9 0.6 0.4 0.4 0.4 0.3 0.3 Methodology 5 6 7 8 9 10 11 12 13 14 15 16 17 18 Emission estimates presented in this chapter utilize a country-specific approach based on Tier 2 methodologies provided by the 2006IPCC Guidelines. 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. 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. The Tier 2 methodology equation is as follows: Eco2 ~ X Ca) ~ X Q) 44 12 where, Ec02 a b Qa ca Qb Emissions from coke, pig iron, EAF steel, or BOF steel production, metric tons Input material a Output material b Quantity of input material a, metric tons Carbon content of input material a, metric tons C/metric ton material Quantity of output material b, metric tons 4-68 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2016 ------- 1 2 3 Cb 44/12 Carbon content of output material b, metric tons C/metric ton material Stoichiometric ratio of CO2 to C 4 The Tier 1 methodology equations are as follows: 5 Es,p = Qs x EFSiP 6 Ed,C02 — Qd X EFd£Q2 1 Ep,co2 — Qp x EFp C02 8 where, 9 10 11 12 13 14 15 16 17 EFp,co2 Ed,C02 Qd EFd,co2 QP Emissions from sinter production process for pollutant p (CO2 or CH4), metric ton Quantity of sinter produced, metric tons Emission factor for pollutant p (CO2 or CH4), metric ton /Vmctric ton sinter Emissions from DRI production process for CO2, metric ton Quantity of DRI produced, metric tons Emission factor for CO2, metric ton CCh/metric ton DRI Quantity of pellets produced, metric tons Emission factor for CO2, metric ton CCh/metric ton pellets produced is Metallurgical Coke Production 19 Coking coal is used to manufacture metallurgical coke that is used primarily as a reducing agent in the production of 20 iron and steel, but is also used in the production of other metals including zinc and lead (see Zinc Production and 21 Lead Production sections of this chapter). Emissions associated with producing metallurgical coke from coking coal 22 are estimated and reported separately from emissions that result from the iron and steel production process. To 23 estimate emissions from metallurgical coke production, a Tier 2 method provided by the 2006IPCC Guidelines was 24 utilized. The amount of carbon contained in materials produced during the metallurgical coke production process 25 (i.e., coke, coke breeze and coke oven gas) is deducted from the amount of carbon contained in materials consumed 26 during the metallurgical coke production process (i.e., natural gas, blast furnace gas, and coking coal). Light oil, 27 which is produced during the metallurgical coke production process, is excluded from the deductions due to data 28 limitations. The amount of carbon contained in these materials is calculated by multiplying the material-specific 29 carbon content by the amount of material consumed or produced (see Table 4-65). The amount of coal tar produced 30 was approximated using a production factor of 0.03 tons of coal tar per ton of coking coal consumed. The amount of 31 coke breeze produced was approximated using a production factor of 0.075 tons of coke breeze per ton of coking 32 coal consumed (AISI 2008; DOE 2000). Data on the consumption of carbonaceous materials (other than coking 33 coal) as well as coke oven gas production were available for integrated steel mills only (i.e., steel mills with co- 34 located coke plants). Therefore, carbonaceous material (other than coking coal) consumption and coke oven gas 35 production were excluded from emission estimates for merchant coke plants. Carbon contained in coke oven gas 36 used for coke-oven underfiring was not included in the deductions to avoid double-counting. 37 Table 4-65: Material Carbon Contents for Metallurgical Coke Production Material kgC/kg Coal Tar 0.62 Coke 0.83 Coke Breeze 0.83 Coking Coal 0.73 Material kgC/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. 38 Although the 2006IPCC Guidelines provide a Tier 1 CH4 emission factor for metallurgical coke production (i.e., 39 0.1 g CH4 per metric ton of coke production), it is not appropriate to use because CO2 emissions were estimated Industrial Processes and Product Use 4-69 ------- 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 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 reports 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 2017) (see Table 4-66). 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 (AISI2004 through 2017) and through personal communications with AISI (AISI 2008) (see Table 4-67). The factor for the quantity of coal tar produced per ton of coking coal consumed was provided by AISI (AISI 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). Currently, data on 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 coking coal, metallurgical coke, coal tar, coke oven gas, and blast furnace gas were provided by the 2006IPCC Guidelines. The C content for coke breeze was assumed to equal the C content of coke. Table 4-66: Production and Consumption Data for the Calculation of CO2 Emissions from Metallurgical Coke Production (Thousand Metric Tons) Source/Activity Data 1990 2005 2012 2013 2014 2015 2016 Metallurgical Coke Production Coking Coal Consumption at Coke Plants 35,269 21,259 18,825 19,481 19,321 17,879 14,955 Coke Production at Coke Plants 25,054 15,167 13,764 13,898 13,748 12,479 10,755 Coal Breeze Production 2,645 1,594 1,412 1,461 1,449 1,341 1,122 Coal Tar Production 1,058 638 565 584 580 536 449 Table 4-67: Production and Consumption Data for the Calculation of CO2 Emissions from Metallurgical Coke Production (Million ft3) Source/Activity Data 1990 2005 2012 2013 2014 2015 2016 Metallurgical Coke Production Coke Oven Gas Production 250,767 114,213 113,772 108,162 102,899 84,336 74,807 Natural Gas Consumption 599 2,996 3,267 3,247 3,039 2,338 2,077 Blast Furnace Gas Consumption 24,602 4,460 4,351 4,255 4,346 4,185 3,741 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). The carbon contained in the pig iron, blast furnace gas, and blast furnace inputs was estimated by multiplying the material-specific C content by each material type (see Table 4-68). 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 C contained in inputs such as natural gas, coke oven gas, fluxes (e.g. burnt lime or dolomite), and pig iron. In each case, the carbon was calculated by multiplying material-specific carbon contents by each material type (see Table 4-68). For EAFs, the amount of EAF anode consumed was 4-70 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2016 ------- 1 approximated by multiplying total EAF steel production by the amount of EAF anode consumed per metric ton of 2 steel produced (0.002 metric tons EAF anode per metric ton steel produced [AISI 2008]). The amount of flux (e.g., 3 burnt lime or dolomite) used in pig iron production was deducted from the "Other Process Uses of Carbonates" 4 source category (CRF Source Category 2A4) to avoid double-counting. 5 Carbon dioxide emissions from the consumption of blast furnace gas and coke oven gas for other activities occurring 6 at the steel mill were estimated by multiplying the amount of these materials consumed for these purposes by the 7 material-specific carbon content (see Table 4-68). 8 Carbon dioxide emissions associated with the sinter production, direct reduced iron production, pig iron production, 9 steel production, and other steel mill activities were summed to calculate the total CO2 emissions from iron and steel 10 production (see Table 4-61 and Table 4-62). 11 Table 4-68: Material Carbon Contents for Iron and Steel Production Material kgC/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 kgC/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. 12 The production process for sinter results in fugitive emissions of CH4, which are emitted via leaks in the production 13 equipment, rather than through the emission stacks or vents of the production plants. The fugitive emissions were 14 calculated by applying Tier 1 emission factors taken from the 2006 IPCC Guidelines for sinter production (see Table 15 4-69). Although the 1995 IPCC Guidelines (IPCC/UNEP/OECD/IEA 1995) provide a Tier 1 CH4 emission factor 16 for pig iron production, it is not appropriate to use because CO2 emissions were estimated using the Tier 2 mass 17 balance methodology. The mass balance methodology makes a basic assumption that all carbon that enters the pig 18 iron production process either exits the process as part of a carbon-containing output or as CO2 emissions; the 19 estimation of CH4 emissions is precluded. A preliminary analysis of facility-level emissions reported during iron 20 production further supports this assumption and indicates that CH4 emissions are below 500 kt CO2 Eq. and well 21 below 0.05 percent of total national emissions. The production of direct reduced iron also results in emissions of 22 CH4 through the consumption of fossil fuels (e.g., natural gas, etc.); however, these emission estimates are excluded 23 due to data limitations. Pending further analysis and resources, EPA may include these emissions in future reports to 24 enhance completeness. EPA is still assessing the possibility of including these emissions in future reports and have 25 not included this data in the current report. 26 Table 4-69: ChU Emission Factors for Sinter and Pig Iron Production Material Produced Factor Unit Sinter 0.07 kg CEU/metric ton Source: IPCC (2006), Table 4.2. 27 Emissions of CO2 from sinter production, direct reduced iron production and pellet production were estimated by 28 multiplying total national sinter production and the total national direct reduced iron production by Tier 1 CO2 29 emission factors (see Table 4-70). Because estimates of sinter production, direct reduced iron production and pellet 30 production were not available, production was assumed to equal consumption. Industrial Processes and Product Use 4-71 ------- 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 Table 4-70: CO2 Emission Factors for Sinter Production, Direct Reduced Iron Production and Pellet Production Metric Ton Material Produced CCh/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 2016 were obtained from AISI's Annual Statistical Report (AISI2004 through 2017) and through personal communications with AISI (AISI2008) (see Table 4-71). 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 2016) and personal communication with the USGS Iron and Steel Commodity Specialist (Fenton 2015). However, data for DRI consumed inEAFs 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. Also, 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 2017) and through personal communications with AISI (AISI 2008) (see Table 4-71 and Table 4-72). Data for EAF steel production, flux, EAF charge carbon, and natural gas consumption were obtained from AISI's Annual Statistical Report (AISI 2004 through 2017) and through personal communications with AISI (AISI 2006 through 2016 and AISI 2008). The factor for the quantity of EAF anode consumed per ton of EAF steel produced was provided by AISI (AISI 2008). Data for BOF steel production, flux, natural gas, natural ore, pellet, sinter consumption as well as BOF steel production were obtained from AISI's Annua! Statistical Report (AISI 2004 through 2017) and through personal communications with AISI (AISI 2008). Data for EAF and BOF scrap steel, pig iron, andDRI consumption were obtained from the USGS Minerals Yearbook-Iron and Steel Scrap (USGS 1991 through 2016). 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 .innual Statistical Report (AISI 2004 through 2017) and through personal communications with AISI (AISI 2008). Data on blast furnace gas and coke oven gas sold for use as synthetic natural gas were obtained from EIA's Natural Gas Annual (EIA 2016b). Carbon contents for direct reduced iron, EAF carbon electrodes, EAF charge carbon, limestone, dolomite, pig iron, and steel were provided by the 2006 IPCC Guidelines. The carbon contents for natural gas, fuel oil, and direct injection coal were obtained from EIA (EIA 2016c) 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 . Innual Statistical Report (AISI 2004 through 2017). Heat contents for coke oven gas and blast furnace gas were provided in Table 37 of AISI's Annua! Statistical Report (AISI 2004 through 2017) and confirmed by AISI staff (Carroll 2016). 4-72 DRAFT Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2016 ------- 1 Table 4-71: Production and Consumption Data for the Calculation of CO2 and ChU Emissions 2 from Iron and Steel Production (Thousand Metric Tons) Source/Activity Data 1990 2005 2012 2013 2014 2015 2016 Sinter Production Sinter Production 12,239 8,315 5,795 5,583 5,521 5,079 4,385 Direct Reduced Iron Production Direct Reduced Iron Production 516 1,303 3,530 3,350 4,790 4,790 4,777 Pellet Production Pellet Production 60,563 50,096 40,622 38,198 37,538 32,146 28,967 Pig Iron Production Coke Consumption 24,946 13,832 9,571 9,308 11,136 7,969 7,124 Pig Iron Production 49,669 37,222 32,063 30,309 29,375 25,436 22,293 Direct Injection Coal Consumption 1,485 2,5"' 2,802 2,675 2,425 2,275 1,935 EAF Steel Production EAF Anode and Charge Carbon Consumption 67 1,127 1,318 1,122 1,062 1,072 1,120 Scrap Steel Consumption 42,691 46,600 50,900 47,300 48,873 44,000 43,211 Flux Consumption 319 695 748 771 771 998 998 EAF Steel Production 33,511 52,194 52,415 52,641 55,174 49,451 52,589 BOF Steel Production Pig Iron Consumption 47,307 34,400 31,500 29,600 23,755 20,349 18,620 Scrap Steel Consumption 14,713 11,400 8,350 7,890 5,917 4,526 4,573 Flux Consumption 576 582 476 454 454 454 408 BOF Steel Production 43,9"' 42,705 36,282 34,238 33,000 29,396 25,888 3 Table 4-72: Production and Consumption Data for the Calculation of CO2 Emissions from 4 Iron and Steel Production (Million ft3 unless otherwise specified) Source/Activity Data 1990 2005 2012 2013 2014 2015 2016 Pig Iron Production Natural Gas Consumption 56,273 59,844 62,469 48,812 47,734 43,294 38,396 Fuel Oil Consumption (thousand gallons) 163,397 16,170 19,240 17,468 16,674 9,326 6,124 Coke Oven Gas Consumption 22,033 16,557 18,608 17,710 16,896 13,921 12,404 Blast Furnace Gas Production 1,439,380 1,299,980 1,139,578 1,026,973 1,000,536 874,670 811,005 EAF Steel Production Natural Gas Consumption 15,905 19,985 11,145 10,514 9,622 8,751 3,915 BOF Steel Production Coke Oven Gas Consumption 3,851 524 568 568 524 386 367 Other Activities Coke Oven Gas Consumption 224,883 97,132 94,596 89,884 85,479 70,029 62,036 Blast Furnace Gas Consumption 1,414,778 1,295,520 1,135,227 1,022,718 996,190 870,485 807,264 Industrial Processes and Product Use 4-73 ------- 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 Uncertainty and Time-Series Consistency The estimates of CO2 emissions from metallurgical coke production are based on 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 C 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; some of this coal may be used for electricity generation. There is also uncertainty associated with the C contents for pellets, sinter, and natural ore, which are assumed to equal the C contents of direct reduced iron, when consumed in the blast furnace. 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 this Inventory report, EPA initiated conversation with AISI to update the qualitative and quantitative uncertainty metrics associated with AISI data elements. EPA has yet to incorporate these changes into this current Public Review draft but will include them in the final Inventory report published in April 2018. The lesiills of I he \pproach 2 t|ii;inlil;ili\ e iiiiccriaiuls ; 111; 11\ sis ;iiv suniniai'i/ed 111 I ahle 4-~' fur niclallurmcal coke production and iron and sieel production Tolal CO emissions Iroin niclalhumcal coke production and iron ;ind steel production for 2d lii were estimated lo he between "5 t> and 4l>.4 \1\11 ( () I !q ;iiihe l>5 percent confidence le\ el This uidic;iles a raime of approximate^ I ~ perceul helow and I ~ perceui aho\ e llie emission esiiniale of 42.2 \1\1T CO I !q Tolal CI I emissions from niclalhumcal coke producliou and iron and sieel production for2<>|()<¦ and 0