EPA-600/7-88-022
November 1988
<&EPA Research and
Development
ANTHROPOGENIC EMISSIONS
DATA FOR THE
1985 NAPAP INVENTORY
United States
Environmental Protection
Agency
Prepared for
The National Acid Precipitation Assessment Program
Prepared by
Air and Energy Engineering Research
Laboratory
Research Triangle Park NC 27711
-------
RESEARCH REPORTING SERIES
Research reports of the Office of Research and Development, U.S. Environmental
Protection Agency, have been grouped into nine series. These nine broad cate-
gories were established to facilitate further development and application of en-
vironmental technology. Elimination of traditional grouping was consciously
planned to foster technology transfer and a maximum interface in related fields
The nine series are:
1. Environmental Health Effects Research
2. Environmental Protection Technology
3. Ecological Research
4. Environmental Monitoring
5. Socioeconomic Environmental Studies
6. Scientific and Technical Assessment Reports (STAR)
7 Interagency Energy-Environment Research and Development
8. "Special" Reports
9. Miscellaneous Reports
This report has been assigned to the INTERAGENCY ENERGY-ENVIRONMENT
RESEARCH AND DEVELOPMENT series. Reports in this series result from the
effort funded under the 17-agency Federal Energy/Environment Research and
Development Program. These studies relate to EPA's mission to protect the public
health and welfare from adverse effects of pollutants associated with energy sys-
tems The goal of the Program is to assure the rapid development of domestic
energy supplies in an environmentally-compatible manner by providing the nec-
essary environmental data and control technology. Investigations include analy-
ses of the transport of energy-related pollutants and their health and ecological
effects; assessments of, and development of. control technologies for energy
systems; and integrated assessments of a wide range of energy-related environ-
mental issues.
EPA REVIEW NOTICE
This report has been reviewed by the participating Federal Agencies, and approved
for publication Approval does not signify that the contents necessarily reflect
the views and policies of the Government, nor does mention of trade names or
commercial products constitute endorsement or recommendation for use
This document is available to the public through the National Technical Informa-
tion Service, Springfield, Virginia 22161.
-------
EP.A-600 7-88-022
November 1983
ANTHROPOGENIC EMISSIONS DATA
FOR THE 1985 NAPAP INVENTORY
Final Report
By
David Zimmerman
Wienke Tax
Mark Smith
Janice Deramy
Rebecca Battye
ALLIANCE TECHNOLOGIES CORPORATION
500 Eastowne Drive
Chapel Hill, North Carolina 27514
EPA Contract 68-02-4274
Work Assignments 26 and 28
EPA Project Officer: Robert C. Lagemann
Air and Energy Engineering Research Laboratory
Research Triangle Park, North Carolina 27711
This study was conducted in cooperation with the
National Acid Precipitation Assessment Program,
Prepared for;
U.S. Environmental Protection Agency
Office of Research and Development
Washington, DC 20460
-------
DISCLAIMER
This report was furnished to the Environmental Protection Agency by
Alliance Technologies Corporation, 213 Burlington Road, Bedford, Massachusetts,
01730, in fulfillment of Contract number 68-02-4274, Work Assignment numbers 26
and 28. The opinions, findings, and conclusions expressed are tnose of the
authors and not necessarily those of the Environmental Protection Agency.
ii
-------
Acknowledgements
The development of a national emissions inventory requires a team of
federal, State, and local environmental professionals to acquire, review, and
computerize engineering data from more than one hundred thousand sources of ai
pollution across the United States. The development of the 1985 emissions
inventory for the National Acid Precipitation Assessment Program was begun in
1985 by EPA Administrator Lee Thomas, the State and Territorial Air Pollution
Program Administrators, the Association of Local Air Pollution Control
Officials, and State Environmental Directors (see Appendix A). Before its
completion in 1988, the inventory represented the work of literally hundreds o
people. Although it is not possible to recognize individually every member of
this team, the United States Environmental Protection Agency gratefully
acknowledges this help as well as the participation of the following
individuals.
State Environmental Agencies
Alabama Department of Environmental Management
Richard Grusnick Chief, Air Division
Sue Robertson NAPAP Coordinator
Arizona Department of Air Quality
Dr. Gerald Teletzke Director
Dennis Siivola NAPAP Coordinator
Anthony Leverock NAPAP Coordinator
Arkansas Department of Pollution Control and Ecology
Wilson Tolefree Chief, Division of Air Pollution
Control
Steven T. Coldwell NAPAP Coordinator
California Air Resources Board
Jannane Sharpless Secretary of Environmental Affairs
Beth Schwehr NAPAP Coordinator
Colorado Department of Health - Air Pollution Control Division
Dr. Thomas M. Vernon Executive Director
Nadine Quigley NAPAP Coordinator
Jim King NAPAP Coordinator
Connecticut Department of Environmental Protection
Stanley J. Pac Commissioner
Ron Freeto NAPAP Coordinator
Delaware Department of Natural Resources and Environmental Control
Richard J. Touhey Director, Division of Air and Waste
Management
James Short NAPAP Coordinator
-------
District of Columbia Department of Consumer and Regulatory Affairs
Anantha Padmanabha Program Manager
Joyce Chandler NAPAP Coordinator
Florida Department of Environmental Regulation
Steven Smallwood Chief, Bureau of Air Quality Management
Larry George NAPAP Coordinator
Georgia Department of Natural Resources
Robert H. Collom Chief, Air Protection Branch
Kenneth Powell NAPAP Coordinator
Idaho Department of Health and Welfare
Kenneth D. Brooks Chief, Bureau of Air Quality
Douglas Hardesty NAPAP Coordinator
Illinois Environmental Protection Agency
Richard J. Carlson Director
Jim Levis NAPAP Coordinator
Indiana Department of Environmental Management
Nancy A. Maloley Commissioner
Ismail Khatri NAPAP Coordinator
Iowa Department of Natural Resources
Larry Wilson Director
John Vedder NAPAP Coordinator
Kansas Department of Health and Environment
David J. Romano Manager, Bureau of Air Quality and
Radiation Control
Donna Dees NAPAP Coordinator
Kentucky Department for Environmental Protection
Roger B. McCann Director, Division of Air Pollution
Control
Diana Parker NAPAP Coordinator
Louisiana Department of Environmental Quality
Tom Coerver Administrator, Air Quality Division
Bill Hopkins NAPAP Coordinator
Maine Department of Environmental Protection
Kenneth C. Young Commissioner, Bureau of Air Quality
Control
Ron Severance NAPAP Coordinator
Gerald Bernier NAPAP Coordinator
iv
-------
Maryland Office of Environmental Programs
George P. Ferreri Director, Air Management Administration
Ed Carter NAPAP Coordinator
Massachusetts Division of Air Quality Control
Bruce K. Maillet Director
Robert Boiselle NAPAP Coordinator
Michigan Department of Natural Resources
Robert P. Miller Chief, Air Quality Division
Paul Shutt NAPAP Coordinator
Minnesota Pollution Control Agency
J. Michael Valentine Director, Division of Air Quality
Kathy Gedde NAPAP Coordinator
Mississippi Department of Natural Resources
Dwight K. Wylie Chief, Air Quality Division
Wayne Anderson NAPAP Coordinator
Missouri Department of Natural Resources
Nick Nikkila Staff Director, Air Pollution Control
Randy Raymond NAPAP Coordinator
Montana Department of Health and Environmental Sciences
Dr. John J. Drynan Director, Air Quality Bureau
Harry Keltz NAPAP Coordinator
Nebraska Department of Environmental Control
Gene Robinson Chief, Air Pollution Control Division
Dennis Burling NAPAP Coordinator
Nevada Division of Environmental Protection
Lowell H. Shifley, Jr. State Air Quality Officer
Gaye McCleary NAPAP Coordinator
New Hampshire Air Resources Agency
Dennis R. Lunderville Director
Linda Spofford NAPAP Coordinator
New Jersey Department of Environmental Protection
Dr. Jorge H. Berkowitz Director, Air Pollution Control Program
John Elston NAPAP Coordinator
Greg Cooper NAPAP Coordinator
New Mexico Environmental Improvement Division
Dennis Fort Director, Air Quality Bureau
Jim Shively NAPAP Coordinator
New York Department of Environmental Conservation
Harry H. Hovey, Jr. Director, Division of Air Resources
Edward Davis NAPAP Coordinator
V
-------
North Carolina Division of Environmental Management
N.O. Gerald Chief, Air Quality Section
Thena Smith NAPAP Coordinator
North Dakota Department of Health
Dana K. Mount Director, Division of Environmental
Engineering
Doug Lipetsky NAPAP Coordinator
Tom Bachman NAPAP Coordinator
Ohio Environmental Protection Agency
Patricia P. Walling Chief, Division of Air Pollution Control
Bill Juris NAPAP Coordinator
Oklahoma Environmental Health Services
John W. Drake Chief, Air Quality Service
Larry Trent NAPAP Coordinator
Oregon Department of Environmental Quality
Fred Hansen Director
Mary Heath NAPAP Coordinator
Pennsylvania Department of Environmental Resources
Dr. James K. Hambright Director
Bob Kulp NAPAP Coordinator
Rhode Island Division of Air and Hazardous Materials
Thomas D. Getz Chief
Chris James NAPAP Coordinator
South Carolina Department of Health and Environmental Control
Otto E. Pearson Chief, Bureau of Air Quality Control
Dennis Ellenwood NAPAP Coordinator
South Dakota Department of Water and Natural Resources
Joel C. Smith Administrator, Office of Air Quality
and Solid Waste
Beth Lockwood NAPAP Coordinator
Tennessee Division of Air Pollution Control
Harold E. Hodges Director
Michael Langreck NAPAP Coordinator
Texas Air Control Board
Allen Eli Bell
Bob Love
Bruce Broberg
Executive Director
NAPAP Coordinator
NAPAP Coordinator
vi
-------
Utah Department of Health, Division of Environment
F. Burnell Cordner Director, Bureau of Air Quality
Robert Dalley NAPAP Coordinator
Vermont Agency of Environmental Conservation
Harold T. Garabedian Air Pollution Control Officer
Neal Jordan NAPAP Coordinator
Virginia Air Pollution Control Board
Richard L. Cook Executive Director
Kirit Chaudhari NAPAP Coordinator
Washington Department of Ecology
Andrea Beatty Riniker Director
Alan Butler NAPAP Coordinator
West Virginia Air Pollution Control Commission
Carl G. Beard Director
Dale Farley NAPAP Coordinator
Wisconsin Department of Natural Resources
Donald Theiler Director, Bureau of¦ Air Management
Paul Yeung NAPAP Coordinator
Wyoming Department of Environmental Quality
Charles Collins Administrator, Air Quality Division
Bernard Dailey NAPAP Coordinator
STAPPA/ALAPCO Officials
State and Territorial Air Pollution Program Administrators
Nick Nikkila President
Dr. James K. Hambright Chair, Interstate Transport and Acid Deposition
Commission
Association of Local Air Pollution Control Officials
Dr. James Lents President
Edgar Chase Chair, Acid Rain Committee
EPA Regional Offices
Region I
Louis Gitto Director, Air Management Division
Bob Judge NAPAP Coordinator
vii
-------
Region II
Conrad Simon
Tom BaLlou
Director, Air and Waste Management Division
NAPAP Coordinator
Region III
Thomas J. Maslany
Ted Erdman
Director, Air Management Division
NAPAP Coordinator
Region IV
Winston A. Smith
Stewart Perry
Region V
David A. Kee
Becky CauLby
Barry Bolka
Region VI
William B. Hathaway
Joseph Winkler
Region VII
William A. Spratlin
Dan Wheeler
Director, Air, Pesticides & Toxic Management
Division
NAPAP Coordinator
Director, Air Management Division
NAPAP Coordinator
NAPAP Coordinator
Director, Air, Pesticides, and Toxics Division
NAPAP Coordinator
Director, Air and Toxic Management Division
NAPAP Coordinator
Region VIII
Irwin L. Dickstein
John Dale
Director, Air and Toxics Division
NAPAP Coordinator
Region IX
David P. Howekamp
Allison Bird
Director, Air Management Division
NAPAP Coordinator
Region X
Gary L. O'Neal
Rich White
Bill Puckett
Director, Air & Toxics Division
NAPAP Coordinator
NAPAP Coordinator
Bill Laxton
Richard Rhoads
John Bosch
John Fink
Chuck Mann
Sue Kimbrough
David Johnson
EPA Office of Air Quality Planning and Standards
Director, Technical Support Division
Director, Monitoring and Data Analysis Division
Chief, National Air Data Branch
Chief, Operations and Maintenance Section
NAPAP Coordinator
NAPAP Coordinator
NAPAP Coordinator
-------
EPA Air and Energy Engineering Research Laboratory
Frank Princiotta
Jim Abbott
Mike Maxwell
Rob Lagemann
Larry Jones
Dale Pahl
Di rector
Director, Engineering Analysis Division
Chief, Acid Deposition Branch
NAPAP Coordinator
NAPAP Coordinator
EPA 1985 Emissions Inventory Program Manager
NAPAP Emissions and Controls Task Group and Advisors
David Beecy
Former Chair
(U.S. Department of Energy)
Denise Swink
Current Chair
(U.S. Department of Energy)
Ed Trexler
Deputy Chair
(U.S. Department of Energy)
Jim Kelly
Member
(U.S. Department of Energy)
Carmen Benkovitz
Member
(Brookhaven National Laboratory)
Paul Schwengels
Member
(U.S. EPA)
Joan Novak
Member
(U.S. EPA)
Peter Mueller
Member
(Electric Power Research
Institute)
Frank Vena
Advisor
(Environment Canada)
Tony Kosteltz
Advisor
(Environment Canada)
Fred Fehsenfeld
Advisor
(National Oceanic & Atmospheric
.
Administration)
Robin Dennis
Advisor
(U.S. EPA)
Brian McLean
Advisor
(U.S. EPA)
William Norris
Advisor
(Tennessee Valley Authority)
Gordon Beales
Advi sor
(Electric Power Research
Institute)
Marylynn Placet
Advi sor
(Argonne National Laboratory)
Paulette Middleton
Advisor
(National Center for Atmospheric
Research)
David Streets
Advisor
(Argonne National Laboratory)
This work was funded and administered by the U.S. Environmental Protection
Agency under EPA Contract Nos. 68-02-3892, 68-02-3997, and 68-02-4274 to
Alliance Technologies Corporation; Contract No. 68-02-3888 to Engineering
Science; Contract No. 68-02-3891 to Midwest Research Institute; Contract No.
68-02-3887 to Pacific Environmental Services; Contract No. 68-02-3890 to E.H.
Pechan and Associates; and Contract No. 68-02-3893 to Scientific Applications
International Corporation.
Preparation of the report itself was a cooperative effort involving the
authors and staff at Alliance Technologies Corporation and numerous EPA
personnel. It was prepared under Contract No. 68-02-4274, Work Assignment
numbers 26 and 28.
ix
-------
ENGLISH TO METRIC CONVERSION FACTORS
1 ton = 907.1860 kilograms
1 foot = 0.30A8 meters
1 gallon = 3.785 liters
1 cubic foot = 28.317 liters
1 megawatt = 1 million Joules/second
= 948.6 BTU/second
1 BTU = 1055 Joules
degrees Fahrenheit = (1.8 x °C) + 32
1 mile/hour = 1.609 kilometers/hour
-------
CONTENTS
Page
DISCLAIMER ii
ACKNOWLEDGEMENTS iii
ENGLISH TO METRIC CONVERSION FACTORS x
LIST OF TABLES xv
LIST OF FIGURES xviii
1.0 EXECUTIVE SUMMARY 1-1
1.1 BACKGROUND 1-1
1.2 METHODOLOGY 1-2
1.3 QUALITY CONTROL 1-4
1.4 SUMMARIES AND ANALYSES OF THE 1985 DATA 1-4
1.5 DOCUMENTATION 1-23
2.0 INTRODUCTION 2-1
2.1 EMISSIONS RESEARCH AND THE NATIONAL ACID PRECIPITATION
ASSESSMENT PROGRAM 2-1
2.2 1985 EMISSIONS INVENTORY OBJECTIVES 2-2
2.3 1985 EMISSIONS INVENTORY DATA 2-5
2.4 OBJECTIVES AND STRUCTURE OF THIS REPORT 2-8
3.0 METHODOLOGY 3-1
3.1 POINT SOURCE DATA 3-2
3.1.1 NEDS Structure 3-2
3.1.2 Priority Data Elements/NAPAP Emissions Inventory
Priorities for 1985 3-3
3.1.3 1985 Emissions Data Collection Effort 3-7
3.1.4 STAPPA Survey 3-8
3.1.5 Written Analysis of Deficiencies 3-9
3.1.6 Guidance 3-9
3.1.6.1 Technical Guidance 3-9
3.1.6.2 Utility Data 3-10
3.1.7 Emissions Estimation Procedures 3-11
3.1.8 Inventory Process at State Level 3-12
3.1.9 Confirmation 3-13
3.1.10 Quality Control 3-15
3.1.11 Transfer to NEDS 3-15
3.2 AREA SOURCES 3-16
3.2.1 Overview 3-16
3.2.1.1 Stationary Sources 3-17
3.2.1.2 Mobile Sources 3-20
3.2.1.3 Solid Waste Disposal 3-21
3.2.1.4 Miscellaneous Area Sources 3-22
3.2.1.5 Additional Area Sources 3-22
3.2.2 Stationary Sources 3-23
3.2.2.1 Residential Fuel 3-23
3.2.2.2 Commercial and Institutional Fuel 3-25
3.2.2.3 Industrial Fuel 3-25
xi
-------
Contents (continued)
Pa^e
3.2.3 Mobile Sources 3-26
3.2.3.1 Highway Vehicles 3-26
3.2.3.2 Off-Highway Vehicles 3-29
3.2.3.3 Railroad Locomotives 3-31
3.2.3.4 Aircraft 3-31
3.2.3.5 Marine Vessels 3-32
3.2.4 Solid Waste Disposal 3-33
3.2.4.1 On-Site Incineration 3-33
3.2.4.2 Open Burning 3-34
3.2.5 Miscellaneous Area Sources 3-35
3.2.5.1 Evaporative Losses From Gasoline
Marketing 3-35
3.2.5.2 Evaporative Losses From Organic
Solvent Consumption 3-36
3.2.5.3 Unpaved Roads 3-39
3.2.5.4 Unpaved Airstrips 3-39
3.2.5.5 Forest Wildfires 3-39
3.2.5.6 Managed Burning 3-40
3.2.5.7 Agricultural Burning 3-40
3.2.5.8 Structural Fires 3-41
3.2.6 Additional Area Sources. 3-41
3.2.6.1 Publicly-Owned Treatment Works
(POTWs ) 3-42
3.2.6.2 Hazardous Waste Treatment, Storage,
and Disposal Facilities (TSDFs) 3-43
3.2.6.3 Fugitive Emissions from Synthetic
Organic Chemical Manufacturing 3-43
3.2.6.4 Bulk Terminals and Bulk Plants 3-44
3.2.6.5 Fugitive Emissions from Petroleum
Refinery Operations 3-44
3.2.6.6 Process Emissions from Bakeries 3-44
3.2.6.7 Process Emissions from Pharmaceutical
Manufacturing 3-45
3.2.6.8 Process Emissions from Synthetic-
Fibers Manufacturing 3-45
3.2.6.9 Crude Oil Natural Gas Production
Fields 3-45
3.2.6.10 Cutback Asphalt Paving Operations 3-46
3.3 NONCRITERIA POLLUTANTS 3-46
3.3.1 Noncriteria Pollutant Inventory 3-46
3.3.2 Emissions Estimation Methods 3-47
3.3.2.1 Primary Sulfates 3-47
3.3.2.2 Hydrogen Chloride 3-43
3.3.2.3 Hydrogen Fluoride 3-50
3.3.2.4 Ammonia 3-50
References for Section 3 3-51
xii
-------
Contents (continued)
Page
4.0 QUALITY CONTROL FOR POINT AND AREA SOURCE DATA 4-1
4.1 BACKGROUND 4-1
4.1.1 EPA Policy 4-1
4.1.2 Emissions Inventory Quality Control 4-3
4.2 OBJECTIVES 4-5
4.2.1 Identification of Key Data Elements and Data
Quality Objectives 4-5
4.2.2 Identification of ProbLems in Existing State
Emissions Inventories 4-7
4.2.3 Standard Estimation Techniques 4-8
4.2.4 Emissions Estimation Procedures. 4-8
4.2.5 Utility Quality Control Checks 4-8
4.3 THE POINT SOURCE QC LOOP 4-9
4.3.1 Overview of QC Loop....... 4-9
4.3.2 State Level - Data Collection and Confirmation. 4-11
4.3.3 EPA Screening Level 4-12
4.3.3.1 NE061 Edit Checks 4-13
4.3.3.2 Completeness Checks. 4-14
4.3.3.3 DOE EIA-767 Utility Fuel Data
Comparison 4-16
4.3.3.4 Additional QC Checks 4-17
4.3.3.5 Audit Trail 4-18
4.3.4 QC Reports 4-19
4.4 RESULTS OF QC PROCEDURES 4-20
4.4.1 Analysis of Quality Control Result 4-20
4.4.2 Resolution of QC Problems 4-21
4.4.3 Remaining QC Questions 4-22
4.5 QA/QC OF AREA SOURCE AREA 4-23
4.5.1 Emission Factors 4-24
4.5.2 Activity Levels 4-24
4.5.3 Emissions 4-25
References for Section 4... 4-27
5.0 5.1 SUMMARY 5-1
5.2 COMPREHENSIVENESS AND QUALITY 5-1
5.2.1 State Participation for 1985 5-3
5.2.2 Emissions Sorting and Confirmation 5-5
5.2.3 Fuel Use 5-13
5.2.4 Data Quality 5-20
5.2.4.1 NEDS Edit Messages 5-20
5.2.4.2 Missing Data Items 5-21
5.2.5 Emissions by Estimation Method Code 5-26
xiii
-------
Contents (continued)
Page
5.3 EMISSIONS SUMMARY AND CONCLUSIONS 5-31
5.3.1 U.S. Emissions By Category and State 5-33
5.3.2 Emissions for Selected Emission Categories 5-33
5.3.2.1 Combustion Sources.. 5-34
5.3.2.2 Primary and Secondary Metals 5-40
5.3.2.3 VOC Emissions 5-40
5.3.2.4 Area Sources 5-47
5.3.2.5 Emissions by Stack. Height 5-48
5.3.2.6 Noncriteria Pollution Emissions 5-58
References for Section 5 5-60
APPENDICES
A. CORRESPONDENCE BETWEEN EPA ADMINISTRATOR AND ASSISTANT
ADMINISTRATOR TO EPA AND STATE PARTICIPANTS IN THE 1985
INVENTORY A-l
B. NE061 EDIT CHECKING MESSAGES B-l
C. STATE EMISSIONS SUMMARIES C-l
D. SCC EMISSIONS LISTING D-l
E. GLOSSARY E-l
xiv
-------
LIST OF TABLES
Number Page
1-1 1985 Stationary Point Source Profile (NEDS) 1-5
1-2 Total 1985 U.S. Emissions by Category for SO2, NOx
and VOC (103 Tons) 1-6
1-3 1985 U.S. Anthropogenic Emissions (Major Categories)..... 1-9
1-4 1985 SO2, NOx, and VOC Emissions (103 Tons) 1-14
1-5 1985 Point Source SO2, NOx, and VOC Emissions
(103 Tons) 1-15
1-6 1985 Area Source SO2, N0X, and VOC Emissions
(103 Tons) 1-16
1-7 Distribution of Emissions by Plant Emissions Size
Classes 1-18
1-8 Summary of 1985 SO2, N0X, and VOC Point Source Emissions
by Stack Height Categories (U.S. Total) 1-20
1-9 Percent Emissions by Year of Record (1000 TPY Plants).... 1-21
1-10 Missing Items for the 1985 NAPAP Emissions
Inventory (1000 Ton Plants, 25 Ton Points) 1-22
2-1 1985 Emissions Inventory Priorities 2-6
3-1 National Emissions Data System (NEDS) Data Structure 3-4
3-2 1985 NAPAP Area Source Categories 3-18
3-3 Area Source Organic Solvent User Categories 3-37
3-4 Area Source Organic Solvents 3-38
3-5 Emissions Sources of Primary Sulfate, Hydrogen Chloride,
Hydrogen Fluoride, and Ammonia in the NAPAP
Inventory 3-49
4-1 Quality Assurance/Quality Control Elements for
Engineering Research and Development Project Plans 4-2
xv
-------
LIST OF TABLES (continued)
Number Pa%e
4-2 Recommended Quality Assurance/Quality Control Elements
for Emissions Inventory Project Plans 4-4
4-3 NAPAP 1985 Emissions Inventory Data Quality Objectives.. 4-6
5-1 Percent Emissions by Year of Emissions
(1000 TPY Plants) 5-4
5-2 Emissions Sorting and Edit Checking Results by State
(Point Sources Only) 5-6
5-3 Coal Consumption Comparison for Utilities 5-15
5-4 Oil Consumption Comparison for Utilities 5-17
5-5 Natural Gas Consumption Comparison for Utilities 5-18
5-6 Comparison of NEDS and DOE 1985 National Fuel Use Totals. 5-19
5-7 Major Rejection and Warning Messages for
Plants Emitting 1000 Tons SO2, NOx, or VOC
(1985 NEDS) 5-22
5-8 Missing Items for the 1985 NAPAP Emissions
Inventory ( 1000 Ton Plants, 25 Ton Points) 5-24
5-9 Missing Items for the 1985 NAPAP Emissions
Inventory (All Points) 5-25
5-10 State SO2 Emissions Totals (10 Tons) by Estimation
Method Code 5-27
5-11 State NOx Emissions Totals (103 Tons) by Estimation
Method Code 5-28
5-12 State VOC Emissions Totals (103 Tons) by Estimation
Method Code 5-29
5-13 Emissions from Utility Boilers of SO2, NOx, and VOC by
State (103 Tons) 5-35
xvi
-------
LIST OF TABLES (continued)
Number Page
5-14 Emissions from Industrial Boilers of SO2, N0X, and
VOC by State (10 Tons) 5-36
5-15 Emissions from Commercial/Institutional Boilers of
SO2, NOx,and VOC by State (10 Tons) 5-37
5-16 1985 Utility Emissions (10^ Tons) of SO2 and N0X by Fuel
Type[[[ 5-39
5-17 Primary and Secondary Smelting SO2 Emissions
(10^ Tons) by State 5—A1
•>
5-18 VOC Point Source Emissions (10 Tons) by State and
Category........ 5-42
5-19 VOC Area Source Emissions (10"* Tons) by State and
Category. 5-43
5-20 VOC Emissions (10^ Tons) by State and Category (Point
and Area)......... 5-44
5-21 1985 State SO2 Emissions (10^ Tons) by Stack Height 5-49
5-22 1985 State NOx Emissions (10^ Tons) by Stack. Height 5-50
5-23 1985 State VOC Emissions (10"* Tons) by Stack. Height...... 5-51
-------
LIST OF FICURES
Number Page
1-1 Comparison of Point and Area Source SO2, N0X and
VOC Emissions 1-10
1-2 U.S. Anthropogenic Point Source Emissions 1-11
1-3 U.S. Anthropogenic Area Source Emissions 1-12
1-4 Distribution of 1985 SO2, N0X, and VOC Emissions
by Major Category 1-13
1-5 Distribution of Emissions by Plant Emissions
Classes 1-19
1-6 1985 Missing Point Source Data Summary (1000 Ton
SO2, N0X, or VOC Plants) 1-26
2-1 Acid Rain Precursors and Products 2-3
4-1 QC Loop for Point Source Emissions Data 4-10
5-1 Distribution of SO2, N0X( and VOC Emissions by Method
Code. 5-30
5-2 1985 Emissions by Boiler Category (SO2 and NOx) 5-38
5-3 1985 VOC Emissions by Major Category (U.S. Total) 5-45
5-4 SO2 Emissions by NEDS Stack. Height 5-52
5-5 NOx Emissions by NEDS Stack. Height 5-53
5-6 VOC Emissions by NEDS Stack. Height 5-54
5-7 Number of Stacks by Height Category for SO2 Emissions 5-55
5-8 Number of Stacks by Height Category for N0X Emissions 5-56
5-9 Number of Stacks by Height Category for VOC Emissions 5-57
xviii
-------
SECTION 1
EXECUTIVE SUMMARY
1.1 Background
The National Acid Precipitation Assessment Program (NAPAP) was established
by Congress in 1980 (Title VII of P.L. 96-294) to coordinate and expand
research on problems posed by acid deposition in and around the United States.
Among the contributors to acid deposition, anthropogenic emissions sources from
both the United States and Canada as well as natural sources are believed to be
of primary importance. This document covers U.S. anthropogenic point and area
source emissions.
A fundamental objective of NAPAP's research program is the investigation
of emissions sources that may contribute to acid deposition. NAPAP1s Task
Group on Emissions and Controls has achieved this objective by developing
historical and current inventories of acid deposition precursor emissions. The
Environmental Protection Agency's Office of Research and Development has the
responsibility for developing the 1985 NAPAP anthropogenic emissions inventory.
Emissions inventories are necessary to assess the impact of various source
types and characteristics on the emissions and abatement of acid precipitation
precursors; to investigate and verify atmospheric process models that simulate
source-receptor relationships; and to assess historical trends in emissions.
The objective of the 1985 NAPAF Emissions Inventory was to meet the needs
of both the NAPAP acid deposition assessment and atmospheric modeling teams.
These NAPAP users had three main requirements of the inventory: (1) that the
data base have a consistent baseline, so that anthropogenic emissions and
operating data reflect as accurately as possible a single and consistent year,
(2) that the data base be complete and that the data be of the highest possible
quality, and (3) that any problems or errors found in the data be accurately
tracked with an Audit Trail. In order to meet these objectives, it was
imperative that cooperation be established between the U.S. EPA, Environment
Canada, and the State air pollution control agencies. During the development
of the 1985 NAPAP Emissions Inventory, budgetary and scheduling constraints
resulted in a need to develop priorities in the inventory effort to ensure that
1-1
-------
Che information of most importance Co NAPAP would be obtained and that the
NAPAP objectives would be mec.
1.2 Methodology
The U.S. anthropogenic emissions inventory is divided inco two major
categories, point and area sources. Poinc sources have precise location data
and emit at least 100 tons per year (TPY) of a criteria pollutant (NOx, SCb,
TSP, VOC, or CO). Area sources comprise both mobile sources and point sources
too small and too numerous to list individually. Point source data are
supplied to EPA by the State agencies in an essentially bottom-up collection
stracegy. The area source emissions estimates are calculated by EPA using a
series of computer programs. This is primarily a cop-down scrategy which
allocates national emissions estimates to the State and county levels.
The data collection efforts for 1985 point source data were prioritized to
reflect the needs of the NAPAP research and assessment programs. Estimates of
SO2, N0XI and VOC emissions were given the highest priority. In addition, the
effort concentrated on facilities emicting at least 1000 TPY because they
represent 97, 90, and 61 percent of the point source SO2, N0X, and VOC
emissions, respectively. Of the 50 data elements in a NEDS record, che data
collection effort focused on the 14 items chat are most imporcanc for the NAPAP
community. These priority items include the annual emissions estimates for
SO2, NOx and VOC; the maximum design and annual operating rates; the Source
and Standard Industrial Classification codes (SCC and SIC); emissions control
equipment and efficiencies; fuel characteristics; stack parameters; location
data; and operating schedules. The final 1985 NAPAP Emissions Inventory will
contain estimates of emissions of five criteria pollutants, SO2, NOx, VOC, CO,
and TSP, and four other chemical species believed to play an important role in
acid deposition. These four other species are primary sulfate, hydrogen
chloride, hydrogen fluoride, and ammonia.
Meeting the objectives of the NAPAP program required the cooperation ot
many different agencies and organizations including the NAPAP Cask groups, the
State and Territorial Air Pollution Program Administrators (STAPPA), che U.S.
1-2
-------
State and Territorial Air Pollution Program Administration (STAPPA), the U.S.
and Regional EPA offices, and Che State air pollution control offices. STAPPA
began the collection effort with a survey to determine the capabilities and
needs of the State agencies to meet the objectives of the 1985 NAPAP Emissions
Inventory. Over the course of the study, EPA offered financial and technical
assistance to States in the development of their inventories. Workshops and
seminars were provided to acquaint State personnel with the priorities and
objectives of the NAPAP Emissions Inventory. In addition, Contractor
assistance was made available to collect, encode, and execute quality control
programs on the State point source data.
1.3 Quality Control
Quality control (QC) procedures consisting of manual and computerized
checking procedures were specifically designed for the 1985 NAPAP Emissions
Inventory. These procedures were designed to ensure that the quality of the
data met the requirements of the NAPAP community as closely as possible, given
the resource constraints of the inventory effort.
QC was conducted at all levels of the inventory development. A
computerized edit checking program (NEG61) was made available to the States,
Regional offices, and Contractors to help identify problems in the data while
the data were still in the hands of the people most knowledgeable about the
sources. As the data were entered into the NEDS, more checks were employed to
ensure that the data were within reasonable limits and were internally
consistent. For each State, a report was developed discussing possible
problems with the data submission and was returned to the State and Regional
offices to afford the people most knowledgeable about the data a chance to
correct errors and comment on the QC findings. The States returned the QC
reports with comments and corrected data. This effort was the first national
emissions inventory in which the data were actually returned to the responsible
agencies for their comments and corrections. As a result, the quality of this
annual inventory is better than that of any previously developed national
inventory.
1-3
-------
1.4
Summaries and Analyses of the 1985 NAPAP Emissions inveniory
The following cables and figures summarize current emissions data in the
1985 NAPAP Emissions Inventory. Table 1-1 gives a State-level profile of the
number of plants, points, and different processes (SCC occurrences). Table 1-2
presents an emissions summary for the three primary pollutants of concern,
broken down into point and area sources by source category. Table 1-3
aggregates these categories into six major categories. These tables clearly
show that over 90 percent of the sulfur dioxide emissions are from point
sources and that approximately 70 percent of the total sulfur dioxide emissions
are from electric utilities. Nitrogen oxide emissions are split between point
and area sources. The electric utilities represent approximately 30 percent of
the total N0X emissions or 70 percent of the point source N0X, while mobile
sources account for 43 percent of the total NOx or 80 percent of the area
source N0X. emissions are dominated by area source categories, with mobile
sources representing 33 percent of the total or 37 percent of the area source
VOC emissions. Solvent evaporation loss is the second largest source,
representing 21 percent of the total or 23 percent of the area source VOC
emissions.
The relative importance of point versus area sources for the three
pollutants is illustrated Figure 1-1. S02 emissions are dominated by point
sources, N0X emissions are almost evenly split between point and area sources,
and VOC emissions are dominated by area sources. Figure 1-2 shows that for
point sources, combustion by electric utilities dominates the S02 and N0X
emissions. Figure 1-3 shows that mobile souces dominate NOx emissions while
mobile sources and solvent evaporation loss (industrial processes) dominate VOC
emissions. Figures 1-4 through 1-6 reillustrate this information by pollutant
using pie charts.
Tables 1-4 through 1-6 break down the primary pollutants by State. The
first table lists total emissions nationwide by State. The second and third
tables illustrate point and area source emissions by State.
Table 1-7 and Figure 1-7 both demonstrate the distribution of point source
emissions by plant size. For S02, 81 percent of emissions are from facilities
emitting at least 10,000 tons per year. These facilities are primarily
1-4
-------
TABLE 1-1. 1985 STATIONARY POINT SOURCE PROFILE (NEDS)
STATE
PLANTS
POINTS
POINT-SCC
Alabama
171
1,100
1,371
Arizona
67
223
311
Arkansas
118
754
804
California
651
8,399
12,599
Colorado
97
380
442
Conneccicuc
55
572
722
Delaware
38
217
275
Disc, af Col.
13
53
121
Florida
127
1,107
1,654
Georgia
184
1,592
2,321
Idaho
23
84
85
Illinois
520
8,703
8,707
Indiana
366
2,211
3,818
Iowa
64
2,622
3,039
Kansas
153
998
1,699
Kentucky
224
627
943
Louisiana
315
7,808
8,976
Maine
59
245
283
Maryland
124
1,323
1,777
Massachusetts
186
2,269
3,440
Michigan
296
3,431
4,114
Minnesota
217
2,817
4,975
Mississippi
130
282
468
Missouri
259
2,466
3,212
Montana
28
385
485
Nebraska
41
153
175
Nevada
22
126
188
New Hampshire
31
194
259
New Jersey
191
2,961
2,971
New Mexico
183
899
900
New York
383
1,469
2,125
North Carolina
348
2,543
3,078
North Dakota
44
111
121
Ohio
430
2,726
3,544
Oklahoma
212
1,967
2,504
Oregon
51
110
151
Pennsylvania
552
3,024
3,987
Rhode Island
24
105
176
South Carolina
126
517
704
South Dakota
18
30
71
Tennessee
298
4,565
5,696
Texas
859
27,657
31,426
Utah
64
213
355
Vermont
17
104
117
Virginia
260
1,706
2,749
Washington
181
1,008
1,520
West Virginia
184
6,555
6,767
Wisconsin
224
1,760
2,649
Wyoming
108
482
577
TOTAL
9,336
111,653
139,451
1-5
-------
3
TABLE 1-2. 1985 U.S. EMISSIONS BY CATEGORY FOR SO2, NOx, AND VOC (10 TONS)
EMISSION CATEGORIES
SO2
AREA
NOx
VOC
S02
POINT
N0X
VOC
FUEL COMBUSTION 1
,070.4
2,022 .2
2,433.7
I 18,132.9
8,570.4
219.7
External Combustion 1
,070.4
2,022.2
2,433.7
118,086.6
7,833.9
136.8
Residential
178.2
407.1
2,395.3
I 0.0
0.0
0.0
Anthracite Coal
7.7
0.8
2.8
I 0.0
0.0
0.0
Bituminous Coal
29.5
2.1
7.5
I 0.0
0.0
0.0
Distillate Oil......
128.2
75.0
2.9
I 0.0
0.0
0.0
Residual Oil
0.5
0.1
0.0
I 0.0
0.0
0.0
Natural Gas
1.2
248.3
13.4
I 0.0
0.0
0.0
Wood
11.0
80.7
2,368.8
I 0.0
0.0
0.0
Electric Generation...
0.0
0.0
0.0
116,166.9
6,701.1
54.0
Anthracite Coal
0.0
0.0
0.0
| 22.4
14.8
0.3
Bituminous Coal
0.0
0.0
0.0
114,835.3
5 ,602.6
35.1
Lignite* •. . . • • - • • • » .
0.0
0.0
0.0
j 627.8
394.7
6.2
Residual Oil
0.0
0.0
0.0
I 597.5
187.2
2.8
Distillate Oil
0.0
0.0
0.0
I 56.0
33.9
0.5
Natural Gas
0.0
0.0
0.0
I 6.5
448.3
7.0
Process Gas
0.0
0.0
0.0
I 0.2
11.5
0.1
Other
0.0
0.0
0.0
I 21.3
8.1
2.1
Industrial
662.4
1,404.1
30.9
I 1,750.3
1,061.2
79.8
Anthracite Coal
0.1
0.1
0.0
I 11.1
2.9
0.0
Bituminous Coal
353.3
130.5
0.6
I 1,080.8
384.2
4.3
Ligni te
0.0
0.0
0.0
| 48.5
9.5
0.2
Residual Oil
236.0
46.5
0.2
! 396.9
138.6
4.3
Distillate Oil
54.7
49.3
1.6
I 19.5
23.3
0.7
Natural Gas
0.7
1,176.9
28.3
i 38.9
296.5
8.5
Process Gas
0.1
0.1
0.0
| 105.4
92.8
3.8
Coke
0.2
0.0
0.0
I 11.9
6.8
0.1
Wood
17.3
0.3
0.2
1 15.1
84.1
49.8
LPG
0.0
0.0
0.0
j 0.0
0.7
0.0
Bagasse
0.0
0.0
0.0
| 0.2
1.7
2.6
Other
0.0
0.0
0.0
| 22.0
20.2
5.3
Commercial/Institution
229.9
211.1
7.5
I 169.4
71.6
3.0
Anthracite Coal
15.5
5.5
0.0
1 3.4
1.2
0.1
Bituminous Coal.....
24.5
6.4
0.2
| 114.6
23.4
0.7
Lignite.
0.0
0.0
0.0
| 1.0
0.5
0.2
Residual Oil
103.6
31.4
0.6
| 43.4
16.4
0.5
Distillate Oil
85.7
51.9
0.7
I 5.7
3.0
0.1
Natural Gas
0.5
115.9
6.0
j 0.7
22.1
0.3
Wood
0.0
0.0
0.0
| 0.1
2.8
0.5
LPG
0.0
0.0
0.0
I 0.0
0.0
0.0
Other.
0.0
0.0
0.0
j 0.5
2.3
0.6
1-6
-------
TABLE 1-2. (continued)
EMISSION CATEGORIES
SO2
AREA
NOx
VOC
SO 2
POINT
NOx
VOC
Internal Combustion ....
O
«
O
0.0
0.0
46.2
736.6
82.9
Electric Generation...
0.0
0.0
0.0
11.0
68.5
2.9
Distillate Oil
0.0
0.0
0.0
10.0
15.4
0.6
Natural Gas
0.0
0.0
0.0
1.1
53.1
2.3
Industrial
0.0
0.0
0.0
32.9
644.5
73.2
Distillate Oil
0.0
0.0
0.0
0.4
5.4
0.4
Natural Gas
0.0
0.0
0.0
31.0
633.9
71.8
Gasoline
0
•
0
0.0
0.0
0.0
0.1
0.2
Diesel Fuel
0.0
0.0
0.0
0.1
2.0
0.1
Other
0.0
0.0
0.0
1.3
3.1
0.7
Commercial/Institution
0.0
0.0
0.0
1.9
17.8
3.3
Engine Testing
0
•
0
0.0
0.0
0.4
5.7
3.5
INDUSTRIAL PROCESS
0.0
0.0
0.0
2,906.9
923.0
2,138.2
Chemical Manufacturing..
0.0
0.0
0.0
501.6
171.7
518.3
Food/Agriculture
0.0
0.0
0.0
3.4
4.8
50.5
Primary Metals
0.0
0.0
0.0
976.2
54.2
55.6
Secondary Metals
0.0
0.0
0.0
50.1
20.6
18.4
Mineral Products
0.0
0.0
0.0
289.3
240.8
17.3
Petroleum Industry
0.0
0.0
0.0
788.5
288.5
231.3
Wood Products
0.0
0.0
0.0
130.6
73.5
45.6
Organic Solvent Evap....
0.0
0.0
0.0
0.9
8.3
825 .2
Petroleum Storage/Trans.
0.0
0.0
0.0
1.5
1.2
290.8
Metal/Fabrication
0.0
0.0
0.0
0.3
1.9
3.3
Textile Manufacturing...
0.0
0.0
0.0
0.0
0.1
8.6
Other/Not Classified....
0.0
0.0
0.0
164.6
57.2
73.2
SOLID WASTE DISPOSAL
14.6
69.2
608.8
20.1
18.3
11.1
Government
0.0
0.0
0.0
5.7
7.8
5.9
Municipal Incineration
0.0
0.0
0.0
5.6
6.4
5.2
Open Burning
0.0
0.0
0.0
0.0
0.4
0.0
Other Incineration....
0.0
0.0
0.0
0.1
0.9
0.7
Residential
9.6
60.3
590.7
0.0
0.0
0.0
On-site Incineration..
1.4
3.0
288.5
0.0
0.0
0.0
Open Burning
8.3
57.3
302.2
0.0
0.0
0.0
Commercial/Institutional
4.6
7.1
9.8
3.6
5.5
1.0
On-site Incineration..
4.6
6.9
8.4
3.6
5.5
1.0
Open Burning
0.0
0.2
1.4
0.0
0.0
0.0
Industrial
0.4
1.7
8.4
10.7
5.0
4.2
On-site Incineration..
0.3
0.3
2.6
10.7
4.9
4.1
Open Burning
0.1
1.4
5.8
0.0
0.1
0.1
1-7
-------
TABLE 1-2. (continued)
EMISSION CATEGORIES
(M
O
to
AREA
NOx
VOC
SO2
POINT
NOx
VOC
TRANSPORTATION
863.6
8,834.4
7,287.7
I 0.0
0.0
0.0
Land Vehicles
690.9
8,549.2
6,667.5
I 0.0
0.0
0.0
Gasoline
274.A
5,139.4
6,102.9
I 0.0
0.0
0.0
Light Duty Vehicles.
185.2
3,368.3
3,643.7
j 0.0
0.0
0.0
Light Duty Trucks...
69.0
1,320.3
1,538.1
1 0.0
0.0
0.0
Heavy Duty Vehicles.
14.1
297 .2
425 .1
0.0
0.0
0.0
Off-Highway
6.2
153.5
496.0
1 0.0
0.0
0.0
Diesel Fuel
416.5
3,409.8
564.6
1 0.0
0.0
0.0
Heavy Duty Vehicles.
242.9
1,825.2
259.7
I 0.0
0.0
0.0
Off-Highway
82.6
994.2
161.6
1 0.0
0.0
0.0
Rail
91.0
590.4
143.3
1 0.0
0.0
0.0
Aircraft
14.1
125.6
173.2
1 0.0
0.0
0.0
Military
4.8
37.4
87.1
j 0.0
0.0
0.0
Civil
1.0
10.5
29.6
1 0.0
0.0
0.0
Commercial
8.3
77.8
56.6
1 0.0
0.0
0.0
Vessels
158.5
159.6
447.0
1 0.0
0.0
0.0
Bituminous Coal
3.6
0.2
0.6
1 0.0
0.0
0.0
Diesel
15.8
118.2
29.6
I 0.0
0.0
0.0
Residual Oil
135.8
22.3
1.3
I 0.0
0.0
0.0
Gasoline
3.3
18.9
415.5
I 0.0
0.0
0.0
MISCELLANEOUS AREA
4.2
129.4
6,184.2
1 0.0
0.0
0.0
Forest Fires
1.2
33.6
161.5
I 0.0
0.0
0.0
Forest Managed Burning..
3.0
82.0
293.1
1 0.0
0.0
0.0
Agricultural Burning....
0.0
8.4
55.0
1 0.0
0.0
0.0
Structural Fires........
0.0
5.4
44.4
I 0.0
0.0
0.0
Gasoline Stn. Evap. Loss
0.0
0.0
999.0
1 0.0
0.0
0.0
So1vent Evap. Loss......
0.0
0.0
4,631.3
1 0.0
0.0
0.0
ADDITIONAL AREA
0.0
0.0
3,385.8
1 0.0
0.0
0.0
POTWs
0.0
0.0
25.4
1 0.0
0.0
0.0
Cutback. A
0.0
0.0
191.9
1 0.0
0.0
0.0
SOCMI Fugitives
0.0
0.0
164.5
I 0.0
0.0
0.0
Bulk Terminals/Plants...
0.0
0.0
398.0
I 0.0
0.0
0.0
Refinery Fugitives
0.0
0.0
762.4
I 0.0
0.0
0.0
Bakeries
0.0
0.0
50.0
I 0.0
0.0
0.0
Pharmaceutical Mfg
0.0
0.0
33.6
I 0.0
0.0
0.0
Synthetic Fiber Mfg
0.0
0.0
160.9
I 0.0
0.0
0.0
Oil/Natural Gas Fields..
0.0
0.0
194.0
1 0.0
0.0
0.0
TSDFs
0.0
0.0
1,405.1
I 0.0
0.0
0.0
GRAND TOTAL 1
952.9
11,055.3
19,900.3
121,059.8
9,511.6
2,369.0
Based on the 48 contiguous States and the District of Columbia
1-8
-------
TABLE 1-3. 1985 U.S. ANTHROPOGENIC EMISSIONS (MAJOR CATEGORIES)
EMISSIONS (103
ton:; )
CATEGORY
TYPE
S02
NOx
VOC
Utility Combustion (UC)
Point
16,177.9
6,769.6
56.9
Industrial Combustion (IC)
Area
Point
662.4
1,783.2
1,404.1
1,705.7
30.9
153.0
Subtotal
2,445.6
3,109.8
183.9
Other Combustion (OC)
Area
Point
408.1
171.7
618.2
95.1
2,402.8
9.8
Subtotal
579.8
713.2
2,412.6
Industrial Process (IP)
Area*
Point
0.0
2,906.9
0.0
923.0
6,200.7
2,138.2
Subtotal
2,906.9
923.0
8,338.9
Transportation (TRAN)
Area
863.6
8,834.4
7,287.7
Other (OTH)
Area
Point
18.9
20.1
198.6
18.3
3,978.2
11.1
Subtotal
38.9
216.8
3,989.3
GRAND TOTAL
23,012.7
20,566.9
22,269.3
* For this analysis, certain area source emissions contained within the
the Miscellaneous and Additional Area Sources have been compared to
the Industrial Process point sources; these area sources include solvent
evaporation loss, synthetic organic chemical manufacturing, bulk,
plants, refinery fugitives, bakeries, pharmaceutical manufacturing,
and synthetic fiber manufacturing.
1-9
-------
so2
1771 POINT SOURCE
N0X
POLLUTANT
VOC
iY\] AREA SOURCE
Figure 1—1. Comparison of point and area source
S02, NOx, and VOC emissions.
1-10
-------
voc
OTH
SO2
1771 uc 1X3 ic
NO,
POLLUTANT
m 'p
ES oc K3
UC = Utility Combustion
IC = Industrial Combustion
IP = Industrial Processes
OC = Other Combustion
OTH = Other Sources
Figure 1-2. U.S. anthropogenic point source emissions.
1-11
-------
19
18 -
17 -
16 -
15 -
14 -
r\
13 -
C
0
12 -
+J
ifl
11 -
0
r~
10 -
(A
9 -
Z
0
8 -
W
(0
7-
i
iii
6 -
5 -
4 -
3 -
2 -
1-
o -L
A
so2
1ZZ3 tran El ip
N0X
POLLUTANT
m IC
voc
ES3 oc M OTH
Tran = Transportation
IP = Industrial Processes*
IC = Internal Combustion
OC = Other Combustion
OTH = Other Sources
For this analysis, some Miscellaneous and Additional area
source categories as listed in Table 1-3 have been included
under Industrial Processes; this is not a NEDS area source
category.
Figure 1-3. U.S. anthropogenic area source emissions.
1-12
-------
S02
NOx
OIHEft COMB (2-3%)
1HDUST PROCESS (12.5X)
IMDUST COMB (10.8.1)
TRANSPORT (3.ax)
OltlEfl (0.2%)
I
h-»
LO
WDUSTPRoSFW"**
IHDUST COMB (13.2%)
UI1UIY (70.2X)
TRANSPORT (42.9X)
VOC
UTIUTY (32.9n)
OTHER (1.18)
OTHER COMB (10.8%) UTMJIY (OJX)
OTHER (17 JX)
IHDUST PROCESS (37.4%)
TRANSPORT (32.7X)
IHDUST COMB (0.9%)
Figure 1-4. Distribution of 1985 S02, NOx, and VOC emissions by major category.
-------
TABLE 1-4. 1985 TOTAL S02, NOXJ AND VOC EMISSIONS (103 TONS)
STATE
SO 2
NO x
VOC
Alabama
720 .6
469.3
510.4
Arizona
699.4
245 .5
225 .3
Arkansas
109.5
238.0
223.6
California
225 .2
1,244.6
2,111.7
Colorado
91.4
291.8
294.2
Connecticut
87.3
127.9
232.9
Delaware
117.6
68.5
68.3
Dist. of Col.
7.4
20.4
28.2
Florida
664.1
681.8
737 .3
Georgia
1,111.5
590.7
587.6
Idaho
36.8
87.8
187.9
111inoi s
1,397.8
971.6
957 .1
Indiana
1,864.2
880.7
573 .7
Iowa
289.7
265 .2
205 .5
Kansas
160.5
443.3
244.2
Kentucky
879.4
520.5
425 .8
Loui siana
402.3
763.4
646.1
Maine
79.2
66.8
171.1
Maryland
307.4
281.7
257 .4
Massachusetts
328.3
277.0
418.7
Michigan
550.6
658.4
789.0
Minnesota
174.8
355 .9
430.6
Mississippi
174.6
238.8
275 .1
Missouri
1,163.8
536.2
507 .8
Montana
91.1
155.4
166.3
Nebraska
65.3
165.3
120.5
Nevada
40.3
112.7
76.2
New Hampshire
85.3
54.6
85.4
New Jersey
183.4
369.0
585 .6
New Mexico
271.4
284.8
171.7
New York
665 .1
626.3
987 .7
North Carolina
484.8
501.1
678.7
North Dakota
221.3
183.2
60.8
Ohio
2,560.8
1,018.8
1,025.9
Oklahoma
150.1
396.4
357 .8
Oregon
44.6
170.5
324.5
Pennsylvania
1,425.0
958.0
925 .1
Rhode Island
9.2
30.3
67.3
South Carolina
236 .8
259.1
355 .7
South Dakota
43.2
72.7
84.0
Tennessee
977 .3
511.4
554.9
Texas
1,476.3
2,477.6
2,324.1
Utah
71.8
170.7
165.1
Vermont
7.3
25.2
52.5
Virginia
311.5
390.1
577 .9
Washington
168.5
274.9
479.5
West Virginia
1,058.4
461.4
410.7
Wi scons in
514.4
348.9
431.9
Wyoming
205 .8
222.4
90.3
TOTAL
23,012.7
20,566.9
22,269.3
1-14
-------
TABLE 1-5. 1985 POINT SOURCE SO2, NOx, AND VOC EMISSIONS (103 TONS)
STATE
so 2
NO x
VOC
Alabama
687 .1
264.2
74.3
Arizona
674.6
75.4
3.7
Arkansas
83.3
71.0
22.8
California
81.2
230.8
97.1
Colorado
76.3
133.0
6.1
Connecticut
71.2
29.3
4.4
Delaware
113.8
42.0
7.8
Dist. of Col.
3.7
1.7
0.6
Florida
613.5
290.6
17.0
Georgia
1,084.5
292.5
35.8
Idaho
24.3
6.4
1.1
111inois
1,361.6
565.4
163.9
Indiana
1,703.7
571.1
95.1
Iowa
272 .0
99.2
7.3
Kansas
144.8
240.3
32.5
Kentucky
827.3
288.9
58.2
Louisiana
260.2
432.3
128.6
Maine
67.1
19.4
6.0
Maryland
254.0
95.8
13.9
Massachusetts
281.5
98.4
37.5
Michigan
519.2
318.5
94.4
Minnesota
154.7
154.5
49.3
Mississippi
158.9
74.5
36.2
Missouri
1,112.3
292.5
84.3
Montana
77.8
39.8
7.6
Nebraska
53.2
40.2
4.1
Nevada
34.3
62.5
0.6
New Hampshire
80.4
24.0
4.3
New Jersey
132.9
114.1
42.0
New Mexico
249.6
184.2
21.5
New York
571.3
183.5
53.2
North Carolina
436.0
233.9
74.1
North Dakota
200.4
115.9
2.9
Ohio
2,460.8
577.1
95.0
Oklahoma
118.4
155.1
31.8
Oregon
12.2
10.9
12.7
Pennsylvania
1,358.1
550.1
75.6
Rhode Island
4.0
2.7
6.0
South Carolina
219.6
118.1
25.9
South Dakota
36.5
17.1
5.6
Tennessee
945.6
277 .6
96.9
Texas
1,255.0
1,163.4
450.4
Utah
53.1
86.8
16.1
Vermont
2.6
1.9
1.8
Virginia
262 .8
139.0
86.6
Washington
137.4
73.3
29.6
West Virginia
1,049.5
381.7
78.9
Wisconsin
493.0
149.5
49.9
Wyoming
184.6
121.2
17.8
TOTAL
21,059.8
9,511.6
2,369.0
1-15
-------
TABLE 1-6. 1985 AREA SOURCE SO2. N°x> AND VOC EMISSIONS (103 TONS)
STATE
SO 2
X
0
7.
VOC
Alabama
33.5
205 .1
436.0
Arizona
24.9
170.0
221.6
Arkansas
26.2
167.0
200.3
Cal if ornia-"
144.0
1,013.8
2,014.6
Colorado
15.1
158.8
228.1
Connecticut
16.1
98.6
228.4
Delaware
3.8
26.5
60.5
Dist. of Col.
3.6
18.3
27.5
Florida
50.6
391.2
720 .3
Georgia
27.1
298.2
551.3
Idaho
12.5
81.3
186.9
Illinois
36.2
406.2
793.1
Indiana
160.5
309.6
478.6
Iowa
17.6
165.9
198.2
Kansas
15.7
203.0
211.7
Kentucky
52.2
231.5
367 .6
Louis iana
142.1
331.1
517.4
Maine
12.1
47.4
165.1
Maryland
53.4
185.9
243.6
Massachusetts
46.9
178.6
381.2
Michigan
31.4
339.3
694.5
Minnesota
20.1
201.4
381.3
Mississippi
15.7
164.3
238.9
Missouri
51.5
243.3
423.5
Montana
13.3
115.6
158.7
Nebraska
12.0
125.1
116.4
Nevada
6.0
50.2
75.7
New Hampshire
4.9
30.6
81.0
New Jersey
50.5
254.9
543.6
New Mexico
21.3
100.7
150.2
New York
93.8
442.8
934.5
North Carolina
48.7
267 .2
604.5
North Dakota
20.9
67.3
57.9
Ohio
100.0
441.3
930.9
Oklahoma
31.3
241.3
326.0
Oregon
32.4
159.7
311.3
Pennsylvania
66.9
407.9
849.4
Rhode Island
5.2
27.7
61.3
Souch CaroLina
17.2
140.9
329.3
South Dakota
6.7
55.6
78.4
Tennessee
31.7
233.3
457 .9
Texas
221.2
1,314.1
1,873.3
Utah
18.7
83.9
149.0
Vermont
4.8
23.3
50.7
Virginia
48.7
251.2
491.2
Washington
31.1
201.6
449.9
West Virginia
8.9
79.7
331.9
Wisconsin
21.4
199.4
331.9
Wyoming
21.2
101.1
72.5
TOTAL
1,952.9
11,055.3
19,900.3
* California maintains an extensive area source inventory that is
not compatible with the NEDS. California area source totaLs from
their State system are 87 x 10* TPY SO2, 826 x 103 TPY NO*, and
1,154 x 103 TPY VOC.
1-16
-------
Table 1-7 and Figure 1-5 both demonstrate the distribution of point source
emissions by plant size. For SO2, 81 percent of emissions are from facilities
emitting at least 10,000 tons per year. These facilities are primarily
utilities and smelters. For NOx, only 60 percent of emissions are from
facilities emitting at least 10,000 tons per year. Again, these facilities are
primarily utilities and large industrial sources. For VOCs, the distribution
according to source size is far more uniform, reflecting the finding that VOC
point source emissions originate in a wide variety of industrial processes.
Table 1-8 presents the relationship between point source emissions and
stack, height. Utility emissions of SO2 and NOx predominate the point source
category; these boilers are typically associated with large stacks. Host
point source VOC emissions, however, are contributed by industrial processes
which typically vent emissions near ground level.
As is seen in Table 1-9, the 1985 NEDS inventory was largely successful in
achieving its goal of a consistent baseline year. With few exceptions, the
data represent 1985 emissions estimates and 1985 operating data. The second
major objective was to compile a complete inventory of the data items that are
most important to the NAPAP community. Table 1-10 illustrates that while some
data elements are still missing in the 1985 NAPAP. Emissions Inventory, the
coverage of the data is excellent for large sources. These remaining missing
items represent three categories: data intentionally left blank by the States,
data not applicable to a particular source, or data which were not available.
The first case includes emissions estimates which are designated by the States
to be calculated by the NEDS software using State-supplied throughputs and
standard emission factors. Inapplicable data include items such as blank stack
heights for ground level sources (e.g., fugitives). As explained in Section 4,
these data have been reviewed using computerized and manual checks at the
State, Regional, and National levels. Large sources of SO2, N0X, and VOC have
received the most scrutiny and most remaining missing items represent the first
and second categories of missing data. However, several data elements, such as
operating rate and maximum design rate, were specifically excluded from some
States' requirements due to confidentiality restrictions, or because the
existing State inventories lacked mechanisms to include the data.
1-17
-------
TABLE 1-7. DISTRIBUTION OF EMISSIONS BY PLANT EMISSIONS SIZE CLASSES
POLLUTANT RANGE NUMBER OF EMISSIONS PERCENT OF
(TONS) PLANTS (103 TONS) EMISSIONS
>10,000
357
17,116
81%
5,000 - 10,000
216
1,548
72
2,500 - 5,000
264
932
51
1,000 - 2,500
457
746
A2
500 - 1,000
449
316
12
100 - 500
1,522
359
22
1 - 100
4,836
43
02
21,060
1002
>10,000
224
5,632
602
5,000 - 10,000
173
1,174
122
2,500 - 5,000
248
873
92
1,000 - 2,500
549
86 7
92
500 - 1,000
564
403
42
100 - 500
2,026
473
52
1 - 100
4,317
90
12
9,512
1002
>10,000
17
231
112
5,000 - 10,000
44
299
132
2,500 - 5,000
120
416
172
1,000 - 2,500
317
492
212
500 - 1,000
473
327
132
100 - 500
2,355
518
212
1 - 100
4,775
86
42
2,369
1002
1-18
-------
100
90
80 H
70
60
50
40 -
30 -
20 -
10
71
'A
rn\
>10,000 5,000-10,0002,500-5,000 1,000-2,500 500-1,000 100-500 1-100
PLANT EMISSIONS RANGE (TONS)
1771 so2 KS N0X voc
Figure 1-5. Distribution of emissions by plant
emissions magnitude classes.
1-19
-------
TABLE 1-8. SUMMARY OF 1985 S02, NOx, AND VOC POINT SOURCE
EMISSIONS BY STACK HEICHT CATEGORIES (U.S. TOTAL)
STACK HEIGHT RANGE
(Feet)
PERCENTAGE
so2
OF EMISSIONS
NOx VOC
0 - 120
7.3
19.1
88.0
121 - 240
12.8
14.6
8.5
241-480
24.2
22.7
2.3
>480
55.7
43.6
1.2
EMISSIONS*
20,974 9
,474
2,374
(103 Tons)
* Do not equal total point source emissions because some point
sources do not have stacks (e.g., process fugitive emissions)
1-20
-------
TABLE 1-9. PERCENT EMISSIONS BY YEAR OF EMISSIONS (:OOO TPY PLANTS)
S02 NO x VOC
STATE 1985 1982 1985 1985 1984
Alabama
1002
1002
1002
Arizona
1002
1002
1002
Arkansas
100%
1002
100%
California
100%
1002
1002
Colorado
100%
1002
1002
Connecticut
100%
100%
1002
Delaware
100%
100%
1002
Dist. of Col.
100%
100%
1002
Florida
98% 2%
100%
1002
Georgia
100%
100%
1002
Idaho
100%
100%
1002
Illinois
100%
100%
1002
Indiana
100%
100%
1002
Iowa
100%
100%
1002
Kansas
100%
100%
1002
Kentucky
100%
100%
1002
Louisiana
100%
100%
1002
Maine
100%
100%
1002
Maryland
100%
100%
1002
Massachusetts
100%
100%
1002
Michigan
100%
100%
1002
Minnesota
100%
1002
1002
Mississippi
100%
1002
1002
Missouri
100%
1002
1002
Montana
100%
100%
1002
Nebraska
100%
1002
1002
Nevada
100%
1002
1002
New Hampshire
100%
100%
100%
New Jersey
100%
1002
1002
New Mexico
100%
1002
1002
New York
100%
1002
1002
North Carolina
100%
100%
100%
North Dakota
100%
100%
100%
Ohio
100%
100%
1002
Oklahoma
100%
100%
1002
Oregon
1002
100%
1002
Pennsylvania
100%
100%
1002
Rhode Island
100%
1002
100%
South Carolina
100%
1002
1002
South Dakota
1002
1002
1002
Tennessee
100%
1002
100%
Texas
100%
1002
1002
Utah
100%
1002
100%
Vermont
1002
1002
100%
Virginia
1002
100%
100%
Washington
1002
100%
1002
West Virginia*
1002
1002
832
Wisconsin
1002
1002
1002
Wyoming
1002
1002
100%
* Submitted 1984 data for certain VOC sources under agreement with EPA
1-21
-------
TABLE 1-10. MISSING ITEMS FOR THE 1985 NAPAP EMISSIONS
INVENTORY (1000 TON PLANTS, 25 TON POINTS)
NAPAP DATA ITEMS
MISSINC
ITEMS
1985*
Percent "*
UTM Zone
B
0.42
UTM Coordinates
4
0.2%
SIC
86
0.32
Stack, Height
158
0.52
Stack. Diameter
178
0.6 2
Stack. Temperature
726
2.4 2
Stack. Gas Flow Rate
220
0.72
Plume Height
158
0.52
BoiLer Capacity
690
3.82
Primary Control Equipment#
18,332
19.92
(S02, NO„, VOC)
Secondary Control Equipment#
19,958
21.62
(S02, NOx, VOC)
Control Efficiency#
16,939
18.4%
(S02, NOx, VOC)
Percent Annual Throughput
172
0.62
Complete Normal Operating Schedule
704
2.32
Emissions Estimate*
0
0.02
(S02, N0X, VOC)
Estimation Method
6,409
6.92
(S02, NOx, VOC)
Year of Record (Emissions)
0
0.02
Annual Operating Rate
1,316
5.92
Maximum Design Rate
1,527
6.92
Fuel Sulfur Content
807
3.62
Fuel Ash Content
1,421
6.42
Fuel Heat Content
1,463
6.6%
Based on 48 contiguous States and the District of CoLumbia
* Excludes exempted operating data from Texas for 1985
+ Excludes emissions estimates that are calculated by NEDS
# Within NEDS, States may report control equipment as blank to
indicate that no information is available to the State or
that the status of the control equipment is uncertain; this
situation overestimates the number of items actually missed
by the inventory process.
1-22
-------
1.5 Documentation
An extensive Audit Trail was developed to document both problems with the
data and the history of the data. All the results of the Quality Control
measures performed on the data have been documented in memos kept in a QC
notebook. When the problems resulted in corrective action, the changes to the
data were listed in computerized files called the Audit Trail. The Audit
Trail lists the source identifiers, the old and new data values, the date of
the change, and the party responsible for the change.
Data tapes containing the final 1985 U.S. anthropogenic point and area
source data are available. These tapes contain a more comprehensive collection
of data than is described in this report. The Audit Trail files are also
available to users of the data to help them identify the source of a data
element. The QC notebook is part of the project docket.
1-23
-------
SECTION 2
INTRODUCTION
2.1 EMISSIONS RESEARCH AND THE NATIONAL ACID PRECIPITATION
ASSESSMENT PROGRAM
The National Acid Precipitation Assessment Program (NAPAP) was established
by Congress in 1980 (Title VII of P.L. 96-294) to coordinate and expand
research on problems posed by acid deposition in and around the United States.
A fundamental objective of NAPAP*s research program is the investigation of
emissions sources that may contribute to acid deposition.
NAPAP's Task Group on Emissions and Controls has achieved this objective
by developing historical and current inventories of acid deposition precursor
emissions. Information about historical trends in emissions is required to
analyze long-term trends in precipitation acidity and dry deposition and to
study these deposition effects on forest, aquatic, agricultural, and material
resources. Current emissions inventories are required to assess the impact of
various source types and characteristics on the emissions and abatement of acid
precipitation precursors. Inventories of current emissions are also required
to investigate and verify atmospheric process models that simulate
source-receptor relationships.
The development of current emissions inventories requires the
investigation of literally hundreds of thousands of sources of air pollution.
These include stationary or point sources such as refineries and utility
boilers as well as area or dispersed sources such as motor vehicle emissions
along highways. The analysis of these sources must include the calculation of
emissions and documentation of those engineering parameters which affect the
atmospheric transport of emissions, such as stack height. Because transport of
pollutants plays a critical role in acid deposition processes, emissions
inventories must encompass geographic areas larger than those in which
acidification and deposition are thought to occur. Thus, the investigation of
acid precipitation in the northeastern United States and Canada requires the
development of emissions inventories for all States and Provinces in both
countries.
2-1
-------
Assembling current emissions inventories across such a broad geopolitical
scale has required careful planning and coordination among the principal
agencies responsible for inventory development. This cooperation was
explicitly anticipated in Public Law 96-294, Sections 7OA(b)(1) and (11), which
described the need for joint research in the States and in interested nations
such as Canada.
In the United States, the Environmental Protection Agency's (EPA's) Office
of Research and Development has the responsibility for completing the
1985 NAPAP Emissions Inventory for NAPAP1s Task Group on Emissions and
Controls. To accomplish this objective, EPA has worked closely with both State
air pollution control agencies and the State and Territorial Air Pollution
Program Administrators to plan, fund, assemble, and ensure the quality of the
1985 inventory data. The EPA and States have the authority to collect required
emissions data under Title 40 of the Code of Federal Regulations, Part 51,
Section 51.321 - 51.323.
In Canada, Environment Canada has the responsibility for developing
national emissions inventories. In a manner analogous to the State-EPA
partnership in the United States, Environment Canada works with Provincial air
pollution control programs to collect emissions data for Canadian industries.
The individuals identified in the acknowledgement in the preface to this report
represent only part of the team of scientists and engineers in the many
agencies that cooperated to develop the 1985 NAPAP Emissions Inventory.
2.2 1985 EMISSIONS INVENTORY OBJECTIVES
The goals and specifications for the 1985 NAPAP Emissions Inventory were
developed to meet NAPAP acid deposition assessment and atmospheric modeling
research objectives. Figure 2-1 summarizes key chemical reactions in the
complex acid deposition process. These indicate that not only SO2 and NOx, but
also oxidants such as ozone and hydrogen peroxide are required for the
production of acids which cause acid precipitation. For example, sulfuric acid
is formed by the oxidation of sulfur dioxide by ozone and by hydrogen peroxide
in acidic clouds. Nitric acid is formed by the oxidation of nitrogen dioxide
with hydroxyl radicals, which in turn are formed from volatile organic
2-2
-------
so,
^2^2 an<* ^5 ^'n c'ou^s)
OH + Qj (In air)
Oxidants (wet surfaces)
bLSO . (sulfuric acid)
N>
I
to
NO.
NOx + VOC
VOC
Sunlight—~OH (In air)
Sunlight (In air)
Sunlight—~H02 (In air)
HNO^ (nitric acid)
O3 (ozone)
HoOo (hydrogen peroxide)
Figure 2-1. Acid rain precursors and products.
-------
compounds and sunlight. Ozone is formed when nitrogen oxides and volatile
organic compounds react in the presence of sunlight. Hydrogen peroxide is
formed by volatile organic compounds reacting in the presence of sunlight.
In summary, the pollutant specifications for the 1985 NAPAP Emissions
Inventory include sulfur dioxide, nitrogen oxides, volatile organic compounds,
and, to a lesser extent, acid gases such as hydrochloric acid and hydrofluoric
acid. The emissions inventory system employed by EPA and the States is
designed to measure or directly estimate emissions of the first three of these
pollutants. Emissions of many other pollutants, including the latter
two listed above, can be estimated indirectiy from other information collected
by the inventory system. In the case of combustion sources such as boilers,
the information required to estimate emissions of these other pollutants
includes data on boiler design, the quantity and quality of fuel fired in the
boiler, and the amount of fuel fired during the year. For other industrial
process sources, required information includes an estimate of the feedstock or
industrial process throughput during a year, and a detailed description of the
industrial process.
Transformation and transportation of these acid precursor emissions begins
almost immediately after they are emitted into the atmosphere. To a large
extent, transformation processes depend on prevailing meteorological
conditions, the atmospheric concentration of the pollutants, and the reaction
rates of Che available chemical species. However, the transport of these
precursors emitted to the atmosphere through stacks or vents is determined in
part by the height of the stack or vent above ground, and by the temperature,
flow, and velocity of the pollutants leaving the stacks or vents.
Thus, specifications for the 1985 NAPAP Emissions Inventory must include
not only pollutant emissions information, but also information about the
location, engineering design, and performance of the industrial processes,
stacks, and vents which inject the pollutants into the atmosphere. Some of
this information is not required, however, for area or dispersed sources.
These types of sources generally do not emit pollution through stacks or vents.
Pollution from these sources is usually released at ground level; atmospheric
transport of these emissions is determined by prevailing meteorological
conditions.
2-4
-------
Table 2-1 summarizes these objectives and specifications tor the 1985
NAPAP Emissions Inventory. The specifications provide a logical prioritization
for the inventory process that reflects the needs of the NAPAP research and
assessment programs. The inventory process focuses on facilities emitting at
least 1000 tons of SO2 , N0X, or VOC. Fourteen data elements, the items of
most importance to NAPAP, are given highest priority from among the 50 data
items contained in each NEDS record. These priority elements include annual
emissions estimates for SO2, NOx, and VOC, the maximum design and operating
rates, the Source and Standard Industrial Classification Codes (SCC and SIC),
emissions control equipment and efficiencies, fuel characteristics, location
data, and operating schedules.
2.3 1985 Emissions Inventory Data
In analyzing and using the emissions inventory data contained in this
report and the associated computer data bases, three factors should be
considered carefully. These factors affect the type of conclusions that should
be drawn from the use and analysis of the 1985 NAPAP Emissions Inventory data.
The first factor is that the emissions data contained in emissions
inventories are based on estimates of emissions rather than on measured values.
The development of an emissions inventory as broad in geographic and temporal
resolution as the 1985 NAPAP Emissions Inventory would not be feasible if
direct measurements were required for each of the hundreds of thousands of
emissions sources that it encompasses.
Because of this constraint, the States and EPA have expended a great deal
of effort to estimate emissions for each of approximately 3300 industrial
processes currently operating in the United States and Canada. During the last
decade, emissions from typical sources in most of the important industrial
process categories have been measured using standardized and EPA-approved
sampling techniques.
After a number of sources in each of the important industrial categories
have been tested, the measured test results are averaged for each category to
develop emission factors. These emission factors result from calculations that
determine the quantity of a pollutant released as a function of some activity
2-5
-------
Table 2-1. 1985 Emissions Inventory Priorities
Complete Emission Estimates
Plant NEDS for SCC
Size Submittal SO2 NOx VOC TSP CO
Control Fuel/
Equip/ Operat'n Locat'n
Effic'y Data Data
Stack Temporal
Data Data
Other
Key
Data
Plants
Confirm
Emiss'n SIC
I
>500
TONS
VOC
5,
>1000
TONS
OTHERS
H H M L
II
>100 M 334MLH 3 3 H M L L LL
TONS
ALL
PRIORITY: H=HIGH M=MEDIUM L=LOW
1 - HIGH for stacks > 100 feet; LOW for stacks < 100 feet.
2 - HIGH for plants with emissions of SO and NO > 2500 tons.
3 - HIGH for combustion sources; MEDIUM for other industries.
4 - HIGH for petroleum refineries and chemical processors; MEDIUM for other industries.
-------
chat is directly associated with its release (such as the number of pounds of
volatile organic compounds released to Lhe atmosphere for each gallon of
solvent used in a dry-cLeaning process). The emission tactor is multiplied by
the activity level of that industrial process over a given period of time to
estimate emissions. To estimate emissions at a given plant during a year, this
calculation is carried out using annual activity estimates and emission factors
for each of the industrial processes in operation during the year.
The second factor than must be considered when using and analyzing
emissions inventory data is the uncertainty surrounding the emissions
estimates. Because the inventory data are estimated rather than measured
values, traditional methods of assuring data quality such as accuracy and
precision cannot be used.
To date, research by NAPAP and EPA experts to develop new methods for
calculating the uncertainty of emission estimates has not been completed. The
research focuses on the most important parameters that must be accounted for in
estimating emissions and transport for each of the industrial processes
currently operating in the United States and Canada:
• the emission factor for each industrial process,
• the annual activity level,
• the variation of industrial feedstock constituents which can
affect emissions over time (e.g., coaL sulfur content),
• the efficiency and operational availability of air pollution
control devices for each industrial process,
• the "allocation" factors used to disaggregate annual estimates
into more resolved estimates such as seasonal or daily values,
and
• data omission or translation errors which are introduced as the
inventory data are first acquired in the field and then encoded
on computer forms and transmitted electronically from
one computer to another.
The third factor that must be considered when using emissions inventory
data is the completeness and accuracy of the information in the data base. For
the 1985 NAPAP Emissions Inventory, both EPA and State air pollution control
agencies worked together to ensure that the data were as complete and accurate
as possible. One indication of the level of attention that this effort
2-7
-------
received in both EPA and State agencies is the communication from EPA
Administrator and Assistant Administrator to both EPA and State participants
in che inventory (Appendix A).
Where funding or schedule constraints on the completion of the 1985 NAPAP
Emissions Inventory prevented complete attainment of the NAPAP inventory
objectives, priorities were established to ensure that the information of most
importance to NAPAP would be obtained. These priorities, presented in
Table 2-1, ensured that 1985 NAPAP Emissions Inventory data would, at a
minimum, be accurate and comprehensive for virtually all of the stationary and
dispersed sources of interest to NAPAP researchers.
2.4 OBJECTIVES AND STRUCTURE OF THIS REPORT
This report on the 1985 NAPAP Emissions Inventory is structured to provide
users with the tools to enable accurate analysis of the computerized inventory
information. It is divided into sections to meet four objectives: first,
documentation of the development of the 1985 NAPAP Emissions Inventory; second,
documentation of the quality control effort to ensure the quality of the
inventory; third, presentations of selected analyses conveying che results of
the quality control effort and summaries of 1985 emissions data with the
conclusions that can be drawn from these data; and fourth, documentation of che
QC and inventory data to provide users with the information necessary to
properly assess the inventory content and data tapes.
Sections 3 and 4 are designed to meet the first two objectives of defining
Che inventory's process and scope. Section 3 describes in detail the methods
which were used to assemble the 1985 NAPAP Emissions Inventory from design to
implementation. Section 4 outlines the quality control plan and data quality
objectives for the inventory.
Section 5 fulfills the third objective of this report by describing the
inventory quality achieved and summarizing the data collected. It presents two
different data analyses important to the user community: first, the degree to
which the inventory met NAPAP objectives for content and completeness is
analyzed, and second, the inventory emissions data are summarized and major
conclusions from these data are drawn.
2-8
-------
Finally, Che final objective of additional documentation for an accurate
assessment of the quality and extent of the NAPAP Emissions Inventory is
provided in the Appendices, which present additional information: Appendix A
contains the communication from the EPA Administrator Assistant
Administrator which defined the impo tance of the inventory effort, Appendix B
provides a list of NEOS edit check software error messages, Appendix C details
each State's emissions of SO2, N0X, and VOC by descriptive category, and
Appendix D lists criteria pollutant emissions by SCC and area source category
to provide an accurate benchmark when assessing the data tapes. Appendix E is
a glossary of terms used in NAPAP inventory work with which readers may not be
familiar.
The 1985 NAPAP Emissions Inventory is an assessment inventory, focusing on
U.S. anthropogenic annual and seasonal emissions data for five criteria
pollutants and four non-criteria pollutants. The criteria pollutant data have
been obtained from the States and the non-criteria estimates have been
calculated based on State-reported source-specific information. This inventory
and report are not relevant to U.S. biogenic emissions, Canadian anthropogenic
or biogenic emissions, or the speciated and temporally/spatially-resolved
estimates which are specific to the modeling inventory. These inventories will
be addressed in subsequent efforts and reports.
2-9
-------
SECTION 3
METHODOLOCY
The purpose of this section is to describe the methods which were used to
develop and collect emissions inventory data for the 1985 NAPAP Emissions
Inventory. For the NAPAP inventory, the National Emissions Data System (NEDS),
an existing EPA inventory computer system, was used to convey and store
information. However, EPA and the States have modified, supplemented, and
improved the NEDS for the 1985 effort, so that the system is substantially
different than the conventional system used in previous years.
The organization of this section reflects the bipartite NEDS file
structure. The first file, which consists of State-reported point source data,
is discussed in Section 3.1. The second file, which contains area source data
calculated by EPA, is described in Section 3.2. This chapter also presents
information on two other types of data collected at the national level:
(1) utility-related fuel use and boiler data (Section 3.1.6), and (2) sources,
procedures, and estimates of noncriteria pollutant emissions of interest to
acid deposition assessment (Section 3.3), including primary sulfates, hydrogen
fluoride, hydrogen chloride, and ammonia. These methods were developed to
calculate NAPAP anthropogenic emission estimates from point and area sources
for acid rain precursors in the U.S.
The point source emissions data base is collected at the State level. It
includes emissions estimates and supporting data gathered by State air
pollution control agencies from point sources within each State. The national
data base is the sum of these State-level source inventories, collected for
each industrial process and compiled successively at the source, State, and
national levels.
The area source data base represents anthropogenic emissions from true
area sources (e.g., mobile sources), from plants which emit less than 100 tons
per year (TPY), and from point sources which emit less than 25 TPY or which are
too difficult to inventory individually. Area source emissions are calculated
by EPA through a process in which county-level emissions are estimated using
category-specific emission factors and activity levels (e.g., published fuel
deliveries), and then summed to produce national emissions estimates. These
3-1
-------
fundamentally different approaches characterize the distinction between the
point and area source data bases, which are described below in detail.
3.1 POINT SOURCE DATA
3.1.1 NEDS Structure
Subpart Q of AO CFR Part 51 (Sections 51.320 through 51.323) authorizes
annual reporting by the States to EPA of emissions inventory data for all
stationary sources (i.e., plants) emitting 100 TPY or more of criteria
pollutants. An exception to the 100-ton limit is made for carbon monoxide
(1000 TPY). For plants meeting these requirements, any points (e.g., stack or
process within the plant) emitting 25 TPY or more of any criteria pollutant
(250 tons for CO) are to be reported. Points with lower emissions may also be
reported by the State if resources are available. Points and sources not
meeting these criteria and not reported in the NEDS point source inventory are
considered area sources. Their emissions are included in county-wide area
source emissions estimates as described in Section 3.2.
Data reported for point sources in NEDS may be categorized according to
the following groups:
General source information — name, address, type(s) of sources(s),
Standard Industrial Classification, year of record, and comments.
Emissions data — operating or production rates and capacities,
estimated emissions, estimation method, and type and efficiency of
control devices for each pollutant.
Modeling parameters — UTM coordinates of source, stack height and
diameter, exhaust gas temperature, and gas flow rate.
Compliance information — allowable emissions, compliance status, and
compliance schedules.
3-2
-------
NEDS point source data are organized into three hierarchical levels:
plant, point, and process.
Plant level data apply to an entire facility.
Point level data apply to individual emission points within a plant. A
plant may contain any number of emission points. A point is that portion of a
facility that may be considered individually for emission purposes. A point
may contain one or more processes or pieces of equipment that contribute to
the emissions from the point. In most cases, a point emits pollutants through
a single confined location such as a stack. A point may also be an aggregation
of two or more similar pieces of equipment which, taken separately, would not
qualify for inclusion due to emissions of SO2, N0X, or VOC of less than 25 TPY.
In addition, defined emissions sources without stacks may be included in NEDS
as point sources (for example, fugitive emissions from plant equipment, storage
piles, and lagoons).
Process level data apply to individual processes within a point and are
utilized to calculate emissions. Each process is defined by a Source
Classification Code (SCC).^ In general, for each SCC there are emission
factors which relate the quantity of pollutants generated by a process to an
annual process operating rate. These emission factors are used to compute
emissions. Multiple processes and multiple SCCs may be grouped under one
emission point, as in the cases of boilers using two fuels or two separate
processes sharing the same stack.
Table 3-1 illustrates the hierarchical structure of the NEDS point source
data file and the individual data items at each level. Individual data items
of importance to NAPAP are discussed in the following section. Table 3-1 has
omitted the "year of record" entries for all levels, as well as other items not
currently in use.
3.1.2 Priority Data Elements/NAPAP Emissions Inventory Priorities for 1985
To help guide data collection efforts for 1985, data collection objectives
were prioritized as summarized in Table 3-1. This prioritization reflected the
needs of the NAPAP research and assessment programs for atmospheric transport,
as well as EPA and State inventory needs. Emissions of SO2, N0X, and VOC
3-3
-------
TABLE 3-1. NATIONAL EMISSIONS DATA SYSTEM (NEDS) DATA STRUCTURE
PLANT LEVEL DATA
State, county, Air Quality Control Region, city and UTM zone codes
Plant identification number
Plant name, address and contact person
POINT LEVEL DATA
Point identification number
Standard Industrial Classification code
UTM coordinates
Stack/plume data (height, diameter, temperature, flow rate)
Points with common stack
Boiler design capacity
Control equipment (primary amd secondary device codes for each
criteria pollutant and overall efficiencies by pollutant)
Operating schedule (by season, hrs/day, days/wk, wks/yr)
Emission estimates for criteria pollutants
(actual, with estimation method, and allowable)
Compliance data (status, schedule, update, regulations)
PROCESS LEVEL DATA
Source Classification Code
Operating rates (annual, maximum hourly design)
Fuel content (sulfur, ash, heat)
-------
were assigned highest priority. Emissions of. TSP were assigned an intermediate
priority level. The highest priority Level was assigned to plants emitting a"
least 1000 tons of these pollutants during 1985. The next level of priority
for these pollutants was assigned to plants emitting 100 to 1000 tons during
1985.
Table 2-1 reflects the NAPAP/EPA consensus on the fourteen highest
priority data elements from among the fifty available in NEDS. To obtain a
nationally consistent baseline, it was critical to obtain actual 1985 data for
emissions estimates and operating rates. The other priority data elements fell
into three categories: emissions estimation data, spatial/temporal data, and
data to identify inconsistencies. All these items were selected for their
importance to the acid deposition research and assessment communities.
Annual Emissions Estimate—Complete reporting of emissions for all
significant emission points (defined as points emitting more than 25 tons in
1985 of SO2, NOx, or VOC) was most important. To establish the best possible
national emissions inventory for 1985, maximum use of standard procedures for
calculating emissions was recommended. Emissions calculation procedures are
discussed in more detail in Section 3.1.7. The inclusion of calendar year 1985
estimates was critical to establish a consistent baseline.
Annual Operating Rate—The annual operating rate is the amount of fuel
consumed, amount of product produced, or other material throughput during 1985,
as defined by the SCC (below). In most cases this information was obtained
from plant records. It was inappropriate to estimate annual operating rates by
simply multiplying the hourly maximum design rate by the hours of operation,
unless it was known that the source operated at or near full capacity
throughout the year. Estimates of annual operating rates took into account a
"capacity utilization factor," i.e., the fraction of full production or
consumption capacity reflecting normaL operations during 1985. Again, 1985
data were critical to the effort.
SCC—To properly classify sources in a standard manner, SCC numbers were
obtained from the NEDS SCC listing.^ An SCC is an 8-digit code divided into
3-5
-------
four levels of identification, signifying 1) the category process, 2) Lhe major
industry group, 3) the major product, and A) different operations at the point
source. The SCC defines specific units for operating rates and carries with it
a set of emission factors, which may be used to caLculate emissions (if
appropriate). A maximum of 10 SCCs could be assigned for any point, and no
duplication of SCCs was allowed at a single point.
Control Equipment and Control Efficiency—Entries were required for all
emission points where control equipment applicable to SO2, NOx, or VOC
emissions was in place. Control efficiencies were taken to be the best
available estimates of annual average actual control efficiencies. NEDS
provides for reporting both primary and secondary control devices and their
control efficiencies for each pollutant.
Fuel SuLfur/Fuel Ash—Fuel sulfur and ash contents were reported as weight
percents, representing weighted annual averages for the fuel burned. Fuel
sulfur was necessary for the calculation of SO2 emissions from coaL, oil,
process gas, and coke combustion; fuel ash was used to calculate particulate
emissions from coal and coke combustion.
UTM Zone and Coordinates—The Universal Transverse Mercator (UTM) zone and
coordinates provided geographical reference within a standard grid system.
These coordinates permitted each point or source co be located to the nearest
0.1 kilometer.
Stack Data—Height (above ground level) and inside diameter of each
emissions stack were reported. Height was ordinarily reported to the nearest
10 feet and diameter to within one-tenth foot. In addition, the exhaust gas
temperature (°F) and flow rate (actual cubic feet per minute) were reported.
Emission points with common stacks were designated where applicable. A plume
height (i.e., release height) was reported in cases where no stack existed
(e.g., storage tanks). These data are essential to estimate plume rise and
atmospheric transport of emissions.
3-6
-------
Normal Operating Schedule—The normal number of hours per day, days per
week., and weeks per year ea.h point operated was indicated. This schedule
allows the temporal resolution of emissions estimates by defining the
variability of operations.
Percent Annual Throughput—Seasonal variations in production rate or
throughput at a point were indicated by percentages for each season, to allow
temporal resolution of emissions estimates among the four seasons.
Fuel Heat Content—Fuel heat content represented the gross or higher
heating value of each fuel. This value was combined with boiler design
capacity and hourly maximum design for a cross-check to identify
inconsistencies between these three items and annual operating rate.
Boiler Design Capacity—Boiler design capacity was defined as the maximum
gross heat input rate for a boiler. It was reported in million BTU/hour. This
rate was applicable only to emission points representing boilers. For all
other points, including other combustion sources, the entry was zero. If a
point represented a number of small boilers grouped together, the boiler
capacity was taken as the total for all boilers at the point.
Hourly Maximum Design Rate—Maximum design rate was defined as the highest
operating rate expected for a source. Where the SCC units were not time
dependent (e.g., capacity of storage tanks), a value of zero was used.
SIC—The Standard Industrial Classification was recorded for the source.
This item provided a cross-check with the SCC to identify inconsistencies, and
is used as a standard identifier in data retrievals and analyses.
3.1.3 1985 Emissions Data Collection Effort
To meet the objectives of the 1985 NAPAP Emissions Inventory, EPA and State
air pollution control agencies developed a plan to improve the quality and
comprehensiveness of the point source data base. A major portion of the plan
3-7
-------
covered data collection efforts. EPA committed substantial resources to
assisting the States in developing and executing the 1985 NEDS. This planning
resulted in standardized data collection and emissions estimation methods. The
organization of the collection effort addressed the following eight areas,
which are discussed individually below:
• Survey by the State and Territorial Air Pollution Program
Administrators (STAPPA) to establish realistic objectives,
• Analysis of data gaps in each State's inventory,
• Written guidance developed for the States and EPA prior to
Che inventory,
• Standardization of emissions estimation procedures,
• Organization of the inventory process to the State level,
• Implementation of emissions confirmation procedures,
• Institution of quality control procedures, and
• Reliable transfer of emissions data from the State to EPA.
3.1.4 STAPPA Survey
The EPA worked with the STAPPA Acid Rain Subcommittee to identify the
capabilities and needs of the States in the development and execution of the
inventory. STAPPA surveyed the States to determine the level of inventory
effort the States could provide for 1985 with existing resources, and the types
and levels of support needed by the States to meet the objectives of the 1985
NAPAP Emissions Inventory. The STAPPA survey demonstrated that States would
encounter resource shortfalls in meeting schedule and data quality objectives.
As a result, EPA committed financial and contractor support during 1986 and
1987 to States requesting assistance in five inventory areas: inventory
development, emissions confirmation, quality control, data coding, and data
editing.
3-8
-------
3.1.5 Written Analysis of Deficiencies
In order to assist States in targeting resources for the 1985 NAPAP
Emissions Inventory, NADB reviewed each State's most recent (i.e., 1984) NEDS
data in early 1986. Each State NAPAP contact and EPA Regional contact received
a written analysis of NEDS data. This analysis identified data problems for
sources of SO2, N0X, or VOC, greater than 1000 TPY which, if not corrected,
would cause data Co fall short of the NAPAP requirements or would require State
resolution during quality control review. These potential problems included
these items:
• Missing high priority data items,
• Allowable emissions reported as actual emissions, and
• Systematic problems such as invalid control codes or
missing emissions estimates.
3.1.6 Guidance
Establishing a common methodology for EPA and the States for data
collection, emissions estimation, and quality control (QC) was an important
early step in assembling an adequate and consistent data base. This step was
important for two reasons. First, the resulting Stace data bases would be
developed on a nationally consistent basis. Second, QC procedures could be
developed to build QC directly into the data collection process, where errors
could be located and corrected most efficiently. Essential elements of
procedural and technical guidance were communicated to EPA, State, and
contractor personnel through two workshops and companion manuals. In addition,
utility data collected by DOE were furnished to the States to aid compilation
of utility data in this inventory.
3.1.6.1 Technical Guidance
The first workshop was held in October 1985 for EPA, NAPAP, and State
personnel. Manuals addressing data collection procedures,^ emissions
3-9
-------
confirmation for major source categories,^ hydrocarbon emissions estimation*
(factors for reactive VOC), and 1985 DOE utility fuel use data^ were
distributed at the workshops and also made available to interested State
inventory personnel. The subjects covered by these documents will be discussed
later in this chapter. Additional written guidance concerning NAPA"' inventory
priorities, confirmation procedures, and QA/QC checks^'^ was distributed to EPA
Regional Offices and States in December, 1985.
A second workshop, specifically for Contractors assisting the States, was
held in May 1986. In addition, States were supplied criteria pollutant
emission factors,* and particulate speciation® and VOC speciation guidance,^
for use in the inventory effort. These workshops and manuals established a
common methodology for collecting, coding, editing, and transferring the 1985
NAPAP Emissions Inventory.
3.1.6.2 Utility Data
The electric utility industry represents a significant source of acid
precipitation precursors in the United States, accounting for approximately
two-thirds of total anthropogenic SO2 emissions and one-third of total
anthropogenic N0X emissions. The increasing trend toward tall stacks at
utilities has increased the likelihood of long-range pollutant transport,
making facility data such as stack parameters important to modeling efforts.
NAPAP priority data were collected from utilities within the framework of the
NEDS inventory; however, there are other data bases which contain similar
information.
The Department of Energy (DOE) requires utilities to report emissions and
operating data on DOE Forms 767, 759, and 423. Form 767 provides boiler-level
data, while 759 and 423 report plant-level information. Historically, DOE
data have been considered to be the most comprehensive available. For 1985,
summaries of the utility data from these forms were made available to the
States at the beginning of the inventory effort. These data were employed to
supplement the 1985 NAPAP Emissions Inventory data, as well as to identify
inconsistencies between data bases on a State, plant, and point basis. In
addition, a comprehensive review of Forms 767, 759, and 423, and of the 1985
3-10
-------
NAPAP Emissions Inventory was conducted in order to identify any systematic
problems in the NAPAP inventory regarding utilities. Further information on
these comparisons is found in Sections 4.0 and 5.0. Based on this review,
NAPAP-reported data appeared as reasonable and accurate as DOE data.
A few States were not able to provide ail the utility data requested by
NAPAP. In these cases, DOE data supplemented State data collected through
NEDS. The State reviewed the DOE information, and updated NEDS information in
light of the DOE results. In each instance, the State chose the data best
representing operations at individual utilities.
3.1.7 Emissions Estimation Procedures
One of the emissions estimation objectives for the 1985 NAPAP Emissions
Inventory was to use calculated estimates of actual emissions during 1985.
NEDS recognizes two distinct estimation procedures: one using individual
source-specific emissions data and one using emission factors. EPA requested
that States use standard emissions estimation methods, which are described
below, whenever applicable, and presented a hierarchy for utilization of the
2
acceptable methods. Each method is documented with a unique code in NEDS and
can be tracked over time for each emission point.
States were asked to report estimated emissions based on individual source
data if available. These calculations are based on continuous emissions
monitoring (CEM) data, source test data, or materials balance information as
reported by the source or calculated by the control agency. Source-specific
data are considered the most accurate data for estimating emissions.
If individual source data were not available, States were asked to
calculate emissions using an emission factor. EPA recommended that the
AP-42*® emission factors and methods be reviewed in selecting an emission
factor. If no AP-42 factor was available, emission factors appearing in the
NEDS SCC and Emission Factor Listing* were to be used if available. This
resource lists AP-42 factors as well as additional factors not from AP-42, and
AP-42 factors assumed to be transferable to other SCC categories. States also
had the option of selecting State emission factors based on knowledge of the
operation of sources within the State.
3-11
-------
If a State was not able to estimate emissions with any of the methods
discussed, two options were still available. For data from sources without
emission factors, the State could code data so that NEDS would calculate
emissions when appropriate emission factors were developed. States could also
use this option to indicate that AP-42 factors on file in NEDS should be
applied to operating data to calculate emissions. As a last option, the State
could use its judgment to estimate emissions based on knowledge of the
particular point (e.g., emissions test data from similar State sources).
3.1.8 Inventory Process at State Level
Point source data were collected by State air pollution control agencies
for the calendar year 1985. The format and logistics for the actual collection
were handled by each State individually. A majority of States used either the
NEDS or the Emission Inventory System/Point Source (EIS/PS) system to store the
collected inventory data. In either case, conducting the inventory entailed
four steps:
• Identification of facilities,
• Questionnaire development and/or distribution,
• Codification of resulting data, and
• Development of a NEDS-compatible computer tape.
The resources for planning and conducting this inventory varied from State
to State. In many cases, States needed to contact all major emission sources.
Where States' resources permitted, the States utilized their normal annual
inventory procedures while emphasizing NAPAP priority data items in their
collection effort. Due to resource constraints, not every State was prepared
to conduct an inventory of this scope for 1985. Financial aid and Contractor
assistance were provided to States, according to each specific State's needs,
as described below.
3-12
-------
When requested, EPA provided assistance to the States in collecting and
coding data. This effort relied on previous inventories, expertise of State
personnel responsible tor air pollution sources, and/or an evaluation of SICs
and number of employees. In general, State expertise was sought through
consulting the central air pollution control agency, working with State
personnel, and accessing current files. Working with a designated State
inventory contact, EPA helped the State devise a questionnaire to request all
key NAPAP data elements, actual emissions estimates, and any additional data of
interest to the State. This questionnaire was sent out by the State and
results were received and filed at the State air pollution control agency.
Facility response was excellent, due in part to efficient State follow-up.
This contact allowed missing or questionable data elements to be reviewed
directly with the facility, and permitted prompt initiation of applicable
confirmation procedures with the facility, as discussed in the following
section.
Most States required assistance to translate the raw data returned from
facilities into a NEDS-compatible format (i.e., NEDS coding forms, State
Emissions Inventory System, or microcomputer data base) and to complete the
necessary QC procedures. Because of the diversity of inventory formats among
States, carrying out the standardized methodology for data coding was a
significant challenge. One efficient method of data coding was to use a
microcomputer and software developed for this purpose. The software allowed a
user with an IBM^ or compatible PC to load data into dBase^"1 data base
management files through a series of menus. These files were then used to
produce a magnetic tape in NEDS card image format.
3.1.9 Confirmation
Where State resources permitted, emissions estimates for large emitters
were confirmed. Confirmation involved affording large emitters an opportunity
to review the reasonableness and acceptability of State emissions estimates and
make comments to the State. A lower limit of 2500 TPY SC>2 or NOx emissions was
chosen for confirmation. The effort was intended to improve the accuracy of
3-13
-------
the inventory for very large SO2 and N0X sources without imposing undue burden
on the State.
Confirmation was accomplished in one ol two ways. In the tirst,
predominant method, the State provided emissions estimates and calculations for
each unit process (SCC) to the facility contact for review. After review of
the State estimate, the source indicated agreement with the estimate or
provided an alternative estimate that was more accurate. In case of
disagreement, the source provided alternative calculations for its estimate.
State and source were to attempt to reach a final agreement on t.he appropriate
emissions estimate, but ultimate responsibility for selecting the most accurate
estimate rested with the State.
A second method involved requesting the facility to provide emissions
estimates and supporting data for each unit process at the time of the yearly
inventory. The State would then review each estimate and either accept or
reject it. As above, the State retained responsibility for submitting the most
accurate estimate.
For each facility emitting at least 2500 tons of SO2 or N0X in 1985, a
confirmation report was prepared by the State or by the Contractor assisting
the State. This report consisted of a cover memorandum from an appropriate
State agency official and a table representing all facilities meeting the
criteria. The table contained source name, NEDS ID number, actual 1985 SO2 and
NOx emissions, and confirmation status. The confirmation status could be one
of the following: (1) agreement of State and source, (2) disagreement —
State estimate chosen, (3) disagreement — source estimate chosen, (4) no
comment by source, or (5) no confirmation attempted. Provisions were made for
distinguishing between major facility-level conflicts and minor point-level
disagreements. A final confirmation report was required from the State by the
date of the final NEDS submittal.
States were encouraged to confirm estimates below the 2500 ton criterion.
Also, large sources of VOC were confirmed in a number of States. Insofar as
States were able to confirm emissions below levels mandated as part of the
effort, confidence in emissions totals was improved.
3-14
-------
3.1.10 Quality Control
For 1985 emission submittals, EPA Regional Offices were responsible for
working with State air pollution control agencies to perform quality control
(QC) of NEDS data on computer tape.^ A computerized edit routine was
specifically revised for this purpose, and assistance was provided to run QC
checks on State data prior to submittal to the Regional Office. The purpose of
these checks was to identify missing and questionable data so that the States
could correct errors and validate questionable data. These efforts are
discussed extensively in Section 4.
The preliminary edit identified two types of problems in the data. First,
data were rejected due to incorrect coding; points having these errors were
unacceptable to NEDS and could not be entered into the system without
correction. These were the most serious errors, and included invalid NEDS
identification codes (i.e., the alphanumeric code that uniquely identifies each
plant and point in the NEDS point source file), invalid estimation method
codes, or invalid SCCs. The second type of problem identified was questionable
data. Such data included inconsistent or missing emissions estimates;
inconsistent or incorrect stack and location parameters; and inconsistent
design, operating, and emissions parameters.
After the State and Regional Office investigated and corrected these error
messages, the tape was forwarded to EPA'S OAQPS and AEERL. These groups then
reviewed the submittal to identify any remaining questions or problems, which
were returned to the Regional Offices for resolution. This quality control
loop for the point source data base is described in much greater detail in
Section 4.0.
3.1.11 Transfer to NEDS
Each State's submittal was entered into NEDS after final QC processing.
The goal was to transfer the State's data to NEDS accurately, ensuring that
NEDS reflected the data on file at the State. This update procedure was
complicated by compatibility problems with the data formats available from some
States. In some cases, these problems arose from unique State storage and
3-15
-------
retrieval systems. At other times, when a State had modified EIS/PS to betLer
serve its own needs, the change created translation difficulties when data were
transferred to NEDS.
These problems were resolved in two ways. First, adjustment or creation of
a translation program was attempted by OAQPS/AEERL in cooperation with the
State. Transfer of high priority data elements was emphasised during these
procedures. If this process was not successful, then either a data translation
was carried out and corrections were made by hand, or the data were entered
directly into NEDS from NEDS forms. In either case, the final NEDS product
faithfuLly transcribed high priority 1985 NAPAP Emissions Inventory data from
the State's data base.
3.2 AREA SOURCES
3.2.1 Overview
For the 1985 NAPAP Emissions Inventory, extensive modifications were made
to the traditional NEDS area source methodologies, which extended the inventory
to sources not previously considered (e.g., treatment, storage, and disposal
facilities), and improved VOC emissions coverage (e.g., fugitives). Area
source data files developed from NEDS serve as the basis for the 1985 NAPAP
area source inventory. These include mobile sources and point sources too
numerous or difficult to classify individually. Historically, NEDS area source
data have been developed mainly by OAQPS from data voluntarily submitted by
State agencies. More complete documentation and references are available in
1 ?
Area Source Documentation for the 1985 NAPAP Inventory. "
NEDS area source emissions estimates are updated annually by a series of
computer programs which multiply each current area source activity level (e.g.,
fuel delivery) by the appropriate emission factor which accounts for emissions
removed by any control technology. County emissions estimates are then summed
to produce national emissions estimates.
Activity levels are derived primarily from related information published by
other Federal agencies, supplemented by special data developed by EPA for the
purpose of developing NEDS area source inventories. Published data such as
3-16
-------
fuel use by State, motor vehicle miles ol travel (VMT) by State and county, and
forest fire acres burned by State are used with related data such as
employment, population, and miscellaneous geographic or economic data to derive
annual county estimates of the activity levels for each of the NEDS area source
categories. The activity levels derived are adjusted to account for point
source activity (such as fuel use by point sources) so that the area source
data reflect only the activity levels (and resulting calculated emissions) that
are not accounted for by point sources.
Area source emissions estimates for five pollutants (particulates, SO2,
NOx, VOC, and CO) are calculated for each area source category utilizing
appropriate emission factors from the NEDS area source emission factor file.
For most categories, emission factors were originally obtained from the EPA
Compilation of Air Pollutant Emission Factors (AP-42).^® For many categories,
the same emission factors are used for all counties; however, for some source
categories, State- or county-specific emission factors account for local
variables that affect emissions. These more specific factors are used in NEDS
calculations for all highway motor vehicle categories, fugitive dust
categories, and for other selected categories in a few counties where data are
available to develop more applicable emission factors than the national
emission factors. Computer-calculated emissions can also be overrridden by
hand-calculated emissions that may be more accurate than values calculated from
emission factors.
As shown in Table 3-2, area sources in the 1985 NAPAP Emissions Inventory
are divided into five major groups: Stationary Sources, Mobile Sources, Solid
Waste Disposal, Miscellaneous Area Sources, and Additional Area Sources.
Additional Area Sources include categories for which methodologies have been
developed to estimate emissions for the 1985 NAPAP Emissions Inventory. Brief
summaries of the methods used for these five major groups are provided below,
followed by detailed descriptions for each group.
3.2.1.1 Stationary Sources
Many stationary emissions sources are point sources which emit less than
25 TPY and which are thus not included individually in the point source
3-17
-------
Table 3-2. 1985 NAPAP AREA SOURCE CATECORY GROUPS
Stationary Sources
Residential Fuel
Commercial and Institutional Fuel
Industrial Fuel
Mobile Sources
Highway Vehicles
Off-Highway Vehicles
Railroad Locomotives
Aircraft
Marine Vessels
Solid Waste Disposal
On-Site Incineration
Open Burning
Miscellaneous Area Sources
Evaporative Losses from Gasoline Marketing
Evaporative Losses from Organic Solvent Consumption
Unpaved Roads
Unpaved Airstrips
Construction
Miscellaneous Wind Erosion
Land Tilling
Forest Wildfires
Managed Burning (Slash/Prescribed Burning)
Agricultural Burning
Frost Control (Orchard Heaters)
Structural Fires
Additional Area Sources
Publicly Owned Treatment Works (POTWs)
Hazardous Waste Treatment, Storage,
and Disposal Facilities (TSDFs)
Fugitive Emissions from Synthetic Organic
Chemical Manufacturing
Bulk Terminals and Bulk Plants
Fugitive Emissions from Petroleum Refinery Operations
Process Emissions from Bakeries
Process Emissions from Pharmaceutical Manufacturing
Process Emissions from Synthetic-Fibers Manufacturing
Crude Oil and Natural Gas Production Fields
Cutback Asphalt Paving Operations
3-18
-------
inventory. The stationary sources category consists of small fuel-burning
sources, and is divided into three major categories: Residential Fuel,
Commercial and Institutional Fuel, and Industrial Fuel. Each category is
further subdivided into fuel types for which consumption data are estimated
using algorithms and published fuel use data. Consumption estimates used as
the measure of activity levels are then multiplied by emission factors from
AP-42 to obtain emissions estimates.
Residential Fuel—The residential fuel category includes estimated
emissions for residential activities which utilize fuel for water heating,
space heating, and cooking. Emissions contributed by residential fuel
consumption are calculated for five fuel types using an algorithm that
considers heating degree-days, number of residential unit«s, and median number
of rooms per dwelling. The estimated county activity levels for each fuel type
are then normalized to be consistent with published State data.
Commercial/Institutional Fuel—Area source emissions from fuel use by
commercial and institutional sources include emissions from hospitals, hotels,
laundries, schools, and universities. County consumption data for five fuel
types are estimated for the five identified coimnercial subcategories.
Algorithms are based on daca such as employment, climatological data,
population, enrollment, and number of beds per institution. Total fuel
consumption is distributed by the subcategories to each fuel type, using
housing data and published State fuel consumption data. Fuel consumption
estimates for the five subcategories are summed and compared with published
State fuel totals. If the estimated fuel use exceeds the published State
total, county estimates are normalized to agree with the State cocal. If the
estimated fuel use is less than the State total, the difference is added to the
estimated subcategory totals and allocated to counties by population.
Industrial Fuel—Emissions for the industrial sector are calculated as
follows. First, State-level industrial area source fuel consumption estimates
are determined by subtracting industrial point source fuel consumption totals
from published State totals. Next, the area source fuel use estimates are
allocated to counties using county employment data for the manufacturing sector
3-19
-------
(Standard Industrial CLassification groups 20 through 39). The county
employment data are adjusted to exclude employment at poinc sources and are
weighted to reflect differences in fuel use per employee for each two-digit SIC
group. The industrial sector includes both external and internal combustion
sources.
3.2.1.2 Mobile Sources
Mobile sources which contribute to area source emissions are divided into
Highway Vehicles, Off-Highway Vehicles, Railroad Locomotives, Aircraft, and
Marine Vessels.
Highway Vehicles—For the purpose of calculating fuel consumption, highway
vehicles are disaggregated into four categories on the basis of use and gross
vehicle weight. The categories include light duty gasoline vehicles, light
duty gasoline trucks, heavy duty diesel vehicles, and heavy duty gasoline
vehicles. NEDS allocates fuel consumption based on vehicle registration data
and published average miles traveled (where available) by vehicle type. Fuel
consumption, average fuel efficiencies, and mileage by road type in each county
are used to determine vehicle miles traveled (VMT) for three road classes:
limited access roads, rural roads, and urban roads. This allocation separates
the total VMT for a county into road speed classes. Emission factors obtained
from the execution of an EPA computer model are applied to determine emissions
for each vehicle type and speed class.
Off-Highway Vehicles—Emissions from off-highway vehicles are generated by
activities of gasoline and diesel vehicles which do not utilize road systems.
Vehicles contributing to off-highway emissions are divided into six general
categories: farm equipment, construction equipment, industrial equipment,
motorcycles, lawn and garden equipment, and snowmobiles. Consumption is
estimated separately for each category by either apportioning national fuel
consumption to counties on the basis of employment, population, etc., or by
calculating county or State totals by applying fuel consumption rates to
average usage figures and equipment populations. Estimated fuel use is
3-20
-------
normalized to agree with published State totals, where available. For each
category, emission factors from AP-42 are weighted using equipment populations.
Off-highway motorcycle emissions are calculated using data representing
uncontrolled emissions from an EPA computer model.
Railroad Locomotives—The activity level for railroad locomotive use of
distillate oil is calculated by allocating published State consumption data to
the county level based on county population statistics. Resulting consumption
data are used with AP-42 emission factors to determine emissions.
Aircraft—Activity level estimates for aircraft (private, military, and
commercial) utilize aircraft operations data and aircraft type populations to
estimate number of landing/takeoff cycles (LTOs) in each group. Emission
factors from AP-42, defined as emissions per LTO, are multiplied by LTOs to
obtain emissions.
Marine Vessels—Estimation of marine vessel consumption for distillate oil,
residual oil, and gasoline is based on published consumption data. Distillate
oil and residual oil used by vessels are allocated to counties using data on
the number of vessels visiting major ports and tonnage of cargo handled in each
port. Gasoline vessel consumption computations utilize inboard and outboard
motorboat registrations and published average consumption data to determine
consumption. County allocation is based on inland water area, coastline, and
the number of months suitable for recreational boating. Consumption data are
multiplied by emission factors from AP-42 to obtain emissions estimates.
3.2.1.3 Solid Waste Disposal
The solid waste disposal category includes on-site refuse disposal
activities by residential, commercial/institutional, and industrial sectors.
Solid waste generation for open burning and on-site incineration is calculated
using population data, per capita generation factors, and information from
related point source categories. Activity levels are multiplied by specific
emission factors from AP-42 to obtain emissions estimates.
3-21
-------
3.2.1.4 Miscellaneous Area Sources
NEDS area sources which are not defined by the other categories are
compiled in the Miscellaneous Area Sources category. Miscellaneous area
sources include Gasoline Marketing, Organic Solvent Consumption, Unpaved Road
Travel, Unpaved Air Strip Use, Forest Wildfires, Managed Burning, Agricultural
Burning, and Structural Fires.
In brief, activity levels estimated using category-specific data are
multiplied by emission factors to obtain emissions estimates. Activity levels
for Gasoline Marketing are determined using county retail service station sales
data. Activity levels for Organic Solvent Consumption are determined by
allocating national estimates of organic solvent consumption by end-use
category to counties according to manufacturing employment data or population.
The Unpaved Road Travel category utilizes unpaved road miles and rural
population as the basis of county allocation. Unpaved Airstrip Use is
estimated by landing/takeoff (LTO) cycles occurring in the county. Number of
acres burned and fuel loading factors are used as activity levels for Forest
Wildfires, Managed Burning, and Agricultural Burning categories. The
Structural Fires category utilizes the number of building fires, allocated to
the county level by population.
3.2.1.5 Additional Area Sources
The 1985 NAPAP Emissions Inventory will provide detailed county-level VOC
emissions estimates for additional area sources which previously have not been
included in the NEDS area source categories. In this section, methods have
been developed for many categories traditionally considered point source
categories, such as Bakeries and Synthetic Fiber Manufacturing. These
categories were included to reconcile the difference between reported total
national air pollutant emissions estimates for these categories and emissions
already accounted for by the NEDS point source data files. The remaining
categories such as Publicly-Owned Treatment Works (POTWs) and Hazardous Waste
Treatment Storage and Disposal Facilities (TSDFs) have been included as area
sources due to the difficulty in measuring emissions from specific points in
3-22
-------
these facilities (e.g., aeration basins).
In the following sections, detailed activity level and emission factor
calculations are described with references.
3.2.2 Stationary Sources
Stationary sources which contribute to area source emissions have been
divided into three major categories: Residential Fuel, Commercial and
Institutional Fuel, and Industrial Fuel. Collectively, these categories
account for all stationary fuel combustion activity not usually reported as
point sources. Methodologies for activity level estimation and emission factor
derivation are discussed for each category and fuel type.
3.2.2.1 Residential Fuel
The residential fuel category estimates emissions for residential
activities which utilize fuel for water heating, space heating, and cooking.
Emissions contributed by residential fuel consumption are broken down by five
fuel types including anthracite coal, bituminous coal, distillate oil, natural
gas, and wood. For each of the listed fuel types, activity levels measured by
fuel quantity consumed in weight or volume units are multiplied by emission
factors from AP-42 to obtain emissions estimates.
In the following methodologies for the calculations of activity levels,
consumption is determined for each type of fuel using two general steps:
o County consumption is calculated using an algorithm based on
significant variables for which county-specific data are available
(i.e., degree-days, number of rooms per dwelling, number of
dwellings, etc.).
o Resulting county consumption estimates are normalized to
reflect published State consumption data by the following
equation:
Normalized Estimated Published State Consumption
County = County X
Consumption Consumption Estimated State Consumption
3-23
-------
Anthracite and Bituminous Coal—The basic methodology for allocating
residential consumption of anthracite and bituminous coal to individual
counties involves the use of an algorithm which calculates coal consumption
from the number of dwelling units and heating degree days. Adjustments are
made to census housing data to account for trends in the number of coal-heated
dwelling units, and to disaggregate the total coal consumption into anthracite
and bituminous components using current coal market data. The results are then
normalized as necessary to reflect published coal consumption.
Distillate Oil—Consumption of distiLlate oil by residential sources is
determined using an algorithm which calculates consumption of fuel for space
heating and water heating using the annual heating degree days and the median
number of rooms of occupied dwelling units for each county for the most recent
census year. The value of each county's consumption is then normalized as
necessary to agree with total State consumption.
Natural Gas—In the NEDS inventory, residential natural gas consumption is
defined as the sum of natural gas consumption and liquefied petroleum gas (LPG)
consumption for the purposes of cooking, water heating, and space heating. In
general, the methodology is designed to produce county consumption estimates
for each use by fuel type using algorithms based on the annual heating degree
days, the number of occupied dwelling units using gas for cooking or water
heating fuel, and the median number of rooms per dwelling. County estimates
are normalized with published data and then combined to produce the final
county estimates for natural gas and LPG.
Mood—Residential wood consumption is estimated by updating published
State figures with annual regional data and then allocating to the county level
based on the number of dwelling units which reported heating with wood in each
county. Emission factors for woodstoves and fireplaces obtained from AP-42
are weighted based on the proportions of wood burned in woodstoves and in
fireplaces. Weighting is accomplished by performing a series of calculations
on computed wood consumption estimates which include the following steps: (1)
estimating the number of stoves based on shipments and imports, (2) calculating
3-24
-------
an obsolescence rate to determine the toLal stove inventory in current use, and
(3) determining the stove population in primary and secondary use based on the
number of dwellings in the county. Stove efficiency is also taken into
account.
3.2.2.2 Commercial and Institutional Fuel
Area source emissions from fuel use by commercial and institutional
sources consist of emissions from ail fuel burned in stationary sources that
are not included under residential sources, industrial sources, power plants,
or commercial point sources. Important commercial/institutional area sources
are hospitals, hotels, laundries, schools, and universities.
Activity levels are estimated for anthracite coal, bituminous coal,
distillate oil, residual oil, and natural gas using the methodology which is
described in the Anthracite Coal section for Residential Fuel.
County commercial/institutional area source activity levels for anthracite
coal, bituminous coal, distillate oil, residual oil, natural gas, and LPG are
calculated for five major subcategories, namely, hospitals, hotels, commercial
laundries, schools, and universities. The methodology obtains consumption data
for each fuel type in the following steps: (1) estimating total county fuel
consumed by the five identified commercial subcategories using algorithms based
on employment, annual heating days, bed counts, and number of rooms; (2)
distributing total fuel consumption to each fuel type by the five subcategories
according to State proportion of occupied residential units using each type for
space heating; (3) determining State total commercial area source fuel
consumption, taking into account point source emissions; (4) normalizing State
estimates against published State consumption; and (5) determining and
allocating county consumption by "other" commercial categories. Emission
factors are taken from AP-42.
3.2.2.3 Industrial Fuel
Area source emissions generated by the industrial sector which are not
accounted for by point source categories are estimated by the following
3-25
-------
methodology. County industrial fuel consumption for bituminous coal,
distillate oil, residual oil, and natural gas is calculated. This step is
accomplished by adjusting county area source employment figures tor SIC
categories 20 through 39 by a fuel incensity factor determined by dividing the
State consumption of fuel for each SIC category by the respective State
employment. County values for fuel consumption are summed and then normalized
with State-published values for respective fuel types.
3.2.3 Mobile Sources
The following section discusses methodologies for activity level and
emission factor estimation for the following five major categories: Highway
Vehicles, Off-Highway Vehicles, Railroad Locomotives, Aircraft, and Marine
Vessels.
3.2.3.1 Highway Vehicles
For the purpose of calculating emissions, NEDS disaggregates highway motor
vehicles into four categories on the basis of use and gross vehicle weight.
Light duty gasoline vehicles are defined as gasoline-powered passenger vehicles
weighing 8500 pounds or less. Light duty gasoline trucks include gasoline
cargo vehicles weighing 8500 pounds or less. Heavy duty vehicle categories
separate diesel and gasoline-powered trucks and buses weighing more than 8500
pounds. Motorcycles, light duty diesel vehicles, and light duty diesel trucks
are assumed to contribute minor emissions relative to the four categories
above.
Fuel consumption and average fuel efficiencies are used to determine
vehicle miles traveled (VMT) for four average speed classes to reflect road
usage, namely, limited access roads (55 mph), rural roads (45 mph), suburban
roads (35 mph), and urban roads (19.6 mph). At the present time, NEDS
calculates emissions for limited access roads, rural roads, and urban roads.
Each speed class includes the following Federal Highway Administration (FiiWA)
assigned functional classes:
3-26
-------
Limited Access Roads Rural
(55 mph) Rural
Other
Rural
Rural (45 mph) Rural
R_ al
Rural
Urban (19.6 mph) Urban
Urban
and Urban Interstate
and Urban Other Principal Arterials
Freeways and Expressways
and Urban Minor Arterials
Major Collector
Minor Collector
Local
Collector
Local
Emission factors in grams per mile obtained from the execution of the EPA
MOBILE3 emission factor model are applied to determine county-level emissions
1 3
for the vehicle types and speed classes described above. County-specific
emission factors for each vehicle class are computed with standard MOBILE3
inputs and default values for most fleet and travel variables. County-specific
inputs include vehicle registration distributions, representative annual
average temperatures for each State, local inspection/maintenance program
characteristics and local VMT per speed class data. For California, MOBILE3 is
run with modified basic exhaust emission rates by model year to reflect
California emissions standards.
For highway vehicles, activity levels include fuel consumption by fuel type
for each vehicle type, and speed-class-specific annual VMT, as discussed for
each vehicle type below.
Light Duty Gasoline Vehicles (LDGV)/Light Duty Gasoline Trucks (LDCT)—
County use of gasoline by LDGV and LDGT is obtained by subtracting county
estimates of gasoline consumed by heavy duty gasoline vehicles (computed as
described below) from the total county consumption. Total gasoline consumption
reported for each State is allocated to counties by one of two methods,
depending on the availability of State-submitted data for vehicle miles
traveled in each county. For States for which county-level-measured VMT data
are available, the total State consumption is distributed to counties based on
the proportion of county to State VMT totals. For States which do not report
annual VMT by county, State consumption is allocated by the number of cars and
trucks weighing less than 6000 pounds registered in each county, adjusted by an
3-27
-------
index of rural/urban miles per vehicle. The light duty vehicle consumption
estimate is then broken down into separate estimates for LDGV and LDGT based on
registration data.
County consumption estimates for each vehicle group are then converted to
total vehicle miles traveled using fuel efficiency figures in miles per
gallon. Total VMT for each vehicle type are then allocated to each of three
speed classes (limited access roads, rural roads, and urban roads), according
to the miles of each road type constructed in the county relative to the miles
of each type constructed in the State.
Heavy Duty Gasoline Vehicles (HDGV)—County gasoline consumption by heavy
duty vehicles is determined by calculating county estimates of gasoline
consumed by three truck weight classes and institutional buses. For trucks,
county truck registrations for each weight class are multiplied by the average
annual miles traveled by each weight class in each State and divided by the
national average weight class fuel efficiency. State gasoline consumption by
institutional buses is calculated using State bus registrations and average
gasoline consumption (gallons per year), and is then allocated to counties by
current county population. County-level truck consumption estimates are
multiplied by average fuel efficiencies and summed to yield total State VMT for
HDGV. Total State HDGV VMT are allocated to county speed classes based on the
county miles of each speed class.
Heavy Duty Diesel Vehicles (HDDV)—Much of the fuel consumption of HDDV is
accrued outside the county of registration. To account for this behavior, the
methodology makes separate HDDV fuel consumption estimates for long-range
travel and short-range travel. Published State consumption is allocated to the
county level on the basis of total, out-of-state, and local VMT, using survey
data on annual miles traveled and percentage of the miles traveled outside the
State for diesel trucks over 8500 pounds.
Long-range VMT estimates for each State are summed to form a national HDDV
long-range VMT pool. The national long-range VMT pool is allocated to counties
by estimated county fractions of total National Network mileage. Short-range
VMT are allocated to the county level on the basis of truck registrations.
3-28
-------
Short-range VMT and long-range VMT are then totaled for each county and
multiplied by the average fuel efficiency to obtain fuel consumption by HDDV.
Each county's long-range HDDV VMT are assumed to occur on limited access roads;
short-range HDDV VMT are divided equally between rural and urban roads.
3.2.3.2 Off-Highway Vehicles
Off-highway vehicles fall into six general categories: farm equipment,
construction equipment, industrial equipment, motorcycles, lawn and garden
equipment, and snowmobiles. Gasoline is consumed by all six categories, while
dieseL fuel is utilized only by farm equipment, construction equipment, and
industrial equipment.
Emission factors for all gasoline and diesel off-highway vehicles are taken
from AP-42, except for motorcycle emission factors, which are estimated by the
MOBILE2 model. Separate reactivity profiles for gasoline and diesel fuel are
used to estimate the reactive portion of total VOC emission estimates. All
off-highway emission factors are based on fuel consumption. Consumption
estimation methodologies are described for each vehicle category and fuel type
below.
Farm Equipment—To estimate State fuel consumption by farm equipment,
consumption values are calculated separately for farm tractors, combines,
motorized balers, forage harvesters, and general purpose large utility engines.
Consumption by fuel type is calculated using State populations for each type of
equipment, average annual usage (hours per year), and average hourly
consumption by fuel type per unit (gallons per hour). For diesel fuel, the sum
of the estimated fuel use for all subcategories is normalized to agree with
published State totals for agricultural diesel fuel use. Total State gasoline
and diesel fuel consumption is then allocated to the county level according co
the ratio of county tractor population to State tractor population.
Construction Equipment—National gasoline consumption for construction
equipment is estimated by OAQPS, while published national totals are available
for diesel fuel. National consumption of each fuel is apportioned to States
3-29
-------
according to total non-buiI ding construction employment in each State. State
totals for diesel fuel are normalized to agree with DOE published totals for
construction equipment. County consumption of fuel is then allocated from
State construction consumption on the basis of county population.
Industrial Equipment—For industrial equipment, national fuel use figures
(obtained as described for construction equipment) are apportioned to counties
according to relative differences between combined county employment and
combined national employment in the manufacturing, mining, and wholesale trade
industries. State totals for diesel fuel are normalized to agree with DOE fuel
use statistics.
Motorcycles—County-level gasoline consumption is estimated with an
algorithm based on population, State motorcycle registrations, average annual
usage, and average fuel consumption rate. The algorithm separates off-road and
combined use motorcycles, and weights the distribution of the two types
according to regional variations. Emission factors are calculated using data
representing uncontrolled emissions in the EPA M0BILE2 model.^
Lawn and Garden Equipment—National consumption of gasoline by lawn and
garden equipment is estimated by NADB and allocated to individual counties
based on the number of single-unit dwelling structures, the number of
freeze-free days annually, the fraction of national snow zone population in the
county, snowthrower fuel consumption rate, average snow removal rate, and
county snowfall.
Snowmobiles—County consumption of gasoline by snowmobiles is derived from
the OAQPS-established national snowmobile gasoline consumption levels, and
allocated on the basis of estimated county snowmobile population. Snowmobile
population is based on algorithms relating the percent of State snowmobiles
used in the county to population and snowfall, and taking into account the
impact of population density on snowmobile usage.
3-30
-------
3.2.3.3 Railroad Locomotives
This category includes fuel utilized by railroad locomotives and fuel used
by railroad stations and workshops for space heating. The latter fuel
consumption has been incLuded primarily because it is difficult to separate
from total railroad fuel use and is considered minor compared to locomotive
consumption. The emission factors for railroad fuel use are taken from AP-42.
The primary fuel consumed by railroad locomotives is distillate oil (diesel
fueL). Published State consumption of diesel fuel by railroad locomotives is
allocated to the county level on the basis of current population distribution.
3.2.3.A Aircraft
Emissions estimates for aircraft are divided into three categories: civil
aircraft, commercial aircraft, and military aircraft. Estimates of aircraft
Landing and tak.e-off cycles (LTOs) by county, based on operation records from
county airports or aircraft registration data, are multiplied by emission
factors based on LTOs to obtain emissions estimates.
Weighted average emission factors are computed for each type of aircraft
within each aviation category. In some categories, flying hours are used as a
unit of measure, under the assumption that the number of flying hours is
proportional to the number of LTOs. Emission factors are then combined using
aircraft type population data to form one factor for each pollutant.
Military and Civilian Aircraft—Initial emission factors are averaged and
weighted by usage and population data for six aircraft types. Emission factors
for each aircraft type are taken from AP-42.
Commercial Aircraft—Emission factors for commercial aircraft are
calculated separately for air taxi and commercial service. Air taxi emission
factors are population-weighted averages of AP-42 emission factors for
turbojets, turboprops, and piston planes. Commercial service aircraft emission
factors from AP-42 are updated and weighted using the previous year's NEDS data
on LTOs and population data. The number of operations is estimated using the
3-31
-------
number of aircraft in service for nine
compared with reported values obtained
(FAA). The weighting factors are appli
average for all plane types.
3.2.3.5 Marine Vessels
plane types. The resulting values are
from the Federal Aviation Administration
ed to the emission factors to produce an
Marine vessel categories include distillate oil (diesel) vessels, residual
oil vessels, and gasoline vessels. Consumption methodologies and emission
factor derivation are presented for each category below.
The diesel vessel category includes large cargo and passenger ships, oil
tankers, tugboats, and other steamships and motorships that are known to
consume distillate oil. Estimates of county-level fuel consumption were
originally based on numbers, types, and sizes of ships, and on time spent in
port and underway. Consumption by vessels at ports was assigned to the port
counties where data were available. The remaining fuel consumption was
distributed to ports and waterways according to tonnage handled. Estimating
current consumption of distillate oil by marine vessels requires the updating
of previous county estimates using DOE State vessel fuel use data, excluding
fuel used by ships outside the U.S. continental limits.
The residual oil vessels category includes large cargo and passenger ships,
oil tankers, and tugboats. Historic county-level residual oil consumption
estimates were based on the number, types, and sizes of ships, and on time
spent in port and underway. Consumption by vessels at ports was assigned to
the port counties where data were available. The remaining fuel consumption
was distributed to ports and waterways according to tonnage handled. Current
consumption estimates required the updating of previous county estimates with
State-level DOE data on residual oil use by bunkering vessels, excluding fuel
used by ships outside the U.S. continental limits.
For distillate oil vesseLs, emission factors are the weighted averages of
AP-42 factors for commercial diesel motorships and steamships. For residual
oil vessels, emission factors are the weighted average of AP-42 emission
factors for commercial residual oil motorships and steamships. For both diesel
and residual fuels, the weighting procedure uses the following assumptions.
3-32
-------
Commercial vessels include 75 percent moiorships and 25 perceni steamships.
Commercial steamships spend 80 percent of the cime hotelling and 20 percent
under full power. DieseL steamships spend 20 percent of the time under
auxiliary power and 80 percent underway.
State gasoline vessel fuel consumption is derived from State boat
registration (inboard and outboard), and average fuel consumption for each boat
type. State consumption is then allocated to counties according to county
inland water area, coastline, and the number of warm months suitable for
recreational boating activities. Average weighted emission factors for
gasoLine vessels are based on inboard and outboard motorboat registrations.
Weighting accounts for higher fuel consumption per hour of operation by inboard
motors.
For all vessel types, VOC emission factors are adjusted to reflect
appropriate species profiles, and efforts are made to exclude operations
conducted outside the continental U.S.
3.2.4 Solid Waste Disposal
The area source category for solid waste disposal includes on-site refuse
disposal activities by residential, commercial/institutional, and industrial
sectors. In this section, emissions from the disposal practices of open
burning and on-site incineration are discussed separately. Solid waste
generation in hundreds of tons is used as a measure of activity level.
3.2.4.1 On-Site Incineration
For the purposes of determining the amount of solid waste generated,
on-site incineration is defined as disposal in a small incinerator,
encompassing the following types of disposal units: backyard burners;
industrial incinerators} and incinerators used by food and department stores,
hospitals, and schools. Since large municipal incinerators are usually
classified as point sources, emissions resulting from disposal in this type of
incinerator have not been included in this category. The quantity of solid
waste generated by each sector was estimated for the base year 1976 using
3-33
-------
population scatistics and per capita generation facLors for each EPA region.
Since 1976, trie previous yea^s estimates of waste generated by each sector
have been updated each year according to the reLative national percentage
increase or decrease in the amount of waste generated (or incinerated) by NEDS
point sources in each respective sector. For the commercial/institutional and
industrial sectors, NEDS calculations use the annual increase or decrease in
waste incinerated by SCC point source categories within each sector. The
annual residential update factor is based on engineering judgment and
calculations by NADB. County allocation is based on population.
Adjustments are made to county estimates based on information about
specific point sources and data submitted by States. If a number of on-site
incinerators have been identified as point sources, it might be appropriate to
reduce or eliminate area source estimates. Also, it is important to note that
State estimates of waste generated replace the extrapolated data for the year
they are submitted. Submitted data are then annually updated by the above
method using the relative percentage increase in waste generated. Emission
factors for intermediate-size incinerators from AP-42 are used for all on-site
incineration.
3.2.4.2 Open Burning
For the purposes of estimating open burning practices, the term "open
burning" refers to uncombined burning of wastes such as leaves, Landscape
refuse, and other rubbish. Large open burning dumps are usually included under
point sources.
The quantity of solid waste burned is computed by updating the previous
year's waste generation for each sector in a manner analogous to updates for
On-Site Incineration. The update factor is determined by engineering judgment.
Estimates of the quantity of solid waste burned by point sources in the most
recent year are obtained from the NEDS point source data. County allocation is
based on population.
The emission faccors for open burning of refuse and organic macerials are
taken directly from AP-42.
3-34
-------
3.2.5 Miscellaneous Area Sources
Area sources which are not defined by Stationary Sources, Mobile Sources,
or Solid Waste categories are compiled in the Miscellaneous Area Sources
category. Although total emissions from each source are relatively small
compared to those from the three major categories, emissions from each
miscellaneous category may be significant because of size, geographic
distribution, or periodic intensity over time.
3.2.5*1 Evaporative Losses From Gasoline Marketing
This source category covers evaporative losses of volatile organic
compounds from gasoline marketing operations, such as filling losses from
loading underground storage tanks at service stations, and spillage and filling
losses from filling automobile tanks. Losses from refineries and bulk
distribution terminals are excluded from this category, because emissions from
refineries and terminals are assumed to be accounted for in point source
categories and in Additional Area Sources. Emissions are calculated by
multiplying emission factors by the activity level for this category, measured
by retail gasoline sales. Retail sales of gasoline include all sales of
gasoline for highway, marine, and aviation use, and for use by the construction
equipment, industrial equipment, and farm equipment off-highway subcategories.
County retail gasoline sales are used directly when reported by States. For
counties for which retail sales of gasoline are not compiled, sales to the
above user categories are estimated separately and summed to generate total
county sales.
State retail sales of gasoline for highway and marine use are allocated to
each county according to the county's proportion of the statewide gross dollar
receipts from gasoline service stations. Published State aviation retail sales
of gasoline are allocated to the county according to ,the total LTO cycles in
the county for each of the military, civilian, and commercial aircraft
categories.
County retail sales of gasoline for off-highway sources are assumed to be
Che same as the consumption derived in the activity levels section of Farm
3-35
-------
Equipment, Construction Equipment, and Industrial Equipment in Off-Highway
Sources (3.2.3.2).
Emission factors for gasoline marketing are obtained from AP-42, weighted
by an assumed distribution of types of filling practices used.
3.2.S.2 Evaporative Losses From Organic Solvent Consumption
Area source evaporation from organic solvent usage is divided into
six major categories: dry cleaning operations, degreasing operations, surface
coating application operations, printing, rubber and plastics, and other
miscellaneous uses. In each category, usage of specific solvents is identified
and enumerated to compute total solvent usage in tons per year. Eventual
evaporation of all solvents is assumed so that solvent usage is equivalent to
VOC emissions.
The methodology for allocating organic solvent consumption by county
consists of apportioning national consumption of nineteen primary solvent
groups by major user category according to county population or user category
employment data. User categories are listed in Table 3-3. Two of the major
user categories, Surface Coatings and Other Uses, are further classified into
subcategories as shown. Table 3-4 contains a list of the primary solvent
groups used to determine losses from organic solvent consumption. The category
"Special Naphthas" includes the aliphatic naphthas such as V. M. & P. naphthas,
Stoddard solvents, rubber solvents, and mineral spirits.
National consumption of the primary solvent groups is distributed to each
of the user categories according to the user category's percent of total
solvent consumption. Percentage usage obtained from published sources is
compiled for each user category. National area source solvent use estimates
are determined by subtracting point source solvent use or emissions for each
user category from total solvent use for each user category.
County consumption for each solvent group and user category is then
computed by allocating calculated national area source consumption on the basis
of county area source employment in applicable SICs (see Table 3-3) or by
population. Area source employment is determined by subtracting point source
employment from total county employment for each SIC category. To reflect
3-36
-------
TABLE 3-3. AREA SOURCE ORGANIC SOLVENT USER CATEGORIES
User Categories Population or Employment Data by
SIC Used For County Allocation
Surface Coatings
Trade Paints
Auto Refinishing
Automotive
Wood Furniture and Fixtures
Metal Furniture and Fixtures
Metal Containers
Sheet Strip and Coil
Appliances
Machinery and Equipment
Paper
Factory-Finished Wood
Transportation (Non-Auto)
Electric Insulation
Other, Exterior, Interior
Marine
Deereasing
Drv Cleaning
Printing
Rubber and Plastics
Other Miscellaneous Use
County Population
7535 (Paint Shops)
371 (Motor Vehicles)
25 (Furniture and Fixtures)
25 (Furniture and Fixtures)
34 (Fabricated Metal Products)
34 (Fabricated Metal Products)
35 and 36 (Machinery, Electrical
Equipment and Supplies)
26 (Paper and Allied Products)
243, 244 (Millwork, Plywood-
Related Supplies, Wooden
Containers)
37 (Transportation Equipment)
Less 371 (Motor Vehicles) and
373 (Shipbuilding Repair)
36 (Electrical Equipment and
Supplies)
19-39 (Total Manufacturing)
373 (Shipbuilding and Repair)
34-39 (Metal Products, Machinery,
Transportation Equipment,
Instruments, Miscellaneous)
2 x 7216, Plus 7215 and 7218 (Dry
Cleaning and Combination with
Wet Laundering)
264, 265, and 27 (Paper
Products, Containers,
Printing and Publish-
ing)
30 (Rubber and Plastics)
1/2 of 19-39 Employment
+ 1/2 County Employment
3-37
-------
TABLE 3-4. AREA SOURCE ORGANIC SOLVENTS
Special Naphthas
Perchloroethylene
Ethanol
Trichloroethylene
Isopropanol
Acetone
Glycol Ethers
Cyclohexanone
Methyl Ethyl Ketone
Ethyl Benzene
Propylene Glycol
Methanol
Butyl Acetate
Ethyl Acetate
Butyl Alcohols
Methyl Isobutyl Ketone
Monochlorobenzene
o-Dichlorobenzene
p-Dichlorobenzene
3-38
-------
unequal solvent use between establishments within SIC groups, consumption is
multiplied by a factor which compares the number of individuals in the county
in each area source user category to the number of individuals in the nation in
each area source user category. County consumptions of all solvent types are
Chen summed to yield a total county consumption.
3.2.5.3 Unpaved Roads
Vehicle traffic over unpaved roads, parking areas, and recreational areas
generates fugitive dust emissions which are estimated in NEDS. Primary factors
which affect the amount of dust generated are vehicle speed, surface type, wind
speed, surface moisture, and type of vehicle. The activity level (in vehicle
miles traveled) is calculated using county population and mileage of unpaved
roads. The emission factor is derived using an equation in AP-42.
3.2.5.4 Unpaved Airstrips
Unpaved airstrip emissions are affected by the same primary factors as
unpaved roads. Fugitive emissions from unpaved airstrip use are measured by
annual landing-take off (LTO) cycles on airstrips made of dirt, sand, gravel,
or gravel pavement, excluding airports with no based aircraft, airports no
longer in operation, heliports, and seaplane bases in each county. The
activity level estimate derived for each county in NEDS is multiplied by an
adjusted emission factor from AP-42 to obtain a particulate emissions estimate.
3.2.5.5 Forest Wildfires
Each year emissions are generated by forest wildfires covering large tracts
of forested land. For this category, emissions estimates are generated by
multiplying the number of acres burned per county by a fuel loading factor and
emission factors from AP-42. VOC emissions are adjusted to include only
reactive species by assuming 79.9 percent by weight of total hydrocarbons are
reactive. Since 1974, the NEDS wildfire activity level for each county from
the previous year has been updated with wildfire statistics from the U.S.
3-39
-------
Forest Service. Regional fuel loading factors in ions per acre for each EPA
Region from AP-42 are applied to State averages within each Region to yield
tons consumed.
3.2.5.6 Managed Burning
Managed burning activities include slash burning and prescribed burning.
In slash burning operations, wastes from logging operations are burned under
controlled conditions to reduce fire hazard and remove brush considered to host
destructive insects. Prescribed burning is used as a forest management
practice to establish favorable seed beds, remove competing underbrush,
accelerate nutrient cycling, control tree pests, and contribute other
ecological benefits.
For this category, emissions estimates are generated by multiplying the
number of acres burned in each county by a fuel loading factor and the emission
factor for each pollutant. Original State estimates of acreage consumed by
both managed burning techniques were determined for the NEDS inventory year of
1974. Individual State officials and the U.S. Forest Service were contacted to
provide estimates of acreage burned, burning technique, and fuel loading
ratios. The 1985 NAPAP Emissions Inventory utilizes State data generated for
1974 which were allocated to the county level according to forest acreage per
county, as obtained from contact with local officials or State land usage maps.
If not provided, fuel loadings for slash burning and prescribed burning are
assumed to be 75 tons per acre and 3 tons per acre, respectively. Particulate
and CO emission factors are obtained from the Source Assessment.^ SO2 and NOx
emission factors are taken directly from AP-42.
3.2.5.7 Agricultural Burning
This miscellaneous area source category estimates emissions from
agricultural burning practices routinely used to clear and/or prepare land for
planting. Specific operations include grass stubble burning, burning of
agricultural crop residues, and burning of standing field crops as part of
harvesting (e.g., sugar cane).
3-40
-------
Emissions estimates are generated by multiplying the number of acres burned
in each county by a fuel loading factor and the emission factor for each
pollutant. The original estimates for 1974 measured activity level in terms of
acres burned per State. It is assumed that the total quantity of agricultural
products burned in 1974 is the same quantity which will be consumed by fire
each year. If no specific crop data were available, it was assumed that the
number of acres burned annually is divided equally between sugar cane and
field crops. Emission factors are taken from the Procedures Document for
Development of National Pollutant Emissions Trends Reports*^ and AP-42.
3.2.5.8 Structural Fires
Structural fires have been included in NEDS because building fires have
been linked to short-term emissions of air contaminants. The activity level
for this category, measured by the total number of fires per councy, is
multiplied by a loading factor and an emission factor to obtain emissions
estimates. In the absence of county-level data, a national average of four
fires per 1,000 population is assumed to occur each year.^ Estimates of the
material burned are obtained by multiplying the number of structural fires by a
fuel factor of 6.8 tons of material per fire.*^ Emission factors are taken
from the OAQPS Technical Tables.^
3.2.6 Additional Area Sources
The 1985 NAPAP Emissions Inventory will provide detailed county level VOC
emissions estimates for additional area sources which previously have not been
included in the NEDS area source categories. This section presents methods
which have been developed for many categories which have been traditionally
considered point source categories, such as Bakeries and Synthetic Fiber
Manufacturing. These categories were included to reconcile the difference
between the total emissions reported in the National Air Pollutant Emissions
gqtimates 1940-1984^ and the emissions already accounted for by the NEDS point
source data files. The remaining categories such as Publicly-Owned Treatment
Works (POTWs) and hazardous waste Treatment, Storage, and Disposal Facilities
3-41
-------
(TSDFs) have been included due to the difficulty of measuring emissions from
specific points within these categories (e.g., aeration basins). Because the
additional categories are believed to generate significant VOC emissions,
existing methodologies and data used by NEDS have been improved to provide
accurate emissions estimates.
In this section, methodologies for estimating VOC emissions are presented
for the following area sources: POTWs and hazardous waste TSDFs; fugitive
emissions from synthetic organic chemical manufacturing; bulk terminals and
bulk plants; fugitive emissions from petroleum refining operations; process
emissions from bakeries, pharmaceutical, and synthetic-fiber manufacturing;
crude oil and natural gas production fields; and cutback asphalt paving
operat ions.
For most categories, national VOC emissions are allocated to the county
level to produce county VOC emissions estimates. Activity levels, emission
factors, and control efficiencies are used to determine emissions for the
remaining sources.
3.2.6.1 Publicly-Owned Treatment Works (POTWs)
The published national VOC emissions estimate for the worst case scenario
for unacclimated treatment systems was selected for use in the calculation of
county VOC emissions in the 1985 NAPAP Emissions Inventory. Because research
on VOC concentrations in POTW influents and effluents indicates that the
removal mechanisms for these pollutants are relatively constant and only a
percentage of incremental loadings is removed by POTWs, the national VOC
emissions estimate for unacclimated treatment systems is allocated to the
county level based on the percentage of industrial flow per county. To
eliminate double counting, emissions accounted for by point source categories
are subtracted from the national emissions estimate before county allocation.
The total VOC emissions from POTWs for the nation are estimated in the EPA
18
Domestic Sewage Study.
3-42
-------
3.2.6.2 Hazardous Waste Treatment, Storage, and Disposal Facilities (TSDFs)
National VOC emissions estimates are developed using facility-specific
process descriptions, waste characterization and quantities, and process-
19
specific emission factors. Emissions from all TSDFs in the U.S. are summed
at Che county Level to form the national emissions estimates. Emissions at a
TSDF are defined as the total VOCs emitted by all plant processes; emissions
from each plant process are calculated as the product of the quant ity of a
specific waste handled and a process-specific emission factor. To eliminate
double counting, emissions accounted for by point source categories are
subtracted from the county-level emissions estimate.
3.2.6.3 Fugitive Emissions from Synthetic Organic Chemical Manufacturing
The fugitive emissions from synthetic organic chemical manufacturing are
estimated using OAQPS estimates of national emissions from the manufacture of
petrochemicals. Assuming that the potential for fugitive emissions (i.e., the
number of pumps, valves, flanges, etc.) increases with the number of employees,
the national VOC emissions are allocated to specific counties based on the
ratio of the county to national employment in SIC 2869 (Industrial Organic
Chemicals - Not Classified).
Currently, NEDS point source entries under several Source Classification
Codes (SCCs) already account for a portion of fugitive emissions, including
3—01-800 (General Processes-Fugitive Leaks), 3-01-888 (Fugitive Emissions-Not
Classified), and many chemical production-specific SCCs designated as Fugitive
Emissions: General. The total VOC emissions for these SCCs are summed by
county and then subtracted from the national emissions estimates. If a county
jjaS greater point source VOC emissions reported in NEDS than the overall
emission, level allocated from the national fugitive VOC emissions estimate, it
is assumed that the fugitive VOC emissions for area source emission levels are
adequately represented by point source VOC emissions data reported in NEDS.
The national sum of all negative emissions is reallocated to the counties
exhibiting positive emission levels based on the relative proportion of
employment for SIC 2869 located in each county.
3-43
-------
3.2.6.4 Balk Terminals and Bulk Plants
Current methodology uses national VOC emissions estimates for gasoline bulk
terminals and bulk plants based on total annual throughput and assumes that
control is negligible at bulk plants. The national VOC emissions estimate is
allocated to the county level based on Che county employment in SIC 5171. The
VOC county emissions totals for the point source data files for SCC 4-04-001
(Bulk Terminals) and SCC 4-04-002 (Bulk Plants) are then subtracted from the
portion of the corresponding national VOC emissions estimate. Negative
emissions are reallocated to the counties as previously described, using
employment data for SIC 5171.
3.2.6.5 Fugitive Emissions from Petroleum Refinery Operations
Annual national VOC emissions estimates for petroleum refinery operations
have been developed by OAQPS using capacity or production data. Under the
assumption that these national VOC emissions estimates represent the sum of the
fugitive and point source emissions for refinery processes, fugitive emissions
are quantified by subtracting the total county point source VOC emissions of
SCC categories 3-06-004 through 3-06-88 (Refinery Processes, excluding process
heaters and catalytic cracking units) from the total emissions estimated for
petroleum refinery processes in each county. National emissions data are
allocated to each county based on refinery capacity. Negative emissions are
reallocated to counties as previously described, using refinery capacity data.
3.2.6.6 Process Emissions from Bakeries
Annual national VOC emissions estimates for bakery operations are
allocated to each county based on the county employment census data for SIC
categories 2051 and 2052. Total county point source VOC emissions for SCC
3-02-032 (Bakeries) listed in NEDS are subtracted from the national VOC
emissions estimate allocated to each county. Negative emissions are
reallocated to counties as previously described, using employment data for SIC
categories 2051 and 2052.
3-44
-------
3.2.6.7 Process Emissions from Pharmaceutical Manufacturing
Annual national VOC emissions estim. les from pharmaceutical manufacturing
operations developed by OAQPS are allocated to specific counties based on the
county employment census data for SIC category 2834. Total point source VOC
emissions for SCC 3-01-060 (Pharmaceutical Preparations) listed in NEDS- for
each county are subtracted from the national VOC emissions estimate allocated
to each county. Negative emissions are reallocated to counties as previously
described, using employment data for SIC 2834.
3.2.6.8 Process Emissions from Synthetic-Fibers Manufacturing
The annual national VOC emissions estimate from synchecic-fibers
manufacturing operations is allocated to the county level based on the
combined county employment for SIC categories 2823 and 2824. NEDS point source
VOC emissions for SCC 3-01-024 (Synthetic Organic Fiber Production) and
3-01-025 (Cellulosic Fiber Production) for each county are then subtracted from
che national emissions estimate allocated to each county. Negative emissions
are reallocated to counties as previously described, using employment data for
SIC categories 2823 and 2824.
3.2.6.9 Crude Oil and Natural Gas Production Fields
Annual national VOC emissions estimates for crude oil and natural gas
production developed by OAQPS are distributed to the State level in proportion
to the volume of State annual production. The State VOC emissions are then
allocated to specific counties based on county employment for SIC 1310. VOC
emissions for these sources currently accounted for in NEDS point source data
files for SCC 3-10-001 (Crude Oil Production) and SCC 3-10-002 (Natural Gas
production) are subtracted from the estimates. If point source emissions
exceed the total emissions estimates, then negative emissions are reallocated
co counties as previously described, using employment data for SIC 1310.
3-45
-------
3.2.6.10 Cutback Asphalt Paving Operations
State VOC emissions for cutback asphalt paving operations are calculated by
multiplying an emission factor by the activity level, measured in total
quantity of cutback asphalt sales. The State emissions totals are then
allocated to specific counties based on employment for SIC 1611. The VOC
emission factor is based on the weight of asphalt used, assuming complete
evaporation of all organic solvents used in paving operations.
3.3 NONCRITERIA POLLUTANTS
3.3.1 Noncriteria Pollutant Inventory
The 1985 NAPAP Emissions Inventory of anthropogenic sources focuses on
three NEDS criteria pollutants: SO2, N0X, and VOC. Although SO2, N0X, and
VOC are considered the primary precursors of acid deposition, other pollutants
are also regarded as significant. Four non-criteria pollutants included in the
O _
1985 Emissions Inventory are primary sulfates (SO4 ), hydrogen chloride (HC1),
hydrogen fluoride (HF), and ammonia (NH3). Estimates for these pollutants are
not supplied by the States, as there is no extant reporting structure.
Historically, EPA has not collected emissions data on these pollutants, and
neither EPA nor the States have the capability to develop source-specific
inventories. Therefore, this sector of the 1985 NAPAP Emissions Inventory
represents a unique effort by EPA to develop emission factors and make
emissions estimates. Estimates for these four noncriteria pollutants were
made at the request of the NAPAP modeling community and the Western Governor's
Association Acid Rain Study Group. In order to create as comprehensive a data
base of acid rain precursors as possible, estimates for the pollutants were
made and have been included in the report. Due to a lack of standardized,
reliable sources of data for these pollutants, EPA developed emission factor
on 21 ?? 21
reports for these pollutants. ''' These reports analyzed existing data
for a variety of source categories and recommended emission factors appropriate
for use in the 1985 NAPAP Emissions Inventory. These factors were rated
qualitatively, in a manner similar to the way that AP-42 factors are rated.
3-46
-------
These documents may be referenced for further information on these factors.
These emission factors and resulting emissions estimates represent the
first time that a comprehensive emissions inventory for primary sulfates,
hydrogen chloride, hydrogen fluoride, and ammonia has been attempted by EPA.
These estimates should not be considered to be as accurate or reliable as the
criteria pollutant inventory. The estimates were not generated or reviewed by
Che States or the sources, and although operating rate data were provided by
the States, errors in the data will affect these emission estimates.
Emissions estimates for noncriteria pollutants from existing NEDS point
sources were developed using SCC-level emission factors applied to NEDS
process-level throughput data. Applicable SCCs were identified during review
of emission factors. It should be emphasized that, in contrast to the criteria
pollutant estimates, the resulting emissions estimates are not the result of
estimates made by the States.
Emissions of SO4 , HC1, HF, and NH3 from area sources were also estimated.
Emission factors were developed to be used with the area source estimation
methodologies, and these factors were combined with activity levels obtained at
the county level.
Where applicable, emission factors account for control practices used
within the relevant source categories. Emission factors consider average
control efficiencies and prevalence of controls in a source category.
Although actual emissions at the plant and point-process level will be under-
or over-estimated to the extent that control practices and efficiencies deviate
from the industry averages, these assumptions were necessary since there is no
structure available for States to report non-criteria pollutant control devices
and efficiencies to NEDS.
3.3.2 Emissions Estimation Methods
3.3.2.1 Primary Sulfates
Primary sulfate is emitted directly from emission sources as SO^2-, unlike
secondary sulfate, which is derived from the atmospheric transformation of SO2.
Source categories that emit primary sulfate include external combustion,
3-47
-------
chemical manufacturing, primary metals, mineral products, and petroleum
refining. Sulfate emission factors were compiled for these categories as part
of the 1985 NAPAP Emissions Inventory effort (Table 3-5).^®
Two previous programs have made assessments of sulfate emission factors:
the Electric Power Research Institute's (EPRI's) sulfate regional experiment
and the United States/Canada Work Group 3B. The NAPAP effort reviewed these
two assessments and also took, advantage of the recent acceptance of a standard
sampling and analysis procedure by the source emissions measurement technical
community. This method, controlled condensation sampling (CCS), is currently
considered to be the most accurate approach to measuring sulfate from
stationary sources. CCS-derived measurements were abstracted from the
literature and primary sulfate emission factors were calculated.
Calculation of source-specific sulfate emission factors was based on a
hierarchical selection process:
1) Where available, field measurements using the CCS
procedure were considered as the prime data set,
2) Sulfate emissions assessments were aggregated for
different point sources within the same category only if
fuel composition and emissions controls were similar, and
3) Non-CCS emissions data were used only if multiple
measurements produced data with minimal variation.
NEDS throughput data were multiplied by appropriate emission factors for
relevant SCCs to produce process-level emissions estimates for the 1985 NAPAP
Emissions Inventory.
3.3.2.2 Hydrogen Chloride
HC1 is emitted from coal combustion, waste incineration, and organic
chemical manufacture (Table 3-5). The primary source is coal combustion.
NAPAP compiled and rated emission factors from the literature for these source
21
categories for use with the 1980 NAPAP Emissions Inventory. Emission
estimates were calculated by EPA based on these emission factors and throughput
data at the SCC-level for the relevant point source categories (SCCs).
3-48
-------
TABLE 3-5. EMISSIONS SOURCES OF PRIMARY SULFATE, HYDROGEN CHLORIDE,
HYDROGEN FLUORIDE, AND AMMONIA IN THE NAPAP INVENTORY.
SOURCE CATEGORY S042" HC1 HF NH;
Combustion
Coal X X X X
Distillate Oil X
Residual Oil X
Natural Gas X
Wood/Bark Waste X
primary Metals
Copper X
Zinc X
Aluminum X X
Iron Production X
Coke Production X X
Petroleum Industry
FCC X X
TCC X
Claus Plants X
Engine Compressors X
Wood Products
Kraft Pulp Mill X
Sulfite Pulp MilL X
Mineral Products
Cement Manufacture X
Gypsum Manufacture X
Chemical Manufacturing
Sulfuric Acid - Contact Process X
Propylene Oxide X
By-product HC1 Production X
HF Production X
Ammonium Nitrate X
Phosphate Fertilizer X
Ammonia Synthesis X
Urea Manufacture X
Ammonium Phosphate X
Incineration X
Area Sources
Mobile Gasoline Combustion X
Mobile Diesel Combustion X
Anhydrous NH3 Fertilizer Application X
Livestock Waste X
3-49
-------
3.3.2.3 Hydrogen Fluoride
The primary source of l!F is coal combustion; HF is also emitted from
hydrogen fluoride manufacture, the primary aluminum industry, and the phosphate
fertilizer industry (Table 3-5). NAPAP compiled and rated emission factors
from the literature for these source categories for use with the 1985 NAPAP
9 1
Emissions Inventory. Emissions estimates were calculated by EPA based on
these emission factors and throughput data at the SCC-Level for the relevant
point source categories.
3.3.2.4 Amnion i a
NH3 is emitted in significant quantities from both point and area sources.
Major anthropogenic source categories include field application of livestock
wastes, beef cattle feedlots, fertilizer manufacture and use, mobile and
stationary fuel combustion, ammonia synthesis, petroleum refining, waste water
treatment, and coke manufacture (Table 3-5). Emission factors for the
22
categories were developed from the literature by NAPAP. AP-42 data were used
to characterize emissions from fertilizer manufacture, ammonia synthesis,
petroleum refining, and coke manufacture.
Before being included in the 1985 NAPAP Emissions Inventory, NH3 emission
factors developed for the inventory were compared uo similar factors developed
for inventories by Environment Canada, EPRI, and the National Aeronautics and
Space Administration (NASA). The factors selected were deemed the most
appropriate available, based on criteria concerning test method validity, and
currentness and representativeness of the data.
Point source category emissions were calculated by EPA using throughput
data at the SCC-level for the relevant categories. Area sources constituted
the majority of NH3 emissions; estimates were derived from activity levels
specific to the category. Where no activity levels could be developed, no NH3
estimate was made.
3-50
-------
REFERENCES FOR SECTION 3
1. NEDS Source Classi£icacion Codes and Emission Factor Listing.
Prepared by U.S. Environmental Protection Agency, Office of ir
Quality Planning and Standards, National Air Data Branch,
October 1985.
2. Report of a Workshop to Review Requirements for the 1985 NAPAP
Emissions Data Base (April 3-4, 1986, Washington, DC). Prepared by
the NAPAP Emissions and Controls Task Group.
3. NEDS/NAPAP Emission Inventory Workshop Handbook. Prepared by U.S.
Environmental Protection Agency, Office of Air Quality Planning and
Standards, National Air Data Branch, October 1985.
4. EPA Emissions Confirmation Guide for Major Source Categories.
Approved by U.S. Environmental Protection Agency, Office of Air
Quality Planning and Standards, National Air Data Branch, October
1985.
5. Inventory of Power Plants in the United States, DOE/EIA-0095(85),
U.S. Department of Energy, Energy Information Administration,
National Energy Information Center, Washington, DC, August 1986.
6. Memorandum dated December 21, 1985, from John Fink., U.S.
Environmental Protection Agency, Office of Air Quality Planning and
Standards, National Air Data Branch, to EPA Regional Office Emission
Inventory Contacts in Regions I-X.
7. Memorandum dated January 17, 1986, from John Fink, U.S.
Environmental Protection Agency, Office of Air Quality Planning and
Standards, National Air Data Branch, to EPA Regional Office Emission
Inventory contacts in Regions I-X.
8. Receptor Model Source Composition Library, EPA-450/4-85-002
(NTIS PB85-228623), 1985.
9. VOC Species Data Manual (Second Edition). EPA-450/4-80-015
(NTIS PB81-119455}, July 1980.
10. Compilation of Air Pollutant Emission Factors, Volume I: Stationary
Point and Area Sources. AP-42. Fourth Edition
(CPO No. 055-000-00251-7), U.S. Environmental Protection Agency,
Research Triangle Park, NC, September 1985.
11. D.A. Pahl and J.D. Mobley, Quality Assurance and Quality Control Plan
for the NAPAP 1985 Emission Inventory. EPA-600/8-86-025 (NTIS
PB86-237682), August 1986.
3-51
-------
12. J.L. Demmy and T.E. Warn, Area Source DocumentalLon for the 1985
National Acid Precipitation Assessment Program Inventory, Final
Report, U.S. Environmental Protection Agency, Office of Air Quality
Planning and Standards, Research Triangle Park, NC, September 1987.
13. User's Guide to MOBILE3 (Mobile Source Emissions Model),
EPA-460/3-84-002 (NTIS PB84-213974), U.S. Environmental Protection
Agency, Office of Mobile Sources, Ann Arbor, MI, June 1984.
14. User's Guide co MOBILE2 (Mobile Source Emissions Model),
EPA-460/3-81-006 (NTIS PB81-205619), U.S. Environmental Protection
Agency, Office of Mobile Sources, Ann Arbor, MI, February 1981.
15. Procedures Document for Development of National Pollutant Emissions
Trends Report. U.S. Environmental Protection Agency, Office of Air
Quality Planning and Standards, Research Triangle Park, NC,
December 1985.
16. C. T. Chi, et al., Source Assessment — Prescribed Burning,
State of the Art. EPA-6>00/2-79-019h (NTIS PB80-181472), U.S.
Environmental Protection Agency, Research Triangle Park, NC,
November 1979.
17. Technical Tables to the Nacional Air Pollutant Emissions Estimates.
1940-1984. EPA-450/4-85-014 (NTIS PB86-121100), U.S. Environmental
Protection Agency, Office of Air Quality Planning and Standards,
Research Triangle Park, NC, January 1986.
18. Report to Congress on the Discharge of Hazardous Waste to Publicly
Owned Treatment Works (the Domestic Sewage Study), EPA/530-SW-86-004
(NTIS PB86-184017), U.S. Environmental Protection Agency, Office of
Water Regulations and Standards, Washington, DC, February 1986.
19. Background Information Document for Chapter 1-6, Hazardous Waste
Treatment, Storage, and Disposal Facilities: Land Treatment, U.S.
Environmental Protection Agency, Office of Air Quality Planning and
Standards, Research Triangle Park, NC, February 1986.
20. J. B. Homolya, Primary Sulfate Emission Factors for the NAPAP
Emissions Inventory. EPA-600/7-85-037 (NTIS PB86-108263), U.S.
Environmental Protection Agency, Air and Energy Engineering Research
Laboratory, Research Triangle Park, North Carolina, September 1985.
21. D. Misenheimer et al., Hydrogen Chloride and Hydrogen Fluoride
Emission Factors for the NAPAP Emission Inventory. EPA-600/7-85-041
(NTIS PB86-134020), U.S. Environmental Protection Agency, Air and Energy
Engineering Research Laboracory, Research Triangle Park, North Carolina,
October 1985.
3-52
-------
D.C. Misenheimer, T.E. Warn, and S. Zelmanowitz, Ammonia Emission
Factors for the NAPAP Emission Inventory. EPA-600/7-87-001 (NTIS
PB87-152336), U.S. Environmental Protection Agency, Air and Energy
Engineering Research Laboratory, Research Triangle Park, North
Carolina, January 1987.
3-53
-------
SECTION U
QUALITY CONTROL FOR POINT AND AREA SOURCE DATA
This Section describes relevant EPA quality assurance/quality control
(QA/QC) policy, clarifies the use of QC for inventory data, and identifies
specific EPA objectives met in order to create a comprehensive and accurate
1985 emissions inventory.
4.1 BACKGROUND
4.1.1 EPA Policy
EPA'S current QA/QC program on emissions measurements was initiated in
May 1979. Detailed program requirements were issued in April 1984 under EPA
Order 5360.1, "Policy and Program Requirements to Implement the Mandatory
Quality Assurance Program." This order was developed to ensure that all
environmental measurements conducted by EPA's Regional Offices, program
offices, laboratories, contractors, and other sources resulted in data that
were both scientifically valid and defensible. Valid and defensible data would
include documentation of measurement precision, accuracy, representativeness,
and comparability, as well as sample custody.
The stringency of these requirements varies with the use of the
measurements being made. The most stringent requirements are reserved for
engineering and environmental measurements that might be incorporated into EPA
regulatory, enforcement, legal, or policy decisions.
Table 4-1 illustrates the elements of a typical QA/QC plan for engineering
research or development projects that might be used for EPA policy decisions.
Almost half of these elements (i.e., elements 3, 5, 7, 8, 10, 11) focus on
topics intrinsic to environmental measurements: measurement equipment,
calibration procedures, analytical procedures, data reduction, equipment
performance audits, and preventive maintenance checks.
However, these elements have no counterparts in nationwide emissions
inventory projects, because inventory data are almost exclusively based on
emissions estimates rather than measured data. In general, the emissions data
contained in emissions inventories are annual estimates that have been
4-1
-------
TABLE 4-1. QUALITY ASSURANCE/QUALITY CONTROL ELEMENTS FOR
ENGINEERING RESEARCH AND DEVELOPMENT PROJECT PLANS
1.
Project objectives and constraints
2.
Project and QA/QC organization
3.
Data quality objectives for measurements
4.
Sampling procedures
5.
Calibration procedures and frequency
6.
Sample custody
7.
Analytical procedures
8.
Data reduction, validation, and reporting
9.
Internal QA/QC checks
10.
Plans for performance and systems audits
11.
Preventive maintenance
12.
Calculation of data quality indicators
13.
Plans for corrective action
14.
Plans for QA/QC reports
4-2
-------
developed by multiplying emission factors by activity indicators such as fuel
consumption, by extrapolating from short-term test data, or by using
engineering judgment. Therefore, the fundamental concepts of data accuracy and
QA (assessing the differences between measured and true values) do not apply to
emissions inventories. The traditional QA/QC approach developed for
environmental measurements such as emissions tests is not applicable to the
1985 NAPAP Emissions Inventory, where QC procedures are limited to verifying
the reasonableness—as opposed to the accuracy—of the emissions inventory.
The development of QC procedures for the 1985 NAPAP Emissions Inventory is
complicated further by the size of the inventory and its component data bases.
For example, the anthropogenic point source data base maintained by EPA's
National Emissions Data System (NEDS) contains 50 data elements for each of
about 200,000 emission points—or about 66 million bytes of data. Ultimately,
the 1985 NAPAP Emissions Inventory, which includes these point source data,
will contain hourly emissions estimates for approximately 35 pollutant species
along with the 100 area source emissions categories for each of 63,000 grid
cells (each cell is a geographic area roughly 20 km square)—or about 4 billion
bytes of data. For data bases of this magnitude, QC procedures must be
maintained and applied on large computers.
The conceptual framework for QC procedures for the 1985 NAPAP Emissions
Inventory was adapted from the QC elements developed by EPA for environmental
measurements and from an EPA study identifying QA/QC approaches to emissions
1 2
inventory activities. The QC program for the 1985 NAPAP Emissions Inventory
represents a pioneering effort in QC procedures for emissions inventory
projects within EPA.
4.1.2 Emissions Inventory Quality Control
Table 4-2 presents eight recommended QA/QC elements for emissions
inventory projects. Note that the elements in this table are similar to the
eight elements which remain in Table 4-1, except that those elements germane
only to environmental measurements have been deleted. These eight elements
constitute the conceptual framework for the ajl hoc EPA emissions inventory QC
program that has been developed for the 1985 NAPAP Emissions Inventory. The QC
4-3
-------
TABLE 4-2. RECOMMENDED QUALITY ASSURANCE/QUALITY CONTROL
ELEMENTS FOR EMISSIONS INVENTORY PROJECT PLANS
1. Project objectives and constraints
2. Project and QA/QC organization
3. Data collection procedures
A. Data custody
5. Data validation and QA/QC checks
6. Internal data handling procedures
7. Calculation of data quality indicators
8. Plans for QA/QC reports and corrective action
4-4
-------
loop developed for the inventory concentrates on elements four through eight in
Table 4-2.
4.2 OBJECTIVES
One of the major objectives of the EPA QC effort for the 1985 NAPAP
Emissions Inventory was to provide a communications process permitting
systematic identification and resolution of problems identified concerning
reported or missing data. Other objectives are listed below.
4.2.1 Identification of Key Data Elements and Data Quality Objectives
A basic objective of the 1985 NAPAP Emissions Inventory was to compile a
comprehensive and accurate inventory of emissions and facility data for
anthropogenic sources for the 1985 base year, in order to serve several needs
articulated by EPA and NAPAP. Both NAPAP and EPA wanted to develop point and
area source data bases with emissions estimates that were confirmed by the
participating sources and States. Additionally, NAPAP planned to use the
inventory developed from these data bases in an atmospheric process model
which analyzes relationships between emissions and acidic deposition.
In order to ensure that the data quality objectives considered important
by NAPAP and EPA received the most attention, NAPAP developed guidelines for
high priority data elements and data quality objectives to help the States
effectively allocate their limited staff resources. High priority data
elements for the 1985 NAPAP Emissions Inventory were indicated in Table 3-1.
Table 4-3 summarizes NAPAP data objectives.
The QC procedures focus on the largest point source emissions categories,
e.g., electric utilities, petroleum refining, cement manufacturing, pulp and
paper mills, motor vehicle coating, and iron and steel mills. In addition, QC
procedures focus on several data elements that have been identified as
priorities for NAPAP research, i.e., emissions estimates, unit process
identifiers (source classification codes), control equipment and efficiencies,
fuel and operating rate data, location, throughput, and temporal profile data.
All five criteria pollutants—nitrogen oxides (N0X), sulfur dioxide (SO2),
4-5
-------
TABLE 4-3. NAPAl* 1985 EMISSIONS INVENTORY DATA QUALITY OBJECTIVES
Emissions Inventory
for Support of
Assessment Activities
Emissions Inventory
for Support of
Eulerian Atmospheric Modeling
Ceographic Domain
Temporal Resolution
Spatial Resolution
Species
Sources
48 U.S. States, District
of Columbia, and Canada
Annual/seasonal
Coordinates for point
sources; area sources at
the county level in the
U.S. and at the province
levels in Canada;
natural sources at county,
State, or province level;
release height
SO2, N0xt reactive VOC,
TSP, CO
Same
Hourly emissions values for
typical weekday, Saturday, and
Sunday for all four seasons
Coordinates for point sources;
area sources assigned to
20 x 20 km grid cells; release
height
SO2, SO4, TSP (Ca, Mg, K, Na),
CO, HC1, HF< NO, NO2, NH3, VOC
(methane, ethane, ethylene,
propane propylene, N-butane,
1.2-butane, isobutane,
isobutene, trans-2 butene
pentane, isopentane,
2.3-dimethyIbutane, other
alkenes, other alkanes, formic
acid, acetic acid, other organic
acids, formaldehyde,
acetaldehyde acetone, other
ketones, other aldehydes,
xylene, benzene, toluene,
ethyl benzene, oLher aromatics)
natural emissions (S, alkaline
dust, NOx, from lightning and
biota, isoprene, 4 major
turpenes, NH3)
Anthropogenic stationary Same
sources emitting >100 tons
of criteria pollutants in
1983; area source estimates
for small stationary and
mobile sources; natural sources
4—6
-------
volatile organic compounds (VOC), total, suspended particulates (TSP), and
carbon monoxide (CO)—are included in the NAPAP inventory. The point source QC
methodology focuses on N0X, SO2, and VOC because they are expected to play
major roles in the acid deposition models.
Another major objective of the point source data base QC checks was to
communicate results effectively to field inventory personnel who are most
knowledgeable about sources in their States. For this reason, the point source
QC process for 1985 is a three-step screening process. The first step, at the
field level, involves State and EPA Regional Office personnel. The second
step, which occurs after the data are submitted to OAQPS/AEERL, is a search for
systematic errors and omissions in the data. The third step, which takes place
after the data are compiled into a preliminary NEDS file, is the return of
detailed QC documents outlining any problems found with the data to the States
so that they may comment and correct the data. A flow chart of the QC
procedures is shown in Figure 4-1.
4.2.2 Identification of Problems in Existing State Emissions Inventories
Early in 1986, the most recent emissions inventory file for each State was
examined by OAQPS/AEERL for missing high priority data items, missing estimated
emissions, and any systematic errors involving invalid coding or processing
(see Section 3.1.5). These reviews indicated several consistent deficiencies
in the EPA data bases: emissions estimation methods were not recorded
correctly, allowable emissions were reported instead of actual emissions, and
blanks rather than zeros were left in the spaces for control equipment/
efficiencies to indicate no control equipment. These data problems did not
necessarily reflect problems specific to the State point source data. For
example, some of the data gaps resulted from incomplete translation of data
from the State systems to NEDS. Subsequently, the Regions and States were able
to identify many of the sources of these errors and institute corrective
measures within the data collection, coding, and transfer steps. This action
alerted EPA to potential problem areas for the 1985 NAPAP Emissions Inventory.
4-7
-------
4.2.3 Standard Inventory Techniques
Establishing a common methodology at the State level for data collection,
emissions estimation, and QC was an important early step in assembling an
adequate and consistent data base. First, the resulting State data bases would
be developed on a nationally consistent basis. Second, QC procedures could be
developed to build QC directly into the data collection process, where errors
could be located and corrected most efficiently. Essential elements of
procedural and technical guidance were communicated to State and contractor
personnel through two workshops and companion manuals (see Section 3.1.6).
4.2.4 Emissions Estimation Procedures
Emissions estimates from the States for the 1985 NAPAP Emissions Inventory
are calculated estimates of actual emissions during 1985. NEDS recognizes
two types of estimation procedures: one using individual source data and one
using emission factors. EPA requested that States, whenever applicable, use
the standard emissions estimation procedures, described in Section 3.7, and
presented a hierarchy for utilization of the acceptable methods. Each method
has a unique code in NEDS sc that the method is documented and can be tracked
over time for each emissions point.
4.2.5 Utility Quality Control Checks
Utilities represent the single largest point source category of N0X and
SO2 emissions. For this reason, a comprehensive methodology of QC checks was
deveLoped specifically for this source category. The focus of this methodology
was on complete coverage of the electric utility industry, proper classifica-
tion of all electric utilities, and agreement between specific cata reported co
EPA and data reported to DOE.
The first part of this methodology was a review of NEDS data to identify
plants that might be electric utilities. The first three digits of the SCC
codes for electric utility boilers are 101 or 102. In addition, the Standard
Industrial Classification (SIC) Code for electric utilities is 4911. The
4-8
-------
review identified both facilities which met these criteria and facilities
which met the criteria in part but appeared to have internally inconsistent
data. The utilities found in NEDS were compared to those provided by DOE. In
order to ensure that all potentially large utilities were covered, all
discrepancies that involved utilities expected to emit at least 100 TPY of NOx,
S02> or VOC were noted.
The second part of the electric utility review was a direct comparison of
data elements reported by the utility plants to EPA through NEDS and to DOE
through Energy Information Administration (EIA) Form 767. The NAPAP high
priority data elements were of particular concern in this comparison.
4.3 THE POINT SOURCE QC LOOP
To meet the objectives of the EPA QC policy for emissions inventories, a
systematic QC loop was developed involving State agencies, EPA Regional
0£fices, and EPA OAQPS/AEERL. Problem resolution, correction, and review
entailed all three levels in a cooperative process. At each major stage in the
QC process, an option existed to refer problems back, to State agencies, to
permit engineers closest to the sources to resolve problems. The QC loop is
outlined generally below, with the components of each phase subsequently
explained. Figure 4.1 provides a graphic presentation of the loop.
4.3.1 Overview of QC Loop
The QC loop begins with the States, which were responsible for the initial
compiling and reporting of 1985 point source data via NEDS to EPA. States
obtained emissions and other source data regarding criteria pollutant emissions
directly from plants. The State role in the QC process continued throughout
each step, as the option existed to refer problems or questions about the data
back to the State agencies generating the data for resolution.
EPA Regional Offices then subjected the State data submittals to the NE061
edit checking program, and enlisted State help in resolving identified problems
or errors in the data. OAQPS then updated the State emissions inventory data
to create a preliminary NEDS point source file. As the data were corrected, a
4-9
-------
SUMMARY OF
FINAL DATA
FINAL
DATA
PRELIMINARY
NEDS DATA
AUDIT
TRAIL
NE061
SOURCE DATA/
CONFIRMATION
STATE ANALYSIS
AND CODING OF
PRELIMINARY
DATA
NE061
ANALYSIS BY
REGIONAL
OFFICES
COMPUTER AND
MANUAL CHECKS;
STATE QC REPORTS
BY
OAQPS/AEERL
Figure 4.1. QC Loop for Point Source Emissions Data
4-10
-------
computerized Audit Trail system kept track of all changes to the original
State-submitted data and ensured that all identified problems had been
addressed.
After the data were screened by EPA Regional Offices, they were subjected
to a second set of screening procedures developed specifically for the 1985
NAPAP Emissions Inventory by OAQPS/AEERL. In this screening process,
four computerized programs and two manual checks were used to indicate
inconsistencies and errors in the data. The results of these programs were
compiled into a six-part QC document for each State, which was sent to the
State and EPA Regional Offices for corrective action or confirmation of
questionable data.
After final corrections and updates to the State's emissions inventory had
been made, a finaL review of the compLete 1985 point and area source data base
constituted a last step in the QC process. A check of the largest sources was
made in an attempt to guarantee that all errors were detected and correcced.
Inconsistencies in confirmation results and the use of NEDS-calculated
emissions were also reviewed. In February 1988, each State received its finaL
NEDS point and area source data for a final review. At this time questionable
data identified in the final QC step were verified or corrected through the
same channels established for the NEDS inventory. AIL changes were documented
by the Audit Trail program.
4.3.2 State Level - Data Collection and Confirmation
Quality controL (QC) of the State emissions inventory submittals proceeded
in several steps. The process began with a State's submittal of 1985 emissions
inventory data. For many States, this effort required contacting each major
source, obtaining detaiLed emissions and facility data for each point within
that source, obtaining confirmation of plant emissions totals, encoding the
data onto NEDS forms, and checking the forms for errors. The State could run a
computerized edit-checking program on the data and make appropriate
corrections. Additionally, States were asked to help NAPAP resolve any
questions that were identified at Later stages of the QC process for sources
that emitted at least 1000 tons of SO2, NOx, or reactive VOC.
4-11
-------
The States then submitted their 1985 NEDS data to EPA Regional Offices.
The Regional Offices broke the data down into standard NEDS subfile and card
image files. If the files passed the general validity scan performed on them,
they were copied to computer disk for further processing. At this point, the
edit program NE061 was run on the card image file, and an error report was
generated. (See Appendix B for a list of rejection and warning messages used
by NE061.) The Regional Offices enlisted State assistance in resolving
identified problems or errors. The Regional Offices then transmitted the
emissions inventory data and confirmation status reports to OAQPS, along with a
State transmittal letter describing the coverage of the inventory.
4.3.3 EPA Screening Level
OAQPS then updated the State data to create a preliminary NEDS point
source file. Once the State data were in a preliminary point source file, a
tape of the file was transferred to AEERL. AEERL entered the file into its
data base and subjected it to a variety of QC checks, which are described in
detail below. These initial checks included the NE061 edit checks to verify
that problems identified by the Regions had been resolved, completeness checks
for missing facilities, comparisons with State utility data submitted to DOE,
and additional QC checks to identify erroneous data.
4.3.3.1 NE061 Edit Checks
The NE061 edit program creates two reports with each execution. The first
report is a reject edit diagnostics report. It identifies data entries that
were rejected for invalid State, Air Quality Control Region (AQCR), plant or
point identifiers (IDs), Source Classification Code (SCC), action code,
transaction type, or card number. These errors must be corrected by the States
and the data must be resubmitted before the entries can be considered valid.
The second report is a warning edit diagnostics report. Cards identified
in this report contain likely errors (certain data items missing, data items
exceeding limits established in external data files, or internally inconsistent
data), but are not necessarily invalid. The warning report was reviewed to
4-12
-------
identify any real errors chat needed Co be corrected or any important missing
data items that needed to be obtained. Warning messages did not always
indicate an error; however, EPA Regional Offices and States were asked to
verify that inventory data flagged with warning messages for facilities
emitting more than 1000 tons per year of SOx, NOx, and VOC were correct.
If no rejection errors or warning messages apply for a particular card
image record, no message is printed. A summary is generated which includes a
count of the total number of card images read, the number that passed the
minimum data quality requirements, and the number that were flagged with
reject messages.
4.3.3.2 Completeness Checks
After the NE061 edit checking program was run, a completeness checking
procedure was performed to identify possible omissions of large sources in the
1985 State NEDS submittals. Data from the 1985 submittals were compared with
data from the 1984 NEDS Large Source File and with plants listed in the
Independent Completeness File. The 1984 NEDS Large Source File contains the
NEDS codes, names, addresses, and emissions for plants in the current 1984 NEDS
point source file emitting more than 1000 tons/year of SO2, NOx, or VOC. The
Independent Completeness File was compiled from several reference sources
including trade association publications, current industrial directories, and
EPA New Source Performance Standards files.
The results of the completeness checks for each State submittal were
summarized in three tables. The first table listed plants in the Independent
Completeness File not found in the 1985 State submittal. The second table
contained a listing of plants in the 1984 Large Source File not found in the
1985 submittal. The third table listed the 1984 NEDS-computed emissions which
differed greatly from the 1985 State-submitted emissions for SO2, NOx, and VOC.
Other computerized and manual data checks were also run on the State NEDS
submittals. These included checks for missing data and checks against standard
data ranges to ensure the completeness and accuracy of the final data base, as
described below.
4-13
-------
Checks for missing data—A pre-update screening of the data was performed
co check chat (1) major sources emicting over 100 tons/year were not improperly
deleted, and (2) plant names matched with current plant IDs to ensure that
plant IDs were not changed.
The data were updated to Che NEDS file and the update reporcs returned to
the Regional Office. The update process stored the new data in NEDS and
calculated emissions or apportioned reported emissions to the SCC level. The
update created the Calculation Validation Report, the Update Rejection Report,
and the Update Validation Report. The Calculation Validation Report identified
inconsistencies in the parameters (estimation methods, emission factors,
operating rates, sulfur and ash content, estimated point emissions, and SCC
emissions) that were rejected because the data were incomplete (missing cards)
or because the update action code (add, change, delete) was incorrect. The
Update Validation Report identified some of the inconsistencies between related
data items.
OAQPS/AEERL procedures include a post-edit inventory of data received.
This inventory was created to determine the amount of data submitted and
whether the coverage agreed with the letter from the State. Major problems
such as incomplete card sets were identified by this procedure.
Checks against standard data ranges—The State NEDS inventory was edited
by the NEDS edit program to identify any items outside the bounds of
"reasonableness" criteria that were not resolved by the State or Regional
Office. Problems for sources above the NAPAP thresholds were referred to the
Regional Office and the State for resolution.
The UTM coordinates for each point source were checked against correct
ranges for the county. For the sources above the NAPAP threshold, the apparent
errors were referred for resolution. For all other problems, NAPAP substituted
the county centroid for the coordinates.
NEDS stack, control device, control efficiency, emissions, and activity
data for a given source underwent statistical checking procedures. Reported
data were checked against mean values for that SIC/SCC combination. Data that
fell more than two standard deviations outside the mean for chat SIC/SCC were
flagged for invescigacion.
4-14
-------
4.3.3.3 DOE EIA-767 Utility Fuel Data Comparison
The electric utility sector is a major source of SO2 and N0X emissions in
the U.S., contributing about two-thirds of total SO2 and one-third of total NOx
emissions. Because this sector represents such a large portion of the total
emissions for these pollutants, it was imperative that the highest possible
quality data be reported to NEDS for the 1985 NAPAP Emissions Inventory. In
addition to running NE061 and other computerized and manual QA checks on State
NEDS submittals, EPA also used utility data submittals to the Department of
Energy (DOE) for QC purposes. Since the utilities are required by Federal law
to report certain information to DOE's Energy Information Administration (EIA),
Chose items should serve as an appropriate basis for comparison with utility
data submitted to NEDS.
Of particular interest for the 1985 NEDS/NAPAP Inventory were the data
reported on Form EIA-767. This report includes data that are very similar or
identical to data required in NEDS. Every steam electric generating plant
with a total generator nameplate rating of 100 megawatts or greater is
required to submit a Form EIA-767. In 1985, this group included about
%00 fossil fuel-fired steam electric plants.
The EIA-767 data were compared with data received from NEDS as a QC check
for utilities above the NAPAP emissions thresholds. For State or local
agencies, EIA-767 could serve either to supplement the data normally collected
from electric utilities or to provide a quality assurance crosscheck. Ideally,
State agencies worked in cooperation with utilities in their domain to ensure
that the data reported to DOE by the utility and the data reported to EPA by
the State were correct and consistent with each other. At this stage in the QC
process, resolution of discrepancies found by OAQPS/AEERL was likely to be less
awkward and time-consuming than it would be later in the QC process.
Significant discrepancies in location, stack parameters, activity levels, and
emissions data were referred to the States for resolution. EPA felt that the
time and effort spent by the States to work with utility companies concerning
their report to EIA were more efficiently used than time spent later trying co
resolve discrepancies between NEDS and EIA data.
4-15
-------
4.3.3.A Additional QC Checks
OAQPS applied a series of additional QC checks to the State NEDS
submittals to indicate systematic errors. These checks included both
computerized and manual checks, and focused on the NAPAP high-priority data
items. Five checks were used: the State Emissions Summary Report, the State
Fuel Summary Report, the County/Plant Emissions Report, Quick Look Reports, and
the Calculation Validation Report.
The State Emissions Summary Report was used for large-scale comparisons of
emissions data. It summarizes emissions data by combustion source type and
fuel type, and presents total statewide emissions for N0X, SO2, and VOC. These
totals were compared to the previous year's totals and to previous emissions
trends to ensure that 1985 data were in line with past emissions totals.
The State Fuel Summary Report was used for large-scale comparison of fuel
consumption data. It was compared to the 1984 State Fuel Summary Report for
each combustion category and for each type of fuel. Data on fuel consumed by
electric utilities were also compared to the DOE Generating Unit Reference File
(GURF). State -consumptions of coal, oil, and gas were compared to fuel
delivery data provided by DOE.^'^'^ Generally, coal consumption by point
sources should have equaled coal delivered. Oil and gas consumption by point
sources should have been less than the fuel deliveries because a large portion
of these fuels is consumed by area and mobile sources. During this analysis,
the most frequent problem encountered was the over-reporting of fuel
consumption in the NEDS submittals.
The County/Plant Emissions Report gives emissions by individual plants
sorted by county. Other data fields, including year of record and location
data, are also listed. This report was quickly scanned to ensure that the new
data had a 1985 year of record.
NEDS can calculate emissions by multiplying the annual operating rate by
an emission factor from the emission factor file. Because these emissions
estimates are computer-calculated, they are not as valid as estimated emissions
data entered by the State agency. The Quick Look Reports were used to identify
two possible problems. First, a check was made for instances where NEDS could
have calculated positive emissions for a source that reported zero emissions.
4-16
-------
Small sources were ignored, but those with potential emissions exceeding 25 TPY
of N0X, S02, or VOC were noted. Second, instances in which the State requested
that NEDS calculate emissions, but no emission factor existed in the emission
factor file, were noted.
The Calculation Validation Report identified any instance where NEDS could
not allocate emissions from the point level to the SCC level. This problem
generally occurred when one of the SCCs assigned to the point had no emission
factor or its operating data were missing. This scan was performed for SCCs
expected to emit 25 TPY or greater of criteria pollutants. If a decrease in
emissions was found during the scan of Che State Emissions Summary Report, then
additional information on the possible miscalculation of emissions was sought
in the Calculation Validation Report, which prints out all instances where
estimated point emissions differ from computed emissions by more than a factor
of three.
4,3.3.5 Audit Trail
The 1985 NAPAP Audit Trail Program was designed to document all changes
made to the NAPAP Emissions Inventory point source data base. The SCC-level
NEDS data were used as input to this program. SCC-level data are defined as
sources having unique State, County, plant, point, and SCC code combinations.
All information resulting from the execution of this program was stored in an
"electronic notebook." disk file.
Each record in the new version of the data base was compared to the
corresponding record in the previous version of the data base. All parameters,
such as stack height, flow rate, emissions estimates, etc., were checked for
matching new and old values. If a match was not found for a particular source,
the record was flagged and the old and new values of the parameter were
printed. If a record was present in the new data base but not the old data
base, the record was flagged as a new (added) record and was printed.
Similarly, if a record appeared in the old data base but not in the new data
base, the record was flagged as an old (deleted) record and was printed.
4-17
-------
4.3.A QC Reports
After NE061 edit checks, completeness checks, and other QC checks were
performed on the State submittal, the resuLts of these efforts were assembled
into a State QC report. This report included the following information:
(1) the results of the NE061 edit checks for plants reported to
emit at least 1000 TPY SO2, N0X, or VOC;
(2) emissions sorting, parameter validations, and large source
completeness reports;
(3) a report showing what portion of total plant emissions were
calculated, rather than reported, for plants where EPA was
requested to calculate emissions from any point by using
the NEDS emission factor file; and
(4) annotation of these reports to indicate, where possible,
which warning messages should be investigated and resoLved
by States.
Once a State QC report had been assembled, OAQPS/AEERL reviewed the
results of the QC checks made on each State NEDS submittal to identify specific
problems in the results, such as misleading or incorrect error/warning
messages, and to identify significant data quality problems revealed by the
checks. After analyzing the results of this review, OAQPS made recommendations
for resolution of any problems found. Appropriate modifications were then
carried out to resoLve the detected problems-
The QC results for each State submittal were then assembled into a formal
report ("State Report") complete with explanatory notes for each section and a
cover letter from OAQPS. An attempt was made to minimize the time required to
generate the State Reports to fewer than 15 working days from receipt of the
State submittals.
The State QC Reports were then submitted to OAQPS and sent on to the State
agencies responsible for the data. The State agencies pursued QC questions and
problems, corrected errors, and provided missing data items for as many sources
as possible. In order to facilitate response to the State Reports, States were
requested to respond to the results of the NE061 edit checks, the completeness
checks, and miscellaneous QC questions on a priority basis. Based on State
4-18
-------
responses, EPA completed final updates to the State emissions inventory data
submissions.
After these data updates to the State NEDS submittals were completed, a
final review by each State of its final NEDS point and area source data was
initiated. As part of this final review, three additional data screens were
made. First, a final check of the largest, sources was made to guarantee that
all errors were located and corrected. In this check, the 1000 largest points
for S02» N0X, and VOG were identified from the NEDS point source tile. (These
points represent 81, 68 and 47 percent of the total point source SO2, NOx, and
VOC emissions, respectively). The data for each point source were examined by
a qualified reviewer for consistency and realistic values. In addition, other
points emitting at least 500 tons per year were screened for missing stack
parameters. Second, inconsistencies between State confirmation reports and
NEDS emissions were identified. Third, NEDS-calculated emissions were again
summarized at the point level for calculated emissions of at least 100 TPY for
SO2, N0X, and VOC.
The complete point and area source NEDS data and the results of these
three additional data screens were reviewed by the States. Questionable data
were verified or corrected where possible through the same channels
established for the previous State reviews. These changes were then updated to
the 1985 NAPAP Emissions Inventory. All changes were documented through the
Audit Trail Program.
A. A RESULTS OF QC PROCEDURES
A.A.I Analysis of Quality Control Results
Through the use of an extensive QC program on the 1985 NEDS State
submittals, the resulting NAPAP interim point source data base represents the
most complete and accurate inventory of acid rain precursors assembled to date.
Essentially all data have a 1935 year of record; by contrast, in a typical NEDS
year only 10 States are able to submit substantially current data. All States
have participated in the 1985 inventory effort and submitted current data;
whereas in general, previous NEDS data represent annual submissions by only
4-19
-------
34 States. Previous inventories contain significant data omissions, even among
priority data items, but in 1985, States delivered substantially all high
priority data items.
Through cooperative work, between States, Regional Offices, and several
branches of EPA, and through regular updates of all data bases, State
inventories and the 1985 NAPAP Emissions Inventory should have substantial
agreement. In the case of records rejected by the NE061 program, complete
replacement submittals of the data alleviated potential continued problems with
erroneous data elements. EPA emphasized that States should submit emissions
estimates for large sources wherever possible rather than allow NEDS to
calculate emissions. EPA requested confirmation of emissions with facilities
emitting over 2500 tons of SO2 and NOx to assure the accuracy of the emissions-
data for the largest point sources; about two-thirds of the States responded
with confirmation letters. Computer checks served to ensure that data were
within standard data ranges, and when data did fall out of range, consultation
with State or local air pollution agencies could often confirm or resolve data
issues.
A.A.2 Resolution of QC Problems
However, the QC program was not without its problems. Time and resource
constraints, present at all levels of QC, made it difficult to track, down
missing data items and data elements which lay outside standard ranges. State
data which had been confirmed with sources sometimes fell outside the data
ranges, and therefore still evoked a warning message each time NE061 was run
on the EPA data base. A more substantial problem occurred when emissions data
were confirmed between a facility and the State, yet these confirmed emissions
were still not in agreement with the data in NEDS. Lack of confirmation was a
problem in over one-third of the States. It was often difficult to resolve
discrepancies in source names and emissions data between 1984 and 1985 NEDS
submittals. For example, Illinois and Michigan renumber their NEDS data
submittals each year, making matchups between data from consecutive years
especially difficult. Problems such as these were referred to the
4-20
-------
participating State agency for resolution, and in almost every case, the Stale
was able to verify or correct these questionable items.
4.4.3 Remaining QC Questions
Several more serious problems remained in the emission inventory.
Individual State confidentiality restrictions prevented the complete reporting
of some priority data elements (e.g., fuel use, operating rate data). In
consideration of this problem, CPA and the States worked together to reach
compromises respecting both these confidentiality restrictions and the need for
accurate and complete data. Specific cases are outlined below.
The State of Texas' Clean Air Act prohibits the public disclosure of
facility operating rate data. Texas agreed to compare summaries of data
submitted by Texas utilities to the Department of Energy with confidential
utility information and to indicate where DOE data were in error, since these
actions would not violate Texas Clean Air Act restrictions. EPA then estimated
fuel consumption for Texas industrial boilers missing this data item as a check,
on the reasonableness of the emissions data provided by Texas. Texas
consequently reviewed EPA's fuel consumption estimates and identified boilers
with inaccurate emissions.
West Virginia had problems due to confidentiality restrictions and a
shortfall of resources. The State agency had obtained 1984 operating rates and
maximum design rates for chemical plants under an agreement which stipulated
that the data remain confidential. In addition, State officials indicated that
because of the effort expended in compiling the 1984 VOC emissions data base,
the State would not have the resources to collect VOC emissions data for these
sources for 1985. West Virginia and EPA agreed that the 1984 VOC data would
fulfill the 1985 inventory objectives.
New Jersey and New York also had specific problems. Initially, State
officials doubted that they could provide operating rate and actual emissions
data, because the State agencies do not collect these data from plants. In
both cases, EPA agreed to allow the State to report allowable emissions for
plants emitting less than 1000 TPY of criteria pollutants but requested actual
emissions data for plants emitting greater than 1000 TPY. New York was able to
4-21
-------
submit actual emissions data as reported by the plant tor plants emitting
greater than 100 TPY. For most points in the New Jersey inventory, operating
rates were not reported. For New Jersey sources reporting over 1000 TPY of
SO2, NOx, or VOC, operating rates were reported, but were not collected or
reviewed according to NAPAP specifications. Consequently, these data do not
meet the inventory criterion for quality.
Ohio and Virginia initially indicated confidentiality problems in
reporting operating rate data. Ohio eventually was able to provide these data.
Virginia did not submit operating rate data for noncombustion sources.
A potentially serious deficiency was recognized during the QC procedures
designed to recognize missing facilities. Department of Defense (DOD)
facilities were reported inconsistently in the 1985 NEDS submittals, and only a
low percentage of those listed were identified as large point sources. Further
investigation with State personnel revealed that most missing DOD facilities
had not submitted data to the State. At that stage of the inventory effort, no
resolution of this problem could be made for 1985. However, DOD facilities
have been identified on a State-by-State basis to indicate those contained in
the inventory, as well as those potentially missing from the inventory.
Another major problem with the entire NEDS data collection and QC effort
was the incompatibility of various State computer systems with the NEDS
system. In addition to delays caused by the extra work, involved in
translating data, this inconsistency caused some incomplete or garbled
translation of data. Consequently, a more extensive data correction effort was
required than would have been necessary if all the computer systems had been
compatible. In cases where States had their emissions inventories stored in
other forms (e.g., on micro-files), data translation and QC was an even
lengthier process. In addition, the fact that State emissions tracking systems
are typically designed for permit enforcement work meant that some data items
were not in the form prescribed for NEDS and required some reworking. Although
these problems were overcome, future inventory efforts will be subject to
similar problems unless long-term solutions for data incompatibilities are
developed and implemented.
4-22
-------
4.5 QA/QC OF AREA SOURCE DATA
Quality assurance and quality control of area source emissions inventory
data, Like QA/QC of point source data, do not involve checks for data accuracy.
Eec&use there is no measured standard with which to compare area source
emissions data nor previously defined QA/QC procedures, QC procedures were
developed which focused on completeness and reasonableness of the. county and
gridded (SO2, NQX and VOC) emissions data, in addition to various input data.
Due to the quantity of data involved, activity Level and emissions data
from significant area source categories were plotted to locate missing data,
gross errors, border discontinuities, and improper Location of urban centers.
Plots (maps) of the final area source emissions were then compared with
appropriate plots of selected surrogate activity levels (e.g., population
distribution). Direct visual comparisons o£ the emissions plots and plots of
the surrogate data used to develop the estimates were used to isolate
potential problems by illustrating dissimilar patterns in the data.
For the purposes of this study, since Regional Acid Deposition Modeling
activities do not include any terrain west of the Rocky Mountains, areas in
the U.S. located east of the Rocky Mountains received more attention, most
specifically the Ohio River Valley. In addition, certain regions which have
exhibited erroneous spatial distributions in previous inventories received more
scrutiny in the QA process. These areas include St. Louis, northern West
Virginia, Pittsburgh, Massachusetts, and Virginia.
In a separate analysis, emission factors for each area source category
were reviewed for reasonableness. Sources of the original factors and their
methods of calculation were sought to verify the factors which were used as
NEDS input data.
Potential problems discovered during the. QA analyses were recorded for
later verification. Errors in the data were corrected when found, but in
general, area source emissions estimates, activity level data, and emission
factor data were found to be reasonable and complete.
4—23
-------
A.5.1 Emission Factors
Emission factors for each area source category were traced back, to the
reference materials. Most emission factors for area sources were adapted from
EPA's Compilation of Air PoLLutant Emission Factors (AP-42). In some source
categories, however, data were available to develop more applicable local
factors. In addition, ten new categories for VOC emissions were added to the
1985 NAPAP Emissions Inventory that were not included in the 1980 inventory.
The National Air Data Branch (NADB) and Alliance developed emission factors for
these additional categories and for ammonia emissions categories. The emission
factors associated with them may require further checks if discrepancies appear
in area source emissions for these new categories.
4.5.2 Activity Levels
Activity levels were derived primarily from related information published
by other Federal agencies, supplemented by special data developed by EPA for
the purpose of developing NEDS area source inventories. Published data such as
fuel use by State and county and forest fire acres burned by State are used
with related data such as employment, population, and miscellaneous geographic
or economic data to derive annual county estimates of the activity levels for
each of the NEDS area source categories.
Surrogate indicators are defined as those variables used to spatially
allocate area source activity levels to the county level in NEDS. Area source
emissions are similarly allocated to the county level by county population,
number of dwelling units, vehicle registration, and employment for various
economic sectors. Surrogates examined in the area source QA efforts were
selected based on the magnitude of national emissions contributed by respective
area source categories. Surrogate information plotted at the county level
included the following:
4-24
-------
Populac i on
Dwelling Units
Vehicle Registration
Manufacturing Employment
Commercial Employment
Solvent User Category Employment
For the most part, allocation variables were easily identified. However,
default surrogates were assigned to those categories where either insufficient
surrogate data were available or the NEDS allocation process used an algorithm
based on more than one significant variable.
Initial comparisons of population, land use, dwelling units, employment,
and vehicle registration plots indicated the general reliability of the data,
as well as identifying potential errors early in the QC process. Correction of
basic information before it was used to allocate State emissions to the county
or grid should reduce errors in the final emissions estimates. Missing and
incorrect data were flagged and corrected where possible.
4.5.3 Emissions
County emissions for the area source categories were estimated by NEDS
using category-specific activity level data and the appropriate emission
factors as shown in Section 3.2. Specific area source categories were
selected for evaluation in the activity level QA efforts based on the magnitude
of emissions contributed by particular area source categories. For each
pollutant (SO2, N0X and VOC), the categories were ranked by emissions magnitude
and the top categories were selected for detailed examination. These
categories included:
Stationary Sources -
Residential
Commercial/Institutional
Industrial
Mobile Sources -
Light Duty Gasoline Vehicles
Light Duty Gasoline Trucks
Heavy Duty Gasoline Vehicles
Off Highway Vehicles
Locomotives
4-25
-------
Ai rcraf t
Marine Vessels
Miscellaneous Sources -
Solvents Purchased
Gasoline Marketed
Additional Sources -
Hazardous Waste TSDFs
Fugitives from Refinery Operations
Bulk Plants/Bulk Terminals
Emissions plots were then generated using county level data from NEDS. Each
plot was reviewed for completeness before comparison with its surrogate
activity level plot. This review included checks for inconsistencies, missing
data, proper location of urban centers, and border problems. Emissions plots
were then compared to the surrogate plots to ascertain inconsistencies in the
data.
Final comparisons of the plots showed general agreement between the NEDS
input data and the final emissions estimates for many of the significant
categories. No major problems were found; however, potential minor concerns
were noted. A small number of missing data items were corrected.
4-26
-------
Section 4 References
1. D.A. Pahl and J.D. Mobley, Quality Assurance and Quality Control Plan
for the NAPAP 1985 Emission Inventory. EPA-600/8-86-025 (NTIS
PB86-23762), U.S. EPA, Air and Energy Engineering Research
Laboratory, Research Triangle Park, NC, August 1986.
2. Development of an Emission Inventory Quality Assurance Program. EPA-
450/4-79-006 (NTIS PB80-112345), U.S. EPA, Office of Air Quality
Planning and Standards, Research Triangle Park, NC, December 1978.
3. Inventory of Power Plants in the United States. DOE (EIA-0095(85),
U.S. Department of Energy, Energy Information Administration,
National Energy Information Center, Washington, DC, April 1986.
4. Coal Distribution January - December 1985. DOE/EIA-0125 (85/4Q),
U.S. Department of Energy, Energy Information Administration, Office
of Oil and Cas, Washington, DC, July 1986.
5. Petroleum Marketing Monthly. DOE/EIA-0380(86-07), U.S. Department of
Energy, Energy Information Administration, Office of Oil and Gas,
Washington, DC, July 1986.
6. Natural Gas Annual 1985. DOE/EIA-0131 (85), U.S. Department of
Energy Information Administration, Office of Oil and Gas,
Washington, DC, November 1986.
7. Compilation of Air Pollutant Emission Factors. AP-42 Vol. I, Supp. A
(NTIS PB87-150959), U.S. EPA, Office of Air and Radiation, Office of
Air Quality Planning and Standards, Research Triangle Park, NC,
October 1986.
4-27
-------
Section 4 References
1. D.A. Pahl and J.D. Mobley, Quality Assurance and Quality Control Plan
for the NAPAP 1985 Emission Inventory, EPA-600/8-86-025 (NTIS
PB86-23762), U.S. EPA, Air and Energy Engineering Research
Laboratory, Research Triangle Park, NC, August 1986.
2. Development of an Emission Inventory Quality Assurance Program, EPA-
450/4-79-006 (NTIS PB80-U2345)t U.S. EPA, Office of Air Quality
Planning and Standards, Research Triangle Park, NC, December 1978.
3. Inventory of Power Plants in the United States. DOE (£IA-0095(85),
U.S. Department of Energy, Energy Information Administration,
National Energy Information Center, Washington, DC, April 1986.
4. Coal Distribution January - December 1985, DOE/EIA-0125 (85/4Q),
U.S. Department of Energy, Energy Information Administration, Office
of Oil and Gas, Washington, DC, July 19G6.
5„ Petroleum Marketing Monthly. DOE/£IA-038G(86-07), U.S. Department of
Energy, Energy Information Administration, Office of Oil and Gas,
Washington, DC, July 1936.
6. Matural Gas Annual 1985. DOE/EIA-0131 (85), U.S. Department of
Energy Information Administration, Office of Oil and Gas,
Washington, DC, November 1986.
7. Compilation of Air Pollutant Emission Factors. AP-42 Vol. I, Supp. A
(NTIS PB87-150959), U.S. EPA, Office of Air and Radiation, Office of
Air Quality Planning and Standards, Research Triangle Park, NC,
October 1986.
4-27
------- |