EPA 240-R-01-003
STATISTICAL METHODOLOGY FOR ASSIGNING
EMISSIONS TO INDUSTRIES IN THE UNITED STATES,
REVISED ESTIMATES: 1970 TO 1997
Benefits Assessment and Methods Development Division
National Center for Environmental Economics
Office of Policy, Economics and Innovation
U.S. Environmental Protection Agency
-------
-------
SUMMARY
This report develops a methodology to assign emissions of criteria air pollutants to the
twenty two-digit standard industrial classification (SIC) code industries of the manufacturing
sector for the years from 1970 to 1997. The methodology employed to assign process and fuel
combustion emissions to specific industries is described and the emission estimates are listed in
an appendix and the accompanying spreadsheet (SIC7097.WK3).
This report develops procedures to estimate the emissions of categories of processes and
fuel combustion emissions that are not available ("NA") in National Air Pollutant Emissions
Trends Update: 1970-1997. As a result, there are cases in which the estimates of emissions used
in this report are higher than the official estimates in National Air Pollutant Emissions Trends
Update: 1970-1997. The Office of Air and Radiation of the U.S. Environmental Protection
Agency does not endorse the emission estimates for the "NA" categories. Hence, the emission
estimates in this report do not supersede the official estimates in National Air Pollutant
Emissions Trends Update: 1970-1997.
This is a technical report whose intended audience consists of individuals interested in
developing environmental accounts.
January 2001
-------
-------
Ill
TABLE OF CONTENTS
SUMMARY i
TABLES v
ACRONYMS AND ABBREVIATIONS vii
1. INTRODUCTION 1
1.1. Purpose of the Project 1
1.2. Organization of this Report 3
2. EMISSIONS FROM INDUSTRIAL PROCESSES 3
2.1. Assigning SIC andNEA Codes to Categories of Process Emissions 4
2.2. Assigning SIC and NEA Codes to Other Categories of Emissions 15
2.3. Assigning Values to Categories of Process Emissions listed as "NA " 17
3. EMISSIONS FROM FUEL COMBUSTION 28
3.1. Fuel Consumption of the Manufacturing Portion of the Industrial Sector 29
3.2. Fuel Consumption of the NonmanufacturingPortion of the Industrial Sector . . 31
3.3. Energy Produced and Consumed at the Same Establishment 38
3.4. Physical units vs. Btu 42
3.5. Concordance Between Fuel Consumption Categories and Emission Categories 43
3.6. Excluded Fuel Combustion Source Categories 46
3.7 Fuel Consumption and Fuel Combustion Source Categories listed as "NA' ... 48
4. OVERVIEW OF THIS REPORT 48
REFERENCES 55
APPENDIX A A-l
APPENDIX B B-l
APPENDIX C C-l
APPENDIX D D-l
APPENDIX E E-l
APPENDIX F F-l
APPENDIX G G-l
January 2001
-------
-------
Assigning Emissions to Industries in the United States, Revised Estimates
Table 1:
Table 2:
TableS:
Table 4:
TableS:
Table 6:
TABLES
Nonmanufacturing Industries Assigned Process Emissions,
1970-1990 Emissions
Manufacturing Industries Assigned Process Emissions,
1970-1990 Emissions
Concordance Between Industrial Process Source Categories
and NEA/SIC Codes
7
Procedures Document Describes the Procedure to Estimate
Emissions of Selected Tier n Categories of Process Emissions 18
Pollutants Associated with Fuel Combustion, by Fuel Type 44
Emission Estimates in the Spreadsheet and the Percentage of
EPA National Estimates of Emissions in the Spreadsheet 50
January 2001
-------
-------
Vll
BEA
BLS
Btu
CO
cu. ft.
DAEMEC
efr
EIA
ex.
ifr
MECS
NA
NAICS
NAMEA
NAPAP
NEA
ACRONYMS AND ABBREVIATIONS
Bureau of Economic Analysis, U.S. Department of Commerce
Bureau of Labor Statistics, U.S. Department of Labor
British thermal unit
carbon monoxide
cubic feet
Derived Annual Estimates of Manufacturing Energy Consumption
external floating roof
Energy Information Administration, U.S. Department of Energy
excluding
internal floating roof
Manufacturing Energy Consumption Survey
not available
North American Industrial Classification System
National Accounting Matrix including Environmental Accounts
National Acid Precipitation Assessment Program
National Energy Accounts
1970-1990 Emissions Statistical Methodology for Assigning Emissions to Industries in the
United States: 1970 to 1990
1940-1990 NAPEE National Air Pollutant Emission Estimates, 1940-1990
1970-1997 NAPET National Air Pollutant Emission Trends Update: 1970-1997
NOX oxides of nitrogen
January 2001
-------
Assigning Emissions to Industries in the United States, Revised Estimates
ACRONYMS AND ABBREVIATIONS
(continued)
Vlll
Pb
PM
PM-2.5
PM-10
pt.
Rec. furnace
sec
SEDR
SIC
SEEA
SO2
SOX
U.S. DOE
VOC
lead
particulate matter
particulate matter less than 2.5 microns in diameter
particulate matter less than 10 microns in diameter
part
recovery furnace
source classification code
State Energy Data Report
Standard Industrial Classification (code)
System for Integrated Environmental and Economic Accounts
sulfur dioxide
oxides of sulfur
U.S. Department of Energy
volatile organic compound(s)
January 2001
-------
1. INTRODUCTION
This report updates the estimates of emissions developed in Statistical Methodology for
Assigning Emissions to Industries in the United States: 1970 to 1990 (U.S. EPA, 2000).
Developing physical data on environmental-economic interactions represents one facet of
environmental accounting. The introductory section of Statistical Methodology for Assigning
Emissions to Industries in the United States: 1970 to 1990 (U.S. EPA, 2000) reviews recent work
on environmental accounting.
Several applications of emissions from industries have been proposed.1 Among the
proposed applications are indicators of environmental performance that measure changes in
emissions to changes in output over time. Emission data can be combined with input-output
tables and used to estimate the pollution-intensity of final demand categories. Finally, adjusted
measures of productivity, which credit an industry for reducing emissions, can be estimated by
linking emission data with data used to estimate traditional measures of productivity.
After National Air Pollutant Emission Estimates, 1940-1990 (U.S. EPA, 1991) was
released, the U.S. EPA changed its methodology for estimating emissions. For its estimates of
emissions after 1984, the "top-down" methodology, which was the basis of the EPA's estimates
in National Air Pollutant Emission Estimates, 1940-1990 (U.S. EPA, 1991), was replaced by a
modified "bottom-up" methodology. In addition, the U.S. EPA revised its estimates of
emissions for 1970 to 1984. National Air Pollutant Emissions Trends Procedures Document,
1900-1996 Projections 1999-2010 (U.S. EPA, 1998a) provides a detailed discussion of both
methodologies.2
1.1. Purpose of the Project
This report describes the procedures used to assign emissions of air pollutants to the
twenty two-digit standard industrial classification (SIC) code industries of the manufacturing
sector for 1970 through 1997. The emission estimates in this report are drawn from National
Air Pollutant Emissions Trends Update: 1970-1997 (U.S. EPA, 1998b).3 The methodology
developed in 1970-1990 Emissions forms the basis of the procedures employed when assigning
1 Nestor and Pasurka (1998, pp. 140-141) discuss additional applications of the data
developed in this report.
2 National Air Pollutant Emissions Trends, Procedures Document 1900-1993 (U.S. EPA,
1994) also discusses the two methodologies.
3 Throughout this report, 1970-1990 Emissions refers to Statistical Methodology for
Assigning Emissions to Industries in the United States: 1970 to 1990 (U.S. EPA, 2000). 1940-
1990NAPEE refers to National Air Pollutant Emission Estimates, J 940-J 990 (U.S. EPA, 1991).
1940-1990 NAPEE contains the data used in 1970-1990 Emissions. Finally, 1970-1997 NAPET
refers to National Air Pollutant Emission Trends Update: J 970-J 997 (U.S. EPA, 1998b).
January 2001
-------
2 Assigning Emissions to Industries in the United States, Revised Estimates
emissions to industries in the industrial sector.4 As was the case for 1970-1990 Emissions, this is
a technical report whose intended audience consists primarily of those individuals interested in
developing environmental accounts.
The 1970-1997NAPET provides information on paniculate matter (PM-10 and PM-2.5),
sulfur dioxide (SO2), nitrogen oxide (NOX), volatile organic compounds (VOCs), carbon
monoxide (CO) and lead (Pb) emissions from fuel combustion and processes. Procedures
developed in 1970-1990 Emissions to allocate fuel combustion emissions among the industries of
the industrial sector are used in this report to allocate fuel combustion emissions. The procedures
developed in 1970-1990 Emissions to allocate particulate matter (PM) emissions among the
industries of the industrial sector are used in this report to allocate PM-10 emissions from
processes.5 The procedures developed in 1970-1990 Emissions to allocate SOX emissions among
industries are used in this report to allocate SO2 emissions from processes. The remaining
pollutants (NOX, VOC, CO, and Pb) are reported by both the 1940-1990 NAPEE and the 7970-
1997 NAPET. The procedures developed in 1970-1990 Emissions to allocate these emissions
among the industries of the industrial sector are also used in this report.
There are five principal reasons for the differences between this report's estimates of
emissions and estimates in 1970-1990 Emissions. First, revised estimates of national emissions
are used in this report. Second, this report assigns emissions from the combustion of wood, coke
oven gas, and still gas to industries. Emissions from these fuels are not assigned to industries in
1970-1990 Emissions. Third, additional sources of VOC emissions are assigned to industries in
this report. Fourth, some emissions from processes assigned to the chemical industry in 7970-
7990 Emissions are now assigned to the mining sector (see Section 2.1.3. for the explanation).
Finally, in 1970-1990 Emissions the emissions from natural gas combustion are allocated
between emissions from boilers and emissions from gas pipelines and plants. Emissions from
boilers are assigned to industries in the industrial sector, and emissions associated with gas
pipelines and plants are assigned to the gas utilities (i.e., production and distribution) sector.
According to 1970-1990 Emissions, between 10 and 15 percent of emissions from natural gas
combustion are from boilers. The remaining emissions are from gas pipelines and plants. The
1970-1997 NAPET does not report separate emission estimates from these two source categories.
As a result, this report assigns all emissions from natural gas combustion by the industrial sector
to industries in the industrial sector. No emissions are assigned to the gas utilities sector.
This report's estimates of emissions differ from those in the 1970-1997NAPET (U.S.
EPA, 1998b) for several reasons. First, this report only assigns emissions to industries that
belong to the industrial sector. Hence, in this report sources such as motor vehicle emissions are
not assigned to industries. Second, this report estimates emissions from some process and fuel
4 The industrial sector consists of industries in the agriculture, mining, construction, and
manufacturing sectors.
5 Since PM-2.5 emissions are not reported until 1990, they are not assigned to industries.
January 2001
-------
Assigning Emissions to Industries in the United States, Revised Estimates 3
combustion categories whose emissions are not available (i.e., listed as "NA") in the 1970-1997
NAPET. Third, this report does not assign emissions from the "Miscellaneous Industrial
Processes" (Tier II) category to industries. Finally, this report excludes emissions from the
combustion of several types of fuel because there is insufficient information about which
industries consume these fuels.
The Office of Air and Radiation of the U.S. Environmental Protection Agency does not
endorse the emission estimates for the "NA" categories in the 1970-1997 NAPET. Hence, the
emission estimates in this report do not supersede the official estimates in National Air Pollutant
Emissions Trends Update: 1970-1997.
1.2. Organization of the Report
The remainder of this report consists of the following sections. In Section 2, the
methodology for allocating process emissions among industries is described. In Section 3, the
procedure for allocating fuel combustion emissions among industries is described. Finally,
section 4 summarizes the findings of this report and potential future extensions.
Appendix A lists the names and SIC codes of the national energy accounts (NBA) sectors
of the nonmanufacturing portion of the industrial sector. Appendix B lists the assignments of
SIC and NBA codes to various subcategories of the Solvent Utilization and Storage & Transport
(Tier I) categories. Appendix C summarizes information on the percentage of emissions in 1985
from categories whose emissions are listed as "NA" in 1984. Appendix D describes the
procedures employed to estimate missing observations for energy consumption and emissions.
Appendix E lists the estimates of emissions of the six criteria air pollutants for 1970 through
1997 for the twenty two-digit SIC manufacturing industries (SIC 20-39).6 Appendix F describes
the organization of the spreadsheet (SIC7097.WK3) whose contents are described in this report.
Finally, Appendix G describes the preparation of the report and the peer review process.
2. PROCESS EMISSIONS FROM THE INDUSTRIAL SECTOR
In the 1970-1997 NAPET, the methodology used to estimate process emissions for 1970
to 1984 is based on the methodology used in the 1940-1990 NAPEE. From 1970 to 1984,
estimates of process emissions are the product of (1) the activity level, (2) the emission factor
and (3) the control efficiency of each source category. In the 1970-1997 NAPET, the emission
estimates for 1985 to 1997 are derived primarily from the Aerometric Information Retrieval
System (AIRS) data, and extrapolations of 1985 data from the National Acid Precipitation
Assessment Program (NAPAP) and 1990 data from the National Emission Trends (U.S. EPA,
1998a, pp. 4-42, 4-44, 4-50, 4-52, and 4-56).
6 Since the most recent Manufacturing Energy Consumption Survey (MECS) reports data for
1994 and the most recent economic census contains data are from 1992, the allocation of fuel
combustion emissions for 1993 to 1997 is preliminary.
January 2001
-------
4 Assigning Emissions to Industries in the United States, Revised Estimates
2.1. Assigning SIC andNEA Codes to Categories of Process Emissions
Unlike the 1940-1990 NAPEE, the 19 70-199 7 NAPET does not assign SIC codes to its
process categories. As a result, information from 1970-1990 Emissions and the National Air
Pollutant Emissions Trends Procedures Document, 1900-1996 Projections 1999-2010 (U.S.
EPA, 1998a) is used to assign SIC and NEA codes to the categories of process emissions in the
1970-1997NAPET. The assignment of SIC and NEA codes is complicated by the fact that the
1970-1997 NAPET and the 1940-1990 NAPEE assign different names to some of the categories
of process emissions.
Table 1 identifies (X) which industries in the nonmanufacturing portion of the industrial
sector are assigned process emissions in 1970-1990 Emissions. The two-digit SIC manufacturing
industries assigned process emissions in 1970-1990 Emissions are identified (X) in Table 2.
Only industries assigned VOC emissions from categories classified as industrial processes in
1970-1990 Emissions and the 1970-1997 NAPET'are marked "X." The following source
categories, which are classified as industrial processes in 1970-1990 Emissions., are not classified
as industrial processes in the 1970-1997NAPET: (1) petroleum marketing & production, (2)
miscellaneous organic solvent evaporation, (3) surface coating, and (4) graphic arts. Section 2.2.
discusses the assignment of VOC emissions from these categories. If the only VOC emissions
assigned to an industry in 1970-1990 Emissions are from categories the 1970-1997 NAPET does
not classify as industrial processes, it is identified with a symbol other than an "X."
The U.S. EPA (1998a, p. 4-2) organizes its categories of emissions of criteria pollutants
into four levels (i.e., tiers). The Tier I and Tier II categories are identical for each of the six
criteria pollutants. The Tier in categories differ among the pollutants. Finally, Tier IV
corresponds to source classification code (SCC). Each SCC represents a "unique process or
function" that produces emissions. The Tier HI categories represent the most disaggregated data
available for 1970 to 1997.7
Table 3 summarizes the assignment of two-digit SIC codes (manufacturing industries)
and NEA codes (nonmanufacturing industries) to the Tier II categories of process emissions in
the 1970-1997 NAPET. For the Chemical & Allied Product Mfg. (Tier I) category, its Tier H
categories are not listed because all emissions are assigned to a single two-digit SIC code.
However, the Tier II categories associated with the Petroleum & Related Industries, Metals
Processing, and Other Industrial Processes (Tier I) categories are listed in Table 3. Information
contained in Table 3 is derived from the methodology used to estimate emissions for the years
prior to 1985 (see U.S. EPA, 1998a, Section 3); however, it serves as the principal guide for
distributing process emissions for 1970 to 1997.
7 Data on emissions from SCCs are available for the years from 1985 to the present.
January 2001
-------
Assigning Emissions to Industries in the United States, Revised Estimates
Table 1
Nonmanufacturing Industries Assigned Process Emissions, 1970-1990 Emissions
NBA
Code
01000
02000
03000
04000
05000
06001
06002
07010
07020
08001
08002
08003
09000
10000
11001
11002
12000
Industry Title
AGRICULTURE
Livestock and livestock products
Other agricultural products
Forestry and forestry products
Agricultural, forestry and fishery services
MINING
Iron and feroalloy ores mining
Uranium - radium - vanadium ores mining
Nonferrous metal ores mining
Anthracite coal mining
Bituminous and lignite coal mining
Crude petroleum
Natural gas
Natural gas liquids
Stone and clay mining and quarrying
Chemical and fertilizer mineral mining
CONSTRUCTION
Oil and gas well drilling
New construction
Maintenance and repair construction
PM
X
X
X
X
X
X
X
sox
NOX
VOCs
(MO)
(PMP)
(PMP)
(MO)
(MO)
CO
Pb
X
Notes:
PMP = VOC emissions from only the Petroleum Marketing & Production category
MO = VOC emissions from only the Miscellaneous Organic Solvent Evaporation category
January 2001
-------
Assigning Emissions to Industries in the United States, Revised Estimates
Table 2
Manufacturing Industries Assigned Process Emissions, 1970-1990 Emissions
SIC
code
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
Industry Groups
Food and kindred products
Tobacco products
Textile mill products
Apparel
Lumber and wood products, except furniture
Furniture and fixtures
Paper and allied products
Printing, publishing, and allied products
Chemicals and allied products
Petroleum refining and related products
Rubber and miscellaneous plastics products
Leather and leather products
Stone, clay, glass, and concrete products
Primary metal products
Fabricated metals products, ex. machinery and
transportation equipment
Industrial and commercial machinery and computer
equipment
Electronic and other electrical equip, and
components, ex. computer equip.
Transportation equipment
Measuring, analyzing and controlling instruments;
photographic, medical and optical goods; watches
and clocks
Miscellaneous manufacturing industries
PM
X
X
X
X
X
X
X
sox
X
X
X
X
X
NOX
X
X
X
X
X
VOCs
X
X
(so
(SC)
(SC)
(GA)
X
X
(SC)
X
X
(SC)
(SC)
(SC)
(SC)
CO
X
X
X
X
X
Pb
X
X
X
X
X
X
Notes: SIC 34 and SIC 35 reported positive quantities of Pb in all years except in
SC = VOC emissions from only the Surface Coating category
GA = VOC emissions from only the Graphic Arts category
1990.
January 2001
-------
Assigning Emissions to Industries in the United States, Revised Estimates
Table 3
Concordance Between Categories of Process Emissions and NEA/SIC Codes
Tier I Category
Chemical & Allied
Product Mfg.
Metals Processing
Metals Processing
Metals Processing
Petroleum & Related
Industries
Petroleum & Related
Industries
Petroleum & Related
Industries
Other Industrial Processes
Other Industrial Processes
Other Industrial Processes
Other Industrial Processes
Other Industrial Processes
Other Industrial Processes
Other Industrial Processes
Other Industrial Processes
Other Industrial Processes
Other Industrial Processes
Tier II Category
—
Nonferrous Metals Processing
Ferrous Metals Processing
Not Elsewhere Classified
Oil & Gas Production
Petroleum Refineries &
Related Industries
Asphalt Manufacturing
Agriculture, Food, & Kindred
Products
Textiles, Leather, & Apparel
Products
Wood, Pulp & Paper, &
Publishing Products
Rubber & Miscellaneous
Products
Mineral Products
Machinery Products
Electronic Equipment
Transportation Equipment
Construction
Miscellaneous Industrial
Processes
SIC codes (NEA codes in
parentheses)
28
33
33
33, (NEA 05000, 06001, 06002)
(NEA 08001, 08002 and 08003)
29
29
20, (NEA 01000, 02000, 03000
and 04000)
22,23,31
24, 25, 26
30
32, (NEA 09000 and 10000)
34
36
37
(NEA 11001, 11002, and 12000)
(see discussion in section
2.1.13.)
January 2001
-------
8 Assigning Emissions to Industries in the United States, Revised Estimates
Several of the Tier n categories in Table 3 include emissions from more than one industry
(e.g., "Textiles, Leather and Apparel Products" includes three two-digit SIC industries).
Some applications of the emission estimates require information about two-digit SIC
manufacturing industries. Hence, rules are devised to allocate the emissions among the industries
associated with each Tier II category. The following "Allocation Rules," which are listed in the
order in which they are invoked, are used to assign process emissions from Tier n categories to
industries:8
(1) If only one SIC or NEA code is assigned to a Tier n category, then emissions of all pollutants
are assigned to that industry. In this case, an industry may be assigned emissions of a pollutant
even if it was not assigned emissions of that pollutant in 1970-1990 Emissions.
(2) If two or more SIC or NEA codes belong to a Tier n category, but only one of the industries
is assigned emissions in 1970-1990 Emissions, then all emissions are allocated to the industry
assigned emissions in 1970-1990 Emissions. In this case, an industry is not assigned emissions
of a pollutant if it was not assigned emissions in 1970-1990 Emissions.
(3) If two or more of the SIC or NEA codes of a Tier II category are assigned emissions in 7970-
1990 Emissions, the shares of emissions assigned to each of those industries in 1970-1990
Emissions are used to assign the emissions from the 1970-1997 NAPETfor 1970 to 1990. The
percentage of the emissions assigned to each industry for 1991 to 1997 is estimated by adjusting
the industry's 1990 share of emissions by changes in its share of the output of the industries
assigned to that Tier II category. For each year from 1991 to 1997, the emission-output ratio for
each industry is scaled so that the sum of the emissions from the industries equals the emissions
from the Tier n category.
(4) If two or more industries are assigned to a Tier II category and none of the industries are
assigned emissions in 1970-1990 Emissions, the emissions are allocated to industries based on
the output shares of the industries assigned to that Tier n category. This allocation rule assumes
that the ratio of process emissions to industry output is identical for all industries that belong to
the Tier II category.
While the percentages vary from year to year, in 1997 more than 99 percent of SO2, CO, and Pb
process emissions are assigned using Allocation Rules 1 and 2. In 1997, more than 97 percent of
NOX emissions, more than 92 percent of PM-10 emissions, and more than 92 percent of VOC
emissions are assigned using Allocation Rules 1 and 2.9
8 These rules are modified in some instances when (1) additional information and/or (2)
emission estimates for Tier in categories are available.
9 If the percentage of VOC emissions includes emissions from the industrial process, solvent
utilization, and storage and transport categories, then more than 78 percent of VOC emissions are
assigned by Allocation Rules 1 and 2.
January 2001
-------
Assigning Emissions to Industries in the United States, Revised Estimates 9
The remainder of Section 2.1. describes the rationale for assigning SIC and NEA codes
to the Tier n and Tier in categories of the four Tier I categories that comprise the industrial
sector: (1) chemical and allied product mfg., (2) metals processing, (3) petroleum and related
industries, and (4) other industrial processes. Specific mention is made when "Allocation Rule 3"
or "Allocation Rule 4" is used to allocate emissions to industries.
2.1.1. Chemical & Allied Product Mfg.
The Chemical & Allied Product Mfg. (Tier I) category consists of seven Tier n categories
(see Table 4). The chemicals and allied products industry (SIC 28) is assigned the emissions
from the Chemical & Allied Product Mfg. category.
2.1.2. Metals Processing
The Metals Processing (Tier I) category consists of three Tier n categories. SIC 33 is
assigned emissions of all pollutants from the Nonferrous Metals Processing (Tier II), and the
Ferrous Metals Processing (Tier II) categories.
The Metals Processing, NEC (Tier II) category reports lead (Pb) emissions from the
"metals mining" and "other" (Tier HI) categories. In 1970-1990 Emissions, Pb emissions are
assigned to the metals mining industries, which are part of the mining sector. A comparison of
the quantity of Pb emissions assigned to the mining sector in the SIC7090.WK3 spreadsheet and
the 1970-1997 NAPETindicates that emissions from the "metals mining" (Tier HI) category
should be assigned to the mining sector. SIC 33 is assigned the Pb emissions from the "other"
(Tier ID) category.
Prior to 1985, PM-10 emissions from the Metals Processing, NEC (Tier n) category are
associated with the mining operations subcategory (U.S. EPA, 1998a, p. 3-124). Although some
of the processes in the Metals Processing, NEC (Tier n) category should be assigned to SIC 33
(see U.S. EPA 1998a, p. 4-12), the mining sector is assigned all PM-10 emissions from this
category.
Since no emissions of SO2, NOX, VOCs, and CO are assigned to the metal mining
industry in 1970-1990 Emissions., Allocation Rule 2 indicates that emissions from the Metals
Processing, NEC (Tier n) category should be assigned to SIC 33. In addition, processes
associated with SIC 33 are included in the Metals Processing, NEC (Tier II) category (see U.S.
EPA 1998a, p. 4-12). However, SIC 33 is assigned emissions of these pollutants from the
Nonferrous Metals Processing (Tier n), and the Ferrous Metals Processing (Tier n) categories.
As a result, the metal mining industry is assigned all SO2, NOX, VOCs, and CO emissions from
the Metals Processing, NEC (Tier II) category.
January 2001
-------
10 Assigning Emissions to Industries in the United States, Revised Estimates
2.1.3. Petroleum & Related Industries
The Petroleum and Related Industries (Tier I) category includes emissions from three
Tier n categories. SIC 29 is assigned emissions of all pollutants from the Petroleum Refineries
and Related Industries (Tier n), and the Asphalt Manufacturing (Tier n) categories.
For SO2 emissions, the Oil & Gas Production (Tier II) category includes the following
Tier HI categories: (1) natural gas and (2) other. Natural gas production and crude oil production
are two of the processes included in the Oil & Gas Production category (U.S. EPA, 1998a, p. 3-
128). The 1970-1990 Emissions assigns emissions from "sulfur recovery plants" processes to SIC
28, and the 1940-1990 NAPEE assigned these emissions to the mining sector. Appendix D of
1970-1990 Emissions discussed this discrepancy between it and the 1940-1990 NAPEE. Since
many of the processes in this Tier II category are related to mining activities and the North
American Industrial Classification System (NAICS) assigns sulfur recovered from natural gas to
the mining sector (see Appendix D of 1970-1990 Emissions), emissions of all pollutants from the
Oil & Gas Production category are assigned to the mining sector. Hence, this represents a
change from 1970-1990 Emissions.
2.1.4. Other Industrial Processes - Agriculture. Food & Kindred Products
The Agriculture, Food & Kindred Products (Tier n) category includes PM-10 emissions
from two Tier HI categories of grain elevators, four Tier in categories of grain mills, and the
"other" (Tier IE) category.
Emissions of PM from the (1) country elevators and (2) terminal elevators categories are
assigned to the agriculture sector in 1970-1990 Emissions. Appendix D of 1970-1990 Emissions
contains a discussion of the allocation of emissions from these categories. As a result, this report
assigns the PM-10 emissions from these two Tier in categories to the agriculture sector.
In this report, emissions of PM-10 from the (1) feed mills, (2) soybean mills, (3) wheat
mills, (4) other grain mills (Tier HI) categories are assigned to SIC 20 (see 1970-1990 Emissions,
p. A-5).
PM-10 emissions from the following processes are assigned to the "other" (Tier HI)
category (U.S. EPA, 2000, p. A-5): cotton ginning, cattle feedlots, and alfalfa dehydrators. In
1970-1990 Emissions, emissions from cattle feedlots are assigned to NEA 01000 and emissions
from cotton ginning are assigned to NEA 04000 (i.e., SIC 0724). The alfalfa dehydrators source
category is assigned to SIC 20 in 1970-1990 Emissions. Hence, "Allocation Rule 3" (see
Section 2.1.) assigns PM-10 emissions from the Agriculture, Food & Kindred Products category
to the agriculture sector and SIC 20. From 1970 to 1990, the cotton ginning and cattle feedlots
source categories generated between 94.5 and 97 percent and the alfalfa dehydrators source
category generated between 3 and 5.5 percent of the PM-10 emissions.
January 2001
-------
Assigning Emissions to Industries in the United States, Revised Estimates 11
From 1985 to 1997, the Agriculture, Food, and Kindred Products (Tier n) category
accounts for between 11 and 16 percent of the PM-10 process emissions from the Other
Industrial Processes (Tier I) category. The Agriculture, Food, and Kindred Products category
accounts for between 7 and 10 percent of the PM-10 process emissions of the industrial sector.
Since no VOC process emissions are assigned to the agriculture sector in 1970-1990
Emissions, all VOC emissions are assigned to SIC 20 (see U.S. EPA 2000, p. A-9). The
following Tier in categories: (1) vegetable oil mfg., (2) whiskey fermentation: aging, (3)
bakeries, and (4) other are associated with VOC emissions from the Agriculture, Food, and
Kindred Products (Tier II) category.
SO2, NOX, and CO process emissions are assigned to neither the agriculture sector nor
SIC 20 in 1970-1990 Emissions. As a result, "Allocation Rule 4" is used to allocate SO2, NOX,
and CO emissions between the agriculture sector and Food and kindred products (SIC 20).
Between 1985 and 1997, the Agriculture, Food, and Kindred Products (Tier n) category
accounts for between 0.5 and 1 percent of SO2 process emissions from the Other Industrial
Processes (Tier I) category. Between 1985 and 1997, this Tier II category accounts for less than
0.5 percent of SO2 process emissions from the industrial sector.
Between 1985 and 1997, the Agriculture, Food, and Kindred Products (Tier n) category
produces less than 1.5 percent of NOX process emissions from the Other Industrial Processes
(Tier I) category. Between 1985 and 1997, this Tier n category accounts for less than 1 percent
of NOX process emissions from the industrial sector.
Between 1985 and 1997, the Agriculture, Food, and Kindred Products (Tier n) category
produces less than 1 percent of CO process emissions from the Other Industrial Processes (Tier I)
category. Between 1985 and 1997, this Tier n category accounts for less than 0.1 percent of CO
process emissions from the industrial sector.
2.1.5. Other Industrial Processes - Textiles. Leather. & Apparel Products
In 1970-1990 Emissions of the three industries (SIC 22, SIC 23, and SIC 31) that
comprise the Textiles, Leather, & Apparel Products (Tier n) category, only SIC 22 is assigned
VOC process emissions. Hence, SIC 22 is assigned all VOC emissions from the Textiles,
Leather, & Apparel Products category.
Emissions of PM-10, SO2, NOX and CO are each less than 0.5 thousand short tons.
Between 1985 and 1997, this Tier II category accounts for less than 0.1 percent of PM-10, SO2,
NOX and CO process emissions from the industrial sector. Since none of these pollutants are
assigned to SIC 22, SIC 23, or SIC 31 in 1970-1990 Emissions, "Allocation Rule 4" is used to
allocate emissions from the Textiles, Leather, & Apparel Products (Tier n) category among the
three industries.
January 2001
-------
12 Assigning Emissions to Industries in the United States, Revised Estimates
2.1.6. Other Industrial Processes - Wood. Pulp & Paper. & Publishing Products
The Wood, Pulp & Paper, & Publishing Products (Tier II) category includes CO
emissions from the following Tier in categories: (1) sulfate pulping: rec. furnace/evaporator, (2)
sulfate (kraft) pulping: lime kiln and (3) other. For PM-10 emissions, the Wood, Pulp & Paper,
& Publishing Products (Tier n) category includes the following Tier HI categories: (1) sulfate
(kraft) pulping: lime kiln and (2) other.
In 1970-1990 Emissions, SIC 24 and SIC 26 are assigned PM emissions. Based on a
comparison of the quantity of PM-10 emissions assigned to SIC 24 and SIC 26 in the
SIC7090.WK3 spreadsheet and emissions of the two Tier HI categories in the 1970-1997
NAPET, SIC 26 is assigned emissions from the sulfate (kraft) pulping (Tier IE) category and SIC
24 is assigned PM-10 emissions from "other" (Tier HI) category.
In 1970-1990 Emissions, SIC 26 is assigned all SO2 and NOX emissions from the Wood,
Pulp & Paper, & Publishing Products category. In this report, SIC 26 is assigned all SO2 and
NOX emissions from the Wood, Pulp & Paper, & Publishing Products category.
In 1970-1990 Emissions, SIC 26 is assigned all CO emissions from the Wood, Pulp &
Paper, & Publishing Products category. Starting in 1985, the "other" (Tier HI) category in the
1970-1997NAPETreports some CO emissions. Section 2.3.3. discusses the possibility that
emissions from the sulfate (kraft) pulping: lime kiln (Tier IE) category were transferred to the
other two Tier in categories starting in 1985. Although some of the "other" (Tier ID) processes
are associated with SIC 24 and SIC 25, "Allocation Rule 2" indicates that emissions from the
"other" (Tier ID) category should be assigned to SIC 26. As a result, SIC 26 is assigned all CO
emissions from Wood, Pulp & Paper, & Publishing Products category.
In 1970-1990 Emissions, no VOC process emissions are assigned to any of the three two-
digit SIC codes that comprise the Wood, Pulp & Paper, & Publishing Products (Tier II) category.
As a result, "Allocation Rule 4" is used to allocate VOC emissions among the three industries of
the Wood, Pulp & Paper, & Publishing Products category. During 1985 to 1997, this Tier II
category produces between 10.5 and 28.5 percent of the VOC emissions from the Other
Industrial Processes (Tier I) category. Between 1985 and 1997, this Tier II category accounts for
between 2 and 8.5 percent of the VOC emissions of the industrial sector.
2.1.7. Other Industrial Processes - Rubber & Miscellaneous Products
SIC 30 is assigned emissions from the Rubber and Miscellaneous products category.
For VOC emissions, the "Rubber and Miscellaneous Plastics Products" (Tier n) category consists
of these Tier HI categories: (1) rubber tire mfg., (2) green tire spray, and (3) other.
January 2001
-------
Assigning Emissions to Industries in the United States, Revised Estimates 13
2.1.8. Other Industrial Processes - Mineral Products
For PM-10 emissions, the "Minerals Products" (Tier II) category includes the following
Tier HI categories: (1) cement mfg., (2) surface mining, (3) stone quarrying/processing, and (4)
other. For NOX emissions, the Minerals Products category consists of the following Tier HI
categories: (1) cement mfg., (2) glass mfg., and (3) other.
SIC 32 is assigned emissions of PM-10 from the cement mfg. (Tier in) category. In
1970-1990 Emissions, PM emissions are assigned to the mineral mining (NEA 10000) and coal
mining (NEA 07020) industries. The Mineral Products category includes emissions from
mineral and coal mining (see U.S. EPA, 1998a, p. 3-156). Hence, this report assumes PM-10
emissions from coal mining are assigned to the "surface mining" (Tier IE) category. This report
also assumes that PM-10 emissions from mineral mining activities are assigned to the "stone
quarrying/processing" (Tier HI) category. The mining sector is assigned emissions from the
"other" (Tier ID) category. This assignment is based on information about the quantities of PM
emissions assigned to SIC 32 and the mining sector by the SIC7090.WK3 spreadsheet and the
quantities of emissions from the four Tier in categories in the 1970-1997 NAPET.
The 1970-1997 NAPET assigns all Pb emissions from the Minerals Products category to
the cement mfg. (Tier IE) category. As a result, SIC 32 is assigned all Pb emissions from the
Mineral Products category.
SO2, NOX, CO, or VOC emissions are assigned to neither the mineral mining nor the coal
mining industries in 1970-1990 Emissions; however, SIC 32 is assigned SO2, NOX, CO, and VOC
emissions. In this report, SIC 32 is assigned emissions of these pollutants.
2.1.9. Other Industrial Processes - Machinery Products
SIC 34 is assigned emissions from the Machinery Products (Tier II) category.
2.1.10. Other Industrial Processes - Electronic Equipment
SIC 36 is assigned emissions from the Electronic Equipment (Tier II) category.
2.1.11. Other Industrial Processes - Transportation Equipment
SIC 37 is assigned emissions from the Transportation Equipment (Tier II) category.
2.1.12. Other Industrial Processes - Construction
For VOC emissions, the 1970-1997 NAPET estimates emissions from the Construction
(Tier II) category. In 1970-1990 Emissions, NEA 11002 and NEA 12000, which are associated
with the construction sector, are assigned VOC emissions.
January 2001
-------
14 Assigning Emissions to Industries in the United States, Revised Estimates
PM-10 is the only other pollutant with a separate estimate of emissions from the
Construction category.
2.1.13. Other Industrial Processes - Miscellaneous Industrial Processes
All industries in the industrial sector assigned emissions in 1970-1990 Emissions are
assigned emissions in this report. This section discusses the rationale for not assigning emissions
from the Miscellaneous Industrial Processes (Tier II) category.
The only SIC code mentioned in the list of processes associated with the Miscellaneous
Industrial Processes category is SIC 35 (U.S. EPA, 1998a, p. 4-15). In addition, this category
includes several processes assigned to SIC 36. There is also a listing for fabricated metal
products - arc welding, which is part of SIC 34. Finally, there is one process (photocopying
equipment mfg., toner grinding) included in this category that is assigned to SIC 38. Since no
additional information is available, industries are not assigned emissions from the Miscellaneous
Industrial Processes (Tier II) category.
Between 1985 and 1997, this Tier n category produces between 12.5 and 28 percent of
the VOC emissions from the Other Industrial Processes (Tier I) category. Between 1985 and
1997, this category accounts for between 1 and 2.5 percent of VOC emissions from all industrial
processes.
Between 1985 and 1997, this Tier n category produces between 0.5 and 1.5 percent of the
CO emissions from the Other Industrial Processes (Tier I) category. Between 1985 and 1997,
this Tier II category accounts for less than 0.5 percent of the VOC process emissions from the
industrial sector.
Between 1985 and 1997, this Tier II category accounts for between 3.5 and 5 percent of
the PM-10 emissions from the Other Industrial Processes (Tier I) category. Between 1985 and
1997, this category produces between 2.5 and 3 percent of the PM-10 process emissions from the
industrial sector.
Between 1985 and 1997, this Tier II category accounts for between 2 and 3 percent of the
NOX emissions from the Other Industrial Processes (Tier I) category. Between 1985 and 1997,
this Tier II category produces between 0.5 and 1.5 percent of the NOX process emissions from the
industrial sector.
Between 1985 and 1997, this Tier II category accounts for between 0.5 and 1.5 percent of
the SO2 emissions from the Other Industrial Processes (Tier I) category. Between 1985 and
1997, this category produces less than 0.5 percent of the SO2 process emissions of the industrial
sector.
January 2001
-------
Assigning Emissions to Industries in the United States, Revised Estimates 15
2.2. Assigning SIC andNEA Codes to Other Categories of Emissions
The assignment of emissions from the "Solvent Utilization" (Tier I) category is discussed
in Section 2.2.1. Section 2.2.2. discusses the assignment of emissions from the "Petroleum and
Petroleum Product Storage," the "Organic Chemical Storage," and the "Inorganic Chemical
Storage" (Tier n) categories, which are part of the "Storage & Transport" (Tier I) category.
Appendix B lists the concordance between SIC and NEA codes and the Tier n and Tier HI
categories of the Solvent Utilization and Storage & Transport (Tier I) categories.
Except for VOCs, emissions of all pollutants from these source categories are reported as
"NA" for the years prior to 1985. According to the post-1984 data, these source categories are
relatively small contributors to non-VOC emissions. As a result, only the VOC emissions from
these categories are assigned SIC or NEA codes.
There are other categories of VOC emissions that could be assigned to nonmanufacturing
industries within the industrial sector. Since the focus of this report is assigning emissions to
manufacturing industries, these remaining categories of VOC emissions are not assigned to
industries.
2.2.1. Solvent Utilization
1970-1990 Emissions assigns VOC emissions ("Miscellaneous Solvent Evaporation -
pesticides") to NEA 02000. The closest match to this category in the 1970-1997 NAPETis the
"Solvent Utilization - pesticide application" (Tier IE) category.
Using assignments of Solvent Utilization categories in 1970-1990 Emissions., the
construction sector is assigned VOC emissions from the "Solvent Utilization - Nonindustrial,
cutback asphalt" (Tier HI) category.
1970-1990 Emissions assigns VOC emissions ("Miscellaneous Solvent Evaporation -
architectural coating") to NEA 11002. The closest match to this category in the 1970-1997
NAPET is the "Surface Coating - architectural" (Tier HI) category.
1970-1990 Emissions assigns VOC emissions ("Surface Coating Operations -
Maintenance Coatings") to NEA 12000. The closest match to this category in the 1970-1997
NAPET is the "Surface Coating - maintenance coating" (Tier HI) category.
Using assignments of Solvent Utilization categories in 1970-1990 Emissions., SIC 22 is
assigned VOC emissions from the "Surface Coating - fabrics" (Tier ID) category.
Using assignments of Solvent Utilization categories in 1970-1990 Emissions, SIC 24 is
assigned VOC emissions from the "Surface Coating - flatwood" (Tier IE) category. SIC 25 is
assigned emissions from the "Surface Coating - wood furniture" and "Surface Coating - metal
January 2001
-------
16 Assigning Emissions to Industries in the United States, Revised Estimates
furniture" (Tier ID) categories. SIC 26 is assigned emissions from the "Surface Coating - paper"
(Tier IE) category. SIC 27 is assigned emissions from the "Solvent Utilization - Graphic Arts"
(Tier ID) category.
Using assignments of Solvent Utilization categories in 1970-1990 Emissions, SIC 30 is
assigned VOC emissions from the "Solvent Utilization - other industrial, plastic parts" and VOC
emissions from the "Solvent Utilization - Surface Coating, rubber & plastics mfg." (Tier ID)
categories.
Using assignments of Solvent Utilization categories in 1970-1990 Emissions, SIC 33 is
assigned VOC emissions from the "Solvent Utilization - Surface Coating, metal coils" and from
the "Solvent Utilization - Surface Coating, magnet wire" (Tier HI) categories.
In 1970-1990 Emissions, Sic 34 is assigned emissions from the following Surface
Coating Operations (Tier ID) categories: (1) Cans, (2) Other Metal Products, and (3)
Miscellaneous Processes. It is assumed that the "metal cans" (Tier IE) category of the Solvent
Utilization (Tier I) category in the 1970-1997 NAPETcorresponds to the "Cans" category.
Neither the "Other Metal Products" nor the "Miscellaneous Processes" categories are listed in the
1970-1997NAPET. It is assumed that the (1) misc. metal parts and (2) steel drums (Tier HI)
categories in the 1970-1997 NAPET correspond to the "Other Metal Products" and the
"Miscellaneous Processes" categories.
Using assignments of Solvent Utilization categories in 1970-1990 Emissions, SIC 35 is
assigned VOC emissions from the "Solvent Utilization - Surface Coating, machinery" (Tier HI)
category.
SIC 36 is assigned VOC emissions from the "Surface Coating Operations - Large
Appliances" source (Tier HI) category. In addition, SIC 36 is assigned VOC emissions from the
"Solvent Utilization - Surface Coating, electronic & other electrical" (Tier ID) category, which is
in the 1970-1997NAPET. The "electronic & other electrical" (Tier HI) category is not listed in
either the 1940-1990 NAPEE or 1970-1990 Emissions.
Using assignments of Solvent Utilization categories in 1970-1990 Emissions, SIC 37 is
assigned VOC emissions from the "Solvent Utilization - Surface Coating, autos & light trucks,"
"Solvent Utilization - Surface Coating, large ships," "Solvent Utilization - Surface Coating,
aircraft," and "Solvent Utilization - Surface Coating, railroad" (Tier HI) categories.
2.2.2. Storage & Transport
In 1970-1990 Emissions, SIC 29 is assigned the emissions of selected Tier in categories
from the "Petroleum & Petroleum Storage" (Tier n) category. First, SIC 29 is assigned
emissions from the "Petroleum Marketing and Production - Crude Oil Storage, Refinery Storage"
category in 1970-1990 Emissions. In the 1940-1990 NAPEE, this source is part of the "Crude Oil
January 2001
-------
Assigning Emissions to Industries in the United States, Revised Estimates 17
Production, Storage and Transfer" category. The 1940-1990 NAPEE assigned this category to
SIC industries 1311 and 4463. Second, SIC 29 is assigned emissions from the "Petroleum
Marketing and Production - Gasoline Storage at Refineries" category in 1970-1990 Emissions.
Appendix D in 1970-1990 Emissions discusses the differences in the assignment of VOC
emissions to industries between 1970-1990 Emissions and the 1940-1990 NAPEE.
The Petroleum and Petroleum Product Storage (Tier II) category includes emissions from
the storage of gasoline and crude oil at refineries (SIC 29) and emissions from the oil field
storage of crude oil (NEA08001) (U.S. EPA, 1998a, p. 3-192). Hence, "Allocation Rule 3" (see
section 2.1.) assigns VOC emissions from the Petroleum and Petroleum Product Storage category
to the Oil and gas extraction industry (SIC 13) and petroleum refining (SIC 29). During 1970 to
1990, between 70 and 78 percent of the VOC emissions from the Petroleum and Petroleum
Product Storage category are assigned to SIC 29. The remainder of the VOC emissions from the
Petroleum and Petroleum Product Storage category are assigned to SIC 13.
The Organic Chemical Storage (Tier n) category includes emissions from (1) waste
solvent recovery and (2) waste disposal from petrochemical manufacturing (U.S. EPA, 1998a, p.
3-200). As a result, VOC emissions from this Tier n category are assigned to SIC 28.
2.3. Assigning Values to Categories of Process Emissions listed as "NA "
In the 1970-1997 NAPET, estimates of emissions from some source categories are not
available (NA) until 1985.10 The availability of data to estimate emissions for 1970 to 1984 is
discussed in Sections of National Air Pollutant Emissions Trends Procedures Document, 1900-
1993 (U.S. EPA, 1994) and Section 3 of National Air Pollutant Emissions Trends Procedures
Document, 1900-1996, Projections 1999-2010 (U.S. EPA, 1998a). Each cell in Table 4 that is
marked with an "X" represents a Tier n category for which information about the activity levels,
emission factors, and control efficiencies are available prior to 1985. The empty cells are those
Tier n categories for which there is no discussion of data sources in either Procedures Document.
Empty cells in Table 4 also correspond to those Tier II categories assigned a "NA" from 1970 to
1984 in the 1970-1997 NACEPT.
Sections of'the National Air Pollutant Emissions Trends Procedures Document, 1900-
1993 (U.S. EPA, 1994), section 3 of the National Air Pollutant Emissions Trends Procedures
Document, 1900-1996, Projections 1999-2010 (U.S. EPA, 1998a), and source categories listed in
10 For a limited number of Tier II categories, emissions are not reported until 1990. These
categories are Paint, Varnish, Lacquer, Enamel Mfg. (CO and SO2 emissions; the SO2 emissions
are not reported until 1992), Electronic Equipment (NOX emissions), Construction (PM-10 and
VOC emissions). When emissions are finally reported for these categories, they produce small
quantities of emissions. As a result, these categories do not affect the discussion in section 2.3.
In addition, an "NA" cell can indicate that the emissions associated with that category are "...
contained in a more aggregate estimate" during the years prior to 1985.
January 2001
-------
18
Assigning Emissions to Industries in the United States, Revised Estimates
Table 4
Procedures Document Describes the Procedure to Estimate
Emissions from Selected Tier II Categories of Process Emissions
("X" = discussion of available data on activity levels, emission factors, and control efficiencies)
Tier I Category
Chemical & Allied
Product Mfg.
Chemical & Allied
Product Mfg.
Chemical & Allied
Product Mfg.
Chemical & Allied
Product Mfg.
Chemical & Allied
Product Mfg.
Chemical & Allied
Product Mfg.
Chemical & Allied
Product Mfg.
Metals Processing
Metals Processing
Metals Processing
Petroleum & Related
Industries
Petroleum & Related
Industries
Petroleum & Related
Industries
Tier n Category
Organic Chemical Mfg.
Inorganic Chemical
Mfg.
Polymer & Resin Mfg.
Agricultural Chemical
Mfg.
Paint, Varnish, Lacquer,
Enamel Mfg.
Pharmaceutical Mfg.
Other Chemical Mfg
Nonferrous Metals
Processing
Ferrous Metals
Processing
Not Elsewhere
Classified
Oil & Gas Production
Petroleum Refineries &
Related Industries
Asphalt Manufacturing
PM-10
X
X
X
X
X
X
X
X
X
SO2
X
X
X
X
X
X
NOX
X
X
X
X
CO
X
X
X
X
X
X
voc
X
X
X
X
X
X
X
X
X
X
January 2001
-------
Assigning Emissions to Industries in the United States, Revised Estimates
19
(Table 4 continued)
Procedures Document Describes the Procedure to Estimate
Emissions from Selected Tier II Categories of Process Emissions
("X" = discussion of available data on activity levels, emission factors, and control efficiencies)
Tier I Category
Other Industrial Processes
Other Industrial Processes
Other Industrial Processes
Other Industrial Processes
Other Industrial Processes
Other Industrial Processes
Other Industrial Processes
Other Industrial Processes
Other Industrial Processes
Other Industrial Processes
Tier n Category
Agriculture, Food, &
Kindred Products
Textiles, Leather, &
Apparel Products
Wood, Pulp & Paper, &
Publishing Products
Rubber &
Miscellaneous Products
Mineral Products
Machinery Products
Electronic Equipment
Transportation
Equipment
Construction
Miscellaneous Industrial
Processes
PM-10
X
X
X
SO2
X
X
NOX
X
X
CO
X
X
voc
X
X
X
January 2001
-------
20 Assigning Emissions to Industries in the United States, Revised Estimates
Appendix A of 1970-1990 Emissions, "NA" cells for Tier n industrial processes indicate the
unavailability of information required to estimate the emissions of those categories.11 Prior to
1985, some Tier II categories marked with an "X" may have Tier HI categories whose emissions
are listed as "NA." Although there is no written discussion of changes in the assignment of
emissions from Tier in categories, it appears that many of the "NA" cells of Tier HI categories
represent cases when emissions are included in other Tier HI categories prior to 1985.12
Since it is assumed that the "NA" cells of Tier n categories represent cases when the
emissions are not estimated, emissions from Tier II categories that list "NA" are estimated using
the backward indexing method. Since it is assumed that "NA" cells of Tier HI categories
represent cases when the emissions are included in another Tier in category, in most instances
emissions for these "NA" cells are not estimated.
The only Tier HI categories whose "NA" cells are estimated are the following: misc.
metal parts, steel drums, and electronic & other electrical categories. These categories are
associated with VOC emissions from the Surface Coating (Tier n) category. Emissions from the
misc. metal parts and steel drums categories are assigned to SIC 34. Metal cans, which is the only
other Surface Coating process assigned to SIC 34, reported an increase in emissions between
1984 and 1985. Emissions from "electronic & other electrical" are assigned to SIC 36. The only
other Surface Coating process assigned to SIC 36 - large appliances - reported a small decrease in
emissions between 1984 and 1985. In addition, between 1984 and 1985 there is a substantial
decline in VOC emissions from the miscellaneous (Tier HI) category, which is not assigned to an
industry. As a result, it is assumed that prior to 1985, VOC emissions from these three Tier HI
categories were included in the miscellaneous (Tier IE) category. If this assumption is correct,
assigning values to the "NA" cells of these three Tier HI categories does not overestimate the
VOC emissions assigned to the industrial sector.
11 An e-mail correspondence (dated May 30, 2000) from Mr. Bill Barnard, of E.H. Pechan and
Associates, indicates that CO emissions from the Agricultural Chemical Mfg. (Tier n) category is
a case when no information existed prior to 1985. He also mentions that CO emissions from
fireplaces and woodstoves (Tier II) categories are contained in the aggregate emissions for
residential burning (Tier n) category.
12 Comparing the trend in emissions of the Tier in category between 1970 and 1984 in the SIC
7090.WK3 spreadsheet (or the NEA7085.WK3 spreadsheet) with the trend in emissions in the
SIC7097.WK3 spreadsheet makes it possible to determine if the emissions of "NA" cells of a
Tier HI category were included in another Tier in category prior to 1985. Since some emission
factors, control efficiencies, and activity levels were revised, the emissions estimates for 1970 to
1984 in the SIC7090.WK3 spreadsheet and the SIC7097.WK3 spreadsheets are not identical.
However, there are similarities in the levels and trends of emissions reported by the two
spreadsheets. If the change in emissions reported by the two spreadsheets diverges between 1984
and 1985, this is interpreted as evidence that starting in 1985 there was a reassignment of
emissions among the Tier III categories.
January 2001
-------
Assigning Emissions to Industries in the United States, Revised Estimates 21
In the 1940-1990 NAPEE, process emissions are the product of (1) the activity level, (2)
the emission factor and (3) control efficiency associated with each source category. For 1970 to
1984, the 1970-1997 NAPET employs the same methodology to estimate emissions as the 1940-
1990 NAPEE. Information on changes in emission factors, control efficiencies, and activity
levels relative to 1985 is required to estimate emissions for Tier n categories whose emissions
are recorded as "NA" for 1970 to 1984. Since it not possible to determine the emission factors
and control efficiencies for the categories whose emissions are recorded as "NA," emissions from
these Tier n categories are estimated by adjusting their 1985 emissions based solely on changes
in the activity level of the category.13 This approach assumes that the emission factor and control
efficiency for these Tier n categories are constant between 1970 and 1985.
Physical outputs of processes were the activity levels used by the U.S. EPA (1998a, pp. 3-
2 to 3-5) to estimate emissions for 1970 to 1984. After 1984, the U.S. EPA (1998a, pp. 4-42 and
4-43) uses the growth of earnings (i.e., payments to labor) by two-digit SIC industry from the
Bureau of Economic Analysis (BEA) of the U.S. Department of Commerce to update point
source emission estimates. This report uses changes in the value of "real" output of the
agriculture sector, the four two-digit SIC mining industries, the construction sector and the
twenty two-digit SIC manufacturing industries.
The industrial production indexes of the Board of Governors of the Federal Reserve
System are used to estimate changes in "real" output for the four two-digit SIC mining industries
and the twenty two-digit SIC manufacturing industries (see Armitage and Tranum, 1990 and
Board of Governors, 1999). Estimates of output of the agriculture and construction sectors are
from the Bureau of Economic Analysis of the U.S. Department of Commerce (Yuskavage, 1996;
Lum and Yuskavage, 1997; and Lum and Moyer, 1998).14
The relative importance of the Tier n categories whose process emissions are listed as
"NA" in the 1970-1997 NAPET varies among the pollutants. Appendix C reports the percentage
13 If the actual emission factor is higher and/or the control efficiency is lower prior to 1985
and only the change in the activity level is used to estimate emissions between 1970 and 1984,
then emissions from a process are underestimated. Since only changes in the activity level are
used to estimate emissions, the estimates probably underestimate the emissions. However, for the
purposes of this report, underestimating emissions is preferable to assuming that those Tier n
categories generated no emissions for 1970 to 1984.
14 BEA data on gross product by industry (i.e., value added) for 1977 to 1997 are derived
using a different methodology than the methodology that it employed to estimate the gross
product by industry for 1970 to 1976. The estimates of output produced by the Board of
Governors of the Federal Reserve System are also generated by a different methodology than that
used by the BEA for 1977 to 1997.
January 2001
-------
22 Assigning Emissions to Industries in the United States, Revised Estimates
of process emissions in 1985 for those categories whose emissions are estimated in this report.15
Appendix D describes the backward indexing method used to estimate emissions for the
categories listed as "NA" prior to 1985.
The remainder of Section 2.3. discusses source categories whose emissions are listed as
"NA" in the 1970-1997 NAPET. In addition, evidence is presented on the accuracy of the
assumption that emission factors and control efficiencies are constant during the years from 1970
to 1985.16 Finally, instances of significant changes in emissions between 1984 and 1985, which
are reported in the SIC7097.WK3 spreadsheet, are discussed. This allows an analysis of errors
resulting from assuming all "NA"cells of Tier n categories must be estimated and "NA"cells of
Tier HI categories reflect emissions included in other Tier HI categories prior to 1985.
2.3.1. N(X emissions
Tier n categories that report "NA" for emissions in 1984, account for 39.3 percent of NOX
process emissions in 1985. The seventeen source categories of NOX process emissions, which are
in the SIC7090.WK3 spreadsheet, report no control efficiencies from 1970 to 1985. In addition,
the National Air Pollutant Emissions Trends Procedures Document, 1900-1996, Projections
1999-2010, (U.S. EPA, 1998a) indicates that no control efficiencies are associated with processes
that generate NOX emissions. Except for the emission factors of the (1) nitric acid and (2) iron &
steel - roll & finish processes, which show higher emission factors from 1970 to 1984 relative to
their 1985 emission factors, there are constant emission factors for the source categories. As a
result, with the exception of two processes previously mentioned, changes in activity levels are
the sole determinants of changes in NOX process emissions.
Between 1984 and 1985, there is a substantial decline in NOX emissions from the
Inorganic Chemicals (Tier n) category. The quantity of NOX emissions from the Agricultural
Chemicals (Tier n) category changed from "NA" in 1984 to a substantial amount in 1985. In
spite of this, Agricultural Chemicals is a Tier n category and it is assumed that the "NA" for its
emissions indicates that data are not available prior to 1985.
15 In Appendix C, the value in the column entitled "1985 Emissions of cells listed as 'NA' in
1984" for the Other Industrial Processes (Tier I) category includes the emissions from the
Miscellaneous Industrial Processes (Tier n) category. Emissions from the Miscellaneous
Industrial Processes (Tier n) category are not assigned to an industry.
16 The SIC7090.WK3 spreadsheet, which contains the data used to estimate the emissions
reported in 1970-1990 Emissions, provides insights into the extent of changes in emission factors
and control efficiencies from 1970 to 1985. Appendix A of 1970-1990 Emissions contains a
complete listing of the source categories in the SIC7090.WK3 spreadsheet that are referred to in
the remainder of Section 2.3.
January 2001
-------
Assigning Emissions to Industries in the United States, Revised Estimates 23
2.3.2. VOC emissions
Tier n categories that report "NA" for emissions in 1984, account for 9.6 percent of the
VOC process emissions in 1985. If the categories from the Solvent Utilization and Storage &
Transport categories assigned to industries in this report are included, then source categories that
report "NA" for VOC emissions in 1984, represent 6.6 percent of the VOC emissions in 1985.
The SIC7090.WK3 spreadsheet and the National Air Pollutant Emissions Trends
Procedures Document (U.S. EPA, 1998a) report constant emission factors for the two processes
assigned to SIC 33 (metals processing). In addition, the SIC7090.WK3 spreadsheet reports no
control efficiencies for either of these source categories. Hence, in the SIC7090.WK3
spreadsheet only changes in activity levels determine changes in VOC emissions from SIC 33.
Of the six processes from the "Other Industrial Processes" category assigned to
industries, no control efficiencies are reported by either the SIC7090.WK3 spreadsheet or the
National Air Pollutant Emissions Trends Procedures Document, 1900-1996 Projections 1999-
2010 (U.S. EPA, 1998a). According to the SIC7090.WK3 spreadsheet, all processes in this Tier
I category had constant emission factors between 1970 and 1985. This category accounts for 57
percent of the VOC process emissions in 1985 that are reported a "NA" in 1984.
Among the fifty processes assigned to SIC 28 (chemicals) in the SIC7090.WK3
spreadsheet, the emission factors changed for the following two processes: (1) manufacture of
petrochemicals - other products and (2) manufacture of petrochemicals - fugitive. From 1970 to
1984, the control efficiencies of five processes in this Tier I category were lower than their 1985
control efficiencies.
According to the SIC7090.WK3 spreadsheet, the eleven processes assigned to SIC 29
(petroleum industry) had constant emission factors between 1970 and 1985. From 1970 to 1984,
the control efficiencies of five of these processes were less than their 1985 levels.
Of the source categories in the SIC7090.WK3 spreadsheet that correspond to processes
within either the Solvent Utilization or Storage & Transport (Tier I) categories, only the gasoline
storage at refineries category is assigned control efficiencies. From 1970 to 1984, the control
efficiencies of this process were less than its 1985 level. According to the SIC7090.WK3, the
Graphics Arts category and the five categories associated with oil production and storage
spreadsheet exhibit constant emission factors between 1970 and 1985. However, the emission
factors were higher from 1970 to 1984, relative to their 1985 levels, for fourteen of the eighteen
categories that belong to Surface Coating Operations category in the SIC7090.WK3 spreadsheet.
There are several instances in which "NA" cells might represent cases when the
emissions of a category are classified in another category prior to 1985. There are five Tier in
categories associated with the Other Chemical Mfg. (Tier II) category. Of these five categories,
the carbon black mfg. (Tier IE) category is the only one that reports emissions prior to 1985. The
January 2001
-------
24 Assigning Emissions to Industries in the United States, Revised Estimates
printing ink mfg., fugitives unclassified, carbon black furnace: fugitives, and the "other" (Tier
IE) categories report "NA" emissions. Until 1984, the changes in the emissions from the carbon
black category in the SIC7090.WK3 spreadsheet are similar to the changes reported for the
carbon black mfg. (Tier HI) category in the SIC7097.WK3 spreadsheet. The SIC7097.WK3
spreadsheet reports a substantial decline in the VOC emissions between 1984 and 1985 from the
carbon black mfg. (Tier HI) category. This decline is not reported for the carbon black category
in the SIC7090.WK3 spreadsheet. The increase in emissions between 1984 and 1985 from the
Other Chemical Mfg. (Tier n) category can be explained if starting in 1985 some emissions were
transferred from the carbon black mfg. (Tier HI) category to the other Tier HI categories.
Five Tier HI categories are associated with VOC emissions from the Petroleum Refineries
& Related Industries (Tier n) category. Prior to 1985, the vacuum distillation, cracking units,
and "other" (Tier HI) categories report emissions. The "process units turnarounds" and
"petroleum refinery fugitives" (Tier HI) categories report "NA." Until 1984, changes in
emissions from the vacuum distillation, cracking units, and "other" (Tier HI) categories in the
SIC7097.WK3 spreadsheet are similar to the changes reported by their corresponding categories
in the SIC7090.WK3 spreadsheet. However, between 1984 and 1985 the SIC7097.WK3
spreadsheet reports substantial declines in the emissions from all three (Tier HI) categories.
These declines are not reported for the comparable categories in the SIC7090.WK3 spreadsheet.
Adding emissions from the "process units turnarounds" and "petroleum refinery fugitives" (Tier
IE) categories to the emissions from the other three Tier in categories yields a larger decline in
the emissions from the Petroleum Refineries & Related Industries (Tier II) category in the
SIC7097.WK3 spreadsheet than the SIC7090.WK3 spreadsheet. Hence, emissions from the
"process units turnarounds" and petroleum refinery fugitives (Tier in) categories may not have
been included in the vacuum distillation, cracking units, and "other" (Tier HI) categories prior to
1985. Since they are Tier HI categories, the "NA" cells for the "process units turnarounds" and
"petroleum refinery fugitives" categories are not estimated.
There are four Tier HI categories of VOC emissions associated with the Agriculture,
Food, & Kindred Products (Tier n) category. Three of these categories report emissions for 1970
to 1984: (1) vegetable oil mfg., (2) whiskey fermentation: aging, and (3) bakeries. The only Tier
in category that does not report emissions is the "other" category. The changes between 1984
and 1985 in the emissions associated with the whiskey fermentation: aging, and bakeries source
categories in the SIC7090.WK3 spreadsheet are similar to the changes in the SIC7097.WK3
spreadsheet. Between 1984 and 1985, the SIC7097.WK3 spreadsheet reports a substantial
decline in emissions from the vegetable oil mfg. (Tier HI) category; however, a small decrease is
reported in the SIC7090.WK3 spreadsheet for its corresponding source categories. The
emissions of the Agriculture, Food, & Kindred Products (Tier n) category in the SIC7097.WK3
spreadsheet and the comparable categories in the SIC7090.WK3 spreadsheet are similar for 1984
and 1985. Hence, it appears that most - maybe all - of the decline in the emissions from the
vegetable oil mfg. (Tier HI) category is the consequence of emissions being reassigned from the
vegetable oil mfg. (Tier HI) category to the "other" (Tier ID) category.
January 2001
-------
Assigning Emissions to Industries in the United States, Revised Estimates 25
Three Tier in categories are associated with VOC emissions from the Rubber &
Miscellaneous Plastic Products (Tier n) category. Among these Tier in categories, only the ruber
tire mfg. (Tier III) category reports emissions prior to 1985. Prior to 1985, emissions from the
green tire spray and "other" Tier HI categories are reported as "NA." Until 1984, the changes in
emissions from the rubber tire category in the SIC7090.WK3 spreadsheet are similar to the
changes reported for the rubber tire mfg. (Tier HI) category of the SIC7097.WK3 spreadsheet. A
significant decline in the emissions between 1984 and 1985 is reported for the rubber tire mfg.
(Tier in) category in the SIC7097.WK3 spreadsheet; however, there is only a slight decline in
emissions reported for the rubber tire category by the SIC7090.WK3 spreadsheet. In the
SIC7097.WK3 spreadsheet, the decline in emissions between 1984 and 1985 from the Rubber &
Miscellaneous Plastic Products (Tier II) category is not as substantial as the decline reported for
the ruber tire mfg. (Tier III) category. Starting in 1985, emissions are reported for the green tire
spray and "other" Tier in categories. Hence, the decline in VOC emissions associated with the
ruber tire mfg. (Tier HI) category seems to be at least partly the result of reassigning emissions
from the rubber tire (Tier HI) category to the other Tier HI categories.
2.3.3. CO emissions
Tier n categories that have "NA" listed for emissions in 1984, account for 1.5 percent of
the CO process emissions in 1985. According to the SIC7090.WK3 spreadsheet, the emission
factors of all twenty-seven sources of CO emissions are constant from 1970 to 1985.
Of the two processes from the "Other Industrial Processes" category assigned to
industries, no control efficiencies are reported by either the SIC7090.WK3 spreadsheet or
National Air Pollutant Emissions Trends Procedures Document, 1900-1996 Projections 1999-
2010 (U.S. EPA, 1998a). Hence, in the SIC7090.WK3 spreadsheet only changes in activity levels
determine changes in CO emissions from the "Other Industrial Processes" category.
Of the thirteen source categories assigned to SIC 28 (chemical industry), control
efficiencies are reported for three categories (charcoal, carbon black production - oil process, and
carbon black production - gas process. During 1970 to 1984, the control efficiencies of these
three processes were lower than their 1985 control efficiencies.
Of the five source categories assigned to SIC 29 (petroleum industry) in the
SIC7090.WK3 spreadsheet, control efficiencies are reported for two categories (Petroleum
refineries, fluid catalytic cracking [FCC] and Petroleum refineries, thermal catalytic cracking
[TCC]). During 1970 to 1984, the control efficiencies of these two processes were lower than
their 1985 levels.
Of the seven source categories assigned to SIC 33 (metals processing) in the
SIC7090.WK3 spreadsheet, control efficiencies are reported for two categories (Iron Foundries
and Steel Manufacturing - B.O.F.). During 1970 to 1984, the control efficiencies of these two
processes were lower than their 1985 control efficiencies.
January 2001
-------
26 Assigning Emissions to Industries in the United States, Revised Estimates
There are several instances in which "NA" cells might represent cases when emissions of
a category are assigned to another category prior to 1985. There are three Tier HI categories
associated with the Other Chemical Mfg. (Tier II) category. Of these three categories, the carbon
black mfg. (Tier HI) category is the only Tier in category that reports emissions prior to 1985.
The carbon black furnace: fugitives, and the "other" (Tier HI) categories report that emissions are
"NA." Until 1984, the changes in emissions from the carbon black category in the
SIC7090.WK3 spreadsheet are similar to the changes in CO emissions reported for the carbon
black mfg. (Tier HI) category. Between 1984 and 1985, the SIC7097.WK3 spreadsheet reports a
larger decline in the CO emissions from the carbon black mfg. (Tier IE) category than is reported
in the SIC7090.WK3 spreadsheet. Between 1984 and 1985, there is an increase in CO emissions
from the Other Chemical Mfg. (Tier n) category as a result of emissions being assigned to the
carbon black furnace: fugitives, and the "other" (Tier HI) categories. Hence, emissions from the
carbon black furnace: fugitives and the "other" (Tier IE) categories might have been excluded
from the carbon black mfg. (Tier HI) category prior to 1985. However, since they are Tier in
categories, the "NA" cells for these two categories are not estimated.
Three Tier in categories are associated with the Nonferrous Metals Processing (Tier II)
category. The aluminum anode baking and prebake aluminum cell (Tier ID) categories report
emissions prior to 1985, while the "other" (Tier ID) category reports "NA." From 1970 through
1984, the changes in CO emissions from the aluminum source category in the SIC7090.WK3
spreadsheet are similar to the changes of the combined emissions from the aluminum anode
baking and prebake aluminum cell (Tier HI) categories in the SIC7097.WK3 spreadsheet. The
change in the CO emissions between 1984 and 1985 of the Nonferrous Metals Processing (Tier
II) category is similar to the change in the aluminum source category in the SIC7090.WK3
spreadsheet. However, between 1984 and 1985 there is a substantial decline in emissions
associated with the aluminum anode baking (Tier HI) category. The amount of the decline is
similar to the emissions reported for the "other" (Tier IE) category. Hence, the decline in
emissions associated with the aluminum anode baking (Tier HI) category is assumed to be the
consequence of a transfer of emissions from it to the "other" source category.
Three Tier in categories are associated with the Wood, Pulp & Paper, and Publishing
Products (Tier n) category. Only the sulfate (kraft) pulping: lime kiln (Tier III) category reports
emissions prior to 1985. The sulfate pulping: rec. furnace/evaporator and the "other" (Tier HI)
categories report "NA." Until 1984, the changes in emissions from the kraft pulping source
category reported in the SIC7090.WK3 spreadsheet are comparable to the changes in CO
emissions reported for the sulfate (kraft) pulping: lime kiln (Tier HI) category. However, the
SIC0797.WK3 spreadsheet reports a substantial decline in the CO emissions between 1984 and
1985 from the sulfate (kraft) pulping: lime kiln (Tier HI) category. This decline in emissions is
not reported in the SIC7090.WK3 spreadsheet. Even after including emissions from all three Tier
in categories, the decline in CO emissions from the Wood, Pulp & Paper, and Publishing
Products (Tier n) category that is reported in the SIC7097.WK3 spreadsheet is not comparable
with the change in emissions from the kraft pulping source category in the SIC7090.WK3
spreadsheet. Hence, it appears that there is a transfer of emissions from the sulfate (kraft)
January 2001
-------
Assigning Emissions to Industries in the United States, Revised Estimates 27
pulping: lime kiln (Tier IE) category to the sulfate pulping: rec. furnace/evaporator and the
"other" (Tier ID) categories.
2.3.4. PM-10 emissions
The 19 70-199 7 NAPET reports no process emissions of PM-10 for all source categories
for 1971 to 1974. This report estimates those emissions through the use of the two-way indexing
method that is discussed in Appendix D.
Tier n categories that report "NA" emissions in 1984, account for 4.9 percent of the PM-
10 process emissions in 1985. Of the 135 processes that are sources of PM emissions in
SIC7090.WK3, only four exhibit higher emission factors for 1970 to 1984 relative to their
emission factors for 1985. The SIC7090.WK3 spreadsheet indicates control efficiencies for a
number of the source categories of PM-10 emissions, fluctuated between 1970 and 1985. For
example, of the fourteen processes assigned to SIC 28 (chemical industry) in the SIC7090.WK3
spreadsheet, the control efficiencies of nine are constant between 1970 and 1985. During 1970 to
1984, the control efficiencies of three processes were both higher and lower than their 1985
emission factors, and the control efficiencies of two processes were lower than their 1985
emission factors.
No "NA" cells appear to represent instances when the emissions of a category are
classified in another category prior to 1985. This reflects the fact that emissions from the
categories whose cells are listed as "NA"in 1984 are a relatively small percentage of PM-10
emissions in 1985.
2.3.5. SO2 emissions
Tier n industrial process categories reporting "NA" emissions in 1984, account for 2.2
percent of the SO2 process emissions in 1985. From 1970 to 1985, no control efficiencies for any
of the source categories that generate SOX emissions are reported by either the SIC7090.WK3
spreadsheet or the National Air Pollutant Emissions Trends Procedures Document, 1900-1996
Projections 1999-2010 (U.S. EPA, 1998a). Hence, only changes in the emission factors and
activity levels influence changes in the twenty-five categories of SOX process emissions.
The SIC7090.WK3 spreadsheet lists four processes, which are assigned to industries, in
the "Other Industrial processes" category. During 1970 to 1985, the emission factors of three of
the categories were constant. During 1970 to 1984, the emission factor of one process (cement
manufacturing) was both higher and lower than its 1985 emission factor.
In the SIC7090.WK3 spreadsheet, four processes are the source of the SOX emissions
from SIC 29 (chemical industry). Among the processes in the Chemical & Allied Products Mfg.
(Tier I) category, "carbon black" is assigned a "NA" in 1970 and 1971. Since the SIC7090.WK3
spreadsheet reports no emission factor for the carbon black source category in 1970 and 1971,
January 2001
-------
28 Assigning Emissions to Industries in the United States, Revised Estimates
this report assigns no emissions to the carbon black process category for 1970 and 1971. The
SOX emission factor from carbon black processes increased between 1972 and 1985. From 1970
to 1984, the emission factors of the four processes are higher than their 1985 levels.
Of the twelve processes assigned to SIC 33 (metals processing) in the SIC7090.WK3
spreadsheet, eight exhibited constant emission factors during 1970 to 1985. During 1970 to
1984, the emission factor for the Iron and steel (roll and finish) process fluctuated both above and
below its 1985 emission factor. During 1970 to 1984, the emission factors of the three copper
smelting processes are higher than their 1985 emission factors.
Of the five processes assigned to SIC 29 (petroleum industry), three exhibited constant
emission factors between 1970 and 1985. During 1970 to 1984, the emission factor of oil
process heaters is higher than its 1985 emission factor. Between 1970 and 1984, the emission
factor of gas process heaters is below its 1985 level.
No "NA" cells appear to represent the case when the emissions of a category are
classified in another source category prior to 1985. This reflects the fact that 1985 emissions
associated with categories whose cells are listed as "NA"in 1984 are a relatively small
percentage of SO2 emissions.
2.3.6. Pb emissions
Those source categories that have "NA" listed for emissions in 1984, account for no Pb
emissions in 1985.
3. EMISSIONS FROM FUEL COMBUSTION
The second step in assigning emissions to industries is to allocate emissions from fuel
combustion. An industry's allocation of fuel combustion emissions is based on its consumption
of fuel for heat and power.17 An industry's share of consumption of a fuel for heat and power is
assumed to equal its share of the emissions from combustion of that fuel. The product of an
industry's share of consumption of a type of fuel by the industrial sector and emissions from
combustion of that fuel by the industrial sector yields the estimate of that industry's emissions
17 The values of fuels purchased for heat, power and electricity generation are used to allocate
fuel combustion emissions among the industries of the industrial sector. The estimates of
emissions assigned to electric utilities in the 1940-1990 NAPEE and the 1970-1997 NAPETXQ
for emissions from electric utility plants (see National Air Pollutant Emission Trends Procedures
Document, 1900-1996Projections 1999-2010, U.S. EPA, 1998). All SCCs associated with
cogeneration are classified in either the Fuel Combustion Industrial or the Fuel Combustion
Other (Tier I) categories. Hence, it is assumed that emissions from electricity generated by
industries in the industrial sector are included in emission estimates from fuel combustion of the
industrial sector.
January 2001
-------
Assigning Emissions to Industries in the United States, Revised Estimates 29
from combustion of that fuel. This procedure, which is identical to the procedure used in 7970-
1990 Emissions, assumes that a standard boiler is used by all industries.
3.1. Fuel Consumption of the Manufacturing Portion of the Industrial Sector
Emissions from fuel combustion by the twenty two-digit SIC industries of the
manufacturing sector are allocated using information on fuel consumption for heat and power.
The two primary sources of information on fuel consumption by the manufacturing sector are the
National Energy Accounts (NEA) and the Derived Annual Estimates of Manufacturing Energy
Consumption (DAEMEC) data (U.S. DOE, 1992 and 1998a). Among the fuels that the NEA
provides information on are the following: (1) coal (NEA energy product code 130), (2) wood for
fuel (NEA energy product code 180), (3) natural gas (dry from gas plants) and other utility gases
(NEA code energy product 221), (4) still gas, (5) coke oven gas, (6) distillate fuel oil (NEA
energy product code 243), and (7) residual fuel oil (NEA energy product code 250).
The DAEMEC (U.S. DOE, 1992 and 1998a) provides data on energy consumption for
heat and power (in Btu) by the twenty two-digit SIC manufacturing industries for 1974 to 1994
for the following fuels: (1) coal (excluding coal coke and breeze), (2) residual fuel oil, (3)
distillate fuel oil, (4) natural gas, (5) LPG, (6) coke and breeze, and (7) "other" fuels.18 The
energy consumption data for 1985, 1988, 1991, and 1994, which are published in the DAEMEC,
are from the "Total Consumption of Offsite-Produced Energy for Heat, Power, and Electricity
Generation" tables in the Manufacturing Energy Consumption Survey (U.S. DOE, 1988a, 1991,
1994, and 1997).19 These data are consistent with the energy consumption data published in the
Annual Survey of Manufactures, which is the basis of the pre-MECS estimates in the DAEMEC
for the manufacturing industries (U.S. DOE, 1992, p. 4).
Although the NEA and DAEMEC data are derived using different assumptions (see U.S.
DOE, 1992, pp. 3-4), there are three advantages to using the NEA data.20 First, the NEA
provides data from 1958 to 1985, while the DAEMEC data starts in 1974. Hence, using the NEA
18 In 1994, over two-thirds of the consumption of "other" fuels (in Btu) is accounted for by the
following three industries (see U.S. DOE, 1997, p. 70): (1) paper and allied products (SIC 26),
(2) chemicals and allied products (SIC 28), and (3) petroleum and coal products (SIC 29).
19 Throughout the remainder of this report, the 1985 MECS, 1988 MECS, 1991 MECS and
1994 MECS refer to the U.S. Department of Energy (U.S. DOE) publications (U.S. DOE, 1988a,
1991, 1994, 1997) that report the results of the triennial Manufacturing Energy Consumption
Survey.
20 Among the differences between the DAEMEC and NEA data, the DAEMEC estimates
include only offsite-produced energy; while NEA data includes off-site produced energy and
some sources of onsite-produced energy (see U.S. DOE, 1992, p. 4 and Price and Wendling,
1986, p. 4).
January 2001
-------
30 Assigning Emissions to Industries in the United States, Revised Estimates
allows an annual time series of emissions to start in 1970. Second, NEA data provide estimates
of fuel consumption by the nonmanufacturing industries of the industrial sector. When the
Annual Survey of Manufactures did not report fuel consumption by an industry, the NEA
estimated the fuel consumption of those industries.
Given the availability of data about energy consumption and emissions from fuel
combustion, emissions from the combustion of the following fuels are assigned to industries:
(1) coal, (2) residual fuel oil, (3) distillate fuel oil, (4) natural gas, (5) wood, and (6) coke oven
gas and still gas. Consumption of (1) wood and (2) coke oven gas and natural gas reflects fuel
produced and consumed at the same establishment. The sources of the estimates of consumption
of these fuels are discussed in Sections 3.3.3. and 3.3.4.
Following the procedures discussed in Sections 3.2.1.2. and 3.2.1.3. in 1970-1990
Emissions, adjustments are made to the fuel consumption data from 1970 to 1984 for selected
industries. The change in the methodology employed by the U.S. EPA to estimate emissions
from fuel combustion provides the rationale for switching from the NEA data to the DAEMEC
data. As a result, the NEA data are used to allocate fuel combustion emissions among industries
for 1970 through 1984. DAEMEC (U.S. DOE, 1992 and 1998a) provides the data on fuel
consumption employed to allocate fuel combustion emissions among the twenty two-digit SIC
manufacturing industries for 1985 to 1994.21 Energy consumption for 1995 to 1997 by
manufacturing industries is estimated using the forward indexing method (see Appendix D).
Although the fuel consumption data for 1985 to 1997 are listed in Btu, these data are used to
distribute emissions in the same manner as the data for 1970 to 1984.
The magnitude of the discrepancies between the NEA and DAEMEC data on fuel
consumption for heat and power by the manufacturing sector are illustrated in the following
table. This table lists the respective totals for 1981, which is the last year the Annual Survey of
Manufactures collected information on the quantities of fuels consumed by the manufacturing
sector:22
21 The shares of coal, residual fuel oil, distillate fuel oil, and natural gas consumption assigned
to industries in Sheets AB, AC, AD, and AG of the SIC7097.WK3 spreadsheet do not appear to
experience a dramatic change between 1984 (NEA data) and 1985 (DAEMEC data).
22 Since 1981 is the last year the NEA data reports survey estimates of fuel consumption by
the manufacturing sector, the NEA estimates of energy consumption for 1982 to 1985 are
generated by a forward indexing method.
January 2001
-------
Assigning Emissions to Industries in the United States, Revised Estimates
Fuel for Heat and Power,
Manufacturing (1981)
Coal (1000 short tons)
Natural Gas (billion cu. ft.)
Distillate Fuel Oil (1000 barrels)
Residual Fuel Oil (1000 barrels)
DAEMEC
52,945
5,437
33,193
120,754
NBA
56,393
5,555
41,713
147,199
It is not possible to determine the source of the discrepancies between the NBA and
DAEMEC data on an industry-by-industry basis. The NEA data include fuels produced and
consumed for heat and power at the same establishment in the petroleum refining, coke ovens,
and blast furnace and steel mills industries (see Price and Wendling, 1986, p. 4). For 1970 to
1984, the U.S. EPA adjusts the consumption of some fuels before it estimates fuel combustion
emissions (see U.S. EPA, 1998a, pp. 3-35, 3-38, 3-39, 3-43 and U.S. EPA, 2000, pp. 21-23).
The adjustments by the U.S. EPA include subtracting residual fuel oil, distillate fuel oil, and
natural gas consumed by the petroleum refining industry, and residual fuel oil and natural gas
consumed by the blast furnace industry. Although Price and Wendling do not supply information
regarding which fuel consumption estimates in the NEA are adjusted, it is likely that most -
maybe all - of the NEA adjusted fuel consumption quantities are subtracted before allocating fuel
combustion emissions (see Section 3.3.2. for a discussion of MECS data on energy produced and
consumed at the same establishment). As a result, the procedure used to assign fuel combustion
emissions is unaffected by the NEA adjustments to fuel consumption in the petroleum refining,
coke ovens, and blast furnace and steel mills industries.
3.2. Fuel Consumption of the Nonmanufacturing Portion of the Industrial Sector
This section discusses the estimates of energy consumption by the agriculture,
construction, and mining sectors.23 The NEA data are used to assign emissions from industrial
fuel combustion for 1970 through 1984. For 1985 to 1997, it is necessary to derive estimates of
fuel consumption for heat and power by these nonmanufacturing sectors. The 1991
Manufacturing Consumption of Energy (U.S. DOE, 1994, pp. 467-478) discusses an attempt to
reconcile estimates of the industrial sector's energy consumption reported in the State Energy
Data Report (SEDR) with the energy consumption estimates reported in the MECS. The data in
the SEDR is considered to be conceptually similar to the "Total Primary Consumption of Energy
for All Purposes" table in the 1985, 1988, and 1991 MECS. This table, which includes energy
consumed for purposes other than heat and power, was renamed "Total First Use of Energy for
23 Energy consumption by the mining sector consists of energy consumption by the following
four industries: (1) metal mining, (2) coal mining, (3) oil and gas extraction, and (4) nonmetallic
minerals, except fuels.
January 2001
-------
32 Assigning Emissions to Industries in the United States, Revised Estimates
All Purposes" in the 1994 MECS. The basis of the derivation of estimates of fuel consumption
by the nonmanufacturing sectors for 1985 to 1997 is the 1991 MECS and Price and Wendling
(1986).
Several publications provide the information required to estimate fuel consumption by
the nonmanufacturing sectors for 1985 to 1997. Fuel Oil and Kerosene Sales (U.S. DOE, various
years) provides data on the annual consumption of distillate fuel oil by the agriculture sector.
Since the Census of Agriculture (natural gas) (U.S. Department of Commerce, 1994), the Census
of Mineral Industries (natural gas, residual fuel oil, distillate fuel oil, and coal) (U.S. Department
of Commerce, 1990b and 1996b), and the Census of Construction (natural gas) (U.S. Department
of Commerce, 1990a and 1996a) are only available for 1987 and 1992, it is necessary to estimate
fuel consumption by these sectors for the remaining years. The two-way indexing method
estimates fuel consumption in 1985, 1986, and 1988 to 1991. The forward indexing method
estimates fuel consumption for 1993 to 1997.24
The following discussion of the estimates of fuel consumption by the nonmanufacturing
sectors, highlights the discrepancies between the fuel consumption estimates derived in the NEA
and the procedure used by this report to estimate fuel consumption for 1985 to 1997. In addition
to this discontinuity in the data, the NEA assigned the fuel consumption of selected SIC codes of
industries in the Census of Mineral Industries to the construction sector. Due to these
inconsistencies, although estimates of fuel consumption by the agriculture, mining, and
construction sectors are in SIC7097.WK3, these estimates are not reported in Appendix E.
3.2.1. Wood. Coke and Breeze, and LPG
According to the NEA, the nonmanufacturing sectors purchased neither (1) wood (NEA
energy product code 180) nor (2) coke and breeze (NEA energy product code 291) for energy
between 1958 and 1985. The lack of consistent historical data results in the SEDR excluding
consumption of the "... pulping liquor; wood chips, bark, wood waste; net steam and hot water;
and miscellaneous energy sources" (U.S. DOE, 1994, p. 477) category, which is included in the
MECS. The 1991 MECS (U.S. DOE, 1994, p. 477) also concludes that the State Energy Data
Report (SEDR) and MECS data for petroleum coke consumption were "... quiet close and judged
to be comparable." As a result, it is assumed that the nonmanufacturing sectors consume neither
wood nor coke and breeze for heat and power from 1986 to 1997.
According to the!991 MECS (U.S. DOE, 1994, p. 476), the difference between the
SEDR and MECS data for purchases of liquified petroleum gas (LPG) can be assigned to the
nonmanufacturing sectors. These purchases of LPG include energy consumed for purposes other
than heat and power. The NEA also assigned LPG consumption for purposes other than heat and
power to the nonmanufacturing sectors.
24 Appendix D describes the procedures used to estimate the missing observations of fuel
consumption for heat and fuel by the nonmanufacturing sectors.
January 2001
-------
Assigning Emissions to Industries in the United States, Revised Estimates 33
3.2.2. Distillate fuel oil
The SEDR reports the following values (in millions of barrels) for consumption of
distillate fuel oil by the industrial sector: 225 in 1982, 204 in 1985, 211 in 1987, 210 in 1988,
196 in 1991, and 196 in 1992 (U.S. DOE, 1996, p. 24). It is useful to remember that the SEDR
data includes energy consumed for purposes other than heat and power. According to the 1991
MECS (U.S. DOE, 1994, p. 476), the difference between the SEDR and MECS estimates of
distillate fuel oil purchases can be assigned to the nonmanufacturing sectors.
According to the MECS, consumption (in 1,000 barrels) of distillate fuel oil for all
purposes by the manufacturing sector was 35,739 in 1985, 37,865 in 1988, and 25,016 in 1991.
The Census of Mineral Industries (U.S. Department of Commerce, 1985, 1990b, 1996b) reports
(in 1,000 barrels) the following values for distillate fuel oil consumption for heat and power by
the mineral industries of 35,118.9 in 1982, 28,943.8 in 1987, and 27,244.5 in 1992. According to
the 1991 MECS (U.S. DOE, 1994, p. 476), distillate fuel oil consumed by the agricultural and
construction sectors account for the discrepancy between the SEDR and MECS estimates.
However, consumption of diesel fuel, a type of distillate fuel oil, for off-highway motor vehicle
use by the agricultural and construction sectors complicates the process of estimating distillate
fuel oil consumption for heat and power by these two sectors (U.S. DOE, Fuel Oil and Kerosene
Sales, 1997, pp. 52-53 ).
In 1985, four industries in the agricultural sector consumed distillate fuel oil for heat and
power: NEA 01000 (803.53 thousand barrels), NEA 02000 (5,107.5 thousand barrels), NEA
03000 (0.04 thousand barrels), and NEA 04000 (186.21 thousand barrels). Fuel Oil and
Kerosene Sales was the source of the NEA estimates of distillate fuel oil consumption by the
agricultural sector (Price and Wendling, 1986). In Fuel Oil and Kerosene Sales, distillate fuel oil
sales to farms are classified as either (1) diesel or (2) "other distillate." This report assumes that
the "other distillate" category measures distillate fuel oil consumption for heat and power by
farms (NEA 01000 and NEA 02000). Fuel Oil and Kerosene Sales 1985 (U.S. DOE, 1986)
reported "other distillate" sales to farms of 148,086 gallons (3,525.86 barrels). The "adjusted"
sales (i.e., deliveries) to farms of "other distillate" were 149,971 gallons (3,570.74 barrels). In
this report, the adjusted value of "other distillate" is the estimate of distillate fuel oil consumed
by farms. As a result, this report's estimates of distillate fuel oil consumption by agriculture
(NEA 01000 and NEA 02000) for 1985 to 1997 are less than the NEA values. Since it is not
possible to determine the source of the NEA estimates of distillate fuel oil consumption by NEA
03000 and NEA 04000, there are no estimates for these industries for 1985 to 1997.
From 1985 through 1988, the results of the "Fuel Oil and Kerosene" survey appeared as
an appendix in two U.S. DOE publications. Distillate oil fuel consumption by farms for 1985
through 1987 appears m Petroleum Marketing Monthly (U.S. DOE, 1986, 1987, and 1988b).
Distillate fuel oil consumption by farms in 1988 is in Petroleum Marketing Annual (U.S. DOE,
1989). For 1989 through 1997, Fuel Oil and Kerosene Sales (U.S. DOE, various issues) is the
source of "other distillate" fuel oil sales to farms.
January 2001
-------
34 Assigning Emissions to Industries in the United States, Revised Estimates
According to the NEA, 42,850.3 thousand barrels of distillate fuel oil were consumed
for heat and power by the mining (26,514.3 thousand barrels) and construction (16,336 thousand
barrels) sectors in 1982. According to the Census of Mineral Industries (U.S. Department of
Commerce, 1985), the mineral industries purchased 35,118.9 thousand barrels of distillate fuel
oil for heat and power in 1982. The NEA assigns the energy consumption of some industries
(i.e., pt. SIC 11, pt. SIC 12, SIC 138, and pt. SIC 148), which is reported in the Census of
Mineral Industries, to the construction sector (NEA 11001, NEA 11002, and NEA 12000).
It is not possible to determine the source of the discrepancy (7,721.4 thousand barrels)
between the NEA and Census of Mineral Industries estimates of residual fuel oil consumption in
1982. Relative to the NEA, this report underestimates residual fuel oil consumption by the
nonmanufacturing sectors from 1985 to 1997 ,25 The relative importance of this discrepancy can
also be illustrated by recalling the amounts of distillate fuel oil consumed by the manufacturing
sector. According to the U.S. DOE (1992, Errata, p. 29), 28,859 thousand barrels were
consumed by manufacturing industries in 1982. However, the NEA estimates that 39,396
thousand barrels were consumed by the manufacturing sector in 1982.
3.2.3. Coal
The EIA's Quarterly Coal Report is the source of the SEDR estimates of coal
consumption. In attempting to reconcile the discrepancy between the SEDR and MECS
estimates of coal consumption, the 1991 MECS (U.S. DOE, 1994, pp. 470-472) estimates
nonmanufacturing coal consumption by subtracting the SEDR estimate of coal consumption by
manufacturing plants from its estimate of coal consumption by "other industrial" plants.26
According to Price and Wendling (1986, pp. 3-4), the primary source of information used
by the NEA to estimate the coal consumption of the mineral industries was the Census of
Mineral Industries21 Aside from the mineral industries, coal was consumed coal for heat and
power by one other nonmanufacturing industry - livestock and livestock products (NEA 01000).
25 For 1985 to 1997, distillate fuel oil consumption by the mining and construction sectors are
estimated by using data from the 1982, 1987 and 1992 Census of Mineral Industries (U.S.
Department of Commerce, 1985, 1990b, and 1996b), and two-way indexing, backward indexing,
and forward indexing methods. The backward indexing is used with estimates from the 7957
Census of Mineral Industries to estimate distillate fuel oil consumption for those industries
whose consumption is not disclosed in the 1982 Census of Mineral Industries.
26 "other industrial" corresponds to "industrial" establishments minus coke plants
27 According to the 1982 Census of Mineral Industries (U.S. Department of Commerce,
1985), the mining sector purchased 3,046.2 thousand short tons of coal for heat and power. The
NEA reported the consumption of 3,110.74 thousand short tons of coal for heat and power by the
mining sector. It is not possible to determine the source of this discrepancy.
January 2001
-------
Assigning Emissions to Industries in the United States, Revised Estimates 35
Livestock and livestock products (NEA 01000) consumed 41.48 thousand short tons of coal in
1982. Price and Wendling (1986) do not cite any source for the estimate of coal consumption by
NEA 01000. In addition, no estimates of coal purchases by NEA 01000 for 1986 to 1997 could
be located. Comparing the estimates of coal purchases by the nonmanufacturing sectors for 1987
and 1992 that are derived from the Quarterly Coal Report with the values reported in the 1987
and 1992 Census of Mineral Industries (U.S. Department of Commerce, 1990b and 1996b)
provides a check on the equivalence of the SEDR and NEA data.
According to the Quarterly Coal Report., coal purchases by the nonmanufacturing sectors
totaled 6,108 thousand short tons in 1987 (Oct-Dec. 1991, p. 39, and Oct-Dec. 1988, p. 46) and
8,877 thousand short tons in 1992 (Oct-Dec. 1995, p. 79, and Oct-Dec. 1993, p. 80 ). The
Census of Mineral Industries (U.S. Department of Commerce, 1990b and 1996b) reported coal
purchases (1000 short tons) of 3,098.7 in 1987 and 2,847.8 in 1992. According to the Census of
Mineral Industries, the following quantities (1000 short tons) were produced and consumed at
the same establishment for heat and power: 276.7 in 1987 and 301.8 in 1992. Hence, the
Quarterly Coal Report assigns more coal consumption for heat or power to the
nonmanufacturing sectors than the estimates derived using the NEA methodology.
The magnitude of this discrepancy can be appreciated by remembering that DAEMEC
estimates 48,004 thousand short tons of coal were consumed by the manufacturing sector in 1982
(see U.S. DOE, 1992, Errata, p. 35) and the NEA estimates the manufacturing sector consumed
52,196 thousand short tons of coal. In this report coal consumption by the mining and
construction sectors for 1985 to 1997, are estimated using data from the 1982, 1987 and 1992
Census of Mineral Industries, and the two-way indexing and forward indexing methods.
3.2.4. Residual fuel oil
The 1991 MECS reported more consumption of residual fuel oil for heat and power by
the manufacturing sector than the SEDR reported for the entire industrial sector (U.S. DOE,
1994, p. 475). There is no explanation for this inconsistency between the MECS and SEDR
estimates of purchases of residual fuel oil by the nonmanufacturing sectors. As a result, this
report utilizes the same sources of information as the NEA.
The NEA reports no consumption of residual fuel oil by NEA 01000 and NEA 02000.
Fuel Oil and Kerosene Sales (U.S. DOE, various years) provides some insights into the quantity
of residual oil consumed by farms (NEA 01000 and NEA 02000). In 1994, the residual fuel oil
consumed by farms is combined with the following consumers of residual fuel oil: (1) residential,
(2) railroad, (3) on-highway diesel use, (4) off-highway use, and (5) all other uses. These six
end-use categories combine for 5.9 percent of residual fuel oil sales to industrial and oil company
end-users in 1986 and 0.2 percent of residual fuel oil sales to industrial and oil company end-
January 2001
-------
36 Assigning Emissions to Industries in the United States, Revised Estimates
users in 1994.28 In 1989, there was a dramatic decline in sales to these six end-use categories.
Since Fuel Oil and Kerosene Sales indicates relatively little residual fuel oil is consumed by
farms, this study assigns no residual fuel oil to NBA 01000 and NEA 02000.
In addition to the mining sector, the NEA reported the consumption of residual fuel oil
for heat and power to the following NEA 03000, NEA 11001, NEA 11002, and NEA 12000. In
1982, NEA 03000 consumed 0.05 thousand barrels, NEA 11001 consumed 1,797.31 thousand
barrels, NEA 11002 consumed 2623.88 thousand barrels, and NEA 12000 consumed 537.43
thousand barrels. The NEA assigned 12,615.4 thousand barrels to the mining (7,656.8 thousand
barrels) and construction (4,958.6 thousand barrels) sectors in 1982. According to the 1982
Census of Mineral Industries (U.S. Department of Commerce, 1985), the mineral industries
purchased 9,241.7 thousand barrels in 1982. The Census of Mineral Industries collects data for
industries (i.e., pt. SIC 11, pt. SIC 12, SIC 138, and pt. SIC 148) assigned to the construction
sector (NEA 11001, NEA 11002 and NEA 12000) by the NEA. It is not possible to determine the
source of the difference (3,373.7 thousand barrels) between the NEA estimate of residual fuel oil
consumed by the mining and construction sectors and the Census of Mineral Industries estimate
for 1982.
In this report, the Census of Mineral Industries provides the estimates of residual fuel oil
consumed by the mining and construction sectors. As a result, this report underestimates residual
fuel oil consumed for heat and power by the nonmanufacturing sectors for 1985 to 1997 relative
to the method employed by the NEA. The magnitude of this discrepancy can be appreciated by
remembering that the DAEMEC reports 92,667 thousand barrels of residual fuel oil were
consumed by manufacturing industries in 1982 (see U.S. DOE, 1992, Errata, p. 31). According
to the NEA, 127,728 thousand barrels of residual fuel oil were purchased by the manufacturing
sector in 1982. For 1985 to 1997, residual fuel oil consumption by the mining and construction
sectors is estimated by using data from the 1987 and 1992 Census of Mineral Industries (U.S.
Department of Commerce, 1990b and 1996b) and the two-way indexing and forward indexing
methods for non-census years. Since data on residual fuel oil consumption are not disclosed in
the 1982 Census of Mineral Industries., data from the 7957 Census of Mineral Industries and the
backward indexing method is used to estimate residual fuel oil consumption in 1985 and 1986.
3.2.5. Natural gas
In attempting to reconcile the SEDR and MECS data on natural gas consumption, the
1991 MECS refers to the Census of Mineral Industries. The Census of Mineral Industries was
also used by the NEA to estimate the natural gas consumption of the mining sector. As a result,
for 1985 to 1997, natural gas consumption by the mineral industries is estimated using data from
the 1982, 1987 and 1992 Census of Mineral Industries, and the two-way indexing and forward
indexing methods.
28 For this survey, the industrial sector consists of mines, smelters, and manufacturing
(excluding oil company use) (see U.S. DOE, Fuel Oil and Kerosene Sales 7997, p. 52).
January 2001
-------
Assigning Emissions to Industries in the United States, Revised Estimates 37
The SEDR estimates of natural gas consumption exclude the agriculture and construction
sectors (see U.S. DOE, 1994, p. 473). According to the NEA, heat and power is the sole use of
natural gas consumed by the agriculture (NEA 01000 and NEA 02000) and construction (NEA
11001, NEA 11002, and NEA 12000) sectors. When estimating natural gas consumption by
these two sectors, this study adopts the NEA methodology. The 1992 Census of Agriculture
(U.S. Department of Commerce, 1994, p. 29) provides estimates of natural gas purchases (in
dollars) by SIC 01 and SIC 02 (i.e., NEA 01000 and NEA 02000) for 1982, 1987, and 1992. The
1982, 1987, and 1992 Census of Construction (U.S. Department of Commerce, 1984, 1990a,
1996a) provide estimates of natural gas purchases (in dollars) by the construction sector. The
1992 Census of Construction reports a combined value for purchases of natural and
manufactured gas, which serves as the proxy for natural gas consumption by the construction
sector. For the non-census years from 1985 to 1997, natural gas consumption by the agriculture
and construction sectors is estimated using the two-way indexing and forward indexing methods.
Since only monetary values of natural gas purchases by the agriculture and construction
sectors are available, an average price for natural gas is used to estimate the quantity of natural
gas purchased by these two sectors in 1987 and 1992. Historical Natural Gas Annual 1930
Through 1997 (U.S. DOE, 1998b) is the source of the average price of natural gas delivered to
U.S. industrial customers, which is used for both the agriculture and construction sectors.29 Prior
to 1996 (see Historical Natural Gas Annual 1930 Through 1997, U.S. DOE, 1998b, p. 12),
natural gas consumption by the agriculture sector was classified as a commercial use. However,
the price of natural gas for industrial consumers, which is less than the price of natural gas paid
by commercial consumers, is used because it yields an estimate of the quantity of natural gas
consumed that is closer to the NEA estimate for the agriculture and construction sectors in 1982.
According to the NEA, 352.93 billion cubic feet of natural gas were consumed by the
mining sector and 75.49 billion cubic feet were consumed by the construction sector.30
According to the 1982 Census of Mineral Industries (U.S. Department of Commerce, 1985),
375.7 billion cubic feet of natural gas were purchased by the mineral industries. The estimate of
natural gas purchases by the construction sector reported in the 1982 Census of Construction
(U.S. Department of Commerce, 1984) and the prices reported in Historical Natural Gas Annual
1930 Through 1997 (U.S. DOE 1998b) are combined to derive an estimate of 69.75 billion cubic
feet of natural gas purchases by the construction sector.
According to the NEA, farms (NEA 01000 and NEA 02000) consumed 94.36 billion
cubic feet of natural gas in 1982 (7.93 billion cubic feet by NEA 01000, and 86.43 billion cubic
feet by NEA 02000). The estimate of natural gas purchases by the agriculture sector in the 1982
29 The NEA converted the cost of purchased natural gas using prices from Gas Facts (Price
and Wendling, p. 4)
30 The NEA excluded the 663.2 billion cubic feet of natural gas produced and consumed for
heat and power by the same establishment in the minerals industries in 1982.
January 2001
-------
38 Assigning Emissions to Industries in the United States, Revised Estimates
Census of Agriculture is combined with the prices in Historical Natural Gas Annual 1930
Through 1997 to derive an estimate of 86.98 billion cubic feet of natural gas purchases by farms.
It is also possible to compare the NEA and DAEMEC estimates of natural gas consumed
by the manufacturing sector in 1982. According to the DAEMEC, 4,577 billion cubic feet of
natural gas were consumed by manufacturing industries (U.S. DOE, 1992, Errata, p. 33). The
NEA estimates the manufacturing sector consumed 4,952 billion cubic feet of natural gas.
3.3. Energy Produced and Consumed at the Same Establishment
In addition to the consumption of energy purchased from off-site sources, there are
instances when an establishment produces and consumes energy for heat and power. Since
emissions are generated by the combustion of the fuels produced and consumed by an
establishment, this consumption is included in the fuel consumption estimates used to assign the
emissions. The remainder of Section 3.3. discusses the sources of the estimates of energy
produced and consumed at the same establishment.
3.3.1. Nonmanufacturing
In the nonmanufacturing sectors, coal is produced and consumed at the same
establishment in NEA 07010 and NEA 07020, and natural gas is produced and consumed at the
same establishment in NEA 08002 and NEA 08003. According to the 7957 Census of Minerals
Industries (U.S. Department of Commerce, 1990), 39.4 percent of the total energy consumed (in
Btu) for heat and power by the mineral industries is from fuels produced and consumed at the
same establishment.31
In 1982, a small quantity of natural gas was produced and consumed at the same
establishment in SIC 138. There was no natural gas produced and consumed at the same
establishment in SIC 138 in 1987 or 1992. Since the NEA did not include this consumption in its
"produced and consumed" category for either NEA 11001 or NEA 12000, this fuel consumption
is excluded from the estimates used to assign fuel combustion emissions.
Estimates of energy produced and consumed at the same establishment by the mining
sector are available in the 1982, 1987, and 1992 Census of Minerals Industries (U.S. Department
of Commerce, 1985, 1990b, and 1996b). Fuel produced and consumed at the same establishment
in the mining sector for 1985, 1986 and 1988 to 1991 is estimated by the two-way indexing
method, and estimates for 1993 to 1997 are generated using the forward-indexing method.
31 The estimate in the NEA for the consumption and production of natural gas by the
establishments in NEA 08002 in 1982 is different than the value in the 1982 Census of Mineral
Industries (U.S. Department of Commerce, 1985). As a result, the NEA estimates for natural gas
produced and consumed at the same establishment in NEA 08002 for 1978 to 1985 are re-
estimated.
January 2001
-------
Assigning Emissions to Industries in the United States, Revised Estimates 39
3.3.2. Manufacturing: Natural gas. Residual fuel oil, and distillate fuel oil
NEA estimates of fuel consumed for heat and power include fuel produced and consumed
at the same establishment in the petroleum refining, coke ovens and blast furnaces, and steel
mills industries (see Price and Wendling 1986, p. 4). However, the NEA reports no separate
estimates of these quantities.
DAEMEC data, which are used to distribute fuel combustion emissions, consist solely of
off-site produced energy. However, the differences between the "Total Inputs of Energy for Heat,
Power, and Electricity" and "Total Consumption of Offsite-Produced Energy for Heat, Power,
and Electricity" tables in the MECS are assumed to represent fuels produced and consumed at the
same establishment.
Combining the MECS estimate of distillate fuel oil and residual fuel oil produced and
consumed at the same petroleum refining (SIC 29) establishment with the DAEMEC estimate of
distillate fuel oil and residual fuel oil purchased by SIC 29 generates an estimate of distillate and
residual fuel oil consumption that is consistent with the NEA data. Fuel produced and consumed
at the same establishment in SIC 29 for 1986, 1987, 1989, 1990, 1992, and 1993 is estimated
using the two-way indexing method. Estimates of fuel consumption for 1993 to 1997 are derived
by the forward-indexing method. Since Price and Wendling (1986, p. 4) did not adjust for energy
produced and consumed at the same establishment in other industries, fuel produced and
consumed at the same establishment in other industries is not used to assign fuel combustion
emissions.32 The remainder of this section discusses evidence from the MECS that fuels may be
produced and consumed at the same plant in other industries.
In addition to SIC 29, residual fuel oil is produced and consumed at the same
establishment in SIC 28 during 1985, 1988, and 1991. In 1988, a small amount of residual fuel
oil is produced and consumed at the same establishment in SIC 33. During 1985, 1988, and
1991, establishments in SIC 29 account for at least 91 percent of the residual fuel oil produced
and consumed at the same establishment. In 1994, SIC 29 account for all residual fuel oil
produced and consumed at the same establishment. Since Price and Wendling (1986, p. 4) did
not adjust for energy produced and consumed at the same establishment in SIC 28 and SIC 33,
residual fuel oil produced and consumed at the same establishment in these industries is not used
when assigning fuel combustion emissions.
In addition to SIC 29, distillate fuel oil is produced and consumed at the same
establishment in SIC 28 during 1985, 1988 and 1991. In 1994, there is a small amount of residual
32 According to Price and Wendling (1986, p. 4), fuel produced and consumed at the same
establishment in SIC 33 are included in the NEA data. Since the estimates in MECS of the
quantities of fuels produced and consumed at the same establishment in SIC 33 are relatively
small, their exclusion from the estimates of fuel consumption does not appear to have a
substantial impact on the assignment of emissions for 1985 to 1997.
January 2001
-------
40
Assigning Emissions to Industries in the United States, Revised Estimates
fuel oil being produced and consumed at the establishment in SIC 37. In 1985, 1988, 1991 and
1994, establishments in SIC 29 account for a minimum of 85 percent of the distillate fuel oil
produced and consumed at the same establishment. Since Price and Wendling (1986, p. 4) did
not adjust for energy produced and consumed at the same establishment in SIC 28 and SIC 37,
distillate fuel oil produced and consumed at the same establishment in these industries is not used
to assign fuel combustion emissions.
In 1985 and 1988, coal was produced and consumed at the same establishment in SIC 28
and SIC 32.33 In 1985 and 1988, SIC 28 accounts for at least 80 percent of coal produced and
consumed at the same establishment. In 1991, SIC 28 and SIC 26 both account for
approximately one-half of the coal produced and consumed at the same establishment. This
consumption is not used when assigning fuel combustion emissions. In 1994, no coal is
produced and consumed at the same establishment in the manufacturing sector.
In 1985 and 1988, natural gas is produced and consumed at the same establishment in
SIC 28. In 1991, natural gas is also produced and consumed at the same establishment in SIC 28,
SIC 35, and SIC 37. Since Price and Wendling (1986, p. 4) did not adjust for energy produced
and consumed at the same establishment in these industries, natural gas produced and consumed
at the same establishment in SIC 28, SIC 35 and SIC 37 is not used to assign fuel combustion
emissions.
The following summarizes the percentage of different fuels consumed by the
manufacturing sector, but not assigned to an industry:
Residual Fuel Oil
Distillate Fuel Oil
Coal
Natural Gas
1985
0.28
0.45
0.08
1.02
1988
0.10
0.24
0.08
0.14
1991
0.22
0.35
0.72
0.13
1994
0.00
0.15
0.00
0.39
These percentages are derived by adding the quantity of off-site produced energy for all
manufacturing industries and the amount of energy produced and consumed by the same
establishment in SIC 29 and dividing the total by the "Total Inputs of Energy for Heat, Power..."
for all manufacturing industries. Hence, the table shows the relative importance of the
unassigned amounts of energy produced and consumed by the same establishment.
33 Coal produced and consumed by SIC 28 may reflect coke oven activities that are typically
assigned to SIC 33.
January 2001
-------
Assigning Emissions to Industries in the United States, Revised Estimates 41
3.3.3. Manufacturing: Wood
In addition to pulping (or black) liquor, the "Total Inputs of Selected Wood and Wood-
Related Products for Heat, Power, and Electricity Generation" table lists four "biomass" fuels
consumed for heat and power (U.S. DOE 1997, pp. 405-406). Excluding agricultural waste, the
remaining three fuels are related to the consumption of wood: (1) wood harvested directly from
trees, (2) wood residues and byproducts from mill processing, and (3) wood-related and paper-
related refuse. The (1) wood harvested directly from trees and (2) wood residues and byproducts
from mill processing constitute the "wood chips, bark" category in the "Total Inputs of Selected
Byproduct Energy for Heat, Power, and Electricity Generation" table in the 1988 MECS (U.S.
DOE 1991, p. 47), the 1991 MECS (U.S. DOE 1994, p. 85), and the 1994 MECS (U.S. DOE
1997, p. 82). The data in these tables include purchased energy, and energy produced and
consumed at the same establishment.
The NEA provides information on wood consumption for heat and power for 1970 to
1984. The MECS "wood chips, bark" category is the source of estimates of wood consumption
in 1988, 1991, and 1994. Estimates of the combustion of wood for 1985-87, 1989-90, 1992-93,
and 1995-97 are derived by combining the MECS data with backward-indexing, two-way
indexing, and forward-indexing methods.
The NEA assigns all consumption of wood for heat and power to NEA 20990 (SIC 24)
and NEA 24990 (SIC 26). The 1988 MECS assigns almost 95 percent of wood consumption to
SIC 24 and SIC 26. However, the 1985 NEA assigns 24 percent of wood consumption to SIC 24,
while the 1988 MECS assigns 41 percent. The 1985 NEA assigns 76 percent of wood
consumption to SIC 26, while the 1988 MECS assigns 54 percent. This results in a discrepancy
between the 1970-1984 and 1985-1997 estimates when assigning emissions from wood
combustion.
3.3.4. Manufacturing: Blast furnace / coke oven gases and still gas
According to the NEA, still gas (NEA energy product code 222) is consumed by the
petroleum refining industry (NEA 31011, which is part of SIC 29) and coke oven gas (NEA
energy product 223) is consumed by the coke oven industry (NEA 37011, which is part of SIC
33).34 The 1988 MECS (U.S. DOE, 1991, p. 47), the 1991 MECS (U.S. DOE, 1994, p. 85), and
the 1994 MECS (U.S. DOE, 1997, p. 82) provides data ("Total Inputs of Selected Byproduct
Energy for Heat, Power, and Electricity Generation" table) on the consumption of blast furnace /
coke oven gases and waste gas. According to the 1994 MECS, waste gas is produced and
consumed primarily by the following two industries: (1) chemicals (SIC 28) - 15 percent and (2)
34 The terms "waste gas" and "still gas" refer to the same fuel.
January 2001
-------
42 Assigning Emissions to Industries in the United States, Revised Estimates
petroleum refining (SIC 29) - 85 percent (U.S. DOE, 1997, p. 82).35 The MECS "Total Inputs"
table also shows that small amounts of waste gas were produced and consumed at the same
establishment by SIC 20, SIC 24, and SIC 32. The primary metals industry (SIC 33) is the sole
producer and consumer of blast furnace / coke oven gases (U.S. DOE, 1997, p. 82).
MECS data, which are reported in Btu, for the consumption of blast furnace / coke oven
gases and waste gas are converted to physical units using the NEA thermal conversion factors for
coke oven gas and still gas, respectively. The backward indexing method is used to estimate
consumption of these fuels from 1985 to 1987. The two-way indexing method is utilized when
estimating consumption in 1989, 1990, 1992, and 1993. The forward indexing method is used to
estimate consumption of these fuels for 1995 to 1997.
3.4. Physical units vs. Btu
Since emission factors are stated in terms of emissions per physical units of fuel
consumption, when assigning fuel combustion emissions, physical units of fuel consumption are
preferred to fuel consumption in Btu. The NEA provides estimates of fuel consumption in
physical units, and these values are used to allocate fuel combustion emissions for 1970 to 1984.
In a given year, the NEA uses the same conversion factor for a fuel across all industries. As a
result, physical units and Btu generate the same assignment of emissions.
The DAEMEC publishes estimates in Btu for 1974 to 1994; however, only the 1974 to
1988 data are reported in physical units. As a result for 1985 to 1997, energy consumption data in
Btu are used to assign fuel combustion emissions. For a given year, the DAEMEC employs the
same thermal conversion factor for a fuel.36 The U.S. DOE conversion factors for natural gas,
residual fuel oil, and distillate fuel oil (U.S. DOE, 1999) are used to convert energy consumption
estimates for the agriculture and construction sectors from physical units to Btu. In addition, the
estimates of fuel consumption (in physical units) reported in the Census of Mineral Industries are
converted to Btu with the conversion factors reported in the Census of Mineral Industries.
Identical thermal conversion factors for distillate fuel oil and residual fuel oil are
specified by the Census of Mineral Industries (U.S. Department of Commerce, 1985, 1990b, and
35 Smaller industries may also produce a waste gas that is consumed for other space heating or
other combustion related use (electricity generation on-site). For example, landfill operations
often produce methane which is a "process gas" and it is sometimes consumed on-site and is
sometimes sold to electric generators for use on the power grid. (10/22/99 e-mail correspondence
with Sharon Nizich)
36 According to the U.S. DOE (1992, p. 94), coal consumed for fuel and coal consumed for
coking have different heat rates (see U.S. DOE 1988a, p. 61; U.S. DOE 1991, p. 148; U.S. DOE
1994, p. 455; and U.S. DOE 1997, p. 431).
January 2001
-------
Assigning Emissions to Industries in the United States, Revised Estimates
43
1996b) and the U.S. DOE.37 There are minor discrepancies between these sources for the thermal
conversion factors for natural gas. Hence, using either physical quantities or Btu has little or no
effect on the allocation of fuel combustion emissions from the consumption of residual fuel oil,
distillate fuel oil, and natural gas. However, the thermal conversion factors reported by the NEA
and the Census of Mineral Industries (U.S. Department of Commerce, 1981 and 1985) in 1977
and 1982 are higher than those reported by the U.S. DOE (1999). This discrepancy in thermal
conversion factors results in a slight increase in the share of coal combustion emissions assigned
to mineral industries for 1984 to 1986.
3.5. Concordance Between Fuel Consumption Categories and Emissions Categories
Table 5 reports which pollutants (marked with an "X") are generated by the "Fuel
Combustion - Industrial" (Tier I) category (U.S. EPA, 1998b).38 Before allocating fuel
combustion emissions among industries, a concordance is established between the fuel
consumption categories listed by the fuel consumption surveys (e.g., NEA and MECS) and fuel
categories listed in the 1970-1997NAPET(U.S. EPA, 1998b).
Coal (bituminous, subbituminous, anthracite & lignite) emissions are aggregated and
assigned to industries based on the quantity of coal consumed by each industry. If this
assumption is a good approximation of reality, then the distribution of coal consumption by
industries is an adequate proxy for the distribution of emissions from the consumption of each
category of coal (e.g., bituminous).
Emissions from the combustion of residual fuel oil and distillate fuel oil are assigned to
industries based on their consumption of residual and distillate fuel oil. However, CO and VOC
emissions are reported for an aggregate fuel oil source category. The 1970-1990 Emissions
spreadsheet (SIC7090.WK3) reports no control efficiencies for CO or VOC emissions from
either residual or distillate fuel oil combustion. In addition, it reports the following constant
emission factors (in millions of pounds of emissions per million gallons of fuel consumed):
VOC
CO
Residual fuel oil
0.3
4.5
Distillate fuel oil
1
7
37 Conversion factors for coal and natural gas consumed by mineral industries in non-census
years are estimated by simple linear interpolation of conversion factors from census years.
38 The "Fuel Combustion - Industrial " source category reports emissions from fuel
combustion by industries that belong to the industrial sector.
January 2001
-------
44
Assigning Emissions to Industries in the United States, Revised Estimates
Table 5
Pollutants Emitted by the "Fuel Combustion - Industrial" Source Category
Coal
bituminous
subbituminous
anthracite & lignite
other
Oil
residual
distillate
other
Gas
natural
process
other
Other
•wood/bark -waste
liquid waste
other
Internal Combustion
PM-10
X
X
X
X
X (NA)
X
X
X
X
X
X
X
X(NA)
X
X
X(NA)
X
X(NA)
SO2
X
X
X
X
X(NA)
X
X
X
X
X
X
X(NA)
NO,
X
X
X
X
X(NA)
X
X
X
X
X
X
X
X
X
X
X(NA)
X
X (NA)
VOCs
X
X
X
X
X(NA)
CO
X
X
X
X
X(NA)
Pb
X
X
X
X
X
X
X
NOTE: An "NA" indicates that estimates of emissions are not reported until 1985. No PM-10
emissions are reported for any fuel combustion category for 1971 to 1974.
January 2001
-------
Assigning Emissions to Industries in the United States, Revised Estimates 45
Since the SIC7090.WK3 spreadsheet reports constant emission factors for these fuels and
pollutants for 1970 to 1990, it is assumed that these emission factors are valid for 1991 to 1997.
The consumption of residual fuel oil is converted into an equivalent amount of distillate fuel oil.
The conversion is based on the emission factors for VOC and CO emissions. For example, for a
gallon of residual fuel oil generates the same quantity of VOC emissions as 0.3 gallons of
distillate fuel oil. VOC emissions from fuel oil combustion are then distributed among industries
based on the consumption of distillate fuel oil and the distillate equivalent of the residual fuel oil.
The same procedure is followed when distributing the CO emissions from the combustion of
residual fuel oil and distillate fuel oil.
The petroleum refining industry (SIC 29) (see U.S. EPA, 1998a, pp. 3-38 and 3-39) is
assigned the PM-10, SO2, and NOX emissions from the "Oil - Other" subcategory.
Natural gas is assumed to be the principal source of the VOC, SO2, and CO emissions
from gas combustion. Hence, these emissions are assigned to industries based on their
consumption of natural gas. PM-10 and NOX emissions from the combustion of natural gas are
assigned to industries based on the share of natural gas consumption by each industry.
PM-10 and NOX emissions from the "Gas-Other" subcategory are assigned to the
petroleum refining industry (SIC 29) (see U.S. EPA, 1998a, p. 3-43).
In addition to emissions from natural gas combustion, PM-10 and NOX emissions are
generated by the consumption of processed gas.39 For 1970 to 1984, the EPA estimates of
emissions from gas combustion include emissions from the combustion of coke-oven gas by the
iron and steel industry and emissions from process heaters in the petroleum refining industry (see
U.S. EPA, 1998a, pp. 3-42 and 3-43). Emission estimates from the "gas - processed" category for
1985 to 1997 includes the following consumers of processed gas: (1) chemical manufacturing,
(2) primary metal production, (3) secondary metal production, (4) petroleum industry, (5) oil and
gas production, and (6) miscellaneous manufacturing industries (U.S. EPA 1998a, pp. 4-7 and 4-
8). As a result, emissions from the gas-processed category are assumed to be generated by the
consumption of "waste gas" and "blast furnace / coke oven gases."
Section 3.3.4 discusses the estimates of fuel consumption used to allocate "gas-
processed" emissions of PM-10 andNOx among industries. From 1970 to 1984, the combustion
of processed gas accounts for a small percentage of PM-10 and NOX emissions. This is due, in
part, to classifying emissions from natural gas consumption by petroleum refining and steel mills
as process emissions prior to 1985 (see 1970-1990 Emissions, sections 3.2.1.2. and 3.2.1.3.).
39 Processed gas refers to still gas, refinery gas, waste gas, or coke oven gas.
January 2001
-------
46 Assigning Emissions to Industries in the United States, Revised Estimates
The U.S. EPA (1990, p. 23) provides the following emission factors:
Petroleum refinery gas
Blast furnace gas
Coke oven gas
Emission factors
(pounds of emissions per million of
cu. ft. of fuel consumed)
PM-10
3.00
2.90
4.35
NOX
140.00
23.00
80.00
The emission factor for petroleum refinery gas is equivalent to the emission factor for still gas.
Emissions from the consumption of blast furnace / coke oven gas are derived using a simple
average of the emission factors for blast furnace gas and coke oven gas. For example, the
average PM-10 emission factor for blast furnace and coke oven gases is 3.625. These emission
factors are assumed to be valid for 1970 through 1997. Consumption of blast furnace / coke
oven gases is converted into a still gas equivalent based on the emission factors for PM-10 and
NOX emissions. For example, for PM-10 emissions a unit of blast furnace / coke oven gas is
equivalent to 1.21 (=3.625/3) units of still gas. PM-10 emissions from the combustion of
processed gas are then assigned to industries based on their share of consumption of still gas and
the still gas equivalent of coke oven gas. The same procedure is followed when distributing NOX
emissions from the combustion of coke oven gas and still gas.
Wood consumed for heat and power by each industry (see discussion in Section 3.3.3.)
determines the assignment of PM-10 and NOX emissions from the "Other - wood/bark waste"
category.
3.6. Excluded Fuel Combustion Source Categories
The 1988 MECS (U.S. DOE, 1991, p. 47), the 1991 MECS (U.S. DOE, 1994, p. 85), and
the 1994 MECS (U.S. DOE, 1997, p. 82) publish estimates ("Total Inputs of Selected Byproduct
Energy for Heat, Power, and Electricity Generation" table) of the consumption of the following
byproduct fuels: (1) blast furnace/coke oven gases, (2) waste gas, (3) petroleum coke, (4) pulping
liquor, (5) wood chips, bark, and (6) waste oils / tars and waste materials. In 1994 (U.S. DOE,
1997, p. 82), the following industries account for most of the consumption of these fuels: Lumber
and Wood Products (SIC 24), Paper and Allied Products (SIC 26), Chemicals and Allied
products (SIC 28), Petroleum Refining (SIC 29), and Primary Metal Industries (SIC 33).
Of these fuels, emissions from the (1) blast furnace/coke oven gases, (2) waste gas, and (3) wood
chips, bark categories are assigned to industries.
January 2001
-------
Assigning Emissions to Industries in the United States, Revised Estimates 47
In addition to a lack of information about the "coal - other" (Tier IE) category, there is
limited information about the following (Tier II) categories in the Fuel Combustion - Industrial
(Tier I) category: (1) Other and (2) Internal Combustion. In National Air Pollutant Emissions
Trends Procedures Document, 1900-1996 Projections 1999-2010 (U.S. EPA, 1998a, p. 3-46),
the "other" (Tier II) category includes emissions from the following fuel sources: (1) coke, (2)
bagasse, (3) kerosene, (4) liquid petroleum gas, and (5) wood.40 Combustion of "other" fuels -
excluding wood - is the source of PM-10 and NOX, emissions from the "Other, other" category.
Unfortunately, data on emissions from "Other" fuels are not available for 1970 to 1990.
The only information available for the consumption of wood and other fuels by the industrial
sector is for 1990 in the SIC7090.WK3 spreadsheet, which provides information on the quantity
of fuel consumed and the emission factors.41 If identical control efficiencies exist on the
combustion of each fuel for each pollutant, the product of the quantities of fuel consumed and the
emission factors provide an estimate of the relative share of emissions associated with these
fuels. Within the "other fuel" category, coke (55.3 percent) and coke-oven gases (36.8) account
for more than 92 percent of SOX emissions. Wood accounts for 93.6 percent of VOC emissions,
and bagasse accounts for 94 percent of the remaining VOC emissions.42 Wood accounts for 85.7
percent of the PM emissions. Bagasse (72.9 percent) and coke (23.5 percent) account for most of
the remaining PM emissions. Wood accounts for 83.6 percent of the NOX emissions. Most of the
remaining NOX emissions are the result of combustion of coke (50.1 percent) and LPG (23.4
percent). Wood accounts for 96.1 percent of the CO emissions, with bagasse (57.2 percent) and
LPG (21.3 percent) accounting for most of the remaining CO emissions from the "other fuel"
category.
Since the 1970-1997 NAPET provides no additional information about the types of fuels
included in the "Other" fuel combustion category for SO2, VOC, and CO emissions, these
emissions are not assigned to industries. In addition, PM-10 and NOX emissions from the "Other
- other" and "Other - liquid waste" categories are not assigned to industries. Finally, industries
are not assigned emissions for the "Internal Combustion" category.
In conclusion, emissions from the following "fuel combustion, industrial" categories are
not assigned to industries: (1) coal - other, (2) other (when listed as a single quantity of
emissions), (3) other - liquid waste, (4) other - other, and (5) internal combustion. As a result,
emissions from the combustion of the following fuels for heat and power are not assigned to
40 "coke" includes both petroleum coke and coal coke (U.S. EPA 1998a, p. 3-46 and 3-37)
41 The 1940-1990 NAPEE (U.S. EPA, 1991, pp. 70-74) provides separate estimates of
emissions from wood and an aggregate of the remaining fuels in the "other" subcategory for
1970, 1975, 1979-1990.
42 Bagasse is consumed at sugar mills (SIC 20) (see U.S. EPA, 1996, Supplement B, Chapter
1, Section 8).
January 2001
-------
48 Assigning Emissions to Industries in the United States, Revised Estimates
industries: (1) coke and breeze, (2) LPG, (3) kerosene, (4) petroleum coke, (5) pulping liquor, (6)
waste oils / tars and waste materials, (7) agricultural waste (includes bagasse), and (8) wood-
related and paper-related refuse.
3.7 Fuel Consumption and Fuel Combustion Source Categories listed as "NA'
When the DAEMEC lists a "NA" or "W," it is necessary to estimate the energy
consumption of that industry. Since there are discrepancies between the NEA and the DAEMEC
estimates of fuel consumption and the heat content of fuels, the DAEMEC estimates of fuel
consumption are used to estimate the missing observations. This allows the estimates to be
consistent with the data from the DAEMEC. Appendix D describes the procedures used to
estimate missing observations in the DAEMEC.
Table 5 lists a "NA" for those fuel combustion categories that do not report emissions
prior to 1985. Emissions from these categories are assumed to represent cases when the
emissions were included in another category prior to 1985. As a result, no emissions are assigned
to these Tier in categories for 1970 to 1984. The only exceptions are NOX emissions from the
"Gas, other" and "Oil, other" category in 1970. Since these two categories reports emissions for
1960 and 1971 to 1984, the backward indexing method is used to estimate the emissions from
these two categories in 1970.
In addition, the 1970-1997 NAPET does not estimate fuel combustion emissions of PM-
10 for 1971 to 1974. Appendix D provides a description of the two-way indexing method, which
in combination with changes in fuel consumption by the industrial sector, used by this study to
estimate PM-10 emissions for 1971 to 1974.
4. OVERVIEW OF THIS REPORT
This report developed procedures to assign emissions of criteria air pollutants among the
twenty two-digit SIC manufacturing industries in the United States for 1970 to 1997.43 For 1993
to 1997, the allocation of fuel combustion emissions is preliminary because the 1994 MECS and
the 1992 economic census for the agriculture, construction, and mineral industries represent the
most recent estimates of energy consumption. In addition, the Board of Governors of the Federal
Reserve System has revised its estimates of industrial production for 1992 to 1997. When the
1997 economic census and the 1998 MECS are published, it will be possible to revise the
assignment of emissions from fuel combustion for 1993 to 1997.
There are several explanations for the divergence between this report's emission
estimates and those in the 1970-1997NAPET (U.S. EPA, 1998b). First, this report only assigns
emissions to industries of the industrial sector. Sources such as motor vehicles emissions are not
43 The 1970-1997 NAPET (U.S. EPA, 1998b), which is the source of the emission estimates
used in this report, estimates emissions from electric utilities for 1970 to 1997.
January 2001
-------
Assigning Emissions to Industries in the United States, Revised Estimates 49
assigned to industries. Second, this report estimates emissions from Tier II categories whose
emissions are listed as "NA" for 1970 to 1984 by the 1970-1997NAPET'(see Section 2.3.).
Third, this report does not assign emissions from the "Miscellaneous Industrial Processes"
category to industries. Finally, this report excludes emissions from the combustion of several
types of fuel because there is insufficient information about the consumption of those fuels by
individual industries (see Section 3.6.).
Table 6 lists the sources of emissions assigned to industries. The "Spreadsheet Estimates"
portion of Table 6 shows the quantity of emissions assigned to the electric utility and industrial
sectors. In Table 6, industrial process emissions include VOC emissions from those Solvent
Utilization and Storage & Transport (Tier I) categories assigned to industries by this report. In
addition to emissions from miscellaneous fuels and processes, and various categories within the
Solvent Utilization and Storage & Transport (Tier I) categories, all emissions from the Waste
Disposal & Recycling, On-road Vehicles, Off-road Vehicles and, Natural Sources, and
Miscellaneous (Tier I) categories are not assigned to industries.
The "Percentage of National Estimates in Spreadsheets" section of Table 6 lists the
percentage of emissions from all sources assigned to the electric utility and industrial sectors. In
1997, the industrial and electric utility sectors generated 88 percent of SO2 emissions, and 59
percent of Pb emissions.
NOX is produced by both stationary and mobile sources, and VOCs and CO are more
closely associated with mobile sources. As a result, the industrial and electric utility sectors
account for a smaller percentage of the emissions of these four pollutants: for NOX, 39 percent;
for VOCs, 22 percent; for CO, 7 percent; and, for PM-10, 4 percent.
The increased share of Pb emissions from the industrial and electric utility sectors results
from a decline in Pb emissions from the On-Road Vehicles (Tier I) category. After 1984, the
decline in the percentage of PM-10 emissions assigned to the industrial and electric utility sectors
results from changes in the quantity of emissions reported for the Natural Resources and
Miscellaneous (Tier I) categories. Emissions from the Fugitive Dust (Tier n) category, which is
in the Miscellaneous category, account for more than 55 percent of PM-10 emissions in 1997.
There are two potential sources of discontinuities in the time series of emissions. The first
occurs as a result of using data from the NEA and DAEMEC to allocate emissions from fuel
combustion. The second source is associated with changes in the methodology used by the U.S.
EPA to estimate process emissions. These changes resulted in one methodology being employed
to estimate process emissions for 1970 to 1984, and another methodology being used to estimate
emissions for 1985 to 1997. The next few paragraphs review these issues.
January 2001
-------
50
Assigning Emissions to Industries in the United States, Revised Estimates
Table 6
Emission Estimates in the Spreadsheet and the Percentage of
EPA National Estimates of Emissions in the Spreadsheet
POLLUTANT
SO2
NOX
VOC
PM-10
CO
Pb
CATEGORY
Industrial Process
Industrial Fuel
Electric Utility Fuel
Total
Industrial Process
Industrial Fuel
Electric Utility Fuel
Total
Industrial Process
Industrial Fuel
Electric Utility Fuel
Total
Industrial Process
Industrial Fuel
Electric Utility Fuel
Total
Industrial Process
Industrial Fuel
Electric Utility Fuel
Total
Industrial Process
Industrial Fuel
Electric Utility Fuel
Total
SPREADSHEET
*
ESTIMATE
1970
7.14
4.50
17.40
29.03
1.00
4.45
4.90
10.36
7.59
0.08
0.03
7.71
7.68
0.62
1.78
10.07
9.89
0.61
0.24
10.73
24.87
0.24
0.33
25.43
1984 1 1997
3.30 1.66
2.67 3.19
16.02 13.08
21.98 17.93
0.80 0.79
3.35 2.33
7.27 6.18
11.42 9.30
7.04 4.18
0.06 0.09
0.04 0.05
7.15 4.32
2.34 0.84
0.61 0.24
0.63 0.29
3.58 1.37
5.18 4.78
0.48 0.54
0.32 0.41
5.98 5.72
2.10 2.23
0.03 0.02
0.09 0.06
2.22 2.31
PERCENTAGE OF EPA
NATIONAL ESTIMATES
IN THE SPREADSHEET**
1970
22.88
14.41
55.75
93.04
4.64
20.62
22.69
47.95
24.61
0.27
0.10
24.98
58.68
4.70
13.56
76.95
7.71
0.47
0.19
8.37
11.26
0.11
0.15
11.51
1984
14.01
11.33
68.12
93.47
3.17
13.29
28.84
45.30
27.02
0.25
0.17
27.44
36.00
9.43
9.76
55.19
4.48
0.42
0.27
5.18
6.16
0.09
0.26
6.50
1997
8.15
15.65
64.23
88.02
3.37
9.87
26.20
39.44
21.74
0.49
0.26
22.50
2.49
0.71
0.86
4.07
5.46
0.61
0.46
6.54
56.94
0.43
1.63
59.01
In millions of short tons, except Pb, which is in thousands of short tons. One short ton equals two
thousand pounds.
Category totals and percentages are calculated from estimates in sheet AR of SIC7097.WK3. Percentages
do not total 100 percent of EPA emission estimates because the spreadsheets exclude emissions from miscellaneous
fuels, commercial/institutional, residential and mobile sources, some solvent utilization VOC emissions, and various
miscellaneous sources.
January 2001
-------
Assigning Emissions to Industries in the United States, Revised Estimates 51
There are differences between the NEA and DAEMEC estimates of energy consumption
by manufacturing industries. There are also differences between the NEA estimates of energy
consumed by the nonmanufacturing sectors and estimates for these sectors generated by the
modified-NEA methodology developed in this report for 1985 to 1997. As a result, some
differences exist between the energy consumption data for 1970 to 1984 and the data for 1985 to
1997. However, the differences between these two sources of data do not appear to have a
substantial impact on the assignment of fuel combustion emissions.
The use of energy consumption data to allocate emissions is further complicated by
changes in industry definitions within the SIC system. The 1972 SIC industry definitions are
used by the 1985 MECS (see U.S. DOE, 1988a, p. 105) and the NEA (see Price and Wendling,
1986, p. 1). However, the MECS data from 1988, 1991, and 1994 use the 1987 SIC industry
definitions. Estimates of energy consumption for 1974 to 1986,which are published in the
DAEMEC, use the 1972 SIC classification scheme, and the DAEMEC estimates of energy
consumption data for the years after 1986 utilize the 1987 SIC classification scheme (see U.S.
DOE, 1992, p. 97). Since the 1972 and 1987 SIC industry definitions differ, in some cases it
might be necessary to combine the Electronic and other electric equipment (SIC 36) and
Instruments and related products (SIC 38) sectors.44
The allocation of fuel combustion emissions can be improved with a complete
reconciliation of the estimates of fuel consumption for 1985 to 1997 from this report and the
NEA estimates, which are used to allocate emissions for 1970 to 1984. A reconciliation of the
energy consumption data would allow a more consistent allocation of emissions.
Allocating emissions from the industrial fuel combustion category among the industries is
complicated by changes in the treatment of the combustion of certain fuels by selected industries.
The differences between the methodology employed to estimate emissions for 1970 to 1984, and
the methodology used to estimate emissions for 1985 to the present are discussed in Section
3.2.1.2. of 1970 to 1990 Emissions (U.S. EPA 2000). Changes in the EPA methodology may
explain the decline in PM-10 process emissions from the cement mfg. (Tier HI) category and the
increase in PM-10 emissions from coal, oil, and gas combustion by the industrial sector. They
may also explain the decline in SO2 process emissions from the petroleum refineries (Tier HI),
44 Links to the documentation for the BEA employment data (FTE, FTPT, and PEP variables)
and the zipped data files are available under "Gross Product by Industry" (Gross Product by
Industry and the Components of Gross Domestic Income that define current dollar GPO) at the
following URL: http://www.bea.doc.gov/bea/dn2.htm. Zipped data files in GPO4797.EXE are
available. Data from 1948 through 1987 use the 1972 SIC codes and are in the spreadsheet file
GPO72SIC.WK1. Data from 1987 through 1997 use the 1987 SIC codes and are in the
spreadsheet file GPO87SIC.WK1. Comparing employment by two-digit SIC manufacturing
industries in 1987 using the 1972 and 1987 code definitions, it can be seen that Electronic and
other electrical equipment (SIC 36) and Instruments and related products (SIC 38) should be
combined to maintain common industry definitions for two-digit SIC codes across the 1972 and
1987 code definitions.
January 2001
-------
52 Assigning Emissions to Industries in the United States, Revised Estimates
cement mfg. (Tier HI), and ferrous metals processing (Tier II) categories. Finally, the increase in
SO2 emissions from the oil and gas combustion (Tier II) categories may be explained by the
change in methodology.
Assigning process emissions to industries, requires assigning SIC codes to the Tier I, Tier
II, and Tier in categories.45 There are differences between the procedures used by the 1970-1997
NAPETio estimate process emissions for the years prior to 1985 and the procedures used
estimate process emissions from 1985 to 1997. Although changing from NEA to DAEMEC data
may account for some discontinuities in the assignment of fuel combustion emissions, it appears
that changes in the methodology used to estimate process emissions account for most of the
major changes in emissions between 1984 and 1985. For example, there were substantial
declines in PM-10 emissions reported for the following categories, which are assigned to the
nonmanufacturing portion of the industrial sector: (1) Agriculture, Food & Kindred Products,
country elevators (Tier in), (2) Agriculture, Food & Kindred Products, terminal elevators (Tier
in), (3) Mineral Products, surface mining (Tier in), and (4) Metals Processing, NEC (Tier II).
For manufacturing industries, most of the substantial changes in emissions between 1984
and 1985 are the result of changes in emissions from processes or solvent utilization. For
example, declines in VOC emissions from SIC 24, SIC 26, SIC 28, and SIC 29 can be explained
as follows. The decline in VOC emissions from SIC 24 results from the decline in emissions
associated with the flatwood products (Tier in) category of the Solvent Utilization (Tier I)
category. The decline in VOC emissions from SIC 26 is the result of a decline in emissions from
the paper (Tier HI) category of the Solvent Utilization (Tier I) category. The decline in VOC
emissions from SIC 28 results from the decline in emissions associated with several Tier H
categories of the Chemical & Allied Product Mfg. (Tier I) category. Finally, the decline in VOC
emissions from SIC 29 is the result of a decline in emissions from the petroleum refineries &
related industries (Tier II) category of the Petroleum & Related Industries (Tier I) category.
Changes in PM-10 emissions from SIC 24 and SIC 26 result from a decrease in emissions
from the wood/bark waste (Tier in) category of the Fuel Combustion Industrial, Other (Tier H)
category. The declines in PM-10 emissions from SIC 28 and SIC 29 are associated with
declines in the process emissions from several Tier H categories within the Chemical & Allied
Product Mfg. (Tier I) category, and the Asphalt Manufacturing (Tier II) category of the
Petroleum & Related Industries (Tier I) category.
Recently, the statistical agencies of several countries have developed estimates of
emissions from industries. For example, Statistics Denmark (1998) developed an inventory of
emissions for Denmark, and Stahmer, Kuhn, and Braun (1998) developed an inventory of
emissions for Germany. These two reports provide information on some of the assumptions
needed to derive the estimates of emissions. Statistics Denmark (1998, p. 11) acknowledge that
while it is preferable to assign emissions from solvents to the end-users, they believed that it was
45 The National Air Pollutant Emissions Trends Procedures Document, 1900-1996
Projections 1999-2010 (U.S. EPA, 1998a, p. 4-2) discusses the difficulty of matching SCCs and
SIC codes.
January 2001
-------
Assigning Emissions to Industries in the United States, Revised Estimates 53
too burdensome. Hence, they do not allocate emissions from solvents. Emissions from fuel
combustion are the product of fuel consumption (in joules) and an emission factor associated
with each sector, fuel type and pollutant combination. In most cases, the emissions factors are the
same for different industries (Statistics Denmark, 1998, p. 13). In addition, energy consumed for
transportation by an industry is included in emissions from that sector (see Statistics Denmark,
1998, p. 13 , "the emission... per unit of gasoline is the same whether the gasoline is used in the
dairy sector or in the sector for book printing"). Stahmer, Kuhn, and Braun (1998, p. 22) also
estimate air emissions by calculating the product of the quantity (tones, joules) of a fuel
consumed for combustion and an emission coefficient.
The procedures developed in this report are comparable to those developed by other
efforts, such as those described in the previous paragraph, to allocate emissions among
industries. The next MECS will report data for the year 1998, and subsequent MECS will be
administered quadrennially. If the methodology specified in this report is acceptable and the
necessary data are available, it should be possible to update these estimates of emissions of
criteria air pollutants from the twenty two-digit SIC manufacturing industries.
The ongoing implementation of the NAICS poses a challenge to maintaining a consistent
time series of emissions from industries. Within the next few years, the NAICS system will
replace the SIC system. It will be possible to derive more than two-thirds of all 4-digit SICs
from the NAICS system (see Executive Office of the President, Office of Management and
Budget, 1998, p. 23). The U.S. Department of Commerce (1999, pp. 9 and 11) provides
preliminary, summary statistics of revenue, payroll, and number of employees for the twenty
three-digit NAICS manufacturing industries that correspond to the twenty two-digit SIC
manufacturing industries.46 Several forthcoming reports will provide additional insights into the
importance of the discrepancies between the SIC and NAICS industries. Converting the time
series of emissions from the twenty two-digit SIC manufacturing industries, which was
developed in this report, to the twenty three-digit NAICS manufacturing industries, requires
historical data on output and fuel consumption by three-digit and perhaps four-digit SIC
industries. Until additional information is available, it is not possible to determine the feasibility
of developing a consistent historical time series of estimates of emissions for the three-digit
NAICS manufacturing industries.47
46 There have been a number of additions and deletions to the manufacturing sector of the
United States. For example, the logging industry (SIC 2411) is part of the Agriculture, Forestry,
Fishing, and Hunting sector (NAICS 11) in NAICS (Executive Office of the President 1998, p.
772), and the newspaper printing and publishing industry (SIC 2711) is moved to the Information
sector (NAICS 51) in the NAICS system (Executive Office of the President 1998, p. 864).
47 The following URL provides links to the implementation plans of various agencies of the
U.S. government: http://www.census.gov/epcd/www/naics.html
The Bureau of the Census of the U.S. Department of Commerce addresses some of the issues
concerning the breaks in time series of data at the following URL:
http://www.census.gov/epcd/www/naicsusr.htmltfBREAKS.
January 2001
-------
54 Assigning Emissions to Industries in the United States, Revised Estimates
There are two potential extensions of the data developed in this report. The first
extension involves generating estimates of fuel consumption by industries for transportation.
This would enable the assignment of emissions from on-road and off-road vehicles to industries.
The second extension involves developing of estimates of emissions at a more
disaggregated level. Estimates of process emissions are available by SCC for 1985 to the
present. It would be necessary to develop a concordance between the SCCs and the SIC or
NAICS codes. Since information on fuel consumption is required to allocate emissions from the
combustion of fuel in the industrial sector, the principal constraint on developing more
disaggregated estimates of emissions is the availability of information on fuel consumption by
industries, and the output data required to estimate fuel consumption for the years in which data
are not collected.
The Bureau of Economic Analysis of the U.S. Department of Commerce (Seskin, Eugene and
David F. Sullivan, 2000, pp. 27-28) has described its plans for implementing the NAICS system.
It appears that neither the Bureau of the Census nor the BEA plan to reclassify their estimates of
industry output (from SIC to NAICS) for the years prior to implementing NAICS. At this time,
the U.S. Department of Energy does not plan to reclassify SIC-based data collected previous to
1998. However, the 1998 MECS will provide information on energy consumption using both
NAICS and SIC codes (see http://www.eia.doe.gov/emeu/mecs/mecs98/naics/naics8.htmn. The
Board of Governors of the Federal Reserve System plans to reclassify some of its estimates of
production by industries from the SIC system to the NAICS. However, it is uncertain how far
back in time the Board of Governors will be able to reclassify its estimates of industrial output.
January 2001
-------
Assigning Emissions to Industries in the United States, Revised Estimates 55
REFERENCES
Armitage, Ken and Dixon A. Tranum (1990), "Industrial Production: 1989 Developments and
Historical Revisions" Federal Reserve Bulletin, 76, (April), 187-204.
Board of Governors, Federal Reserve System (1999), "Industrial Production and Capacity
Utilization, G-17," (downloaded on 9/3/99 from the following URL:
http://www.bog.frb.fed.us/releases/G17/).
Executive Office of the President, Council of Economic Advisers (1987), Economic Report of
the President, 1987, Washington, DC: U.S. Government Printing Office.
Executive Office of the President, Office of Management and Budget (1987), Standard Industrial
Classification Manual, National Technical Information Service, Springfield, Virginia PB-87-
100012. (The following URLs have information on the 1987 SIC codes:
http://www.osha.gov/oshstats/sicser.html and http://www.lib.virginia.edu/socsci/sic.htmn
Executive Office of the President, Office of Management and Budget (1998), North American
Industrial Classification System: United States, 1997, National Technical Information Service,
Springfield, Virginia, PB98-127293.
Lum, Sherlene K.S. and Brian C. Moyer (1998), "Gross Product by Industry," Survey of Current
Business, 78, No. 11 (November), 20-40. (see "Industry Estimates" at the following URL:
http://www.bea.doc.gov/bea/anl.htm for PDF and HTML versions of the article. The data are
available at http://www.bea.doc.gov/dn2/gpo.htm under "Detailed Data Files.")
Lum, Sherlene K.S. and Robert E. Yuskavage (1997), "Gross Product by Industry, 1947-96,"
Survey of Current Business, 77, No. 11 (November), 20-34. (HTML and PDF files are located at
the following URL under "Industry Estimates": http://www.bea.doc.gov/bea/anl.htm).
Nestor, Deborah Vaughn and Carl A. Pasurka, Jr. (1998), "USA: Environmental Protection
Activities and Their Consequences," in Environmental Accounting in Theory and Practice, Vol.
11 of Economy and Environment series, eds. Kimio Uno and Peter Bartelmus, pp. 131-142,
Boston: Kluwer Academic Publishers.
Price, Lori H. and Robert M. Wendling (1986), "National Energy Accounts: An Overview,"
Capital, Energy and Productivity Studies Division, Office of Business Analysis, U.S. Department
of Commerce, Draft, March 1986.
Seskin, Eugene P. and David F. Sullivan (2000), "Annual Revision to the National Income and
Product Estimates," Survey of Current Business, 80, No. 8 (August), 6-33.
Stahmer, C., M. Kuhn, andN. Baum (1998), "Physical Input-Output Tables for Germany, 1990,"
Eurostat Working Paper No. 2/1998/B/l, European Commission.
January 2001
-------
56 Assigning Emissions to Industries in the United States, Revised Estimates
Statistics Denmark (1998), Danish NAMEA, 1980-1992, in cooperation with Eurostat.
U.S. Department of Commerce, Bureau of the Census (1981), 7977 Census of Mineral
Industries, Washington, DC: U.S. Government Printing Office.
U.S. Department of Commerce, Bureau of the Census (1984), 7952 Census of Construction,
Washington, DC: U.S. Government Printing Office.
U.S. Department of Commerce, Bureau of the Census (1985), 7952 Census of Mineral
Industries, Washington, DC: U.S. Government Printing Office.
U.S. Department of Commerce, Bureau of the Census (1990a), 7957 Census of Construction,
Washington, DC: U.S. Government Printing Office.
U.S. Department of Commerce, Bureau of the Census (1990b), 7957 Census of Mineral
Industries, Washington, DC: U.S. Government Printing Office.
U.S. Department of Commerce, Bureau of the Census (1994), 7992 Census of Agriculture,
Washington, DC: U.S. Government Printing Office.
(http://www.census.gOv/prod/2/agr/92area/aca51 .pdf)
U.S. Department of Commerce, Bureau of the Census (1996a), 7992 Census of Construction,
Washington, DC: U.S. Government Printing Office.
(http ://www. census. gov/prod/1 /constr/92area/cca 1 Of .pdf)
U.S. Department of Commerce, Bureau of the Census (1996b), 7992 Census of Mineral
Industries, Fuels and Electric Energy Consumed, MIC92-S-2, it was last modified in September
1996 (HTML file is located at: http://www.census.gov/econ/www/minesum.htmltfp2)
U.S. Department of Commerce, Bureau of the Census (1999), 1997 Economic Census, Advance
Report, Core Business Statistics Series, EC97X-CS1, (PDF file is located at:
http://www.census.gov/epcd/www/advance2.htm).
U.S. Department of Energy, Energy Information Administration (1986), Petroleum Marketing
Monthly, DOE/EIA-03 80(86), July.
U.S. Department of Energy, Energy Information Administration (1987), Petroleum Marketing
Monthly, DOE/EIA-03 80(87), July.
U.S. Department of Energy, Energy Information Administration (1988a), 1985 Manufacturing
Energy Consumption Survey, DOE/EIA-0512(85), Washington, D.C. (HTML, PDF, and WK1
files are located at: http://www.eia.doe.gov/emeu/consumption)
U.S. Department of Energy, Energy Information Administration (1988b), Petroleum Marketing
Monthly, DOE/EIA-0380(88), June.
January 2001
-------
Assigning Emissions to Industries in the United States, Revised Estimates 57
U.S. Department of Energy, Energy Information Administration (1989), Petroleum Marketing
Annual, DOE/EIA-0487(88).
U.S. Department of Energy, Energy Information Administration (1991), 1988 Manufacturing
Energy Consumption Survey, DOE/EIA-0512(88), Washington, D.C. (HTML, PDF, and WK1
files are located at: http://www.eia.doe, gov/emeu/consumption)
U.S. Department of Energy, Energy Information Administration (1992), Derived Annual
Estimates of Manufacturing Energy Consumption 1974-1988. DOE/EIA-0555(92)/3,
Washington, D.C. (HTML file is located at: http://www.eia.doe.gov/emeu/consumption)
U.S. Department of Energy, Energy Information Administration (1994), 1991 Manufacturing
Energy Consumption Survey, DOE/EIA-0512(91), Washington, D.C. (HTML, PDF, and WK1
files are located at: http://www.eia.doe.gov/emeu/consumption)
U.S. Department of Energy, Energy Information Administration (1996), State Energy Data
Report: 1995, DOE/EIA-0214(95), Washington, D.C. (updated files are at
http://www.eia.doe. gov/emeu/states/_states.html).
U.S. Department of Energy, Energy Information Administration (1997), 1994 Manufacturing
Energy Consumption Survey, DOE/EIA-0512(94), Washington, D.C. (HTML, PDF, and WK1
files are located at: (http://www.eia.doe.gov/emeu/consumption)
U.S. Department of Energy, Energy Information Administration (1998a), Derived Annual
Estimates of Manufacturing Energy Consumption 1988-1994, Energy Consumption Series, it
was last modified on August 14, 1998 (HTML file is located at:
http://www.eia.doe.gov/emeu/consumption)
U.S. Department of Energy, Energy Information Administration (1998b), Historical Natural Gas
Annual 1930 Through 1997, Report, DOE/EIA-0110(97)
(http ://www. eia. doe. gov/oil_gas/natural_gas/nat_frame.html).
U.S. Department of Energy, Energy Information Administration (1999), Monthly Energy Review
On-line, (thermal conversion factors for various fuels were downloaded on 1/16/99 and 3/2/99,
and are located under "Thermal Conversion Factors" at http://tonto.eia.doe.gov/mer/)
U.S. Department of Energy, Energy Information Administration (various issues), Quarterly Coal
Report, DOE/EIA-0121.
U.S. Department of Energy, Energy Information Administration (various issues), Coal Industry
Annual, DOE/EIA-0584.
U.S. Department of Energy, Energy Information Administration (various issues), Fuel Oil and
Kerosene Sales, DOE/EIA-0535.
January 2001
-------
58 Assigning Emissions to Industries in the United States, Revised Estimates
U.S. Environmental Protection Agency, Office of Air and Radiation, Office of Air Quality
Planning and Standards (1990) AIRS Facility Subsystem Source Classification Codes and
Emission Factor Listing for Criteria Air Pollutants, EPA 450/4-90-003 (NTIS code: PB90-
207242).
U.S. Environmental Protection Agency, Office of Air and Radiation, Office of Air Quality
Planning and Standards (1991), National Air Pollutant Emission Estimates., 1940-1990. EPA-
450/4-91-026.
U.S. Environmental Protection Agency, Office of Air and Radiation, Office of Air Quality
Planning and Standards (1994), National Air Pollutant Emissions Trends, Procedures Document
1900-1993, EPA 454/R-95-002.
U.S. Environmental Protection Agency, Office of Air and Radiation, Office of Air Quality
Planning and Standards (1996), AP-42, Fifth Edition, Supplement B,
(http://www.epa.gov/ttn/chief/ap42supp.html#sup B)
U.S. Environmental Protection Agency, Office of Air and Radiation, Office of Air Quality
Planning and Standards (1998a), National Air Pollutant Emission Trends Procedures Document,
1900-1996. Projections 1999-2010, EPA 454/R-98-008.
(http ://www. epa. gov/ttn/chief/ei_data.html#ETDP)
U.S. Environmental Protection Agency, Office of Air and Radiation, Office of Air Quality
Planning and Standards (1998b), National Air Pollutant Emission Trends Update: 1970-1997.
EPA 454/E-98-007. (http://www.epa.gov/ttn/chief/trends97/emtrnd.htmn
U.S. Environmental Protection Agency, Office of Policy, Economics and Innovation, National
Center for Environmental Economics (2001), Statistical Methodology for Assigning Emissions to
Industries in the United States: 1970 to 1990, EPA 240-R-01-002.
(http ://www. epa. gov/economics/)
Yuskavage, Robert E. (1996), "Improved Estimates of Gross Product by Industry, 1959-94,"
Survey of Current Business, 76, No. 8 (August), 133-155. (HTML and PDF files are located at:
http://www.bea.doc.gov/bea/anl.htm).
January 2001
-------
APPENDIX A
NEA SECTORING PLAN
-------
-------
Assigning Emissions to Industries in the United States, Revised Estimates
A-l
NEA
Code
01000
02000
03000
04000
05000
06001
06002
07010
07020
08001
08002
08003
09000
10000
11001
11002
12000
Industry Title
AGRICULTURE
Livestock and livestock products
Other agricultural products
Forestry and forestry products
Agricultural, forestry and fishery services
MINING
Iron and feroalloy ores mining
Uranium - radium - vanadium ores mining
Nonferrous metal ores mining
Anthracite coal mining
Bituminous and lignite coal mining
Crude petroleum
Natural gas
Natural gas liquids
Stone and clay mining and quarrying
Chemical and fertilizer mineral mining
CONSTRUCTION
Oil and gas well drilling
New construction
Maintenance and repair construction
SIC Code
02 (ex. 0254, pt. 0219, pt. 0259,
pt. 0291), pt. 0191
01 (ex. pt. 0191), pt. 0219,
pt. 0259, pt. 0291
08 (ex. 085), 09 (ex. 092)
0254, 07 (ex. 074), 085, 092
101, 106
1094
102, 103, 104, 105, pt. 108, 109
(ex. 1094)
pt. 11
pt. 12
pt. 131
pt. 131
132
141, 142, 144, 145, pt. 148, 149
147
pt. 138
pt. 15, pt. 16, pt. 17, pt. 108,
pt. 11, pt. 12, pt.148
pt. 15, pt. 16, pt. 17, pt. 138
Source: Price and Wendling (1986, Appendix B)
January 2001
-------
-------
APPENDIX B
CONCORDANCE BETWEEN THE SOLVENT UTILIZATION AND
STORAGE & TRANSPORT CATEGORIES,
AND THE NEA AND SIC CODES
-------
-------
Assigning Emissions to Industries in the United States, Revised Estimates
SOLVENT UTILIZATION
Degreasing
Graphic Arts
Dry Cleaning
Surface Coating
industrial adhesives
fabrics
paper
large appliances
magnet wire
auto & light trucks
metal cans
metal coil
wood furniture
metal furniture
flatwood products
plastic parts
large ships
aircraft
misc. metal parts
steel drums
architectural
traffic markings
maintenance coatings
railroad
auto refinishing
SIC code (NEA codes in parentheses)
NA
27
721 (not part of the industrial sector)
NA
22
26
36
33
37
34
33
25
25
24
30
37
37
34
34
(NEA 11 002)
NA
(NEA 12000)
37
7532 (not part of the industrial sector)
January 2001
-------
B-2
Assigning Emissions to Industries in the United States, Revised Estimates
SOLVENT UTILIZATION
(continued)
machinery
electronic & other electrical
general
miscellaneous
thinning solvents
other
Other Industrial
miscellaneous
rubber & plastics mfg.
other
Nonindustrial
cutback asphalt
other asphalt
pesticide application
adhesives
consumer solvents
other
Other
STORAGE & TRANSPORT
Bulk Terminals & Plants
Petroleum & Petroleum Product
Storage
fixed roof gasoline
fixed roof crude
floating roof gasoline
SIC code (NEA codes in parentheses)
35
36
NA
NA
NA
NA
NA
28
NA
(NEA 12000)
NA
(NEA 02000)
NA
NA
NA
NA
NA
29
29
29
January 2001
-------
Assigning Emissions to Industries in the United States, Revised Estimates
B-3
STORAGE & TRANSPORT
(continued)
floating roof crude
efr / seal gasoline
efr / seal crude
ifr / seal gasoline
ifr / seal crude
variable vapor space gasoline
area source: crude
other
Petroleum & Petroleum Product
Transport
Service Stations: Stage I
Service Stations: Stage II
Service Stations: Breathing &
Emptying
Organic Chemical Storage
Organic Chemical Transport
Inorganic Chemical Storage
Inorganic Chemical Transport
Bulk Materials Storage
SIC code (NEA codes in parentheses)
29
29
29
29
29
29
29
29
NA
NA
NA
NA
28
NA
28
NA
NA
NA = not assigned to an industry
NOTE: Tier I category names are in BOLD, Tier II category names are in the regular font, and
Tier HI category names are in italics.
January 2001
-------
-------
APPENDIX C
PERCENTAGE OF 1985 PROCESS EMISSIONS
WITH MISSING DATA FROM 1970 TO 1984
-------
-------
Assigning Emissions to Industries in the United States, Revised Estimates
CO Emissions
(in thousand short tons)
Coal
Oil
Gas
Other
TOTAL - Fuel Comb. Industrial
Chemicals
Metals
Petroleum
Other
TOTAL - Industrial Processes
1985
Emissions
86
47
257
390
1845
2223
462
694
5224
1985 Emissions
of cells listed as
"NA" in 1984
34
6
13
24
77
Percent of 1985
Emissions listed
as "NA" in 1984
1.9
0.3
2.8
3.4
1.5
NOx Emissions
(in thousand short tons)
Coal
Oil
Gas
Other - wood/bark waste
TOTAL - Fuel Comb. Industrial
Chemicals
Metals
Petroleum
Other
TOTAL - Industrial Processes
1985
Emissions
608
309
1520
118
2554
262
87
124
327
800
1985 Emissions
of cells listed as
"NA" in 1984
203
29
69
14
315
Percent of 1985
Emissions listed
as "NA" in 1984
77.4
33.0
55.4
4.4
39.3
January 2001
-------
C-2
Assigning Emissions to Industries in the United States, Revised Estimates
VOC Emissions
(in thousand short tons)
Coal
Oil
Gas
Other
TOTAL - Fuel Comb. Industrial
Chemicals
Metals
Petroleum
Other
TOTAL - Industrial Processes
Solvent Utilization
Storage & Transport
GRAND TOTAL
1985
Emissions
7
17
57
81
881
76
703
390
2049
2056
257
4362
1985 Emissions
of cells listed as
"NA" in 1984
11
19
165
196
93
289
Percent of 1985
Emissions listed
as "NA" in 1984
1.3
25.5
0.0
42.4
9.6
4.5
6.6
SO2 Emissions
(in thousand short tons)
Coal
Oil
Gas
Other
TOTAL - Fuel Comb. Industrial
Chemicals
Metals
Petroleum
Other
TOTAL - Industrial Processes
1985
Emissions
1818
862
397
3076
456
1042
505
425
2428
1985 Emissions
of cells listed as
"NA" in 1984
27
18
1
8
53
Percent of 1985
Emissions listed
as "NA" in 1984
5.8
1.7
0.2
1.8
2.2
January 2001
-------
Assigning Emissions to Industries in the United States, Revised Estimates
C-3
PM-10 Emissions
(in thousand short tons)
Coal
Oil
Gas
Other - wood/bark waste
TOTAL - Fuel Comb. Industrial
Chemicals
Metals
Petroleum
Other
TOTAL - Industrial Processes
1985
Emissions
71
52
47
67
236
58
220
63
611
952
1985 Emissions
of cells listed as
"NA" in 1984
4.9
0.0
0.3
41.2
46.3
Percent of 1985
Emissions listed
as "NA" in 1984
8.4
0.0
0.5
6.7
4.9
January 2001
-------
-------
APPENDIX D
TREATMENT OF MISSING OBSERVATIONS
-------
-------
Assigning Emissions to Industries in the United States, Revised Estimates D-l
This appendix discusses the treatment of missing observations (i.e., when the data are
listed as withheld, W, or are not available, NA, for energy consumption and emissions). Rather
than assign a value of zero to an observation with a "W" or "NA," many missing observations are
assigned values. In addition, observations recorded as "*" (less than 0.5 trillion Btu) in Derived
Annual Estimates of Manufacturing Energy Consumption (U.S. DOE 1992 and 1998a) are
treated as zeros. The procedures described in this appendix mirror those used by Derived Annual
Estimates of Manufacturing Energy Consumption (U.S. DOE) when it estimates energy
consumption for years between the MECS survey and the estimates of missing observations in
the NEA (see Price and Wendling 1986).
D. 1 . MANUFACTURING ENERGY CONSUMPTION
Some observations in the DAEMEC are withheld (W) or not available (NA). The
preferred procedure for estimating these missing observations is the two-way indexing method
(see U.S. DOE 1992, p. 8 and Appendix A). The two-way indexing method estimates fuel
consumption for a missing year y by using data from a known starting-year (i.e., the year prior to
the first missing year) and a known ending-year (i.e., the first year after the last missing year). A
"known" year is one in which consumption (in Btu) of that fuel is reported by the U.S. DOE. The
two "known" observations are adjusted by changes in the level of production of the two-digit SIC
industry between the known years and the missing year y. A weighted average (i.e., a linear
interpolation) of the two estimates of fuel consumption in year y yields the estimate of fuel
consumption (in Btu) in year y. Weights are determined by the relative distance of the known
starting-year and known ending-year from the missing year. The two-way indexing method for
estimating missing observations for consumption (in Btu) of the ith type of fuel (BTUyS) by a two-
digit SIC manufacturing industry (s) for each missing year (y) can be specified as:
where Ays is the Federal Reserve Board's production index for the two-digit SIC manufacturing
industry (Armitage and Tranum 1990) for year y, and Aos is the index/ output for the starting year
(y=0) and A^ is the index / level of output for the ending year (y=L).
If an estimate of fuel consumption (in Btu) for a year prior to 1985 is required to serve as
the known starting-year, the Btu value from the DAEMEC is used. DAEMEC data are used in
order to maintain consistency with the post- 1984 observations.
When a "W" or "NA" is reported in the 1994 MECS, there is no known ending-year. As a
result, it is not possible to utilize the two-way indexing method to estimate the missing
observations. For these cases, the forward indexing method is used to estimate the missing
observations (see U.S. DOE 1992, p. 8 and Appendix A). The forward indexing method can be
specified as:
January 2001
-------
D-2 Assigning Emissions to Industries in the United States, Revised Estimates
For these cases, fuel consumption (in Btu) for 1988 or 1991, which is then adjusted by changes in
the output of the two-digit SIC manufacturing industry associated with the process source
category, is used to estimate fuel consumption for all subsequent years to 1994.
It is also not possible to utilize the two-way indexing method to estimate fuel
consumption for 1995 through 1997 for any industries. As a result, energy consumption for 1995
to 1997 are estimated using the forward indexing method. In this case, fuel consumption (in Btu)
for 1994, adjusted by the change in the output of the industry, is used to estimate the fuel
consumption (in Btu) of each fuel type by an industry for 1995 through 1997. Until the 1998
MECS becomes available, all estimates of fuel consumption derived using the forward indexing
method are preliminary.
D.2. NONMANUFACTURING ENERGY CONSUMPTION
During 1985 to 1997, the Census of Agriculture, the Census of Mineral Industries, and
the Census of Construction are only available for 1987 and 1992. Depending on whether data are
available in the 1982 economic census, either the two-way indexing method or the backward
indexing method is used to estimate fuel consumption for 1985 and 1986. The two-way indexing
method to estimate the consumption of coal, and natural gas for heat and power for 1988 to 1991.
First, fuel consumption coefficients (i.e., the fuel-output ratios) for 1988 to 1991 are derived by
linearly interpolating the fuel consumption coefficients from the 1987 and 1992 economic
census. These fuel consumption coefficients are then multiplied by the U.S. BEA constant dollar
estimate of output of the agriculture, mining, and construction sectors, respectively. This yields
estimates of fuel consumption by these sectors for 1988 through 1991. The forward indexing
method is also used to estimate fuel consumption for 1993 to 1997. Fuel consumption by the
three sectors within the nonmanufacturing portion of the industrial sector is estimated by
multiplying their 1992 fuel consumption coefficients by their constant dollar output for 1993 to
1997. Until the results of the 1997 economic census are available, the post-1992 estimates of
fuel consumption are preliminary.
The Census of Mineral Industries provides information on the conversion factors used to
convert physical quantities of energy consumed for heat and power by the mining industries into
Btu. The conversion factors for natural gas, residual fuel oil, and distillate fuel oil from Monthly
Energy Review On-line (U.S. DOE 1999) are used to convert physical quantities fuel
consumption by the agriculture and construction sectors to Btu.
D.3. PROCESS EMISSIONS - INDUSTRIAL SECTOR
In the 1970-1997 NAPET, no industrial process emissions of PM-10 are reported from
1971 to 1974. As a result, PM-10 process emissions are estimated using the two-way indexing
method (see U.S. DOE 1992, p. 8 and Appendix A). This is described in section D.2. Industrial
January 2001
-------
Assigning Emissions to Industries in the United States, Revised Estimates D-3
process emissions for a missing year y, are estimated by the two-way indexing method, which
uses industrial process emissions for a "known" starting-year and a "known" ending-year. PM-
10 industrial process emissions in 1970 and 1975 are the "known" starting-year and ending-year
observations for the interpolation. These observations are adjusted by changes in the level of
production of the industry between the know-year and the missing-year. A weighted average (i.e.
a linear interpolation) of the two estimates of industrial process emissions in year y yields an
estimate of industrial process emissions in year y.
Emissions of the pollutant (p) generated by industrial process source category "c"
(EMITycp) for each missing year (y) can be specified as:
EM'T
L"
ALc)
where Ayc is the Federal Reserve Board production index for the two-digit SIC manufacturing
industry (Armitage and Tranum 1990) for year y, and Aos is the index/ output for the starting
year (y=0) and A^ is the index/ level of output for the ending year (y=L).
In addition to the missing PM-10 emissions for 1971 to 1974, there are other missing
observations for various Tier n categories. Most of the missing observations are from 1970
to!984. As a result, the two-way indexing method cannot be implemented. Instead, the
backward indexing method, in which 1985 is the known ending-year, is used to estimate
emissions for the preceding years:
Lc
D.4. FUEL COMBUSTION EMISSIONS - INDUSTRIAL SECTOR
The 19 70-199 7 NAPET report does not estimate the PM-10 emissions from industrial fuel
combustion for 1971 to 1974. The two-way indexing method is employed to estimating these
missing PM-10 emissions. In this case, PM-10 emissions in 1970 and 1975 are the "known"
starting-year and ending-year observations.
The "known" observations are adjusted by changes in the level of fuel consumption of the
industrial sector, which is reported by the NEA, between the know-year and the missing-year. A
weighted average (i.e., a linear interpolation) of the two estimates of industrial fuel combustion
emissions in year y yields an estimate of industrial fuel combustion emissions in year y. This
procedure is followed for each type of fuel. This is the same procedure used to estimate missing
fuel consumption observations for the manufacturing industries after 1985 (see section D. 1 .).
The two-way indexing method for estimating missing emissions of the pollutant (p)
generated by combustion of the ith type of fuel (EMITyp) by the industrial sector for each missing
year (y) can be specified as:
January 2001
-------
D-4 Assigning Emissions to Industries in the United States, Revised Estimates
EMIT,1
Lp
where Fy is the quantity of the ith fuel consumed for heat and power by the industrial sector in
year y, and Fg is the quantity of the ith fuel consumed by the industrial sector in the starting-year
(y=0) and FL is the quantity of the ith fuel consumed by the industrial sector in the ending-year
(y=L).
There is one difference between the "index" used to estimate fuel combustion emissions
and the one that estimates missing observations in sections D.I.through D.3.. Industrial fuel
combustion emissions are estimated using fuel consumed for heat and power as the activity
indicator; while fuel consumption and process emissions are derived using output as the activity
indicator.
January 2001
-------
APPENDIX E
EMISSIONS FROM TWENTY TWO-DIGIT SIC
MANUFACTURING INDUSTRIES*
* The highlighted cells in the tables of Appendix E represent emissions that include estimates of
the "NA" cells. Most of the highlighted cells represent cases when "NA" values for certain Tier
II categories of process emissions are estimated. The other cases involve estimates of PM-10
emissions from fuel combustion for 1971 to 1974, and the estimates of VOC emissions from
several Tier in categories (misc. Metal parts, steel drums, and electronic & other electrical)
within the Surface Coating (Tier n) category. Aside from the cases associated with estimating
PM-10 emissions from fuel combustion for 1971 to 1974, instances when the estimate of the
"NA" cell is less than 0.5 thousand short tons are noted at the bottom of each table.
-------
-------
E-2 Assigning Emissions to Industries in the United States, Revised Estimates
January 2001
-------
Assigning Emissions to Industries in the United States, Revised Estimates
E-l
Table E-l
Emissions from SIC 20 (Food and Kindred Products)
(thousand short tons)1
Year
1970
1971
1972
1973
1974
1975
1976
1977
1978
1979
1980
1981
1982
1983
1984
1985
1986
1987
1988
1989
1990
1991
1992
1993
1994
1995
1996
1997
PM-10
69.03
65.56
65.01
62.07
55.19
52.50
44.77
50.78
47.55
50.11
43.10
44.34
54.18
36.27
43.95
43.05
44.22
44.53
45.56
45.82
48.80
43.75
40.80
43.14
43.10
43.73
43.47
44.52
SO2
421.15
365.29
380.44
368.77
280.15
283.13
282.34
290.26
321.73
296.46
276.58
264.04
279.99
236.15
251.58
254.85
270.40
273.96
285.68
287.42
324.60
313.57
331.62
330.63
322.37
330.00
326.56
318.17
NOx
384.35
379.21
373.88
365.70
320.04
313.46
323.66
322.49
327.92
325.96
293.11
289.35
335.53
296.17
314.57
179.98
179.60
182.52
189.53
194.01
157.93
163.54
177.76
185.00
187.86
188.93
185.10
186.32
voc
216.15
216.78
214.21
209.02
206.12
188.32
190.07
190.29
194.42
196.77
197.28
203.29
200.72
197.40
174.36
174.87
177.78
180.85
183.40
181.21
144.05
135.70
132.47
152.15
151.87
154.17
142.23
148.46
CO
55.68
59.39
58.27
57.20
50.40
49.22
50.50
50.35
50.93
50.53
46.71
46.10
52.23
47.03
49.40
31.53
31.09
31.34
32.13
32.75
35.24
39.43
42.90
45.64
46.64
49.84
49.08
49.64
Pb
23.38
16.00
12.69
11.59
7.35
6.88
5.54
5.37
5.98
6.21
6.24
6.10
5.49
2.79
2.77
2.59
2.30
2.06
1.94
1.88
1.89
2.00
2.10
2.14
2.14
2.07
1.82
1.85
'except Pb, which is in short tons.
NOTE: the combined estimates of the "NA" cells for CO emissions from SIC 20 and the Agriculture sector for
1970 to 1984 never exceed 0.26 thousand short tons in any year
January 2001
-------
E-2
Assigning Emissions to Industries in the United States, Revised Estimates
Table E-2
Emissions from SIC 21 (Tobacco Products)
(thousand short tons)1
Year
1970
1971
1972
1973
1974
1975
1976
1977
1978
1979
1980
1981
1982
1983
1984
1985
1986
1987
1988
1989
1990
1991
1992
1993
1994
1995
1996
1997
PM-10
0.73
0.60
0.62
0.59
0.49
0.46
0.54
0.59
0.49
0.39
0.36
0.35
0.37
0.30
0.31
0.64
0.76
0.87
0.96
0.92
1.07
0.97
1.04
0.84
0.97
1.03
1.03
0.97
SO2
17.53
13.65
12.97
12.94
13.32
11.87
12.57
15.61
17.43
15.98
16.14
16.66
17.53
17.99
18.20
14.67
17.82
21.58
24.57
23.45
23.61
23.50
24.10
19.91
23.15
24.40
24.65
22.99
NOx
6.79
5.68
5.28
5.37
5.93
4.86
5.04
5.91
6.06
6.17
6.25
6.52
6.96
7.04
7.22
5.35
6.37
7.50
8.47
8.33
7.49
7.85
8.19
6.98
8.42
8.75
8.71
8.34
voc
0.09
0.08
0.08
0.08
0.09
0.07
0.07
0.08
0.07
0.07
0.07
0.07
0.08
0.08
0.07
0.09
0.11
0.12
0.12
0.12
0.12
0.11
0.13
0.10
0.14
0.12
0.12
0.12
CO
0.90
0.80
0.74
0.75
0.83
0.69
0.71
0.84
0.84
0.85
0.87
0.91
0.96
0.97
0.99
0.78
0.91
1.08
1.21
1.20
1.36
1.46
1.52
1.27
1.50
1.54
1.56
1.48
Pb
1.00
0.61
0.43
0.39
0.35
0.30
0.27
0.32
0.36
0.38
0.44
0.45
0.40
0.23
0.22
0.17
0.17
0.18
0.18
0.18
0.17
0.18
0.18
0.15
0.18
0.19
0.17
0.16
'except Pb, which is in short tons.
January 2001
-------
Assigning Emissions to Industries in the United States, Revised Estimates
E-3
Table E-3
Emissions from SIC 22 (Textile Mill Products)
(thousand short tons)1
Year
1970
1971
1972
1973
1974
1975
1976
1977
1978
1979
1980
1981
1982
1983
1984
1985
1986
1987
1988
1989
1990
1991
1992
1993
1994
1995
1996
1997
PM-10
8.49
7.71
7.83
6.57
4.28
4.52
5.52
6.20
5.62
4.50
3.96
3.45
3.33
2.44
2.48
4.16
4.53
4.40
4.37
4.19
4.57
3.75
4.14
4.51
4.61
4.71
4.61
4.55
SO2
166.43
154.15
143.50
127.37
104.85
101.94
116.34
125.45
115.74
103.55
91.95
92.19
88.67
82.06
86.43
78.25
85.55
85.51
84.87
80.71
83.63
74.42
81.03
88.13
89.58
92.30
90.75
87.31
NOx
97.23
99.45
95.53
89.04
78.08
72.28
75.44
76.42
75.16
79.79
69.79
71.27
71.50
69.26
73.52
43.71
45.36
45.82
45.81
46.08
37.12
36.71
39.32
42.95
43.98
42.91
41.74
41.95
voc
171.52
179.42
203.51
218.14
202.67
186.86
207.38
204.43
203.02
210.86
197.27
191.36
175.99
192.66
197.75
45.51
45.49
45.96
46.99
46.85
35.96
33.72
36.06
36.18
35.39
35.37
33.56
34.83
CO
13.44
14.99
14.47
13.54
11.89
11.06
11.40
11.56
11.39
11.97
10.77
11.08
10.80
10.69
11.23
7.27
7.45
7.54
7.49
7.58
7.87
8.48
9.12
10.10
10.38
10.32
10.11
10.25
Pb
8.26
6.00
4.15
3.43
2.55
2.38
2.32
2.35
2.18
2.16
2.00
2.12
1.78
1.06
1.03
0.84
0.79
0.71
0.60
0.52
0.47
0.45
0.50
0.58
0.62
0.56
0.48
0.51
'except Pb, which is in short tons.
NOTE: the combined estimates of the "NA" cells for PM-10 emissions from SIC 22, 23, and 31 for 1970 to 1984
never exceed 0.01 thousand short tons in any year
NOTE: the combined estimates of the "NA" cells for SO2 emissions from SIC 22, 23, and 31 for 1970 to 1984
never exceed 0.01 thousand short tons in any year
NOTE: the combined estimates of the "NA" cells for NOX emissions from SIC 22, 23, and 31 for 1970 to 1984
never exceed 0.06 thousand short tons in any year
January 2001
-------
E-4
Assigning Emissions to Industries in the United States, Revised Estimates
Table E-4
Emissions from SIC 23 (Apparel and Other Textile Products)
(thousand short tons)1
Year
1970
1971
1972
1973
1974
1975
1976
1977
1978
1979
1980
1981
1982
1983
1984
1985
1986
1987
1988
1989
1990
1991
1992
1993
1994
1995
1996
1997
PM-10
1.04
0.91
0.82
0.66
0.48
0.55
0.64
0.74
0.69
0.56
0.53
0.52
0.48
0.38
0.37
0.38
0.41
0.50
0.56
0.56
0.56
0.52
0.54
0.55
0.55
0.58
0.57
0.54
SO2
21.37
18.65
16.06
12.23
8.61
9.55
10.36
11.83
11.14
9.18
9.03
9.90
8.63
7.20
6.74
5.24
5.56
6.86
8.29
8.36
8.20
7.73
8.22
8.29
8.11
8.64
8.39
7.66
NOx
20.33
19.93
18.65
16.64
14.79
13.76
13.47
15.03
16.58
14.37
12.58
13.10
12.06
10.54
10.87
4.10
4.60
5.75
7.04
6.94
4.90
4.77
5.14
5.42
5.74
5.64
5.43
5.23
voc
0.42
0.44
0.42
0.39
0.36
0.32
0.31
0.35
0.39
0.33
0.29
0.29
0.27
0.24
0.25
0.17
0.22
0.25
0.30
0.28
0.22
0.22
0.21
0.21
0.27
0.29
0.28
0.28
CO
3.08
3.27
3.07
2.78
2.50
2.25
2.18
2.44
2.69
2.31
2.08
2.17
1.93
1.72
1.77
0.77
0.87
1.06
1.28
1.25
1.12
1.20
1.31
1.43
1.57
1.59
1.54
1.50
Pb
1.18
0.80
0.49
0.33
0.18
0.20
0.18
0.19
0.18
0.16
0.17
0.20
0.16
0.09
0.07
0.05
0.04
0.05
0.05
0.05
0.04
0.04
0.04
0.05
0.05
0.04
0.03
0.04
'except Pb, which is in short tons.
NOTE: see the notes at the bottom of Table E-3
January 2001
-------
Assigning Emissions to Industries in the United States, Revised Estimates
E-5
Table E-5
Emissions from SIC 24 (Lumber and Wood Products)
(thousand short tons)1
Year
1970
1971
1972
1973
1974
1975
1976
1977
1978
1979
1980
1981
1982
1983
1984
1985
1986
1987
1988
1989
1990
1991
1992
1993
1994
1995
1996
1997
PM-10
165.26
166.13
179.88
186.05
174.20
158.66
180.46
181.57
192.80
190.48
184.27
171.79
156.48
164.44
172.37
57.89
60.93
62.71
63.94
62.65
65.14
55.39
56.84
57.87
53.48
54.69
55.72
57.76
SO2
42.33
35.56
39.76
41.78
36.25
28.86
35.40
29.08
30.90
28.96
24.23
19.59
15.01
11.82
12.91
18.82
18.91
16.70
14.29
14.40
19.46
17.31
18.33
21.03
22.32
22.95
23.19
23.03
NOx
95.98
90.69
91.35
92.62
83.53
78.46
83.03
77.09
82.60
82.07
67.82
60.98
53.66
54.18
54.95
49.65
51.90
52.63
51.56
52.69
46.93
44.62
45.09
45.75
46.18
46.55
47.65
49.05
voc
74.74
84.39
92.58
92.74
89.40
86.28
97.25
100.33
103.40
104.94
93.55
80.58
66.67
67.66
68.55
17.70
18.95
19.05
19.00
18.69
38.32
36.29
40.04
43.33
42.78
47.99
49.42
51.33
CO
10.88
10.94
10.62
10.70
9.23
8.65
8.54
7.47
7.86
7.69
5.71
4.83
3.99
3.53
3.55
2.97
3.12
3.18
2.95
3.12
3.78
3.51
3.76
4.48
4.89
5.13
5.22
5.32
Pb
1.70
1.07
0.92
0.94
0.76
0.49
0.55
0.43
0.46
0.44
0.36
0.29
0.22
0.15
0.15
0.16
0.16
0.13
0.09
0.08
0.07
0.06
0.07
0.08
0.11
0.11
0.10
0.11
'except Pb, which is in short tons.
January 2001
-------
E-f
Assigning Emissions to Industries in the United States, Revised Estimates
Table E-6
Emissions from SIC 25 (Furniture and Fixtures)
(thousand short tons)1
Year
1970
1971
1972
1973
1974
1975
1976
1977
1978
1979
1980
1981
1982
1983
1984
1985
1986
1987
1988
1989
1990
1991
1992
1993
1994
1995
1996
1997
PM-10
1.08
0.80
0.85
0.58
0.43
0.39
0.39
0.41
0.39
0.35
0.34
0.29
0.27
0.18
0.20
1.42
1.45
1.56
1.56
2.16
2.85
2.92
2.48
2.17
1.45
1.47
1.50
1.55
SO2
29.09
20.28
20.84
14.04
11.41
10.19
8.97
10.54
10.28
9.23
8.27
8.15
7.30
5.50
6.03
5.77
5.86
7.35
7.40
8.80
10.19
9.30
8.16
8.16
6.33
6.36
6.40
6.47
NOx
15.97
16.88
18.75
16.07
15.95
12.68
12.90
14.47
15.23
13.73
11.22
10.77
11.65
10.27
11.70
7.27
7.09
8.23
8.68
9.73
8.14
8.38
7.63
7.40
6.47
6.53
6.57
6.77
voc
251.31
314.48
354.13
334.81
307.35
277.98
294.72
285.71
301.38
316.17
290.64
249.37
223.44
246.29
261.26
182.52
192.83
195.12
196.01
193.26
225.92
219.23
227.87
245.52
261.77
256.64
276.80
289.22
CO
2.26
2.63
2.95
2.58
2.59
2.04
2.07
2.33
2.45
2.22
1.86
1.78
1.88
1.69
1.91
1.11
1.07
1.27
1.34
1.38
1.31
1.29
1.27
1.39
1.38
1.42
1.42
1.47
Pb
1.77
0.98
0.72
0.43
0.29
0.24
0.16
0.19
0.18
0.17
0.17
0.18
0.13
0.06
0.06
0.05
0.04
0.05
0.05
0.05
0.05
0.06
0.05
0.05
0.04
0.04
0.03
0.03
'except Pb, which is in short tons.
January 2001
-------
Assigning Emissions to Industries in the United States, Revised Estimates
E-7
Table E-7
Emissions from SIC 26 (Paper and Allied Products)
(thousand short tons)1
Year
1970
1971
1972
1973
1974
1975
1976
1977
1978
1979
1980
1981
1982
1983
1984
1985
1986
1987
1988
1989
1990
1991
1992
1993
1994
1995
1996
1997
PM-10
1022.28
963.61
941.76
883.39
762.59
590.56
599.05
614.54
591.60
591.86
596.52
547.51
543.20
538.08
552.42
137.22
142.80
141.48
143.25
142.71
154.75
122.36
121.49
123.32
121.78
126.85
127.39
131.51
SO2
1018.59
1001.03
1016.50
974.60
1002.22
869.68
943.08
960.77
918.39
897.50
897.66
860.03
900.00
859.62
948.18
767.79
824.14
812.51
810.81
811.77
869.16
823.93
844.35
827.76
801.92
852.11
845.98
830.10
NOx
532.50
545.99
526.37
500.88
521.30
464.54
499.65
499.00
494.12
515.18
473.67
484.87
510.68
499.58
538.99
402.42
420.28
415.38
421.48
428.31
416.09
409.89
420.20
422.93
426.88
425.78
420.34
432.24
voc
672.33
666.61
753.76
754.93
674.63
568.19
599.34
606.45
692.34
695.90
649.72
505.11
461.85
529.01
658.53
134.00
138.49
138.98
143.85
143.68
131.16
119.46
120.80
124.69
122.10
123.88
124.58
130.61
CO
668.85
671.14
714.03
734.66
737.10
655.67
744.92
752.17
773.06
784.19
848.27
845.68
832.53
888.96
935.59
670.66
692.65
691.06
694.15
701.49
527.01
519.57
509.41
515.00
524.04
547.74
556.00
579.45
Pb
41.67
32.32
24.61
22.00
21.08
17.10
15.57
14.97
14.04
14.89
14.77
14.87
14.05
8.41
8.67
7.18
6.60
5.84
4.96
4.42
4.35
4.40
4.71
5.00
5.01
4.55
3.80
4.22
'except Pb, which is in short tons.
January 2001
-------
E-S
Assigning Emissions to Industries in the United States, Revised Estimates
Table E-8
Emissions from SIC 27 (Printing and Publishing)
(thousand short tons)1
Year
1970
1971
1972
1973
1974
1975
1976
1977
1978
1979
1980
1981
1982
1983
1984
1985
1986
1987
1988
1989
1990
1991
1992
1993
1994
1995
1996
1997
PM-10
0.83
0.62
0.65
0.57
0.38
0.44
0.35
0.43
0.42
0.43
0.42
0.38
0.45
0.31
0.31
0.19
0.31
0.26
0.28
0.20
0.25
0.21
0.22
0.23
0.23
0.22
0.22
0.23
SO2
10.19
9.19
9.94
10.23
8.70
9.14
7.59
8.33
8.15
8.33
7.24
6.24
7.13
5.65
6.30
3.72
5.55
4.22
4.42
3.05
4.95
4.81
4.89
4.81
4.59
4.55
4.47
4.60
NOx
25.91
25.04
24.43
23.04
18.03
18.95
16.27
20.05
19.91
21.05
18.36
18.80
22.74
20.41
24.17
7.50
8.48
9.34
10.80
10.57
7.22
7.64
7.64
7.56
7.38
7.27
7.17
7.37
voc
319.57
301.54
344.51
351.74
325.61
254.95
280.99
292.49
346.78
353.51
373.90
257.86
238.26
274.50
360.59
317.51
325.33
339.96
362.17
363.54
274.92
301.90
308.82
322.43
333.28
339.37
389.39
412.11
CO
3.89
4.10
3.99
3.78
2.96
3.09
2.64
3.26
3.23
3.42
3.10
3.18
3.73
3.41
3.99
1.49
1.68
1.85
2.12
2.08
1.83
2.20
2.24
2.31
2.34
2.37
2.34
2.42
Pb
0.26
0.16
0.19
0.22
0.19
0.18
0.12
0.12
0.12
0.12
0.10
0.08
0.09
0.05
0.05
0.02
0.03
0.02
0.01
0.00
0.00
0.00
0.00
0.00
0.01
0.01
0.01
0.01
'except Pb, which is in short tons.
January 2001
-------
Assigning Emissions to Industries in the United States, Revised Estimates
E-9
Table E-9
Emissions from SIC 28 (Chemicals and Allied Products)
(thousand short tons)1
Year
1970
1971
1972
1973
1974
1975
1976
1977
1978
1979
1980
1981
1982
1983
1984
1985
1986
1987
1988
1989
1990
1991
1992
1993
1994
1995
1996
1997
PM-10
281.15
265.39
259.29
241.22
209.04
154.74
164.60
177.62
188.84
178.16
174.45
156.04
130.98
140.72
163.94
86.74
88.24
87.83
92.78
92.66
109.89
94.99
99.47
95.11
105.09
96.93
97.45
101.25
SO2
1753.36
1664.10
1659.41
1598.19
1468.30
1216.40
1128.50
1170.70
1150.34
1129.98
996.58
1010.82
924.04
894.93
946.27
1028.16
1010.63
1013.09
1051.55
1017.84
921.38
842.69
858.93
853.51
848.34
880.09
889.57
900.71
NOx
1386.38
1350.70
1390.62
1441.99
1516.83
1353.20
1458.21
1433.14
1384.19
1420.01
1236.25
1241.83
1189.79
1090.38
1198.28
700.04
696.73
727.38
779.61
791.54
576.26
587.22
604.60
608.61
616.84
615.58
619.40
641.69
voc
1702.81
1695.21
1817.68
1923.24
1910.52
1656.87
1893.46
2011.77
2105.71
2165.38
1996.74
1907.47
1647.69
1890.52
2055.56
954.02
994.47
1001.80
1064.96
1062.00
708.10
789.63
796.89
784.30
780.10
749.83
509.92
535.60
CO
3561.66
3338.41
2756.39
2904.41
2703.36
2383.68
2551.38
2832.94
2729.51
2730.12
2316.74
2374.99
1865.93
1995.51
2246.99
1917.07
1924.22
1875.59
2000.59
2010.72
1263.92
1220.59
1212.22
1199.64
1282.43
1335.97
1337.40
1405.85
Pb
172.90
167.49
157.26
165.39
173.26
142.08
141.74
151.11
157.18
161.33
120.30
144.83
134.40
146.16
140.63
124.17
113.17
127.16
139.75
139.34
139.09
135.52
96.69
95.40
99.42
166.28
169.94
162.09
'except Pb, which is in short tons.
NOTE: the estimates of the "NA" cells of SO2 emissions for 1985 to 1991 never exceed 0.01 thousand short tons in
any year
NOTE: the estimates of the "NA" cells of CO emissions for 1985 to 1989 never exceed 0.01 thousand short tons in
any year
January 2001
-------
E-10
Assigning Emissions to Industries in the United States, Revised Estimates
Table E-10
Emissions from SIC 29 (Petroleum and Coal Products)
(thousand short tons)1
Year
1970
1971
1972
1973
1974
1975
1976
1977
1978
1979
1980
1981
1982
1983
1984
1985
1986
1987
1988
1989
1990
1991
1992
1993
1994
1995
1996
1997
PM-10
289.98
273.96
260.57
247.08
211.38
182.36
157.36
148.30
151.97
158.28
141.63
125.33
114.74
122.46
126.69
91.81
84.30
82.67
81.66
79.10
68.50
54.55
55.29
51.17
51.09
53.89
53.98
55.82
SO2
985.04
671.76
707.58
778.02
774.96
750.67
752.08
786.80
782.99
765.29
729.26
663.56
640.31
627.77
635.15
579.16
466.74
461.49
470.36
453.07
482.82
437.78
466.02
427.05
419.42
423.36
424.49
433.49
NOx
461.94
280.81
287.53
299.03
301.02
292.34
302.04
321.05
321.77
310.26
292.38
258.08
249.68
241.23
244.00
462.00
425.46
407.77
416.88
416.57
345.10
369.39
400.57
374.50
367.38
359.24
360.09
370.83
voc
2041.67
2004.65
2043.75
2037.01
1945.44
1927.22
1764.66
1822.83
1773.64
1746.59
1612.61
1471.25
1401.73
1297.32
1254.08
953.52
922.64
935.18
938.92
937.46
636.93
698.00
707.72
718.26
721.25
727.68
485.38
507.73
CO
2183.00
2252.00
2291.71
2327.27
2245.94
2224.96
2135.36
2032.37
1930.67
1838.11
1734.47
1202.35
750.26
491.22
388.76
487.56
474.80
477.22
464.96
462.43
325.86
365.62
388.91
388.61
342.86
354.13
354.29
369.90
Pb
2.15
1.14
0.83
0.83
0.72
0.59
0.41
0.37
0.42
0.52
0.44
0.36
0.45
0.32
0.35
1.65
0.45
0.49
0.48
0.35
0.31
0.27
0.32
0.38
0.40
0.33
0.26
0.33
'except Pb, which is in short tons.
January 2001
-------
Assigning Emissions to Industries in the United States, Revised Estimates
E-ll
Table E-ll
Emissions from SIC 30 (Rubber and Miscellaneous Plastic Products)
(thousand short tons)1
Year
1970
1971
1972
1973
1974
1975
1976
1977
1978
1979
1980
1981
1982
1983
1984
1985
1986
1987
1988
1989
1990
1991
1992
1993
1994
1995
1996
1997
PM-10
5.91
5.47
7.01
6.88
5.51
4.90
5.84
6.74
6.15
5.46
4.69
4.29
4.31
4.02
4.41
4.98
5.12
5.26
5.38
5.25
5.52
5.25
5.71
5.10
4.85
4.91
4.95
5.09
SO2
110.43
95.19
120.93
119.16
104.76
84.87
85.81
94.51
79.34
69.08
57.56
49.75
44.96
33.68
32.50
31.70
32.74
33.30
31.52
31.76
36.66
31.05
30.60
31.87
29.81
33.26
33.99
32.46
NOx
66.11
68.94
80.78
76.68
69.62
60.01
61.68
68.33
65.67
63.63
51.67
53.08
57.39
51.36
55.77
28.88
28.11
30.36
30.39
30.67
22.60
21.51
21.56
23.17
22.97
23.18
23.53
23.92
voc
77.52
79.71
86.74
84.72
81.65
70.76
73.91
89.02
88.57
85.18
70.27
78.01
74.99
80.11
90.63
53.50
55.06
55.63
57.86
57.71
85.79
82.00
87.18
84.68
84.57
78.88
78.80
82.44
CO
9.41
10.68
12.36
11.66
10.56
9.23
9.37
10.50
10.16
9.79
8.19
8.54
9.04
8.33
9.00
5.37
5.19
5.59
5.57
5.59
5.19
5.48
5.59
6.23
6.45
6.68
6.79
7.00
Pb
6.38
4.40
4.13
3.79
2.91
2.11
1.73
1.77
1.47
1.40
1.22
1.09
0.82
0.41
0.36
0.28
0.27
0.25
0.19
0.16
0.15
0.14
0.14
0.16
0.16
0.13
0.11
0.13
'except Pb, which is in short tons.
NOTE: the estimates of the "NA" cells of NOX emissions for 1970 to 1984 never exceed 0.04 thousand short tons in
any year
NOTE: the estimates of the "NA" cells of CO emissions for 1970 to 1984 never exceed 0.07 thousand short tons in
any year
January 2001
-------
E-12
Assigning Emissions to Industries in the United States, Revised Estimates
Table E-12
Emissions from SIC 31 (Leather and Leather Products)
(thousand short tons)1
Year
1970
1971
1972
1973
1974
1975
1976
1977
1978
1979
1980
1981
1982
1983
1984
1985
1986
1987
1988
1989
1990
1991
1992
1993
1994
1995
1996
1997
PM-10
1.07
0.77
0.73
0.54
0.39
0.41
0.38
0.49
0.41
0.36
0.38
0.33
0.36
0.26
0.24
0.26
0.37
0.34
0.33
0.35
0.31
0.18
0.17
0.26
0.22
0.23
0.23
0.20
SO2
19.30
14.50
13.19
10.39
8.19
7.49
6.68
7.98
7.14
6.14
6.14
5.37
5.58
4.13
3.96
4.44
6.13
5.81
5.58
5.77
5.22
3.10
3.00
4.21
3.55
3.82
3.82
3.17
NOx
10.70
8.52
7.59
6.16
5.24
5.39
5.34
5.59
5.28
5.22
4.35
4.07
4.45
4.01
3.57
2.26
2.70
2.78
2.75
3.00
2.19
1.58
1.57
1.87
1.62
1.50
1.49
1.33
voc
0.18
0.16
0.14
0.12
0.10
0.11
0.11
0.11
0.11
0.11
0.09
0.08
0.09
0.08
0.08
0.08
0.11
0.11
0.11
0.12
0.09
0.08
0.08
0.08
0.08
0.08
0.08
0.07
CO
1.51
1.32
1.17
0.95
0.81
0.85
0.85
0.86
0.82
0.80
0.67
0.64
0.68
0.63
0.55
0.38
0.43
0.46
0.45
0.49
0.52
0.42
0.43
0.51
0.46
0.44
0.44
0.41
Pb
0.90
0.54
0.38
0.28
0.18
0.16
0.12
0.14
0.13
0.12
0.11
0.10
0.10
0.06
0.06
0.05
0.06
0.06
0.04
0.03
0.02
0.01
0.01
0.02
0.02
0.02
0.01
0.01
'except Pb, which is in short tons.
NOTE: see the notes at the bottom of Table E-3
January 2001
-------
Assigning Emissions to Industries in the United States, Revised Estimates
E-13
Table E-13
Emissions from SIC 32 (Stone, Clay, and Glass Products)
(thousand short tons)1
Year
1970
1971
1972
1973
1974
1975
1976
1977
1978
1979
1980
1981
1982
1983
1984
1985
1986
1987
1988
1989
1990
1991
1992
1993
1994
1995
1996
1997
PM-10
1737.06
1596.45
1514.73
1361.07
1067.45
706.75
668.43
677.95
689.31
573.92
419.72
323.02
249.36
270.16
304.13
233.21
233.53
223.94
216.51
213.20
213.84
167.34
164.49
159.11
157.03
158.13
156.39
161.20
SO2
744.97
741.66
766.06
767.87
730.84
627.99
699.17
788.16
834.16
813.54
735.48
702.05
614.08
650.61
703.98
755.15
752.47
724.51
719.66
732.04
797.01
742.48
711.36
708.65
708.36
710.09
738.53
748.76
NOx
339.42
350.52
364.40
367.88
344.98
304.28
340.71
349.56
364.78
361.51
307.16
287.85
260.35
269.47
285.11
473.79
472.87
461.38
466.35
468.87
481.07
448.28
450.06
464.63
481.92
487.84
500.50
516.82
voc
5.63
5.90
6.10
6.09
5.70
5.20
5.95
6.00
6.23
6.13
5.11
4.76
4.47
4.46
4.59
21.12
21.69
21.18
21.47
20.99
23.95
22.37
31.56
32.51
36.15
37.52
38.00
39.31
CO
35.79
54.87
56.71
62.49
56.39
50.75
55.59
61.82
64.92
64.86
53.56
47.10
41.34
45.74
49.19
81.71
83.01
82.15
83.85
84.06
97.13
118.96
125.72
174.17
174.51
171.68
175.56
182.09
Pb
541.69
484.63
410.28
262.31
198.58
217.85
200.53
198.67
209.43
121.11
93.21
53.76
39.98
44.36
47.73
48.22
29.40
31.03
25.96
26.13
29.09
27.51
29.19
30.15
31.13
32.13
32.02
33.01
'except Pb, which is in short tons.
January 2001
-------
E-14
Assigning Emissions to Industries in the United States, Revised Estimates
Table E-14
Emissions from SIC 33 (Primary Metal Industries)
(thousand short tons)1
Year
1970
1971
1972
1973
1974
1975
1976
1977
1978
1979
1980
1981
1982
1983
1984
1985
1986
1987
1988
1989
1990
1991
1992
1993
1994
1995
1996
1997
PM-10
812.08
720.88
750.64
787.45
701.76
512.58
687.18
613.10
608.65
562.63
457.81
450.96
313.62
274.00
297.80
227.97
209.30
200.15
214.78
217.66
219.61
181.92
175.67
175.54
179.83
206.25
205.96
213.91
SO2
5357.69
4673.81
5015.66
5018.89
4324.95
3171.50
2811.17
2672.97
2484.55
2588.71
2051.81
2256.10
1586.08
1522.37
1552.31
1230.63
1054.40
808.51
865.15
847.52
905.63
767.49
775.97
772.83
736.39
720.56
723.99
740.95
NOx
492.94
455.22
467.43
480.09
447.94
376.30
390.20
393.08
423.36
396.59
325.50
313.69
260.20
250.12
263.65
305.54
268.43
273.16
293.96
296.13
269.10
244.21
252.44
265.94
281.61
290.35
292.05
302.09
voc
447.25
335.23
417.92
441.09
422.53
386.02
399.65
374.09
353.91
375.44
323.64
302.09
208.84
201.82
227.90
132.59
128.76
125.21
131.21
130.49
168.15
174.45
170.34
171.80
174.42
174.00
121.68
127.04
CO
3701.74
3607.62
3753.48
3707.86
3482.43
2540.81
2692.71
2583.28
2662.17
2650.08
2284.95
2254.41
1550.94
1589.59
1764.83
2253.65
2105.63
2012.23
2130.52
2161.26
2632.14
2581.65
2506.53
2552.73
2494.46
2400.58
2399.25
2487.61
Pb
23907.68
22484.45
17804.67
14492.21
12582.17
9664.46
6859.53
4411.74
4284.34
3881.41
2823.62
2423.35
2153.92
1880.30
1791.44
1957.64
1715.39
1704.84
1803.30
1919.82
1986.46
1777.08
1574.26
1707.93
1829.01
1866.92
1863.78
1850.88
'except Pb, which is in short tons.
January 2001
-------
Assigning Emissions to Industries in the United States, Revised Estimates
E-15
Table E-15
Emissions from SIC 34 (Fabricated Metal Products)
(thousand short tons)1
Year
1970
1971
1972
1973
1974
1975
1976
1977
1978
1979
1980
1981
1982
1983
1984
1985
1986
1987
1988
1989
1990
1991
1992
1993
1994
1995
1996
1997
PM-10
12.11
10.98
11.63
11.68
10.11
9.00
9.87
10.63
10.72
10.69
9.84
9.53
8.96
8.58
9.24
9.58
9.80
10.04
10.38
10.59
10.84
9.75
10.28
8.43
8.36
8.43
8.46
8.91
SO2
102.54
88.66
86.13
75.60
58.20
51.23
52.84
55.03
51.68
49.04
41.63
39.04
41.51
34.04
33.76
30.26
31.56
30.06
31.58
28.70
35.77
28.82
30.10
30.95
30.69
32.27
32.83
32.91
NOx
152.38
145.08
151.10
145.97
125.79
116.26
124.61
125.14
129.20
122.03
101.14
100.83
113.98
96.80
103.11
45.84
44.59
46.07
50.48
49.52
35.89
32.56
34.87
37.99
43.07
44.30
44.94
46.22
voc
67.97
75.30
80.92
83.00
80.41
76.44
86.24
90.86
96.93
102.40
95.34
85.98
81.71
87.99
87.57
119.31
117.50
116.78
118.06
117.38
164.91
191.44
197.60
200.86
207.23
210.72
215.46
224.28
CO
22.49
23.21
24.15
23.48
20.26
18.72
19.89
20.02
20.67
19.54
16.76
16.68
18.34
15.86
16.75
8.79
8.51
8.66
9.40
9.20
8.18
8.81
9.42
10.48
11.38
11.84
12.04
12.42
Pb
4.78
3.10
2.26
1.85
1.21
0.98
0.82
0.83
0.75
0.78
0.66
0.65
0.63
0.33
0.29
0.20
0.20
0.17
0.15
0.13
0.12
0.09
0.09
0.10
0.10
0.10
0.09
0.09
'except Pb, which is in short tons.
NOTE: the estimates of the "NA" cells of PM-10 emissions for 1970 to 1984 never exceed 0.11 thousand short tons
in any year
NOTE: the estimates of the "NA" cells of SO2 emissions for 1970 to 1984 never exceed 0.27 thousand short tons in
any year
January 2001
-------
E-16
Assigning Emissions to Industries in the United States, Revised Estimates
Table E-16
Emissions from SIC 35 (Industrial Machinery Equipment)
(thousand short tons)1
Year
1970
1971
1972
1973
1974
1975
1976
1977
1978
1979
1980
1981
1982
1983
1984
1985
1986
1987
1988
1989
1990
1991
1992
1993
1994
1995
1996
1997
PM-10
5.37
4.52
4.55
3.91
2.72
2.55
2.62
2.62
2.57
2.27
2.03
1.91
2.02
1.34
1.37
1.91
1.86
1.77
1.80
1.71
1.84
1.39
1.38
1.40
1.46
1.69
1.86
2.02
SO2
114.56
100.76
101.69
92.71
73.37
68.40
60.18
62.40
66.70
66.39
57.22
53.36
52.85
41.58
43.32
39.28
38.44
37.92
38.76
35.77
38.08
30.97
30.04
29.83
30.46
35.73
39.29
41.93
NOx
116.98
115.12
118.74
117.13
104.56
93.32
94.80
94.42
99.96
97.31
80.69
78.42
87.33
68.70
72.47
33.46
31.96
34.06
37.14
36.67
25.67
23.43
22.68
22.88
23.65
26.68
29.15
32.22
voc
41.69
46.59
50.32
52.95
53.38
53.12
62.34
64.53
68.18
70.50
64.24
55.49
63.84
54.96
52.84
38.38
29.75
29.84
30.10
29.76
29.29
27.27
27.29
27.44
28.16
26.75
28.00
29.44
CO
17.09
18.19
18.73
18.62
16.68
14.87
15.03
15.02
15.84
15.43
13.21
12.83
13.94
11.17
11.66
6.22
5.92
6.30
6.86
6.73
6.02
6.05
5.94
6.19
6.52
7.53
8.26
9.19
Pb
6.14
4.32
3.29
2.81
1.89
1.63
1.11
1.11
1.18
1.33
1.23
1.15
0.96
0.48
0.46
0.37
0.31
0.27
0.24
0.21
0.18
0.16
0.15
0.16
0.17
0.18
0.17
0.20
'except Pb, which is in short tons.
January 2001
-------
Assigning Emissions to Industries in the United States, Revised Estimates
E-17
Table E-17
Emissions from SIC 36 (Electric and Electronic Equipment)
(thousand short tons)1
Year
1970
1971
1972
1973
1974
1975
1976
1977
1978
1979
1980
1981
1982
1983
1984
1985
1986
1987
1988
1989
1990
1991
1992
1993
1994
1995
1996
1997
PM-10
3.90
3.22
3.33
2.89
1.87
1.68
1.86
1.90
1.89
1.70
1.53
1.43
1.59
1.14
1.18
1.35
1.39
1.17
1.27
1.26
1.43
1.21
1.33
1.27
1.36
1.72
2.11
2.50
SO2
80.12
65.48
69.69
67.73
52.63
42.42
44.42
45.57
45.28
45.22
35.54
34.92
36.12
30.46
30.63
25.02
25.91
21.91
24.01
23.85
27.02
25.66
27.83
27.05
28.76
36.73
45.10
52.82
NOx
79.37
77.30
79.59
76.98
64.23
56.79
61.42
62.25
64.38
62.79
53.03
52.08
60.65
53.40
56.12
23.96
23.31
21.89
25.59
24.03
17.70
17.91
18.77
19.08
20.39
25.23
30.72
37.17
voc
80.90
91.93
101.93
101.53
91.32
78.12
82.48
85.30
90.61
96.19
93.57
89.49
87.00
94.32
108.64
102.51
99.39
99.31
99.56
98.60
101.88
97.79
100.48
104.19
109.70
108.64
121.75
128.30
CO
18.43
19.14
20.45
21.16
19.08
16.69
18.31
19.84
21.23
22.17
21.38
21.82
23.14
23.27
26.43
22.31
22.29
21.58
17.49
16.46
5.56
6.00
6.23
6.47
6.87
8.27
9.78
11.62
Pb
4.18
2.54
2.16
2.05
1.41
0.99
0.84
0.82
0.79
0.89
0.70
0.71
0.63
0.35
0.32
0.22
0.20
0.15
0.14
0.13
0.12
0.13
0.15
0.15
0.17
0.21
0.22
0.28
'except Pb, which is in short tons.
NOTE: the estimates of the "NA" cells of PM-10 emissions for 1970 to 1984 never exceed 0.08 thousand short tons
in any year
NOTE: the estimates of the "NA" cells of SO2 emissions for 1970 to 1989 never exceed 0.45 thousand short tons in
any year
NOTE: the estimates of the "NA" cells of NOX emissions for 1970 to 1984 never exceed 0.07 thousand short tons in
any year
January 2001
-------
E-18
Assigning Emissions to Industries in the United States, Revised Estimates
Table E-18
Emissions from SIC 37 (Transportation Equipment)
(thousand short tons)1
Year
1970
1971
1972
1973
1974
1975
1976
1977
1978
1979
1980
1981
1982
1983
1984
1985
1986
1987
1988
1989
1990
1991
1992
1993
1994
1995
1996
1997
PM-10
8.56
7.80
7.98
7.23
5.47
4.81
5.33
5.55
5.67
5.35
4.87
4.46
4.40
3.84
4.14
6.25
6.44
6.04
6.08
6.08
6.69
5.77
5.67
3.86
3.59
3.61
3.62
3.77
SO2
184.46
164.54
169.86
153.31
132.27
120.12
119.54
116.26
120.07
117.95
99.65
99.53
94.84
87.26
92.76
86.93
91.46
85.79
83.69
83.30
92.01
79.64
75.25
74.02
67.29
68.43
68.91
70.37
NOx
118.00
118.10
120.73
113.71
105.22
97.67
109.54
106.92
111.63
105.12
81.92
80.54
83.64
80.49
86.00
53.09
54.34
52.91
53.58
54.71
43.65
41.24
40.82
42.43
41.28
40.04
39.95
42.49
voc
193.63
216.79
232.07
238.07
236.81
233.86
261.63
268.36
258.56
243.08
197.68
146.64
126.25
148.46
169.81
132.66
136.55
134.82
139.22
142.29
120.70
118.44
121.54
121.46
125.54
123.81
148.46
158.06
CO
16.76
18.06
18.39
17.39
16.14
15.02
16.81
16.46
17.17
16.09
12.82
12.61
12.81
12.57
13.29
9.21
9.36
9.13
9.20
9.36
9.52
9.73
9.80
10.54
10.58
10.47
10.50
11.26
Pb
10.97
8.11
6.23
5.22
3.94
3.24
2.57
2.34
2.34
2.62
2.37
2.44
1.95
1.08
1.07
0.92
0.82
0.69
0.58
0.52
0.50
0.47
0.45
0.46
0.43
0.39
0.34
0.38
'except Pb, which is in short tons.
NOTE: the estimates of the "NA" cells of NOX emissions for 1970 to 1984 never exceed 0.01 thousand short tons in
any year
NOTE: the estimates of the "NA" cells of CO emissions for 1970 to 1984 never exceed 0.01 thousand short tons in
any year
January 2001
-------
Assigning Emissions to Industries in the United States, Revised Estimates
E-19
Table E-19
Emissions from SIC 38 (Instruments and Related Products)
(thousand short tons)1
Year
1970
1971
1972
1973
1974
1975
1976
1977
1978
1979
1980
1981
1982
1983
1984
1985
1986
1987
1988
1989
1990
1991
1992
1993
1994
1995
1996
1997
PM-10
1.87
1.92
2.08
1.63
1.31
1.13
1.10
1.11
1.01
0.96
0.91
0.84
0.94
0.71
0.76
1.61
1.62
1.62
1.63
1.71
2.00
1.72
1.82
1.85
1.74
1.82
1.86
1.84
SO2
46.19
48.18
52.86
43.78
42.67
35.82
33.45
22.92
25.56
41.51
37.16
35.20
38.44
36.78
38.96
35.92
36.33
37.57
38.15
39.42
42.15
40.26
41.21
41.83
39.26
40.93
42.31
41.01
NOx
21.73
22.87
24.03
20.65
20.10
18.40
20.25
18.71
18.52
22.77
20.34
20.00
23.17
22.12
24.37
15.78
15.86
17.61
17.49
18.95
15.87
15.70
16.55
17.14
16.72
16.93
17.27
17.31
voc
0.33
0.36
0.37
0.32
0.31
0.29
0.35
0.38
0.35
0.36
0.32
0.32
0.38
0.36
0.39
0.38
0.40
0.48
0.46
0.50
0.43
0.38
0.41
0.38
0.44
0.45
0.45
0.46
CO
2.99
3.32
3.46
2.97
2.86
2.68
3.00
2.87
2.81
3.32
3.04
3.00
3.41
3.30
3.61
2.47
2.47
2.84
2.77
3.02
3.19
3.26
3.47
3.58
3.49
3.56
3.67
3.68
Pb
2.69
2.35
1.94
1.48
1.26
0.96
0.73
0.42
0.49
0.98
0.97
0.92
0.84
0.46
0.46
0.41
0.34
0.30
0.28
0.27
0.27
0.29
0.29
0.30
0.29
0.29
0.26
0.27
'except Pb, which is in short tons.
January 2001
-------
E-20
Assigning Emissions to Industries in the United States, Revised Estimates
Table E-20
Emissions from SIC 39 (Miscellaneous Manufacturing Products)
(thousand short tons)1
Year
1970
1971
1972
1973
1974
1975
1976
1977
1978
1979
1980
1981
1982
1983
1984
1985
1986
1987
1988
1989
1990
1991
1992
1993
1994
1995
1996
1997
PM-10
1.02
0.94
1.08
0.83
0.53
0.51
0.59
0.51
0.48
0.39
0.36
0.31
0.30
0.23
0.24
0.28
0.30
0.34
0.28
0.28
0.27
0.24
0.24
0.25
0.24
0.25
0.25
0.25
SO2
15.91
16.01
17.36
14.30
10.99
8.32
10.02
8.83
8.45
7.57
7.14
6.88
6.20
5.78
5.54
5.10
5.45
6.52
5.59
5.57
5.33
4.92
5.06
5.01
4.64
5.00
5.10
4.89
NOx
14.67
15.27
16.96
14.90
13.18
11.23
11.63
11.46
12.00
11.53
10.01
9.75
9.69
8.84
9.44
4.39
4.67
5.28
5.80
5.64
3.56
3.42
3.66
3.90
3.84
3.88
3.94
3.95
voc
0.30
0.33
0.37
0.33
0.30
0.26
0.26
0.26
0.28
0.27
0.23
0.22
0.22
0.20
0.22
0.18
0.21
0.21
0.26
0.25
0.19
0.18
0.17
0.17
0.20
0.22
0.23
0.23
CO
2.13
2.42
2.69
2.37
2.11
1.81
1.84
1.84
1.92
1.86
1.66
1.61
1.55
1.43
1.52
0.81
0.86
0.95
1.09
1.05
0.90
0.93
1.00
1.10
1.11
1.15
1.17
1.19
Pb
0.61
0.46
0.41
0.32
0.22
0.15
0.17
0.14
0.14
0.13
0.13
0.13
0.11
0.07
0.06
0.05
0.05
0.05
0.03
0.03
0.02
0.02
0.02
0.02
0.02
0.02
0.02
0.02
'except Pb, which is in short tons.
January 2001
-------
APPENDIX F
ORGANIZATION OF SPREADSHEET
-------
-------
Assigning Emissions to Industries in the United States, Revised Estimates
F-l
Contents of Spreadsheet SIC7097. WK3
SIC7097.WK3
Sheet A
Sheet B
Sheet C
Sheet D
Sheet E
Sheet F
Sheet G
Sheet H
Sheet I
Sheet J
Sheet K
Sheet L
Sheet M
Sheet N
Sheet O
Sheet P
Sheet Q
Sheet R
Sheet S
Sheet T
Sheet U
Description of contents of Sheet
Contents of spreadsheet
CO emissions (original spreadsheet)
NOX emissions (original spreadsheet)
VOC emissions (original spreadsheet)
SO2 emissions (original spreadsheet)
PM-10 emissions (original spreadsheet)
PM-2.5 emissions (original spreadsheet)
Ammonia emissions (original spreadsheet)
Pb emissions (original spreadsheet)
Process emissions (agriculture, 4 mining industries, construction, and 20
manufacturing industries) (sum of sheets B through F, and sheet I) -
assigned by Allocation Rules 1 and 2
Adjusted CO emissions (assign values to "NA" cells)
Adjusted NOX emissions (assign values to "NA" cells)
Adjusted VOC emissions (assign values to "NA" cells)
Adjusted SO2 emissions (assign values to "NA" cells)
Adjusted PM-10 emissions (assign values to "NA" cells)
Adjusted Pb emissions (assign values to "NA" cells)
Output indexes : 1970 -1997 (agriculture, 4 mining industries,
construction, and 20 manufacturing industries)
Federal Reserve Board, value added, 1992, 1987, 1982
Federal Reserve Board, production indexes, 1919-1985
Federal Reserve Board, production indexes, 1986-1999
Process emissions (agriculture, 4 mining industries, construction, and
manufacturing industries) (sum of sheets K through P)
January 2001
-------
F-2
Assigning Emissions to Industries in the United States, Revised Estimates
SIC7097.WK3
Sheet V
Sheet W
Sheet X
Sheet Y
Sheet Z
Sheet AA
Sheet AB
Sheet AC
Sheet AD
Sheet AE
Sheet AF
Sheet AG
Sheet AH
Sheet AI
Description of contents of Sheet
BEA, gross output (1947-1987) : agriculture and construction
BEA, gross output (1987-1997) : agriculture and construction
Coal consumption (1982-1997) : agriculture, 4 mining industries, and
construction)
Natural Gas consumption (1982-1997) : agriculture, 4 mining industries,
and construction
Distillate Fuel Oil consumption (1982-1997) : agriculture, 4 mining
industries, and construction
Residual Fuel Oil consumption (1982-1997) : agriculture, 4 mining
industries, and construction
Share of Coal consumption of agriculture, mining, construction, and 20
manufacturing industries (1970-1997)
Share of Residual Fuel Oil consumption of agriculture, mining,
construction, and 20 manufacturing industries (1970-1997)
Share of Distillate Fuel Oil consumption of agriculture, mining,
construction, and 20 manufacturing industries (1970-1997)
Share of Fuel Oil consumption (aggregate - assign CO emissions) of
agriculture, mining, construction, and 20 manufacturing industries (1970-
1997)
Share of Fuel Oil consumption (aggregate - assign VOC emissions) of
agriculture, mining, construction, and 20 manufacturing industries (1970-
1997)
Share of Gas (natural, processed, and other) consumption of agriculture,
mining, construction, and 20 manufacturing industries (1970-1997)
Share of Wood consumption of agriculture, mining, construction, and 20
manufacturing industries (1970-1997)
Share of Coke Oven Gas and Still Gas consumption (PM-10) of
agriculture, mining, construction, and 20 manufacturing industries (1970-
1997)
January 2001
-------
Assigning Emissions to Industries in the United States, Revised Estimates
F-3
SIC7097.WK3
Sheet AJ
Sheet AK
Sheet AL
Sheet AM
Sheet AN
Sheet AO
Sheet AP
Sheet AQ
Sheet AR
Sheet AS
Description of contents of Sheet
Share of Coke Oven Gas and Still Gas consumption (NOX) of agriculture,
mining, construction, and 20 manufacturing industries (1970-1997)
Emissions from coal combustion (agriculture, mining, construction, and
20 manufacturing industries)
Emissions from residual fuel oil combustion (agriculture, mining,
construction, and 20 manufacturing industries)
Emissions from distillate fuel oil combustion (agriculture, mining,
construction, and 20 manufacturing industries)
Emissions (VOC and CO) from fuel oil combustion (agriculture, mining,
construction, and 20 manufacturing industries)
Emissions from natural gas combustion (agriculture, mining,
construction, and 20 manufacturing industries)
Emissions from wood combustion (agriculture, mining, construction, and
20 manufacturing industries)
Fuel combustion emissions (agriculture, mining, construction, and 20
manufacturing industries) (sum of sheets AK through AP)
Process and Fuel combustion emissions (agriculture, mining,
construction, and 20 manufacturing industries) (sum of sheets U through
AQ)
Process and Fuel combustion emissions (agriculture, mining,
construction, and 20 manufacturing industries) (linked to Appendix E)
January 2001
-------
-------
APPENDIX G
PREPARATION OF REPORT AND PEER REVIEW PROCESS
-------
-------
Assigning Emissions to Industries in the United States, Revised Estimates G-l
G. 1. PREPARING THE REPORT
This report was prepared by Carl A. Pasurka, Jr. in the EPA's National Center for
Environmental Economics, Office of Policy, Economics and Innovation.
Dwight French, Christy Hall, Alice Lipert, and Mark Schipper of the U.S. Department of
Energy provided information on the availability of energy consumption data for the industrial
sector.
Chris Perrien of the U.S. Bureau of the Census provided information on thermal
conversion factors used in the Census of Mineral Industries.
Wiley Barbour (Office of Air and Radiation, U.S. EPA) and Anton Steurer (Eurostat)
provided encouragement and insights throughout the process of writing this report.
Bill Barnard of E.H. Pechan and Associates provided useful insights concerning the
estimates of emissions. Douglas Solomon and David Misenheimer (U.S. EPA, Office of Air and
Radiation) also provided assistance. Special thanks go to Sharon Nizich (Office of Air and
Radiation, U.S. EPA) and Tom McMullen (Office of Air and Radiation, U.S. EPA) for
providing information on the procedures employed to estimate air emissions. It would have been
impossible to complete this project without their assistance.
G.2. DESCRIPTION OF THE REVIEW PROCESS
The draft of what has emerged as Statistical Methodology for Assigning Emissions to
Industries in the United States, Revised Estimates: 1970 to 1997 was subjected to both inter- and
intra-agency review. In addition to a representative of the Statistical Office of the European
Communities, representatives from the Office of Air and Radiation of the U.S. EPA were
contacted for comment.
The following individuals (in alphabetical order) provided helpful comments on the draft
of this report:
Wiley Barbour (U.S. EPA, Office of Air and Radiation)
Tom McMullen (U.S. EPA, Office of Air and Radiation)
Sharon Nizich (U.S. EPA, Office of Air and Radiation)
Anton Steurer (Statistical Office of the European Communities)
In addition to making numerous suggestions for changes in style, the reviews also made
the following suggestions regarding the content of the report. Upon receipt of the comments,
Carl Pasurka attempted to address and incorporate all comments to the maximum extent possible.
The comments listed in this section do not always represent the exact wording of the original
comments and may reflect a combination of more than one comment. Specific comments and
responses are listed below.
January 2001
-------
G-2 Assigning Emissions to Industries in the United States, Revised Estimates
G.3. SPECIFIC COMMENTS AND RESPONSES
Comment: Report should highlight the fact that emissions from motor vehicles are not
assigned to various industries.
Response: Section 1.1. now mentions that emissions from motor vehicles are not assigned to
industries.
Comment: The use of fuel consumption data to assign emissions from result from fuel
combustion ignores differences in control efficiencies, age of boilers, and
variation in standards across industries.
Response: The first paragraph of Section 3 acknowledges this assumption. Section 4
(Overview) points out that this same assumption has been employed by the
statistical agencies of other governments that have developed inventories of
emissions by industry.
Comment: There should be some explanation of why certain source categories are not
assigned to industries.
Response: Portions of Sections 1.1, 2.1.13, 3.6, and 4 discuss which source categories are
excluded and explain why some of the categories are excluded.
Comment: The report should provide a list of Acronyms and Abbreviations.
Response: The report now includes a list of relevant Acronyms and Abbreviations.
Comment: The report should define "industrial sector."
Response: The report provides a definition of "industrial sector" in footnote 4 on page 2.
Comment: There should be a discussion of the potential uses of the data developed in the
report.
Response: Within the Introduction section, the "Purpose of the Project" now contains an
overview of some of the proposed uses of these data.
Comment: Will it be fairly easy to revise the data in this study from an SIC basis to the
NAICS codes?
Response: Footnotes 46 and 47 discuss some of the problems associated with maintaining a
consistent time series using data collected by the SIC and NAICS systems. This is
an important issue confronting the statistical agencies of the United States. It
appears there will be efforts to maintain consistent time series for at least some
data. However, it will be several years before it becomes apparent which data and
which years will be included in such a time series. If the change from SIC to
January 2001
-------
Assigning Emissions to Industries in the United States, Revised Estimates
G-3
NAICS codes makes it difficult to assign the Tier I, n, and in categories of
process emissions, it would be possible to use the estimates of emissions by SCC
codes and develop a consist set process emissions for the years starting with 1985.
Comment: The "Allocation Rules" in Section 2.1 of need to be clarified.
Response: The discussion of the "Allocation Rules" in Section 2.1 has been revised.
Comment: The study should clarify how emissions from the co-generation of electricity are
treated.
Response: Footnote 17 discusses this issue
Comment: How significant is the discrepancy between the NEA and DAEMEC estimates of
fuel consumption?
Response: The shares of coal, residual fuel oil, distillate fuel oil, and natural gas
consumption assigned to industries in Sheets AB, AC, AD, and AG of
SIC7097.WK3 do not appear to experience a dramatic change between 1984
(NEA data) and 1985 (DAEMEC data). This is mentioned in footnote 21. Section
3.3.3. discusses the discrepancies that exist between the NEA and DAEMEC
estimates of consumption of wood for heat and power. Sheets AH of
SIC7097.WK3 presents the distribution of wood consumption among industries.
Section 3.3.4. discusses the discrepancies that exist between the NEA and
DAEMEC estimates of consumption of blast furnace / coke oven and still gases
heat and power. Sheets AI of SIC7097.WK3 presents the distribution of blast
furnace / coke oven and still gases consumption among industries. In addition,
Section 4 contains two paragraphs illustrating that most of the substantial declines
in VOC and PM-10 emissions between 1984 and 1985 are the consequence of
factors other the switch from the NEA and DAEMEC estimates of fuel
consumption.
Comment: Since Statistical Methodology for Assigning Emissions to Industries in the United
States, Revised Estimates: 1970 to 1997 assigns values to some cells listed as
"NA" in the 1970-1997 NAPET, the reader should be told which values are
affected by these estimates.
Response: In Appendix E, all estimates of emissions affected by estimates of "NA" cells are
highlighted.
Comment: Is it possible to develop more disaggregated estimates of emissions than the two-
digit SIC level?
Response: Estimates of process emissions are available by SCC for 1985 to the present. It
would be necessary to develop a concordance between the SCCs and the SIC or
NAICS codes. Since fuel consumption by industry is used to assign emissions
January 2001
-------
G-4 Assigning Emissions to Industries in the United States, Revised Estimates
from the combustion of fuel in the industrial sector, the only other constraint on
developing more disaggregated estimates of emissions is the availability of
information on fuel consumption by industry.
Comment: What share of total emissions are assigned to industries in this report?
Response: Table 6 has been added and shows the quantity of emissions included in this
report, and the share of total emissions assigned to the electric utility sector and
industries in the industrial sector.
January 2001
------- |