* — \
*1 PROt^
Documentation for the Final 1999 National
Emissions Inventory (Version 3.0) for Criteria
Air Pollutants and Ammonia, Point Sources
-------
-------
EPA-454/B-20-020
January 2004
Documentation for the Final 1999 National Emissions Inventory (Version 3.0) for Criteria Air
Pollutants and Ammonia, Point Sources
Prepared by:
E.H. Pechan & Associates, Inc.
5528-B Hempstead Way
Springfield, VA 22151
Prepared for:
Roy Huntley
Environmental Engineer
Office of Air Quality Planning and Standards
Emission Factor and Inventory Group
U.S. Environmental Protection Agency
Research Triangle Park, NC 27711
U.S. Environmental Protection Agency
Office of Air Quality Planning and Standards
Air Quality Assessment Division
Research Triangle Park, NC
-------
-------
CONTENTS
Page
TABLES AM) FIGURES v
ACRONYMS vi
A. WHAT IS THE PURPOSE OF THIS DOCUMENT? 1
B. WHO SUBMITTED POINT SOURCE INVENTORIES TO EPA? 1
C. WHAT IS EPA'S POLICY FOR ACCEPTING STATE/LOCAL/TRIBAL DATA? 1
D. WHAT DID EPA DO WITH THE STATE/LOCAL/TRIBAL DATA? 1
E. HOW DO I FIND AND REVIEW MY DATA? 5
F. HOW DID EPA MERGE THE 1999 NEI VERSION 2.0 AND STATE/LOCAL/TRIBAL
INVENTORIES? 6
G. HOW DID EPA MERGE LOCAL AND TRIBAL INVENTORIES INTO STATE INVENTORIES? 6
H. HOW DID EPA MERGE STATE/LOCAL ELECTRICITY GENERATING UNIT (EGU)
EMISSIONS DATA WITH EPA DATA? 7
I. WHAT IS THE RELATIONSHIP BETWEEN THE NEI AND eGRID? 10
J. HOW DID EPA ADD PM-RELATED POLLUTANT EMISSIONS TO MY INVENTORY? 12
K. HOW DID EPA ADD NH3 EMISSIONS TO STATE/LOCAL INVENTORIES? 14
L. HOW DID EPA FILL IN MISSING OR CORRECT INVALID INFORMATION FOR
MANDATORY AND NONMANDATORY FIELDS? 16
M. DID EPA REJECT ANY INFORMATION IN STATE/LOCAL INVENTORIES? 20
N. WHERE DO I FIND DOCUMENTATION OF EPA POINT SOURCE METHODOLOGIES? ... 20
Documentation for the Final 1999 National Emissions
Inventory (Vgrsion 3) for Criteria Air Pollutants and Ammonia 111
Point Source Methodologies
-------
APPENDIX A
DOCUMENTATION OF CHANGES MADE TO INDIVIDUAL
STATE/LOCAL/TRIBAL INVENTORIES A-l
APPENDIX B
1999 NEI METHODS FOR ELECTRIC UTILITY STEAM GENERATED FOSSIL FUEL
COMBUSTION UNITS B-l
APPENDIX C
METHODOLOGY FOR AUGMENTING STATE/LOCAL/TRIBAL 1999 POINT SOURCE
EMISSIONS INVENTORIES WITH PM10 AND PM25 EMISSIONS FOR THE 1999 NEI.. . C-1
1. INTRODUCTION C-l
2. APPROACH C-l
a. Step 1: Prepare S/L/T PM and PM10 Emissions for Input to the PM Calculator C-3
b. Step 2: Develop and Apply Source-Specific Conversion Factors C-4
c. Step 3: Prepare Factors from PM Calculator C-l 1
d. Step 4: Develop and Apply Algorithms to Estimate Emissions from S/L/T Inventory
Data C-l2
e. Step 5: Review Results and Update the NEI with Emission Estimates and Control
Information C-12
3. UNCERTAINTIES/FUTURE RESEARCH C-13
Documentation for the Final 1999 National Emissions
Inventory (Vgrsion 3) for Criteria Air Pollutants and Ammonia IV
Point Source Methodologies
-------
TABLES AND FIGURES
Tables Page
1. Summary of State/Local/Tribal Point Source Inventories for the Criteria Air Pollutants 2
2. Summary of State/Local/Tribal Point Source Comments for the Criteria Air Pollutants 3
3. Utilities with Different County FIPS Codes in Draft Version 3 of the NEI and eGRID 9
4. Utilities to Which One ORIS ID is Assigned to Two Site IDs in Draft Version 3 of the NEI 10
5. Field Names and Definitions in ORIS Facility Codes Database 12
6. Field Names and Definitions in NH3_V2_to_DV3_Xwalk Database 15
B-l. Methods for Developing Annual Emission Estimates for Steam Generated Fossil-Fuel Utility Boilers for
the Years 1989-1999 B-l 1
B-2. Steam Electric Utility Unit Source Classification Code Relationships B-13
B-3. Boiler Emissions Data Sources (Other than EIA-767)
for NOX, S02, and Other Pollutant Emissions Data by Year, 1985-1999 B-15
B-4. Emission Factors B-l6
B-5. Rules for Assigning Primary and Secondary PM Control Device NEDS Codes B-l8
B-6. PM Condensable Emission Factor (PMCDEF) Rules B-l9
B-7. Algorithms Used to Estimate EIA-Based VOC, NOX, CO, S02, PM10, PM2.5, and NH3
Annual Emissions from Electric Utility Boilers B-20
B-8. Algorithms Used to Estimate EIA-Based Condensible PM, Primary PM10 and Primary
PM2.5 Annual Emissions for Electric Utility Boilers B-21
B-9. Algorithms Used to Disaggregate ETS/CEM Boiler Data to the Boiler-SCC Level B-22
Figures
1. Procedures for Calculating PM2.5 and PM10 Emissions for the NEI C-2
Documentation for the Final 1999 National Emissions
Inventory (Vgrsion 3) for Criteria Air Pollutants and Ammonia V
Point Source Methodologies
-------
ACRONYMS
CAMD EPA's Clean Air Markets Division
CEM Continuous Emissions Monitoring
CHIEF Clearinghouse for Inventories and Emission Factors
CHP Combined heat and power
CON condensible
DOE Department of Energy
eGRID Emissions & Generation Resource Integrated Database
EIA Energy Information Administration
EPA U.S. Environmental Protection Agency
ETS Emissions Tracking System
FGD flue gas desulfurization
FIL filterable
FIPS Federal Information Processing Standards
FTP File Transfer Protocol
GT gas turbine
HC hydrocarbon
IC internal combustion
H EP Institute of Tribal Environmental Professionals
MW megawatt
NEDS National Emission Data System
NEI National Emissions Inventory
NH3 ammonia
NIF NEI Input Format
NMOC nonmethane organic compound
NMOG nonmethane organic gases
NOX oxides of nitrogen
ORIS Office of the Regulatory Information System
OSD ozone season day
PM particulate matter
PMCD condensible PM
PM10 particulate matter with an aerodynamic diameter less than or equal to a nominal 10 micrometers
in diameter
PM2.5 particulate matter with an aerodynamic diameter less than or equal to a nominal 2.5 micrometers
in diameter
PRI primary
QA quality assurance
QC quality control
S/L/T State/Local/Tribal
SCC Source Classification Code
SIC standard industrial classification
Documentation for the Final 1999 National Emissions
Inventory (Vgrsion 3) for Criteria Air Pollutants and Ammonia VI
Point Source Methodologies
-------
S02
sulfur dioxide
S04
sulfate
SOX
oxides of sulfur
TOG
total organic gases
TSP
total suspended particulate
U.S.
United States
voc
volatile organic compound
Documentation for the Final 1999 National Emissions
Inventory (Vgrsion 3) for Criteria Air Pollutants and Ammonia VII
Point Source Methodologies
-------
A. WHAT IS THE PURPOSE OF THIS DOCUMENT?
This document summarizes the procedures the U.S. Environmental Protection Agency (EPA) used in
developing Version 3 of the 1999 National Emission Inventory (NEI) and incorporating State/Local/Tribal
(S/L/T) agency-submitted point source data. For S/L/T agencies that did not submit an inventory for Version 3
of the NEI, emissions data from final Version 2 of the 1999 NEI are used in Version 3.
B. WHO SUBMITTED POINT SOURCE INVENTORIES TO EPA?
Table 1 identifies the agencies that submitted 1999 point source inventories for the criteria air pollutants.
Table 1 also provides information on the geographic and pollutant coverage of each inventory, and identifies
States that previously submitted a 1999 inventory that has been included in final Version 2 of the NEI. All of
the inventories were submitted to EPA in NEI Input Format (NIF) 2.0. Table 2 identifies states that provided
comments on Version 3.
C. WHAT IS EPA'S POLICY FOR ACCEPTING STATE/LOCAL/TRIBAL DATA?
EPA preferentially uses S/L/T point source data in the NEI, except for utility NOX and sulfur dioxide
(S02) emissions and PM emissions. For utilities, EPA uses oxides of nitrogen (NOX) and S02 and PM
emissions that facilities report to the Emissions Tracking System/Continuous Emissions Monitoring (ETS/CEM)
Scorecard database. Due to resource constraints associated with preparing Version 3 of the 1999 NEI, the
Version 3 inventories for CT, LA, NH, and WV were not included in Version 3 of the NEI because the
Version 3 inventories did not substantially differ from their inventories included in Version 2 of the 1999 NEI.
D. WHAT DID EPA DO WITH THE STATE/LOCAL/TRIBAL DATA?
Upon receipt of each inventory, EPA ran its quality control/quality assurance (QC/QA) program on the
inventory to identify format errors, missing data required in mandatory fields, duplicate records, and other
referential integrity problems. EPA contacted the S/L/T agencies to resolve many of these issues in the Access
database submittals. Then the S/L/T inventories were loaded into EPA's Oracle database, and combined into
a single data set in NIF 2.0. The following procedures were then applied to the S/L/T inventories prior to
merging the NEI into the S/L/T inventories to fill data gaps.
1. Removed all non-criteria pollutant emissions from the Emission table and associated parent and child
records from other tables.
Documentation for the Final 1999 National Emissions
Inventory (Vgrsion 3) for Criteria Air Pollutants and Ammonia 1
Point Source Methodologies
-------
Table 1. Summary of State/Local/Tribal Point Source
Inventories for the Criteria Air Pollutants
State
NIF
Versio
n
Transactio
n Type
Inventory
Type
Geographic
Coverage1
Number
of
Counties
Number of
Counties in
State
Pollutants in
Inventory
Submitted
Version 2?
Arizona
2.0
Original
CRITHAP
Statewide
10
15
CO, NH3, NOX, PM10-
PRI, S02, VOC
No
Arizona - Salt
River Pima-
Maricopa Indian
Community
(ITEP Tribe ID =
433)
2.0
Original
CRITHAP
Tribal
1
15
CO, NMOC, NOX, PM10-
PRI, S02, SOX, VOC
No
Connecticut2
2.0
Original
CRIT
Statewide
8
8
CO, NOX, PM10-PRI,
SOX, VOC
Yes
Delaware
2.0
Original
CRITHAP
Statewide
3
3
CO, NH3, NOX, PM10-
PRI, PIM-PRI, S02, VOC
No
Georgia
2.0
Original
CRIT
Partial State
29
159
CO, NOX, VOC
No
Idaho
2.0
Original
CRITHAP
Partial State
20
44
CO, NH3, NMOG, NOX,
PM10-FIL, PIM-FIL, S02,
SOX, VOC, TOG
No
Illinois?
2.0
Original
CRIT
Statewide
102
102
CO, NOX, PM10-PRI, PIM-
PRI, S02, VOC
Yes
Louisiana2
2.0
Original
CRITHAP
Statewide
61
64
CO, NH3, NOX, PM10-FIL,
S02, VOC
Yes
Michigan
2.0
Replace
CRITHAP
Statewide
80
83
CO, NOX, PM10-PRI,
PM25-PRI, PIM-PRI, S02,
VOC
Yes
Minnesota
2.0
Original
CRIT
Statewide
83
87
CO, NOX, PM10-PRI,
S02, VOC, 7439921
Yes
North Carolina -
Beaufort
County (County
FIPS 013)
2.0
Original
CRITHAP
Local
1
100
CO, NH3, NOX, PM10-
PRI, PIM-PRI, S02, VOC
Yes
North Carolina -
Forsyth County
(County FIPS
067)
2.0
Original
CRITHAP
Local
1
100
CO, NH3, NOX, PM10-
PRI, PIM-PRI, S02, VOC
Yes
New
Hampshire2
2.0
Original
CRITHAP
Statewide
10
10
CO, NOX, PIM10-FIL,
PM25-FIL, PIM-FIL, S02,
VOC
Yes
New Jersey
2.0
Original
CRIT
Statewide
8
21
CO, NOX, PIM-PRI, PM10-
PRI, S02, VOC
No
New York
2.0
Original
CRITHAP
Statewide
62
62
CO, NH3, NMOC, NOX,
PIM-PRI, PM10-PRI, S02,
VOC, 7664417
No
Nevada
2.0
Original
CRITHAP
Statewide
13
17
CO, NH3, NOX, PM10-
PRI, PIM-PRI, S02, VOC
No
Oregon4
2.0
Original
CRITHAP
Statewide
34
36
CO, NOX, PM10-PRI,
SOX, VOC, 7664417
Yes
Oregon -
Umatilla (ITEP
Tribe ID = 80)
2.0
Original
CRITHAP
Tribal
1
36
PM10-PRI
No
Documentation for the Final 1999 National Emissions
Inventory (Vgrsion 3) for Criteria Air Pollutants and Ammonia 2
Point Source Methodologies
-------
Table 1 (continued)
State
NIF
Versio
n
Transactio
n Type
Inventory
Type
Geographic
Coverage1
Number
of
Counties
Number of
Counties in
State
Pollutants in
Inventory
Submitted
Version 2?
South Carolina
2.0
Original
CRITHAP
Statewide
46
46
CO, HC, NH3, NOX,
PM10-FIL, PM25-FIL, PM-
FIL, S02, VOC
Yes
Tennessee
2.0
Original
CRIT
Statewide
80
95
CO, NH3, NOX, PM25-FIL,
PM-FIL, S02, VOC
No
Tennessee -
Shelby County
(County FIPS
157 )5
2.0
Original
CRITHAP
Local
1
95
CO, HC, NH3, Nox, PM,
PM10, PM10-FIL, PM10-
PRI, PM2.5, PM2.5-FIL,
PM25-FIL, PM25-PRI, PM-
FIL, PM-PRI, S02, S02,
SOX, TSP, VOC
No
Tennessee -
Davidson
County (County
FIPS 037)
2.0
Original
CRITHAP
Local
1
CO, NOX, PM10-PRI,
S02, VOC
No
West Virginia2
2
Original
CRITHAP
Statewide
54
55
CO, HC, NH3, NOX,
PM10-FIL, PM25-FIL, PM-
FIL, S02, VOC
Yes
1 Statewide means the inventory was submitted by the State rather than a Local or Tribal agency. It does not mean that the inventory
contains data for all counties in a State. Thus, a statewide inventory may exclude counties for which the Local agency prepares its own
inventory for the National Emission Inventory (NEI).
2 Due to time and resource constraints associated with preparing Version 3 of the 1999 NEI, the Version 3 inventories for CT, LA, NH, and
WV were not included in Version 3 of the NEI because the Version 3 inventories did not substantially differ from their inventories included
in Version 2 of the 1999 NEI.
3 IL's inventory was for the 2000 base year, and, therefore was not included in Version 3 of the 1999 NEI. IL submitted a 1999 base year
inventory that was included in Version 2 of the 1999 NEI.
4 At the direction of the State of Oregon and Lane County authorities, data for Lane County (FIPS 039) included in the State's inventory
were not included in draft Version 3 of the 1999 NEI. The authorities will provide data for Lane County during the comment period. In
addition, the State's inventory for other counties contained sites missing from the inventory they provided for Version 2 of the NEI.
Therefore, the sites in their Version 3 submittal were added to the sites in their Version 2 submittal for draft Version 3 of the NEI.
5 The inventory submitted by Shelby County, TN was not included in Version 3 of the NEI due to time and resource constraints associated
with resolving referential integrity issues in the inventory submittal.
Table 2. Summary of State/Local/Tribal Point Source
Comments for the Criteria Air Pollutants
State
NIF
Version
Transaction
Type
Inventory
Type
Geographic
Coverage1
Number of
Counties
Number of
Counties in
State
Pollutants in Inventory
Submitted
Version 2?
Louisiana2
2.0
Original
CRITHAP
Statewide
61
64
CO, NH3, NOX, PM10-FIL,
S02, VOC
Yes
Michigan
2.0
Replace
CRITHAP
Statewide
80
83
CO, NOX, PM10-PRI,
PM25-PRI, PM-PRI, S02,
VOC
Yes
Massachusetts
2.0
Minnesota
2.0
Nevada
2.0
17
No
NC-Mecklenburg
1
Ohio
2.0
Documentation for the Final 1999 National Emissions
Inventory (Vgrsion 3) for Criteria Air Pollutants and Ammonia 3
Point Source Methodologies
-------
Table 2 (continued)
State
NIF
Version
Transaction
Type
Inventory
Type
Geographic
Coverage1
Number of
Counties
Number of
Counties in
State
Pollutants in Inventory
Submitted
Version 2?
PA-Allegheny
County
2.0
Yes
South Carolina
2.0
Original
CRITHAP
Statewide
46
46
CO, HC, NH3, NOX,
PM10-FIL, PM25-FIL, PM-
FIL, S02, VOC
Yes
Tennessee
2.0
Original
CRIT
Statewide
80
95
CO, NH3, NOX, PM25-FIL,
PM-FIL, S02, VOC
No
Tennessee -
Shelby
2.0
Original
CRITHAP
Local
1
CO, NOX, PM10-PRI,
S02, VOC
No
Virginia
West Virginia
2
Original
CRITHAP
Statewide
54
55
CO, HC, NH3, NOX,
PM10-FIL, PM25-FIL, PM-
FIL, S02, VOC
Yes
1 Statewide means the inventory was submitted by the State rather than a Local or Tribal agency. It does not mean that the inventory
contains data for all counties in a State. Thus, a statewide inventory may exclude counties for which the Local agency prepares its own
inventory for the National Emission Inventory (NEI).
2. Changed Pollutant Codes as follows: oxides of sulfur (SOX) to S02, nonmethane organic
compounds (NMOC) or hydrocarbon (HC) to volatile organic compounds (VOC), Chemical
Abstract Number 7439921 to ammonia (NH3), and made all other pollutant codes upper case. In
the NEI, VOC and S02 are used exclusively, and all pollutants are upper case. If an agency
provided NMOC or HC emissions for the same data key (excluding the pollutant code) for which it
provided VOC emissions, the NMOC or HC emissions were removed from the NEI. Otherwise,
the NMOC or HC codes were changed to VOC. The routines for merging Version 2 of the NEI
with S/L/T data in the Emission and Control Equipment tables match on the data key for the tables
which include Pollutant Code. Where records matched on the data key, S/L/T data were used. The
changes to the pollutant codes were made to avoid inserting NEI records that would otherwise
double count emissions. In addition, the pollutant codes need to be consistent to generate reports of
national emissions or by State.
3. Identified and analyzed referential integrity issues (i.e., orphan and widow [childless parent] records).
Corrected issues with mandatory field data to resolve referential integrity issues. Then resolved
remaining issues by adding parent records for orphans and adding the minimum required data to
make the record valid. Widow records were removed from the tables, and stored in a separate file.
A final analysis was conducted to confirm compliance with referential integrity. Additionally, for the
final inventory submissions QA was performed to ensure compliance with submittal flag rules. In
general, submittal flag issues were resolved as follows. If a record was coded with a submittal flag of
"A" (add), but the record existed in the inventory, the submittal flag was altered to "RA". If the
submittal flag was coded as "RA" but the record did not exist in the inventory, the submittal flag was
recoded to "A". If records within a table were coded "A" and "D" by key values, these were
changed to "RA" and "RD" as appropriate. If duplicates were found, and the record was a true
Documentation for the Final 1999 National Emissions
Inventory (Vgrsion 3) for Criteria Air Pollutants and Ammonia 4
Point Source Methodologies
-------
duplicate (all fields were equal), the first record added to the inventory is kept. If it is an emission
record, the record with the higher emission value is kept.
4. Ran additional routines to identify remaining QA issues. Corrections to QA issues were coordinated
with the S/L/T agencies when necessary.
E. HOW DO I FIND AND REVIEW MY DATA?
Final Version 3 of the 1999 NEI is available on EPA's File Transfer Protocol (FTP) server
(ftp://ftp.epa.gov/EmisInventory/finalnei99ver3/criteria/). The inventory is provided in MF 2.0 in a separate
Access97 database for each State. Note that EPA converted all emissions in S/L/T inventory submittals to
"TON" in order to facilitate QA and generation of emission summaries. In the Access97 databases, the units of
the emissions are now reported as "TON" for all S/L/T agency inventories.
EPA entered a flag in the second blank field of the Emissions table to identify the source of the emission
records included in your inventory. The flags are defined as follows:
Code Definition
S State data
S-C State supplied activity data that EPA used to calculate emissions for pollutants missing
from State's inventory
L Local data
R Tribal data
E EPA NEI data
GENPARENTParent record added to comply with referential integrity
ETS Augmentation Codes
ETS-I Record created and inserted into draft Version 3 to hold ETS emissions not included
in S/L/T inventory. Note that "ETS _F' was also used; this will be corrected to ETS-I
in final Version 3.
ETS-I-NPC Record created and inserted into draft Version 3 to hold utility pollutant emissions
from EPA's 1999 utility inventory that were missing from S/L inventory for the same
site and emission unit
ETS-U Record in S/L/T inventory for which NOX or S02 emissions are replaced with ETS
emissions
Ammonia Augmentation Codes
NH3-N0X
NH3-C0
NH3-S02
NH3-V0C
NH3-PM10-F
Record created by Ammonia Augmentation - calculated off of NOX
Record created by Ammonia Augmentation - calculated off of CO
Record created by Ammonia Augmentation - calculated off of S02
Record created by Ammonia Augmentation - calculated off of VOC
Record created by Ammonia Augmentation - calculated off of PM10-FIL
Documentation for the Final 1999 National Emissions
Inventory (Vgrsion 3) for Criteria Air Pollutants and Ammonia 5
Point Source Methodologies
-------
PMAugmentation Codes
AUG-C
AUG-A
Record added for PM10/PM25 emissions estimated using the PM Calculator
Record added for PM10/PM25 emissions estimated using individual estimations
EPA has also prepared the following summaries to assist S/L/T agencies in their review:
Summary
File Name
Annual emissions by State, County, and Sector
Annual emissions by State, County, and Source
County SumCrit_99F V3_03 03
SCC SumCrit 99FV3 0303
Classification Code (SCC)
Annual emissions by State, County, and
TierSumCrit 99FV3 0303
Tier 1, 2, and 3 Categories
All Facilities List for 1999 NEI Version 3
SiteSumPointCrit 99FV3 0303
Each of these summaries are in an Access97 database located on EPA's FTP server (ftp://ftp.epa.gov/
Emislnventory/).
F. HOW DID EPA MERGE THE 1999 NEI VERSION 2.0 AND STATE/LOCAL/TRIBAL
INVENTORIES?
In general, all of the data in final Version 2 were replaced with S/L/T data on a county-by-county basis.
Augmentation procedures were then run to add ETS NOX and S02 and PM, NH3, and non-utility PM-
related pollutants missing from S/L/T inventories. Thus, if a State did not include a county in its inventory for
which final Version 2 contained sites, the sites in final Version 2 are included in draft Version 3 of the NEI.
The EPA compared the number of sites by county in each S/L/T inventory to the number of sites by
county in final Version 2 of the NEI. The number of sites in some counties in AZ, GA, SC, and NV differed
significantly. These States reviewed and identified sites in final Version 2 that were not in their Version 3
inventory submittal to keep in Version 3.
GA did not include any S02 or PM-related pollutants in their inventory. Since GA provided throughput
data with the VOC, NOX, and CO emissions they included in their inventory, EPA selected emission factors
from Factor Information Retrieval (FIRE) and AP-42 to calculate S02 and PM-related pollutant emissions.
G. HOW DID EPA MERGE LOCAL AND TRIBAL INVENTORIES INTO STATE
INVENTORIES?
The following explains how Local and Tribal inventories were merged with State inventories:
Salt River Pima-Maricopa Indian Community, Arizona: The site data in this inventory were added to
the site data in final Version 2 of the NEI submitted by Maricopa County.
Documentation for the Final 1999 National Emissions
Inventory (Vgrsion 3) for Criteria Air Pollutants and Ammonia 6
Point Source Methodologies
-------
Umatilla Tribe, Oregon: The site data in this inventory were added to the site data that Oregon
submitted for final Version 2 and draft Version 3 of the NEI.
Beaufort County, North Carolina: The county's inventory was added to the inventory that the State
of North Carolina supplied for this county for final Version 2 of the NEI.
Forsyth County, North Carolina: The county's inventory replaced the inventory it submitted for final
Version 2 of the NEI.
Davidson County, Tennessee: The county's inventory replaced the county inventory in final Version
2 of the NEI. The State of Tennessee did not provide any data for this county in its Version 3
inventory submittal.
Shelby County, Tennessee: The county's inventory replaced the county inventory in final Version 2
of the NEI.
Mecklenburg County, North Carolina: The county's inventory replaced the county inventory in final
Version 2 of the NEI.
H. HOW DID EPA MERGE STATE/LOCAL ELECTRICITY GENERATING UNIT (EGU)
EMISSIONS DATA WITH EPA DATA?
NOX and S02 and PM emissions data collected by ETS/CEM procedures by EPA's Clean Air Markets
Division (CAMD) have been deemed more reliable and accurate than other reported data. Thus, for
ETS/CEM units, EPA replaced S/L-supplied NOX and S02 and PM emissions with the emissions from
ETS/CEM when the ETS/CEM emission units could be identified in S/L inventories. This procedure affected
only S/L inventories because Tribal inventories did not contain any emission units for utilities that report NOX
or S02 or PM emissions to ETS/CEM.
The EPA prepared a crosswalk to match boilers (for which emissions are reported to ETS/CEM) in its
EGU inventory to sites and emission units in the S/L inventories. Process-level NOX and S02 and PM
emissions in S/L inventories were adjusted by the ratio of ETS-to-S/L emission unit-level emissions. By using
this approach, site, emission unit, and process IDs and SCCs are not changed in S/L inventories. In addition,
EPA included in the crosswalk boilers in its EGU inventory that report to the Energy Information Administration
(EIA) but not to ETS/CEM in order to identify the EGU's in the NEI. For the purpose of the crosswalk, these
units are termed EIA-Only units. The EPA retained S/L NOX and S02 emissions in the NEI for the EIA-Only
units. PM was updated for all units.
This crosswalk was then used to add the ORIS identification (ID) code to the ORIS Facility Code field in
the Site table (if missing from S/L inventories), and the EPA boiler ID to the ORIS Boiler ID field in the
Emission Unit table. The EPA is reserving the ORIS Boiler ID field for EPA use to include the boiler ID code
from EPA's EGU inventory. Note that EPA removed S/L information from the ORIS Boiler ID field in
Version 3 (and final Version 2) of the NEI and added the boiler ID from its inventory. By doing this, EPA can
Documentation for the Final 1999 National Emissions
Inventory (Vgrsion 3) for Criteria Air Pollutants and Ammonia 7
Point Source Methodologies
-------
quickly identify and compare S/L data in the NEI to EPA data for EGUs. Appendix B provides the methods
EPA used to prepare the 1999 EGU inventory.
In addition, if a S/L/T agency did not include emissions for pollutants that are included in EPA's EGU
inventory for a specific boiler/emission unit, EPA added the emissions for the missing pollutants in its EGU
inventory to Version 3. These records are identified with a data source flag of ETS-I-NPC in the EM table of
draft Version 3 of the NEI.
1. Utility Crosswalk File
The crosswalk is provided in an Access database named "Utility_Blr_to_NEI_EU_XWalk_DV3. mdb".
This file includes the following relationships to link boiler IDs in EPA's EGU inventory and emission units in S/L:
One-to-one match: One boiler in EPA's EGU inventory matches to one emission unit ID in the
NEI;
One-to-many match: One boiler in EPA's EGU inventory matches to two or more emission unit
IDs in the NEI; and
Many-to-one match: Two or more boilers in EPA's EGU inventory match to one emission unit ID in
the NEI.
The crosswalk contains the following fields:
ORIS FACILITY CODE From S/L inventory if supplied by S/L agency; otherwise, EPA entered the
ORIS ID
FACILITY_NAME From S/L inventory if supplied by S/L agency
STATE FIPS From S/L inventory if supplied by S/L agency
COUNTY FIPS From S/L inventory if supplied by S/L agency
SITE ID From S/L inventory if supplied by S/L agency
EMISSION UNIT ID From S/L inventory if supplied by S/L agency
ORISBOILERID An ID assigned by EPA that is used in EPA's 1999 utility inventory.
ORIS BOILER ID FLAG Yes means the ORIS boiler ID is entered in the ORIS BOILER ID field
in the Emission Unit table of the NEI; No means the boiler ID was not
entered in the ORIS BOILER ID field in the Emission Unit table of the
NEI because more than one boiler matched to only one emission unit in the
NEI (many-to-one relationship).
REPORTS TO This field uses the following codes: EIA = reports only to the Energy
Information Administration (EIA); ETS Only = reports only to the
ETS/CEM Scorecard database; ETSEIA = reports to both ETS/CEM and
EIA. Note that records with ETSEIA and ETS Only identify the emission
units in S/L inventories for which EPA replaced S/L NOX and S02
emissions with the emissions that are reported to ETS/CEM except when
the S/L emissions were within one ton or less of the ETS/CEM emissions.
Documentation for the Final 1999 National Emissions
Inventory (Vgrsion 3) for Criteria Air Pollutants and Ammonia 8
Point Source Methodologies
-------
Process-level NOX and S02 and PM emissions in S/L inventories were
adjusted by the ratio of ETS-to-S/L emission unit-level emissions.
2. QA Issues
The EPA has identified the following QA issues with utility data in draft Version 3 of the 1999 NEI:
Table 2 shows utilities in Version 3 of the NEI that have a county FIPS code that is different than the
code reported in eGRID and EPA's EGU inventory. The utility crosswalk file contains the county
FIPS code reported in draft Version 3 of the NEI. The EPA requests that the States identified in
Table 2 correct the county FIPS codes in draft Version 3 or provide information that indicates that
the county FIPS codes in draft Version 3 are correct.
Table 3. Utilities with Different County FIPS Codes in Draft Version 3 of the NEI and eGRID
ORIS
PL ID
State
FIPS
State
Count
y FIPS
in NEI
Count
y in
NEI
Site ID
in NEI
Facility Name in
NEI
County
FIPS in
eGRID
County
in
eGRID
Plant Name
in eGRID
1382
21
Kentuc
ky
233
Web st
er
2123300
001
Western KY Energy
Corp
101
Henders
on
HMP&L
Station Two
2721
37
North
Carolin
a
161
Rutherf
ord
3716100
028
Duke Energy Corp -
Cliffside Stea
045
Clevelan
d
Cliffside
For Basin Electric Laramie River Station (WY, State FIPS 56, County FIPS 031, site ID 00001),
ORIS ID 6204 is used in the NEI and in the utility crosswalk database. In eGRID, the site is divided
between two ORIS IDs (6204.1 and 6204.2) because boiler 1 goes to the eastern grid and boilers 2
and 3 go to the western grid. For final Version 3, EPA added the two IDs to make the IDs
consistent with eGRID.
For RG&E ALLEGANY STATION #133 (NY, State FIPS 36, County FIPS 003, site ID
9025800018), the ORIS ID in eGRID is 7784 and the ORIS ID in draft Version 3 of the NEI and
the crosswalk database is 10619. The correct ORIS ID for this facility is 7784; therefore, this
correction was made in final Version 3 of the NEI.
Table 3 identifies cases where one ORIS ID has been assigned to two site IDs in draft Version 3 of
the NEI.
Documentation for the Final 1999 National Emissions
Inventory (Vgrsion 3) for Criteria Air Pollutants and Ammonia 9
Point Source Methodologies
-------
Table 4. Utilities to Which One ORIS ID is Assigned to Two Site IDs in Version 3 of the
NEI
ORISPL
ID
State
FIPS
State
County
FIPS in
NEI
County
Name in
NEI
Site ID in
NEI
Facility Name in NEI
Comments1
1572
24
Maryland
031
Montgome
ry
0019
PEPCO - DICKERSON
1572
24
Maryland
031
Montgome
ry
1822
PEPCO - STATION H
6076
30
Montana
087
Rosebud
0008
PPL, MONTANA COLSTRIP UNITS #
& #2
6076
30
Montana
087
Rosebud
0014
MPC - COLSTRIP UNITS #3 & #4
2434
34
New
Jersey
011
Cumberla
nd
75507
Vineland Municipal Electric Utility -
Down St
2434
34
New
Jersey
011
Cumberla
nd
0020
UNKNOWN
Record in Site table was
generated to maintain
parentless record in Emission
table. This site ID is not
included in the utility crosswalk.
2531
36
New York
017
Chenango
708220001
4
AES JENNISON
2531
36
New York
017
Chenango
0083
UNKNOWN
Record in Site table was
generated to maintain
parentless record in Emission
table. This site ID is not
included in the utility crosswalk.
8906
36
New York
081
Queens
263010018
5
ASTORIA (ORION) GENERATING
STATION
8906
36
New York
081
Queens
263010019
1
ASTORIA GAS TURBINE POWER-
31-01 20TH AVE
This site ID is not included in the
utility crosswalk.
1 unless otherwise indicated in comments, the site ID is included in the utility crosswalk.
3. Unmatched ETS-Only Unit File
The ETS/CEM Scorecard database contains emission units for which 1999 NOX and S02 emissions
were reported to ETS/CEM but not to the Energy Information Administration (EIA). For the purpose of the
NEI, these emission units are referred to as ETS-only units. The EPA matched many of the ETS-only units to
S/L site and emission unit IDs in the NEI. This spreadsheet file named "ETS-Only Units to Match to V3.xls"
identifies the ETS-only units for which EPA was unable to match to S/L inventories.
Documentation for the Final 1999 National Emissions
Inventory (Vgrsion 3) for Criteria Air Pollutants and Ammonia 10
Point Source Methodologies
-------
I. WHAT IS THE RELATIONSHIP BETWEEN THE NEI AND eGRID?
The EPA's CAMD prepares the Emissions & Generation Resource Integrated Database (eGRID) which
is a comprehensive source of data on the environmental characteristics of all electric power generated in the
United States.1 The latest version, eGRID2000, is the first complete database of emissions and resource mix
for virtually every power plant and company that generates electricity in the United States.
The data fields in this database are provided in Table 4. Further definition of the field names is provided in
the eGRID technical support documentation (File Name = EGRID00v2TSD.PDF) available at
http://www.epa.gov/airmarkets/egrid/index.html. The data included in this database is a subset of the data
provided in the eGRID2000 plant-level summary (File Name = EGRDOO.xls, spreadsheet name =
EGRDPLNT) which is available at http://www.epa.gov/ airmarkets/egrid/index.html. The EPA removed the
ORIS IDs for sites that operated in 2000 but not 1999 and added ORIS IDs for sites that operated in 1999
but not 2000. Note that eGRID2002 data for 1996-2000 (in spreadsheet and PC-based data browser
format) should be completed by March 31, 2003.
1 E-GRID Fact Sheet: http://www.epa.gov/airmarkets/egrid/factsheet.html
Documentation for the Final 1999 National Emissions
Inventory (Vgrsion 3) for Criteria Air Pollutants and Ammonia 11
Point Source Methodologies
-------
Table 5. McUl Names :ind Definitions in OUISKaoililvCodes Database
Field Name
Field Definition
ORISPL
ORIS Plant Code in eGRID1999
FIPST
State FIPS code (text field for matching to NEI)
FIPSCNY
County FIPS code (text field for matching to NEI)
ORISPL_ln_NEI_V300_Draft
Identifies the codes that have been entered in the ORIS Facility Code field in the Site table of draft
Version 3 of the 1999 NEI
SitelD_ln_NEI_V300_Draft
Identifies the site ID associated with the ORIS Facility Code in the Site table of draft Version 3 of the 1999
NEI
FIPST_NUM
State FIPS code (numeric field)
FIPSCNY_NUM
County FIPS code (numeric field)
PSTATABB
Postal code abbreviation of the State where the plant is located
CNTYNAME
County name
PNAME
Plant name
OPRNAME
Plant operator name
OPPRNAME
Operator parent company name
PCANAME
Physical (operator) based power control area name
PLTYPE
Plant type (utility or nonutility)
PREVUTIL
Previously a utility plant flag (This field has a value of 1 if the plant has been sold by a utility to a nonutility;
it has a value of 0 otherwise)
LAT
Plant latitude
LON
Plant longitude
NUMBLR
Number of utility boilers within a plant
NUMGEN
Number of generators within a plant
SOURCEM
The source(s) of emissions data for the plant. The choices are:
T = ETS/CEM NOx, S02, and C02 emissions reported to EPA;
E = Emissions estimated by applying EPA AP-42 emission factors to fuel data from EIA-767, EIA-
759/FERC-423, EIA-860B, or default values; and
Z = Plant utilizes energy resources with zero emissions.
PLPRIMFL
Plant's primary fuel based on maximum heat input or assignment (if plant is solar, wind, nuclear,
geothermal, or hydro) (see eGRID technical support document for definition of codes used in this field).
PLFFLCTG
Plant Fossil Fuel Category (field contains a "C" if PLPRIMPL is derived from coal, "O" if it is derived from oil,
or "G" if it is derived from gas. The value is blank otherwise. Fossil Fuel refers to any naturally occurring
organic fuel, such as petroleum, coal, and natural gas)
CAPFAC
Plant capacity factor
BOILCAP
Utility plant boiler capacity (MMBtu/hr)
NAMEPCAP
Plant generator capacity (MW)
CHPFLAG
Combined heat and power (CHP) (cogenerator) 1998 plant flag
USETHRMO
CHP plant 1998 useful thermal output (MMBtu)
PWRTOHT
CHP plant 1998 power to heat ratio
ELCALLOC
CHP plant 1998 electric allocation factor
PLNUCONN
Nonutility plant connected to grid flag (Y=Yes, N = No)
J. HOW DID EPA ADD PM-RELATED POLLUTANT EMISSIONS TO MY INVENTORY?
The NEI requires reporting of primary (PRI) PM10/PM25 emissions, which is the sum of the filterable
(FIL) PM10/PM25 and condensible (CON) fractions of PM10/PM25 for each emissions source of
PM10/PM25. S/L/T agencies, however, may report any fraction of total PM and/or PM10/PM25 emissions
Documentation for the Final 1999 National Emissions
Inventory (Vgrsion 3) for Criteria Air Pollutants and Ammonia 12
Point Source Methodologies
-------
for the 1999 NEI. The S/L/T inventories included emissions data for either PM-FIL, PM-PRI, PM 10-FIL,
PM10-PRI, PM25-FIL, PM25-PRI, and/or PM-CON. From any one of these pollutants, the augmentation
procedure estimates PM10-PRI, PM10-FIL, PM25-PRI, PM25-FIL, and PM-CON as shown in the
following table. If not included in a S/L/T inventory, PM10-PRI and PM25-PRI were calculated by summing
the emissions for PM10-FIL and PM-CON or PM25-FIL and PM-CON, respectively. For utilities, PM
information was added using the method in section I.
PM Category
Description
PM-PRI
Primary PM (includes filterables and condensibles)
PM-CON
Primary PM, Condensible Portion Only (all less than 1 micron)
PM-FIL
Primary PM, Filterable Portion Only
PM10-PRI
Primary PM10 (includes filterables and condensibles)
PM10-FIL
Primary PM10, Filterable Portion Only
PM25-PRI
Primary PM25 (includes filterables and condensibles)
PM25-FIL
Primary PM25, Filterable Portion Only
The following explains the QA issues identified and resolved in S/L/T inventories prior to applying the
augmentation procedures, followed by a summary of the methods for augmenting the point source NEI with
PM10-PRI, PM10-FIL, PM25-PRI, PM25-FIL, and PM-CON emissions missing from S/L/T inventories.
1. Resolution of QA Issues Identified for PM-Related Pollutants in S/L/T Inventories
For S/L inventories that contained inconsistent emissions (e.g., PM25>PM10 or PM10>PM), the
agencies provided corrections to resolve the inconsistencies prior to applying the augmentation procedures.
All PM-PRI and PM-FIL records were removed from the input file for the PM augmentation procedure if
an emission source had any form of PM10, PM25, or PM-CON. PM-CON records with emissions greater
than zero were used in the PM augmentation procedure.
2. Methods
The procedures for point source SCCs include the following five steps which are documented in detail in
Appendix C to this report:
Step 1: Prepare S/L/T PM and PM10 Emissions for Input to the PM Calculator
Step 2: Develop and Apply Source-Specific Conversion Factors
Step 3: Prepare Factors from PM Calculator
Step 4: Develop and Apply Algorithms to Estimate Emissions from S/L/T Inventory Data
Step 5: Review Results and Update the NEI with Emission Estimates and Control Information
Documentation for the Final 1999 National Emissions
Inventory (Vgrsion 3) for Criteria Air Pollutants and Ammonia 13
Point Source Methodologies
-------
The EPA has reviewed the SCCs for which PM-CON is estimated and eliminated PM-CON from the
calculation for any SCCs it decides are not sources of PM-CON.
Additionally, there are a number of SCCs that are not included in the augmentation procedure. For these
SCCs the following logic was used to create a complete set of PM values. If the state supplied PM10-PRI,
PMIO-FIL was set to PM10-PRI. If the state supplied PMIO-FIL, PM10-PRI was set to PM10-FIL.
Similar logic was followed for PM25. If the state supplied only PM10 or only PM25 value(s), PM10 was set
to PM25 or vice versa.
K. HOW DID EPA ADD NH3 EMISSIONS TO STATE/LOCAL INVENTORIES?
In addition to the criteria pollutants, the NEI also houses estimates of NH3 emissions. In 1999, only a few
of the S/L/T agencies submitted NH3 emissions. As a consequence, the NH3 emissions from final Version 2 of
the 1999 NEI were added. Tribal point source inventories were not affected by this procedure since the sites
in Tribal inventories are new additions to the NEI.
For the NH3 augmentation procedure, EPA prepared a crosswalk to match sites with NH3 emissions in
final Version 2 of the NEI to sites in the S/L inventories. If the S/L site had NH3 emissions, the S/L NH3
emissions were maintained in Version 3. If a S/L site did not have NH3 emissions but it did have NH3
emissions in final Version 2 of the NEI, the NH3 emissions in final Version 2 were added to the S/L site in
Version 3. This required adding process-level records to the Emission table. Site-level emissions from final
Version 2 were allocated to the same site in draft Version 3 by using the S/L process-to-site level ratio for the
pollutants in the S/L inventory in the following priority order: NOX, CO, VOC, S02, and PM-related
pollutants. The pollutant used to perform the allocation for an individual site is identified in the data source code
for the NH3 augmentation procedure in the second blank field of the Access97 database.
The crosswalk that EPA developed to match sites with NH3 emissions in S/L inventories with those in the
1999 NEI is provided in an Access database named "NH3_V2_to_DV3_Xwalk.mdb." Table 5 provides the
names and definitions of the fields in the database. This database contains a table NH3_V2_to_DV3_Xwalk
that provides links between the site-level emissions in final Version 2 and Version 3 of the 1999 NEI. The
matchtype field (see the field definitions below) contains the information on the type of match between the final
Version 2 site and the Version 3 site. "300_HAS_NH3" Match Type indicates that a site in draft Version 3
has NH3 emissions. If the Version 3 site has NH3 emissions, its NH3 emissions were maintained in the NEI;
however, a site match with final Version 2 may still be documented. "300_NO_NH3" Match Type indicates
that the Version 3 site does not have NH3 emissions and that the Version 2 site does have NH3 emissions.
The Version 3 site was augmented with the final Version 2 emissions as described above. "NONE" Match
Type indicates that a final Version 2 site with NH3 emissions did not have a match with a Version 3 site.
Additional information concerning multiple matches is stored in the matchmultiple field (see field definition in
Table 5).
Documentation for the Final 1999 National Emissions
Inventory (Vgrsion 3) for Criteria Air Pollutants and Ammonia 14
Point Source Methodologies
-------
'I'iihie 6. Meld N nines mid Definitions in MI3_Y2_lo_l)V3_\w:ilk Dsilnhsise
Field Name
Field Definition
P200_STAT E_FIPS
State Fips for the 1999 final Version 2 site
This field and other Version 2 fields below are null when a match was not
found for a Version 3 site.
P200_COU NTY_FI PS
County Fips for the Version 2 site
P200 SITE ID
Site ID for the Version 2 site
P200_ORIS_FACILITY_CODE
ORIS Facility Code for the Version 2 site
P200 SIC PRIMARY
Primary SIC for the Version 2 site
P200_FACILITY_NAME
Facility Name of the Version 2 site
P200 SITE DESCRIPTION
Description of the Version 2 site
P200_STREET_LINE_1
Street Address of the Version 2 site
P200 STREET LINE 2
Street Address of the Version 2 site
P200_ST R E ET_L IN E_3
Street Address of the Version 2 site
P200_CITY
City of the Version 2 site
P200_STATE
State of the Version 2 site
P200_ZIP_CODE
Zip Code of the Version 2 site
P200_NH3_EMISSION_TON_SUM
Site Level Sum of Annual NH3 Emissions from Version 2
MATCH_TYPE
Type of Match between the 1999 Version 2 site and the 1999 Version 3 site.
Valid Data for this field are as follows:
300_HAS_NH3
This indicates that the Version 3 has NH3 emissions associated with it.
If the Version 3 site has NH3 emissions, it will not be augmented, however, a
Site match with Version 2 may still be documented.
300_NO_NH3
This indicates that the Version 3 site does not have NH3 emissions.
NONE
This indicates that a Version 2 site with NH3 emissions did not have a match
With a Version 3 site.
MATCH_MULTIPLE
Indicates whether the match is a multiple match between the 1999 Version 2 site
and the 1999 Version 3 site.
Valid Data for this field are as follows:
200_M_to_1_300
This indicates multiple Version 2 site match to a single Version 3 site.
200_1_to_M_300
This indicates a single Version 2 site matches to multiple Version 3 sites.
MATCH_RATIO
Indicates the ratio used to calculate the NH3 to apply value where a single Version 2 site is
applied to multiple Version 3 Sites
N H3_TO_APPLY_EMISSION_TON_SUM
Amount of NH3 emissions to be distributed over a Version 3 Site. This value is
null in cases where there is not Version 2 match for a Version 3 Site. It is also null
in the case where there is a multiple Version 2 site match to a single Version 3 site
match (200_M_to_1_300). In the event of a 200_M_to_1_300 case, the emissions are
summed for the multiple sites and place on a single occurrence of the Version 3 site
match.
P300_STAT E_FIPS
State Fips for the 1999 Version 3 site
This field and other Version 3 fields below are null when a match was not
found for a Version 2 site.
P300_COU NTY_FI PS
County Fips for the Version 3 site
P300_SITE_ID
Site ID for the Version 3 site
P300 ORIS FACILITY CODE
ORIS Facility Code for the Version 3 site
P300_SIC_PRIMARY
Primary SIC for the Version 3 site
P300 FACILITY NAME
Facility Name of the Version 3 site
P300_SITE_DESCRIPTION
Description of the Version 3 site
P300_ST R E ET_L IN E_1
Street Address of the Version 3 site
Documentation for the Final 1999 National Emissions
Inventory (Vgrsion 3) for Criteria Air Pollutants and Ammonia 15
Point Source Methodologies
-------
'I'iihie 6. Meld N nines mid Definitions in MI3_Y2_lo_l)V3_\w:ilk Dsilnhsise
Field Name
Field Definition
P300 STREET LINE 2
Street Address of the Version 3 site
P300_STREET_LINE_3
Street Address of the Version 3 site
P300 CITY
City of the Version 3 site
P300_STATE
State of the Version 3 site
P300 ZIP CODE
Zip Code of the Version 3 site
P300 NH3 EMISSION TON SUM
Site Level Sum of Annual NH3 Emissions from Version 3
L. HOW DID EPA FILL IN MISSING OR CORRECT INVALID INFORMATION FOR
MANDATORY AND NONMANDATORY FIELDS?
After incorporating S/L/T comments on draft Version 2 of the NEI, some inventories contained invalid MF 2.0
codes. The following explains how EPA changed invalid codes to make the codes MF 2.0 compliant:
Standard Industrial Classification (SIC) Code in SI and EU Tables
Invalid pre-1987 codes were corrected to valid 1987 codes if possible (e.g., where there was a one-to-
one conversion). Otherwise, invalid codes were changed to the miscellaneous SIC code of 9999.
North American Industry Classification System CNAICS) Code in SI and EIJ Tables
Invalid codes or unknown codes were changed to 99, 999, 9999, or 99999.
Invalid MACT Code in SI Table
Invalid codes in the MACT Code field were changed to null (blank) values because this is not a
mandatory field for the criteria air pollutant NEI.
Design Capacity Unit Numerator/Denominator in the EU Table
Where non-standard units were given (i.e. not in the EPA lookup table) - an effort was made to determine
an appropriate translation. If no valid unit replacement was found, the values were nulled.
SCCs in EP Table
During preparation of Version 2 of the NEI, EPA replaced SCCs in a number of states that are not in
EPA's master list of SCCs with the generic SCC 79900101. This SCC also appears in the Version 3
inventory submittals for MN, NJ, and SC. Any generic SCCs that were not replaced during the comment
period were removed, with the exception of NJ SCCs which were identified and replaced with an existing
SCC.
Documentation for the Final 1999 National Emissions
Inventory (Vgrsion 3) for Criteria Air Pollutants and Ammonia 16
Point Source Methodologies
-------
MACT Compliance Status Code in EP Table
Invalid codes in the MACT Compliance Status Code field were changed to null (blank) values because
this is not a mandatory field for the criteria air pollutant NEI.
Stack Parameters
In S/L/T inventories processed for Version 3 of the NEI, the Emission Release Point Type field was
frequently left blank. In addition, checks on stack parameters (when provided) frequently did meet EPA QA
criteria. Due to time and resource constraints for preparing final Version 3, EPA was unable to address and
resolve all of the QA issues identified. Thus, EPA applied the following steps to resolve QA issues with stack
parameters:
If a record contained stack parameters and the Emission Release Point Type field was designated as
a "stack", EPA applied the NEI Quality Assurance and Data Augmentation Steps to adjust stack
parameters to fall within specified ranges, and to fulfill consistency checks.
If a site had NOX emissions of 1,000 tons or more and the Emission Release Point Type field was
null, the Emission Release Point Type field was set to "stack" and the stack parameter fields were
filled in with default values using the NEI Quality Assurance and Data Augmentation Steps.
If the Emission Release Point Type field was null and the site had NOX emissions of less than 1,000
tons, the Emission Release Point Type field was set to "fugitive", the stack parameters were erased if
they existed, and the fugitive release point parameters were filled in with default values.
In final Version 3, it was determined that certain states had previously submitted data misidentifying the
emission release type as fugitives for all emission release points. For these states (NC ), their original stack
information was reapplied and their emission types reset.
Coordinates in ER Table
For final Version 3, EPA incorporated comments that S/L/T agencies supplied to correct coordinates in
draft Version 3. Then EPA used GIS software to map the coordinates, identified coordinates that did not map
within the boundary of the designated county, and then either corrected the coordinates using in-house data or
corrected the county FIPS code. If the coordinates could not be corrected with in-house data or were missing,
coordinates for the county centroid were used in final Version 3 of the 1999 NEI. All coordinates were
converted to LATLON.
EM Reliability Indicator. Emission Calculation Method, and Rule Effectiveness Method Code in EM
Table
Documentation for the Final 1999 National Emissions
Inventory (Vgrsion 3) for Criteria Air Pollutants and Ammonia 17
Point Source Methodologies
-------
Invalid codes in these field were changed to null values because they are not mandatory fields.
Factor Unit Numerator/Factor Unit Denominator in the EM Table
Where non-standard units were given (i.e. not in the EPA lookup table) - an effort was made to determine
an appropriate translation. If no valid unit replacement was found, the values were nulled.
Throughput Unit Numerator in PE Table
Where non-standard units were given (i.e. not in the EPA lookup table) - an effort was made to
determine an appropriate translation. If no valid unit replacement was found, the values were nulled.
Material Codes and Material I/O codes in PE and EM Tables
Where material information was provided that was not consistent with EPA standard units this information
was altered as appropriate. Where an appropriate value was not able to be determined, these values were
nulled out.
Control Efficiency Data and Control Device Codes in the CE Table
The following assumptions were applied to added control efficiency data to the "Total Capture Control
Efficiency" field in the CE table:
Primary PCT Control
Efficiency (PPCE)
PCT Capture Efficiency
(PCE)
Total Capture Control
Efficiency (TCCE)
Assumption
Data
Data
No data
TCCE=PPCE x PCE
Data
No data
No data
TCCE=PPCE
No data
Data
No data
Not enough information,
assumed uncontrolled
Data
No data
Data
Leave as is
No data
Data
Data
Leave as is
No data
No data
Data
Leave as is
Primary Device Type and Secondary Device Type codes that did not exist in the MF 2.0 code list were
reviewed to determine if the codes could be corrected to those in the MF 2.0 code list. If the codes could not
be corrected, and the S/L/T agency did not provide information on how to correct the codes, the codes were
set to 099 for miscellaneous control devices.
Documentation for the Final 1999 National Emissions
Inventory (Vgrsion 3) for Criteria Air Pollutants and Ammonia 18
Point Source Methodologies
-------
2. Scrubbing of Seasonal Throughput and Operating Parameter Values
Prior to running the scripts to calculate OSD emissions, the fields holding data used in the calculations are
reviewed to identify QA issues. The following explains the QA issues that occurred in S/L/T inventories and
how the QA issues were resolved (or scrubbed).
Operating Parameters
QA issues with data in the Annual Avg Days Per Week, Annual Avg Hours Per Day, Annual Average
Hours Per Year, and Annual Avg Weeks Per Year fields are diagnosed and corrected as follows if the annual
emissions are greater than zero. If annual emissions are zero, the values in these fields are set to zero.
Days per week:
Hours per day:
Hours per year:
Weeks per year:
Value is set to 7 if the existing value is greater than 7, equal to zero, or null.
Value is set to 24 if the existing value is greater than 24, equal to zero, or null.
Value is set to 8760 (non-leap year) or 8784 (leap year) if the existing value is
greater than 8760 or 8784, equal to zero, or blank (null).
Value is calculated using the following equation:
Annual average weeks per year =
(annual average hours per year / annual average hours per day) /
annual average days per week
The fields used in the calculation are scrubbed to correct QA issues before
calculating annual average weeks per year. The calculation is completed if the
values in each of the three fields are greater than zero, the weeks per year field
has a zero or blank (null) value, and the calculated value is less than or equal to
52 weeks per year.
Seasonal Throughputs
QA issues with data in the Winter Throughput Pet, Spring Throughput Pet, Summer Throughput Pet,
and Fall Throughput Pet fields in the EP table are diagnosed and corrected as follows:
Step 1 - Sets Winter, Spring, Summer, and Fall Throughput Percentages to positive values if they
contain negative values.
Step 2 - If the seasonal throughput percentages are equal to zero or a blank (null) value in all four
fields, the seasonal throughput percentages are taken from EPA's Temporal Allocation Factor File
(TAFF). The TAFF provides default seasonal throughput fractions by SCC. The fields in the TAFF
that are used are as follows:
Documentation for the Final 1999 National Emissions
Inventory (Vgrsion 3) for Criteria Air Pollutants and Ammonia 19
Point Source Methodologies
-------
Winter: SEACODE = 4, and DAY CODE = 1
Spring: SEA CODE = 1, and DAY CODE = 1
Summer: SEA CODE = 2, and DAY CODE = 1
Fall: SEA CODE = 3, and DAY CODE = 1
Step 3 - If the TAFF does not contain an SCC in the inventory and all four seasonal throughput values
are zero or blank (null), the values are set to 25% for each of the four seasons.
Step 4 - If the sum of the seasonal throughput values for all four seasons does not equal 100%, this
step normalizes the values to sum to 100%.
Step 5 - All throughput percentages are set to zero if annual emissions are zero.
M. DID EPA REJECT ANY INFORMATION IN STATE/LOCAL INVENTORIES?
Yes. EPA does not recognize "777" as a valid county FIPS code. All records for this FIPS code
were removed from the inventories that contained them.
N. WHERE DO I FIND DOCUMENTATION OF EPA POINT SOURCE
METHODOLOGIES?
Documentation of S/L inventory submittals prior to this year and methods for growing emissions to
prepare the 1999 NEI are provided in "Current Methods Used to Estimate Emissions, 1985-1999 Procedures
Document for National Emission Inventory, Criteria Air Pollutants 1985-1999," March 2001, EPA-454/R-01-
006. This document is available from the following EPA Web Site:
http://www.epa.gov/ttn/chief/trends/index.html.
-------
APPENDIX A
DOCUMENTATION OF CHANGES MADE TO INDIVIDUAL
STATE/ LOCAL/TRIBAL INVENTORIES
Documentation for the Final 1999 National Emissions
Inventory (Version 3) for Criteria Air Pollutants and Ammonia A-l
Point Source Methodologies
-------
Arizona: Criteria Pollutants
AZ included annual emissions and related data for carbon monoxide (CO), ammonia (NEB), oxides of
nitrogen (NOX), PM10-PRI, sulfur dioxide (S02), and volatile organic compounds (VOC) for sites in 10 of
the 15 counties in AZ. AZ's initial inventory submittal did not include NOX and PM-related pollutant
emissions. AZ indicated that their initial submittal excluded these pollutants due to a data base conversion
error. Subsequently, AZ provided an inventory that included NOX and PM-related pollutant emissions to
replace their original submittal. This replacement inventory was processed and incorporated into Version 3 of
the NEI.
A. Initial Submittal Processing
The United States (U.S.) Environmental Protection Agency (EPA) compared the number of sites by
county in AZ's inventory to the number of sites by county in final Version 2 (originating from Version 1.5) of the
NEI and noticed a significant difference in the number of sites in some counties. AZ noted that its inventory for
Version 3 did not include all of the sites that operated in 1999 in some counties. Thus, AZ provided further
instructions on what sites in final Version 2 to keep in draft Version 3 of the NEI.
AZ provided corrections to coordinates after submitting their inventory. The coordinates were
incorporated into draft Version 3 of the NEI. ORIS IDs were added to the ORIS Facility Code field.
Per instruction from AZ a number of SCC changes were made.
B. Comments Processing
AZ did not provide further comments.
Documentation for the Final 1999 National Emissions
Inventory (Version 3) for Criteria Air Pollutants and Ammonia A-2
Point Source Methodologies
-------
Arizona - Salt River Pima-Maricopa Indian Community (ITEP Tribe ID = 433): Criteria Pollutants
The Salt River Pima-Maricopa Indian Community included annual emissions and related data for CO,
nonmethane organic compounds (NMOC), NOX, PM10-PRI, S02, oxides of sulfur (SOX), and VOC for 16
sites.
A. Initial Submittal Processing
Through consultation with the Institute of Tribal Environmental Professionals (ITEP) who submitted this
inventory to EPA, it was determined that all of the sites in this inventory are located in Maricopa County
(Federal Information Processing Standards [FIPS] code 013). These sites were added to the sites that
Maricopa County submitted last year that are included in Final Version 2 of the 1999 NEI. However, the site
IDs in the two inventories were similar. Therefore, an "R-" was added as a prefix to site IDs in the inventory
for the Salt River Pima-Maricopa Indian Community to avoid duplicating sites in Version 3 of the 1999 NEI.
In addition, the years in the start date and end date of the Salt River Pima-Maricopa Indian
Community's inventory were 1997 and 1998. Through consultation with ITEP who prepared the inventory and
EPA, it was agreed that the inventory was representative of a 1999 base year.
Added county FIPS code 013. Added an "R-" to beginning of Site IDs. Changed the Inventory Year
from 1998 to 1999. Changed the year in the Start Date and End Date fields from 1997 and 1998 to 1999.
B. Comments Processing
Arizona - Salt River Pima-Maricopa Indian Community did not provide further comments.
Documentation for the Final 1999 National Emissions
Inventory (Version 3) for Criteria Air Pollutants and Ammonia A-3
Point Source Methodologies
-------
Delaware: Criteria Pollutants
DE included annual and ozone season day (OSD) emissions and related data for CO, NH3, NOX,
PM10-PRI, PM-PRI, S02, and VOC.
A. Initial Submittal Processing
The inventory submitted by DE replaced the inventory in final Version 2 of the NEI for all three
counties in the State. ORIS IDs were added to the ORIS Facility Code field. Per instruction from DE
several SCC changes were made.
B. Comments Processing
Delaware did not provide further comments.
Documentation for the Final 1999 National Emissions
Inventory (Version 3) for Criteria Air Pollutants and Ammonia A-4
Point Source Methodologies
-------
Georgia: Criteria Pollutants
GA included annual emissions and related data for CO, NOX, and VOC for 29 of its 159 counties.
A. Initial Submittal Processing
For the 29 counties for which GA provided data, GA included annual throughput data in its inventory.
Therefore, EPA used the throughput data and selected emission factors from Factor Information Retrieval
(FIRE) and AP-42 to calculate S02 and PM-related pollutant emissions. As a part of this process, GA
assisted EPA in resolving SCC issues where the original SCC units did not match emission factor units. GA's
inventory was updated to correct the SCCs. In addition, GA identified sites in Version 2 of the NEI to keep in
Version 3 because the sites were missing from its original inventory submittal. A limitation associated with the
S02 and PM-related pollutant emission estimates is that GA was unable to provide Control Equipment data for
processes that had S02 or PM controls.
GA supplied a number of updates to facility names and zip codes which were implemented. ORIS IDs
were added to the ORIS Facility Code field. Per instruction from GA several SCC changes were made.
There were a few SCC and PM combinations missed during the PM Augmentation procedures. In many
cases, these PM combinations occurred during the generation of the PM for the state as describe above.
These PM combinations were corrected for final Version 3.
B. Comments Processing
Georgia did not provide further comments.
Documentation for the Final 1999 National Emissions
Inventory (Version 3) for Criteria Air Pollutants and Ammonia A-5
Point Source Methodologies
-------
Idaho: Criteria Pollutants
ID included annual emissions and related data for CO, NH3, NMOG, NOX, PMIO-FIL, PM-FIL,
S02, SOX, total organic gases (TOG), and VOC
A. Initial Submittal Processing
The inventory submitted by ID replaced the inventory in final Version 2 of the NEI for all of the
counties included in ID's inventory submittal. ORIS IDs were added to the ORIS Facility Code field. Per
instruction from ID several SCC changes were made. Three duplicate records were removed from the EP
table. The first record inserted into the system was kept. Per instructions from ID, the EM table was revised
to correct cases where PMIO-FIL emissions were greater than PM-FIL emissions.
B. Comments Processing
Documentation for the Final 1999 National Emissions
Inventory (Version 3) for Criteria Air Pollutants and Ammonia A-6
Point Source Methodologies
-------
Louisiana: Criteria Pollutants
A. Initial Submittal Processing
LA initial submittal was not processed.
B. Comments Processing
Documentation for the Final 1999 National Emissions
Inventory (Version 3) for Criteria Air Pollutants and Ammonia A-7
Point Source Methodologies
-------
Massachusetts: Criteria Pollutants
A. Initial Submittal Processing
MA did not submit an initial file.
B. Comments Processing
Documentation for the Final 1999 National Emissions
Inventory (Version 3) for Criteria Air Pollutants and Ammonia A-8
Point Source Methodologies
-------
Michigan: Criteria Pollutants
MI included annual emissions and related data for CO, NOX, PM10-PRI, PM25-PRI, PM-PRI,
S02, and VOC.
A. Initial Submittal Processing
The inventory submitted by MI replaced the inventory in final Version 2 of the NEI for all of the
counties included in Mi's inventory submittal. ORIS IDs were added to the ORIS Facility Code field. Per
instruction from MI a number of SCC changes were made. Per instructions from MI, the EM table was
revised to correct cases where PM25-PRI emissions were greater than PM10-PRI emissions and cases where
PM-PRI emissions were provided which were less than PM10-PRI.
B. Comments Processing
Documentation for the Final 1999 National Emissions
Inventory (Version 3) for Criteria Air Pollutants and Ammonia A-9
Point Source Methodologies
-------
Minnesota: Criteria Pollutants
MN included annual emissions and related data for CO, NOX, NH3, PM10-PRI, S02, and VOC.
A. Merging of State and NEI Version 2 Inventories
The inventory submitted by MN replaced the inventory in final Version 2 of the NEI for all of the
counties included in MN's inventory submittal. ORIS IDs were added to the ORIS Facility Code field.
Per instruction from MN several SCC changes were made.
B. Comments Processing
Documentation for the Final 1999 National Emissions
Inventory (Version 3) for Criteria Air Pollutants and Ammonia A-10
Point Source Methodologies
-------
New Jersey: Criteria Pollutants
NJ included annual, OSD, and CO season day emissions and related data for CO, NOX, PM-PRI,
PM10-PRI, S02, and VOC.
A. Initial Submittal Processing
The inventory submitted by NJ replaced the inventory in final Version 2 of the NEI for all of the
counties included in NJ's inventory submittal. The file NJ submitted had several referential integrity
inconsistencies. It was determined that the best approach would be to rebuild the inventory from the
information in the EM table and then update the resulting parent tables (and CE table). Per direction from
NJ, several updates were made to the primary SIC. ORIS IDs were added to the ORIS Facility Code field.
Parent EP records were generated for orphaned EM records. Set primary device type equal to 099 where left
blank.
B. Comments Processing
New Jersey did not provide further comments
Documentation for the Final 1999 National Emissions
Inventory (Version 3) for Criteria Air Pollutants and Ammonia A-11
Point Source Methodologies
-------
New York: Criteria Pollutants
NY included annual and OSD emissions and related data for CO, NH3, NMOC, NOX, PM-PRI,
PM10-PRI, S02, and VOC.
A. Initial Submittal Processing
The inventory submitted by NY replaced the inventory in final Version 2 of the NEI for all of the
counties included in NY's inventory submittal. Per direction from NY, several updates were made to the
primary SIC. ORIS IDs supplied by NY were added to the ORIS Facility Code field. Parent EP records
were generated for orphaned EM records. Duplicate records were inactivated. The record inserted into the
system first was retained. Several EP records had blank emission release point ID fields. These fields were
filled in with the following combination "S-" + Emission Unit ID + + Process ID in order to facilitate the
generation of ER parent records for orphaned EP records. Duplicate ER records were deleted. The ER
record with the highest exit gas flow rate was kept. Duplicate records were inactivated. The record with the
lowest primary key was retained. Pollutant code 7664417 was changed to NH3. Duplicate emission records
were removed, keeping the emission record with the highest value. Set primary device type equal to 099
where left blank.
B. Comments Processing
New York did not provide further comments
Documentation for the Final 1999 National Emissions
Inventory (Version 3) for Criteria Air Pollutants and Ammonia A-12
Point Source Methodologies
-------
Nevada: Criteria Pollutants
NV included annual emissions and related data for CO, NH3, NOX, PM-PRI, PM10-PRI, S02, and
voc.
A. Initial Submittal Processing
The inventory submitted by NV partially replaced the inventory in final Version 2 of the NEI for all of
the counties included in NV's inventory submittal. NV provided guidance on what sites in final Version 2 to
keep in draft Version 3 because the sites were not included in their submittal for draft Version 3. ORIS IDs
were added to the ORIS Facility Code field. Per instruction from NV a number of SCC changes were
made.
B. Comments Processing
Documentation for the Final 1999 National Emissions
Inventory (Version 3) for Criteria Air Pollutants and Ammonia A-13
Point Source Methodologies
-------
Ohio: Criteria Pollutants
A. Initial Submittal Processing
OH did not submit an initial file.
B. Comments Processing
Documentation for the Final 1999 National Emissions
Inventory (Version 3) for Criteria Air Pollutants and Ammonia A-14
Point Source Methodologies
-------
North Carolina - Beaufort County (County FIPS 013): Criteria Pollutants
Beaufort County, NC included annual emissions and related data for CO, NH3, NOX, PM10-PRI,
PM-PRI, S02, and VOC for one site (Site ID 3701300071).
A. Initial Submittal Processing
The data for the site were added to the emissions data included in the State of NC's inventory included
in final Version 2 of the 1999 NEI. ORIS ID 50509 was added to the ORIS Facility Code field for the
one site.
B. Comments Processing
North Carolina - Beaufort County did not provide further comments.
Documentation for the Final 1999 National Emissions
Inventory (Version 3) for Criteria Air Pollutants and Ammonia A-15
Point Source Methodologies
-------
North Carolina - Forsyth County (County FIPS 067): Criteria Pollutants
Forsyth County, NC included annual emissions and related data for CO, NH3, NOX, PM10-PRI,
PM-PRI, S02, and VOC.
A. Initial Submittal Processing
The emissions data in final Version 2 of the NEI were completely replaced with the emissions data
included in Forsyth County's inventory. ORIS IDs were added to the ORIS Facility Code field.
B. Comments Processing
North Carolina - Forsyth County did not provide further comments.
Documentation for the Final 1999 National Emissions
Inventory (Version 3) for Criteria Air Pollutants and Ammonia A-16
Point Source Methodologies
-------
North Carolina - Mecklenburg County: Criteria Pollutants
A. Initial Submittal Processing
Pennsylvania - Allegheny County: Criteria Pollutants
A. Initial Submittal Processing
Pennsylvania - Allegheny County did not submit an initial file.
B. Comments Processing
Documentation for the Final 1999 National Emissions
Inventory (Version 3) for Criteria Air Pollutants and Ammonia A-17
Point Source Methodologies
-------
South Carolina: Criteria Pollutants
SC included annual emissions and related data for CO, hydrocarbon (HC), NH3, NOX, PM10-FIL,
PM25-FIL, PM-FIL, S02, and VOC.
A. Initial Submittal Processing
The inventory submitted by SC replaced the inventory in final Version 2 of the NEI for all three
counties in the State. The SC left a few sites out of their V3 inventory submittal, in that case the Version 2 sites
were kept. ORIS IDs were added to the ORIS Facility Code field. Per instruction from SC a number of
SCC changes were made. Pollutant code 7664417 was changed to NH3. PM emissions were adjusted
as directed by SC where PM25-FIL was greater than PM10-FIL or PM-FIL was less than PM10-FIL or
PM25-FIL.
B. Comments Processing
Documentation for the Final 1999 National Emissions
Inventory (Version 3) for Criteria Air Pollutants and Ammonia A-18
Point Source Methodologies
-------
Tennessee: Criteria Pollutants
TN included annual and OSD emissions and related data for CO, NH3, NOX, PM25-FIL, PM-FIL,
S02, and VOC. After its initial inventory submittal, TN provided a second inventory for 12 companies. This
second inventory was combined with TN's initial inventory submittal.
A. Initial Submittal Processing
For the counties that TN included in its inventory, the emissions data in final Version 2 of the NEI were
completely replaced with the emissions data included in TN's inventory. ORIS IDs were added to the ORIS
Facility Code field. Per instruction from TN a number of SCC changes were made. Additionally a number of
area source SCCs were left in the EP table to be reviewed for the final Version 3. Per instruction from TN
PM-FIL pollutants were updated to PM10-FIL. PM emissions were adjusted where PM25-FIL was greater
than PM10-FIL.
B. Comments Processing
Documentation for the Final 1999 National Emissions
Inventory (Version 3) for Criteria Air Pollutants and Ammonia A-19
Point Source Methodologies
-------
Tennessee - Davidson County (County FIPS 037): Criteria Pollutants
Davidson County, TN included annual and OSD emissions and related data for CO, NOX, PM10-
PRI, S02, and VOC.
A. Initial Submittal Processing
The emissions data in final Version 2 of the NEI were completely replaced with the emissions data
included in Davidson County's inventory submittal for Version 3. ORIS IDs were added to the ORIS Facility
Code field.
B. Comments Processing
Tennessee - Davidson County did not provide further comments.
Documentation for the Final 1999 National Emissions
Inventory (Version 3) for Criteria Air Pollutants and Ammonia A-20
Point Source Methodologies
-------
Tennessee - Shelby County (County FIPS 157): Criteria Pollutants
Shelby County, TN included annual emissions and related data for CO, HC, NH3, NOX, PM, PM10,
PM10-FIL, PM10-PRI, PM2.5, PM2.5-FIL, PM25-FIL, PM25-PRI, PM-FIL, PM-PRI, S02, S02, SOX,
total suspended particulate (TSP), and VOC.
A. Initial Submittal Processing
Due to referential integrity issues that could not be resolved within the timeframe this inventory was not included
in Draft Version 3.
B. Comments Processing
Shelby County resubmitted the inventory for final version 3. There were referential integrity issues; however,
the focus was on retaining accurate emissions. Therefore referential integrity issues and duplicates were
managed as described in section D.
Documentation for the Final 1999 National Emissions
Inventory (Version 3) for Criteria Air Pollutants and Ammonia A-21
Point Source Methodologies
-------
West Virginia: Criteria Pollutants
A. Initial Submittal Processing
West Virginia did not submit an initial file.
B. Comments Processing
Documentation for the Final 1999 National Emissions
Inventory (Version 3) for Criteria Air Pollutants and Ammonia A-22
Point Source Methodologies
-------
APPENDIX B
1999 NEI METHODS FOR ELECTRIC UTILITY STEAM GENERATED
FOSSIL FUEL COMBUSTION UNITS
Documentation for the Final 1999 National Emissions
Inventory (Vgrsion 3) for Criteria Air Pollutants and Ammonia B -1
Point Source Methodologies
-------
A. FUEL COMBUSTION - ELECTRIC UTILITY
1. Which sources does the United States (U.S.) Environmental Protection Agency (EPA) include in
the Fuel Combustion - Electric Utility category?
The point and area source categories under the "Electric Utility" heading include the following Tier I and
Tier II categories:
Tier I Category Tier II Category
(01) FUEL COMBUSTION - ELECTRIC UTILITY (01) Coal
(02) Oil
(03) Gas
(04) Other Fuels including
wood, waste, etc.
The emissions from the combustion of fuel by electric utilities are divided into two classifications: (1) steam
generated fossil-fuel units (boilers) with Source Classification Codes (SCCs) = lOlxxxxx; and (2) non-steam
generated fossil-fuel units such as gas turbines (GT) and internal combustion (IC) engines with SCCs =
201xxxxx. Estimating emissions for these two classes requires two very different methodologies, each of which
is described separately. Section 2 describes the methodology for fossil-fuel steam utility boilers. The
methodology used to estimate emissions for nonsteam generated fossil-fuel units (except for those who report
data to EPA's Emissions Tracking System/Continuous Emissions Monitoring (ETS/CEM) - they are included
in section 2) is described in section 3.
2. What emissions data for electric utilities are included in the Trends inventory?
The Trends databases for fossil-fuel steam electric utility boilers include emission estimates of volatile
organic compounds (VOC), oxides of nitrogen (NOX), carbon monoxide (CO), sulfur dioxide (S02),
particulate matter with an aerodynamic diameter less than or equal to a nominal 10 or 2.5 micrometers in
diameter (PM10 and PM2.5) for the years 1985 through 1999. In addition, ammonia (NH3) emissions were
added in 1996 and primary PM10 and PM2.5 were added in 1996. Table B-l summarizes the methods
applied to estimate emissions for each pollutant for 1989 through 1999. Table B-2 identifies the SCCs by fuel
type and boiler firing and bottom type for which emissions were estimated (prior to overlaying non-ETS/CEM
data with State data for 1999). Estimates for fossil-fuel steam electric utilities do not include emissions from the
combustion of anthracite coal because anthracite coal accounts for less than 1 percent of the overall emissions
from fuel combustion by fossil-fuel steam electric utility units. EPA does not develop emissions estimates for
sulfates (S04) because no known utility emission factors exist for this pollutant.
3. How does EPA develop emission estimates for fossil-fuel fired steam electric utilities?
Seven basic factors are used to estimate emissions for fossil-fuel steam electric utility units for the years
1985 through 1999: (1) fuel consumption; (2) the latest (September 6, 2002) EPA-approved emission factor,
Documentation for the Final 1999 National Emissions
Inventory (Version 3) for Criteria Air Pollutants and Ammonia B-2
Point Source Methodologies
-------
which relates the quantity of fuel consumed to the quantity of pollutant emitted; (3) fuel characteristics, such as
sulfur content, ash content, and heating value of fuels; (4) control efficiency, which indicates the percent of
pollutant emissions not removed through control methods; (5) rule effectiveness (which, according to EPA,
measures a regulatory program's ability to achieve all the emissions reductions that could be achieved by full
compliance with the applicable regulations at all sources at all times and is used through 1995); (6) whether
ETS/CEM data exist for S02, NOX, and heat input (from 1995 on); and (7) whether State data were
submitted (for 1999). Fuel consumption characteristics and control efficiencies are determined at the boiler-
level, whereas emission factors are specified at the SCC-level.
To derive 1999 emissions estimates, EPA first estimates the 1999 boiler-level emissions and heat input
from EIA data and AP-42 emission factors. ETS/CEM S02, NOX, and heat input values, if they exist,
overlay the EIA-767-based data; and State data, if they were submitted, overlay the EIA-767 data that were
not overlaid by ETS/CEM data. An additional 74 boilers with ETS/CEM data were added to the EIA-767
frame from the list of 1999 ETS/CEM-only boilers. These 74 ETS/CEM-only boilers had the largest NOX
emissions (> 100 tons) or were additional boilers within a plant that had boilers with NOX emissions >100
tons.
4. Where does EPA obtain the utility data necessary for emissions estimates?
Primary utility data collected by the Department of Energy's (DOE) Energy Information Administration
(EIA) serves as the basis for the fossil-fuel-fired steam electric utility component of the Trends inventory. The
EIA uses Form EIA-767 (,Steam-Electric Plant Operation and Design Report1) to collect monthly boiler-
level data on a yearly basis. Currently, data from Form EIA-767 are available for the years 1985 through
1999. Note that beginning in 1998, EIA had determined that plants that have previously reported to Form
EIA-767 must continue to do so — even if they have been sold to a nonutility, so that this inventory component
does contain some fossil-fuel steam utility boilers that are presently owned by nonutilities. This steam
component does not include data from GT or IC engines (which account for a very small share of electric utility
fuel use and corresponding emissions) unless companies report that data to EIA, or if the units are among the
74 ETS/CEM-only additional data added (see above).
The steam emission inventory data for 1985 through 1999 are initially based on the aggregated monthly
electric utility steam boiler-level data provided by Form EIA-767. All plants of at least 10 megawatts (MW)
that have at least one operating boiler are required to provide this information to EIA, although the amount of
data required from plants with less than 100 MW of steam-electric generating capacity is not as extensive as
the amount required from those plants of at least 100 MW. For plants with a generator nameplate rating from
10 MW to less than 100 MW, only those pages of Form EIA-767 containing identification (ID) information
(i.e., plant Office of the Regulatory Information System (ORIS) code, State name, county name, plant name,
operator name, boiler ID), boiler fuel quantity and quality, and flue gas desulfurization (FGD) information must
be completed. Other sources of data for NOX, S02, and heat input are used in place of the EIA-based
estimated data when the data are known to be better: EPA's ETS/CEM annual Scorecard NOX and S02
emissions and heat input overlay the EIA-based data for affected acid rain utility boilers beginning in 1995 (the
data are also available for Phase 1 units for 1994).2 For 1999, submitted State data were used for those
boilers for whom ETS/CEM data were unavailable. These sources are summarized in Table B-3.
Documentation for the Final 1999 National Emissions
Inventory (Vzrsion 3) for Criteria Air Pollutants and Ammonia B-3
Point Source Methodologies
-------
a. What data does Form EIA-767 contain?
The EIA requires that the operating utility for each plant with fossil-fuel steam utility boilers of 10 MW or
greater submit at least some sections of Form EIA-767. This form is designed so that information for each
plant is reported on separate pages that relate to different levels of data. The relevant levels of data include the
following:
! Plant-level: Delineation of the plant configuration, which establishes the number of boilers and the
IDs for each boiler, as well as the associated generator(s), FGD unit(s) (S02 scrubbers), flue gas
particulate collectors, flue(s), and stack(s). These do not necessarily have a one-to-one
correspondence. In addition, plant name, location, and operating utility are provided.
! Boiler-level: Monthly fuel consumption and quality data (for coal, oil, gas, and other), regulatory
data, and design parameters (including NOX control device and annual S02 operating efficiency).
! Generator-level: Monthly generator and maximum nameplate capacity.
! FGD-level: One page per five FGD units for annual operating data (including S02 control efficiency)
and design parameter data (including type of S02 control device).
! Flue gas particulate collector-level: One page each for (up to five) collectors with annual operating
data [including total suspended particulate (TSP) control efficiency] and design specifications
(including type of particulate control device).
! Flue- and stack-level: Design parameter data.
Form EIA-767 data for 1985 through 1999 are processed in a series of steps aimed at converting the
mainframe-level computerized data into usable database form. Only certain information is extracted. For
example, Form EIA-767 includes fuel-related boiler data such as monthly values for each fuel burned, along
with the fuel's associated sulfur, ash, and heat content. Only information regarding coal, oil, and gas fuel type
data is processed for the Trends inventory and only data from the first stack associated with a boiler is used.
Beginning with the 1998 data, EIA provided 15 database files to include the EIA-767 data, and for the first
time, all fuel types' data (including wood and refuse) were processed.
The data are aggregated for each fuel to produce annual estimates for each boiler before they are
combined with other data (such as control devices and efficiencies, plant location data, associated generator
generation, and associated stack parameters). Once SCCs are assigned to each boiler's fuel data in a given
plant, the SCC-specific data are then separated so that each new database record is on the plant-boiler-SCC
level.
Documentation for the Final 1999 National Emissions
Inventory (Version 3) for Criteria Air Pollutants and Ammonia B-4
Point Source Methodologies
-------
5. How does EPA develop the necessary data not supplied by the EIA forms?
To obtain data not contained in the computerized EIA data files, or converted to other measurement units,
algorithms (utilized since the 1980s) are used to develop values for SCC, heat input, pollutant emissions, and
NOX control efficiency.
Although Form EIA-767 reports generator nameplate capacity, this information cannot be used to
represent the boiler size when a one-to-one correspondence does not exist between boiler and generator
(referred to as a multiheader situation-for example, if one boiler is associated with two or more generators or if
several boilers are reciprocally associated with several generators). Therefore, EPA developed a boiler design
capacity variable (in MMBtu/hr) based on the reported maximum continuous boiler steam flow at 100 percent
load (in thousand pounds per hour) by multiplying the steam flow value by a units conversion of 1.36. (EPA
revised the boiler capacity methodology and updated the previous value of 1.25 to 1.36 beginning with the
1997 data year.)
AP-423 emission factors are used to calculate emissions (see Table B-4). The emission factor used
depends upon the SCC and pollutant, as explained below.
! The appropriate SCC is assigned to each source based on its fuel and boiler characteristics. For
sources using coal, the SCC is based on the American Society for Testing and Materials criteria for
moisture, mineral-free matter basis (if greater than 11,500 Btu/lb, coal type is designated to be
bituminous; if between 8,300 and 11,500 Btu/lb, coal type is designated to be subbituminous; and if
less than 8,300 Btu/lb, coal type is designated to be lignite) and the boiler type (firing configuration
and bottom type) as specified by AP-42. Fluidized bed combustion boilers have SCCs assigned
based on the fuel type. If both coal and oil are burned in the same boiler, it is assumed that the oil is
distillate; if coal is not burned, the oil burned is assumed to be residual. See Table B-2 for a
complete list of the relationships among fuel type, firing type, bottom type, and SCC.
Since Form EIA-767 does not provide control efficiencies for NOX, PM10, and PM2.5, control
efficiencies are derived using the following methods:
! NOX control efficiency is based on the assumption that the boiler would be controlled so that its
emission rate would equal its emission limit, expressed on an annual equivalent basis. After
calculating the heat input, EPA back-calculates controlled emissions assuming compliance with the
applicable standard. The NOX net control efficiency is calculated by dividing the controlled by the
uncontrolled NOX emissions.
! Since Form EIA-767 only reports TSP control efficiency, EPA uses the (updated) PM10
Calculator4 to derive PM10 and PM2.5 control efficiencies. (The PM Calculator estimates PM10
and PM2.5 control efficiencies based on the SCC and the primary and secondary control devices.
These device codes are National Emission Data System (NEDS) codes that are derived from EIA-
767 reported PM control devices and efficiencies rules as explicated in Table B-5. The control
efficiencies from the PM Calculator are based on particle size distribution data from AP-42 for
Documentation for the Final 1999 National Emissions
Inventory (Vzrsion 3) for Criteria Air Pollutants and Ammonia B-5
Point Source Methodologies
-------
specific SCCs, where available. These control efficiencies were revised beginning with the 1998
data file.)
EPA computes the S02 emissions as controlled emissions assuming 100 percent rule effectiveness and
using the sulfur content of the fuel as specified in the EIA-767 data. The PM10 and PM2.5 emissions are also
computed as controlled emissions assuming 100 percent rule effectiveness. The ash content of the fuel used to
calculate uncontrolled PM10 and PM2.5 emissions is also specified in the EIA-767 data. The NOX emissions
are computed as controlled emissions assuming 80 percent rule effectiveness for 1985-1994 data; beginning
with 1995 data, NOX rule effectiveness is assumed to be 100 percent. The CO and VOC emissions are
calculated as uncontrolled emissions. Although no NH3 AP-42 emission factors officially exist for utility fossil-
fuels, in 1998 EPA developed coal, oil, and gas NH3 emission factors that are applied to the specified quantity
of fuel used (see Table B-4). Thus, beginning with the 1996 data year, NH3 estimates are included in the
Trends database.
The PM10 and PM2.5 emissions included in the Trends inventory for all years through the 1999 data year
represent filterable PM10 and PM2.5 emissions. Beginning with data year 1996, condensible PM (PMCD)
emissions were estimated and summed with filterable PM10 and PM2.5 emissions to estimate primary PM10
and primary PM2.5 emissions. The rules used to determine the PM condensable emission factors are described
in Table B-6. The algorithms to compute all pollutant emissions are presented in Tables B-7 and B-8.
Since fewer required data elements (identification data, boiler fuel quantity and quality data, and FGD
data, if applicable) exist for those plants with a total capacity between 10 MW and 100 MW, many values are
missing. Most data elements are assigned a default value of zero; however, if values for boiler firing and bottom
type are missing (these are needed in the SCC assignment), the default values for wall-fired and dry bottom
types are assigned. In the past, discrepancies have occurred in the boiler bottom and firing type data as
reported to EIA and CAMD. Based on a coordinated effort in 1996, all differences in bottom and firing types
for coal boilers were resolved for previous years (i.e., 1985 through 1995).
6. What EIA data have been replaced with data from other sources?
EPA replaced the 1985 S02 emissions and heat input calculated from the 1985 Form EIA-767 data with
corresponding boiler-level data (disaggregated to the SCC level) from the National Allowance Data Base
Version3.11 (NADBV311).5 These data underwent two public comment periods in 1991 and 1992 and are
considered the best available data for 1985. Aggregations at the fuel levels (Tier IE) are approximations only
and are based on the methodology described in Section 1.
In 1996, CAMD completed research on utility coal boiler-level NOX rates. Approximately 90 percent of
the rates were based on relative accuracy tests performed in 1993 and 1994 as a requirement for continuous
emissions monitor (CEM) certification, while the remaining boilers' rates were obtained from utility stack tests
from various years. These coal boiler-specific NOX rates were considered, on the whole, to be significantly
better than those calculated from EPA's NOX AP-42 emission factors, which are SCC-category averages.
Documentation for the Final 1999 National Emissions
Inventory (Vzrsion 3) for Criteria Air Pollutants and Ammonia B-6
Point Source Methodologies
-------
Thus, whenever these new NOX rates were available, EPA recalculated NOX coal emissions at the coal
SCC level, using the heat input (EIA's 767 fuel throughput multiplied by the fuel heat content) and adjusting
units, according to the following equation:
These new NOX SCC-level coal emissions replaced the AP-42 calculated emissions for most of the coal
SCCs in the 1985-1994 data years (when ETS/CEM data were unavailable).
As of January 1, 1994, Title IV (Acid Deposition Control) of the Clean Air Act Amendments of 1990
(CAAA) required Phase I affected utility units to report heat input, S02, and NOX data to EPA. Beginning
January 1, 1995, all affected units were required to report heat input and S02 emissions; most also had to
report NOX emissions, although some units received extensions until July 1, 1995 or January 1, 1996 for NOX
reporting.
The ETS/CEM data files contain actual, rather than estimated, data. Thus, if a complete set of ETS/CEM
annual S02 and/or NOX emissions and/or heat input data existed for 1994 and 1995, those data values
replaced the data estimated from EIA-767 data. This process involved the following steps:
! Aggregation of ETS/CEM hourly or quarterly data to annual data.
! Assignment of ETS/CEM data, reported on a monitoring stack or pipe level, to the boiler level.
! Matching the ETS/CEM boiler-level annual data to the processed EIA-767 annual data.
! Disaggregating the boiler-level ETS/CEM data to the boiler SCC level based on each SCC's
fractional share of the boiler EIA-based heat input, S02, and NOX, respectively. The algorithms
used are included in Table B-9.
Beginning with 1996 data, the ETS/CEM annual Scorecard data replaced EIA-derived S02 and NOX
emissions and heat input for all boilers included in EIA-767 and in ETS/CEM. For those records in which the
ETS/CEM heat input replaces the EIA-calculated value, the heat input does not equal the product of the EIA-
reported fuel throughput and heat content. Additionally, C02 and PMCD values are recalculated using the
ETS/CEM heat input value, thus also changing the values of TOTPMIO and TOTPM25.
Beginning with 1999 data, submitted State data replaced all other data except for S02 and NOX data
from ETS/CEM.
NOXCOAL^ = NQXPT^ * Hnm * ^
(Eq. 1)
where: NOXCOAL
NOXRT
HTI
NOX emissions for the boiler coal SCC (in tons)
CAMD's coal NOX rate for the given boiler (in lbs/MMBtu)
heat input for the boiler's coal SCC (in MMBtu)
Documentation for the Final 1999 National Emissions
Inventory (Vzrsion 3) for Criteria Air Pollutants and Ammonia B - 7
Point Source Methodologies
-------
7. How does EPA calculate ozone season daily emissions?
Ozone season daily (OSD) emissions are estimated for data years 1990-1999 by assuming the day to be
a typical or average summer July day. Emissions for VOC, NOX, CO, S02, PM10, PM2.5, and NH3 (S04
is zero) are calculated at the SCC level by taking the ratio of the Form EIA-767 July monthly to annual heat
input, dividing it by 31, and then multiplying this value by the already calculated annual emissions. Beginning in
data year 1998, a weighted average of the heat inputs for the five ozone season months (July-September) was
used in place of the July month heat input. The equation is:
HTISUMgm
EQKDgfp = — rjnvAwm— + (Eq. 2)
31 + H71ANNjyy. v 4 '
where: EOSD
HTISUM
HTIANN
EANN
Ozone season daily emissions for a given pollutant at the SCC level (in tons)
July monthly or ozone season monthly average Form EIA-767 calculated heat
input for the given boiler's SCC (in MMBtu)
annual Form EIA-767 calculated heat input for the given boiler's SCC (in
MMBtu)
Annual emissions for a given pollutant at the SCC level (in tons) for that year
Beginning with 1999 data, for the OSD S02 and NOX emissions for the 74 ETS/CEM-only boilers, EPA's
hourly data files aggregated to the daily hour were used. The data for July 21 was first chosen for the OSD
values; if 0, the next day on either side was chosen (July 20 then July 22) and so on until non-zero values, if
they existed during the five months, were found.
8. What additional emissions estimates adjustments does EPA make?
To derive VOC emissions estimates, an adjustment is made due to the underestimation of aldehydes which
are not accounted for in the VOC emission factors for the following SCCs: 10100401, 10100404, 10100501,
10100601, and 10100604. The VOC emissions are augmented according to the methodology used in the
Hydrocarbon Preprocessor (HCPREP) of the Flexible Regional Emissions Data System (FREDS).6 This
augmentation was made on steam emission inventories for the years 1985 through 1999.
9. How does EPA perform its calculations?
The following provides an example calculation for estimating S02 emissions for a tangentially-fired dry-
bottom utility boiler burning bituminous coal. This example shows how the emissions are initially calculated
using data reported to EIA-767 and an AP-42 emission factor, and then overlaid with S02 emissions reported
to ETS/CEM. The methods shown in the example calculation are used to estimate emissions for all steam
generated fossil-fuel boilers and pollutants. See section 7 for details on what EIA-767 data are replaced with
ETS/CEM data for calculating emissions.
Documentation for the Final 1999 National Emissions
Inventory (Vzrsion 3) for Criteria Air Pollutants and Ammonia B - 8
Point Source Methodologies
-------
! 1995 boiler SCC data:
Variable
Variable Description
Name
Value
Units
Source classification code
SCC
10100212
-
Annual fuel throughput
thruput
1300000
SCC units
Heat content of fuel
heatcon
23.18 (really 23.1849046)
MMBtu/SCC units
Sulfur content of fuel
sulfcon
3.17 (really 3.1716)
%
S02 control efficiency
coneff4
89.3
%
Final emissions for inventory
emiss4
9332.559
tons
Final heat input for inventory
htinpt
31782453.38
MMBtu
Annual heat input calculated from EIA-767 data
eiahti
30140376
MMBtu
Annual S02 emissions calculated from EIA
eiaso2
8382.2216
tons
767 data
S02 emission factor
emf4
39 (38 beginning with 1996 data)
lbs S02/ton coal
Annual S02 emissions reported to ETS/CEM
so2ets
9332.559
tons
Annual heat input reported to ETS/CEM
htiets
31782453.38
MMBtu
Equation:
eiaso2 =
coal thruput * EMF4 * sujfccm. * (1 - (ponqff4l 1 DP))
2000
(Eq. 3)
! Calculation:
HMQ _ (1,300,000) (38) (3.1716) (1-0.893)
2 2000
Result:
XIASQ) ¦ 8,382 (fansJytafy to ntartst M»gtr
But rtpia£td by 1995 STSICSM SO^ mdutenu. (SO^M) ¦ 9,332.5590 (tons/ytnfy ¦ final tmUttmu. (XMISS4)
Ihtr^brs XIASO^ ¦ 8^82 (povulyaafy, and ¦ "rurssu. ¦ 9333 (pomulytat) in tfw fumitory
Note that the AP-42 S02 emission factor for SCC 10100212 was changed from 39 to 38 lbs/ton of coal
beginning with data year 1996, reflecting the updated emission factor value.
10. References
1. Steam-Electric Plant Operation and Design Report, Form EIA-767, data files for 1985 - 1999, U.S.
Department of Energy, Energy Information Administration, Washington, DC, 2000.
2. Acid Rain Program CEMS Submissions Instructions for Monitoring Plans, Certification Test
Notifications, and Quarterly Reports, U.S. Environmental Protection Agency, Washington, DC, May
1995.
Documentation for the Final 1999 National Emissions
Inventory (Vzrsion 3) for Criteria Air Pollutants and Ammonia B - 9
Point Source Methodologies
-------
3. July 14, 2000 EPA-approved updated Compilation of Air Pollutant Emission Factors, Volume I:
Stationary Point and Area Sources, Fifth Edition, AP-42, U.S. Environmental Protection Agency,
Research Triangle Park, NC.
4. Enhanced Particulate Matter Controlled Emissions Calculator, Draft User's Manual, Emission
Factor and Inventory Group, Emissions Monitoring and Analysis Division, Office of Air Quality Planning
and Standards, U.S. Environmental Protection Agency, Research Triangle Park, NC. Prepared by E.H.
Pechan & Associates, Inc., Durham, NC under EPA Contract No. 68-D7-0067, Work Assignment No.
3-09, November 1999.
5. The National Allowance Data Base Version 3.11: Technical Support Document, Acid Rain Division,
Office of Atmospheric Programs, U.S. Environmental Protection Agency, Washington, DC, March 1993.
6. The Flexible Regional Emissions Data System (FREDS) Documentation for the 1985 NAPAP
Emission Inventory: Preparation for the National Acid Precipitation Assessment Program.
Appendix A. EPA-600/9-89-047. U.S. Environmental Protection Agency, Office of Research and
Development, Air and Energy Engineering Research Laboratory, Research Triangle Park, NC, May
1989.
Documentation for the Final 1999 National Emissions
Inventory (Vzrsion 3) for Criteria Air Pollutants and Ammonia B -10
Point Source Methodologies
-------
Table B-1. Methods for Developing Annual Emission Estimates for Steam Generated
Fossil-Fuel Utility Boilers for the Years 1989-1999
For the data years For the pollutant(s) EPA estimated emissions by
1989-1993
NOX
If coal is burned, EIA data and EPA/ARD emission factors
and heat input are used; if coal is not burned, EIA data and
AP-42 emission factors applied to fuel quantity are used.
1994-1995
NOX
If the boiler reports to both EIA-767 and ETS/CEM, and the
ETS/CEM NOX data are complete for the year, then the
ETS/CEM data are used. Otherwise, if a boiler burned coal,
EIA data and EPA/ARD emission factors and heat input are
used; if coal is not burned, EIA data and AP-42 emission
factors applied to fuel quantity are used.
1996-1999
NOX
If the boiler reports to both EIA-767 and ETS/CEM, then the
ETS/CEM data are used. Otherwise, EIA data and AP-42
emission factors applied to fuel quantity are used. Note that
AP-42 emission factors for some SCCs changed from data
years 1985-1995 to data year 1996, and again in data year
1997.
1989-1993
S02
EIA data and AP-42 emission factors applied to fuel quantity
are used.
1994-1999
S02
If the boiler reports to both EIA-767 and ETS/CEM, then the
ETS/CEM data are used. Otherwise, EIA data and AP-42
emission factors applied to fuel quantity are used. Note that
AP-42 emission factors for some SCCs changed from data
years 1985-1995 to data year 1996.
1989-1998
VOC, CO
EIA data and AP-42 emission factors applied to fuel quantity
are used. Note that AP-42 emission factors for some SCCs
changed from data years 1985-1995 to data year 1996 for
VOC and CO.
1989-1997
PM10, PM2.5 (Filterable)
EIA data and AP-42 emission factors applied to fuel quantity
are used. Note that AP-42 emission factors for some SCCs
changed from data years 1985-1995 to data year 1996 for
PM10.
1998
PM10, PM2.5 (Filterable)
EIA data and AP-42 emission factors applied to fuel quantity
are used. Note that AP-42 emission factors for some SCCs
changed from data year 1996 to data year 1998 for PM10 and
PM2.5. Since the PM10 Calculator Program was updated in
1999-2000, updated PM efficiencies are derived for emissions
calculations.
Documentation for the Final 1999 National Emissions
Inventory (Version 3) for Criteria Air Pollutants and Ammonia B -11
Point Source Methodologies
-------
Table B-1 (continued)
For the data years
For the pollutant(s)
EPA estimated emissions by
1996-1999
PM Condensible
(PMCD), Primary PM10,
Primary PM2.5
EIA data and AP-42 emission factors applied to heat input
are used to estimate PMCD. PMCD is summed with
filterable PM10 and PM2.5, respectively, to estimate primary
PM10 and PM2.5. However, if the boiler reports to both EIA-
767 and ETS/CEM, then the ETS/CEM heat input overlays
ElA-based heat input, PM condensible is recalculated, and
primary PM10 and PM2.5 emissions are updated.
1999
VOC, CO, Filterable and
Primary PM10 and
PM2.5, NH3
Emissions specified are overlaid by State submitted data
when available.
1999
NOX and S02
Emissions specified are overlaid by State submitted data
when available, and if no ETS/CEM data available.
1996-1999
NH3
EIA data and emission factors applied to heat input are used
to estimate NH3 emissions. However, if the boiler reports to
both EIA-767 and ETS/CEM, then the ETS/CEM heat input
overlays ElA-based heat input, NH3 is recalculated, and the
emissions are updated. For data years prior to 1996, NH3
emissions were not estimated for utility boilers.
1999
NOX, S02, VOC, CO,
PM10, PM2.5, NH3
State submitted data overlays all emissions except S02 and
NOX ETS/CEM data.
Documentation for the Final 1999 National Emissions
Inventory (Version 3) for Criteria Air Pollutants and Ammonia B -12
Point Source Methodologies
-------
Table B-2. Steam Electric Utility Unit Source Classification Code Relationships
Prime
Type
Fossil-Fuel
Firing Type
Bottom Type
see
No data
No data
10100202
Wet
10100201
Dry
10100202
Wall**
No data
10100202
Wet
10100201
Dry
10100202
Opposed
No data
10100202
Wet
10100201
Dry
10100202
Tangential
No data
10100212
Wet
10100201
Dry
10100212
Stoker
All
10100204
Cyclone
All
10100203
Fluidized Bed
N/A
10100217
No data
No data
10100222
Wet
10100221
Dry
10100222
Wall
No data
10100222
Wet
10100221
Dry
10100222
Opposed
No data
10100222
Wet
10100221
Dry
10100222
Tangential
No data
10100226
Wet
10100221
Dry
10100226
Stoker
All
10100224
Cyclone
All
10100223
Fluidized Bed
N/A
10100238
No data
All
1.01003011016
Wall
All
Opposed
All
Tangential
All
Stoker
All
Cyclone
All
Fluidized Bed
N/A
STEAM
Coal
Bituminous
Subbituminous
Lignite
ST
ST
Documentation for the Final 1999 National Emissions
Inventory (Vzrsion 3) for Criteria Air Pollutants and Ammonia B -13
Point Source Methodologies
-------
Table B-2 (continued)
Prime
Type Fossil-Fuel Mover* Firing Type Bottom Type SCC
Residual Oil
ST
No data
All
1.0100401101e+47
Wall
All
Opposed
All
Tangential
All
Stoker
All
Cyclone
All
Distillate Oil
ST
No data
All
1.0100501101e+47
Wall
All
Opposed
All
Tangential
All
Stoker
All
Cyclone
All
Natural Gas
ST
No data
All
10100601***
Wall
All
10100601
Opposed
All
10100601
Tangential
All
10100604
Stoker
All
10100601
Cyclone
All
10100601
Process Gas
ST
N/A
N/A
10100701
Petroleum Coke
ST
N/A
N/A
10100801
Biomass/Wood/Wood Waste
ST
N/A
N/A
10100902
Paper Pellets
ST
N/A
N/A
10100903
Propane Oil/LPG
ST
N/A
N/A
10101002
Other Solid Waste
ST
N/A
N/A
10101201
Refuse (Tire) Derived Fuel
ST
N/A
N/A
10101202
Other Liquid/Oil Waste
N/A
N/A
10101302
NONSTEAM
Oil
GT, CT
N/A
N/A
2010010120100102
IC
N/A
N/A
Gas
GT, CT
N/A
N/A
2010020120100202
IC
N/A
N/A
*ST=Steam, GT=gas turbine, CT=combustion turbine, IC=internal combustion engine (reciprocating)prime mover.
**Wall firing includes front, arch, concentric, rear, side, vertical, and duct burner firing.
***An SCC of 10100601 is used for those few units that are classified as IGCC (internal gasification combined
cycle); EPA is planning on creating a new SCC for IGCCs.
Documentation for the Final 1999 National Emissions
Inventory (Vzrsion 3) for Criteria Air Pollutants and Ammonia B -14
Point Source Methodologies
-------
Table B-3. Boiler Emissions Data Sources (Other than EIA-767)
for NOX, S02, and Other Pollutant Emissions Data by Year, 1985-1999
Year
NOX
S02
Other Pollutant Emissions
1985
1986
1987
1988
1989
1990
1991
1992
1993
1994
1995
1996
Overlaid with CAMD coal NOX rate
calculations when possible
Overlaid with CAMD coal NOX rate
calculations when possible
Overlaid with CAMD coal NOX rate
calculations when possible
Overlaid with CAMD coal NOX rate
calculations when possible
Overlaid with CAMD coal NOX rate
calculations when possible
Overlaid with CAMD coal NOX rate
calculations when possible
Overlaid with CAMD coal NOX rate
calculations when possible
Overlaid with CAMD coal NOX rate
calculations when possible
Overlaid with CAMD coal NOX rate
calculations when possible
Overlaid with CAMD coal NOX rate
calculations when possible; overlaid
ETS/CEM data when possible
Overlaid with ETS/CEM data when available
Overlaid with ETS/CEM data when available
NADBV311 data
Calculated from EIA-767 data
Calculated from EIA-767 data
Calculated from EIA-767 data
Calculated from EIA-767 data
Calculated from EIA-767 data
Calculated from EIA-767 data
Calculated from EIA-767 data
Calculated from EIA-767 data
Overlaid with ETS/CEM data when
possible
Overlaid with ETS/CEM data when
possible
Overlaid with ETS/CEM data when
possible
1997
Overlaid with ETS/CEM data when available Overlaid with ETS/CEM data when
possible
1998
Overlaid with ETS/CEM data when available Overlaid with ETS/CEM data when
possible
1999
Overlaid with ETS/CEM data when
available; otherwise overlaid by State
submitted data when available
Overlaid with ETS/CEM data when
available; otherwise overlaid by
State submitted data when
available
Calculated from EIA-767 data
Calculated from EIA-767 data
Calculated from EIA-767 data
Calculated from EIA-767 data
Calculated from EIA-767 data
Calculated from EIA-767 data
Calculated from EIA-767 data
Calculated from EIA-767 data
Calculated from EIA-767 data
Calculated from EIA-767 data
Calculated from EIA-767 data
Calculated from EIA-767 data; if
ETS/CEM heat input are
available, PMCD is recalculated
using ETS/CEM heat input, and
then primary PM10 and primary
PM2.5 are recalculated, too
Calculated from EIA-767 data; if
ETS/CEM heat input are
available, PMCD is recalculated
using ETS/CEM heat input, and
then primary PM10 and primary
PM2.5 are recalculated, too
Calculated from EIA-767 data; if
ETS/CEM heat input are
available, PMCD is recalculated
using ETS/CEM heat input, and
then primary PM10 and primary
PM2.5 are recalculated, too
Calculated from EIA-767 data; if
ETS/CEM heat input are
available, PMCD is recalculated
using ETS/CEM heat input, and
then primary PM10 and primary
PM2.5 are recalculated, too.
Overlaid with State submitted
data when available.
CAMD = EPA's Clean Air Markets Division
NADBv311 = National Allowance Data Base Version 3.11
ETS/CEM = EPA's Emissions Tracking System/Continuous Emissions Monitoring data
Documentation for the Final 1999 National Emissions
Inventory (Vzrsion 3) for Criteria Air Pollutants and Ammonia B -15
Point Source Methodologies
-------
Table B-4. Emission Factors
see
COEF
NOXEF
VOCEF
PM10EF
PM25EF
S02EF
NH3EF
PMFLAG
S02FLAG
10100101
0.6000
18.0000
0.0700
2.3000
0.6000
39.0000
0.000565
A
S
10100102
0.6000
9.0000
0.0700
4.8000
2.4000
39.0000
0.000565
S
10100201
0.5000
31.0000
0.0400
2.6000
1.4800
38.0000
0.000565
A
S
10100202
0.5000
22.000*
0.0600
2.3000
0.6000
38.0000
0.000565
A
S
10100203
0.5000
33.0000
0.1100
0.2600
0.1100
38.0000
0.000565
A
S
10100204
5.0000
11.0000
0.0500
13.2000
4.6000
38.0000
0.000565
S
10100205
6.0000
7.5000
0.0500
6.0000
2.2000
38.0000
0.000565
S
10100211
0.5000
14.0000
0.0400
2.6000
1.4800
38.0000
0.000565
A
S
10100212
0.5000
15.000*
0.0600
2.3000
0.6000
38.0000
0.000565
A
S
10100215
0.5000
31.0000
0.0600
2.3000
0.6000
38.0000
0.000565
A
S
10100217
18.0000
15.2000
0.0500
7.8000
3.2000
31.0000
0.000565
S
10100218
18.0000
5.0000
0.0500
7.8000
3.2000
31.0000
0.000565
S
10100221
0.5000
24.0000
0.0400
2.6000
1.4800
35.0000
0.000565
A
S
10100222
0.5000
12.000*
0.0600
2.3000
0.6000
35.0000
0.000565
A
S
10100223
0.5000
17.0000
0.1100
0.2600
0.1100
35.0000
0.000565
A
S
10100224
5.0000
8.8000
0.0500
13.2000
4.6000
35.0000
0.000565
S
10100225
6.0000
7.5000
0.0500
6.0000
2.2000
35.0000
0.000565
S
10100226
0.5000
8.400*
0.0600
2.3000
0.6000
35.0000
0.000565
A
S
10100235
0.5000
14.0000
0.0600
2.3000
0.6000
35.0000
0.000565
A
S
10100238
18.0000
5.0000
0.0500
7.8000
3.2000
31.0000
0.000565
S
10100300
-9.0000
-9.0000
-9.0000
-9.0000
-9.0000
-9.0000
0.000565
10100301
0.2500
13.000*
0.0700
1.8170
0.5214
30.0000
0.000565
A
S
10100302
0.6000
7.1000
0.0700
2.3000
0.6600
30.0000
0.000565
A
S
10100303
0.6000
15.0000
0.0700
0.8700
0.1100
30.0000
0.000565
A
S
10100304
6.0000
6.0000
0.0700
1.6000
0.5600
30.0000
0.000565
A
S
10100306
5.0000
5.8000
0.0700
1.6000
0.5600
30.0000
0.000565
A
s
10100316
0.1500
3.6000
0.0300
12.0000
1.4000
10.0000
0.000565
s
10100317
0.1500
3.6000
0.0300
12.0000
1.4000
10.0000
0.000565
s
10100318
0.1500
3.6000
0.0300
12.0000
1.4000
10.0000
0.000565
s
10100401
5.0000
47.0000
0.7600
5.9000
4.3000
157.0000
0.800000
s
10100404
5.0000
32.0000
0.7600
5.9000
4.3000
157.0000
0.800000
s
10100405
5.0000
47.0000
0.7600
5.9000
4.3000
157.0000
0.800000
s
10100406
5.0000
32.0000
0.7600
5.9000
4.3000
157.0000
0.800000
s
10100501
5.0000
24.0000
0.2000
1.0000
0.2500
143.6000
0.800000
s
10100504
5.0000
47.0000
0.7600
5.9000
4.3000
150.0000
0.800000
s
10100505
5.0000
32.0000
0.7600
5.9000
4.3000
150.0000
0.800000
s
10100601
84.0000
280.000*
5.5000
1.9000
1.9000
0.6000
3.200000
10100602
84.0000
100.0000
5.5000
1.9000
1.9000
0.6000
3.200000
10100604
24.0000
170.0000
5.5000
1.9000
1.9000
0.6000
3.200000
10100701
84.0000
280.0000
5.5000
1.9000
1.9000
950.0000
-9.000000
s
10100702
84.0000
100.0000
6.0000
1.9000
1.9000
950.0000
-9.000000
s
10100801
0.6000
21.0000
0.0700
7.9000
4.5000
39.0000
-9.000000
A
s
10100901
13.6000
1.5000
1.4000
16.8000
10.0000
0.0750
-9.000000
10100902
13.6000
1.5000
1.4000
6.4800
5.4700
0.0750
-9.000000
10100903
13.6000
1.5000
1.4000
7.9000
7.9000
0.0750
-9.000000
Documentation for the Final 1999 National Emissions
Inventory (Vzrsion 3) for Criteria Air Pollutants and Ammonia B -16
Point Source Methodologies
-------
Table B-4 (continued)
see
COEF
NOXEF
VOCEF
PM10EF
PM25EF
S02EF
NH3EF
PMFLAG
S02FLAG
10101001
3.6000
21.0000
0.2600
0.6000
0.6000
86.5000
-9.000000
S
10101002
3.2000
19.0000
0.2500
0.6000
0.6000
86.5000
-9.000000
S
10101101
2.0000
1.2000
2.0000
5.6000
1.4000
0.0000
-9.000000
10101201
0.0165
3.8000
2.0000
11.4000
7.8000
3.9000
-9.000000
10101202
3.6000
5.0000
2.0000
44.0000
11.0000
1.7000
-9.000000
10101301
5.0000
19.0000
1.0000
51.0000
13.0000
147.0000
-9.000000
A
S
10101302
5.0000
19.0000
1.0000
51.0000
13.0000
147.0000
-9.000000
A
S
20100101
0.4587
122.3200
0.0570
1.6680
1.6680
140.3900
-9.000000
S
20100102
130.0000
604.0000
0.0570
11.2000
10.8000
39.7000
-9.000000
S
20100201
83.6400
326.4000
2.1420
6.7320
6.7320
3.4680
-9.000000
20100202
399.0000
2840.0000
116.0000
10.0000
10.0000
0.6000
-9.000000
20100901
6.7200
97.7000
2.3800
8.5400
8.5400
6.2000
-9.000000
20100902
130.0000
604.0000
49.3000
42.5000
42.5000
6.2000
-9.000000
* For these 6 SCCs there are two NOx emission factors, one representing pre-NSPS and one representing post-NSPS. Pre-NSPS is
before 1974, while post-NSPS is 1974 and beyond. The pre-NSPS is used as the default when the date of operation is not known. The
post-NSPS is below for reference.
Post-NSPS NOx Emission Factors:
10100202
12.0000
10100212
10.0000
10100222
7.4000
10100226
7.2000
10100301
10100601
-9 indicates a missing emission factor.
Documentation for the Final 1999 National Emissions
Inventory (Vzrsion 3) for Criteria Air Pollutants and Ammonia B -17
Point Source Methodologies
-------
Table B-5. Rules for Assigning Primary and Secondary PM Control Device NEDS Codes
If the PM device is missing, then it is defaulted to an electrostatic precipitator (ESP)
If the PM device is an electrostatic precipitator (ESP) and the PM control efficiency is at least
95 percent, then the NEDS control device code = 10.
If the PM device is an electrostatic precipitator (ESP) and the PM control efficiency is at least
80 but less than 95 percent, then the NEDS control device code =11.
If the PM device is an electrostatic precipitator (ESP) and the PM control efficiency is less
than 80 percent, then the NEDS control device code = 12.
If the PM device is a wet scrubber and the PM control efficiency is at least 95 percent, then
the NEDS control device code = 1.
If the PM device is a wet scrubber and the PM control efficiency is at least at least 80 but less
than 95 percent, then the NEDS control device code = 2.
If the PM device is a wet scrubber and the PM control efficiency is less than 80 percent, then
the NEDS control device code = 3.
If the PM device is a baghouse, then the NEDS control device code = 17.
If the PM device is a single cyclone, then the NEDS control device code = 75.
If the PM device is a multiple cyclone, then the NEDS control device code = 76.
If the PM device is other, then the NEDS control device code = 99.
Documentation for the Final 1999 National Emissions
Inventory (Vzrsion 3) for Criteria Air Pollutants and Ammonia B -18
Point Source Methodologies
-------
Table B-6. PM Condensable Emission Factor (PMCDEF) Rules
Coal
If SCC = 10100204, 10100205, 10100224, 10100225, 10100304, 10100306, then PMCDEF = .04.
Else, if SCC = 10100217,10100218, then PMCDEF = .01.
Else, if SCC = 10100201,10100202,10100212,10100221,10100222,10100223,10100226, AND
there is either an S02 FGD or a PM wet scrubber, then PMCDEF = .02.
Else, if SCC = 10100201, 10100202, 10100203,10100212, 10100221, 10100222,10100223,
10100226, 10100301,10100302,10100303 and there is any PM control other than a wet scrubber and
there is no S02 control, OR SCC = 10100222 and there is no PM control, then PMCDEF = .1 * sulfur
content [as a decimal] - .03.
And if PMCDEF is less than .01, then PMCDEF=.01.
Light Oil (Distillate)
If SCC between 10100401 and 10100499 inclusive, then PMCDEF=.01.
Heavy Oil (Residual)
If SCC between 10100501 and 10100599 inclusive, then PMCDEF=.009.
Natural Gas
If SCC between 10100601 and 10100699 inclusive, then PMCDEF=.0057.
Processed Gas
If SCC between 10100701 and 10100799 inclusive, then PMCDEF=.0056.
Petroleum Coke
If SCC between 10100801 and 10100899 inclusive, then PMCDEF=.01.
Wood/Biomass
If SCC between 10100901 and 10100999 inclusive, then PMCDEF=.017.
LPG (Propane, Butane)
If SCC between 10101001 and 10101099 inclusive, then PMCDEF=.0056.
Solid Waste/Refuse
If SCC between 10101201 and 10101299 inclusive, then PMCDEF=.017.
Liquid Waste
If SCC between 10101301 and 10101399 inclusive, then PMCDEF=.009.
Documentation for the Final 1999 National Emissions
Inventory (Vzrsion 3) for Criteria Air Pollutants and Ammonia B -19
Point Source Methodologies
-------
Table B-7. Algorithms Used to Estimate EIA-Based VOC, NOX, CO, S02, PM10, PM2.5,
and NH3 Annual Emissions from Electric Utility Boilers
SNDit see = PCsec * tee* C1 " * ucp
^CDbtFOC^SCC = ^ SCC + ^CO or VOC, SCC
or »c,3, SCC = PC SCC* or SCC * Af * 0 " CBj^ „ + UCF
Eso,, see = PC see* ^so^scc* sr* O " cssoitb) * UCF
smSi scc = PC sec* PPfmit sec * UCF
where: E = annual estimated emission (in tons/year)
FC = annual fuel consumption (in units/yearf)
EF = emission factor (in Ibs/unitf)
S = sulfur content (expressed as a decimal)
A = ash content (expressed as a decimal)
CE = control efficiency (expressed as a decimal)
b = boiler
f = fuel type
UCF = units conversion factor (1 ton/2000 lbs)
unitcoai = tons burned
unitoil = 1000 gallons burned
unitgas = million cubic feet burned
f Note that VOC also undergoes an augmentation procedure.
Documentation for the Final 1999 National Emissions
Inventory (Version 3) for Criteria Air Pollutants and Ammonia B-20
Point Source Methodologies
-------
Table B-8. Algorithms Used to Estimate EIA-Based Condensible PM, Primary PM10 and
Primary PM2.5 Annual Emissions for Electric Utility Boilers
ST§ia\ sex; - HHscc * sftucd%scc*
STotOiu or iCC ~ ^ oSOS * SJ9iOK SOS
where: PMCD = particulate matter condensible
E = annual estimated emissions (in tons/year)
HTI = annual heat input (in MMBtu/year)$
EF = emission factor (in tons/MMBtu)
f Note that VOC also undergoes an augmentation procedure.
$ Calculate using EIA fuel consumption and heat content values, but use ETS/CEM heat input data if available and
recalculate PMCD, TOTPM10, and TOTPM25.
Documentation for the Final 1999 National Emissions
Inventory (Vzrsion 3) for Criteria Air Pollutants and Ammonia B -21
Point Source Methodologies
-------
Table B-9. Algorithms Used to Disaggregate ETS/CEM
Boiler Data to the Boiler-SCC Level
CEMS02
nee
16HS02
XCCfr
767S02*
* CEMS02,
CEMNOXscc =
1&1N0X.
sacjt
767JVOK
+ CEMNOXh
CEMHUi
see
767H77-
HCCfi
767fl77v
* CEMH77h
where: b
CEMS02, CEMNOX, CEMHTI
767S02, 767NOX, 767HTI
= boiler-level
= ETS/CEM annual boiler data for given parameter
= Form EIA-767-based calculated data for given parameter
Documentation for the Final 1999 National Emissions
Inventory (Vzrsion 3) for Criteria Air Pollutants and Ammonia B -22
Point Source Methodologies
-------
APPENDIX C
METHODOLOGY FOR AUGMENTING STATE/LOCAL/TRIBAL 1999
POINT SOURCE EMISSIONS INVENTORIES WITH PM10 AND PM25
EMISSIONS FOR THE 1999 NEI
1. INTRODUCTION
The NEI requires reporting of primary (PRT) PM10/PM25 emissions, which is the sum of the filterable
(FIL) PM10/PM25 and condensible (CON) fractions of PM10/PM25 for each emissions source of
PM10/PM25. S/L/T agencies; however, may report any fraction of total PM and/or PM10/PM25 emissions
for the 1999 NEI. The S/L/T inventories included emissions data for either PM-FIL, PM-PRI, PM10-FIL,
PM10-PRI, PM25-FIL, PM25-PRI, and/or PM-CON. From any one of these pollutants, the augmentation
procedure estimates PM10-PRI, PM10-FIL, PM25-PRI, PM25-FIL, and PM-CON as shown in the
following table. If not included in a S/L/T inventory, PM10-PRI and PM25-PRI were calculated by adding
PM10-FIL and PM-CON or PM25-FIL and PM-CON, respectively.
PM Category
Description
PM-PRI
Primary PM (includes filterables and condensibles)
PM-CON
Primary PM, Condensible Portion Only (all less than 1 micron)
PM-FIL
Primary PM, Filterable Portion Only
PM10-PRI
Primary PM10 (includes filterables and condensibles)
PM10-FIL
Primary PM10, Filterable Portion Only
PM25-PRI
Primary PM25 (includes filterables and condensibles)
PM25-FIL
Primary PM25, Filterable Portion Only
2. APPROACH
Figure 1 provides a flow diagram of the procedures for calculating PM10/PM25. The procedures were
developed to take advantage of the PM Calculator which calculates uncontrolled and controlled PM10-FIL
and PM25-FIL emissions using either uncontrolled PM-FIL or PM10-FIL emissions as inputs. For controlled
sources, up to two control devices can be used as input to the PM Calculator. The following steps were taken
to prepare S/L/T data for input to the PM Calculator:
Documentation for the Final 1999 National Emissions
Inventory (Version 3) for Criteria Air Pollutants and Ammonia C -1
Point Source Methodologies
-------
Figure 1
Procedures for Calculating PM2 5 and
PM10 Emissions for the NEI
Uncontrolled or
Controlled Emissions
Uncontrolled Emissions
S/L/T
Inventory
Emissions
PM-FIL
PM-PRI
PM-CON
PM10-FIL
> / Results
. / PM10-PRI
/ PM25-PRI
PM
Calculator
PM-FIL
PM10-FIL
PM25-FIL
PM10-FIL
PM10-PRI
PRI to FIL
Conversion
Factor
PRI to FIL
Conversion
Factor
FIL to CON
Conversion
Factor
-------
Step 1. Prepare S/L/T PM and PM10 Emissions for Input to the PM Calculator;
Step 2. Develop and apply source-specific factors to convert S/L/T-supplied PM-PRI to PM-FIL or
PM10-PRI to PMIO-FIL based on uncontrolled data;
Step 3. Prepare Factors from PM Calculator;
Step 4: Develop and apply algorithms to estimate emissions from S/L/T inventory data; and
Step 5: Review Results and Update the NEI with emission estimates and control information.
The remainder of this appendix explains each of these steps in detail. Tables C-l through C-12, which
are referenced in the following text, are provided in a separate Excel Workbook file named
"Point Appendix C Tables l_12.xls."
a. Step 1: Prepare S/L/T PM and PM10 Emissions for Input to the PM Calculator
The first step in the procedure involved back-calculating uncontrolled emissions to prepare the emissions
for input to the PM Calculator. Initially, the control efficiencies included in S/L/T inventories were used to
back-calculate uncontrolled emissions. However, this approach resulted in inconsistent results for the following
reasons:
(1) The PM Calculator uses default control efficiencies from AP-42 for each control device type or
combination of two control devices in series. Because the S/L/T control efficiency data typically did
not match the default control efficiencies in the PM Calculator that are applied to the back-calculated
uncontrolled emissions, the PM Calculator produced estimates for PM25-FIL in some cases that
were higher than the PMIO-FIL emissions included in S/L/T inventories.
(2) Some S/L/T inventories contained control device codes but did not report control efficiency data for
the control devices, or reported efficiencies of 100% for emission sources with emissions above zero.
Thus, the default control efficiencies in the PM Calculator were used to back-calculate uncontrolled
emissions by SCC and control device code. Using the default control efficiencies in the PM Calculator ensured
that the calculator produced the same PMIO-FIL emission values reported in a S/L/T inventory, and provided
a consistent basis for estimating PM25-FIL and PM-CON emissions from S/L/T PMIO-FIL emissions.
Prior to back-calculating uncontrolled emissions provided in S/L/T inventories, the PM, PM10, and
PM25 emissions in S/L/T inventories were reviewed to identify cases where PM10 emissions were less than
PM25 emissions. The following rules were applied to correct issues with the different fractions of PM:
If PM25-PRI > PM10-PRI and PM10-PRI is not null or zero, set PM25-PRI = PM10-PRI.
If PM25-FIL > PMIO-FIL and PMIO-FIL is not null or zero, set PM25-FIL = PMIO-FIL.
All PM-PRI and PM-FIL records were removed from the input file for the PM augmentation procedure if an
emission source had any form of PM10, PM25, or PM-CON. PM-CON records with emissions greater than
zero were used in the PM augmentation procedure. PM-CON records with null or zero emissions were
removed and calculated using the PM augmentation procedure.
Documentation for the Final 1999 National Emissions
Inventory (Version 3) for Criteria Air Pollutants and Ammonia C - 3
Point Source Methodologies
-------
The PM Calculator does not include particle-size profiles for several control device codes for which S/L/T
agencies reported PM or PM10 emissions. Therefore, it was necessary to map records with control codes not
in the PM Calculator to codes in the PM Calculator as shown in Table C-l. In addition, some agencies used
control device codes that do not comply with the NEI Input Format (NIF) Version 2.0 control device codes.
Therefore, EPA changed the codes in S/L/T inventories to comply with MF 2.0 codes and updated the NEI
with the MF 2.0 compliant codes.
b. Step 2: Develop and Apply Source-Specific Conversion Factors
The PM Calculator was used to estimate missing PM10-FIL and/or PM25-FIL emissions from the S/L/T
submitted emissions for PM-PRI, PM-FIL, or PM10-PRI. For this methodology, the emission factors for
PM-FIL, PM10-FIL, and PM-CON are most important because the PM calculator can produce the following
output from the indicated input:
Uncontrolled Input Uncontrolled Output Controlled Output
PM-FIL
PM10-FIL
PM10-FIL
PM25-FIL
PM25-FIL
PM10-FIL
PM10-FIL
PM10-FIL
PM25-FIL
PM25-FIL
Using the PM Calculator to produce these emissions estimates narrows the required conversion factors to the
following:
Convert S/L/T Emissions From Convert To (Input to PM Calculator)
PM-PRI
PM-FIL
PM10-PRI
PM10-FIL
PM10-FIL
PM-CON
These conversion factors are calculated with the following equations:
PM-PRI,,, ,,, -> PM-FIL,
PM-PRIstate X
PM-FIL
AP-42
PM-PRI
= PM-FIL STATF, conversion factor is
AP-42 J \ . iv. . j-->.iAp_42
pm-fil AP42 I
PM-PRI. 1 1 ;
PM10-PRIstate -» PM10-FILstate:
PM10-FILap 42^ ( PM10-FILAP42
PM10-PRJstate X Ipm10_PRIapJ = PM10.FILstaie, conversion factor is [pM10pRIipJ (2)
PM10-FILSTATE -» PM-CONstate :
. PM-CONAP42l ( pm-conAP42 ,
PM10-FILSTATF X = PM-CONSTATF, conversion factor is (3)
STATE ',PM10-FILAP 42J state I PMIO-FILaj, 42
Documentation for the Final 1999 National Emissions
Inventory (Version 3) for Criteria Air Pollutants and Ammonia C - 4
Point Source Methodologies
-------
With these conversion factors and the PM Calculator, all of the needed emissions values can be produced
from S/L/T submitted inputs for each Source Classification Code (SCC) and control device combination. The
uncontrolled conversion factors are provided by SCC in the Access database named "Uncontrolled PM
Factors.mdb."
Conversion Factors for Stationary External and Internal Combustion Sources (SCCs Starting with
1 and 2)
The emission factors and supporting data available in Chapters 1 and 3 of AP-42 were used to prepare
conversion factors for stationary external and internal combustion sources. All AP-42 PM emission factors that
were not specified as being CON, FIL, or PRI were assumed to be FIL, When necessary, for coal, sulfur
content was always assumed to be 1%, and ash content was always assumed to be 8%. Generally, AP-42 is
thorough in allocating the emissions factors for specific SCCs, but it is slightly ambiguous when describing
control devices. When there was ambiguity, the emission factor presented was assumed to apply to any NEI
control device code. For example, if AP-42 listed the control device for an emission factor as "ESP," the
following NEI codes were assumed to apply: 010 (high efficiency ESP), Oil (medium efficiency ESP), 012
(low efficiency ESP), and 128 (ESP).
Analysis of the factors presented in Chapters 1 and 3 of AP-42 revealed that Flue Gas Desulfurization
(FGD), NEI control device code 039, was the only control that affected the PM-CON value. For this control
only, a separate conversion factor from PM10-FIL to PM-CON is required that differs from the uncontrolled
conversion factor. For all other control devices, the conversion of the uncontrolled PM10-FIL emissions to
PM-CON emissions will result in the PM-CON value to use for both uncontrolled and controlled PM
emissions.
Table C-2 identifies the AP-42 emission factors (in lb/MMBtu) and references used to prepare the
conversion factors. For SCCs with complete information in AP-42, emissions factors for PM10-FIL, PM-
FIL, PM10-PRI, PM-PRI, and PM-CON were presented individually or could be derived from each other by
performing the necessary addition or subtraction. When PM-PRI or PM10-PRI emission factor values were
not listed in AP-42, the values were estimated by adding the appropriate FIL and CON factors if they were
available. Back calculations to derive missing FIL or CON factors from the PRI value were also performed.
An example of calculating conversion factors for SCCs with complete information is as follows:
SCC 10100201 with no controls (000) has the following reported values in AP-42:
PM Category
Emission Factor (lb/MMBtu)
AP-42
Page No.
PM10-FIL
PM-FIL
PM-CON3
2.1538
0.07
0.8
1.1-21
1.1-21
1.1-24
a Since the value applied to "All PM controls (without FGD controls)" this value was assumed to apply to
uncontrolled emissions as well.
Documentation for the Final 1999 National Emissions
Inventory (Version 3) for Criteria Air Pollutants and Ammonia C - 5
Point Source Methodologies
-------
PM-PRIap_42:
PM-PRIap_42 = PM-FILA|J_4, + PM-CONap.42
PM-PRIap_42 = 2.1538 ^ + 0.07 ^=2.2238 ^
PM10-PRIAP42:
PM10-PRIAP42 = PM10-FTLap_42 + PM-CON AP_42
PM10-PRIad^ =0.8 tJ^-+ 0.07 _J1_ = 0.87 lb
AP-42 " MMBtu ' MMBtu ' MMBtu
(PM-PRIstate -> PM-FILstate ) conversion factor:
PM-FILAP42 2.1538 ^
— ii — U.^Ooj
PM-PRIap^ 2.2238 „
(PM10-PRIstate -> PM10-FILSTATE) conversion factor:
PM10-FILAP 42 0.8 ^ Q9i95
PM10PRIap ~ 0.87 * ~
(PM10-FILstate -> PM-CONstate) conversion factor:
tth /r r^r\tvt n m lb
PM-CON A|1_42 = 0.07 MMBtu = 0 08?5
PM10-FILAP49 08
AP-42 u o MMBtu
AP-42 does not contain comprehensive data for each 8-digit SCC that is a source of PM10 and PM25
emissions. Thus, for 8-digit SCCs for which emission factor data were not available, average factor values
were calculated for the first 6 digits of the SCC code combined with the control type. For example,
uncontrolled SCC 10100212 has no reported PM-CON value, therefore, its PM-CON value was assumed to
be the average of the 11 SCCs that begin with "101002" and have a reported PM-CON value. SCCs
unaccounted for by averaging the known emissions factors values of the first 6 digits of the SCC were
accounted for by taking the average of the known emissions factor values' for all SCCs with the same first 5
digits as the SCC, then the first 4 digits, then the first 3 digits, then the first 2 digits, and finally the first digit.
Missing FIL (PM10-FIL and PM-FIL) values are accounted for in a similar fashion. The PM-PRI and PM10-
PRI values could then be obtained by adding the appropriate FIL value to the CON value.
To account for the effects that the FGD (039) control device has on the PM-CON value, conversion
factors also had to be compiled to calculate controlled PM-CON from an uncontrolled PM10-FIL, Analysis
of the State data revealed that FGD was used as one of the control devices for only three SCCs: 10100222,
10100202, and 10100501. Only SCC 10100501 did not have a PM-CON value reported in AP-42, so the
average value (0.02) of reported PM-CON values with an SCC that began with "10100" was used. The
reported values for these factors are as follows:
Documentation for the Final 1999 National Emissions
Inventory (Version 3) for Criteria Air Pollutants and Ammonia C - 6
Point Source Methodologies
-------
Control PM-CON AP-42
SCC Code (Ib/MMBtu) Page No.
10100201
039
0.02
1.1-24
10100202
039
0.02
1.1-24
10100203
039
0.02
1.1-24
10100221
039
0.02
1.1-24
10100222
039
0.02
1.1-24
10100223
039
0.02
1.1-24a
10100301
039
0.02
1.7-13
10100302
039
0.02
1.7-13
10100303
039
0.02
1.7-13b
a From AP-42, page 1.1-25: "No data are available for cyclone boilers (SCCs 1-01-
002-03/23, 1-02-002-03/23, 1-03-002-23) ... use the factors provided for
pulverized coal-fired boilers and applicable control devices."
b From AP-42, page 1.7-13: "No data available for cyclone boilers (SCCs 1-01-003-
03, 1-02-003-03) ... use the factors provided for pulverized coal-fired boilers and
applicable controls."
Once the necessary gaps had been filled, conversion factors could easily be calculated by using Equations
(1), (2), and (3) as described earlier. Final conversion factors for uncontrolled External Combustion Sources
and Stationary Internal Combustion Sources are shown in Table C-3. The conversion factors for uncontrolled
PM10-FIL to controlled PM-CON for an FGD control were calculated by using Equation (3), but substituting
in the controlled PM-CON value. The final values for the FGD control conversion factors are displayed in
Table C-4.
Conversion Factors for Industrial Processes (SCCs Starting with 3)
The procedures for preparing PM factors for SCCs starting with 3 are similar to those for SCCs starting
with 1 and 2. However, relative to the procedures for SCCs starting with 1 and 2, fewer particle-size-specific
emission factors are available and the units of the emission factors vary considerably more for SCCs starting
with 3. Thus, for example, factors developed for an SCC for natural gas burned in a utility or industrial boiler
cannot be applied to an industrial process burning natural gas because of the differences in the emission factor
units. The procedures for SCCs relied on the use of particle-size-specific emission factors for specific source
categories when available in Factor Information Retrieval Information (FIRE) system and AP-42. If particle-
size-specific emission factors in FIRE or AP-42, the procedures relied on the use of the generic particle size
profiles presented in Appendix B of AP-42.
In FIRE, only 25 SCCs beginning with 3 have emission factors for PM-FIL, PM10-FIL, and PM-CON
(see Table C-5). To compute the conversion factors for those 25 SCCs, the emission factors from FIRE were
utilized in equations 1, 2, and 3.
AP-42 contains emission factors for 13 SCCs for asphalt concrete operations that have not yet been
included in FIRE (see Table C-6). To compute the conversion factors for the 13 SCCs, the emission factors
were utilized in equations 1, 2, and 3. If one of the SCCs also appeared in Table C-5, the information from
Table C-6 was used to calculate the conversion factors and the Table C-5 information was not used.
Documentation for the Final 1999 National Emissions
Inventory (Version 3) for Criteria Air Pollutants and Ammonia C - 7
Point Source Methodologies
-------
For SCCs that did not have enough information in FIRE or AP-42 to generate the conversion factors, the
generic weight percent PM profiles located in AP-42 were used to generate the conversion factors. The
generic PM profile information is located in Appendix B.l, Appendix B.2, and the individual sections of AP-
42. If the relative percent composition of PM emissions for a group of similar SCCs (i.e., processes) is known,
then conversion factors can be calculated for that group of SCCs based on the percentages.
Using the generic PM profiles, the necessary assumptions to proceed from this point are as follows:
Emissions less than or equal to 1 |im are PM-CON emissions.
Emissions less than or equal to 10 |im are PM10-PRI emissions (since they include emissions less
than or equal to 1 |im as well).
• The weight percent for PM-PRI will always be 100%.
PM-FIL and PM10-FIL weight percentages can be derived by subtracting the PM-CON weight
percentages from the PM-PRI and PM10-PRI weight percentage values, respectively.
The compilation of generic PM profiles resulted in two distinct types of profiles. Some of the PM profiles
originated from specific sections of AP-42 and were SCC specific. These profiles can be viewed in Table C-
7. Other PM profiles originated from the generic profiles found in Appendices B.l and B.2 of AP-42. These
profiles were allocated to more general, 6-digit SCCs; they can be viewed in Table 8. For the 6-digit profiles,
if more than one category number in the AP-42 appendix was cited for a given 6-digit SCC, the average value
of all the categories was used for cumulative weight percent of PM that was less than 1 |im and less than 10
|im, respectively.
After compiling the generic PM profiles from AP-42, there were still a large number of SCCs beginning
with 3 that did not have an attributed profile. To account for these SCCs, the first step was to further augment
the 6-digit generic PM profiles. This was accomplished by using the average of the 8-digit generic PM profiles
where they had the first 6 digits in common and those 6 digits were not already accounted for by a 6-digit
generic PM profile.
The next step was averaging of the 6-digit PM generic profiles to produce more general PM profiles to
account for even more SCCs. For example, the following 6-digit SCCs were used to produce a 5-digit SCC
profile for '30101' via averaging:
SCC Cumulative % 10 |im Cumulative % 1 |im
301012
51
4
301014
85
6
301015
85
6
301019
94
60
30101
78.75
19
Following this methodology, generic PM profiles at the 5, 4, 3, 2, and 1-digit level can be derived from the 6-
digit generic PM profiles.
Documentation for the Final 1999 National Emissions
Inventory (Version 3) for Criteria Air Pollutants and Ammonia C - 8
Point Source Methodologies
-------
Once completed, the generic profile information was then mapped to the appropriate SCCs beginning with
'3'. Those SCCs with complete PM information in FIRE and SCCs with 8-digit profiles were excluded from
the mapping process. The process started at the 6-digit SCC level and ended at the 1-digit SCC level. If an
SCC had its first 6-digits matched to a 6-digit SCC PM generic profile, then it was considered to have the PM
percentage composition of that generic profile. For instance, SCC 30101202, Chemical Manufacturing of
Hydroflouric Acid (Rotary Kiln: Acid Reactor), was assumed to have the percentage composition of the
generic '301012' PM profile from Appendix B.2: 51% <10 |im and 4% < 1 |im. If the SCC did not match on
the 6-digit level, then a match was attempted at the 5-digit level to the calculated 5-digit generic profiles. The
process was continued to the 4-digit, 3-digit, 2-digit, and even 1-digit level until a match was made.
Once all of the SCCs beginning with '3' had a PM profile in place, the calculation of conversion factors
was performed using the aforementioned assumptions. The equations utilized were as follows:
Conversion Factor from PM-PRI to PM-FIL:
PM-FILap_42 100% - (Cumulative % < 1 um)
PM-PRI AP_42 " 100% ( )
Conversion Factor from PM10-PRI to PM10-FIL:
PMIO-FILap.^ (Cumulative % < 10/im) - (Cumulative % < 1 fJtm)
PMIOPRIap.42 (Cumulative % < 10 /im) ^
Conversion Factor from PM10-FIL to PM10-CON:
PM-CONap 42 (Cumulative % < 1 /im)
= (6)
PM10-FILAp.42 (Cumulative % < 10 /im) -(Cumulative % < 1 /im)
Utilizing these equations, the conversion factors for the example SCC, 30101202, Chemical Manufacturing of
Hydroflouric Acid (Rotary Kiln: Acid Reactor) would be as follows:
Conversion Factor from PM-PRI to PM-FIL:
PM-FIL AP 42 100%-4%
= = 0.96
PM-PRI ^.42 100%
Conversion Factor from PM10PRI to PM10-FIL:
PM10-FIL Al, 42 51% - 4% „ _
^2- = = 0.92
PM10-PRIap_42 51%
Conversion Factor from PM10-FIL to PM10-CON:
PM-CONAP42 4%
ap^= = 00g5
PM10-FIL 51%-4%
The conversion factors for the other SCCs beginning with 3 were also calculated in this manner. The
results were then combined with those conversion factors calculated for the SCCs with 8-digit generic PM
profiles, the SCCs with complete PM information in FIRE, and the SCCs with the emission factor information
from the AP-42 text. All conversion factors can be viewed in Table C-9.
Documentation for the Final 1999 National Emissions
Inventory (Version 3) for Criteria Air Pollutants and Ammonia C-9
Point Source Methodologies
-------
Conversion Factors for Petroleum and Solvent Evaporation (SCCs Starting with 4)
For Version 2.0 of the NEI, condensible PM emissions were estimated for coating oven heaters and fuel
fired equipment used in surface coating operations. Because of the lack of data for estimating PM-CON
emissions for these sources, conversion factors for electric utility external combustion boilers were used as
surrogate factors. Fuel type was used to match the conversion factors for external combustion boilers to the
SCCs for coating oven heaters and fuel fired equipment. Table C-10 shows how the SCCs were matched and
the conversion factors.
Conversion Factors for Solid Waste Disposal (SCCs Starting with 5)
An additional complication was presented by the information contained in Solid Waste Disposal (Chapter
2) which generally pertained to point source SCCs beginning with 5. Units for the emission factors were in
lbs/ton of fuel. Since the fuel could differ for each SCC, there is no definitive way to interpolate factor values
for SCCs that did not have the values directly presented. This was not an issue for the External Combustion
Sources and Stationary Internal Combustion Sources, since enough information was provided to convert all of
those emissions factors to units of lb/MMBtu. However, the most significant problem for the Solid Waste
Disposal emission factors was a complete lack of information on PM-PRI, PM10-PRI, or PM-CON emissions
factors. Therefore, the Solid Waste Disposal conversion factors were estimated from the conversion factors
for External Combustion Sources and Stationary Internal Combustion Sources.
To estimate conversion factors for the Solid Waste Disposal SCCs, average emissions factors from the
External Combustion Sources and Stationary Internal Combustion Sources were used. The average values
used (displayed below in lbs/MMBtu) were those calculated directly from information presented in AP-42,
previous to filling in information via the truncated SCC averaging method. Conversion factors were calculated
as follows:
Average Values Used for Solid Waste Disposal Conversion Factor Calculations
PM Category Average Emissions Factor (lb/MMBtu)
PM-FIL
1.0728
PM10-FIL
0.3383
PM-CON
0.0342
Documentation for the Final 1999 National Emissions
Inventory (Version 3) for Criteria Air Pollutants and Ammonia C-10
Point Source Methodologies
-------
(PM-PRIstate -» PM-FILSTATE) conversion factor:
PM-FILap.42 1.0728 i
= ir 5-— = 0.9691
PM-PRI„.42 1.0728 „+ 0,0342 „
(PM10-PRIstate -> PM10-FILSTATE) conversion factor:
PM10-FILAP42 0.3383 ^
PM10-PRIap_42 " 0.3383 0.0342 " "
(PM10-FILstate -> PM-CONSTATE) conversion factor:
1U
PM-CONap_42 _ 0.0342 lb
PM10-FIL.p 47 0 3383
AP-42 MMBtu
AP-42 MMjjtu q IQll
These factors were for all point source SCCs beginning with 5. For documentation purposes, the SCCs of
interest are listed along with these factors in Table C-l 1.
c. Step 3: Prepare Factors from PM Calculator
S/L/T inventory data for PM/PM10/PM25 emissions include thousands of records. The NEI is prepared
in a large Oracle data base in MF 2.0. Consequently, because the PM Calculator is a computer desk-top
tool, the PM Calculator could not be used to efficiently process the thousands of records of data in the NEI.
Therefore, the PM Augmentation procedure was conducted in Oracle. Oracle scripts were written to use the
underlying particle size data in the PM Calculator to produce uncontrolled and controlled emissions for every
SCC and control device combination available in the PM Calculator to estimate PM10-FIL and PM25-FIL
emissions using 100 tons of uncontrolled PM-FIL as input to the calculator. The emissions output was then
used to prepare ratios to apply to S/L/T inventory data. The output table is too large to present in a table in this
attachment. Therefore, the output table is stored in an Access 2000 data base named "PM Calculator Generic
Output.mdb". This table was subdivided in the Access data base by SCC groups. The table contains the
following fields:
SCC
Primary Device Type Code
Secondary Device Type Code
PMCalc PMFIL Uncontrolled: Uncontrolled PM-FIL emissions (100 tons)
PMCalc PMlOFIL Uncontrolled: Uncontrolled PM10-FIL emissions (tons)
PMCalc_PM25FIL_Uncontrolled: Uncontrolled PM25-FIL emissions (tons)
PMCalc PMlOFIL Controlled: Controlled PM10-FIL emissions (tons)
PMCalc_PM25FIL_Controlled: Controlled PM25-FIL emissions (tons)
PMCalc PMlOFIL Control EFF: Overall PM10-FIL control efficiency for control device combinations
specified
PMCalc_PM25FIL_Control_EFF: Overall PM25-FIL control efficiency for control device combinations
specified
Documentation for the Final 1999 National Emissions
Inventory (Version 3) for Criteria Air Pollutants and Ammonia C -11
Point Source Methodologies
-------
d. Step 4: Develop and Apply Algorithms to Estimate Emissions from S/L/T
Inventory Data
The S/L/T data were grouped to determine the different combinations of pollutants S/L/T submitted for
individual process (SCC) and control device combinations. A total of 14 groups of data were identified, and
some of the groups were further divided into uncontrolled and controlled emissions. Table C-12 identifies the
14 groups of data supplied by S/L/T agencies. For each group of data, the second column of Table C-12
identifies the pollutant used as input to each algorithm; all of the pollutants supplied for a process are listed in
parenthesis. The third column identifies the control status, and the fourth column identifies the algorithms
applied to estimate emissions for missing pollutants. Note that, although not shown in the fourth column in
Table C-12, the procedure sums PMIO-FIL and PM-CON to estimate PM10-PRI, and sums PM25-FIL and
PM-CON to estimate PM25-PRI.
For example, Group 1 identifies emission processes for which S/L/T agencies supplied uncontrolled PM-
PRI and PM25-PRI emissions. For this group, PM25-PRI emissions are used as the starting point in the
algorithms to calculate uncontrolled PM25-FIL, PM IO-FIL, and PM-CON emissions. The following equation
was used to estimate PM25-FIL emissions from PM25-PRI emissions for a given SCC:
pm25fil = pm25pri * pmlOpri_to_pmlOfil
In this equation, PM25-PRI emissions are from the S/L/T inventory. The term "pmlOpri_to_pmlOfil" identifies
where the conversion factor (SCC specific) were used to calculate PM25-FIL from PM25-PRI. Factors to
convert PM25-PRI to PM25-FIL were not developed; therefore, S/L/T-supplied PM25-PRI emissions were
converted to PM25-FIL emissions using the conversion factors for PM10-PRI_to_PM10-FIL.
The following equation was used to estimate PM IO-FIL emissions:
pmlOfil = pm25fil * (pmcalc_pmlOfil_uncontrolled / pmcalc_pm25fil_uncontrolled)
In this equation, PM25-FIL calculated from the previous equation is multiplied by the term
"pmcalc_pmlOfil_uncontrolled / pmcalc_pm25fil_uncontrolled" which identifies the fields from the PM
Calculator output file ("PM Calculator Generic Output.mdb") used to calculate a ratio to adjust uncontrolled
PM25-FIL emissions to uncontrolled PMIO-FIL emissions.
The following equation is then used to estimate PM-CON emissions by subtracting calculated PM25-FIL
emissions from S/L/T-supplied PM25-PRI emissions.
pmcon = pm25pri - pm25fil
For the Group 1 controlled scenario, the equations are the same as those for the uncontrolled scenario
except that the controlled emission fields in the PM Calculator output file are used to adjust controlled PM25-
FIL emissions to controlled PMIO-FIL emissions for each SCC and control device combination.
e. Step 5: Review Results and Update the NEI with Emission Estimates and
Control Information
Documentation for the Final 1999 National Emissions
Inventory (Version 3) for Criteria Air Pollutants and Ammonia C-12
Point Source Methodologies
-------
After completing the emission estimates for pollutants missing from S/L/T inventories, the emission
estimates are compared to the S/L/T inventory data to identify data issues. For example, the procedures
resulted in negative PM25 emissions for a small number of emission units even though the algorithms key off of
S/L/T-supplied emissions for other pollutants. For these records, negative values are set to zero.
The PM 10-FIL to PM-CON conversion factors were developed using uncontrolled emission factors or
particle-size data. However, when the factors were applied to uncontrolled PM10-FIL emissions that were
back-calculated using control efficiencies, the PM-CON emissions and resulting PM10-PRI and PM25-PRI
emissions were overestimated considerably. As a result, EPA applied the conversion factors to controlled
PM 10-FIL emissions if an emission source was controlled which produced much more realistic emission
estimates.
S/L/T-supplied filterable, condensible, and primary PM10 and PM25 emissions are maintained in the
NEI. Records for missing pollutants are inserted into the Emission table and populated with the pollutant code
and annual emissions estimated using the augmentation procedure. If an emission source is controlled, a
records for the missing pollutants are inserted into the Control Equipment table and populated with the pollutant
code, overall control efficiency, and control device codes from the PM Calculator. Thus, for a given emission
source, the control efficiencies and control device codes associated with S/L/T-supplied pollutants may be
different from the pollutants generated by the augmentation procedure and added to the Control Equipment
table. This decision was made in order to report the control efficiencies and control device codes upon which
the augmented emission estimates are based.
3. UNCERTAINTIES/FUTURE RESEARCH
The most significant uncertainty associated with the PM augmentation procedures applied to populate the
NEI with condensible, filterable, and primary PM10 and PM25 is the lack of particle-size-specific emission
factors for uncontrolled sources and controlled sources for various types of control equipment combinations.
The particle-size-specific emission factors for uncontrolled and controlled stationary external and internal fuel
combustion sources in AP-42 are good. However, good data for other SCCs are lacking. Thus, S/L/T
agencies should conduct research to prioritize source categories of fine PM emissions and focus on improving
the data needed to estimate fine PM emissions for the highest priority categories.
The PM augmentation procedure relies on the use of the PM Calculator to estimate PM10-FIL and
PM25-FIL emissions from S/L/T-supplied PM-PRI, PM-FIL, or PM10-PRI emissions. Although the PM
Calculator uses particle size data available in AP-42 and FIRE, it will not provide estimates that will be as
accurate as applying the emission factors to the activity, operating schedule, and control equipment data for
individual sources. Thus, S/L/T agencies are encouraged to use the emission factors in AP-42 and FIRE with
source-specific activity, operating schedule, and control equipment data to prepare their inventories in the
future.
The PM Calculator does not contain particle size data for several SCCs and control devices for which
S/L/T agencies reported PM, PM10, and/or PM25 emissions in their inventories. The augmentation
procedures can be improved by updating the PM Calculator to include particle size data for SCCs and control
equipment currently not included in the calculator.
Documentation for the Final 1999 National Emissions
Inventory (Version 3) for Criteria Air Pollutants and Ammonia C -13
Point Source Methodologies
-------
S/L/T inventories contain several inaccuracies in the PM data. For example, for NEI Version 2, some
agencies did not report the form of PM, PM10, and PM25 emissions requiring EPA to make a decision on if
the emissions included condensibles or not when the agency could not be reached to provide clarification. For
NEI Version 2 and draft Version 3, some agencies did not report control device codes that complied with NIF
2.0 codes which made it difficult to determine how to match the codes to control equipment codes in the PM
Calculator. Mismatching of S/L/T codes to codes in the PM Calculator probably occurred resulting in
increasing the inaccuracy of the emission estimates. Several agencies reported PM25 emissions as being
significantly higher than PM10 emissions for a given process. For these cases, PM25 emissions were set equal
to PM10 emissions if the agency did not provide corrections. However, this correction probably resulted in
inaccurate emission estimates. Thus, S/L/T agencies need to compare PM25 and PM10 emission estimates to
make sure that PM25 emissions are less than or equal to PM10 emissions.
S/L/T agencies should note that the PM augmentation procedures need further refinement. For example,
EPA is aware that the procedures calculate PM-CON for industrial SCCs (e.g., industrial paved and unpaved
roads, material handling operations in the mineral products and primary and secondary metal production
industries) that should not be sources of condensible emissions. The EPA will review the Industrial Processes
(SCCs starting with 3) and Petroleum and Solvent Evaporation (SCCs starting with 4) sectors shown in Tables
C-9 and C-10, respectively, for which PM-CON is estimated. For final Version 3, EPA will remove PM-
CON emissions for any SCCs it decides are not sources of PM-CON. The EPA requests comments on this
issue.
The conversion factors for Solid Waste Disposal PM emissions were produced from the limited data
provided in the External Combustion Sources and Stationary Internal Combustion Sources sections of AP-42.
Further work is needed to identify alternative sources of data to develop condensible PM emission factors for
Solid Waste Disposal SCCs.
Documentation for the Final 1999 National Emissions
Inventory (Version 3) for Criteria Air Pollutants and Ammonia C -14
Point Source Methodologies
-------
United States Office of Air Quality Planning and Standards Publication No. EPA-454/B-20-020
Environmental Protection Air Quality Assessment Division January 2004
Agency Research Triangle Park, NC
------- |