* _ \
KWJ
*1 PRO1^
NEI Quality Assurance and Data Augmentation
for Point Sources
-------
-------
EP A-454/R-20-001
February 2006
NEI Quality Assurance and Data Augmentation for Point Sources
U.S. Environmental Protection Agency
Office of Air Quality Planning and Standards
Air Quality Assessment Division
Research Triangle Park, NC
-------
TABLE OF CONTENTS
Sections Page
1. Introduction 1
2. Inventory Information 16
2.1 Record Type 16
2.2 Submittal Flag 16
2.3 Transaction Type 16
2.4 Inventory Year 17
2.5 Inventory Type Code 17
2.6 Source Type 17
2.7 Affiliation Type 17
2.8 Format Version 17
2.9 Revision Date 17
3.
Geographic Information
18
3.1
State and County FIPS Code
18
3.2
Tribal Code
19
3.3
Urban/Rural Classification
20
3.4
MSA Name and Code
21
3.5
Country
21
4.
Facility Information
22
4.1
NEI Unique Facility ID
23
4.2
NEI Standardized Facility Name
26
4.3
NEI Standardized Address
26
4.4
NEI Standardized City
26
4.5
NEI Standardized State
26
4.6
NEI Standardized Zip Code
26
4.7
Location Address
26
4.8
City
27
4.9
State
27
4.10
Zip Code
27
4.11
ORIS Facility Code
27
4.12
Facility Registry ID
28
4.13
TRIID
28
4.14
Dun and Bradstreet Number
28
4.15
BART Source and BART Unit
28
4.16
IPM Facility and IPM Unit
29
February 01, 2006
li
-------
5. Geocoordinates 30
5.1 Geocoordinates QA and Default Procedures 30
5.2 X Coordinate (longitude) 34
5.3 Y Coordinate (latitude) 34
5.4 UTM Zone 34
5.5 XY Coordinate Type 34
5.6 Location Default Methodology 35
5.7 Latitude/Longitude EPA Data Standards 35
Stack Latitude 35
Stack Longitude 36
Horizontal Collection Method Code 37
Horizontal Reference Datum Code 37
Coordinate Data Source Code 37
Reference Point Code 37
Horizontal Accuracy Measure 37
Source Map Scale Number 37
5.8 NEI Site Latitude 39
5.9 NEI Site Longitude 39
5.10 Latitude/Longitude QC Report 39
6 Unit and Process IDs 42
6.1 Emission Data Level 42
6.2 Emission Unit ID 42
6.3 Process ID 43
6.4 Emission Release Point ID 43
7 Source Description 44
7.1 NAICS Code 44
7.2 SIC Code 45
7.3 NAICS/SIC Code Default Methodology 46
7.4 SCC 47
7.5 Process MACT and Section 112(k) Area Source Standards Code 47
7.6 MACT and Section 112(k) Area Source Standards Code
Assignment 50
7.7 MACT and Section 112(k) Area Source Standards Compliance
Code 51
7.8 Facility Category 52
8 Emission Release Point Description and Parameters 53
8.1 Emission Release Point Type 53
8.2 Stack Parameters QA and Default Procedure 53
8.3 Stack Parameter Default Flag 58
8.4 Stack Parameters QC Report 58
February 01, 2006
in
-------
9 Unit Codes 61
9.1 Unit Codes 61
9.2 Design Capacity, Design Capacity Numerator and Denominator 61
10. Activity Data Fields 63
10.1 Emission Type 64
10.2 Start and End Dates 64
10.3 Start and End Times 64
10.4 Year of Emission Estimate 64
10.5 Annual Average Days Per Week 65
10.6 Annual Average Hours Per Day 65
10.7 Annual Average Hours Per Year 65
10.8 Annual Average Weeks Per Year 65
10.9 Period Days Per Week 65
10.10 Period Hours Per Day 65
10.11 Period Hours Per Period 65
10.12 Period Weeks Per Period 66
10.13 Winter, Spring, Summer, and Fall Throughput Percentages 66
11. Control Equipment 67
11.1 Control Status 67
11.2 Primary Device Type Code, Secondary Device Type Code,
Third Device Type Code, and Fourth Device Type Code 67
11.3 Total C apture C ontrol Effl ci ency 68
11.4 Total Capture Control Efficiency Methodology Flag 69
12. Emissions 70
12.1 Pollutants 70
NEI Pollutant Code Table 70
CAPHAP Data Field 72
Chemical Data Standards Lookup Table in the NEI 72
HAPs Reporting 73
Preferred Reporting of HAPs 74
Q A of HAP Pollutant Codes 78
CAPs Reporting 79
12.2 PM Augmentation 80
Steps in PM Augmentation 81
Creation of PM Ratios 87
References 93
12.3 VOC and PM Augmentation from HAPs 93
12.4 CAP Augmentation for Missing CAPs 94
12.5 HAP Augmentation from VOC and PM 94
12.6 HAP Performance Level 94
12.7 QC of HAP Category Group Emissions 95
February 01, 2006
iv
-------
12.8 Emissions Type Period 100
12.9 Annual Emissions 101
12.10 Non-Annual Emissions 104
12.11 Emission Ton Value 105
12.12 Emission QC 105
Emission Content QC Report 107
Other Summary Files to Assist in Emission Content QC 108
Toxicity Weighting of HAPs 111
12.13 Emission Data Quality Rating for HAPs 113
12.14 Data Source 116
12.15 Maintaining an Emissions History Table 118
13. Other Fields that undergo QA but are not augmented 120
14. Default Flags 121
February 01, 2006
v
-------
List of Tables
Tables Page
1. NIF V3.0 Point Source Table Description 3
2. Fields in NIF V3.0 Point Source Tables 4
3. Additional Point Source Data Fields in the NOF that are
not in the NIF V3.0 14
4. Additional Point Source Data Fields in NEI Auxiliary Files that
are not in NIF V3.0 or NOF 15
5. MS As Assigned to Counties in New England States Associated
with More Than One MSA for Use In The County U1/U2/R
Classification Process 21
6. Location Default Flags 35
7. EPA Latitude/Longitude Data Standards. 36
8. Geocoder Default Flags and Default Values for
Latitude/Longitude Data Standard Elements 38
9. HAP Emission Data Levels 42
10. NAICS/SIC Code Default Flags 46
11. MACT Code Facility List Flags 49
12. Stack Parameter Data Replacement Matrix 56
13. Fields in NIF Requiring Unit Codes 61
14. Total Capture Control Efficiency Methodology 68
15. NEI Pollutant Code Table 71
16. NEI Chemical Data Standards Table 73
17. POM Compounds 76
18. List of Compounds Often Mistaken as Glycol Ethers 78
19. PM Compounds Reported in the 2002 NEI 79
20. PM Cases and Required Steps to Augment PM 84
21. PM Augmentation: SCCs Starting with "402" and the
Surrogate SCCs 90
22. PM Augmentation: SCCs Starting With "5" That Have
Combustion Sources (Incinerators and Flares) That Can Use
Average Emission Factors from External Combustion Sources
and Stationary Internal Combustion Sources 91
23. Methodology Used to Generate Annual Emissions 101
24. Unit Conversions for Tons 105
25. Table Structure for ALL NEI 109
February 01, 2006
vi
-------
SECTION 1. INTRODUCTION
The Emission Inventory and Analysis Group (EIAG) in the Environmental Protection Agency
(EPA) compiles the National Emission Inventory (NEI) for hazardous air pollutants (HAPs) and
criteria air pollutants (CAPs). The NEI plays an important role in air quality management
activities such as emission trends, rule and policy development and risk assessment. The Clean
Air Act (CAA) includes many mandates for the EPA related to CAPs and HAPs. The NEI is a
tool that EPA can use to meet the CAA mandates. To support many of the these functions, the
NEI must contain the data necessary for air quality modeling.
The NEI is a critical component of the EPA's National Air Toxics Program. The CAA presents a
list of 188 HAPs for which EPA is to identify their sources, quantify their emissions by source
category, develop regulations for each source category, and assess public health and
environmental impacts after the regulations are put into effect. The NEI for HAPs is compiled in
order to determine if CAA programs are successful in reducing emissions and human health and
environmental risk due to HAPs emissions. The initial objective is to make the NEI HAP data
available to EPA modelers for use in the National Air Toxics Assessment (NATA) and for use in
residual risk assessments conducted by EPA.
The NEI for CAPs are used in State Implementation Plans (SIPs), compliance demonstrations,
emissions trading, and in modeling activities designed to evaluate ambient air concentrations.
Title I, Section 110 of the CAA requires states to submit emission inventories for CAPs as part
of their SIPs. EPA collects NEI CAP data under the Consolidated Emissions Reporting Rule
(CERR) (40 CFR Part 51). The CERR specifies two sets of reporting thresholds for CAPs.
Type A (large sources) must report annually, while Type B sources must report every three
years. For the 2002 NEI, EPA collected information on both Type A and Type B sources. (For
more information on CERR, see: /www.epa.gov/ttnchiel/cerr/index.htmn
The NEI contains estimates of facility-specific HAP and CAP emissions and their
source-specific parameters necessary for modeling such as location and facility characteristics
(stack height, exit velocity, temperature, etc.). Complete source category coverage is needed,
and the NEI contains estimates of emissions from stationary point and nonpoint (stationary
sources such as residential heating that are inventoried at the county level) and mobile source
categories. The NEI contains individual stack and fugitive estimates at individual
geocoordinates for point sources. County level estimates are provided in the NEI for nonpoint
and mobile sources. Point source categories of HAPs include major and area sources as defined
in Section 112 of the CAA. Nonpoint source categories of HAPs include area sources and other
stationary sources that may be more appropriately addressed by other programs rather than
through regulations developed under certain air toxics provisions (Sections 112 or 129) in the
CAA.
The major steps involved in compiling the 2002 NEI include:
Submittal of 2002 inventory data by state and local agencies, tribes, industry, and EPA
February 01, 2006
1
-------
offices;
Blending/Merging of data from multiple data sources;
Augmentation of data for missing data elements;
QC/QA of data;
• Preparation of draft 2002 NEI for external and internal review;
• Incorporation of external and internal review comments on the draft 2002 NEI and
incorporation of new inventory data submitted during review period; and
• Preparation of final 2002 NEI.
Important steps in preparing point source files are Quality Assurance (QA)/Quality Control (QC)
and augmentation of data. QA and data augmentation are needed to prepare point source files for
use in air quality and risk and exposure modeling. In addition, the QA of data and data
augmentation are needed in order for the NEI to meet recent EPA data standards and Office of
Management and Budget's (OMB) Information Quality Guidelines. In response to guidelines
issued by OMB under Section 515(a) of the Treasury and General Government Appropriations
Act for Fiscal Year 2001 (P.L. 106-554; H.R. 5658), EPA issued the Guidelines for Ensuring
and Maximizing the Quality, Objectivity, Utility, and Integrity of Information Disseminated by
the Environmental Protection Agency. This report focuses on the QA and augmentation of data
in the NEI point source inventory.
We conduct a variety of QA activities to identify point source records with referential integrity
problems, duplicate records, and records with missing or out-of-range parameters which are
needed for air quality and exposure modeling. After conducting QA, we augment data using the
methodology in this report. We summarize the errors found and provide reports back to the data
providers on the QA findings. We track errors, their resolution and all communications on a
QA/QC form, through emails, and in a phone log as part of documentation. These tracking
mechanisms help to ensure the transparency and reproducibility of the NEI. This is a
requirement of OMB Information Quality Guidelines, but it also helps us establish an electronic
trail for each record in the NEI. We have created a QA/QC process and tracking database to
provide feedback reports to data providers at regular intervals during the QA of the data. The
feedback reports include the following 4 QC reports.
Data QC Report - summarizes structural problems with the data, e.g, data errors,
incorrect codes, data integrity issues, etc.
Latitude/Longitude QC Report - lists latitude/longitudes that failed QC review and
provides EPA defaulted coordinates.
Stack Parameters QC Report - lists stack parameters that failed QC review and
provides EPA defaulted parameters.
Emission Content QC Report - compares 2002 emissions data to 1999 emissions data,
and highlights potential outliers, missing facilities, and missing and erroneous pollutants.
February 01, 2006
2
-------
We archive all files submitted by data providers, records removed during QA, records
augmented, and each iteration of the draft and final NEI.
We first resolve records with referential integrity problems and duplicates. Then after
identifying parameters and data fields with missing or out-of-range values during the QA of
files, we augment the data using the methodology in this document. The NEI data files identify
all fields of data that have defaulted parameters. Data augmentation occurs at different times in
the compilation of the NEI. For example, augmentation of location coordinates occurs prior to
blending/merging of data, while augmentation of stack parameters occurs after blending/merging
of data.
Data providers submit their data to EPA using the NEI Input Format (NIF) or an extensible
markup language (XML) format. Both formats are described and located at
http://www.epa.gov/ttn/chief/nif/index.html. The 2002 point source data were submitted to EPA
in NIF V3.0 or NEI XMLV3.0 for incorporation into the 2002 NEI. Table 1 describes the eight
tables in the NIF for point sources.
Table 1. NIF V3.0 Point Source Table Description
NIF Table
Abbreviation
Description
Transmittal
TR
organization submitting to EPA, data year, source types
Site
SI
plant location and IDs
Emission Unit
EU
unit or device generating point emissions
Emission Process
EP
operational process generating emissions
Emission Period
PE
activity data and time period for releases
Emission Release
Point
ER
mechanism releasing emissions to the air including stack
parameters and geographic coordinates
Control Equipment
CE
control equipment and efficiency
Emission
EM
amount of emissions per pollutant
Table 2 presents a list of all fields in the NIF V3.0 Point Source Tables. Table 2 indicates if the
field has a methodology for data augmentation in this document. For all fields of data in Table 2
that do not have a data augmentation procedure, data submitters will need to populate the fields
of data. Table 2 also indicates those fields of data for which we will only conduct QA and will
be unable to default invalid values.
Table 3 shows additional data fields in the NEI that are included in the NEI Output Format
(NOF) Tables. All NIF data fields are included in the NOF. Table 4 shows additional data fields
in the NEI that are included in auxiliary files. Data for all fields in Tables 3 and 4 will be
generated by EIAG.
February 01, 2006
3
-------
The QA and Augmentation Procedure for fields of data in Tables 2 - 4 are presented in Sections
2 - 14 of the report. Fields of data have been grouped by function into the following sections of
the report.
Section 2
Inventory Information
Section 3
Geographic Information
Section 4
Facility Information
Section 5
Geocoordinates
Section 6
Unit and Process IDS
Section 7
Source Description
Section 8
Emission Release Description and Parameters
Section 9
Unit Codes
Section 10
Activity
Section 11
Control Equipment
Section 12
Emissions
Section 13
Other Fields that undergo QA but are not augmented
Section 14
Default Flags
This document revises the previous report, "NEI Quality Assurance and Data Augmentation for
Point Sources", February 1, 2005. Any comments or questions concerning this report should be
sent to Anne Pope, pope.anne@epa.gov.
Table 2. Fields in NIF V3.0 Point Source Tables
Ml Table
Abbrc\ iitlion
Held Name
Held Description
Augmentation
Procedure
Report
Section
PE
Actual
Throughput
Throughput of process activity
TR
Affiliation
Type
Report certifier
NIF Code Table
Augmentation
Procedure
2.7
EP
Annual
Average Days
per Week
Average days per week process is
operating
Augmentation
Procedure
10.5
EP
Annual
Average Hours
per Day
Average hours per day process is
operating
Augmentation
Procedure
10.6
EP
Annual
Average Hours
per Year
Average hours per year process is
operating
Augmentation
Procedure
10.7
February 01, 2006
4
-------
Table 2 continued. Fields in NIF V3.0 Point Source Tables
Ml Table
Abbrc\ iitlion
Held Name
Held Description
Augmentation
Procedure
Report
Section
EP
Annual
Average
Weeks per
Year
Average weeks per yr process is
operating
Augmentation
Procedure
10.8
EP
Ash Content
Ash content of fuel
SI
City
City of physical location
Augmentation
Procedure
4.8
TR
Contact Person
Name
EPA's single pont of contact
regarding questions on inventory
submittal
TR
Contact Phone
Number
Phone number for contact person
EM
Control Status
Controlled or Uncontrolled
Augmentation
Procedure
11.1
CE
Control System
Description
Description of control devices
ER
Coordinate
Data Source
Code
Data source providing latitude and
longitude
NIF Code Table
Augmentation
Procedure
5.7
SI
Country
Country Name
Augmentation
Procedure
3.5
EU
Design
Capacity
Operational Capacity for emission
unit
Augmentation
Procedure
9.2
EU
Design
Capacity
Denominator
Unit of measure for design
capacity
NIF Code Table - Unit codes
Augmentation
Procedure
9.1
9.2
EU
Design
Capacity
Numerator
Unit of measure for design
capacity
NIF Code Table - Unit codes
Augmentation
Procedure
9.1
9.2
SI
Dun and
Bradstreet
Number
Dun and Bradstreet number for
facility
Augmentation
Procedure
4.14
EM
EF Reliability
Indicator
AP-42 Rating
NIF Code Table
QA
13
February 01, 2006
5
-------
Table 2 continued. Fields in NIF V3.0 Point Source Tables
Ml Table
Abbrc\ iitlion
Held Name
Held Description
Augmentation
Procedure
Report
Section
TR
Electronic
Address Text
Email address
TR
Electronic
Address Type
Name
Email, Internet, Intranet, HTTP,
ftp, etc
NIF Code Table
EM
EM Reliability
indicator
Quality of emissions estimate
EM
Emission
Calculation
Method Code
How were emissions derived?
NIF Code Table
QA
13
EM
Emission Data
level
Level HAP emissions reported at -
site, unit, stack, process
Augmentation
Procedure
6.1
EM
Emission
Numeric Value
Augmentation
Procedure
12
EP
Emission
Process
Description
Emission process description
ER
Emission
Release Point
Description
EP, ER, EM
Emission
Release Point
ID
ID for point/location where
emissions are released to the air
Augmentation
Procedure
6.4
ER
Emission
Release Point
Type
Code for physical configuration of
the release point
NIF Code Tables
Augmentation
Procedure
8.1
EM
Emission Type
Temporal description of emissions
reported (annual, weekday, etc.)
NIF Code Table
Augmentation
Procedure
10.1
EU
Emission Unit
Description
Emission unit description
EU, EP, PE,
CE, EM
Emission Unit
ID
ID assigned to identify emission
units within a facility
Augmentation
Procedure
6.2
February 01, 2006
6
-------
Table 2 continued. Fields in NIF V3.0 Point Source Tables
Ml Table
Abbrc\ iitlion
Held Name
Held Description
Augmentation
Procedure
Report
Section
EM
Emission Unit
Numerator
Unit of measure for emissions
NIF Code Table - Unit codes
Augmentation
Procedure
9.1
PE, EM
End Date
End date of period of emissions
Augmentation
Procedure
10.2
PE, EM
End Time
End time of period
Augmentation
Procedure
10.3
ER
Exit Gas Flow
Rate
Flow rate of an exit gas (cu ft/sec)
Augmentation
Procedure
8.2
ER
Exit Gas
Temperature
Temperature of an exit gas stream
(degrees F)
Augmentation
Procedure
8.2
ER
Exit Gas
Velocity
Velocity of an exit gas (ft/sec)
Augmentation
Procedure
8.2
SI
Facility
Category
Major or Area
NIF Code Table
Augmentation
Procedure
7.8
SI
Facility Name
Name of Facility
SI
Facility
Registry ID
EPA Facility Registry ID assigned
by EPA's OEI
Augmentation
Procedure
4.12
EM
Factor
Numeric Value
EM
Factor Unit
Denominator
Unit of measure for emission
factor denominator
NIF Code Table - Unit codes
Augmentation
Procedure
9.1
EM
Factor Unit
Numerator
Unit of measure for emission
factor numerator
NIF Code Table - Unit codes
Augmentation
Procedure
9.1
EP
Fall
Throughput
Percent
Percentage of process that operates
during the fall months (Sept, Oct,
Nov)
Augmentation
Procedure
10.13
TR
Format
Version
NIF version
Augmentation
Procedure
2.8
CE
Fourth Device
Type Code
Fourth control device
NIF Code Tables
Augmentation
Procedure
11.2
February 01, 2006
7
-------
Table 2 continued. Fields in NIF V3.0 Point Source Tables
Ml Table
Abbrc\ iitlion
Held Name
Held Description
Augmentation
Procedure
Report
Section
ER
Fugitive
Dimensions
Unit
Unit of measure
NIF Code Table - Unit codes
Augmentation
Procedure
9.1
EM
HAP
Performance
Level
Actual, Allowable, Maximum,
Potential
NIF Code Table
Augmentation
Procedure
12.5
EP
Heat Content
Heat content of fuel in MMBTU
per tons of coal, per 1000 barrels
of oil, or million SCG of natural
gas
ER
Horizontal
Accuracy
Measure
Measure in meters of accuracy of
the latitude and longitude
coordinates
Augmentation
Procedure
5.7
ER
Horizontal
Area Fugitive
Horizontal area of fugitive
emissions
ER
Horizontal
Collection
Method Code
Method used to determine latitude
and longitude
NIF Code Table
Augmentation
Procedure
5.7
ER
Horizontal
Reference
Datum Code
Reference date used to determine
latitude and longitude coordinates
NIF Code Table
Augmentation
Procedure
5.7
TR
Incremental
Submission
Number
Number of submission
TR
Inventory year
Year of inventory
Augmentation
Procedure
2.4
TR
Inventory Type
Code
Does data contain HAPs, CAPs or
both?
NIF Code Table
Augmentation
Procedure
2.5
SI
Location
Address
Physical location of the main
entrance of facility
Augmentation
Procedure
4.7
PE, EM
Material
Type of material processed
NIF Code Tables
QA
13
PE, EM
Material I/O
Is material input or output?
NIF Code Tables
QA
13
February 01, 2006
8
-------
Table 2 continued. Fields in NIF V3.0 Point Source Tables
Ml Table
Abbrc\ iitlion
Held Name
Held Description
Augmentation
Procedure
Report
Section
EU
Maximum
Nameplate
Capacity
Rated design capacity at 100%
operation
SI, EU
NAICS Code
North American industry
Classification Code
NIF Code Table
Augmentation
Procedure
7.1
SI
NTI Site ID
1996 NTI ID
TR
Organization
Name
Name of organization submitting
data
EU
ORIS Boiler
ID
Unique ID for electric generating
facilities
SI
ORIS Facility
Code
Unique ID for electric generating
facilities
Augmentation
Procedure
4.11
CE
Pet Capture
Efficiency
Percent capture efficiency of
control devices
PE
Period Days
per Week
Average number days per week
process is active within period;
Parameter is applied when
emission period is less than annual
Augmentation
Procedure
10.9
PE
Period Hours
per Day
Average number hours per day
process is active within period;
Parameter is applied when
emission period is less than annual
Augmentation
Procedure
10.10
PE
Period Hours
per Period
Average number hours per period
process is active within period;
Parameter is applied when
emission period is less than annual
Augmentation
Procedure
10.11
PE
Period Weeks
per Period
Average number weeks per period
process is active within period;
Parameter is applied when
emission period is less than annual
Augmentation
Procedure
10.12
CE, EM
Pollutant Code
Pollutant Code
NIF Code Table
Augmentation
Procedure
12.1
February 01, 2006
9
-------
Table 2 continued. Fields in NIF V3.0 Point Source Tables
Ml Table
Abbrc\ iitlion
Held Name
Held Description
Augmentation
Procedure
Report
Section
CE
Primary
Device Type
Code
Primary control device
NIF Code Tables
Augmentation
Procedure
11.2
CE
Primary PCT
Control
Efficiency
Percent effectiveness of primary
control system
EP, PE, CE,
EM
Process ID
ID assigned to identify emission
processes within a facility
Augmentation
Procedure
6.3
EP
Process MACT
Code
Maximum Achievable Control
Technology source category code
NIF Code Tables
Augmentation
Procedure
7.5
EP
Process MACT
Compliance
Code
Major/Area classification under
Sections 112 and 129
NIF Code Table
Augmentation
Procedure
7.7
TR, SI, EU,
EP, PE, ER,
CE, EM
Record Type
Type of NIF record
Augmentation
Procedure
2.1
ER
Reference
Point Code
Place where latitude and longitude
are established (individual stack,
plant entrance, plant centroid, etc.)
NIF Code Table
Augmentation
Procedure
5.7
ER
Release Height
Fugitive
Release height above terrain of
fugitive emissions
TR
Reliability
Indicator
Score for quality of inventory
EM
Rule
Effectiveness
Measure of percent effectiveness
of control device
EM
Rule
Effectiveness
Method
Rule effectiveness method
NIF Code Table
QA
13
EP
see
Source Classification Code
NIF Code Tables
Augmentation
Procedure
7.4
CE
Secondary
Device Type
Code
Secondary control device
NIF Code Tables
Augmentation
Procedure
11.2
February 01, 2006
10
-------
Table 2 continued. Fields in NIF V3.0 Point Source Tables
Ml Table
Abbrc\ iitlion
Held Name
Held Description
Augmentation
Procedure
Report
Section
SI, EU
SIC Code
Standard Industrial Classification
Code - being replaced by NAICS
NIF Code Tale
Augmentation
Procedure
7.2
SI
Site
Description
Comments on facility
ER
Source Map
Scale Number
Map scale
Augmentation
Procedure
5.7
TR
Source type
Point
NIF Code Table
Augmentation
Procedure
2.6
EP
Spring
Throughput
Percent
Percentage of process that operates
during the spring months (Mar,
Apr, May)
Augmentation
Procedure
10.13
ER
Stack Diameter
Stack Diameter (ft)
Augmentation
Procedure
8.2
ER
Stack Fence
line Distance
Distance from stack to fence line
in ft
ER
Stack Height
Stack Height (ft)
Augmentation
Procedure
8.2
PE, EM
Start Date
Start date of period of emissions
Augmentation
Procedure
10.2
PE, EM
Start Time
Start time of period
Augmentation
Procedure
10.3
SI
State
State of physical location
Augmentation
Procedure
4.9
TR, SI, EU,
EP, PE, ER,
CE, EM
State and
County FIPS
FIPS code for the state and county
NIF Code Table
Augmentation
procedure
3.1
SI, EU, EP,
PE, ER, CE,
EM
State Facility
ID
ID assigned by data submitter to
identify a facility
SI, EU, EP,
PE, ER, CE,
EM
Submittal Flag
Submittal Status of record - 0, A,
D, R
Augmentation
Procedure
2.2
February 01, 2006
11
-------
Table 2 continued. Fields in NIF V3.0 Point Source Tables
Ml Table
Abbrc\ iitlion
Held Name
Held Description
Augmentation
Procedure
Report
Section
EP
Sulfur Content
Sulfur content of fuel
EP
Summer
Throughput
Percent
Percentage of process that operates
during the summer months (June,
July, Aug)
Augmentation
Procedure
10.13
TR
Telephone
Number Type
Name
Fax, Mobile, Pager, Office
NIF Code Table
QA
13
CE
Third Device
Type Code
Third control device
NIF Code Tables
Augmentation
Procedure
11.2
PE
Throughput
Unit
Numerator
Unit of measure for throughput
NIF Code Table - Unit codes
Augmentation
Procedure
9.1
CE
Total Capture
Control
Efficiency
Collective efficiency for all
controls
Augmentation
Procedure
11.3
TR
Transaction
Comments
General comments regarding
transmittal
TR
Transaction
Creation Date
Creation date of transmittal data
QA
13
TR
Transaction
Type
Original or Revision
NIF Code Table
Augmentation
Procedure
2.3
SI
TRIID
TRI ID for facility
Augmentation
Procedure
4.13
TR, SI, EU,
EP, PE, ER,
CE, EM
Tribal Code
Codes for American Indian tribes
and Alaskan Native entities
NIF Code Table
Augmentation
Procedure
3.2
ER
UTM Zone
Zone number
Augmentation
Procedure
5.4
EP
Winter
Throughput
Percent
Percentage of process that operates
during the winter months (Dec,
Jan, Feb)
Augmentation
Procedure
10.13
ER
X Coordinate
Longitude (- decimal degrees)
Augmentation
Procedure
5.2
February 01, 2006
12
-------
Table 2 continued. Fields in NIF V3.0 Point Source Tables
Ml Table
Abbrc\ iitlion
Held Name
Held Description
Augmentation
Procedure
Report
Section
ER
XY Coordinate
Type
Type of coordinate system
NIF Code Table
Augmentation
Procedure
5.5
ER
Y Coordinate
Latitude (+ decimal degrees)
Augmentation
Procedure
5.3
SI
Zip Code
Zip Code of physical location
Augmentation
Procedure
4.10
February 01, 2006
13
-------
Table 3. Additional Point Source Data Fields in the NOF that are not in the NIF V3.0
Field Name
Field Description
Ml Table
Relationship
Report
Section
Augmented Data
Default flags added to indicate defaulted
fields of data
all
14
BART Source
Does facility contain BART units?
SI
4.15
BART Unit
Is unit a BART unit?
EU
4,15
CAP HAP
Is pollutant CAP or HAP?
EM
12.1
Data Rating
Qualitative rating for emission estimate
EM
12.13
Data Source
Source of Data
EM
12.14
Emission Ton
Value
Create emissions in tons
EM
12.11
Emissions Type
Period
Are emissions annual or non-annual?
EM
12.8
IPM Facility
Does facility contain IPM units?
SI
4.16
IPM Unit
Is unit an IPM unit?
EU
4.16
Location Default
Flag
How latitude/longitude coordinates are
defaulted
ER
5.6
MACT Code Flag
Method used to assign MACT Codes
ER
7.6
NAICS Code
Default Flag
How NAICS and SIC Codes are defaulted
SI, EU
7.3
Revision Date
Latest date of record - will be creation or
revision date
all
2.9
SIC Code Default
Flag
How NAICS and SIC Codes are defaulted
SI, EU
7.3
Stack Parameter
Default Flag
Indicates which stack parameters have been
defaulted
ER
8.3
Total Capture
Control Efficiency
Default Flag
Indicates how total capture control
efficiencies have been defaulted
CE
11.4
Year
Year of emission estimate
EM
10.4
February 01, 2006
14
-------
Table 4. Additional Point Source Data Fields in NEI Auxiliary Files that are not in NIF
V3.0 or NOF
Field Name
Held Description
Ml- Table
Relationship
Report
Section
MSA Code
Metropolitan Statistical Area Code
SI
3.4
MSA Name
Metropolitan Statistical Area Name
SI
3.4
NEI Unique
Facility ID
NEI Unique Facility ID assigned to all
facilities in NEI - needed for merged HAP
and CAP
SI, EU, EP, PE,
ER, CE, EM
4.1
NEI Site Latitude
Latitude of Facility
SI
5.8
NEI Site Longitude
Longitude of Facility
SI
5.9
NEI Standardized
Facility Name
Facility Name associated with NEI Site ID
SI
4.2
NEI Standardized
Address
Address associated with NEI Site ID
SI
4.3
NEI Standardized
City
City associated with NEI Site ID
SI
4.4
NEI Standardized
State
State associated with NEI Site ID
SI
4.5
NEI Standardized
Zip Code
Zip Code associated with NEI Site ID
SI
4.6
Stack Latitude
Convert Coordinates to Latitude if in UTM
ER
5.7
Stack Longitude
Convert Coordinates to Longitude if in
UTM
ER
5.7
Urban/Rural
Classification
Is county Urban or Rural using Section
112k/l 12c3 definition - Ul, U2, R
SI
3.3
February 01, 2006
15
-------
SECTION 2. INVENTORY INFORMATION
This section includes the following data fields.
Record Type
Submittal Flag
Transaction Type
• Inventory Year
Inventory Type Code
Source Type
Affiliation Type
• Format Version
Revision Date
2.1 Record Type
The Record Type indicates the type of record in each table. The Record Type for each record
within a table is the table abbreviation (TR, SI, EU, EP, PE, ER, CE, EM). If the Record Type is
missing or invalid, we will default the field based on the type of table.
2.2 Submittal Flag
The Submittal Flag indicates the status of each record in the NIF tables. Acceptable values
include the following codes.
O - Original
A-Add
D - Delete
R - Revise
If the Submittal Flag is missing in the original submission of data in June 2004, we will populate
the field with "O". If the Submittal flag is missing in the revision of data in May 2005, we will
populate the field for new records with "A" or "RA" for revised records. The deleted records
would not appear in the final NEI tables as these records would have been removed.
2.3 Transaction Type
The Transaction Type indicates if the inventory is an original submittal or a revision to the draft
2002 NEI. Acceptable NIF code values include the following codes.
00 - Original
05 - Replace
If this field is missing in the original submission of data in June 2004, we will populate the field
with "00". If the Submittal flag is missing in the revision of data in May 2005, we will populate
the field for new records with "05" for revised inventory.
February 01, 2006
16
-------
2.4 Inventory Year
The Inventory Year indicates the year of the inventory submitted to EPA. If the Inventory Year is
not included in the Transaction Table, then we will assume that the year of the data is 2002 and
augment this data field.
2.5 Inventory Type Code
The Inventory Type indicates whether the submittal contains CAPs, HAPs or both HAPs and
CAPs. Acceptable values include the following codes.
NIF Code Pollutants
We will verify that the inventory type matches the pollutants submitted in the Emissions Table. If
the code is in error, we will correct the Inventory Type Code.
2.6 Source Type
The Source Type indicates whether the submittal contains point, nonpoint, onroad or nonroad
records. The Source Type should be coded as "POINT" for point source submittals. If Source
Type is missing or in error, we will populate this field with "POINT".
2.7 Affiliation Type
The only valid code for Affiliation Type is "Report Certifier" in the NIF. This indicates that the
Contact Person identified in the Transmittal Table is EPA's point of contact to answer questions
about the data submitted. If Affiliation Type is blank or in error, we will populate this field with
"Report Certifier".
2.8 Format Version
The Format Version is the Version of NIF that is used to format and submit NIF files. For the
2002 NEI, the format version should be reported as "3.0". If the Format Version is blank or in
error, we will augment this field as "3.0".
2.9 Revision Date
We will add this field to all tables to indicate the date of the data for each record. For example for
the emissions table, the draft emissions estimate would contain a date of 20050201. If the
estimate is not revised, the date would remain 20050201. If the estimate was revised during the
external review period, the date would be revised to 20050501 in the final inventory. As
revisions are made in time, then the revision date will be revised to reflect date of revision.
CRIT:
HAP:
CRITHAP:
CAPs
HAP
CAPs and HAPs
February 01, 2006
17
-------
SECTION 3. GEOGRAPHIC INFORMATION
This section includes the following data fields.
State/County FIPS Code
Tribal Code
Urban/Rural Classification
MSA Name and Code
Country
We will prepare a State/County FIPS Code Table with the following data fields.
State/County FIPS Code
State FIPS
County FIPS
State/County FIPS Code Description
Start Date - date FIPS code became valid
End Date - date FIPS code became invalid for use
Valid for NIF 2.0
Valid for NIF 3.0
Comments - Notes about changes
• EPA Region
County Name
State Abbreviation
Urban/Rural Classification
2002 Population (number of people) - estimates from US Census Bureau
MSA Name - from U.S. Census Bureau
MSA Code - from U.S. Census Bureau
County Centroid Latitude
County Centroid Longitude
County Maximum Latitude
County Minimum Latitude
County Maximum Longitude
County Minimum Longitude
State Maximum Latitude
State Minimum Latitude
State Maximum Longitude
State Minimum Longitude
This table is found in the 2002 State Lookup File, 02nei_lkup_states.zip, found at:
ftp://ftp.epa.gov/EmisInventorv/2002finalnei/documentation/point/ .
3.1 State and County FIPS Codes
State and County Federal Information Processing Standards (FIPS) codes are used by a variety of
organizations to identify the state and county within databases. The State and County FIPS Code
February 01, 2006
18
-------
is a mandatory data field in the NEI. The State and County FIPS code represents the county or
county equivalent and the state or state equivalent of the United States.
For State and County FIPS codes, we will first compare the State and County FIPS Codes
reported by data submitters to 2002 State/County FIPS NIF Code Table and identify records with
errors in FIPS code. 2002 revisions to FIPS codes in the NEI include the following.
CO: Adding FIPS code for Broomfield County, CO (08014). Broomfield County
(08014) was created from parts of Adams (08001), Boulder (08013), Jefferson
(08059), and Weld (08123) counties.
VA: Removing independent city (county-equivalent) of Clifton Forge, VA (51560).
Clifton Forge reverted to town status and is now part of Alleghany County, VA
(51005) rather than a separate county-equivalent surrounded by Alleghany County.
For invalid FIPS codes in CO, we will contact the data providers and correct invalid FIPS codes
after consultation with the data provider. For invalid VA FIPS Code in Clifton Forge, we will
augment data by deleting the old FIPS code and replacing it with the FIPS code for Alleghany
County.
If a facility has records with valid FIPS codes, we will default records having missing or invalid
FIPS code with the FIPS code from the records within the facility having valid FIPS code. For
facilities that do not have any records with valid FIPS codes, we will first use the 2002 NEI
Facility Table to see if the facility is present in the table. If it is present, we will default the State
and County FIPS code to the FIPS Code in the 2002 NEI Facility Table.
For facilities missing State and County FIPS Codes that are not in the 2002 NEI Facility Table,
we will contact the data submitter to correct the FIPS code.
3.2 Tribal Code
The Tribal Code, a mandatory data field in NIF, represents American Indian Tribes and Alaskan
Native entities. The tribal code is used to identify facilities that are located on tribal lands.
For Tribal Codes, we will compare the Tribal Codes reported by data submitters to 2002 NIF
Tribal Code Table and identify records with invalid or missing Tribal Codes. For invalid or
missing Tribal Codes, we will first use the 2002 NEI Facility Table and see if the facility is
present in the file. If it is present, we will default the Tribal Code using the 2002 NEI Facility
Table. For facilities that are not in the 2002 NEI Facility Table, we will augment Tribal Codes
for point sources in the NEI by mapping site latitude and longitude of all facilities and identifying
facilities on tribal lands using Tribal GIS, a GIS software which contains boundaries for federally
recognized tribes. We will augment Tribal Codes for facilities on tribal lands in the NEI that do
not have a Tribal Code after tribes review the Tribal Code assignment during the review of the
draft NEI. We will specifically ask tribes to review the assignment of Tribal Codes to facilities
during the external review process.
February 01, 2006
19
-------
Because the Tribal Code is a primary key in the NEI, we will default non-tribal data to "000".
3.3 Urban/Rural Classification
The Integrated Urban Air Toxics Strategy is an important part of EPA's national air toxics
program. The NEI is used in analyses relating to Section 112(k) and Section 112(c)(3)
requirements in the CAA. Please note that the definition of "urban"does not necessarily apply for
regulatory or implementation purposes. For more information on EPA's Urban Air Toxics
Program, go to the following site, www.epa.gov/ttn/atw/urban/urbanpg.html .
We will classify all counties in the 2002 NEI as Urban-1, Urban-2, or Rural, based on 2002
Census data. The classifications are defined as:
Urban-1 (Ul): Counties included in an MSA with MSA population greater than 250,000
(for counties associated with more than one MSA in the six New England states, the
largest MSA population was used);
Urban-2 (U2): Counties in which the U.S. Census Bureau designates more than 50 percent
of the total county population as urban; and
Rural (R): Counties in which the U.S. Census Bureau designates less than 50 percent of
the total county population as urban. Counties in which the urban population exactly
equaled the rural population were classified as R.
The following population and geographic information available from the U.S. Census Bureau (via
their Internet web page) will be used in the U1/U2/R classifications for each of the counties
defined in the 2002 NEI
1. Total county, Minor Civil Division (MCD), and incorporated place population estimates
(U.S. Census Bureau).
2. Total Metropolitan Statistical Area (MSA) population, and county and place populations
associated with each MSA (U.S. Census Bureau).
Urban-1 (Ul) counties are included in an MSA with MSA population greater than 250,000.
The U.S. Census Bureau does not estimate county urban population between censuses (1990 and
2000). Therefore a surrogate for county urban population will be calculated by summing the
population from all incorporated places within a county with population of 2,500 or more (the
only part of the urban area definition for which 2002 estimates are available). If over 50% of the
county population live in incorporated places with population of 2,500 or more, the county will be
classified as Urban-2 (U2).
Rural (R) counties have a surrogate urban population that is less than 50 percent of the total
county population.
February 01, 2006
20
-------
Counties will be classified for 2002 as Ul, U2, or R using the same criteria as used for previous
classifications. For counties in the six New England states that are associated with more than one
MSA, the MSA with the largest population is assigned for use in the U1/U2/R classification
process. Table 5 shows the MSAs that were assigned in the 1999 NEI for these areas.
Table 5. MSAs Assigned to Counties in New England States Associated with More Than
One IV
SA for Use In The County U1/U2/R
Classification Process
I-IPS Sialic
Cock'
(Sialic
Al)hiv\ iiiiion)
1 IPS
(oil II l\
Cock'
Coning \aimc
FIPS
MSA
Cock'
(1 •>«>«>)
Mclmpolilain Slailislicail Arcai \aimc
09
(CT)
005
Litchfield County
5602
New York-Northern New Jersey-Long Island,
NY-NJ-CT-PA CMS A
007
Middlesex County
5602
New York-Northern New Jersey-Long Island,
NY-NJ-CT-PA CMS A
011
New London County
3280
Hartford, CT MSA
015
Windham County
1122
Boston-Worcester-Lawrence, MA-NH-ME-CT
CMSA
23
(ME)
031
York County
1122
Boston-Worcester-Lawrence, MA-NH-ME-CT
CMSA
25
(MA)
005
Bristol County
1122
Boston-Worcester-Lawrence, MA-NH-ME-CT
CMSA
013
Hampden County
1122
Boston-Worcester-Lawrence, MA-NH-ME-CT
CMSA
44
(RI)
009
Washington County
6480
Providence-Fall River-Warwick, RI-MA MSA
3.4 MSA Name and Code
We will identify the MSA of counties in the 2002 NEI. We will include the MSA Name and
MSA Code in the State/County FIPS Code table.
3.5 Country
Country is not a mandatory data field in NIF. If NEI data are provided to other countries, then it
is useful to populate this data field. The Country name is the name that represents a primary
geopolitical unit of the world. The default for the 2002 NEI is "USA".
February 01, 2006
21
-------
SECTION 4. FACILITY INFORMATION
This section includes the following data fields.
NEI Unique Facility ID
NEI Standardized Facility Name
NEI Standardized Address
NEI Standardized City
NEI Standardized State
NEI Standardized Zip Code
ORIS Facility Code
Location Address
City
State
Zip Code
Facility Registry ID
TRIID
Dun and Bradstreet Number
BART Source and Unit
• IPM Facility and Unit
The 2002 NEI will combine HAPs and CAPs into a single inventory. In order to compile a
merged file, we will prepare a 2002 NEI Facility Table. For each facility in the NEI, the table
will contain the following data fields.
State/County FIPS Code
Tribal Code
State FIPs Abbreviation - State abbreviation for State/County FIPS Code
County FIPS Name - County Name for State/County FIPS Code
• Tribal Name - Name of Tribe for Tribal Code
State Facility ID - Local identifier provided in original state, local agency, tribal, industry,
TRI, or MACT data submittal
Facility Name (associated with State Facility ID)
Location Address (associated with State Facility ID) - Physical location of the main
entrance of the site
City (associated with State Facility ID) - City of the physical location of the site
State (associated with State Facility ID) - State abbreviation of the physical location of
the site
Zip Code ( associated with State Facility ID) - U.S. Postal Service zip code of the physical
location of the site
NEI Unique Facility ID - Common identifier assigned to all sites determined to be the
same facility
NEI Standardized Facility Name - Preferred name when one or more names are available
for the same facility
NEI Standardized Address - Preferred zip code when one or more location addresses are
available for the same facility
February 01, 2006
22
-------
NEI Standardized City - Preferred city when one or more cities are available for the same
facility
NEI Standardized State - Preferred state abbreviation when one or more states are
available for the same facility
NEI Standardized Zip Code - Preferred zip code when one or more zip codes are available
for the same facility
NEI Site Latitude - Average of all latitudes for the site
NEI Site Longitude - Average of all longitudes for the site
NEI Primary SIC Code
• NEI Primary NAICS Code
ORIS Facility ID - assigned by EPA to electric generating units
FRS State/County FIPS Code - Facility Registry Identifier state and county FIPS assigned
to facility by OEI
FRS ID - Facility Registry Identifier assigned to facility by OEI
FRS Facility Name - Facility Registry Identifier facility name assigned to facility by OEI
FRS Address - Facility Registry Identifier location address assigned to facility by OEI
FRS City - Facility Registry Identifier city assigned to facility by OEI
FRS State - Facility Registry Identifier state abbreviation assigned to facility by OEI
FRS Zip Code - Facility Registry Identifier zip code assigned to facility by OEI
FRS Latitude - Facility Registry Identifier latitude assigned to facility by OEI
FRS Longitude - Facility Registry Identifier longitude assigned to facility by OEI
• TRI ID - assigned by EPA's Office of Environmental Information (OEI)
Dun and Bradstreet Number - Dun and Bradstreet number for the facility
Historical Names for Facility and Ownership
Facility Status - indicates if facility is closed or inactive
• 1999 CAP - denotes if the site is in the 1999 NEI for CAPs
• 1999 HAP -denotes if the site is in the 1999 NEI for HAPs
• 2002 Preliminary NEI - denotes if the site is in the 2002 Preliminary NEI for CAPs
• 2002 Draft NEI - denotes if the site was in the 2002 NEI 02012005 Draft
• 2002 NEI CAP 02012006 - denotes if the site is in the current version (Version 1.0,
02012006) of the 2002 NEI for CAPs
• 2002 NEI HAP 02012006 - denotes if the site is in the current version (Version 1.0,
02012006) of the 2002 NEI for HAPs
Comments - provides additional information about some facilities
The 2002 NEI Facility Table will evolve into a historical facility database where changes in
names/ownerships and subsidiaries will be tracked over time. The 2002 Facility Table is located
in a zipped file, nei2002_point_facility_ids_history_date, at:
ftp://ftp.epa.gov/EmisInventorv/2002finalnei/point sector data/ .
4.1 NEI Unique Facility ID
The NEI Unique Facility ID provides a single unique ID for each facility in the NEI. Data
providers sometimes use different site IDs for CAPs and HAPs emitted from the same facility.
February 01, 2006
23
-------
The NEI also merges data provided by different sources for pollutants emitted from individual
processes within a facility . The NEI Unique Facility ID is assigned to facility records submitted
from different data sources in order to detect duplicate estimates among the records. The NEI
Unique Facility ID is critical to blending HAP and CAP emissions from different data sources
and for augmenting VOC and PM emissions from HAP emission estimates and HAP emissions
from VOC and PM emissions.
We will first determine if a facility is in the 1999 NEI Facility Table. If it is in the 1999 NEI
Facility Table, then we will use the existing NEI Unique Facility ID. If the facility is not in the
table, then we will assign a NEI Unique Facility ID and add all of the information for the facility
to the 2002 NEI Facility Table. This information will be used to augment other data fields
associated with facilities.
We matched facilities in the 2002 Preliminary NEI for CAPs and in the 1999 NEI for HAPs and
prepared a crosswalk with between the two data sets to generate a preliminary 2002 NEI Facility
Table that was used as a starting point for the 2002 NEI. We used a combined automated and
manual approach for matching facilities in the 2002 Preliminary NEI for CAPs and the 1999 NEI
for HAPs. We will use this same approach for facilities that are submitted for inclusion in the
2002 NEI that are not in the 2002 Preliminary NEI Facility Table.
The process of matching facilities in the point source NEI consists of two activities:
matching facilities in the NEI and assigning a common NEI Unique Facility ID, and
matching NEI facilities to EPA's Facility Registry System (FRS).
Matching Facilities in the NEI
An automated facility-matching program is first run to identify the best matches of common
facilities using state/county FIPS code, facility name, facility address, and site latitude/longitude
coordinate pairs (a site average of individual emission release point coordinates). The automated
matching assigns a common ID to facilities if they matched on all of these parameters. The
program then looks for additional high probability matches by matching exactly on all parameters
except the site latitude/longitude coordinate pairs. In this case, the program searches first for
coordinate pairs located within 0.001 degrees of one another. If no match is found, the program
interactively relaxes this condition for additional matches until it reaches an upper bound of
coordinates that are 0.1 degrees apart. 0.1 degrees was found empirically to be the limit of 100%
confidence in automated matching for the 1999 NEI. The procedure starts with exact matches (0
degrees difference) and works up to 0.1 degrees in increments of 0.0001 degrees to insure that the
closest facility is always matched in the case where multiple plants may exist. However, some
facilities that might be co-located within 0.1 degrees are not automatically matched if there is
more than one potential match between facilities. The automated facility matches are compiled
into a facility file and removed from the facility universe.
For the remaining unmatched facilities, an algorithm is applied that strips out punctuation and
leading/trailing spaces, drops insignificant punctuation, and standardizes corporate names, from
the facility name. Additional searches using these standardized names are then conducted by:
February 01, 2006
24
-------
comparing facility names on a case sensitive basis, identifying similar sounding facility names in
each county, and matching on similar name and address and exact site latitude/longitude
coordinate pairs. After using exact coordinate pairs, the program identifies additional potential
matches using similar names, similar address, and similar coordinate pairs with latitude/longitude
varying from 0.001 to 0.1 degrees. The matching stops when any ambiguity is found. The
program generates candidate lists of potential matches for manual review. A number of these
candidate lists are generated by using an artificial intelligence technique called heuristic fuzzy
pattern matching to match facility names and addresses with small typographical differences.
Candidate lists of potential facility matches are then prepared and reviewed manually to
determine if a valid match exists. Examples of candidate lists include the following.
Same state/county FIPS code, facility name and address, but coordinates differ by > 0.1
degrees
Same state/county FIPS code, facility name and coordinates, but different or missing
address
Same state/county FIPS code, address and coordinates, but different facility name
Same state FIPS code, facility name, address and coordinates, but different county FIPS
code
First 5 letters of facility name are same with same state/county FIPS code, address and
coordinates
The same type of candidate lists are also generated for potential "fuzzy" facility name and address
matches. Candidate lists include SIC and NAICS codes.
These candidate lists are evaluated one at a time in order of highest to least certainty. To verify
matching of facilities in the candidate lists, we researched individual companies, their locations,
acquisition history and name changes on the internet. We consulted several external databases
including company websites, Mapquest, internet yellow pages, TRI explorer, FRS, and business
directories. These web searches help to identify closures and matches. The person reviewing the
candidate lists verifies that the list has been manually validated and then compiles matches from
each candidate facility list into the file with automated facility matches.
Assigning FRS IDS
EPA is required to include Facility Registry System (FRS) IDs in the NEI in order to meet the
new EPA data standards for Facility Identification. (For more information on EPA data
standards, go to: http://oaspub.epa.gov/edr/epastd$.startup#! ) The standards relevant to the NEI
are the: SIC/NAICS, Latitude/Longitude, Chemical Identification, Facility Identification and
Contact Information. To implement the Facility Identification standard, EIAG must map NEI
facilities to the FRS facilities maintained by EPA's Office of Environmental Information (OEI).
The assignment of FRS IDS to NEI facilities involves the following steps.
EIAG submits NEI facility list to OEI
OEI assigns draft FRS IDs to NEI facilities
• EIAG incorporates FRS IDs into NEI facility file
The process of assigning FRS IDs to the 2002 NEI will not occur until after external revisions are
received and the final list of facilities in the 2002 NEI is compiled.
February 01, 2006
25
-------
4.2 NEI Standardized Facility Name
The NEI Standardized Name provides a single public or commercial name for a facility. The
Facility Name provided by different data submitters for the same facility may differ in HAP and
CAP NIF Site Tables. This often occurs because of the use of subsidiary names, abbreviations, or
old names that do not reflect current name and ownership of facilities in inventories. We will use
the 2002 NEI Facility Table to augment this data field.
4.3 NEI Standardized Address
The NEI Standardized Address provides a single physical address for a NEI facility. The Location
Address provided by different data submitters for the same facility may differ in HAP and CAP
NIF Site Tables. This often occurs because a road may be identified using a street name and as a
highway number in different inventories. It also occurs when an address changes for a facility.
We will use the 2002 NEI Facility Table to augment this data field.
4.4 NEI Standardized City
The NEI Standardized City provides a single name of the city, town, or village or other locality in
the physical address for a facility. The City provided by different data submitters for the same
facility may differ in HAP and CAP NIF Site Tables. This sometime occurs if a new city/town is
incorporated or city limits are extended and a facility has new address and inventory files are not
updated. We will use the 2002 NEI Facility Table to augment this data field.
4.5 NEI Standardized State
The NEI Standardized State data field indicates the state associated with the NEI Unique Facility
ID. We will use the 2002 NEI Facility Table to augment this data field.
4.6 NEI Standardized Zip Code
The NEI Standardized Zip Code data field indicates the zip code associated with the NEI Unique
Facility ID. The Zip Code provided by different data submitters for the same facility may differ
in HAP and CAP NIF Site Tables. This sometime occurs when new zip codes are assigned by the
US Postal Service and the inventory files are not updated. We will use the 2002 NEI Facility
Table to augment this data field.
4.7 Location Address
The Location Address describes the physical (geographic) location of the front door or main
entrance of a facility site, including urban-style street address or rural address. It is needed to
default location coordinates for a facility if they are missing or invalid. If the Location Address
is missing and the facility is in the 2002 NEI Facility Table, we will use the NEI Standardized
Address to default the Location Address.
February 01, 2006
26
-------
4.8 City
The City is the name of the city, town, or village or other locality of the physical location of the
facility. The City is needed to verify the reported zip code. The zip code is needed to default
location coordinates for a facility if they are missing or invalid. If the City is missing and the
facility is in the 2002 NEI Facility Table, we will use the NEI Standardized City to default this
field.
4.9 State
The State is the state abbreviation of the physical location of the facility. The state abbreviation
can be used to QA the Zip Code, State and County FIPS Code, and Latitude and Longitude.
We will QA the State by comparing it to the State and County FIPS Code. If the State is missing
or invalid and the facility is in the 2002 NEI Facility Table, we will use the NEI Standardized
State to default this field. If the facility is not in the 2002 NEI Facility Table, we will use the
State and County FIPS Code to default the state abbreviation. If the State is incorrect, it is likely
that the Address, City and Zip Code are also incorrect and that a mailing address was used rather
than the physical location of the facility.
4.10 Zip Code
The Zip Code is needed to default geocoordinates for a facility if they are missing or invalid. The
Zip Code is the combination of the 5-digit Zone Improvement Plan (ZIP) code and the four-digit
extension code (if available) that represents the geographic segment that is a subunit of the Zip
Code, assigned by the U.S. Postal Service.
We will QA the quality of the Zip Code using a Zip Code QA file available at the following
address:
ftp://ftp.epa.gov/EmisInventorv/2002finalnei/documentation/augmentation point sector/.
The source of the Zip Code QA file is ESRI and the file uses zip code information from the US
Postal Service and FIPS code information from the US Department of Commerce, National
Institute of Information and Technology. If the reported State and County FIPS Code does not
match the FIPS Code in Zip Code QA file, then the reported Zip Code is assumed to be invalid.
If Zip Codes are incorrect or missing, we will use the NEI Standardized Zip Code in the 2002
NEI Facility Table to default the Zip Code. If global errors such as dropping leading zeroes for a
state or transposition errors exist, then we will correct the Zip Codes. We will also use the Zip
Code QA file to replace Zip Codes if the facility is not in the 2002 NEI Facility Table. If we are
unsure of how to correct errors (e.g., zip code does not match town or latitude/longitude), then we
will contact the data providers.
4.11 ORIS Facility Code
The ORIS Facility Code is a unique ID for electric generating units. The ORIS Facility Code is
February 01, 2006
27
-------
assigned by the Department of Energy (DOE)'s Energy Information Administration(EIA). We
will compare the reported ORIS Facility Code for a facility to the ORIS Facility Code in the 2002
NEI Facility Table. If the ORIS Facility Code is missing for electric generating units, then we
will use the 2002 Electric Generating Unit (EGU) ORIS Facility Code in the 2002 NEI Facility
Table to default the missing ORIS Facility Codes. If the comparison indicates that the ORIS
Facility Codes are different in the 2002 NEI Facility Table and the data submitted in NIF tables to
EPA, then we will use DOE information to correctly assign ORIS Facility Codes in the NIF Site
Table and to revise the ORIS Facility Codes in the 2002 NEI Facility Table.
4.12 Facility Registry ID
EPA's OEI assigns Facility Registry IDs to provide a single ID to identify facilities across EPA
databases. We will use the 2002 NEI Facility Table to augment this data field if data submitters
do not provide it.
4.13 TRIID
The Toxic Release Inventory (TRI) ID is assigned by EPA's OEI to facilities that report under
SARA 313 Right To Know Act. TRI is one of the sources of NEI data for point sources emitting
HAPs and ammonia. We will use the 2002 NEI Facility Table to augment this data field for TRI
facilities if data submitters do not provide it.
4.14 Dun and Bradstreet Number
The Dun and Bradstreet Number is a number assigned to facilities in the Dun and Bradstreet
database. The Data Universal Numbering System (DUNS) number is assigned by Dun and
Bradstreet to uniquely identify business establishments. The Dun and Bradstreet Number is
reported in the TRI. We will use the 2002 NEI Facility Table to augment this data field if the
data are not available.
4.15 BART Source and BART Unit
The identification of Best Available Retrofit Technology (BART) emission units in the NEI is
needed to support modeling for EPA's regional haze rule. The BART requirements apply to
units within facilities built between 1962 and 1977 in 26 source categories that have the potential
to emit 250 tons or more of visibility- impairing pollutants. (For more information on BART, go
to: http://www.epa.gov/air/visibilitv/factsheet.html .)
State and local agencies, tribes, and Regional Planning Organizations (RPOs) have begun to
identify BART units in their inventories. We will flag BART facilities and emission units within
facilities in the 2002 NEI from available information provided by state and local agencies, tribes
and RPOs.
February 01, 2006
28
-------
4.16 IPM Facility and IPM Unit
EPA uses the Integrated Planning Model (IPM) to analyze the projected impact of environmental
policies on the electric power sector. EPA uses the IPM to evaluate the cost and emissions
impacts of proposed policies to limit emissions of sulfur dioxide (S02), nitrogen oxides (NOX) ,
carbon dioxide (C02), and mercury (Hg) from the electric power sector. (For more information
on IPM, go to: http://www.epa.gov/airmarkets/epa-ipm/.)
The identification of IPM emission units in the NEI is needed to support IPM analyses. We will
flag IPM facilities and emissions units within facilities in the 2002 NEI using information
provided by EPA staff.
February 01, 2006
29
-------
SECTION 5. GEOCOORDINATES
This section includes the following data fields:
X coordinate (longitude)
Y coordinate (latitude)
• UTM zone
XY Coordinate Type
Location Default Methodology
Stack Latitude
Stack Longitude
Horizontal Collection Method Code
Horizontal Reference Datum Code
Coordinate Data Source Code
Reference Point Code
Horizontal Accuracy Measure
Source Map Scale Number
NEI Site Latitude
NEI Site Longitude
5.1 Geocoordinates QA and Default Procedure
The first step in blending and merging of point sources is to QA geocoordinates and correct
erroneous coordinates. Latitude and longitude are needed to correctly place facility emission
release points and associated emissions into specific geographic domains (grid cell, census tract,
etc.) for proper emissions modeling.
All UTM coordinates will first be converted to latitudes and longitudes based on 1984 World
Geodetic System datum (WGS84).
Longitude should be reported in units of decimal degrees with a negative sign and latitude should
be reported in units of decimal degrees with a positive sign. We will QA the reported latitude and
longitude to determine if units are reported in degrees-minutes-seconds. If the latitude and
longitude are reported in degrees-minutes-seconds, we will convert the latitude and longitude to
decimal degrees using the following equations.
Latitude (decimal degrees) = Latitude degrees + Latitude minute/60.0 + Latitude
second/3600.0
Longitude (decimal degrees) = Longitude degrees + Longitude minute/60.0 + Longitude
second/3600.0
If latitude and longitude are reversed, a common error in NEI submittals, we will transpose the
latitude and longitude. If the reported value does not have a decimal, we will use the minimum
and maximum latitude and longitude fields in the State/County FIPS Code Table to determine if
February 01, 2006
30
-------
placing a decimal in the reported values will result in a valid latitude and longitude. If we cannot
correct the reported geocoordinates, we will contact the data provider.
After converting UTMs to latitude and longitude and correcting latitudes and longitudes reported
in wrong units, we will use a routine to assess the validity of the latitude and longitude values, to
replace values if necessary, and to fill-in missing data points. The QA of location coordinates is
a multi-step process. The first step is to make sure that all of the emission release points within a
facility are within a reasonable distance of one another. The second step uses geographic
information system (GIS) overlays to evaluate each coordinate pair with respect to its county
boundary. The stages of the routine are described below.
Step 1. Find and replace the latitude/longitude of emission release points within a facility that are
located at distances greater than 3.0 km of other release points in the facility.
This step includes determining if an individual release point within a facility is within 3.0 km of
other emission release points, (i.e., to ensure no stacks within a single facility are located miles
apart).
If there is only one emission release point, the process is complete and we will proceed to Step 2.
If there is more than one emission release point, we will identify the emission release points with
the highest and the lowest latitudes and emission release points with the highest and lowest
longitudes at the site. We will calculate the distances between the highest and lowest latitudes
and between the highest and lowest longitudes. Please refer to the diagram below for more details
on how we will conduct this analysis.
A)
(minx, miny)
fmaxx, max y)
'
B) (x, max yX>—-t]
(minx, q/ 0(maxx.¥)
V £.<>
(x, min y)
This method may but does not necessarily measure the true distance between emission release
points. For example, if a facility has 2 emission release points, the true distance is measured by
examining minimum and maximum latitudes/longitudes at all emission release points described
by Example A in the diagram. If, for example, a facility has 4 emission release points as in
Example B of the diagram, the distance calculated does not represent the distance between any
two specific emission release points, but rather defines the maximum distance that could exist
between any two emission release points given all of the points in the set. This analysis identifies
February 01, 2006 31
-------
sites whose emission release points are potentially far apart and whose coordinates need
correction, but it is not designed to identify the specific outlying emission release point(s).
A. If the greatest distance between latitudes and between longitudes is less than 3.0 km, the
process is complete and we will proceed to Step 2.
B. If the distances between the highest and lowest latitudes or the highest and lowest
longitudes are greater than 3.0 km, the latitudes and longitudes of all emission release
points within a facility are evaluated. If there are records whose latitudes and longitudes
are found to be at a distance of more than 3.0 km, the SIC or NAICS codes are examined
to determine if the distance between emission release points is feasible.
~ For a source category, if it is acceptable that the distance between emission release
points is greater than 3.0 km, the process is complete and we will proceed to Step
2. For a list of SIC Codes and NAICS Codes that the distance between emission
release points may be greater than 3.0 km, download the table "SIC NAICS 3km
ERP Acceptable" from:
ftp://ftp.epa.gov/EmisInventorv/2002finalnei/documentation/augmentation point
sector/ .
~ For a source category, if it is determined that the distance between emission
release points should not be more than 3.0 km, then the records are returned to the
data provider for assistance in identifying the source of error. We will then
proceed to Step 2.
Step 2. Find and replace latitude/longitude of emission release points that are out-of-countv
boundary or missing.
This step includes the use of a GIS overlay to plot each latitude/longitude value and compare it to
the physical boundaries of the State/County FIPS Code to which the value is associated (i.e., to
ensure no stacks are located in the oceans or in far away states). Detailed county boundaries
using a scale of 1 to 100,000 or better are used in the GIS overlay.
If the plotted release point is within 10 km of the county, the point is assumed to be valid and
neither latitude/longitude nor State/County FIPS Code are corrected. The process is complete.
If the plotted release point is found to exist more than 10 km outside of the county or if the
latitude/longitude is missing, then the latitude/longitude of each erroneous emission release point
is replaced using the following hierarchy. Only one method is used for missing or
out-of-boundary latitudes/longitudes within a facility, i.e., method A is not used for one emission
release point at a facility and method D for another emission release point at the same facility.
A. Use Geocoding software - If the reported emission release point latitudes/ longitudes
failed any of the above steps and revised information from the data source is not provided,
February 01, 2006
32
-------
geocoding software is used. More information on Tele Atlas North America "EZ- Locate"
geocoding software is located at: www.geocode.com/.
The first step in using Geocoding software is to check the quality of the zip code provided
in the inventory with a Zip Code QA file. The Zip Code QA file can be found at the
following address:
ftp://ftp.epa.gov/EmisInventorv/2002finalnei/documentation/augmentation point sector/
. The source of the Zip Code QA file is ESRI and the file uses zip code information from
the US Postal Service and FIPS code information from the US Department of Commerce,
National Institute of Information and Technology.
Zip Codes and State and County FIPS Codes reported in the Site records will be compared
to Zip Codes and FIPS codes in the Zip Code QA file. If the reported State and County
FIPS Code does not match the FIPS Code in Zip Code QA file, then we will not use the
reported Zip Code in the Geocoding process. If global errors such as dropping leading
zeroes for a state or transposition errors exist, then we will correct the Zip Codes to the
extent possible. If we are unsure of how to correct errors (e.g., zip code does not match
town or latitude/longitude), then we will contact the data providers.
After correcting Zip Codes where possible, we will compile a list of sites that have
emission release points with missing or erroneous latitude/longitude. The file contains
site name, the physical address of the sites, and state and county FIPS code. The file is
submitted to the geocoding software. Using the inventory record's street address, the
geocoding software provides matches using the following hierarchy:
1. First the software standardizes the street address and looks for an exact address
match. The latitude/longitude are located as an exact match.
2. If an exact match cannot be found, the software then tries to match at the single
street block, as defined by Geocoder's documentation. The latitude/longitude are
located at the centroid of the single street block.
3. If a single street block match cannot be found, then the software tries to match on
the 5-digit zip code plus 2 digits. The latitude/longitude are located at the centroid
of the 5 digit zip code plus 2 digits.
4. If a match to 5-digit zip code plus 2 digits cannot be found, then the software tries
to match to the 5-digit zip code alone. The latitude/longitude are located at the
centroid of the 5-digit zip code.
5. If a match to 5-digit zip code cannot be found, then the software tries to match to
3-digit zip code. The latitude/longitude are located at the centroid of the 3-digit zip
code.
February 01, 2006
33
-------
6. If a match to a 3-digit zip code cannot be found, then the software provides an
"ambiguous" match which is a match to multiple non-standardized street segments.
If the geocoding software finds valid latitude/longitude data that match the recorded FIPS,
the process is considered to be complete and the geocoded latitude/longitude pairs are
assumed to be valid. If the latitude/longitude data do not match the recorded FIPS, the
geocoded coordinates are not used. The State and County FIPS code will be changed only
after we contact the data provider and verify the coordinates and FIPS code. If geocoded
latitude/longitude are used, the process is considered complete.
B. Use 2002 NEIFacility Table - If Geocoder cannot find a valid latitude/longitude for the
facility, then the 2002 NEI Facility Table will be used. If the identified state and county
in 2002 NEI Facility Table for a facility does not match the reported state and county in
the inventory, the 2002 NEI Facility Table data will not be used. If the 2002 NEI Facility
Table contains valid latitude/longitude data with corresponding valid county, the process
is considered complete and all the latitude/longitude data are assumed to be valid.
C. Assign Site Release Point at County Centroid - If, after each of these stages, an emissions
release point latitude/longitude data set is still found to be missing or invalid, the latitude
and longitude of the county centroid will be used to default the coordinates. County
centroids are found in the State/County FIPS Code Table.
5.2 X Coordinate (longitude)
The longitude should be reported in decimal degrees with a negative sign. We will use the
Geocoordinate QA and Default Procedure to QA and default longitudes.
5.3 Y coordinate (latitude)
The latitude should be reported in decimal degrees with a positive sign. We will use the
Geocoordinate QA and Default Procedure to QA and default latitudes.
5.4 UTM Zone
The UTM Zone number will be verified and corrected where needed based on the State and
County FIPS code.
5.5 XY Coordinate Type
We will verify the XY Coordinate Type and correct erroneous types. If latitude and longitude are
reported, the XY Coordinate Type is LATLON. If UTM coordinates are reported, the XY
Coordinate Type is UTM. After we convert all coordinates to latitude and longitude, we will
populate this field with coordinate type of LATLON in the draft and final 2002 NEI.
February 01, 2006
34
-------
5.6 Location Default Methodology
All defaulted latitudes and longitudes will be identified in the Emissions Release Point records in
the NEI. We will use the coordinate default flags shown in Table 6.
Table 6. Location Default Flags
Code
Code Description
Exact
Match is to within a unique intersection or within a single side of a single
street block.
Near
Match is to a single street block but the correct placement within block is
unknown.
Zipcode+2
Match is to a 5-digit zip code, plus the first two digits of the 4-digit
extension.
Zipcode3
Match is to multiple 3-digit zip codes based on postal service Sectional
Center Facility (SCF).
Zipcode5
Match is to a 5-digit zip code.
CNTYCENT
County centroid is used.
NEIHFT
Coordinate is found in the 2002 NEI Facility Table.
SITEAVG
A Site Average of accurate coordinates of other emission release points at the
same site is used.
StateRev
The coordinates were revised by state or local agency or tribe.
5.7 Latitude/Longitude EPA Data Standard Elements
The EPA's Latitude/Longitude Data Standard consists of the group of data elements used for
recording horizontal and vertical coordinates and associated metadata that define a point on earth.
Table 7 summarizes these elements. This standard will help users gauge the accuracy and
reliability of a given set of coordinates. (For more information on EPA data standards, go to:
http://oaspub.epa.gov/edr/epastdS.startup# 1 ) The primary responsibility for populating these
fields lies with the data submitters, as it is difficult if not impossible to discern the origin of a
latitude/longitude without being the primary author of the data.
Stack Latitude
We will report stack latitude in decimal degrees with a positive sign after QA and defaulting of
geocoordinates have occurred.
Stack Longitude
We will report stack longitude in decimal degrees with a negative sign after QA and defaulting of
geocoordinates have occurred.
February 01, 2006
35
-------
Table 7. EPA Latitude/Longitude Data Standards
Latitude/Longitude
Standard
Description
Comments
Latitude Measure
Y Coordinate - The measure of the
angular distance on a meridian north
or south of the equator.
+78.123456
The number of decimal positions
recorded is determined by the
precision of the measurement.
Longitude Measure
X Coordinate - The measure of the
angular distance on a meridian east
or west of the prime meridian.
-123.234561
The number of decimal positions
recorded is determined by the
precision of the measurement
Source Map Scale
Number
The number that represents the
proportional distance on the ground
for one unit of measure on the map
or photo.
Only used when a map has been
used to determine
latitude/longitude, e.g., 125,000.
Horizontal Collection
Method Code
Method used to determine the
latitude and longitude coordinates
for a point on the earth.
e.g., 001 = address-matching house
number, 018 on interpolation map,
028 = Global Positioning Method,
with unspecified parameters.
Horizontal Accuracy
Measure
The measure of the accuracy (in
meters) of the latitude and longitude
coordinates.
Indicates the maximum distance on
the ground from coordinates to true
physical location, e.g., 3.
Horizontal Reference
Datum Code
The code that represents the
reference datum used in determining
latitude and longitude coordinates.
001= North American Datum of
1927
002 = North American Datum of
1983
003 = World Geodetic System of
1984
Reference Point Code
The code that represents the place
for which geographic coordinates
were established.
e.g., 101= Entrance point of a
facility or station; 105 = Point
where substance is processed,
treated, settled, or stored; 106 =
Point where a substance is released.
Coordinate Data Source
Code
The code that represents the party
responsible for providing the
latitude and longitude coordinates
e.g., EPA Headquarters, a state
agency, tribal organization, EPA
regional office etc.
Horizontal Collection Method Code
If the Horizontal Collection Method code is missing, we will populate this field code as "027"
(i.e., the information is not known) when latitude and longitude coordinates that are not defaulted
are retained in the 2002 NEI. When we default latitude and longitude using Geocoding software,
February 01, 2006
36
-------
we will be able to populate these fields based on documentation from TeleAtlas Geocoding EZ
Locator Service (www.geocode.com/ ). Table 8 presents default flags from Geocoder software
and default values for the latitude and longitude data standards.
Horizontal Reference Datum Code
We will only populate the Horizontal Reference Datum Code with codes in Table 8 when
latitude/longitudes are obtained from the TeleAtlas Geocoding EZ Locator Service
(www, geocode. com/Y If the Horizontal Reference Datum Code is not reported by data
submitters and we do not default latitude and longitude using Geocoder software, we will not be
able to populate this field unless EPA's OEI provides us a methodology to populate this data
field.
Coordinate Data Source Code
If the Coordinate Data Source Code is missing, we will populate this data field based on the data
source using the codes in the NIF code tables.
Reference Point Code
If the Reference Point Code is missing, we will populate this field code as "108" (Points not
represented by 101-107) when latitude and longitude coordinates that are not defaulted are
retained in the 2002 NEI. We will populate the Reference Point Code with codes in Table 8 when
latitude/longitudes are obtained from the TeleAtlas Geocoding EZ Locator Service
(www, geocode. com/Y
Horizontal Accuracy Measure
We will only populate the Horizontal Accuracy Measure with values in Table 8 when
latitude/longitudes are obtained from the TeleAtlas Geocoding EZ Locator Service
(www, geocode. com/Y If the Horizontal Accuracy Measure is not reported by data submitters
and we do not default latitude and longitude, we will not be able to populate this field unless
EPA's OEI provides us a methodology to populate this data field.
Source Map Scale Number
This data field is only applicable when a map has been used to determine latitude and longitude.
We will only populate the Source Map Scale Number with codes in Table 8 when
latitude/longitudes are obtained from the TeleAtlas Geocoding EZ Locator Service
(www, geocode. com/Y If the Source Map Scale Number is not reported by data submitters and
we do not default latitude and longitude, we will not be able to populate this field unless EPA's
OEI provides us a methodology to populate this data field.
February 01, 2006
37
-------
Table 8. Geocoder Default Flags and Default Values for Latitude/Longitude Data Standard Elements
Code
Description
Source
Map
Scale
Horizontal Collection Method Code
& Description
Horizontal
Reference
Datum
Horizontal
Accuracy
(meters)
Coordinate
Data Sou rce
Code
Exact
Match is to within a unique
intersection or within a single side of a
single street block.
24000
002 - Determination method based on
address matching-block face.
001 - North
American Datum
of 1927
12
080 or 084*
Near
Match is to a single street block but
the correct placement within block is
unknown.
24000
003 - Determination method based on
address matching-street centerline.
001 - North
American Datum
of 1927
50
080 or 084*
Zipcode+2
Match to a 5-digit zip code, plus the
first two digits of the 4-digit
extension.
24000
038 - Determination method based the
center of an area defined by the 5-digit
ZIP code and its 2-digit geographic
segment extension.
001 - North
American Datum
of 1927
100
080 or 084*
Zipcode3
Match is to a 3-digit zip code.
24000
021 - Determination method based on
interpolation-other.
001 - North
American Datum
of 1927
1000
080 or 084*
Zipcode5
Match to a 5-digit zip code.
24000
026 - Determination method based on
zipcode-centroid.
001 - North
American Datum
of 1927
10000
080 or 084*
CNTYCENT
County centroid.
24000
021 - Determination method based on
interpolation-other; 030 - based on a
digital map source (TIGER).
001 - North
American Datum
of 1927
N/A
080 or 084*
SITEAVG
Site average.
021 - Determination method based on
interpolation-other.
001 - North
American Datum
of 1927
N/A
084
* Coordinates are derived from USPS, Census Bureau Tiger server, or Eagle's TeleAtlas. These correspond to codes 080 (government agency) and 084
(contracting organization). The Reference Point Code for all defaults is 108 (points not represented by codes 101-107).
February 01, 2006
38
-------
5.8 NEI Site Latitude
An average NEI Site Latitude will be estimated using the latitudes of the group of valid non-
defaulted latitudes of emission release points within a facility. If all coordinates are defaulted,
then the defaulted site latitude will be used.
5.9 NEI Site Longitude
An average NEI Site Longitude will be estimated using the longitudes of the group of valid non-
defaulted longitudes of emission release points within a facility. If all coordinates are defaulted,
then the defaulted site longitude will be used.
5.10 Latitude/Longitude QC Report
We will prepare a QC report that contains a list of emission release point records that failed our
QA of latitude and longitude and provides our defaulted coordinates. We will send these reports
to the data providers and will ask data providers to review the defaults and provide corrected
addresses or latitudes/longitudes. Replacing coordinates or correcting bad address information
could improve the quality of the data needed for modeling..
The QC report will contain the following two files.
LATLON Report will contain 3 sections. Section I will provide a count of the affected
records by error type. Section II will summarize the corrections made by default method.
Section III will provide definitions of some of the errors and default types.
• LATLON QA database will contain 3 tables. The table, "LatLonRevisions", will contain
all coordinates that failed one or more of our tests and the default applied to correct the
error. Field explanations for "LatLonRevisions" as well as full default definitions will be
found in tables "tblFields" and "DefaultCodes," respectively.
Data providers will be able review the defaulted latitude/longitude and provide corrected
addresses and/or coordinate pairs if desired. Revisions will be provided by data providers by
filling in the fields starting with "REVISED" at the end of the LatLonRevisions table. We will
ask data providers to focus their review on defaulted coordinates labeled as "CNTYCENT" as
these are the least accurate coordinate type. We will also ask data submitters to look more closely
at those sites with emissions release points that were greater than 3km apart (i.e., those with
"Fail" in the "ERPComment" field) as we will not be able to identify which of the points are
outliers.
If data providers do not provide changes, we will use the defaulted values in the 2002 NEI.
The LatLon Revisions Table will contain the following fields of data and revision columns for
data providers to complete.
State Name
State FIPS
February 01, 2006
39
-------
State County FIPS
State Facility ID
Emission Release Point ID
Original X Coordinate
Original Y coordinate
Facility Name
Street Address
City
State
Zip Code
NullZero - identifies records submitted with null coordinates, 0 coordinates or UTM
coordinates without a UTM zone
Converted - indicates if the X and Y Coordinates were originally provided in UTM or as
DDMMSS and were converted to decimal degrees
ERP Comment - result of emission release point test to indicate whether ERPs are within
3 km of each other. Results are pass/fail; although some SIC/NAICS codes indicate
facilities that could legitimate, and these are marked "Pass due to SIC/NAICS".
GISCOMatch - QA Test: Indicates if coordinates when plotted in a GIS fall into the
county associated with the emissions release point. Results are pass/fail.
Within 10 km - QA Test: Indicates where coordinates are when plotted in a GIS with
respect to the county associated with the emissions release point. Value of 0 indicates
that the point falls into its assigned county.
GIStestFIN - QA Test Results: Indicates whether coordinates ultimately pass/fail GIStest
given "GISCoMatch" and "WithinlOkm."
ZipCode Check - QA Test: Indicates if the zip code associated with the site matches the
county associated with the site. Also indicates any corrections made to zipcode based on
NEI Historical Facility Table.
Bad Address - Identifies records with addresses that are not physical addresses or
addresses that can be used for geocoding.
GISFIPS - The county FIPS associated with X and Y coordinates when plotted.
Dist2County - Distance from plotted XY coordinates to the county within which it is
reported to lie.
Default - Code indicating how the X and Y coordinates were defaulted.
New X Coordinate - Defaulted X coordinate. We will use defaulted coordinate unless a
revision is provided.
New Y Coordinate - Defaulted Y coordinate. We will use defaulted coordinate unless a
revision is provided.
XY Type - Defaulted coordinate type. We will use defaulted coordinate type unless a
revision is provided.
Latitude/Longitude EPA Data Standard Elements - Includes Horizontal Collection
Method, Horizontal Accuracy Measure, Horizontal Reference Datum, Reference Point
Code, Source Map Scale, and Coordinate Data Source Code
Original Data Provider - person who provided original coordinates
February 01, 2006
40
-------
Revised Data Fields: Reviewers may provide corrections for EPA to evaluate in these
data fields.
~ Data provider of revised coordinates - required data field for all revisions
~ Revised Street Address
~ Revised City
~ Revised State
~ Revised Zip Code
~ Revised X Coordinate
~ Revised Y Coordinate
~ Revised UTM Zone
~ Revised XY Type
~ Revised Latitude/Longitude EPA Data Standard Elements - Horizontal Collection
Method, Horizontal Accuracy Measure, Horizontal Reference Datum, Reference
Point Code, Source Map Scale, and Coordinate Data Source Code
February 01, 2006
41
-------
SECTION 6. UNIT AND PROCESS IDS
This section includes the following data fields.
• HAP Emission Data Level
• Emission Unit ID
Process ID
• Emission Release Point ID
Due to the CERR not including HAPs, some of the key fields that are mandatory for CAPs are not
mandatory for state and local agencies and tribes reporting HAPs. For the 2002 NEI, process-
level data are required for CAPs, but not for HAPs. While data are also preferred at the process-
level for HAPs, if process-level data are unavailable, then an agency or tribe may report HAP data
at the site, unit, or emission release point level, and we will process the data for incorporation in
the 2002 NEI. The possible reporting options for HAPs and defaults are shown in Table 9.
Table 9. HAP Emission Data Levels
Emission Data
Level
Mandatory Reported Fields
EIAG Defaulted Fields
Site
Site ID, Emission Release
Point ID
Emission Unit ID, Process ID
Unit
Site ID, Emission Unit ID,,
Emission Release Point ID
Process ID
Stack
Site ID, Emission Unit ID,
Emission Release Point ID
Process ID
Process
Site ID, Emission Unit ID,
Process ID, Emission
Release Point ID
6.1 Emission Data Level
If the Emission Data Level is missing, we will default the field based on the level that emissions
are reported.
6.2 Emission Unit ID
If the HAP Emission Data Level is reported as Site and Emission Unit IDs are not reported, we
will assign Emissions Unit IDs to all of the records in the EU, EP, PE, CE and EM NIF tables.
February 01, 2006
42
-------
6.3 Process ID
If HAP Emission Data Level is reported as Site, Unit, or Stack and Process IDs are not reported,
we will assign Process IDs to all records in the EP, PE, CE, and EM NIF tables.
6.4 Emission Release Point ID
If the Emission Release Point IDs are not reported, we will assign Emission Release Point IDs to
records after we contact data providers.
February 01, 2006
43
-------
SECTION 7. SOURCE DESCRIPTION
This section includes the following data fields.
NAICS Code
SIC Code
NAICS/SIC Code Default Flag
SCC
Process MACT Code
Process MACT Compliance Code
MACT Code Assignment
Facility Category
7.1 NAICS Code
Standard Industrial Classification (SIC) codes are being replaced by the North American Industry
Classification (NAICS) codes that were adopted by Canada, Mexico, and the United States in
1997. The NAICS is a classification of business establishments by economic activity. It
supercedes the SIC code. The NAICS code consists of 6 digits which are arranged hierarchically.
• Two digits - Economic sector (NAICS Sector Code)
Three digits - Economic sub sector (NAICS Sub sector Code)
Four digits - A group of related industries within the economy (NAICS Industry
Code)
Five digits - An industry within the economy (NAICS Industry Code)
Six digits - A subdivision of an industry (NAICS Code)
NAICS codes may be reported in the NEI as 2, 3, 4, 5, or 6 digits. NAICS Codes are reported in
the NEI at the site level and at the emission unit level. The Primary NAICS Code is reported in
the SI table for site level and the NAICS Code Unit Level is reported in the EU table for emission
unit level. The Primary NAICS Code is mandatory in the NIF, and the NAICS Code Unit Level
is not mandatory.
To meet the EPA's NAICS Data Standard, we adapted the Census Bureau's 1987 SIC to 2002
NAICS crosswalk (see www.census.gov/epcd/naics02A and applied it to the 1999 NEI for HAPs.
NAICS codes supplied by the data submitters were not overwritten in the 1999 NEI. We prepared
a NAICS/SIC crosswalk for the NEI by modifying the Census Bureau crosswalk to accommodate
those situations in which one SIC code maps to multiple NAICS codes. Where all the NAICS
codes associated with one SIC code shared the first 5 digits, the SIC code was mapped to this 5
digit NAICS code. If no common 5 digit NAICS code existed, we applied the common 4 digit
NAICS code, and so on. In those cases where all of the NAICS codes associated with an SIC code
did not share the same 5, 4, 3 or 2 digit NAICS code, then the most common 5, 3, 4, 2 digit
NAICS code was selected. The SIC/NAICS crosswalk is contained in the table,
NAICS02toSIC87_NEI99, is found in the 2002 State Lookup File, 02nei_lkup_states.zip, at:
ftp://ftp.epa.gov/EmisInventorv/2002finalnei/documentation/point/ .
February 01, 2006
44
-------
We will first QA the reported NAICS codes by using the Census Bureau's list of NAICS codes
(see www.census.gov/epcd/naics02A A common error that we found in the 99 NEI was the
reporting of the SIC Code as the NAICS Code. We will compare the SIC and NAICS codes and
identify records where the SIC and NAICS codes are the same and determine if the SIC Code is
valid.
We will default all Primary NAICS codes in the Site Table. We will use the following hierarchy
to default the Primary NAICS Code.
1. If valid SIC codes are reported, we will use the SIC/NAICS crosswalk to augment the
NAICS Codes.
2. For facilities that have invalid or missing SIC codes or SIC codes that are not in the
SIC/NAICS crosswalk , we will use facility-specific NAICS codes from third-party
services such as PowerBusiness by InfoUsa. The PowerBusiness database is available on
a subscription basis and contains census-derived data for over 14 million U.S. businesses
and is searchable by name, address, business type, SIC code, and NAICS code.
3. For facilities that have invalid or missing SIC codes and are not in the Power Business
database but are in the 2002 NEI Facility Table, we will use the 2002 NEI Facility Table
to augment the Primary NAICS Code.
We will only default invalid NAICS Code Unit Level if valid SIC Code Unit Level is reported
and a NAICS Code match exists for the SIC Code in the SIC/NAICS crosswalk. We will not
default missing NAICS Code Unit Level.
We will assign NAICS/SIC default flags shown in Table 10 to indicate how the NAICS codes are
defaulted. Default flags are included for NAICS codes in the Site and Emission Unit Tables.
7.2 SIC Code
SIC codes may be reported in the NEI as 2, 3, or 4 digits.
• Two digits - Major Group Number
• Three digits - Industry Group Number
• Four digits - Industry Number
SIC Codes are reported at the site level and at the emission unit level. The Primary SIC Code is
reported in the SI table for site level and the SIC Code Unit Level is reported in the EU table for
emission unit level. SIC codes are not mandatory in the 2002 NEI.
We will first QA the reported SIC codes by using the Census Bureau's list of SIC codes (see
www.census.gov/epcd/naics02A) A common error found in the 1999 NEI was the use of outdated
SIC codes. If SIC codes are invalid, we will first use the 1987 SIC manual to revise old SIC
codes. We will default Primary SIC Codes if possible.
We will use the following hierarchy to default the Primary SIC Code.
1. For facilities included in the 2002 NEI Facility Table, we will use the 2002 NEI Facility
Table to augment the Primary SIC Code.
February 01, 2006
45
-------
2. For new facilities not included in the 2002 NEI Facility Table, we will use facility-specific
SIC codes from third-party services such as PowerBusiness by InfoUsa. The
PowerBusiness database is available on a subscription basis and contains census-derived
data for over 14 million U.S. businesses and is searchable by name, address, business type
and SIC and NAICS codes.
3. For new facilities not in the 2002 NEI Facility Table and not in the Power Business
database, if valid NAICS codes are reported and there is a one-to-one match between
NAICS and SIC codes, we will use the SIC/NAICS crosswalk, NAICS02toSIC87_NEI99,
in the 2002 State File, 02nei_lkup_states.zip, is found in the 2002 State Lookup File,
02nei_lkup_states.zip, at:
ftp://ftp.epa.gov/EmisInventorv/2002finalnei/documentation/point/ .
to augment the SIC Codes.
For the SIC Code Unit Level, we will only default missing SIC Codes if valid NAICS codes are
reported and there is a one-to-one match between NAICS and SIC codes by using the SIC/NAICS
crosswalk.
We will assign the SIC Code default flags shown in Table 10 to indicate how the SIC codes are
defaulted. Only codes 01 and 10 are valid as SIC code default flags. Default flags are included
for SIC codes in the Site and Emission Unit Tables.
7.3 NAICS/SIC Code Default Methodology
Table 10 presents the NAICS and SIC Code Default Flags that will be used in the Site and
Emission Unit Tables.
TABLE
L0. NAICS/SIC Code Default Flags
Cock'
Miilcli
Tj po
( ode Description
01
one to one
One SIC maps to only one NAICS code.
02
one to
many
One SIC maps to many NAICS code all of which share the first 5-digits
03
one to
many
One SIC maps to many NAICS code. Have chosen the most common 5
digit NAICS among these.
04
one to
many
One SIC maps to many NAICS code all of which share the first 4-digit.
05
one to
many
One SIC maps to many NAICS code. Have chosen the most common 4
digit NAICS among these.
06
one to
many
One SIC maps to many NAICS code all of which share the first 3-digits.
February 01, 2006
46
-------
Table 10 continued. NAICS/SIC Code Default Flags
Code
Match
Type
Code Description
07
one to
many
One SIC maps to many NAICS code. Have chosen the most common 3
digit NAICS among these.
08
one to
many
One SIC maps to many NAICS code all of which share the first 2-digits.
09
one to
many
One SIC maps to many NAICS code. Have chosen the most common 2
digit NAICS among these.
10
exact
1999 NEI facility file
7.4 SCC
Source Classification Codes (SCCs) are reported at the process level. SCCs are mandatory for
CAPs and must be reported at the 8 digit level for CAPs. SCCs are not mandatory for HAPs and
may be reported in the NEI as 3, 6, or 8 digits.
We will QA the reported SCCs by using the SCC Code Lookup Table posted at:
http://www.epa.gov/ttn/chief/code/index.html . For missing or invalid SCCs, we will return
records to the data provider for assistance in revising SCCs during the review of the draft 2002
NEI. If data providers do not revise invalid SCCs, we will use the Emission Process Description
to default the SCCs. If we revise the SCCs, we will indicate that SCCs have been revised.
7.5 Process MACT and Section 112(k) Area Source Standards Code
The MACT and Section 112(k) Area Source Standards Codes identify processes that are or will
be subject to MACT standards and new Section 112 (k) area source standards. MACT and
Section 112(k) Area Source Standards codes (hereafter, referred to as MACT codes), flags and
compliance codes are applied to processes in the NEI. For a list of MACT standards,
please see www.epa. gov/ttn/atw/eparules.html. For a list of area source categories, please see
www.epa.gov/ttn/atw/urban/urbanpg.html .
We will prepare a MACT Code table with the following fields:
MACT Source Category
MACT Code
proposed new code (Yes or No)
112k area source category (Yes or No)
Group (EPA group responsible for category)
• 10 year MACT (Yes or No)
Residual Risk (Yes or No)
Contact (EPA staff lead)
February 01, 2006
47
-------
Contact Email
Promulgation Date
Implementation Date
Major Source MACT Compliance Code
Area Source MACT Compliance Code
• Comment
The MACT Code Table will be included in the 2002 State Lookup File, 02nei_lkup_states.zip,
found at: ftp://ftp.epa.gov/EmisInventorv/2002finalnei/documentation/point/ .
MACT codes will be used in the blending/merging step to identify potential duplicate processes
for a facility where data are provided by multiple data sources. MACT Codes are assigned at the
process level in the NEI.
The first preference in assigning MACT codes will be given to the code assigned by the original
data provider. Thus, when the state/local agency or tribe supplies a code, we will retain the
MACT code and identify the data source as "STATE- BASED". When we receive data from an
EPA Emissions Standard Division (ESD) lead, we will assign the appropriate MACT Code and
identify the data source "ESD-BASED". We will identify EGU data generated by EPA as "EIG-
BASED". Please note that EPA's Emissions Standards Division is now known as the Sector
Policies and Programs Division.
If state and local agencies and tribes do not provide MACT codes in their 2002 submittals, we
will default MACT codes by using Facility lists, or applying SCC, SIC code, and NAICS code
defaults.
MACT codes will be assigned to processes based on the following hierarchy.
Emissions data provided by EPA
Emissions data provided by State or local agency or tribe
EGU data provided by EPA
Facility list provided by EPA's ESD
SCC/MACT Code dictionary
SIC/MACT Code dictionary
NAICS/MACT Code dictionary
The complete steps for assigning MACT Codes are described below.
1. If ESD-BASED file, apply MACT Code supplied by ESD and flag as ESD-
BASED.
2. If State-BASED file, retain MACT Code supplied by the state or local agency or
tribe and flag as STATE-BASED.
3. If EGU-based file, retain MACT or Area Source Code supplied by the EIAG and
flag as EIG-BASED.
February 01, 2006
48
-------
4. Apply Facility list codes to all processes at the facility. Do not overwrite ESD-,
STATE-, or EIG-BASED codes. The facility lists were supplied by ESD and
ESD's contractor, RTI. In some cases, the list was derived from ESD-BASED
MACT data in the 1999 NEI. Codes assigned by these lists are flagged as
"FACILLIST*." Table 11 describes the specific source of each facility list and its
corresponding flag.
Flags beginning with "FACILLIST-MLT" indicate that the facility appeared on
one or more (i.e., multiple) category lists. The look-up table, "Multiple MACT
Codes", includes a list of all MACT Codes assigned to a Multi-MACT Category.
The table, "Multipl-MACT Source of Facility Lists", provides a key for the MACT
flags beginning with "FACILLIST-MLT" The two tables will be included in the
2002 State Lookup File, 02nei_lkup_states.zip, found at:
ftp://ftp.epa.gov/EmisInventorv/2002finalnei/documentation/point/ .
TABLE 11. MACT Code Facility List Flags
Facility List Flag
Description
FACILLIST-99NEI
Facility list derived from 1999 NEI
FACILLIST-ESD02
Facility supplied by ESD in 2002
F ACILLIS T-RTI
Facility list supplied by RTI
FACILLIST-MLTx
Facility appears on multiple lists
5. Apply SCC-Default list to assign MACT and Area Source Flags to non-coded
processes. If SCC-default list assigns a different MACT or Area source category
than a Facility list, then let that code overwrite the Facility list supplied code. Use
"SCC-DEFAULT" as the flag.
6. Apply SIC-Default list to assign MACT Flags to all processes at a site. However,
if any one process already has a MACT code, do not assign ANY SIC-based codes
to that site. If SIC-default list assigns a different MACT category than a Facility
list, then let that code overwrite the Facility list supplied code. Use "SIC-
DEFAULT" as the flag.
7. Apply NAICS-Default list to assign MACT and Area Source Flags to all processes
at a site. However, if any one process already has a MACT code, do not assign
ANY NAICS-based codes to that site. If NAICS-default list assigns a different
MACT category than a Facility list or SIC-default, then let that code overwrite the
previous codes. Use "NAICS-DEFAULT" as the flag.
In all cases, any one process and thus any emission record will be associated with one and only
February 01, 2006
49
-------
one MACT category. If data are supplied by the EPA for a particular MACT category or a
facility list is provided, then the appropriate MACT code will be assigned and SIC codes, NAICS
codes, and SCCs will not be used to default the MACT code. If SCC, SIC code, or NAICS code
defaults are used, we will either use SCC, SIC code, or NAICS code defaults for a process within
a facility and will not mix SCC, SIC code, and NAICS code defaults at a single facility. Thus
once we default a process within a facility based on SCC, we will not default MACT codes for
other records missing SCCs that have a SIC or NAICS code. For example, the three following
processes are present at a facility.
Process 1 SCC 10100101 SIC 4911
Process 2 SIC 4911
Process 3 SIC 4911
We would default Process 1 to MACT Code 1808-1, utility coal boiler, but we would not default
Process 2 and 3 to a MACT Code.
The SCC/SIC Code/NAICS Code/MACT Code default dictionary used in the 2002 NEI was
developed by comparing all of the SCCs, SIC Codes, and NAICS Codes with information on
types of sources that may be subject to each MACT and Section 112(k) Area Source standard.
EPA staff will review the draft 2002 NEI to revise the dictionary and to revise the facilities
associated with individual categories. For 2002 NEI, we revised the 1999 SCC/SIC Code/MACT
Code dictionary to include the new area sources categories. SCC/MACT Code, SIC Code/MACT
Code and NAICS Code/MACT Code defaults used in the 2002 NEI are found in the tables, "SCC
to MACT Defaults", "SIC to MACT Defaults", and "NAICS to MACT Defaults", in the 2002
State Lookup File, 02nei_lkup_states.zip, found at:
ftp://ftp.epa.gov/EmisInventorv/2002finalnei/documentation/augmentation point sector/ .
Separate SCC MACT default tables are available for the Hazardous Organic NESHAP and
petroleum Refinery MACT Categories in the Lookup file.
We will assign MACT Codes to both HAP and CAP records in the 2002 NEI.
7.6 MACT and Section 112(k) Area Source Standards Code Assignment
The basis of MACT and Section 112(k) Area Source Standards Code (hereafter, referred to as
MACT Code) assignment in the 2002 NEI will be indicated in the data field, MACT Flag
Assignment, using the following values.
EFIG-Based
ESD-Based
FACILLIST-99NEI
F ACILLIST-ESD002
FACILLIST-MLT50
FACILLIST-MLT51
FACILLIST-MLT52
FACILLIST-MLT53
FACILLIST-RTI
NAICS Default
February 01, 2006
50
-------
SCC Default
SIC Default
State-Based
7.7 MACT and Section 112(k) Area Source Standards Compliance Code
The MACT and Section 112(k) Area Source Standards Compliance Code (hereafter, referred to as
MACT Compliance Code) will be used to identify whether a source is subject as a major or an
area source of HAPs to CAA Section 112 and Section 129 standards. The MACT Compliance
Code and MACT Code can be used by modelers to identify for a MACT or Section 112(k) Area
Source Standard source category major and area sources of HAPs that are subject to standards.
For projections, major and area sources of HAPs within a MACT or Section 112(k) Area Source
Standards category can have different percent reductions.
Acceptable NIF MACT Compliance Codes are:
01 Major Source (>10/25 tpy), Compliance date has not yet occurred;
02 Major Source (>10/25 tpy), Compliance date has occurred;
03 Area Source (<10/25tpy) category listed in, and subject to, Sections 112 &129 standards;
04 Area Source (<10/25tpy) category listed in, but not subject to, Sections 112 &129
standards as a synthetic minor;
05 Area Source (<10/25tpy) category listed in, but not subject to, Sections 112 &129
standards as true area or natural minor source;
06 Rule only affects major sources; area may be flagged; and
07 Rule only covers certain HAPs, all HAPs flagged.
The MACT Compliance Code will be assigned as follows.
1. If the process is associated with an Section 112(k) Area Source category code and
this category is only a Section 112(k) Area Source category, assign compliance
code "03".
2. If a process is associated with a category code that could either be a MACT or
Section 112(k) Area Source category code and the facility is a major source, assign
the compliance code listed under "Major Source compliance code" in the MACT
lookup table.
3. If a process is associated with a category code that could either be a MACT or
Section 112(k) Area Source category and the facility is an area source, assign
compliance code "03".
4. If a process is associated with a category code that is only a MACT Code and the
facility is a major source, assign the compliance code listed under "Major Source
MACT Compliance Code" in the MACT Code lookup table.
5. If a process is associated with a category code that is only a MACT Code and the
February 01, 2006
51
-------
facility is a area source, assign the compliance code listed under "Area Source
MACT Compliance Code" in the MACT Code lookup table.
7.8 Facility Category
The Facility Category indicates whether a facility is a major or an area source of HAP emissions.
HAP emissions are defined as either major or area in the Clean Air Act. Major sources are
defined as stationary sources that:
have the potential to emit 10 tons per year (tpy) or more of one HAP; or
have the potential to emit 25 tpy or more of any combination of HAPs.
Area sources are defined as sources with annual emissions below these thresholds.
For more information about the definition of major and area sources, please see:
www.epa.gov/ttn/atw/pollsour.html .
If the Facility Category is missing for a facility that emits HAPs, we will calculate the total
emissions of HAPs for the facility by summing the reported emissions of individual HAPs to
determine if a facility is a major or an area source. If more than one HAP Performance Level is
reported, we will select an emissions value for only one HAP Performance Level to use in
calculating the total emissions of the facility. The value of the reported emissions using the
following hierarchy will determine if a facility is major or area.
1. Potential
2. Actual
3. Allowable
4. Maximum
After calculating the facility emissions, we will populate missing fields with valid NIF Codes.
The valid NIF codes for the Facility Category are:
01 - Major HAP source, and
02 - Area HAP source
February 01, 2006
52
-------
SECTION 8. EMISSION RELEASE DESCRIPTION AND PARAMETERS
This section includes the following data fields.
Emission Release Point Type
Stack Parameters (Stack Height, Stack Diameter, Exit Gas Temperature, Exit Gas
Velocity, Exit Gas Flow Rate)
Stack Parameter Default Flag
In preparing emissions for grid modeling, valid parameters for the physical characteristics of each
release point (stack height, diameter, temperature, velocity, and flow) are necessary to correctly
place facility release points and associated emissions into vertical layers for proper air quality
modeling. Gaussian dispersion models need stack parameters to characterize the plume, which is
needed to estimate proper concentrations from these models. The first step is to QA the Emission
Release Point Type. After we augment for any invalid or missing Emission Release Point Types,
we will use a routine to assess the validity of the stack parameters, to replace values if necessary,
and to fill-in missing data points.
8.1 Emission Release Point Type
The Emission Release Point Type identifies whether emissions are released as a stack or fugitive.
The valid NIF codes for the Emission Release Point Type are:
01 - Fugitive,
02 - Vertical,
03 - Horizontal,
04 - Goose Neck,
05 - Vertical with Rain Cap, and
06 - Downward Facing Vent
The Emission Release Point Type is needed to determine how to augment missing or out-of-range
stack parameters.
We will evaluate the SCCs of Emission Release Point Types to determine if the Emission Release
Point Type is reported correctly as a fugitive or a stack by using the SCC Emission Release Point
Type Crosswalk. Two tables are in the crosswalk - SCCs that are always fugitive and SCCs that
are always non-fugitive (stack). The Crosswalk is available at the following address:
ftp://ftp.epa.gov/EmisInventorv/2002finalnei/documentation/augmentation point sector/ .If the
Emission Release Point Type is found to be inconsistent with the reported SCC, we will revise the
Emissions Release Point Type. When we change a non-fugitive(stack) value to fugitive, we will
use NIF Code 01 for Emissions Release Point Type. When we change a fugitive value to non-
fugitive, we will use NIF Code 02 for a vertical stack.
8.2 Stack Parameters QA and Default Procedure
Stack parameters include stack height, stack diameter, exit gas temperature, exit gas velocity, and
exit gas flow rate.
February 01, 2006
53
-------
We will employ a routine that compares each emission release point parameter to a minimum and
maximum range of values and when that parameter is missing or is found to exist outside of that
range, we will augment the parameter. We will also check non-fugitive stack parameters for
internal consistency between:
stack height and diameter, and
stack diameter, exit gas velocity, and exit gas flow rate.
When internal consistency is not met, we will replace the parameters.
The following steps summarize the process of finding and replacing missing, out-of-range, or
internally inconsistent stack parameters.
Step 1: For fugitive emission release points, replace stack parameters
For fugitive emission release points, we will first compare the existing height against the
following range thought to be representative of the minimum and maximum values allowable for
most fugitive emission release points. Although the values must be reported in English units and
will be defaulted to English units, we have provided metric conversions for convenience.
Fugitive Release Height: 0.1 to 100 ft 0.03048 to 30.48 m
If the height is valid, we will keep the height and replace all other stack parameters with the
defaulted values listed below. If the height is invalid, we will replace all stack parameters with
the defaulted values.
Stack Height: 10 ft 3.048 m
Stack Temperature: 72 °F 295.37 °K
Stack Diameter: 0.003ft 0.00091 m
Stack Velocity: 0.0003 ft/sec 0.000091 m/sec
Stack Flow: 0 cu ft/sec 0 cu m/sec
Step 2: For non-fugitive emission release points, find out-of-range or missing stack parameters
For non-fugitive emission release points, we will first compare existing stack parameters against a
set of the following ranges thought to be representative of the minimum and maximum values
allowable for most emission release points. Although the values must be reported in English units
and will be defaulted to English units, we have provided metric conversions for convenience.
StackHeight: 0.1 to 1250 ft 0.03048 to 381.00 m
Stack Temperature* 50 to 1,800 °F 283.15 to 982.37 °K
Stack Diameter 0.1 to 50 ft 0.03048 to 15.24 m
Stack Velocity 0.1 to 150 ft/sec 0.03048 to 45.72 m/sec
* Stack Temperature should be greater than 100 °F for the following SCCs and MACT Codes if
controls are not present.
SCCs: 10*, 20*, 501001*, 501005*, 502001*, 502005*, 503001*, 503005*
MACT codes: 0105, 0107, 0107-1, 0107-2, 0107-3, 0107-4, 0108, 0801, 0801-1, 0801-2,
0801-3, 0801-4, 1801, 1802, 1808-1, 1808-2, 1808-3
February 01, 2006
54
-------
First we will identify missing or out-of range parameters. Then we will evaluate the source
category to determine if out-of-range parameters may be plausible. If any parameter is missing or
out-of range, the parameter will be replaced using the procedures described in Step 4.
If all parameters are found to exist within the bounds of the emission release point ranges, we will
proceed to Step 3.
Step 3: For non-fugitive emission release points, find inconsistencies in stack parameters
We will determine any inconsistencies in stack parameters by conducting the following two steps.
A. For stack diameter, we will compare the stack diameter to the stack height. For non-
fugitive emission release points, the stack height may not be less than stack diameter.
B. We will determine the internal consistency between diameter, velocity and flow rate using
the following equation.
Stack Flow [cu ft/sec] = (II [Pi] * (Stack Diameter [ft] / 2) A 2) * Stack Velocity [ft/sec]
If the calculated flow and the reported flow are within 10 % of one another, then internal
consistency is assumed to be valid.
If all parameters are found to exist within the bounds of the emission release point ranges in Step
2, and the consistency checks (A) and (B) in Step 3 are satisfied, no additional steps will be taken.
If any parameter is missing or out-of range, or if the parameters fail the internal consistency tests,
the parameter will be replaced using the procedures described in Step 4.
Step 4: Replace stack parameters for non-fugitive emission release points
The first step in replacing stack parameters is to determine if there are problems with stack height
or diameter. Because stack height and diameter are the physical parameters that are most easily
measured or estimated, when there are problems with these parameters, then the entire set of stack
parameters are deemed questionable. If either height or diameter is missing or out-of range, or if
the stack diameter is greater than stack height, then all 5 parameters will be defaulted using
national default sets of physical parameter data. No additional steps are taken once all 5
parameters are defaulted.
If stack height and diameter do not need replacement, then velocity and flow rate are evaluated
next. If velocity and flow rate are not internally consistent, we will QA the flow rate to determine
if it was reported in cubic feet per minute rather than cubic feet per second as required in the NIF.
In the 1999 NEI, we found that several data submitters reported flow rate in cubic feet per minute
rather than cubic feet per second. We will correct flow rates reported in cubic feet per minute to
cubic feet per second and than evaluate the flow rate and velocity for internal consistency.
If the internal consistency is not met for velocity, flow rate, and diameter, Table 12 provides
instructions on how we will replace missing, out-of-range values, or internally inconsistent values
February 01, 2006
55
-------
for velocity and flow rate based on different reported scenarios. Velocity and flow rate are
augmented either by calculation or the use of national defaults.
Table 12. Stack Parameter Dal
a Replacement Matrix (X = Data value present)
Diameter
Velocity
Flow
Rate
Action
X
X
X
1. Check that velocity is within range.
A. If velocity is within range and flow rate does
not meet internal consistency for diameter,
velocity and flow rate, then:
~ Calculate flow rate using internal
consistency formula.
B If velocity is not within range, then:
~ Calculate velocity using internal
consistency formula.
~ Check that calculated velocity is within
range. If so, then default to calculated
velocity.
~ If calculated velocity is not within
range, then default all 5 parameters
using national default set.
X
X
1. Calculate velocity using internal consistency formula.
2. Check that calculated velocity is within range.
A. If calculated velocity is not within range, then:
~ Default all 5 parameters using national
default sets.
X
X
1. Check that velocity is within range.
A. If velocity is within range, then:
~ Calculate flow rate using internal
consistency formula.
B. If velocity is not within range, then:
~ Default all 5 parameters using national
default sets.
X
-
-
1. Default velocity using national default sets.
2. Calculate flow rate using internal consistency formula.
-
X
X
1. Default all 5 parameters using national default sets.
February 01, 2006
56
-------
Finally, in cases where all 5 parameters have not been defaulted, and velocity and flow rate have
been evaluated and replaced if necessary, temperature is evaluated. If temperature is missing or
out-of-range, then the temperature is defaulted using national default sets of physical parameter
data.
The following four National default sets of physical parameter data are available for non-fugitive
emissions.
1. SCC match
2. MACT Code match
3. facility level SIC Code match
4. national default for release points, if no SCC, MACT Code or SIC Code match is
possible
From Version 3 of the 1999 NEI for HAPs and 1999 NEI for CAPs, default look-up tables were
generated by SCC, MACT Code and SIC code to report the average value calculated for stack
height, diameter, temperature and velocity for all emission release point types that are coded as
stacks. Only valid values were used in the averaging. The following records were removed from
the 1999 NEI files prior to preparing SCC, MACT Code, and SIC Code stack parameter default
look-up tables.
Records with SCCs incorrectly reported as stack
Records with national default
Records with velocity greater than 100 ft/sec
• Records with temperatures < 250 °F with SCC 10*, 20*, 501001*, 501005*, 502001*,
502005*, 503001*, and 503005*
• Records with temperatures < 250 °F with MACT codes 0105, 0107, 0107-1, 0107-2, 0107-
3, 0107-4, 0108, 0801, 0801-1, 0801-2, 0801-3, 0801-4, 1801, 1802, 1808-1, 1808-2, and
1808-3
The stack flow rate was calculated by using the average diameter and average velocity and the
equation in Step 3.
Default stack parameters are available for more than 3,600 SCCs, 125 MACT codes and more
than 800 SIC codes. Separate look-up tables are prepared for SCCs, MACT Codes, and SIC
Codes. These tables are in the file, 2002 NEI Stack Parameter Default, and can be found at the
following address:
ftp://ftp.epa.gov/EmisInventorv/2002finalnei/documentation/augmentation point sector/ .
The following hierarchy is used to replace missing or invalid stack parameters for non-fugitive
emissions..
1. SCC match unless the SCC is 39999999
2. MACT Code match
3. facility level SIC Code match
4. national default for release points, if no SCC, MACT Code or SIC Code match is
possible
February 01, 2006 57
-------
If multiple SCCs (or MACT Codes or SIC Codes) are available for a single Emission Release
Point, we will use the default record having the lowest stack height to modify and replace that out-
of-range release parameters.
If no SCC, MACT Code, or SIC code match is possible, we will use the following national default
values for the stack parameters. Although the values must be reported in English units and will be
defaulted to English units, metric conversions are provided for convenience.
Stack Height: 10 ft 3.048 m
Stack Temperature: 72 °F 295.37 °K
Stack Diameter: 1ft 0.3048 m
Stack Velocity: 15 ft/sec 5.572 m/sec
Stack Flow: 12 cu ft/sec 339.802 cu m/sec
We will send a Stack Parameter QA report to data providers. The report contains all of the
emissions release point records submitted and identifies which parameters were defaulted as a
result of our QA. We will request that data providers review the defaulted records and revise the
records if they do not agree with the defaulted values. If we do not receive any comments from
data providers, we will retain the defaulted values.
8.3 Stack Parameter Default Flag
All defaulted stack parameters will be identified in the Emission Release Point record. We will
use the following coding system to identify the source of default stack parameters.
0 = Original value (not a default)
1 = SCC default
2 = SIC code default
3 = National default
4 = Calculated value
5 = MACT Code default
6 = State/Local Agency or Tribe revision to the defaulted value
8 = CAMD Value
A single default field will be used to represent the source of all five stack parameters. The codes
will be presented in this field in the following order: stack height, exit gas temperature, stack
diameter, exit gas velocity, exit gas flow rate. Thus, the code "00014" indicates that stack height,
exit gas temperature, and stack diameter are original values, exit gas velocity is based on SCC
defaults, and exit gas flow rate is calculated based on the stack diameter and exit gas velocity
values.
8.4 Stack Parameters QC Report
We will prepare a QC report that contains a list of emission release point records that failed our
QA of stack parameters and provides our defaulted parameters. We will send these reports to the
data providers and will ask data providers to review the defaults and provide corrected stack
February 01, 2006
58
-------
parameters. Replacing stack parameters could improve the quality of the data needed for
modeling.. Data providers can review the defaults and provide corrected stack parameters if
desired. If we do not receive corrections, we will use the defaulted values in the 2002.
We will prepare a file, "StackParameterQA", that contains tables and forms for submitting
revisions.
The following forms will be included in the file.
Stack Parameter QC Report - Summarizes QC of all stack parameters and provides
information on the number of records defaulted for each stack parameter
View/Revised Defaulted Emission Release Type Records - contains a list of all Emission
Release Point records that have the Emissions Release Type defaulted based on SCCs; the
report presents the original and defaulted Emission Release Type and reviewers may revise
these records by revising the SCCs of the records
View/Revised Converted Exit Gas Flow Rate Records - contains a list of all Emission
Release Point records where the Exit Gas Flow Rate have been defaulted using the stack
flow equation in Section 8.2; the report presents the original and defaulted parameters and
reviewers may revise these parameters
View/Revised Stack Parameters Records - contains a list of all Emission Release Point
records where the stack parameters have been defaulted; the report presents the original
and defaulted parameters and reviewers may revise these parameters
The following tables will be included in the StackParameterQA file.
Default tables
~ National defaults for fugitive releases
~ National defaults for stacks
~ MACT Code defaults
SCC defaults
~ SIC Code defaults
Stack Default Codes for all 5 stack parameters
Stack Parameter Revision Table - contains all stack records (both defaulted and non-
defaulted records)
The Stack Parameter Revision contains the following data fields. Revisions will be provided by
data providers by filling in the fields ending with "REVISED".
StateCountyFIPS
StateFacilityldentifier
• EmissionReleasePointID
TribalCode
FacilityName
LocationAddress
City
State
ZipCode
February 01, 2006
59
-------
StackDefault Codes
Emi s si onRel easePointT y peORIGINAL
Emi s si onRel easePointTy peDEF AULTED
EmissionReleasePointTypeRE VISED
StackHeightORIGINAL
StackHeightDEF AULTED
StackHeightRE VISED
StackDiameterORIGINAL
StackDi ameterDEF AULTED
StackDi ameterRE VISED
ExitGasT emperatureORIGINAL
ExitGasT emperatureDEF AULTED
ExitGasT emperatureRE VISED
ExitGasVel ocity ORIGINAL
ExitGas Vel ocity_DEF AULTED
ExitGas V el ocity RE VISED
ExitGasFlowRateORIGINAL
ExitGasFlowRateDEF AULTED
ExitGasFlowRateRE VISED
MACT Code
SCC
SIC Code
ConvertFlowConsistency - Indicates if calculated flow rate is inconsistent with other
velocity
Data Source
Data Provider - person who provided data
State File Name
February 01, 2006
60
-------
SECTION 9. UNIT CODES
This section includes the following data fields.
Design Capacity Numerator
Design Capacity Denominator
• Emission Unit Numerator
• Factor Unit Denominator
Factor Unit Numerator
Fugitive Dimensions Unit
Throughput Unit Numerator
Unit codes are used in a variety of data fields in the NIF. Table 13 associates NIF Data Fields with
the Unit Fields. For example, if Design Capacity is reported, then the Design Capacity
Numerator and Denominator must be reported.
Table 13. Fields in NIF Requiring Unit Codes
Fields Needing Unit Codes
Unit Code Fields
Design Capacity
Design Capacity Numerator
Design Capacity Denominator
Emission Numeric Value
Emission Unit Numerator
Factor Numeric Value
Factor Unit Numerator
Factor Unit Denominator
Horizontal Area Fugitive
Release Height Fugitive
Fugitive Dimensions Unit
Actual Throughput
Throughput Unit Numerator
9.1 Unit Codes
We will QA the reported Unit Codes. If the reported Unit Code values are incorrect, then we will
make corrections using valid codes. This type of error occurred frequently in the development of
the 1999 NEI. If a new code is reported that is not in the NIF Code Table, we will add the new
code to the NIF Unit Code Table. If we cannot determine what the Unit Code is or if the Unit
Code is missing for a reported value, we will contact the data provider to obtain the correct unit
and then we will augment the data field with the correct Unit Code.
9.2 Design Capacity, Design Capacity Numerator and Denominator
We will conduct additional QA of Design Capacity and Design Capacity Numerator and
Denominator for industrial boilers, turbines and IC engines. Design Capacity information is
needed for modeling that considers regional control strategies.
February 01, 2006
61
-------
The steps of the process include the following.
1. Identify industrial boilers, turbines and IC engines by using SCCs of 102* and 202*.
2. QA reported Design Capacity Numerator and Denominator. The Design Capacity
Numerator/Design Capacity Denominator should be MMBTU/HR. If Design Capacity
Numerator/Design Capacity Denominator is reported as BTU/HR, we will QA the Design
Capacity Value and replace with correct value as reported in units of MMBTU/HR.
3. If Design Capacity and Design Capacity Numerator and Denominator are missing and
capacity information is reported in the Emission Process Description or in the Emission
Unit Description, we will use this information to populate the Design Capacity, Design
Capacity Numerator, and Design Capacity Denominator fields,
4. For records that are missing Design Capacity and Design Capacity Numerator and
Denominator with SCCs of 102* and 202* except 10200402, 10200403, 10200502,
10200503, 10200602, and 10200603, and emissions greater than 100 tpy NOX, we will
send a list of these facilities to the data provider and encourage them to provide capacity
information.
February 01, 2006
62
-------
SECTION 10. ACTIVITY DATA FIELDS
This section includes the following data fields.
Emission Type
Start Date
End Date
Start Time
• End Time
Year of Emission Estimate
Annual Average Days per Week
Annual Average Hours per Day
Annual Average Hours per Year
Annual Average Weeks per Year
Period Days per Week
Period Hours per Day
Period Hours per Period
Period Weeks per Period
Winter Throughput Percent
Spring Throughput Percent
Summer Throughput Percent
Fall Throughput Percent
In addition to the correct placement of emissions, air quality modeling attempts to represent the
actual physical and chemical processes as they occur over a specific duration of time. As such, it is
important that the temporal allocation of emissions be as accurate as possible. Temporal allocation
can be thought of as an accounting of emissions variation over time. The simplest temporal
allocation is for a steady-state emissions source that continually releases emissions at the same rate
all the time. Under actual conditions, however, steady-state emission sources are quite rare.
Instead, emissions sources may operate only in the winter, not operate on Sundays, or their activity
may peak during certain hours of the day.
Temporal allocations based on seasonal throughput percentages, weeks per year, days per week,
and hours per day allow emissions variability to be correctly modeled during the desired modeling
periods. For example, an emission source reporting activity for fifty-two weeks per year, five days
a week, and eight hours a day would be allocated a temporal emission release between the hours of
9 AM and 5 PM, Monday through Friday, each week of the year.
If operating parameters are not reported by data providers to assign emissions to month, week,
day, or hour, default SCC-based temporal profiles are used to represent when emissions from
source types typically occur. Documentation of the procedures and allocation profiles currently
used in this allocation can be found at the following address:
http://www.epa.gov/ttn/chief/emch/temporal/index.html . It is the intent of EPA to use actual
reported operating parameters to assign when individual sources release their emissions. Currently,
however, many instances exist where data submitted to EPA are outside of acceptable ranges or
February 01, 2006
63
-------
missing completely. This results in the assignment of a flat profile of twenty four hour a day,
seven days a week, fifty two weeks a year operation (i.e., the same emission release each hour of
each day of the year). Having operating parameters provided with the emissions inventory will
result in a better, more accurate, source specific distribution of emissions on a diurnal, daily,
weekly, and seasonal level.
10.1 Emission Type
The Emission Type is a temporal description of the emissions reported. Acceptable values include
the following codes.
20
Average Sunday
21
Average Monday
22
Average Tuesday
23
Average Wednesday
24
Average Thursday
25
Average Friday
26
Average Saturday
27
Average Weekday
28
Average Weekend Day
29
Average Day
30
Entire Period
We will QA the data field and default missing or invalid codes to "30".
10.2 Start and End Dates
If the Start Date and End Date are not included in data submittals, we will default the Start and
End Date using the Inventory Year. For example if the Inventory Year is 2002, we will default the
Start Date to 20020101 and the End Date to 20021231.
10.3 Start and End Times
If the Start Time and End Time are invalid in data submittals, we will default the Start Time to
0000 and the End Time to 2359.
10.4 Year of Emission Estimate
The Year of the emission estimate indicates the actual year of the estimate whereas the Inventory
Year for the 2002 NEI is 2002. For example, some agencies collect HAP data every two years
(1999, 2001, 2003, 2005, etc.), and they have reported 2001 data for the 2002 NEI. In this
example, the Inventory Year in the 2002 NEI is 2002 and the Year of Emission Estimate is 2001.
We will populate the Year of the emission estimate using the Start and End Dates.
February 01, 2006
64
-------
10.5 Annual Average Days Per Week
If Annual Average Days Per Week is invalid (greater than 7), we will default invalid values to 7.
10.6 Annual Average Hours Per Day
If Annual Average Days Per Week is invalid (greater than 24), we will default invalid values to
24.
10.7 Annual Average Hours Per Year
We will QA this data field by using the following equation.
Annual Ave Hrs Per Yr = Annual Ave Hrs Per Day * Annual Ave Days Per Week * Annual Ave Weeks Per Yr
The maximum value is 8760. If Annual Average Hours Per Year is invalid, we will default invalid
values by using the equation if valid Annual Average Hours Per Day, Annual Average Days Per
Week, and Annual Average Weeks Per Year are reported. If these data fields are not reported,
then we will default invalid values to 8760.
10.8 Annual Average Weeks Per Year
If Annual Average Weeks Per Year is invalid (greater than 52), we will default invalid values to
52.
10.9 Period Days per Week
We will QA this data field using Start and End Dates for records. If the period is one week or less
based on the Start and End Dates, we will default missing or invalid values to the number of days
in a week based on the Start and End Dates. If the period is greater than one week based on Start
and End Dates, we will default missing and invalid values to 7.
10.10 Period Hours per Day
We will QA this field using Start and End Times. If Start and End Times are present, we will
default missing and invalid values based on Start and End Times. If Start and End Times are
missing, we will default missing and invalid values to 24.
10.11 Period Hours per Period
We will QA this field using Start and End Dates and Start and End Times. We will first calculate
the number of days of the period for the record. If Start and End Times are reported, we will use
the Start and End Times to calculate the number of hours in a day the process operates. If the Start
and End Times are not reported, we will assume the facility operates 24 hrs a day. We will default
missing or invalid values by multiplying the number of days per period by the number of hours per
day.
February 01, 2006
65
-------
10.12 Period Weeks per Period
We will QA this field using the Start and End Dates. We will default invalid values based on Start
and End Dates.
10.13 Winter, Spring, Summer, and Fall Throughput Percentages
These data fields are needed by EPA and others to calculate emissions for an average day in a
specific season from annual emissions. If seasonal throughput percentages are not reported, then
we will use SCC-based temporal profiles found at the following address:
http://www.epa.gov/ttn/chief/emch/temporal/index.html . NOTE: EPA will not retain seasonal
emissions from data submitters unless seasonal throughput percentages are included in NIF
submission.
Throughput Percentages should be reported as an integer between 0 and 100. We will first QA
the reported values and change any percentages reported as decimal to integer. We found some
values being reported as a decimal less than 100.
Then we will use the following methodology to QA Throughput Percentages.
1. Determine if the reported seasons of the Throughput Percentage match the Start and End
Date period.
2. Sum the reported Seasonal Throughput Percentages. The sum of Winter, Spring, Summer
and Fall Throughput Percentages should equal 100.
When the Throughput Percentages are invalid, we will use the following methodology to default
these data fields.
If the sum is greater than 100, we will default Throughput Percentages by allocating
percentages based on the Start and End Dates. For example, if the Start and End Dates are
April 2002 - August 2002, then only Spring and Summer Throughput Percentages should
be reported and we would default the invalid fields to Spring Throughput Percentage of 40
and Summer Throughput Percentage to 60.
If the sum is less than 100%, we will default missing percentages by summing reported
percentages and equally allocating remaining percentages based on Start and End Dates.
For example, if the reporting period is April 2002 - November 2002 and Summer
Throughput Percentage is reported as 90, we would default Spring Throughput Percentage
to 5 and Fall Throughput Percentage to 5.
February 01, 2006
66
-------
SECTION 11. CONTROL EQUIPMENT
This section includes the following data fields.
Control Status
Primary Device Type Code
Secondary Device Type Code
Third Device Type Code
Fourth Device Type Code
Total C apture C ontrol Effi ci ency
Total Capture Control Efficiency Methodology Flag
Information about control equipment and efficiencies is needed for modeling. This information
can be used in regulatory assessments as well as projections. Control information allows one to
look at the potential for future reductions within a facility or source category.
11.1 Control Status
We will QA and default the Control Status to indicate if emissions are controlled or uncontrolled.
The valid values are "CONTROLLED", "UNCONTROLLED", and "UNKNOWN".
We will QA the Control Status field by examining the control device codes in the CE table.
Invalid values occur for the following two cases.
All control device codes are "000" and the control status is "CONTROLLED"
All control device codes are not "000" and the control status is "UNCONTROLLED".
For invalid Control Status values, we will contact the data provider.
For missing values, we will use the following methodology.
If a primary control device code is not reported in the CE table, we will default the status as
"UNKNOWN".
If any control device code is reported as any code other than 000 in the CE table, we will
default this field as "CONTROLLED".
If the control device code of 000 is reported in the CE table, we will populate this field as
"UNCONTROLLED".
11.2 Primary Device Type Code, Secondary Device Type Code, Third Device Type Code,
and Fourth Device Type Code
We will QA the Device Type Codes. For missing or invalid Device Type Codes, we will use the
Control System Description to default the Device Type Codes. For control equipment not
included in the Control Device Code Table, we will add new codes.
February 01, 2006
67
-------
11.3 Total Capture Control Efficiency
To facilitate the use of data in dispersion and exposure modeling, we will default missing Total
Capture Control Efficiencies. The Total Capture Control Efficiency represents the collective
(aggregate) value for all control devices and is reported as a percent decimal.
We will QA the numerical value and revise if a decimal value instead of a percent is used. We
found that some states reported the numerical value as decimal only, for example they reported
0.985 rather than 98.5. We will also QA the numeric value for a range of 1 - 99.99.
We will augment the Total Capture Control Efficiency by reviewing the Primary Percent Control
Efficiency, Percent Capture Efficiency, and Total Capture Control Efficiency fields. Table 14
presents the methodology that will be used to populate this data field.
Table 14. Tota
Capture Control Efficiency Methodology
Total Capture
Control
Flag Code
Reported Data Fields
Total Capture Control Efficiency
Methodology
01
Primary Percent Control Efficiency = 0
Percent Capture Efficiency = 0
Total Capture Control Efficiency = 0
Total Capture Control Efficiency = 0
02
Total Capture Control Efficiency
Total Capture Control Efficiency = Reported Total
Capture Control Efficiency
03
Percent Capture Efficiency
Total Capture Control Efficiency = Percent Capture
Efficiency
04
Percent Capture Efficiency
Total Capture Control Efficiency
Total Capture Control Efficiency = Total Capture
Control Efficiency
05
Percent Capture Efficiency
Primary Percent Control Efficiency
Total Capture Control Efficiency = Percent Capture
Efficiency * Primary Percent Control Efficiency
06
Primary Percent Control Efficiency
Total Capture Control Efficiency = Primary Percent
Control Efficiency
07
Primary Percent Control Efficiency
Total Capture Control Efficiency
Total Capture Control Efficiency = Total Capture
Control Efficiency
08
Total Capture Control Efficiency
Percent Capture Efficiency
Primary Percent Control Efficiency
1. Calculate Total Capture Control Efficiency
by multiplying Percent Capture Efficiency
and Primary Percent Control Efficiency
2. Compare reported and calculated Total
Capture Control Efficiencies. Use the
greater value.
February 01, 2006
68
-------
11.4 Total Capture Control Efficiency Methodology Flag
All Total Capture Control Efficiencies will be identified in the Control Equipment record using
the Total Capture Control Flag codes listed in Table 14.
February 01, 2006
69
-------
SECTION 12. EMISSIONS
This section includes the following topics.
Pollutant Code
HAPs Emissions Reporting
CAPs Emissions Reporting
• PM Filterable and Condensible Augmentation for PM2.5 and PM10
• VOC and PM Augmentation from HAPs
CAP Augmentation for Missing CAPs
• HAPs Augmentation from VOC and PM
• HAP Performance level
QC of HAP Category Group Emissions
Emissions Type Period
Annual and Non-Annual Emissions
Emissions Ton Value
Emissions QC
Emission Data Quality Rating
Emissions Data Source
Historical Emissions Table
• Toxicity Weighting of HAPs
A variety of QA and data augmentation occurs related to the pollutant emissions. We will conduct
QA of reported pollutant codes and their emissions values. We will augment emissions of some
pollutants, and we will generate annual emission estimates from reported data. In addition, we
will assign an emission data quality rating to every HAP record and identify the source of data for
HAPs and CAPs. A Historical Facility Emissions Table will be compiled to show original
emissions submitted by all data sources, draft 2002 estimates, revisions received in Spring 2005,
final estimate in 2005 version of the 2002 NEI, and any revisions received over time. The Data
Source indicates who provided the emission estimate (state or local agency, EPA, Industry, Tribes,
etc.)
12.1 Pollutants
The 2002 NEI includes Hazardous Air Pollutants (HAPs) and Criteria Air Pollutants (CAPs).
Data providers should report pollutants using NIF pollutant codes. We will add any additional
pollutants that are HAPs submitted by data providers to the NIF pollutant code table.
NEI Pollutant Code Table
We will prepare a NEI Pollutant Code Table (Table 15). The NEI Pollutant Code Table is
contained in the 2002 State Lookup File, 02nei_lkup_states.zip, found at:
ftp://ftp.epa.gov/EmisInventorv/2002finalnei/documentation/point/ .
More information about specific fields in Table 15 will be discussed later in this section.
February 01, 2006
70
-------
Table 15. NEI Pollutant Code Table
Field Name
Description
Pollutant Code
NEI pollutant code Chemical Abstracts Services (CAS)
number for individual compounds; 2 and 3 digit numerical
codes for HAP Category groups and 3 letter Abbreviation for
CAPs
Pollutant Code Description
NEI pollutant description
HAP Category Name
For HAP pollutant groups, HAP group name
Chemical Formula
Chemical formula for pollutants that are not compound groups
Molecular Weight
Molecular Weight for pollutants that are not compound groups
Active
Yes or No: Is Pollutant Code Active?
Last Modification Date
Date code was added or deleted from NIF Code Tables
Valid for NIF 2.0
Yes or No
Valid for NIF 3.0
Yes or No
Comments
33 Urban HAP
Yes or No: Is Pollutant listed under Section 112k and 112c3
Urban Area Source Category Strategy that was published in
July 1999 in Federal Register, 64FR38705?
For more information, please see
http://www.epa.aov/ttn/atw/urban/urbanbackaround.html#Al .
33 Urban HAP 1990 baseline
NTI
Yes or No
NATO TEQ
NATO TEQ Factor
WHO TEQ
WHO TEQ Factor
I TEQ
I TEQ Factor
Ozone Precursor
Yes or No: IsHAPaVOC?
PM Precursor
Yes or No: Is HAP a PM?
HAP or CAP
Identify whether pollutant is a HAP or CAP
Pollutant Rank
Hierarchy used to select pollutants and emissions within a HAP
pollutant group where emissions of both individual compounds
within a group and a compound group are reported for an
emissions record
February 01, 2006
71
-------
CAPHAP Data Field
In the NOF EM Table, we will identify whether the pollutant in each record is a HAP or a CAP.
We will populate this field as CAP or HAP and will use the NEI Pollutant Code Table to populate
this field. Because the NEI contains both CAPs and HAPs, EIAG made the decision that
emissions data for lead, which is both a CAP and a HAP, will be treated in the NEI as a HAP.
Chemical Data Standards Lookup Table in the NEI
The EPA's Chemical Data Standard provides for the use of common identifiers throughout the
Agency for all chemical substances regulated or monitored by EPA environmental programs. This
standard provides unique, unambiguous, chemically correct common names for all chemicals,
substances and groupings in EPA's system and enables users to search for chemical substances
across EPA programs and databases. (For more information on EPA data standards, go to:
http://oaspub.epa.gov/edr/epastdS.startup# 1) The Chemical Identification Standards Database is
an excellent source of CAS numbers for HAPs. The Chemical Identification Standard consists of
the following data elements:
• Chemical Abstracts Service Registry Number - the unique number assigned by Chemical
Abstracts Service (CAS) to a chemical substance;
• Chemical Substance Systematic Name (9th Collective Index Name) - the name assigned to
a chemical substance that describes it in terms of its molecular composition;
• EPA Chemical Identifier - The unique number assigned in Chemical Registry System
(CRS) by EPA when CAS number is not available; and
• EPA Chemical Registry Name - the name EPA has selected as the name to be commonly
used by EPA in referring to a chemical substance.
We will compile a NEI Chemical Data Standards Table (Table 16) to meet EPA's Chemical Data
Standards. The NEI Chemical Data Standards Table contains each of these elements for every
NEI pollutant code and each NEI HAP Category. EPA's Office of Environmental Information
(OEI) assigns the following fields of data in Table 16.
CRS ID
Chem Sy stemati cName
• EPA Chemical ID
• EPAChemRegistryName
OEI also verifies the CASRN.
We will use the 99 NEI Chemical Data Standards Table for pollutants as a starting point and will
add any new pollutants to the Table. For new pollutants in 2002, we will submit a list of pollutants
to OEI for assignment of Chemical Data Standards information. The process of assigning
Chemical Registry System data fields in the 2002 NEI will not occur until after external revisions
are received and the final list of pollutants in the 2002 NEI is compiled.
The NEI Chemical Data Standards Table is contained in the 2002 State Lookup File,
02nei_lkup_states.zip, found at:
ftp://ftp.epa.gov/EmisInventorv/2002finalnei/documentation/point/ .
February 01, 2006
72
-------
Table 16. NEI Chemical Data Standards Table
Field Name
Description
CRSID
Chemical Registry System ID
NEI Pollutant Code
Unique code assigned to NEI pollutant
NEI Pollutant Name
HAP name for NEI pollutant
NEI HAP Category
Grouping of related NEI pollutants
CASRN
Chemical Abstracts Service Registry Number
CASRN_compact
Chemical Abstracts Service Registry Number without
dashes
ChemSystematicName
Chemical Substance Systematic Name (9th Collective
Index Name)
EPAChemicallD
EPA Chemical Identifier
EP AChemRegi stryN ame
EPA Chemical Registry Name
HAPCAT CRS ID
Chemical Registry System ID (assigned to HAP
category)
HAPCAT CASRN
Chemical Abstracts Service Registry Number (assigned
to HAP category)
HAPCATCASRNcompact
Chemical Abstracts Service Registry Number without
dashes (assigned to HAP category)
HAPCATChemSystematicName
Chemical Substance Systematic Name (9th Collective
Index Name)(assigned to HAP category)
HAPCATEPAChemciallD
EPA Chemical Identifier (assigned to HAP category)
HAPCAT EP AChemRegi stryName
EPA Chemical Registry Name (assigned to HAP
category)
HAPs Reporting
Section 112(b) of the CAA contains a list of 188 HAPs. HAPs are generally defined as those
pollutants that are known or suspected to cause serious health problems, including cancer. Section
112(b) of the Clean Air Act currently identifies a list of 188 pollutants as HAPs.
(http://www.epa. gov/ttn/atw/orig 189.html ) EPA's ATW web site presents more information on
HAPs, their effects, and EPA's programs to reduce HAPs. ( www.epa.gov/ttn/atw/pollsour.html)
February 01, 2006
73
-------
The NEI includes emissions data for all 188 HAPs. In addition to numerous specific chemical
species and compounds, the list of 188 HAPs includes several compound groups (e.g., individual
metals and their compounds, polycyclic organic matter (POM), and glycol ethers). Many of the
uses of the NEI depend upon data for individual compounds within these groups rather than
aggregated data on each group as a whole.
For risk assessments, individual speciated HAPs are needed because the toxicity associated with
an individual compound within a compound group varies widely. For risk, exposure and
dispersion modeling, we prepare files containing speciated pollutants using the NEI Pollutant Code
and Description. For trends purposes, we are also asked to track the emissions over time for the
Section 112b list. For trends analyses and summary NEI data, we prepare files using the Section
112b HAPs and aggregate speciated NEI pollutants using the HAP Category Name. The HAP
Category Name is the same name for HAPs that are not listed as compound groups in Section
112b, for example, formaldehyde. For a pollutant such as methylmercury, the HAP Category
Name is mercury and compounds and the NEI Pollutant Description is methylmercury.
The 2002 NEI NEI Pollutant Code Table lists all of the specific pollutants and compound groups
included in the 2002 NEI along with their Chemical Abstract Services (CAS) numbers for
individual compounds; information about if the pollutant is a constituent of VOC, PM or both;
TEQ factors; and other information. The NEI Pollutant Code Table is contained in the 2002 State
Lookup File, 02nei_lkup_states.zip, found at:
ftp://ftp.epa.gov/EmisInventorv/2002finalnei/documentation/point/ .
Please refer to this table for the following discussion of HAPs in the NEI.
Preferred Reporting of HAPs
We encourage data providers to report emissions for specific compounds, both for individual HAP
species and for pollutants within compound groups. CAS numbers are preferred to identify
pollutants reported to the EPA. If data providers cannot report pollutants using CAS numbers,
aggregated compound group emissions will be accepted. If emissions are reported for groups of
compounds, then we will use simplifying assumptions regarding speciation within the group in
order to use these data as inputs to models.
Recommendations for reporting data for specific groups of compounds are summarized below in a
hierarchy of most preferred method to least preferred. For pollutant groups, only one reporting
strategy per HAP group per source should be used. Simultaneous use of more than one reporting
strategy (e.g., reporting both individual chromium compounds and total chromium for the same
source) may result in the same emission being counted twice and potential overestimation of
emission levels and risk. CAS numbers referenced to in the text are also shown in 2002 NEI
Pollutant Code Table with their associated pollutant names.
Metal and cyanide groups:
1. Report emissions and associated CAS numbers of all individual metal and cyanide
compounds; e.g., report emissions and associated CAS numbers of arsenic oxide, lead
arsenate, etc., rather than emissions of arsenic compounds as a whole. All individual
February 01, 2006
74
-------
compounds should be reported as the mass of the total compounds, not just the metal
within the compound.
2. If individual metal compounds cannot be reported, a less preferred method for chromium,
lead, mercury and nickel is to separately report two forms of widely-varying toxicity. If
you use this approach, report only the mass of emissions of the metal, not of the entire
metal compound.
Chromium {NEI pollutant codes 7440473 and 136) - Separate chromium compounds into
hexavalent chromium (NEI Pollutant Code 18540299) and trivalent (NEI Pollutant
Codel6065831).
Lead {NEI pollutant code 195) - Separate lead compounds into organic lead (NEI
pollutant code 603) and inorganic lead (NEI pollutant code 602).
Mercury {NEIpollutant codes 7439976, 199, and 22967926)- Separate mercury
compounds into particulate divalent mercury (NEI Pollutant Code 202), elemental gaseous
mercury (NEI Pollutant Code 200) and gaseous divalent mercury (NEI Pollutant Code
201). Default mercury speciation factors are found at:
ftp://ftp.epa.gov/EmisInventorv/2002finalnei/documentation/point/ .
Nickel Compounds {NEIpollutant codes 226)- Separate nickel compounds into nickel
subsulfide (NEI Pollutant Codel2035722) and other nickel (CAS #7440020).
Other metal and cyanide groups - Report total emissions of the group in terms of the mass
of the metal or cyanide alone, and report under the CAS number of the metal or cyanide.
3. Alternatively, but far less preferred, report total emissions of the group in terms of mass of
total emissions, and report under the pollutant group number for "metal and compounds" or
"cyanide and compounds". Do not include metals or cyanide already reported using the
more preferred methods above, in order to avoid the possibility of double counting
emissions.
• POM : POM compounds are formed primarily from combustion and are present in the
atmosphere in particulate and gaseous forms. Table 17 identifies individual POM
Compounds of interest to EPA. POM compounds are generally grouped into POM reported
as 7-PAH, POM reported as 15-PAH, and POM reported as non-15-PAH. In order to
generate a total POM estimate, you must sum 15-PAH and non-15-PAH together. You
should not sum 7-PAH and 15-PAH together as 15-PAH includes 7-PAH. Table 17 shows
the specific compounds constituting 7-PAH and 15-PAH groups.
1. Report emissions and associated CAS numbers of as many individual POM compounds as
possible, rather than as total PAH or total POM. Most important to report individually are
the 7-PAH compounds listed in Table 17. We also encourage the reporting of other
February 01, 2006
75
-------
individual POM compounds for which cancer assessments are available (also listed in
Table 17).
2. If you cannot report all individual PAHs, then report 7-PAH as a subgroup.
3. If you cannot report emissions of 7-PAH, then report total POM (total POM includes total
PAH). Since naphthalene is listed individually as a HAP, do not include any individually-
reported naphthalene as total POM.
4. If you follow any other scheme than one listed in this hierarchy, clearly identify what it is.
Table 17. POM Compounds
7-PAII
15-PAII (includes
7-PAII coinpounds)
Oilier POM Compounds lor which
we have cancer assessments
B enz(a)anthracene
B enz(a)anthracene
Carbazole
Benzo(a)pyrene
Benzo(a)pyrene
Dibenz(a,h)acridine
B enzo(b)fluoranthene
Benzo(b)fluoranthene
Dibenz(a,j)acridine
B enzo(k)fluoranthene
B enzo(k)fluoranthene
7H-Dibenzo(c,g)carbazole
Chrysene
Chrysene
Dib enzo(a, e)py rene
Dibenz(a,h)anthracene
Dibenz(a,h)anthracene
Dibenzo(a,i)pyrene
Indeno( 1,2,3 -cd)pyrene
Indeno( 1,2,3 -cd)pyrene
Dibenzo(a,l)pyrene
Acenaphthene
7,12-Dimethylbenz(a)anthracene
Acenaphthylene
1,6-Dinitropyrene
Anthracene
1,8-Dinitropyrene
Benzo(ghi)perylene
3 -Methyl chol anthrene
Fluoranthene
5-Methyl chry sene
Fluorene
5 -Nitroacenaphthal ene
Phenanthrene
6-Nitrochrysene
Pyrene
2-Nitrofluorene
1-Nitropyrene
4-Nitropyrene
February 01, 2006
76
-------
• Dioxins/Furans: Dioxins and Furans can be grouped by 2,3,7,8-tetrachlorodibenzo-p-
dioxin (TCDD) equivalents (TEQ). TEQ factors are multipliers for some dioxin and furan
congeners to estimate a common basis of toxicity for chlorinated dioxins and furans. In the
NEI, we group dioxins and furans into TEQ and non-TEQ groups (See 2002 NEI Pollutant
Code Table). There are 3 sets of TEQ factors used to estimate TEQ -1 TEQ factors,
NATO TEQ factors and WHO TEQ factors. TEQ factors change as better toxicity data
become available.
1. Report mass emissions and associated CAS numbers of all individual congeners of both
chlorinated dibenzodioxins (CDDs) and chlorinated dibenzofurans (CDFs).
2. If you cannot report individual CDD and CDF congeners, report dioxins and furans as TEQ
under the HAP name "dioxins/furans as TEQ". (Note: Although the Clean Air Act
specifically lists only 2,3,7,8-TCDD as a HAP, other CDDs and CDFs qualify as HAPs
within the POM group. Because some of these other congeners are also potent
carcinogens, EPA will use the TEQ approach to evaluate CDDs and CDFs as a group.). Do
not include dioxins/furans in the reporting of POM emissions. Identify the TEQ approach
that you are using -1 TEQ, NATO TEQ or WHO TEQ. Separate pollutant codes are in the
NEI Pollutant Code Table for I TEQ (NEI pollutant code 626), NATO TEQ or WHO
TEQ (NEI pollutant code 627).
3. If you cannot report emissions using the TEQ approach, report individual congener
emissions where possible and report any remaining emissions as total "dioxins" or total
"dibenzofurans".
• Glycol Ethers: Glycol ethers include mono- and di- ethers of ethylene glycol, diethylene
glycol, and triethylene glycol. Polymers are excluded from the glycol ether category.
1. Report emissions for individual glycol ethers with their associated CAS numbers. Ethylene
Glycol Butyl Ether (EGBE) has been delisted as a HAP and should not be included in the
glycol ethers group. A good reference for determining if a compound is a glycol ether is:
http://daq.state.nc.us/toxics/glvcol. Another reference is the Toxic Release Inventory
(TRI) guidance on glycol ethers, http://www.epa.gov/tri/guide docs/2000/glvcol2000.pdf.
Note that, historically, many compounds have been mistakenly included in the glycol
ethers compound group. Table 18 identifies compounds that were reported to EPA as
glycol ethers in the development of the 1999 NEI that are not actually glycol ethers and
should not be included in 2002 NEI reporting.
2. If you cannot report individual glycol ether emissions, report total emissions of glycol
ethers as a group under "glycol ethers".
February 01, 2006
77
-------
Table 18. List of Compounds Often Mistaken as Glycol Ethers
Compound
CAS Number
1,1 -Dimethoxy ethane
534156
1 -Ethoxy-2-propanol
1569024
3 -Ethoxy-1 -propanol
111353
Diethylene glycol
111466
Diethylene glycol di(3-aminopropyl) ether
4246519
Dipropylene glycol monomethyl ether
34590948
Glycols, polyethylene, polypropylene
monobutyl ether (nonionic)
9038953
Isopropyl glycol
109591
Nonyl phenyl polyethylene glycol ether
9016459
Propylene glycol methyl ether acetate
108656
Propylene glycol monomethyl ether
107982
Propylene glycol t-butyl ether
57018527
Triethylene glycol
112276
• Xylenes and Cresols: Xylenes include ortho-, meta-, and para- isomers. Cresols include
ortho-, meta-, and para- isomers and cresylic acid.
1. Report emissions for individual xylene and cresol isomers with their associated CAS
numbers. Do not report any emissions for total xylenes or cresols to avoid double counting
2. If you cannot report individual emissions of xylenes or cresols, report total emissions of
xylenes or cresols as a group under "xylenes (mixture of o, m, and p isomers)" (NEI
Pollutant Codel330207) or "cresols/cresylic acids" (NEI Pollutant Code 331).
OA of HAP Pollutant Codes
We will QA the reported Pollutant Codes. If the reported codes are incorrect, we will contact the
data provider for a crosswalk of HAPs reported to CAS Numbers and pollutant names. If a
reported HAP is not in the current NIF Pollutant Code Table, we will add the pollutant to the NIF
Code Table, NEI Pollutant Code Table, and NEI Chemical Data Standards Table.
February 01, 2006
78
-------
We will augment HAPs and their emissions in the 2002 NEI if VOC and PM are reported for a
facility, but no HAPs are reported, and a reliable estimation methodology exists.
CAPs Reporting
CAPs include: Volatile Organic Compounds (VOC), Nitrogen Oxides (NOX), Carbon Monoxide
(CO), Sulfur Dioxide (S02), and Particulate Matter (PM). Ammonia (NH3) will also be included
in the 2002 CAP NEI. Lead is both a HAP and a CAP and has been included in previous
discussions on HAPs Reporting. Table 19 describes PM compounds that may be reported to EPA
by data providers for the 2002 NEI. We will generate missing emission estimates for filterable
and primary PM-2.5, filterable and primary PM-10 and condensible PM in the 2002 NEI.
Table 19. PM Compounds Reported in the 2002 NEI
NEI Code
Pollutant
Pollutant Description
PM-CON
Primary PM
Condensible portion
only (all < 1 micron)
Material that is vapor phase at stack conditions, but
which condenses and/or reacts upon cooling and
dilution in the ambient air to form solid or liquid PM
immediately after discharge from the stack.
PM-FIL
Primary PM, Filterable
portion only
Particles that are directly emitted by a source as a solid
or liquid at stack or release conditions and captured on
the filter of a stack test train.
PM-PRI
Primary PM, includes
filterables and
condensibles
PM-PRI=
PM-FIL + PM-CON
Particles that enter the atmosphere as a direct emission
from a stack or an open source. It is comprised of two
components: Filterable PM and Condensible PM.
PM10-FIL
Primary PM10,
Filterable portion only
Particles with an aerodynamic diameter equal to or less
than 10 micrometers that are directly emitted by a
source as a solid or liquid at stack or release conditions
and captured on the filter of a stack test train.
PM10-PRI
Primary PM10,
includes filterables and
condensibles,
PM10- PRI =
PM0-FIL + PM-CON
Particles with an aerodynamic diameter equal to or less
than 10 micrometers that enter the atmosphere as a
direct emission from a stack or an open source. It is
comprised of two components: Filterable PM and
Condensible PM. (As specified in § 51.15 (a)(2), these
two PM components are the components measured by a
stack sampling train such as EPA Method 5.)
February 01, 2006
79
-------
Table 19, continued. PM Compounds reported in the 2002 NEI
M l ( ode
Pollutant
Pollulanl Description
PM25-FIL
Primary PM2.5,
Filterable portion only
Particles with an aerodynamic diameter equal to or less
than 2.5 micrometers that are directly emitted by a
source as a solid or liquid at stack or release conditions
and captured on the filter of a stack test train.
PM25-PRI
Primary PM2.5,
includes filterables and
condensibles
PM25-PRI=
PM25-FIL + PM-CON
Particles with an aerodynamic diameter equal to or less
than 2.5 micrometers that enter the atmosphere as a
direct emission from a stack or an open source. It is
comprised of two components: Filterable PM and
Condensible PM. (As specified in § 51.15 (a)(2), these
two PM components are the components measured by a
stack sampling train such as EPA Method 5.)
We will QA the reported Pollutant Codes. If the reported codes are incorrect, we will correct
invalid codes.
We will QA reported CAP emissions to identify missing CAPs and invalid emissions. For each
individual emission record, the sum of PM-condensibles and PM2.5-filterables cannot be greater
than the sum of PM2.5-primary, and the sum of PM-condensibles and PMlO-filterables cannot be
greater than the sum of PMlO-primary.
We will augment CAPs and their emissions for two cases in the 2002 NEI.
1. PM augmentation for filterable and primary PM-10, filterable and primary PM-2.5, and
condensible PM
2. CAPs augmentation if a partial submission of CAPs is provided in a data submittal, e.g., a
state agency only submits NOX.
12.2 PM Augmentation
All PM emissions estimates that are defaulted in the 2002 NEI will include data source codes to
identify that the emissions are defaulted.
PM terms are related to each other via the following three equations.
PM-PRI = PM-FIL + PM-CON
PM10-PRI = PM10-FIL + PM-CON
PM25-PRI = PM25-FIL + PM-CON
February 01, 2006
80
-------
Steps in PM Augmentation
From those three equations, we developed seven math steps and 14 ratios that will be applied to
augment the reported PM terms in the 2002 point source NEI.
Before we run PM augmentation math steps, we will first review the data for inconsistencies. If
values are found to be inconsistent, they will be replaced. These replacement values will have a
data source code of "A-R." The consistency checks and replacement actions are as follows:
1. If PM10-PRI >0 and PM25-PRI > PM10-PRI (and PM10-FIL, PM25-FIL and PM-CON
are null or 0), then set PM25-PRI = PM10-PRI.
2. If PM10-PRI = 0 and PM25-PRI > PM10-PRI (and PM10-FIL, PM25-FIL and PM-CON
are null or 0), then set PM10-PRI = PM25-PRI.
3. If PM10-FIL > 0 and PM25-FIL > PM10-FIL (and PM10-PRI, PM25-PRI and PM-CON
are null or 0), then set PM25-FIL = PM10-FIL.
4. If PM10-FIL = 0 and PM25-FIL > PM10-FIL (and PM10-PRI, PM25-PRI and PM-CON
are null or 0), then set PM10-FIL = PM25-FIL.
5. If PM10-PRI >0 and PM10-FIL > PM10-PRI (and PM25-PRI, PM25-FIL and PM-CON
are null or 0), then set PM10-FIL = PM10-PRI.
6. If PM10-PRI = 0 and PM10-FIL > PM10-PRI (and PM25-PRI, PM25-FIL and PM-CON
are null or 0), then set PM10-PRI = PM10-FIL.
7. If PM25-PRI > 0 and PM25-FIL > PM25-PRI (and PM10-PRI, PM10-FIL and PM-CON
are null or 0), then set PM25-FIL = PM25-PRI.
8. If PM25-PRI = 0 and PM25 FIL > PM25-PRI (and PM10-PRI, PM10-FIL and PM-CON
are null or 0), then set PM25-PRI = PM25-FIL.
9. If PM10-PRI > 0 and PM25-FIL > PM10-PRI (and PM10-FIL, PM25-PRI and PM-CON
are null or 0), then set PM25-FIL = PM10-PRI.
Math Steps 1-7 will be used to augment PM.
Math Step 1
PM-CON term is needed. PM-PRI and PM-FIL are available
PM-CON = PM-PRI - PM-FIL
Math Step 2
PM-CON term is needed. PM10-PRI and PM10-FIL are available
PM-CON = PM10-PRI - PM10-FIL
February 01, 2006
81
-------
Math Step 3
PM-CON term is needed. PM25-PRI and PM25-FIL are available
PM-CON = PM25-PRI - PM25-FIL
Math Step 4
PM10-FIL term is needed. PMIO-PRI and PM-CON are available
PMIO-FIL = PMIO-PRI - PM-CON
Math Step 5
PMIO-PRI term is needed. PMIO-FIL and PM-CON are available
PMIO-PRI = PMIO-FIL + PM-CON
Math Step 6
PM25-FIL term is needed. PM25-PRI and PM-CON are available
PM25-FIL = PM25-PRI - PM-CON
Math Step 7
PM25-PRI term is needed. PM25-FIL and PM-CON are available
PM25-PRI = PM25-FIL + PM-CON
If Math Steps 1 through 7 are not adequate to augment all of the missing PM term(s), additional
augmentation steps will be needed that require the use of ratios developed from available
emissions and particle size distribution data. These ratios will be needed when only one PM term
is available, and two or more terms need to be augmented. Examples of how we will use the PM
ratios are shown below.
FIL x RatioCON/nL = CON
PRI x RatiOcoN/PM = CON
CON x RatioFIL/CON = FIL
CON x RatioPRI/CON = PRI
Ratios 1 through 14 will be used to complete PM augmentation.
Ratio 1
PM25-FIL term is needed. PMIO-FIL is available
PM25-FIL — PMIO-FIL x RatioPM25-FiL/PMio-FiL
Ratio 2
PMIO-FIL term is needed. PM25-FIL is available
PMIO-FIL — PM25-FIL x Ratio])Vti\i25-rn.
Ratio 3
PM-CON term is needed. PMIO-PRI is available
PM-CON — PMIO-PRI x RatioPM.CQN/pM10.PRI
February 01, 2006
82
-------
Ratio 4
PM25-PRI term is needed. PM10-PRI is available
PM25-PRI — PM10-PRI x RatiOpM25_pRI/pM10_pRI
Ratio 5
PM-CON term is needed. PM25-PRI is available
PM-CON — PM25-PRI x RatioPM.CON/pM25-PRi
Ratio 6
PM10-PRI term is needed. PM25-PRI is available
PM10-PRI — PM25-PRI x RatiOpM10.pRI/pM25-pRI
Ratio 7
PM-CON term is needed. PM10-FIL is available
PM-CON — PMIO-FIL x RatioPM.CON/pMio-FiL
Ratio 8
PM25-FIL term is needed. PMIO-FIL is available
PM25-FIL — PMIO-FIL x Ratio1)M25-Hii>\no-Hi
Ratio 9
PM-CON term is needed. PM25-FIL is available
PM-CON — PM25-FIL x RatiO],XKON i>m25-mi,
Ratio 10
PM10-PRI is needed. PM-CON is available
PM10-PRI = PM-CON x RatioPM10_PRI/pM_CON
Ratio 11
PMIO-FIL is needed. PM-CON is available
PMIO-FIL — PM-CON x RatioPM10.FIL/pM.CON
Ratio 12
PM-CON is needed. PM-PRI is available
PM-CON — PM-PRI x RatioPM_CQN/pM_PRj
Ratio 13
PMIO-FIL is needed. PM-FIL is available
PMIO-FIL — PM-FIL x RatioPM10.FIL/pM.FIL
Ratio 14
PM25-FIL is needed. PM-FIL is available
PM25-FIL — PM-FIL x RatiO],M25-Hii>\[-Hi
February 01, 2006
83
-------
Prior to the application of these math steps and ratios, we will analyze the data and separate the
data into cases. The cases will determine which combination of ratios and math steps to apply.
Table 20 presents an overview of these cases and their augmentation methodology.
Table 20. PM
Cases and Required Si
teps to Augment PM
PM Reported
PM to Default
Au^mentation Methodology
PM25-PRI
PM25-FIL
PM10-PRI
PM10-FIL
PM-CON
PM-PRI
PM-CON
PM25-PRI
PM25-FIL
PM10-PRI
PM10-FIL
PM-CON = PM-PRI x Ratio 12
PM-FIL = PM-PRI - PM-CON, Math Step 1
PM10-FIL = PM-FILx Ratio 13
PM25-FIL = PM-FIL x Ratio 14
PM25-PRI = PM-CON + PM25-FIL, Math Step 7
PM10-PRI = PM-CON + PM10-FIL, Math Step 5
PM10-PRI
PM-CON
PM25-PRI
PM25-FIL
PM10-FIL
PM-CON = PM10-PRI x Ratio 3
PM10-FIL = PM10-PRI - PM-Con. Math Step 4
PM25-FIL = PM10-FIL x Ratio 8
PM25-PRI = PM-CON + PM25-FIL, Math Step 7
PM10-FIL
PM-CON
PM25-PRI
PM25-FIL
PM10-PRI
PM-CON = PM10-FIL x Ratio 7
PM25-FIL = PM10-FIL x Ratio 1
PM25-PRI = PM-CON + PM25-FIL, Math Step 7
PM10-PRI = PM-CON + PM10-FIL, Math Step 5
PM25-PRI
PM-CON
PM25-FIL
PM10-PRI
PM10-FIL
PM-CON = PM25-PRI x Ratio 5
PM25-FIL = PM25-PRI - PM-Con, Math Step 6
PM10-FIL = PM25-FIL x Ratio 2
PM10-PRI = PM-CON + PM10-FIL, Math Step 5
PM25-FIL
PM-CON
PM25-PRI
PM10-FIL
PM10-PRI
PM-CON = PM25-FIL x Ratio 9
PM10-FIL = PM25-FIL x Ratio 2
PM25-PRI = PM-CON + PM25-FIL, Math Step 7
PM10-PRI = PM-CON + PM10-FIL, Math Step 5
PM-CON
PM25-PRI
PM25-FIL
PM10-PRI
PM10-FIL
PM10-FIL = PM-CON x Ratio 11
PM25-FIL = PM10-FIL x Ratio 1
PM25-PRI = PM-CON + PM25-FIL, Math Step 7
PM10-PRI = PM-CON + PM10-FIL, Math Step 5
PM-FIL
PM-CON
PM25-PRI
PM25-FIL
PM10-PRI
PM10-FIL
PM 10 -FIL=PM-FIL x Ratio 13
PM-CON= PM10-FIL x Ratio7
PM25-FIL = PM10-FIL x Ratio 1
PM25-PRI = PM-CON + PM25-FIL, Math Step 7
PM10-PRI = PM-CON + PM10-FIL, Math Step 5
PM10-FIL
PM10-PRI
PM-CON
PM25-PRI
PM25-FIL
PMCON= PM10-PRI - PM10-FIL, Math Step 2
PM25-FIL = PM10-FIL x Ratio 1
PM25-PRI = PMCON + PM25-FIL, Math Step 7
February 01, 2006
84
-------
Table 20 continued. PM Cases and Required Steps to Augment PM
PM Reported
PIVI to Default
Augmentation Methodology
PM10-FIL
PM-CON
PM10-PRI
PM25-PRI
PM25-FIL
PM25-FIL = PM10-FIL x Ratio 1
PM25-PRI = PM-CON + PM25-FIL, Math Step 7
PM10-PRI = PM-CON + PM10-FIL, Math Step 5
PM10-FIL
PM25-PRI
PM-CON
PM25-FIL
PM10-PRI
PM-CON= PM10-FIL x Ratio7
PM25-FIL = PM25-PRI - PM-Con, Math Step 6
PM10-PRI = PM-CON + PM10-FIL, Math Step 5
PM10-PRI
PM25-FIL
PM-CON
PM25-PRI
PM10-FIL
PM-CON= PM10-PRI x Ratio 3
PM25-PRI = PM-CON + PM25-FIL, Math Step 7
PM10-FIL = PM10-PRI - PM-CON, Math Step 4
PM10-PRI
PM-CON
PM10-FIL
PM25-PRI
PM25-FIL
PM10-FIL= PM10-PRI - PM-CON, Math Step 4
PM25-FIL = PM10-FIL x Ratio 1
PM25-PRI = PM-CON + PM25-FIL, Math Step 7
PM25-FIL
PM25-PRI
PM-CON
PM10-PRI
PM10-FIL
PM-CON= PM25-PRI - PM25-FIL, Math Step 3
PM10-FIL = PM25-FIL x Ratio2
PM10-PRI = PM-CON + PM10-FIL, Math Step 5
PM25-FIL
PM-CON
PM10-PRI
PM25-PRI
PM10-FIL
PM10-FIL = PM25-FIL x Ratio 2
PM25-PRI = PM-CON + PM25-FIL, Math Step 7
PM10-PRI = PM-CON + PM10-FIL, Math Step 5
PM-CON
PM25-PRI
PM25-FIL
PM25-PRI
PM10-FIL
PM25-FIL= PM25-PRI - PM-CON, Math Step 6
PM10-FIL = PM25-FIL x Ratio 2
PM10-PRI = PM-CON + PM1025-FIL, Math Step 5
PM10-FIL
PM25-FIL
PM-CON
PM25-PRI
PM10-PRI
PM-CON = PM10-FIL x Ratio 7
PM25-PRI = PM-CON + PM25-FIL, Math Step 7
PM10-PRI = PM-CON + PM10-FIL, Math Step 5
PM10-PRI
PM25-PRI
PM-CON
PM25-FIL
PM10-FIL
PM-CON = PM10-PRI x Ratio 3
PM25-FIL= PM25-PRI - PM-CON, Math Step 6
PM10-FIL= PM10-PRI - PM-CON, Math Step 4
PM-PRI
PM-FIL
PM-CON
PM25-PRI
PM25-FIL
PM10-PRI
PM10-FIL
PM-CON = PM-PRI - PM-FIL, Math Step 1
PM10-FIL = PM-FILx Ratio 13
PM25-FIL = PM-FIL x Ratio 14
PM25-PRI = PM-CON + PM25-FIL, Math Step 7
PM10-PRI = PM-CON + PM10-FIL, Math Step 5
PM10-FIL
PM10-PRI
PM25-FIL
PM-CON
PM25-PRI
PM-CON = PM10-PRI - PM10-FIL, Math Step 2
PM25-PRI = PM-CON + PM25-FIL, Math Step 7
PM10-FIL
PM10-PRI
PM25-PRI
PM-CON
PM25-FIL
PM-CON = PM10-PRI - PM10-FIL, Math Step 2
PM25-FIL= PM25-PRI - PM-CON, Math Step 6
February 01, 2006
85
-------
Table 20 continued. PM Cases and Required Steps to Augment PM
PM Reported
PIVI to Default
Augmentation Methodology
PM10-PRI
PM25-PRI
PM25-FIL
PM-CON
PM10-FIL
PM-CON= PM25-PRI - PM25-FIL, Math Step 3
PM10-FIL= PM10-PRI - PM-CON, Math Step 4
PM10-FIL
PM25-FIL
PM-CON
PM25-PRI
PM10-PRI
PM25-PRI = PM-CON + PM25-FIL, Math Step 7
PM10-PRI = PM-CON + PM10-FIL, Math Step 5
PM10-FIL
PM25-FIL
PM25-PRI
PM-CON
PM10-PRI
PM-CON= PM25-PRI - PM25-FIL, Math Step 3
PM10-PRI = PM-CON + PM10-FIL, Math Step 5
PM10-FIL
PM10-PRI
PM-CON
PM25-PRI
PM25-FIL
PM25-FIL = PM10-FIL x Ratio 1
PM25-PRI = PM-CON + PM25-FIL, Math Step 7
PM10-PRI
PM25-PRI
PM-CON
PM25-FIL
PM10-FIL
PM25-FIL= PM25-PRI - PM-CON, Math Step 6
PM10-FIL= PM10-PRI - PM-CON, Math Step 4
PM10-FIL
PM10-PRI
PM25-FIL
PM25-PRI
PM-CON
PM-CON= PM25-PRI - PM25-FIL, Math Step 3
PM10-FIL
PM10-PRI
PM25-PRI
PM-CON
PM25-FIL
PM25-FIL= PM25-PRI - PM-CON, Math Step 6
PM10-FIL
PM10-PRI
PM25-FIL
PM-CON
PM25-PRI
PM25-PRI = PM-CON + PM25-FIL, Math Step 7
PM10-PRI
PM25-FIL
PM25-PRI
PM-CON
PM10-FIL
PM10-FIL= PM10-PRI - PM-CON, Math Step 4
PM10-FIL
PM25-FIL
PM25-PRI
PM-CON
PM10-PRI
PM10-PRI = PM-CON + PM10-FIL, Math Step 5
Creation of PM Ratios
We developed the ratios 1 through 14 from the following sources.
PM calculator
February 01, 2006 86
-------
AP-42
2002 Electric Generating Unit (EGU) NEI data developed by EIAG and used to
supplement the state and local agency and tribal inventories.
Some of these ratios are SCC and control device specific. We will use a hierarchy when we use
the ratios. Unfortunately, in some cases, particularly for PM-CON, only generic ratios can be used.
Information about the ratios and the hierarchy of their use in PM Augmentation are shown below.
1. EGU Data
a. PM10 and PM25 primary and filterable
b. SCC and control device specific
c. Developed by EIAG using PM Calculator and AP-42 data
d. Applicable only to SCCs that start with 101 and 201
2. PM Calculator
a. Applicable only to filterable PM; Does not include PM-CON ratios
b. SCC and control device specific
3. AP-42
a. Only source of PM-CON ratios; Not limited to filterable PM
b. SCC specific and sometimes control device specific
c. Limited number of SCCs;
d. General/generic PM ratios based on SCC 1-digit least preferred
e. Developed using emission factors (preferred) or particle distribution
The file, PM Augmentation Ratios, contains the ratios and is available at at:
ftp://ftp.epa.gov/EmisInventorv/2002finalnei/documentation/augmentation point sector/ .
A detailed description of how we created PM ratios is presented below.
EGU Data
Ratios developed using EGU data are SCC and control device specific. We developed
ratios by using the available PM emissions in the EGU data disaggregated by SCC. We
then averaged ratios according to SCCs. However EGU ratios only applied to SCCs
starting with 101 and 201. Details on the development of the PM emission estimates for
EGUs can be found in the "Documentation for the 2002 Electric Generating Unit (EGU)
National Emissions Inventory (NEI)" (Eastern Research Group and E.H. Pechan, 2004).
The2002 NEI EGU Point Source Documentation can be found at:
ftp://ftp.epa.gov/EmisInventorv/2002finalnei/documentation/point.
• PM-Calculator
Ratios developed using the PM Calculator are both SCC and control device specific. The
PM Calculator is EPA's software for estimating various PM terms from other PM terms
(U.S. EPA, 2004a). The PM Calculator is applicable to point sources only, and requires
February 01, 2006
87
-------
the user to input uncontrolled emissions (either total PM-FIL or PM10-FIL) for each
source, the SCC and the control device, if any. The program then calculates controlled
emissions for PM25-FIL and PM10-FIL. This method is preferred because it is SCC and
control device specific. Because the PM Calculator is limited in the PM terms it can
calculate, PM ratios will have to be found using other sources of data.
The PM Calculator can only calculate controlled or uncontrolled PM10-FIL and PM25-FIL
using uncontrolled PM10-FIL or PM-FIL. The PM Calculator does not directly import NIF
files, so we developed generic ratios from the output file by inputting a simple dummy file.
We uploaded the dummy file, with all of the SCCs and control code device codes in the
2002 point source inventory, into the PM Calculator. Emissions of 100 tons were entered
as the emissions in the dummy file for simplicity. The PM Calculator then created an
output file that listed for SCC and control device codes the following information.
PM uncontrolled
PM10 uncontrolled
PM25 uncontrolled
PM10 controlled
PM25 controlled
PM10 control efficiency (CE)
PM25 CE.
Then we developed ratios by simply dividing the term needed by the term available.
To the extent possible, we developed PM-FIL ratios using the PM Calculator for all of the
SCCs and control device codes in the 2002 point source state, local, and tribal agency
submittals. When a ratio based on SCC and control device could not be developed using
PM Calculator, we used AP-42 emission factors or particle size distributions.
Emission Factors from AP-42
We developed PM ratios using the emission factors from the EPA's AP-42 (U.S. EPA,
2004b). There were, however, a few drawbacks associated with the use of AP-42 emission
factors in this manner. In some cases, the ash content or the sulfur content in the fuel used
in the process were needed to calculate the emission factors. Unfortunately, most of the
state, local, and tribal records did not include sulfur content or ash content.
To overcome this shortcoming, we used assumptions from the 1999 NEI for CAPs. (Strait
et al., 2003). For coal combustion, we assumed that the sulfur content was 1% and ash
content was 8% for SCCs that start with 1 and 2. We also assumed that PM-CON was only
controlled with flume gas desulfurization. To develop the PM ratios, we grouped PM
emission factors for various PM terms based on their SCCs. This required PM emission
factors for both the available PM and the needed PM term.
In some cases, ash content and sulfur content were not needed to develop PM ratios. We
developed PM ratios for these SCCs by looking for emission factors for the various PM
February 01, 2006
88
-------
terms available. However, there still were situations when there were not enough data, and
we had to use other surrogate data to develop PM ratios.
Cumulative Particle Distribution from AP-42
AP-42 also has particle distribution data for some SCCs based on the 6 and 8 digit level.
These are less specific than the ratios developed from the PM Calculator and AP-42
emission factors, and will only be used when there are no corresponding ratios from EGU
Data, PM Calculator, or AP-42 emission factors.
To develop these ratios, we made the following assumptions:
• Percent distribution less than or equal to 1 micron are PM-CON emissions
• Percent distribution less than or equal to 2.5 microns are PM25-PRI emissions
(since they would also include emissions less than or equal to 1 micron)
• Percent distribution less than or equal to 10 microns are PM10-PRI emissions
(since they would also include emissions less than or equal to 1 micron)
• The percent for PM-PRI will always be 100%
The data available from the AP-42 includes the percent distribution less than or equal to 1
micron, the percent distribution less than or equal to 2.5 microns, and the percent
distribution less than or equal to 10 microns. Therefore, the data were easily converted into
ratios. First, we labeled the percent distributions as PM-CON, PM25-PRI, and PM10-PRI.
Then we calculated PM-FIL, PM25-FIL, and PM10-FIL using the available terms. Finally,
we calculated the ratios by dividing the needed term by the available term. For example, if
PM-PRI is available and PM10-FIL is needed, you can simply multiply PM-PRI by the
ratio of PM10-FIL/PM-PRI, to calculate PM10-FIL.
Other SCC Emission Factors
For SCCs that start with 3 and 4, PM emissions can occur from fuel fired equipment,
process heaters, incinerators, and coating ovens. For SCCs that start with 3 or 4 and have
combustion sources, we developed surrogate ratios based on emission factors for industrial
external combustion boilers. These surrogates were matched on the fuel type used. Table
21 shows some of the SCCs that were used as surrogates for the SCCs that start with 402
and have combustion sources. For coating ovens, we assumed that natural gas is the fuel
used.
SCCs that start with 5, solid waste disposal, did not have the necessary emission factors (or
cumulative particle distribution data discussed below) in AP-42. Therefore, we used
average emission factors for external combustion sources and stationary internal
combustion sources. A list of these SCCs is provided in Table 22.
Table 21. PM Augmentation: SCCs Starting with "402" and the Surrogate SCCs
February 01, 2006 89
-------
see
SCCL3
SCCL4
SCC to Use
Description of SCC to Use
402008
Coating Oven - General
102006
Industrial External Fuel Combustion, Natural Gas
40200801
Coating Oven - General
General
102006
Industrial External Fuel Combustion, Natural Gas
40200803
Coating Oven - General
Baked > 175F **
102006
Industrial External Fuel Combustion, Natural Gas
40200810
Coating Oven - General
General
102006
Industrial External Fuel Combustion, Natural Gas
40200820
Coating Oven - General
Prime/Base Coat Oven
102006
Industrial External Fuel Combustion, Natural Gas
40200830
Coating Oven - General
Topcoat Oven
102006
Industrial External Fuel Combustion, Natural Gas
40200840
Coating Oven - General
Two Piece Can Curing Ovens:
General
102006
Industrial External Fuel Combustion, Natural Gas
40200845
Coating Oven - General
Three Piece Can Curing Ovens
102006
Industrial External Fuel Combustion, Natural Gas
40200870
Coating Oven - General
Color Coat Oven
102006
Industrial External Fuel Combustion, Natural Gas
40200898
Coating Oven - General
General
102006
Industrial External Fuel Combustion, Natural Gas
40200899
Coating Oven - General
See Comment **
102006
Industrial External Fuel Combustion, Natural Gas
40201001
Coating Oven Heater
Natural Gas
102006
Industrial External Fuel Combustion, Natural Gas
40201002
Coating Oven Heater
Distillate Oil
102005
Industrial External Fuel Combustion, Distillate
Oil
40201122
Fabric Coating/Printing
Fabric Print:Dryer: Fuel-fired
102006
Industrial External Fuel Combustion, Natural Gas
40204340
Fabric Coating, Dip Coating
Drying/Curing
102006
Industrial External Fuel Combustion, Natural Gas
40290013
Fuel Fired Equipment
Natural Gas:
Incinerator/Afterburner
102006
Industrial External Fuel Combustion, Natural
Gas
40290023
Fuel Fired Eauioment
Natural Gas: Flares
102006
Industrial External Fuel Combustion. Natural Gas
Table 22. PM Augmentation: SCCs Starting With "5" That Have Combustion Sources (Incinerators and Flares) That Can Use
Average Emission Factors from External Combustion Sources and Stationary Internal Combustion Sources
February 01, 2006
90
-------
see
SCCL2
SCCL3
SCCL4
501001
Government
Municipal Incineration
50100101
Government
Municipal Incineration
Starved Air: Multiple Chamber
50100102
Government
Municipal Incineration
Mass Burn: Single Chamber
50100103
Government
Municipal Incineration
Refuse Derived Fuel
50100104
Government
Municipal Incineration
Mass Burn Refractory Wall Combustor
50100105
Government
Municipal Incineration
Mass Burn Waterwall Combustor
50100106
Government
Municipal Incineration
Mass Burn Rotary Waterwall Combustor
50100107
Government
Municipal Incineration
Modular Excess Air Combustor
50100108
Government
Municipal Incineration
Fluidized Bed: Refuse Derived Fuel
50100410
Government
Landfill Dump
Waste Gas Destruction: Waste Gas Flares
50100411
Government
Landfill Dump
Waste Gas Destruction: Incinerators
50100505
Government
Other Incineration
Medical Waste Incinerator, unspecified type,
Infectious wastes only
50100506
Government
Other Incineration
Sludge
50100515
Government
Other Incineration
Sludge: Multiple Hearth
50100516
Government
Other Incineration
Sludge: Fluidized Bed
502001
Commercial/
Institutional
Incineration
50200101
Commercial/
Institutional
Incineration
Multiple Chamber
50200102
Commercial/
Institutional
Incineration
Single Chamber
50200501
Commercial/
Institutional
Incineration: Special Purpose
Med Waste Controlled Air Incin-aka Starved
air, 2-stg, or Modular comb
50200502
Commercial/
Institutional
Incineration: Special Purpose
Med Waste Excess Air Incin - aka Batch,
Multiple Chamber, or Retort
50200503
Commercial
/Institutional
Incineration: Special Purpose
Medical Waste Rotary Kiln Incinerator
50200504
Commercial/
Institutional
Incineration: Special Purpose
Medical Waste Incinerator, unspecified type
(use 502005-01, -02, -03)
50200505
Commercial/
Institutional
Incineration: Special Purpose
Medical Waste Incinerator, unspecified type,
Infectious wastes only
50200507
Commercial/
Institutional
Incineration: Special Purpose
VOC Contaminated Soil
50200515
Commercial/
Institutional
Incineration: Special Purpose
Sewage Sludge Incinerator: Multiple Hearth
50200518
Commercial/
Institutional
Incineration: Special Purpose
Sewage Sludge Incinerator: Single Hearth
Cyclone
50200601
Commercial/
Institutional
Landfill Dump
Waste Gas Flares ** (Use 5-01-004-10)
February 01, 2006
91
-------
Table 22. PM Augmentation: SCCs Starting With "5" That Have Combustion Sources
(Incinerators and Flares) That Can Use Average Emission Factors from External
Combustion Sources and Stationary Internal Combustion Sources
SCC
SCCL2
SCCL3
SCCL4
50300101
Industrial
Incineration
Multiple Chamber
50300102
Industrial
Incineration
Single Chamber
50300107
Industrial
Incineration
Trench Burner: Tires
50300112
Industrial
Incineration
Mass Burn Waterwall Combustor
50300501
Industrial
Incineration
Hazardous Waste
50300502
Industrial
Incineration
Hazardous Waste Incinerators: Fluidized
Bed
50300503
Industrial
Incineration
Hazardous Waste Incinerators: Liquid
Injection
50300506
Industrial
Incineration
Sludge
50300520
Industrial
Incineration
Sewage Sludge Incinerator: High Pressure,
Wet Oxidation
50300599
Industrial
Incineration
Fuel Not Classified
50300601
Industrial
Landfill Dump
Waste Gas Flares
50300789
Industrial
Liquid Waste
Sludge Digester Gas Flare
1st Digit SCC AP-42 Data
At the most generic level, we calculated ratios based on the 1st digit of the SCC if specific
SCC ratios could not be found. We developed these from AP-42 emission factors and
cumulative particle distribution. Again, ratios based on emission factors are preferred over
those based on cumulative particle distribution. These 1st digit SCC based ratios also have
very generic control efficiencies that can be applied. For this method, we used the same
assumptions and methods for the 6 and 8 digit SCCs. In addition, we also developed
multipliers that can be used to convert the uncontrolled emissions into controlled
emissions. We will only use these very generic ratios as a last resort to augment PM.
The EPA has revised the PM emissions from natural gas combustion in the final version of the
2002 NEI. The revision decreased the emissions of PM10-PRI, PM2.5-PRI, and PM-CON. The
reason for this adjustment is that EPA believes that the current AP-42 factors for condensible
emissions are too high. The EPA is basing this adjustment on some limited data from the draft
EPA dilution method that is similar to conditional test method (CTM) 39
(http://www.epa.gov/ttn/emc/ctm.html ) that measures PM10-PRI and PM2.5-PRI directly.
The data that this adjustment is based on can be found at:
http://www.nvserda.Org/programs/Environment/emepreports.asp#Fine Particulates .
The file, "Ratios to adjust PM values in NEI for natural gas combustion, August 2005", contains
the PM natural gas ratios by SCC and is available at:
February 01, 2006
92
-------
ftp://ftp.epa.gov/EmisInventorv/2002finalnei/documentation/augmentation point sector/
The Data Source codes for emissions that have PM augmentation are "A-R" and "A". "A" denotes
augmented PM data that used original reported PM in augmentation procedures. "A-R" denotes
augmented PM data that required the replacement of existing PM data submitted to NEI due to
inconsistency in the reporting of PM by dat submitters.
References
Eastern Research Group and E.H. Pechan, 2004. "Documentation for the 2002 Electric
Generating Unit (EGU) National Emissions Inventory (NEI)." Prepared for the Emission
Inventory Group, U.S. Environmental Protection Agency, Research Triangle Park, North Carolina.
Strait et al. 2003. PM Augmentation Procedures for the 1999 Point and Area Source NEI.
Presented at the 12th International Emission Inventory Conference "Emission Inventories -
Applying New Technologies," San Diego, CA. April 29 - May 1, 2003.
U.S. EPA, 2004a. PM Calculator. Emission Inventory Group, U.S. Environmental Protection
Agency, Research Triangle Park, NC. Accessed February 23, 2004.
U.S. EPA, 2004b. Compilation of Air Pollutant Emission Factors, AP-42, Fifth Edition, Volume I:
Stationary Point and Area Sources, http://www.epa.gov/ttn/chief/ap42/index.html
12.3 VOC and PM Augmentation from HAPs
We will compare reported PM and VOC emissions with sum of VOC HAPs and sum of PM HAPs
at the facility level. Facilities will be identified where the sum of VOC HAPs emissions is greater
than reported VOC emissions and the sum of PM HAPs emissions is greater than reported PM
emissions. We will use the NEI Pollutant Code Table to identify which HAPs are VOC and/or PM
constituents.
For facilities that have HAP VOC and HAP PM emissions but do not have VOC or PM emissions,
we will not augment CAP inventories to include the HAP VOC emissions and HAP PM emissions
in the 2002 NEI. This is an area of ongoing methodological development that will be addressed in
the2008 NEI.
If the sum of HAP VOC or the sum of HAP PM emissions are more than 20% greater than
reported VOC or PM emissions, we will compile a list of these facilities with their emissions data
for further QA. It is not known at this time if we will be able to resolve the discrepancy. At a
minimum, the table will need to contain the following information.
• Facility information
Source category information
• VOC emissions from NEI for CAPs
• PM emissions from NEI from CAPs
February 01, 2006
93
-------
• Sum of VOC HAPs emissions from NEI for HAPs
• Sum of PM HAPs emissions from NEI for HAPs
% difference between VOC emissions from NEI for CAPs and sum of VOC emissions
from NEI for HAPs
• % difference between PM emissions from NEI for CAPs and sum of PM emissions from
NEI for HAPs
Resolution if any
If VOC emissions are greater than HAP VOC emissions or PM emissions are greater than HAP
PM emissions, we will assume that VOC and PM emissions include all VOC HAP and PM HAP
emissions plus emissions of additional non-HAPs.
12.4 CAP Augmentation for Missing CAPs
When a data provider submits a partial submission of CAPs for a facility, we will determine if
emissions of other CAPs should be reported. If the data provider reports emissions of other CAPs
as "0" and indicates in the documentation that "0" means no emissions as opposed to being a
placeholder, we will not add missing CAPs. For missing CAPs, we will use the reported
Throughput and SCC with available emission factors to generate estimates for other CAPs. If
Throughput and SCCs are not reported, or emission factors are not available for a process, then we
will use the 2002 Preliminary NEI for CAP Point Source estimates released in January 2004 to
default emissions of missing CAPs.
12.5 HAP Augmentation from VOC and PM
We will augment HAPs emitted from boilers for facilities that have reported VOC or PM boiler
emissions, but no HAP boiler estimates. If throughput is reported, we will use emission factors and
reported throughput to generate HAP emissions. If throughput is not available, we will develop
ratios of HAP emission factors to CAP emissions factors and apply the ratios to reported CAP
emissions to develop HAP emission estimates.
Any HAP boiler emission estimates defaulted from VOC and PM emissions will be assigned a
data source code of "A-B".
12.6 HAP Performance Level
Data providers may report HAP emissions as actual, allowable, potential or maximum. In Trends
analyses, actual emissions are preferred whereas in risk assessments, maximum emissions are
preferred.
More than one performance level may be reported for each pollutant and process. For example, an
agency may reported actual and potential emissions of a single HAP from a process. Acceptable
NIF code values are:
01 Actual
February 01, 2006
94
-------
02 Allowable
03 Maximum
04 Potential
When we prepare files for various uses, we use a hierarchy to select only one type of emissions to
avoid double counting. For 2002 NATA modeling, the following hierarchy will be used.
Actual
Allowable
Potential
Maximum
If this field is invalid or missing for HAP emission records, we will default the reported emissions
as actual.
12.7 QC of HAP Category Group Emissions
After we have blended and merged point source data, we will evaluate emissions records
associated with facilities that have individual HAP compounds reported and HAP category groups
reported (e.g., where chromium and compounds and hexavalent chromium are reported for the
same facility and process). If duplicate emissions are identified, we will delete records with the
compound group emissions and retain speciated compounds. In cases where a single data source
reports emissions of individual compounds and compound group, the following actions will be
taken.
• Antimony:
1. If speciated Antimony emissions are greater than or equal to Antimony and Compounds
(NEI pollutant code 92) emissions, delete Antimony and Compounds record.
2. If Antimony and Compounds emissions are greater than speciated Antimony emissions,
delete speciated Antimony records.
3. If emissions are speciated and no Antimony and Compounds emissions are reported
(Antimony and Antimony Trioxide for example), keep all records.
• Arsenic:
1. If speciated Arsenic emissions are greater than or equal to Arsenic and Compounds (NEI
pollutant code 93) emissions, delete Arsenic and Compounds record.
2. If Arsenic and Compounds emissions are greater than speciated Arsenic emissions, delete
individual pollutant records.
3. If emissions are speciated and no Arsenic and Compounds emissions are reported(Arsenic
and Arsine for example), keep all records.
February 01, 2006
95
-------
• Beryllium:
1. If speciated Beryllium emissions are greater than or equal to Beryllium and Compounds
(NEI pollutant code 109) emissions, delete Beryllium and Compounds record.
2. If Beryllium and Compounds emissions are greater than speciated Beryllium emissions,
delete speciated Beryllium records.
3. If emissions are speciated and no Beryllium and compounds emissions are reported
(Beryllium and Beryllium oxide for example), keep all records.
Cadmium:
1. If speciated Cadmium emissions are greater than or equal to Cadmium and Compounds
(NEI pollutant code 125) emissions, delete Cadmium and Compounds record.
2. If Cadmium and Compounds emissions are greater than speciated Cadmium emissions,
delete Cadmium record.
3. If emissions are speciated and no Cadmium and compounds emissions are reported
(Cadmium and Cadmium oxide for example), keep all records.
• Chromium:
A. If Chromium VI (NEI pollutant code 18540299) and Chromium III (NEI pollutant code
16065831) emissions are greater than or equal to Chromium and Compounds (NEI
pollutant code 136) emissions, delete Chromium and Compounds record.
Cr VI + Cr III > Cr and Compounds; delete Cr and Compounds
B. If Chromium VI and Chromium III emissions are greater than or equal to Chromium (NEI
pollutant code 7440473) emissions, delete Chromium record.
Cr VI + Cr III > Cr; Delete Cr
C. If emissions are speciated and no Chromium and compounds emissions are reported
(Chromium VI and Chromium III for example), keep all records.
No revision
D. If Chromium and Chromium VI emissions are reported and Chromium VI emissions are
less than Chromium emissions, generate Chromium III emission record by subtracting
Chromium VI emissions from Chromium emissions. Delete Chromium record.
Cr - CrVI = CrIII; delete Cr and add CrIII
E. If Chromium and Compounds and Chromium VI emissions are reported and Chromium VI
emissions are less than Chromium and Compounds emissions, generate Chromium III
emission record by subtracting Chromium VI emissions from Chromium and Compounds
emissions. Delete Chromium and Compounds record.
Cr and Compounds - CrVI = CrIII; delete Cr and Compounds and add CrIII
February 01, 2006
96
-------
F. If Chromium and Chromium III emissions are reported and Chromium III emissions are
less than Chromium emissions, generate Chromium VI emission record by subtracting
Chromium III emissions from Chromium emissions. Delete Chromium record.
Cr - CrIII = CrVI; delete Cr and add CrVI
G. If Chromium and Compounds and Chromium III emissions are reported and Chromium III
emissions are less than Chromium and Compounds emissions, generate Chromium VI
emission record by subtracting Chromium III emissions from Chromium and Compounds
emissions. Delete Chromium and Compounds record.
Cr and Compounds - CrIII = CrVI; delete Cr and Compounds and add CrVI
Cobalt:
1. If speciated Cobalt emissions are greater than or equal to Cobalt and Compounds (NEI
pollutant code 139) emissions, delete Cobalt and Compounds record.
2. If Cobalt and Compounds emissions are greater than speciated Cobalt emissions, delete
speciated Cobalt records.
3. If emissions are speciated and no Cobalt and compounds emissions are reported(Cobalt and
Cobalt oxide for example), keep all records.
Cresols:
In the 2002 NEI, data providers either reported the 3 isomers of Cresols (ortho-cresol, NEI
pollutant code 95487; meta- cresol, NEI pollutant code 108394; and para-cresol, NEI pollutant
code 106445) or they reported Cresols as a group (NEI pollutant code 1319773). The toxicity is
the same for all isomers or cresol and cresylic acid.
1. If speciated Cresol emissions are greater than or equal to Cresols emissions, delete Cresols
records.
2. If Cresols emissions are greater than individual isomers of Cresol emissions, delete
individual isomer Cresol records. This may happen if Cresols are reported and only one or
two isomers of cresols are reported.
3. If emissions are speciated and no Cresols emissions are reported, keep all records.
Cyanide:
1. If speciated Cyanide emissions are greater than or equal to Cyanide and Compounds (NEI
pollutant code 144) emissions, delete Cyanide and Compounds record.
2. If Cyanide and Compounds emissions are greater than speciated Cyanide emissions, delete
speciated Cyanide records.
3. If emissions are speciated (Cyanide and Hydrogen Cyanide for example), keep all records.
February 01, 2006
97
-------
Glycol Ethers:
In our analysis of point sources, glycol ethers emissions were always much greater than individual
ethers reported for same emission release point records. We assumed data providers did not
double count individual ethers and glycol ether group. Glycol ethers all have same toxicity which
is not very high.
The procedure for reconciling glycol ethers emissions includes the following steps.
A. If speciated glycol ether emissions are greater than or equal to Glycol Ethers (NEI pollutant
code 171) emissions, delete Glycol Ethers records.
B. If Glycol Ethers emissions are greater than speciated Glycol Ethers emissions, keep all
records.
C. If emissions are speciated and no Glycol Ethers emission are reported, keep all records.
Lead:
1. If speciated Lead emissions are greater than or equal to Lead and Compounds (NEI
pollutant code 195) emissions, delete Lead and Compounds record.
2. If Lead and Compounds emissions are greater than speciated Lead emissions, delete
speciated Lead records.
3. If emissions are speciated and no Lead and compounds emissions are reported (Lead and
Lead Oxide for example), keep all records
Manganese:
1. If speciated Manganese emissions are greater than or equal to Manganese and Compounds
(NEI pollutant code 198) emissions, delete Manganese and Compounds.
2. If Manganese and Compounds emission are greater than speciated Manganese emissions,
delete speciated Manganese records.
3. If emissions are speciated and no Manganese and compounds emissions are
reported(Manganese and Manganese Trioxide for example), keep all records.
• Mercury:
1. If speciated Mercury emissions are greater than or equal to Mercury and Compounds (NEI
pollutant code 199) emissions, delete Mercury and Compounds record.
2. If Mercury and Compounds emissions are greater than speciated Mercury emissions, delete
speciated Mercury records.
3. If emissions are speciated and no Mercury and Compounds emissions are reported (Mercury
and Mercuric Oxide for example), keep all records.
February 01, 2006 98
-------
4. If Mercury (NEI pollutant code 7439976) emissions, Mercury (organic) (NEI pollutant code
22967926) emissions and Mercury and Compounds emissions are all equal, delete Mercury
(organic) emissions and Mercury and Compounds records.
• Nickel:
1. If speciated Nickel emissions are greater than or equal to Nickel and Compounds (NEI
pollutant code 226) emissions, delete Nickel and Compounds record.
2. If Nickel and Compounds emissions are greater than speciated Nickel emissions, delete
speciated Nickel records.
3. If emissions are speciated and no Nickel and compounds emissions are reported(Nickel and
Nickel Subsulfide for example), keep all records.
POM:
In the 2002 NEI, data providers report POM for a variety of PAHs. Please refer to the NEI
Pollutant code table for a complete list of POM compounds. The NEI Pollutant Code Table is
contained in the 2002 State Lookup File, 02nei_lkup_states.zip, found at:
ftp://ftp.epa.gov/EmisInventorv/2002finalnei/documentation/point/ .
1. If speciated PAH emissions are greater than or equal to POM (NEI pollutant codes
130498292, 249, 246, 75, 40, or 284) emissions, delete POM records.
2. If POM emissions are greater than speciated PAH emissions, delete speciated PAH records.
3. If emissions are speciated and no POM emissions are reported (benzoapyrene and
anthracene for example), keep all records.
Selenium:
1. If speciated Selenium emissions are greater than or equal to Selenium and Compounds (NEI
pollutant code 253 emissions, delete Selenium and Compounds record.
2. If Selenium and Compounds emissions are greater than speciated Selenium emissions,
delete speciated Selenium records.
3. If emissions are speciated and no Selenium and compounds emissions are
reported(Selenium and Selenium oxide for example), keep all records.
Xylenes:
In 2002 NEI, data providers either reported emissions of the 3 isomers of Xylenes (ortho- xylene,
NEI pollutant code 95476; meta-xylene, NEI pollutant code 108383; and para-xylene, NEI
February 01, 2006
99
-------
pollutant code 106423) or they reported Xylenes as a group (NEI pollutant code 1330207). The
toxicity is the same for all isomers or xylenes.
1. If individual Xylene isomer emissions are greater than or equal to Xylenes emissions, delete
Xylenes records.
2. If Xylenes emissions are greater than the sum of individual isomers of Xylene emissions,
delete isomers of Xylene records.
3. If emissions are speciated and no xylenes emissions are reported, keep all records.
When we prepare files for various uses, we will use the Pollutant Rank data field in the NEI
Pollutant Code Table to select pollutants and emissions within a HAP Category where emissions of
both individual compounds within a group and a compound group are reported for an emission
record.
Any Chromium VI or Chromium III emission estimates that we have created will be assigned a data
source code of "A-C".
12.8 Emissions Type Period
The 2002 NEI will contain annual emission estimates for HAPs and CAPs and non-annual
emissions if reported for CAPs. If non-annual emissions are reported and annual emissions are not
reported, we will generate an annual emission estimate.
The Emissions Type Period in the NOF EM Table distinguishes between annual and non-annual
emission estimates in the 2002 NEI.
We will use the Emission Type, Start Date, and End Date to determine the Period Type. In
determining if an emission is annual or non-annual, we will use the following procedure.
~ If there is only one emissions record for a given pollutant for a unique emission release
point (tribal code, state and county FIPS code, site id, unit id, process id, emissions release
point id), we will base the Emissions Type Period on the Emission Type and period of
emissions.
If the Emissions Type is 30 and the Start Date is 01012002 and End Date is
12312002, the Emissions Type Period is annual.
If the Emission Type is 30 and emissions are reported for a period of a season or
less, then the Emissions Type Period is assumed to be non-annual.
If the Emission Type is 30 and emissions are reported for a period greater than a
season, then the Emissions Type Period is assumed to be annual.
If the Emission Type is 20 - 29, the Emissions Type Period is non-annual.
February 01, 2006
100
-------
~ When multiple records with Emission Type 30 and other Emission Types are reported
for a unique emission release point, we will evaluate the Emission Types and Start and End
Dates. If an Emission Type of 30 is reported and the Start and End Dates are 20020101 and
20021231, we will assume the Emissions Type Period is annual for this record and the
Emissions Type Period of other reported records is non-annual. In most cases in which the
data provider supplied a partial year (ozone season) in the 1999 NEI, they also supplied an
annual record.
~ When multiple records with Emission Type 30 and multiple Start and End Dates are
reported for a unique emission release point, we will evaluate the Start and End Dates. If
the Start and End Dates for a record are 20020101 and 20021231, we will identify the
Emissions Type Period as annual for this record and non-annual for all other records. When
emissions are provided for each season and no annual estimate is provided, the Emissions
Type Period is non-annual for each of the records.
12.9 Annual Emissions
The 2002 NEI will contain annual emission estimates for HAPs and CAPs. We will retain the
reported emissions and units in the NIF tables.
We will use the methodology in Table 23 to calculate the Annual Emissions. We will only use one
methodology for each emission release point. We will give preference to complete years, type 30.
In most cases in which a state supplied a partial year estimate (e.g., ozone season), they also
supplied an annual emission estimate. However, if only a partial year estimate is supplied, it will
be retained in the annual file and expressed as annual emissions. Retaining this partial year
estimate, prevents the loss of the emissions data from the annual file.
Table 23. Methodology Used to Generate Annual Emissions
Period
Type
K mission
Type
.Methodology
Annual
30
Use reported emissions
February 01, 2006
101
-------
Table 23, continued. Methodology Used to Generate Annual Emissions
Period
"I'vpc
K mission
Type
Methodology
Non-
Annual
30
All Individual Periods are present for a year; or
If all individual periods are not present for a year and the data provider
indicates Non-Annual Emissions for reported periods are the same as
Annual Emissions.
Annual Emissions = Sum of Numeric Emissions Value of each non-
overlapping period
Ex. Emissions are reported for each of the 12 months,
Annual Emissions = Sum of monthly emissions
Non-
Annual
30
Start Date and End Dates in Period are Not 20020101 and 20021231
Data provider indicates Non-Annual Emissions are not the Annual
Emissions, but can be scaled to the entire year:
~ Start Time = 0000 and End Time = 2359 or Start and End
Times are missing
Annual emissions = Emissions Numeric Value * (365/ Number
of Days in Period)
~ Start and End Times are for less than a 24 hr period
Annual Emissions = Numeric Emissions * (365/Number of
Days in Period) * (Hrs/Day)
If the data provider cannot be reached, set Annual emissions =
Emissions Numeric Value.
Non-
Annual
29
Data provider indicates Non-Annual Emissions are the same as Annual
Emissions.
~ Start Time = 0000 and End Time = 2359 or Start and End Times are
missing
Annual Emissions = Numeric Emissions * (Number of Days/Period)
~ Start and End Times are for less than a 24 hr period
Annual Emissions = Numeric Emissions * Number of Days/Period)
* (Hrs/Day)
February 01, 2006
102
-------
Table 23 continued. Methodology Used to Generate Annual Emissions
Period
"I'vpc
K mission
Type
Methodology
Non-
Annual
29
Start Date and End Dates in Period are Not 20020101 and 20021231
Data provider indicates Non-Annual Emissions are not the Annual
Emissions, but can be scaled to the entire year:
~ Start Time = 0000 and End Time = 2359 or Start and End
Times are missing
Annual emissions = Emissions Numeric Value * (365/ Number
of Days in Period)
~ Start and End Times are for less than a 24 hr period
Annual Emissions = Numeric Emissions * (365/Number of
Days in Period) * (Hrs/Day)
Ex. Emissions are for 20020501 - 20020831 for 24 hr/day
Annual Emissions == Emissions (365/123)
If the data provider cannot be reached, set Annual Emissions =
Numeric Emissions * (Number of Days/Period).
Non-
Annual
27
(average
weekday)
Data provider indicates Non-Annual Emissions are the same as
Annual Emissions.
~ Start Time = 0000 and End Time = 2359 or Start and End
Times are missing
Annual Emissions = Numeric Emissions * (Number of
Days/Period) * 5/7
~ Start and End Times are for less than a 24 hr period
Annual Emissions = Numeric Emissions * Number of
Days/Period) * (Hrs/Day) * 5/7
February 01, 2006
103
-------
Table 23 continued. Methodology Used to Generate Annual Emissions
Period
"I'vpc
K mission
Type
.Methodology
Non-
Annual
28
(average
weekend)
Data provider indicates Non-Annual Emissions are the same as
Annual Emissions.
~ Start Time = 0000 and End Time = 2359 or Start and End
Times are missing
Annual Emissions = Numeric Emissions * (Number of
Days/Period) * 2/7
~ Start and End Times are for less than a 24 hr period
Annual Emissions = Numeric Emissions * Number of
Days/Period) * (Hrs/Day) * 2/7
12.10 Non-Annual Emissions
We will not retain non-annual emissions in the final NEI from data submitters unless seasonal
throughput percentages are included in NIF submission. EPA will not augment for invalid or
missing non-annual emissions in the NEI.
We will QA emissions for emission release point records that contain both annual and non-annual
emission estimates to identify cases where non-annual emissions exceed annual emission values.
Because emissions may be reported and/or calculated for different time periods, e.g., annual, daily,
seasonal, etc., and for the same or different pollutants, verification is needed to ensure that non-
annual emissions are not higher than annual emissions. We will delete non-annual emission records
where non-annual emissions exceed annual emissions.
12.11 Emission Ton Value
We will convert emissions to tons if reported in other units and report the ton value in the Emission
Ton Value field in the NOF EM Table. Table 24 shows the conversions factors that we will use to
generate emissions in tons.
Table 24. Unit Conversions for Tons
Reported Unit
Conversion Factors for Tons
E3TON
1000
MILLIGRAM
1.1025E-09
February 01, 2006
104
-------
Reported I nil
Conversion Ksiclors lor I ons
G
1.1025E-06
KG
0.0011025
LB
0.005
TON
1
UG
1.1025E-12
E2LB
0.05
NG
1.1025 E-15
TONNE
1.1025
12.12 Emission Content QC
After we have compiled a draft point source inventory, we will conduct a variety of emissions
content QC on an internal draft inventory prior to preparing the draft inventory files for external
review. The purpose of Emissions Content QA/QC is to identify and correct erroneous emissions
and add missing emissions. We will identify potential problems including:
Outliers
Missing facilities
Missing pollutants from facilities
Pollutants that should not be reported for facilities
Our Emissions Content QA/QC will consist of both Comparative and Statistical QC. The
Comparative QC includes the evaluation of significant changes between the 1999 NEI and 2002
NEI data, and/or extreme variation within the 2002 data. We will compare 1999 emission estimates
to 2002 estimates by pollutant for each facility, source category, county, and state. These big
picture summaries will highlight source categories, states, and facilities with potential problems.
Statistical QC includes using statistical tables to identify top emitters for each pollutant
nationwide, ranking each facility based on its emissions of each pollutant on a national basis and
listing the top emitters for pollutant/source category combination nationwide. Comparative and
Statistical QC will be used to identify outliers, missing pollutants from facilities and erroneous
pollutants reported for facilities. Comparative QC will be used to identify missing facilities.
Specific Comparative QC activities will include:
"Top 10" lists by pollutant and source category - include highest emitting facilities and
source categories nationally and within a state and county or region;
Emissions ranked lists and maps by pollutant - include counties within a state ranked by
pollutant and source category emissions;
Previous Year Emissions Comparison - compare pollutant emissions nationally and for
state, county, source category and facility from 1999 NEI with the draft 2002 NEI; and
February 01, 2006
105
-------
99-2002 Site List - provides a list of sites found in the 1999 base year NEI, sites in the 2002
draft NEI, and sites common to both the 1999 and 2002 inventories.
Specific Statistical QC activities will include:
Statistical tables (min, max, average, standard deviation, and percentile values) by source
category - compare pollutant specific statistical values nationally and for state to individual
counties for source categories and individual facilities;
Emissions distribution by pollutant - prepare source category, national, county level, and
state level emissions values frequency histogram;
Contribution of facilities and source categories to multiple pollutants - compare list of
pollutants emitted from a source category to individual facilities; and
Pollutant/Source Category Crosswalk Comparison - compare emissions of pollutants to
Pollutant/Source Category crosswalks generated from the 1999 NEI..
Outliers are often difficult to spot - what may appear to be a high emissions value may in fact be
acceptable for a particular facility or source category. To aid in detecting these errors, we will use
both comparative and statistical QC activities. We will compare emissions data to the range of
values in the draft 2002 NEI and the percent contribution to total emissions. A summary table with
the list of facilities that appear multiple times as top emitters for different pollutants will help
identify point sources with outliers. These high values may be due to a series of outliers or
duplicated emission records. The high emissions may also be correct for that facility and category.
Thus, these summary data will need to be closely reviewed before any records are deleted. In some
cases, the state/local agency submitting the data will be contacted to discuss the quality of the
estimates.
To identify missing facilities, we will use the 99-2002 Site List. By sorting this list by state,
county, and facility name, we will evaluate the sites listed in each county and detect missing
facilities. If a site is found in the 1999 version but is not in the 2002 draft, reviewers can verify that
the facility closed or supply data for the missing facility during the draft review period.
We will use identify missing and erroneous pollutants for individual facilities using the
pollutant/source category crosswalk tables and statistical tables. We prepared pollutant/source
category crosswalk tables to QA pollutants emitted from source categories. For CAPs, the
crosswalk identifies source categories by SCC. For HAPs, the crosswalk identifies source
categories by MACT Code, SIC Code, and SCC. We will explore expanding the crosswalk table to
include emissions ranges. The two pollutant source category crosswalk files, HAP Source Category
Crosswalk and CAP Source Category Crosswalk, are available at:
ftp://ftp.epa.gov/EmisInventorv/2002finalnei/documentation/augmentation point sector/ . The
HAP Source Category Crosswalk contains 3 tables: Facility MACT, Facility SCC, and Facility SIC.
The CAP Source Category Crosswalk contains 1 table - Facility SCC. The following data fields are
included in the pollutant/source category crosswalk tables.
Source Category - MACT Code, SCC, or SIC Code
Pollutant - HAP Category Name for HAPS or Pollutant Name for CAPs
February 01, 2006
106
-------
Number of Facilities - number of facilities reporting emissions of pollutant for a source
category
Number of Total Facilities - number of total facilities for a source category
% Facilities - % of Number Facilities reporting emissions of pollutant/Total Number of
Facilities for a source category
• Max oF Emissions(TPY) - maximum emissions for facility emitting a pollutant within the
category
• Min OF Emissions(TPY) - minimum emissions for facility emitting a pollutant within the
category
99 Total Emissions(TPY) - total emissions for pollutant from source category
Emission Content QC Report
We will prepare a Emission Content QC report for each state and tribe. The report will compare
2002 emissions data to 1999 emissions data and highlight potential outliers, missing facilities,
missing pollutants and erroneous pollutants. We will send these reports to the data providers and
also post the reports during the review period of the draft 2002 NEI. We will ask data providers to
review the reports and provide revisions to us. The following tables will be included in the state
emissions QC files.
• Missing Facility Report - facilities present in 1999 NEI but missing from 2002 NEI
Missing County Report - counties with facilities present in 1999 NEI but missing from 2002
NEI
Counties with No Data - No facilities were present in these counties in 1999 NEI and 2002
NEI
Missing CAPs at State Level - CAPs missing from 2002 state inventory
Missing Facility CAPs- CAPs present for these facilities in 1999 NEI but missing from
2002 NEI
• 1999 and 2002 Facility CAP Comparison - comparison of emissions at facility level in 1999
NEI and 2002 NEI
• 1999 and 2002 Facility HAP Comparison - comparison of emissions at facility level in
1999 NEI and 2002 NEI
1999 and 2002 State CAP Comparison - comparison of emissions at state level in 1999 NEI
and 2002 NEI
1999 and 2002 State HAP Comparison - comparison of emissions at state level in 1999 NEI
and 2002 NEI
State Facility CAP Counts - Comparison of number of facilities emitting CAPs in 1999 NEI
and 2002 NEI
• State Facility HAP Counts - Comparison of number of facilities emitting HAPs in 1999 NEI
and 2002 NEI
Emissions Percentage CAPs - Facilities where % of total state emissions of a pollutant
(facility emissions/state point source emissions) are greater than 25%
Emissions Percentage HAPs - Facilities where % of total state emissions of a pollutant
(facility emissions/state point source emissions) are greater than 25%
Top 10 Facilities of CAPs
• Top 10 Facilities of HAPs
February 01, 2006
107
-------
Data from the 2002 NEI Facility File - to assist in identifying locations and facility IDs
Other Summary Files to Assist in Emission Content QC
The following emission summary files will also be compiled from the draft 2002 NEI to help
facilitate QC of the NEI, and to help put the emission estimates into perspective by state, county,
source category, and facility.
All NEI
HAPs, Annual
CAPs Annual
CAPs Non Annual
NEON HAP Summary Tables
NEI Emissions History Table (See Section 12.15)
For NEI point source data users who are interested in more than one state data set, three ALL NEI
summary files will be available that combine key data elements from all of the state and tribal data
sets in a single database.
• ALL_NEI_CAP_ANNUAL - This file will contain annual emissions data for the 2002 NEI
for CAPs. Emissions data will be presented in tons/year. However, records will retain their
original start/end dates and emission types for reference.
• ALL_NEI_CAP_NONANNUAL - This file will contain nonannual emissions data for the
2002 NEI for CAPs. This file will contain the non-annual emissions records supplied by
state and local agencies and tribes. The emissions will be presented in their original units
and will not be converted to tons.
• ALL_NEI_HAP_ANNUAL - This file will contain annual emissions data for the 2002 NEI
for HAPs. Emissions data will be presented in tons/year. However, records will retain their
original start/end dates and emission types for reference.
Table 25 contains the table structure for the ALL NEI files.
Table 25. Table Structure for ALL NEI
Name
CAP
HAP
POP FLAG
X
X
EPARegion
X
X
strTribalCode
X
X
Tribal_Name
X
X
STABBR
X
X
Cnty_Name
X
X
str StateCountyFIP S
X
X
February 01, 2006
108
-------
Name
CAP
HAP
strFacilityRegi stryldentifier
X
X
strStateF acilityldentifier
X
X
strNEI Unique Facility ID
X
X
strFacilityName
X
X
strSICPrimary
X
X
strNAIC SPrimary
X
X
ergNAICSflag
X
X
strEmissionUnitID
X
X
strProcessID
X
X
strSCC
X
X
strEmissionReleasePointID
X
X
strEmissionReleasePointType
X
X
IngStartDate
X
X
IngEndDate
X
X
strEmissionType
X
X
strPollutantCode
X
X
PollutantCodeDesc
X
X
CAP_Category_Name
X
H AP_C ategory _Nam e
X
Emissions(TPY)
X
X
Emissions(TPY)-TEXT
X
X
strControl Status
X
X
strLocationAddress
X
X
strCity
X
X
strState
X
X
strZipCode
X
X
sngStackHeight
X
X
sngExitGasT emperature
X
X
sngStackDiameter
X
X
sngExitGas V el ocity
X
X
February 01, 2006
109
-------
Table 25 continued. Table Structure for ALL NEI
Name
CAP
HAP
sngExitGasFlowRate
X
X
ergStackDefaultFlag
X
X
ergDataSource
X
X
strHAPEmissionsPerformanceLevel
X
strCAPEmissionsPerformanceLevel
X
dblXCoordinate
X
X
dblYCoordinate
X
X
strHorizontalCollectionMethodCode
X
X
strHorizontalAccuracyMeasure
X
X
strHorizontalReferenceDatumCode
X
X
strReferencePointCode
X
X
str SourceMap S cal eNumb er
X
X
strCoordinateDataSourceCode
X
X
ergLocati onD efaultFl ag
X
X
The NEON HAP files contain emission summaries for each 188 HAP category, the sum of the 188
HAPs, and the sum of the 33 urban HAPs. Emissions are also presented for each individual HAP
species in all files except for the county emission summary, the source category summary and the
point source facility summary files. The data structure for the NEON summary files is in the file,
neon_table_specs092003.xls, and is available at:
ftp://ftp.epa.gov/pub/EmisInventorv/finalnei99ver3/haps/summaries/. The 7 NEON HAP files
include the following summaries.
• County Emission Summary - The county emission summary presents the NEI HAP
emissions by state, and county for major, area, onroad, and nonroad sources. Major and area
sources are also summarized as MACT vs. non-MACT source categories.
• Source Category Summary - The source category summary presents the NEI HAP
emissions by state, and county for major, area, onroad, and nonroad sources. The area
sources are delineated as point or nonpoint. Each stationary source category is presented by
MACT code, SIC Code, or just source category name if there is no applicable MACT or
SIC code.
• Point Source Facility Summary -The point source facility summary presents the NEI HAP
point source emissions by NEI Unique Facility ID and individual site ID for major and area
February 01, 2006
110
-------
point sources. Included with each facility record is the address, site latitude/ longitude,
emission type (actual, allowable, potential, etc.), MACT and/or SIC code, the source of the
emission estimate.
• Point Source Stack Summary - The point source stack summary presents the NEI HAP
emissions by NEI Unique Facility and individual site for major and area point sources.
Included with each record is the emission type (actual, allowable, potential, etc.), emission
unit ID, process ID, emission release point ID, SCC, MACT and/or SIC code, emission
release point type (stack/vent or fugitive), and latitude/longitude of the emission release
point and the source of the emission estimate.
• Nonpoint Stationary Source Summary - The nonpoint stationary source summary
presents the NEI HAP emissions by state, county, and area source category. Included with
each record is the emission type (actual, allowable, potential, etc.), SCC, MACT, and/or SIC
code.
• Nonroad Summary - The nonroad summary presents the NEI HAP emissions by state,
county, and aggregated nonroad source categories.
• Onroad Summary - The onroad summary presents the NEI HAP emissions by state,
county, and aggregated nonroad source categories.
Toxicity Weighting
Toxicity weighting of HAP emissions is a useful tool to identify potential outliers and problems
with HAP emissions. Emissions of HAPs can be toxicity weighted for cancer and noncancer
effects. The National Air Toxics Assessment contains more information on risk characterization at
the following address, http://www.epa.gov/ttn/atw/natal999/riskbg.html .
Exposure modeling should be conducted to provide an accurate risk characterization. Toxicity
weighting of emissions provides a quick and less accurate relative ranking of emissions by risks.
Toxicity weighting does not account for dispersion and transport of HAPs, secondary formation of
HAPs, or exposure within micro-environments to HAPs. Because toxicity weighting does not have
units, the results are dimensionless and only a relative ranking of pollutants is possible using this
approach. We toxicity weighted draft NEI HAP emissions to identify facilities with the largest risk
in order to conduct additional QC.
The methodology to toxicity weight emissions data includes the following steps.
1. For all individual NEI pollutants except Chromium and Compounds (pollutant
codel36) and Chromium (pollutant code 7440473): Multiply emissions by Metal CN
Speciation Factor to extract metal and cyanide mass for toxicity weighting. The Metal CN
Speciation Factors are found in the file, "Tox wt factors for 2002 NEI 06282006", available
at: ftp://ftp.epa.gov/EmisInventorv/2002finalnei/documentation/point/ . Note: Some NEI
HAPs are speciated into more than one metal or into a metal and cyanide (CN). The
February 01, 2006
111
-------
Metal_CN Speciation factors were developed by using the molecular weight of the
pollutants to extract only the mass of metal element or HCN from compounds. The toxicity
weighting factors for metals are assigned to elemental metal and the toxicity weighting
factors for cyanides are assigned to hydrogen cyanide (HCN). NEI estimates for metals
must first be multiplied by the Metal CN Speciation Factor to provide emission estimates
for toxicity weighting. For all non-metal and non-cyanide compounds, the metal speciation
factor is 1. In addition, the table groups NEI POM compounds into the 8 POM groups that
have toxicity factors.
2. For NEI pollutants Chromium and Compounds (pollutant codel36) and Chromium
(pollutant code 7440473): Use chromium speciation file, "Cr Speciation 06282006.mdb"
available at: ftp://ftp.epa.gov/EmisInventorv/2002finalnei/documentation/point/
to speciate source category emissions into hexavalent chromium (Cr(VI)) and trivalent
chromium (Cr(III)) emissions. Cr(VI) emissions are generated by multiplying Chromium
and Compounds (pollutant codel36) and Chromium (pollutant code 7440473) by the
Hexavalent Chromium Speciation Multiplier; and Cr(III) emissions are generated by
multiplying Chromium and Compounds (pollutant code 13 6) and Chromium (pollutant code
7440473) by the Trivalent Chromium Speciation Multiplier. Emissions by source category
are needed in order to use chromium speciation file. In order to toxicity weight chromium
emissions, all emissions must be allocated to Cr(VI) and Cr(III). The following hierarchy
is used to assign default speciation profiles to chromium source categories.
A. Use MACT Code default speciation profiles first
B. Use SCC speciation profiles if MACT code default speciation profiles are
unavailable. NOTE nonroad SCCs are truncated in table.
C. Use SIC Code speciation profiles if MACT code and SCC default speciation
profiles are unavailable.
D. Use default speciation multiplier of 0.34 for Cr(VI) and 0.66 for Cr(III) if
MACT Code, SCC and SIC Code default speciation profiles are unavailable.
3. For Cancer Tox weighting: Multiply emissions generated from steps 1 and 2 by the Unit
Risk Estimate (URE) found in the file, "Tox wt factors for 2002 NEI 06282006".
NOTE: Cancer risks between pollutants are additive. Cancer and NonCancer risks are not
additive.
4. For NonCancer Tox weighting: Divide emissions generated from steps 1 and 2 by the
Reference Concentration (RfC) found in the file, "Tox wt factors for 2002 NEI 06282006".
NOTE: NonCancer risks between pollutants are additive. Cancer and NonCancer risks are
not additive.
An example toxicity weighted emissions calculation is shown below..
NEI: 1 ton of lead arsenate is emitted by a facility.
February 01, 2006
112
-------
Metal_CN Speciation Factors: 0.2146 Arsenic and 0.5935 lead
Lead emissions for toxicity weighting: 1 ton x 0.5935 = 0.5935 tons lead
Arsenic emissions for toxicity weighting: 1 ton x 0.2146 = 0.2146 tons arsenic
UREs:
RfCs:
Arsenic 0.0043
Arsenic 0.00003; Lead 0.0015
Cancer Risk:
Noncancer Risk:
Arsenic: 0.2146 x 0.0043 = 0.0092278
Arsenic: 0.2146/0.00003 = 7153.3
Lead: 0.5935/0.0015 = 395.6
From this example, while the arsenic emission are smaller than lead, the toxicity weighted
emissions results in greater risks for arsenic than lead.
12.13 Emissions Data Quality Rating for HAPs
The Data Rating will help to provide users of the NEI some sense of the reliability of an emission
estimate. We developed a simplified rating scheme so that a qualitative grade from A to E can be
assigned to each HAP estimate. This enhances the transparency of the data and also satisfies the
requirement that EPA perform an assessment of the data. The scoring may help us determine which
data point to retain in the final 2002 NEI when there are multiple estimates for the same facility. It
may also help us to understand which source categories need improvement.
This rating scheme will not be in-depth as other systems such as DARS (Data Attribute Rating
System), but will consider the following factors in assigning a score.
The Emissions Data Quality Rating is based on the following factors..
Completeness of Data
Emission Calculation Methods
Age of Data
Emission Data Level
Specificity of Data
To evaluate these six inventory factors, we will first extract the following data fields from NOF to
make a single table:
State County FIPS
Tribal Code
State Facility Identifier
Emission Unit ID
Process ID
Pollutant Code
HAP Category Name
Emission Release Point ID
see
February 01, 2006
113
-------
Start Date
End Date
Emission Type
Data Source
Emissions (TPY)
Control Status
EM Reliability Indicator
Factor Numeric Value
Factor Unit Numerator
Factor Unit Denominator
Emission Calculation Method Code
EF Reliability Indicator
Emission Data Level
see
Actual Throughput
Throughput Unit Numerator
The methodology for assigning a rating is described below.
• Completeness of Data
This factor addresses whether the submitter provided enough information to enable the reviewer to
repeat the calculation, assess emission factors and/or calculation methods. The fields that provide
this information are:
- Actual Throughput,
- Throughput Unit Numerator,
- Emissions Reliability Indicator,
- Emission Calculation Method,
- Factor Numeric Value,
- Factor Unit Numerator,
- Factor Unit Denominator, and
- Emission Factor Reliability Indicator.
A maximum of 10 points could be earned for completeness of data. If the following fields are not
null then 10 points will be awarded. The sum of the points earned is divided by 5 to obtain a
completeness of data number.
Actual Throughput & Throughput Unit Numerator (both must be non-null)
• EM Reliability Indicator
Factor Numeric Value, Factor Unit Numerator, Factor Unit Denominator (all 3 must be
non-null)
• EF Reliability Indicator
Control Status (only uncontrolled and controlled are given points)
• Emission Calculation Method Code
February 01, 2006
114
-------
This factor addressees the quality of the emission estimate based on estimation method. For
example, estimates based on continuous monitoring receive higher scores than data based on less
accurate methods, such as "engineering judgment."
Theemissions calculation method code determines the amount of points used in the rating. The
maximum number of points given for emission calculation method code is 10. If the field is null or
invalid, no points are given.
Emission Calculation Emission Calculation Method Description Points Given
Method Code
01 CEMS - Continuing Monitoring System 10
02 Engineering Judgment 2
03 Material Balance 5
04 Stack Test 8
05 EPA Speciation Profile 4
06 State/Local/Tribal Speciation Profile 4
07 Manufacturer Specification 4
08 EPA Emission Factor 5
09 State/Local/Tribal Emission Factor 5
10 Site-Specific Emission Factor 5
11 Vendor Emi s si on F actor 5
12 Trade Group Emission Factor 5
• Age of Data
In some cases, the NEI will have an emissions estimate from an earlier year, not the current
inventory year. Having "old" data is preferred to having a data gap. This factors accounts for the
age of the estimate.
A maximum of 5 points will be given for the age of data. If the data has a start date or end date of
2002, then 5 points will be given. Any other year will be given zero.
• Emission Data Level
Emission Data Level is very important for modeling. An estimate which provides process level
emissions is better than aggregated facility level HAP emissions. A maximum of 5 points will be
given for the emission data level.
Emission Data Level Points Given
Process 5
Stack 3
Unit 3
Site 1
February 01, 2006
115
-------
• Specificity of Data
This factor addresses pollutant groups. Reporting of individual HAPs is preferred to reporting HAP
compound groups such as arsenic and compounds.
A maximum of 5 points will be given for the specificity of the data. If the pollutant is ranked
greater than 1 in the pollutant code table, no points will be given. All others will receive 5 points.
see
SCCs are very important for modeling. This factor addresses whether SCCs are present.
A maximum of 5 points will be given for the SCC. If the SCC is not null and not 39999999, then 5
points will be given. All others will receive zero points.
• Composite Score and Grading
All of the different scores will be added together and each record will receive the following grades
based on the score:
Score Grade
33-40 A
25-32 B
17-24 C
9-16 D
0-8 F
We will assign the Data Rating for all HAP point source records in the final 2002 NEI.
12.14 Data Source
The Data Source in the NOF Emissions Table indicates the source of the reported emission value.
The following codes will be used.
767 EGU estimates based on 2002 Department of Energy (DOE) Form EIA-767
information (utility related data)
767/CAMD EGU record in both 2002 DOE Form EIA-767 and 2002 CAMD ETS/CEM;
with S02, NOx, and heat input values from CAMD ETS/CEM; condensible
PM, and primary PM10 and PM2.5 are recalculated using EPA's CAMD
heat input values
767/CAMD 1 EGU Combined cycle record (heat recovery steam generators (HRSG) +
combustion turbine) in 2002 CAMD ETS/CEM used; the HRSG record in
2002 DOE Form EIA-767 eliminated. S02, NOx, and heat input values from
CAMD ETS/CEM; other emissions estimated
767/CAMD2 EGU Combined cycle record (steam turbine + combustion turbine) in 2002
ETS/CEM used; the steam record in 2002 Form EIA-767 eliminated. S02,
NOx, and heat input values from ETS/CEM; other emissions estimated.
A Augmented PM data
February 01, 2006
116
-------
A-B
A-C
A-R
B
BCAMD1
C-Ol-F
CAMD
CAMD1
I
L
LCAMD1
LCAMD3
M
N
P
PCAMDHG
PI
PLHG
PSHG
R
S
SCAMD1
SCAMD2
SCAMD3
SCAMD4
T
February 01, 2006
Augmented boiler data
Augmented chromium Data
Augmented PM data replaced existing PM data submitted to NEI
Tribal data
EGU data were received from the tribal agency. The tribal agency's NOX
and S02 emission values were replaced with the ETS values.
Wyoming did not have any PM and NOX data in their submittal. Data was
brought forward from the 2001 inventory.
EGU record only in CAMD 2002 ETS/CEM for S02, NOx, and heat input
values; other emissions estimated.
EGU data was not received from the state, local agency, or tribal agency.
The ETS records were used.
Industry data
Local agency data
EGU data were received from the local agency. The local agency's NOX
and S02 emission values were replaced with the ETS values.
EGU data were received from the local agency. The local agency's NOX
and S02 value was used where the ETS value was zero.
MACT, Residual Risk, or Area Source data provided by ESD
99 NEI
Preferred MACT, Residual Risk, or Area Source data provided by ESD
EGU Mercury data . No data were received from state and local agencies
and tribes. The ESD mercury values were given to the ETS record.
Preferred industry data.
EGU Mercury data were received from the local agency. The local agency
provided a total mercury record. The record was replaced with three
individual mercury species records from ESD.
EGU Mercury data were received from the state. The state provided a total
mercury record. The record was replaced with three individual mercury
species records from ESD.
Regional Planning Organization data
State agency data
EGU data were received from the state. The state's NOX and S02 emission
values were replaced with the ETS values.
EGU Data were received from the state. The state did not provide NOX,
S02, or either. NOX and/or S02 records were created with the same IDs and
the ETS values were used.
EGU data were received from the state. The state's NOX and S02 value was
used where the ETS value was zero.
EGU Data were received from the state. It was not clear which of the state's
units matched to the ETS units. The state values were maintained, because
the emissions were close to the ETS values.
2002 Toxics Release Inventory data
117
-------
12.15 Maintaining an Emissions History Table
We will create and maintain an Emissions History Table which tracks the changes in total
emissions for each facility/site/pollutant combination in iterative versions of the 2002 NEI. This
table will enable users to see past and current total facility emissions for each pollutant. The
Emissions History Table will contain the following elements.
Status - Indicates if a facility has closed or was added after the initial version
State and County FIPS Code
Tribal Code
• NEI Facility ID
Pollutant Code
Pollutant Name
HAP Category Name
Original P (tpy) - Emissions supplied for Preferred category by ESD for MACT, Residual
Risk or Area Source
Original S (tpy) - Emissions supplied by State
Original L (tpy) - Emissions supplied by Local Agency
Original B (tpy) - Emissions supplied by Tribe
Original R (tpy) - Emissions supplied by Regional Planning Organization
Original T (tpy) - Emissions supplied by TRI
Original M (tpy) - Emissions supplied by ESD for MACT, Residual Risk or Area Source
OrigCAMD_NS (tpy) - Emissions supplied by CAMD for NOX and S02 for EGUs
OrigE (tpy) - Emissions for EGUS for all other pollutants except NOX and S02 provided by
EPA
OrigA (tpy) - Emissions augmented by EPA
• 02012005 Draft Emissions (tpy)
• 02012005 Draft Data Source
• 02012006 Final Emissions (tpy)
• 02012005 Final Data Source
Revised Version Date - date of future revisions
Revised Emissions Value (tpy) - emissions value of future revisions
Revised Data Source - data source of future revisions
Emissions Change (tpy) - difference between revised and current emissions
Comments
This table will be aggregated to the site/pollutant code level and will provide an emission estimate
for all processes associated with each site. As new versions of the NEI are created, a new column
will be added to this table providing the revised estimates and the data source for the newest
version.
If a site closed (and is deleted from the inventory), the historical record will be retained, and the
status column will be updated to "D " If a site is added that did not exist in previous versions, the
February 01, 2006
118
-------
status will be updated to "A" and the columns related to the most recent version (and every version,
thereafter) will be filled in. Maintaining a derived table like this can quickly become unwieldy if
too many data elements are added or the level of disaggregation is too detailed. This table is not
meant to provide all the information that the NEI contains, but to provide a quick overview of the
changes in emissions values over the course of the revision process. It is meant to help the user
target facilities for further investigation, not provide all of the supporting information.
The Emissions History Table will be supplemented by the "holding" tables which contain the
specific and individual changes submitted during each review period. These holding tables mirror
the NIF tables and contain complete NIF records. These records have a "submittal flag" which
indicates the change type ("A", "RA" or "D") and contain a source field which describes the
submitter of the change. We will improve this table by adding a comment field. This field would
be populated in those cases in which a user has provided a specific reason for a change.
In addition to the "holding" tables, when we use the "merge" algorithm to choose one value from
among multiple data sources for the same pollutant/facility, we will save the "rejected values" to
subsidiary tables along with the list of data source choices. We will experiment with merging this
information with the Emissions History Table. If supplementing the Emissions History Table with
the "rejected values" is not efficient, we will retain them as separate stand-alone tables.
The 2002 NEI Emissions Histyory Table will be located in a zipped file,
nei2002_point_facility_emissions_history, at:
ftp://ftp.epa.gov/EmisInventorv/2002finalnei/point sector data/ .
February 01, 2006
119
-------
SECTION 13. OTHER FIELDS THAT UNDERGO QA BUT ARE NOT AUGMENTED
There are data fields that we will QA reported values and codes. If the codes are invalid, we will
delete the values and not augment the values.
These following data fields will undergo QA but will not be augmented.
• EF Reliability Indicator
Emission Calculation Method Code
Material
Material I/O
Rule Effectiveness Method
Telephone Number Type Name
Transaction Creation Date
February 01, 2006
120
-------
SECTION 14. DEFAULT FLAGS
Default Flags will be added to all records in each table to indicate values of fields that have been
deleted, revised or augmented. Specific default flags that were discussed in previous sections will
be added for the following fields.
Coordinate Default Methodology (Latitude/Longitude)
NAICS/SIC Code Default Flag
MACT Code Assignment
Stack Parameter Default Flag
Total Capture Control Efficiency Methodology Flag
For all fields of data that are defaulted, a Default Flag will be added to each NOF Table to indicate
the fields of data that are defaulted. Each data field in a table will be assigned a number from 01 to
xx, and we will indicate if a field is defaulted by reporting its number. For example, the following
fields are in the Control Equipment Table:
01
Record Type
02
State and County FIPS Code
03
State Facility Identifier
04
Emission Unit ID
05
Process ID
06
Pollutant Code
07
Primary PCT Control Efficiency
08
PCT Capture Efficiency
09
Total Capture Control Efficiency
10
Primary Control Device Type Code
11
Secondary Control Device Type Code
12
Control System Description
13
Third Control Device Type Code
14
Fourth Control Device Type Code
15
Submittal
16
Tribal Code
If we defaulted the Emission Unit ID, Pollutant Code, and Primary Control Device Code, then we
would populate the Default Flag as "04,06,10".
February 01, 2006
121
-------
United States Office of Air Quality Planning and Standards Publication No. EPA-454/R-20-001
Environmental Protection Air Quality Assessment Division February 2006
Agency Research Triangle Park, NC
------- |