* _ \
iWJ
PRO^
Evaluate 2008 NEI to Identify Areas of
Improvement that Benefit Use in Residual Risk
Assessments
-------
-------
EP A-454/R-19-013
December 2012
Evaluate 2008 NEI to Identify Areas of Improvement that Benefit Use in Residual Risk
Assessments
U.S. Environmental Protection Agency
Office of Air Quality Planning and Standards
Air Quality Assessment Division
Research Triangle Park, NC
-------
*ERG
Evaluate 2008 NEI to Identify Areas of
Improvement that Benefit Use in Residual
Risk Assessments
Prepared for:
U.S. Environmental Protection Agency
Rebecca Lee Tooly
U.S. Environmental Protection Agency
109 T.W. Alexander Drive
Mail Code C339-02
Research Triangle Park, NC 27711
Prepared by:
Eastern Research Group, Inc.
1600 Perimeter Park
Suite 200
Morrisville, NC 27560
December 14, 2012
EPA Contract No. EP-D-11-006
Work Assignment 2-08
-------
TABLE OF CONTENTS
Page
1.0 Introduction 1
2.0 Background - 2008 NEI 2
3.0 Prioritize the Type of Improvements of Most Significant Benefit to RTR
and Establish Assessment Focus 3
3.1 Key Emission Inventory Data for RTR Sector Modeling 3
3.2 Improvements Made in the 2008 NEI Cycle With Expected Benefit for the
RTR Program 6
3.3 Priority Source Categories 9
3.3.1 RTR and MACT Source Categories 10
3.3.2 Source Categories Identified by 2008 NEI Toxicity Weighted
Emissions 16
3.3.3 Recommended Priority Source Category List 20
4.0 Usefulness of the 2008 NEI for Recent RTR Work 27
5.0 Assess 2008 NEI for Missing HAPs 34
5.1 Matrix of Expected HAPs 34
5.1.1 Data References 35
5.1.2 Master Matrix Re suits 38
5.1.3 Other Data References Not Considered 39
5.2 Estimate Potentially Missing HAP-YOC and HAP-PM in 2008 NEI .40
5.3 Evaluate TRI Data Usage in RTR 43
5.4 Missing HAP Results 45
6.0 Recommended Improvement activities for future NEI development
USING THESE RESULTS 48
7.0 References 51
Appendix A. 1 - Expected HAPs Matrix
Appendix A.2 - Summary of SCC and HAP Counts by EIS Sector
Appendix A.3 - Comparison of Expected HAPs to HAPs Reported to the 2008 v2 Point
Sources NEI
Appendix B - Crosswalk of VOC HAPs and PM HAPs
Appendix C - RTR TRI Data Not Used in 2008 NEI
i
-------
LIST OF TABLES
Page
Table 1. Key Emission Inventory Fields for RTR Emissions Modeling 4
Table 2. Improvements in the 2008 NEI Cycle 6
Table 3. RTR Priority Source Categories With Final Rule Publication Date 10
Table 4. RTR Priority Source Categories With Final Rule Signature or Consent Decree (Not
Published) 11
Table 5. RTR Priority Source Category With No Consent Decree 12
Table 6. Non-RTR MACT Source Categories 12
Table 7. Initial Recommendation of Priority Source Categories 15
Table 8. Top Source Categories Ranked by Cumulative Cancer Toxicity Weighting 17
Table 9. NATA05 National and Regional Pollutant Drivers and Contributors 18
Table 10. Listing of Source Categories Contributing to 90% of Noncancer Toxicity Weighting
By Select Target System 21
Table 11. Recommended Assessment Priority for Source Categories Based on RTR
Analysis Schedule and Toxicity-weighting Results 24
Table 12. Charge Questions Evaluating Usefulness of 2008 NEI in RTR Activities 28
Table 13. SPPD MACT/Regulatory Websites 35
Table 14. Example Format for Expected HAPs Matrix - NEI Point Sources 38
Table 15. EPARegional Office Permitting Websites 39
Table 16. Non-Point Stationary Sources - Top 5 HAP Emissions By 2011 Ratio Method For
Select SCCs Not Included in 2008 NEI 42
Table 17. Top 10 TRI HAP Emissions Not Included in 2008 NEI Point Sources 45
Table 18. Comparison of Expected HAPs to Actual HAP Reporting 46
-------
Evaluate 2008 NEI to Identify Areas of Improvement that Benefit Use in Residual Risk Assessments
1.0 Introduction
The Office of the Inspector General (OIG) recommended that the U.S. Environmental
Protection Agency (EPA) improve air toxics emissions data needed to conduct residual risk
assessments. The original OIG report is available at http://www.epa. gov/oig/reports/2008/20071031-08-
P-0020.pdf. In response to the OIG recommendation and to further continuous improvement
activities for the EPA's National Air Emissions Inventory (NEI), ERG assessed the 2008 NEI for
use by the residual risk program under EPA Contract No. EP-D-11-006, Work Assignment
(WA) 2-08. The purpose/objectives of the work assignment are the following:
1) Evaluate the 2008 NEI v2 to identify areas of improvements in HAP emissions that would
benefit the EPA Risk and Technology Review (RTR) and National Air Toxics Assessment
(NATA) programs, the expected degree of benefit, and the recommended implementation
priority (Task 2).
2) Establish the usefulness of the 2008 NEI in support of RTR recent activities. Specifically,
EIAG would like to understand if these inventories were considered in recent RTR
rulemakings (Task 3).
3) Perform analyses to assess the NEI for missing HAPs (Task 4).
4) Describe some next steps to accomplish suggested improvements during future NEI cycles,
including if feasible the 2011 NEI development cycle that has already started.
This report summarizes the technical approach and focus for assessing the 2008 NEI, results
from the analyses performed, and recommendations for improvement activities to implement for
future NEI development cycles.
The following describes how the information in this report is organized:
Section 2 - background information on the 2008 NEI
Section 3 - priority of improvements with significant benefit to RTR and focus of this
assessment
Section 4 - usefulness of the 2008 NEI for recent RTR work
Section 5 - analyses results from assessing the NEI for missing HAPs
Section 6 - recommendations for improvements based on these results for future NEI
development cycles
Section 7 - references
1
-------
Evaluate 2008 NEI to Identify Areas of Improvement that Benefit Use in Residual Risk Assessments
2.0 Background - 2008 NEI
The NEI was created to provide EPA, federal, state, local, tribal (SLT) agency decision
makers, as well as the U.S. public and other countries, the best and most complete estimates of
criteria and hazardous air pollutant (CAP and HAP) emissions for the United States. While EPA is
not directly obligated to create the NEI under the Clean Air Act (CAA), the CAA authorizes the
EPA Administrator to implement data collection efforts needed to properly administer the National
Ambient Air Quality Standards (NAAQS) program. Therefore, the EPA Office of Air Quality
Planning and Standards (OAQPS) implements the NEI program in support of the NAAQS.
Furthermore, the CAA requires states to submit emissions to EPA as part of their State
Implementations Plans (SIPs) that describe how they will meet the NAAQS, and the NEI is used as
one mechanism for states to meet some of those emission requirements, particularly for 3-year
reporting requirements.
While the NAAQS program is the basis on which EPA collects CAP emissions from the
SLT air agencies, it does not require collection of HAP emissions. The HAP reporting requirements
are voluntary. Nevertheless, compiling the HAP emissions are an essential part of the NEI program.
These emission estimates allow EPA to assess progress in meeting HAP reduction goals described
in the CAA Amendments of 1990 such as evaluation of risks remaining after EPA's application of
specific industry standards, e.g., Maximum Achievable Control Technology (MACT) standards,
and to determine whether additional standards are needed to reduce residual risks. The EPA also
conducts a national-scale air toxics assessment (NATA). The purpose of NAT A is to identify and
prioritize air toxics, emission source types, and locations that are of greatest potential concern in
terms of contributing to population risk. NATA is based on an inventory of air toxics emissions
including data available in the NEI. NATA typically follows the 3-year cycle of available NEI data.
Although the 2008 NEI is the most recent complete and comprehensive inventory available, NATA
2008 was not done due to resource limitations. Thus, 2005 is the most recent assessment. Results
are located at http://www.epa.gov/ttn/atw/nata2005. The EPA plans to conduct NATA 2011 which
will include use of the 2011 NEI.
For the 2008 NEI cycle, the Emission Inventory System (EIS), was used for the first time to
collect, compile, and store the emissions data. This new system greatly improved the collection
approach from less structured approaches used in the past. The numerous automated data checks in
EIS have undoubtedly improved the data quality and allowed EPA more time to review the data
prior to publication. Other data quality checks performed on the 2008 NEI have also improved the
HAP emissions such as specific correction of location coordinates for emission releases,
augmentation to complete apparent missing data, and priority review of specific facilities due to
high-risk potential.
2
-------
Evaluate 2008 NEI to Identify Areas of Improvement that Benefit Use in Residual Risk Assessments
3.0 Prioritize the Type of Improvements of Most Significant Benefit to RTR and
Establish Assessment Focus
In Task 2 of the WA, EPA identifies and prioritizes the types of improvements that could be
made to the NEI for point sources that are expected to provide the highest degree of benefit to the
RTR program. That established the focus of the assessment and includes a recommended list of
source categories and pollutants to assess for potentially missing HAPs.
3.1 Key Emission Inventory Data for RTR Sector Modeling
Performing a risk assessment for an RTR source category is a multi-step, collaborative
process involving multiple stakeholders. Stakeholders can include EPA regulatory and inventory
staff, risk assessors, SLT agencies, trade associations, and industry. This can be especially true of
complicated RTR sources categories, such the Pulp and Paper Production and Chemical Sectors.
Assessments generally include the following components (EPA, 2012a):
• Inhalation Assessment using the Human Exposure Model
• Multipathwav Assessment using a tiered approach:
o Tier 1: Compare facility emissions to risk-based thresholds
o Tier 2: Refine Tier 1 results using site-specific meteorological and location of
fishable lakes near the facility
o Tier 3: Refine Tier 2 results using dietary distributions of dietary fractions
o Tier 4: Model with Total Risk Integrated Methodology (TRIM)
¦ Focus on Persistent Bioaccumulative and Toxic (PBT) HAPs
• Ecological HAP Assessment (still being developed)
For the inhalation assessment, the approach generally follows these steps:
1. Inhalation modeling file development
2. Model preprocessing
3. Air dispersion modeling using AERMOD
4. Model-generated ambient HAP concentrations
5. HEM inhalation exposure modeling
6. Develop exposure concentrations
7. Risk characterization, using dose-response values
SPPD is responsible for the inhalation modeling file development, which primarily
includes the development of the emissions inventory. As such, the data elements that are
important for risk modeling require additional scrutiny prior to finalizing the emissions inventory.
EPA's HEM AERMOD equation is presented below:
3
-------
Evaluate 2008 NEI to Identify Areas of Improvement that Benefit Use in Residual Risk Assessments
c=
-{-HO
2<7V(X
Where:
C = model concentration
Q = emission rate
h = plume height release
x = downwind distance
As presented above, the most important emission inventory variables affecting the modeled
concentration (and thereby risk), from a modeling sensitivity standpoint, are: 1) the emission rate;
2) plume height release; and 3) downwind distance. Table 1 further describes these and related variables,
in addition to other emission inventory data that are important for risk modeling and which are
components of the risk modeling file. Additionally, Table 1 sorts these variables by their relative
importance and identifies which of them are data fields in the 2008 NEI.
Table 1. Key Emission Inventory Fields for RTR Emissions Modeling
Priority
Data Field(s)
Field(s) in
NEI2008
Description
1
Emission Rate
Yes
Emissions (Q) are directly proportional to modeled
concentrations (C), and thereby risk. The emission rate
included used in RTR modeling files is in units of tons
per year.
2
Emission
Release Point
Type, Stack
Parameters,
Fugitive
Dimensions
Yes
Important for understanding the fate and transport of
emitted pollutants. The emission release point type
indicates whether a release is fugitive or stack. If the
release is from a stack, the emission release point type
code indicates the type of stack (vertical, horizontal,
goose neck, vertical with rain cap, or downward-facing
vent). All of these parameters affect the plume height
release (h), which is inversely proportional to modeled
concentrations (C), and thereby risk.
3
Geographical
Coordinates
Yes
HAP emission release points are mapped to specific
census tracts, and the impacts on the affected nearby
population are estimated. The downwind distance (x) is
inversely proportional to modeled concentration (C),
and thereby risk.
4
-------
Evaluate 2008 NEI to Identify Areas of Improvement that Benefit Use in Residual Risk Assessments
Table 1. Key Emission Inventory Fields for RTR Emissions Modeling
Priority
Data Field(s)
Field(s) in
NEI2008
Description
4
MACT Codes
(or Regulatory
Codes)
Yes
The identification of as many facilities as possible that
are subject to a rule is the first step in the process.
MACT codes must be assigned at the process-level and
can also be pollutant-specific. Regulatory codes are
used in the 2008 NEI. Regulatory codes can be mapped
to MACT codes.
5
SCCs
Yes
MACT Standards are always process-level.
6
Pollutant
Codes
Yes
Especially important for distinguishing high risk
pollutants, such as hexavalent chromium (highly toxic)
vs. trivalent chromium (non-toxic). Some metals such
as mercury are not speciated in the 2008 NEI. Mercury
is speciated for RTR modeling files.
7
Emission
Process Group
No
Assigned by SPPD project lead to further identify the
types of processes that are assigned to the MACT code;
allows for review of risk results and evaluation of
control options by groups of processes (e.g., storage
tanks) rather than at the detailed SCC-level or broader
facility-level.
8
Emissions
Type
Yes
Actual, allowable, maximum, etc.
9
Facility
Category
Code
Yes
Code which describes if the facility is Major or Area.
10
Control
Technology
Information
(e.g.,
device(s),
control
efficiency)
Yes
Important in understanding emissions and for
technology option selection (i.e., technology review).
11
Facility
Identification
Yes
NEI ID and/or State Facility ID used by SPPD Leads
12
NAICS Codes
Yes
In lieu of regulatory codes, the NAICS code assignment
and/or facility type provides an indication of the type of
the type of operations and processes at a facility.
13
Data Source
Yes
Indicates source of data, such as SLT, SPPD, TRI,
augmented, or carried-forward from prior inventory.
14
Year Closed
Yes
Year in which the facility closed; primarily for landfills
or facilities maintained in the 2005 NEI for historical
purposes. Included in 2008 NEI as Facility Site Status
Code and Facility Site Status Code Year.
5
-------
Evaluate 2008 NEI to Identify Areas of Improvement that Benefit Use in Residual Risk Assessments
RTR emissions inventories are developed using data from a variety of sources including
Information Collection Requests (ICR), Section 114 letters, voluntary surveys, emissions tests, site
visits, stakeholder meetings, Trade Associations, air permits, 2005 NEI, 2008 NEI, and TRI for current
and prior years. After the information from these variety of sources are compiled, the data are reviewed
in detail in order to identify the range of differences, incorrect codes, or other inconsistencies between
these data sources. Significant errors observed in the RTR emissions inventory development process can
skew risk results. These differences are examined for potential error, with examples listed below:
• Incorrect entry of emissions data, such as entering emissions in pounds instead of tons;
• Unreasonable stack parameters, such as stack temperature entered in °C rather than °F;
• Stack coordinates at the center of a facility or outside of the fenceline, rather than at specific
release points;
• Improper designation of MACT processes, SCCs, and Emission Process Groups due to
miscoding and/or misunderstanding of processes;
• Incomplete or over-estimation of emissions inventory data due to misunderstanding of
facility configuration (i.e., is the neighboring facility part or separate);
• Maximum allowable emissions entered rather than actual emissions;
• Improper use of pollutant codes by entering total pollutant rather than speciated pollutant;
• Incorrectly designating a facility as "Area" when it's "Major"; and
• Identifying facilities (or processes) as active, when they are actually closed or inactive.
3.2 Improvements Made in the 2008 NEI Cycle With Expected Benefit for the RTR Program
As stated before, for the 2008 NEI cycle, the EIS was used for the first time to collect, compile,
and store the emissions data submitted by SLT agencies. Some of the improvements made during the
2008 NEI development process for both stationary and mobile sources are described in the EPA's 2008
NEI v2 Technical Support Document (TSD)
(http://www.epa.gov/ttn/chief/net/2008neiv2/2008_nei v2_tsd_draft.pdf) and are presented in Table 2.
Improvements for stationary sources expected to have direct benefit to the RTR program are noted.
Table 2. Improvements in the 2008 NEI Cycle
#
Improvement
RTR
Benefit
1
More structured, automated collection of data, including automated data audit.
y
2
Numerous data checks of submitted data prior to acceptance in EIS.
V
3
Less time formatting data into common data structure allowing more time for
QA.
-------
Evaluate 2008 NEI to Identify Areas of Improvement that Benefit Use in Residual Risk Assessments
Table 2. Improvements in the 2008 NEI Cycle
#
Improvement
RTR
Benefit
8
EIS is more dynamic in nature than previous NEIs. As SLT agencies correct
data and are approved by EPA, data pulls will reflect current information.
V
9
The 2008 NEI v2 uses updated emissions factors for several metal HAPs and
acid gases from coal-fired utility boilers.
V
10
In EIS, another category called "event" has been added and is used to compile
day-specific data from fires.
11
Consolidated the number of HAP compounds significantly for metals and
cyanides and provided conversion factors to enable SLT agencies to provide
them as the metal or cyanide that is important for risk.
S
12
For all data categories only speciated chromium and specific allowable
chromium species by speciating agency-reported total chromium are reported.
This was done to allow easier toxicity weighting of the inventories and more
streamlined risk modeling.
S
13
EIS allows the NEI to have both facility latitude/longitude coordinates as well
as release point (e.g., stack) coordinates. Previous NEI databases could store
only coordinates at release points. The two separate sets of values allowed EPA
to assess whether the facility coordinates and the release point coordinates were
in the same vicinity and make adjustments to resolve inconsistencies in
collaboration with the SLT agencies.
S
14
Data were not carried forward from previous NEIs, in response to SLTs
comments about potential double-counting and facility closures (however, this
may lead to incomplete source categories if facilities were not reported).
S
15
For onroad mobile sources, the 2008 NEI v2 used the MOVES model for the
first time. The MOVES-based emissions have been compiled using daily
meteorology data for 2008 rather than monthly averages used in past
approaches, and then summed to an annual value.
16
For nonroad mobile sources, emissions at airports are treated comprehensively
as point sources. In past inventories, some airports were point sources while
others were aggregated to a total nonpoint county estimate.
17
The emissions for aircraft ground support equipment and aircraft auxiliary
power units associated with aircraft-specific activity were estimated by the
Federal Aviation Administration (FAA) Emissions and Dispersion Modeling
System (EDMS) using the assumptions and defaults incorporated in the model.
This is a significant change from the previous NEI emissions, for which ground
support equipment estimates came from the NONROAD model and auxiliary
power unit emissions were not included in EPA's estimates.
18
The in-flight lead emissions have been included in the 2008 NEI for the first
time and are reflected in the totals for the "Mobile Sources - Aircraft" sector.
19
For fires, EPA has used the SMARTFIRE2 system for the first time in the 2008
NEI v2. This system eliminated a shortcoming in the 2005 NEI that did not
assign all fires to either wildfire or prescribe burning categories.
7
-------
Evaluate 2008 NEI to Identify Areas of Improvement that Benefit Use in Residual Risk Assessments
Table 2. Improvements in the 2008 NEI Cycle
#
Improvement
RTR
Benefit
20
In 2005 NEI, emissions from vessels out to 200 nautical miles (nm) were
allocated to "state" emissions, whereas in the 2008 NEI, emissions only in state.
21
The 2008 NEI also includes emission estimates for aircraft auxiliary power units
(APUs) and aircraft ground support equipment (GSE) typically found at
airports, such as aircraft refueling vehicles, baggage handling vehicles, and
equipment, aircraft towing vehicles, and passenger buses. These APUs and GSE
are located at the airport facilities as point sources along with the aircraft
exhaust emissions.
22
EPA developed emissions estimates associated with an aircrafts' landing and
takeoff (LTO) cycle. During each mode of operation, an aircraft engine operates
at a fairly standard power setting for a given aircraft category. Emissions for one
complete cycle are calculated using emission factors for each operating mode
for each specific aircraft engine combined with the typical period of time the
aircraft is in the operating mode.
23
The approach for matching EIS units with the Mercury Air Toxics Standard
(MATS) data so that the EIS software used only one emissions estimate for a
process-pollutant combination, rather than one estimate from each data supplier
prevented double-count emissions.
24
The 2008 NEI v2 data contains two sets of alternate unit identifiers related to
the ORIS plant and CAMD boiler IDs (as found in the CAMD heat input
activity dataset) for export to the SMOKE modeling file. ORIS IDs typically are
assigned to electric generating units (EGUs).
25
EPA developed data for industrial nonpoint fuel combustion as a means of
assisting SLT to develop their own nonpoint estimates by accounting for the
point source contribution that they submitted, and the total fuel available for
combustion tracked by the Energy Information Administration.
26
In addition, the "Where you live" feature of the Air Emissions website allows
users to select states and EIS sectors to create KMZ files used by Google Earth.
27
Other significant emissions sectors which have seen improvements and
therefore inconsistent trend data through the years include paved and unpaved
road PM emissions, animal waste ammonia emissions, and residential wood
combustion emissions.
28
EIS Sectors are being used for the first time with the release of the 2008 NEI.
These sectors have been developed to better group emissions for both CAP and
HAP summary purposes. The sectors are based simply on grouping the
emissions by the emissions process based on the source classification code
(SCC) to the EIS sector.
29
SLT agency submission of particulate matter (PM) emissions to the NEI are
required to include primary PMio (called PMio-PRI in EIS and NEI outputs) and
primary PM2.5 (PM25-PRI). In addition, EPA requests states provide filterable
PM (PM10-FIL and PM25-FIL) along with condensable PM (PM-CON).
8
-------
Evaluate 2008 NEI to Identify Areas of Improvement that Benefit Use in Residual Risk Assessments
Table 2. Improvements in the 2008 NEI Cycle
#
Improvement
RTR
Benefit
30
As part of the quality assurance, EPA examined whether some of these
categories had VOC but not HAP VOC. Since many of these sectors are known
and important emitters of HAP VOC, when VOC is provided without HAP
VOC this is a clear case of missing emissions.
V
31
The emission factors for some counties in the CMU Ammonia Model files were
zero. To ensure that all counties with animal populations were assigned
emissions factors, the emission factor input files provided with the CMU
Ammonia Model were modified. For all counties with an emission factor of
zero, the emission factor was replaced with the state average emission factor. If
all counties in the state had emission factors of zero, then the county emission
factor was replaced with the national average emission factor.
32
Another quality assurance method conducted for mercury (Hg) was to look at
boiler SCCs and check for Hg emissions. Other than for natural gas
consumption, Hg is expected. As it turned out, some boilers even after gapfilling
using TRI and HAP augmentation did not have Hg emitted. EPA computed that
the 2008 NEI were missing 0.5 tons of Hg. Note that this issue included all
boilers, not just from the industrial sector.
S
33
If a state reported a zero value for any of the HAPs, that zero was retained in the
2008 NEI.
S
3.3 Priority Source Categories
The next stage of this task establishes assessment focus through prioritization of key HAPs
and/or sectors. Key HAPs may include recent regulatory actions, such as for mercury, lead, and/or the
NATA priority pollutants (e.g., benzene, formaldehyde, acrolein, naphthalene, and manganese). Key
sectors may focus on RTR sectors, such as:
• Boilers;
• Coke Ovens;
• Portland Cement;
• Polymers and Resins;
• Secondary Aluminum;
• Phosphoric Acid/Phosphate Fertilizers;
• Mercury-cell Chlor-alkali Plants;
• Secondary Lead; and
• Oil and Natural Gas Activities.
An example of a non-RTR sector is Municipal Solid Waste Landfills.
9
-------
Evaluate 2008 NEI to Identify Areas of Improvement that Benefit Use in Residual Risk Assessments
3.3.1 RTR and MACT Source Categories
One practical approach to prioritizing key HAPs/key sectors is to identify those that have gone
through recent or are undergoing RTR rulemaking activities (e.g., Oil and Natural Gas) and non-RTR
rulemaking activities (e.g., Mercury Air Toxics Standards or LeadNAAQS). The priority may be the
sectors which may have more immediate impact on the 2011 NEI cycle. Table 3 presents the RTR
source categories that have been published (EPA, 2012b). Of the 34 published rules, eighteen have
been published since 2008.
Table 3. RTR Priority Source Categories With Final Rule Publication Date
Source Category
Proposal
Signature
Date/Consent
Proposal
Publication
Date
Final Rule
Signature
Date/Consent
Final Rule
Publication
Date
Decree Date
Decree Date
Hard Chromium Electroplating
1/27/2012
2/8/2012
8/15/2012
9/19/2012
Decorative Chromium
1/27/2012
2/8/2012
8/15/2012
9/19/2012
Electroplating
Chromium Anodizing Tanks
1/27/2012
2/8/2012
8/15/2012
9/19/2012
Steel Pickling-HCL Process
1/27/2012
2/8/2012
8/15/2012
9/19/2012
Pulp and Paper I & III
12/15/2011
12/27/2011
7/31/2012
9/11/2012
Oil and Natural Gas
7/28/2011
8/23/2011
4/3/2012
8/16/2012
Production
Natural Gas Transmission and
7/28/2011
8/23/2011
4/3/2012
8/16/2012
Storage
Secondary Lead Smelters
4/29/2011
5/19/2011
12/16/2011
1/5/2012
Shipbuilding and Ship Repair
12/3/2010
12/21/2010
11/4/2011
11/21/2011
Wood Furniture
12/3/2010
12/21/2010
11/4/2011
11/21/2011
Primary Lead Smelting
1/31/2011
2/17/2011
11/4/2011
11/15/2011
Marine Vessel Loading
9/14/2010
10/21/2010
3/31/2011
4/21/2011
Pharmaceuticals
9/14/2010
10/21/2010
3/31/2011
4/21/2011
Printing and Publishing
9/14/2010
10/21/2010
3/31/2011
4/21/2011
Epichlorohydrin Elastomers
9/14/2010
10/21/2010
3/31/2011
4/21/2011
Production P&R I
Nitrile Butadiene Rubber
9/14/2010
10/21/2010
3/31/2011
4/21/2011
Production, P&R I
Polybutadiene Rubber
9/14/2010
10/21/2010
3/31/2011
4/21/2011
Production, P&R I
Styrene Butadiene Rubber and
9/14/2010
10/21/2010
3/31/2011
4/21/2011
Latex Production, P&R I
Polysulfide Rubber-Pc&i? /
12/6/2007
12/12/2007
12/10/2008
12/16/2008
Ethylene Propylene-Rubber
12/6/2007
12/12/2007
12/10/2008
12/16/2008
P&R I
Butyl Rubber- P&R I
12/6/2007
12/12/2007
12/10/2008
12/16/2008
Neoprene-/\f7^ /
12/6/2007
12/12/2007
12/10/2008
12/16/2008
10
-------
Evaluate 2008 NEI to Identify Areas of Improvement that Benefit Use in Residual Risk Assessments
Table 3. RTR Priority Source Categories With Final Rule Publication Date
Source Category
Proposal
Signature
Date/Consent
Decree Date
Proposal
Publication
Date
Final Rule
Signature
Date/Consent
Decree Date
Final Rule
Publication
Date
Epoxy Resins- P&R II
12/6/2007
12/12/2007
12/10/2008
12/16/2008
Non-Nylon Polyamides- P&R
II
12/6/2007
12/12/2007
12/10/2008
12/16/2008
Acetal Resins GMACTI
12/6/2007
12/12/2007
12/10/2008
12/16/2008
Hydrogen Fluoride GMACT I
12/6/2007
12/12/2007
12/10/2008
12/16/2008
Halogenated Solvent Cleaners
8/9/2006
8/17/2006
4/16/2007
5/3/2007
Hazardous Organics NESHAP
(HON)
6/1/2006
6/14/2006
12/15/2006
12/21/2006
Dry Cleaners
12/9/2005
12/21/2005
7/13/2006
7/27/2006
Industrial Process Cooling
Towers
10/18/2005
10/24/2005
3/31/2006
4/7/2006
Hospital Sterilizers
10/18/2005
10/24/2005
3/31/2006
4/7/2006
Magnetic Tape
10/18/2005
10/24/2005
3/31/2006
4/7/2006
Gasoline Distribution
8/4/2005
8/10/2005
3/31/2006
4/6/2006
Coke Ovens
7/29/2004
8/9/2004
3/31/2005
4/15/2005
Table 4 presents the seventeen RTR source categories that have upcoming consent decree
dates (Pope, 2012).
Table 4. RTR Priority Source Categories With Final Rule Signature or Consent Decree
(Not Published)
Source Category
Proposal
Signature
Date/Consent
Proposal
Publication
Date
Final Rule
Signature
Date/Consent
Final Rule
Publication
Date
Decree Date
Decree Date
Portland Cement
6/15/2017
NA
6/15/2018
—
Aerospace
3/15/2014
NA
1/15/2015
—
Acrylic/ Modacrylic Fibers
12/11/2013
NA
9/15/2014
—
Polycarbonates Production
12/11/2013
NA
9/15/2014
—
Polymers and Resins III
12/11/2013
NA
9/15/2014
—
Off-Site Waste Recovery
12/11/2013
NA
9/15/2014
—
Operations
Phosphoric Acid
11/14/2013
NA
8/15/2014
—
Phosphate Fertilizers
11/14/2013
NA
8/15/2014
—
Flexible Polyurethane Foam
10/30/2013
NA
7/30/2014
—
Production
Secondary Aluminum
1/30/2012
2/14/2012
3/14/2014
—
Primary Aluminum
11/4/2011
12/6/2011
3/14/2014
—
11
-------
Evaluate 2008 NEI to Identify Areas of Improvement that Benefit Use in Residual Risk Assessments
Table 4. RTR Priority Source Categories With Final Rule Signature or Consent Decree
(Not Published)
Source Category
Proposal
Signature
Date/Consent
Decree Date
Proposal
Publication
Date
Final Rule
Signature
Date/Consent
Decree Date
Final Rule
Publication
Date
Pesticide Active Ingredient
Production
11/30/2011
1/9/2012
1/31/2014
—
Polyether Polyols Production
11/30/2011
1/9/2012
1/31/2014
—
Polymers and Resins IV
11/30/2011
1/9/2012
1/31/2014
—
Ferroalloys Production
11/4/2011
11/23/2011
12/10/12a
—
Mineral Wool
11/4/2011
11/25/2011
12/10/12a
—
Wool Fiberglass
11/4/2011
11/25/2011
12/10/12a
—
a To be determined, but currently scheduled for this date.
Table 5 presents one RTR source category that does not yet have a consent decree, but is on-
going due to a settlement agreement.
Table 5. RTR Priority Source Category With No Consent Decree
Source Category
Proposal
Signature
Date/Consent
Decree Date
Proposal
Publication
Date
Final Rule
Signature
Date/Consent
Decree Date
Final Rule
Publication
Date
Petroleum Refineries
No consent decree
Finally Table 6 presents seventy-nine MACT source categories that are not on the RTR
Schedule and have no consent decree or proposal date.
Table 6. Non-RTR MACT Source Categories
MACT Source Category
MACT
Code
Ammonium Sulfate - Caprolactam By-Product Plants
1401
Asphalt Processing and Asphalt Roofing Manufacturing
0418
Asphalt/Coal Tar Application - Metal Pipes
0402
Auto & Light Duty Truck (Surface Coating)
0702
Boat Manufacturing
1305
Brick and Structural Clay Products Manufacturing
0414
Carbon Black Production
1415
Carbonyl Sulfide (COS) Production
1604
Cellulose Products Manufacturing
1349
12
-------
Evaluate 2008 NEI to Identify Areas of Improvement that Benefit Use in Residual Risk Assessments
Table 6. Non-RTR MACT Source Categories
MACT Source Category
MACT
Code
Clay Ceramics Manufacturing
0415
Commercial and Industrial Solid Waste Incineration
1807-1
Commercial Sterilization Facilities
1609
Cyanide Chemicals Manufacturing
1405
Engine Test Facilities
0101-1
Ethylene Processes
1635
Flexible Polyurethane Foam Fabrication Operations
1341
Friction Materials Manufacturing
1636
Hazardous Waste Incineration: Cement Kilns
0801-3
Hazardous Waste Incineration: Commercial
0801-1
Hazardous Waste Incineration: HC1 Production Furnaces
0801-7
Hazardous Waste Incineration: Lightweight Aggregate Kilns
0801-4
Hazardous Waste Incineration: Liquid Fuel Boilers
0801-6
Hazardous Waste Incineration: On-Site
0801-2
Hazardous Waste Incineration: Solid Fuel Boilers
0801-5
Hydrochloric Acid Production
1407
Industrial/Commercial/Institutional Boilers & Process Heaters - coal
0107-1
Industrial/Commercial/Institutional Boilers & Process Heaters - gas
0107-2
Industrial/Commercial/Institutional Boilers & Process Heaters - oil
0107-3
Industrial/Commercial/Institutional Boilers & Process Heaters - wood or waste
0107-4
Integrated Iron & Steel Manufacturing
0305
Iron and Steel Foundries
0308
Large Appliance (Surface Coating)
0704
Leather Tanning & Finishing Operations
1634
Lime Manufacturing
0408
Manufacture of Nutritional Yeast
1101
Medical Waste Incinerators
1801
Mercury Cell Chlor-Alkali Plants
1403
Metal Can (Surface Coating)
0707
Metal Coil (Surface Coating)
0708
Metal Furniture (Surface Coating)
0709
Miscellaneous Coating Manufacturing
1642
Miscellaneous Metal Parts & Products (Surface Coating)
0710
Miscellaneous Organic Chemical Manufacturing
1641
Municipal Landfills
0802
Municipal Waste Combustors: Large
1802-2
Municipal Waste Combustors: Small
1802-1
Organic Liquids Distribution (Non-Gasoline)
0602
Other Solid Waste Incineration
1807-2
Paint Stripping Operations
1621
Paper & Other Webs (Surface Coating)
0711
13
-------
Evaluate 2008 NEI to Identify Areas of Improvement that Benefit Use in Residual Risk Assessments
Table 6. Non-RTR MACT Source Categories
MACT Source Category
MACT
Code
Plastic Parts & Products (Surface Coating)
0712
Plywood and Composite Wood Products
1624
Polyvinyl Chloride & Copolymers Production
1336
Primary Copper Smelting
0203
Primary Magnesium Refining
0207
Printing, Coating & Dyeing Of Fabrics
0713
Publicly Owned Treatment Works (POTWs)
0803
Pulp & Paper Production II
1626-2
Refractory Products Manufacturing
0406
Reinforced Plastic Composites Production
1337
Rocket Engine Test Firing
0101-2
Rubber Tire Production
1631
Semiconductor Manufacturing
1629
Site Remediation
0805
Solvent Extraction for Vegetable Oil Production
1103
Spandex Production
1003
Stationary Combustion Turbines - Natural Gas
0108-1
Stationary Combustion Turbines - Oil
0108-2
Stationary Reciprocating Internal Combustion Engines - Natural Gas
0105-1
Stationary Reciprocating Internal Combustion Engines - Oil
0105-2
Taconite Iron Ore Processing
0411
Uranium Hexafluoride Production
1414
Utility Boilers: Coal
1808-1
Utility Boilers: Natural Gas
1808-2
Utility Boilers: Oil
1808-3
Utility Boilers: Wood or Waste
1808-4
Viscose Process Manufacturing
1348
Wet-Formed Fiberglass Mat Production
0413
Wood Building Products (Surface Coating)
0703
Based on the above tables, it is recommended that priority focus of source categories be for the
18 source categories with final rules published since 2008, the 17 source categories with
signature/consent decree dates established, and the RTR category identified in Tables 4 and 5. Table 7
presents a composite list of these priority categories. Those with rules published since 2008 provide the
more recent RTR data that will be used to develop the matrix of expected HAPs. Those with established
consent decrees are recognized as 'soon to start' and for which RTR data are not yet available to help
develop the matrix of expected HAPs but for which NEI improvements are also desired as the NEI will
be consulted as one of many data sources.
14
-------
Evaluate 2008 NEI to Identify Areas of Improvement that Benefit Use in Residual Risk Assessments
Table 7. Initial Recommendation of Priority Source Categories
Source Category
Rule Published
Since 2008
Consent
Decree, Soon to
Start
Acrylic/Modacrylic Fibers
Y
Aerospace
Y
Chromium Anodizing Tanks
Y
Decorative Chromium Electroplating
Y
Epichlorohydrin Elastomers Production, P&R I
Y
Ferroalloys Production
Y
Flexible Polyurethane Foam Production
Y
Hard Chromium Electroplating
Y
Marine Vessel Loading
Y
Mineral Wool
Y
Natural Gas Transmission and Storage
Y
Nitrile Butadiene Rubber Production, P&R I
Y
Off-Site Waste Recovery Operations
Y
Oil and Natural Gas Production
Y
Pesticide Active Ingredient Production
Y
Petroleum Refineries
YA
Pharmaceuticals
Y
Phosphate Fertilizers
Y
Phosphoric Acid
Y
Polybutadiene Rubber Production, P&R I
Y
Polycarbonates Production
Y
Polyether Polyols Production
Y
Polymers and Resins III
Y
Polymers and Resins IV
Y
Portland Cement
Y
Primary Aluminum
Y
Primary Lead Smelting
Y
Printing and Publishing
Y
Pulp and Paper I & III
Y
Secondary Aluminum
Y
Secondary Lead Smelters
Y
Shipbuilding and Ship Repair
Y
Steel Pickling-HCL Process
Y
15
-------
Evaluate 2008 NEI to Identify Areas of Improvement that Benefit Use in Residual Risk Assessments
Table 7. Initial Recommendation of Priority Source Categories
Source Category
Rule Published
Since 2008
Consent
Decree, Soon to
Start
Styrene Butadiene Rubber and Latex Production,
P&RI
Y
Wood Furniture
Y
Wool Fiberglass
S
a No consent decree issued yet, but work background work is on-going.
3.3.2 Source Categories Identified by 2008 NEI Toxicity Weighted Emissions
A pollutant emitted in high quantities does not necessarily present a higher risk to human health
than a pollutant emitted in very low quantities. The more toxic the pollutant, the more risk associated
with its emissions in ambient air. The development of various health-based risk factors is used to apply
weight to the emissions of pollutants based on toxicity rather than mass emissions.
The toxicity-weighted emissions approach developed by EPA (EPA, 2007) consists of the
following steps:
1. Obtain HAP emissions data from the 2008 NEI for point sources.
2. Apply the mass extraction speciation profiles to extract metal and cyanide mass.
3. Apply weight to the emissions derived from the steps above based on their toxicity.
a. To apply weight based on cancer toxicity, multiply the emissions of each pollutant by its
cancer unit risk estimate (URE). The URE is an upper bound estimate of an individual's
probability of contracting cancer over a lifetime of exposure to a concentration of one
microgram of the pollutant per cubic meter of air.
b. To apply weight based on noncancer toxicity, divide the emissions of each pollutant by
its noncancer reference concentration (RfC). The RfC is a concentration of the compound
in air thought to be without adverse effects even if a person is exposed continuously.
While the absolute magnitude of the pollutant-specific toxicity-weighted emissions is not
meaningful, the relevant magnitude of toxicity-weighted emissions is useful in identifying the order of
potential priority. Higher values suggest greater priority; however, even the highest values may not
reflect potential cancer effects greater than the level of concern (100 in-a-million) or potential noncancer
effects above the level of concern (e.g., HQ = 1.0).
In support of an EPA Enforcement project, the 2008 NEI for point sources was recently toxicity-
weighted for each HAP-emitting facility. Additionally, for EPA's 2010 Urban Air Toxics Monitoring
16
-------
Evaluate 2008 NEI to Identify Areas of Improvement that Benefit Use in Residual Risk Assessments
Program (UATMP) Final Report, a source category for each NEI facility was assigned based on its SCC,
Regulatory Code, or NAICS Code. Together, this information can be used to prioritize the top pollutants
by toxicity and by source category. The tables below use results from those two EPA projects.
Table 8 presents the source categories ranked by percent (%) contribution of the nationwide
cancer-toxicity weighting. Cancer weightings can be summed across pollutants. Rank order is based
on categories with the higher toxicity-weighted emission values - both in terms of the cumulative
toxicity weight across all HAPs analyzed and the amount of those toxicity weighted emissions. The
source categories which contribute to the cumulative Top 90% are listed. The bottom two entries in
the table summarize that 105 categories are in the remaining 10% of which 80 are potential RTR
categories and 25 are not.
Table 8. Top Source Categories Ranked by Cumulative Cancer Toxicity Weighting
Source Category
Potential
MACT or
RTR
Sources
Cancer
Toxic ity-
Weighting
(unitless)
%
Contribution
to Total Tox-
Weighted
Emissions
Cumulative
%
Contribution
Electricity Generation via
Combustion
Y
0.859
30%
30%
Coke Batteries
Y
0.254
9%
39%
Steel Mills
Y
0.245
9%
47%
Pulp and Paper Production
Y
0.129
5%
52%
Oil and/or Gas Production
Y
0.119
4%
56%
Electrical and Electronics
Y
0.083
-3 0/
59%
Equipment: Finishing Operations
J /O
Aircraft s
N
0.083
3%
62%
Miscellaneous Manufacturing
Industries
N
0.068
2%
64%
Rubber and Miscellaneous
Y
0.065
2%
66%
Plastics Products
Petroleum Refineries
Y
0.051
2%
68%
Iron and Steel Foundries
Y
0.051
2%
70%
Pipeline Compressor Stations
N
0.050
2%
72%
Electroplating, Plating, Polishing,
Anodizing, and Coloring
Y
0.049
2%
74%
Bulk Terminals/Bulk Plants
Y
0.046
2%
75%
Surface Coating Facility (Other)
N
0.043
2%
77%
Chemical Manufacturing
Y
0.037
1%
78%
Tank Batteries
Y
0.037
1%
79%
Primary Metal Production (Other)
N
0.034
1%
80%
Secondary Metal (Other)
Processing
N
0.033
1%
82%
Aerospace Industries
Y
0.033
1%
83%
17
-------
Evaluate 2008 NEI to Identify Areas of Improvement that Benefit Use in Residual Risk Assessments
Table 8. Top Source Categories Ranked by Cumulative Cancer Toxicity Weighting
Source Category
Potential
MACT or
RTR
Sources
Cancer
Toxic ity-
Weighting
(unitless)
%
Contribution
to Total Tox-
Weighted
Emissions
Cumulative
%
Contribution
Synthetic Organic Chemical
Manufacturing (HON)
Y
0.030
1%
84%
Industrial/Commercial/Institution
al Boilers & Process Heaters
Y
0.027
1%
85%
Plywood, Particleboard, OSB
Manufacturing
Y
0.027
1%
86%
Organic Chemical
Storage/Transportation
Y
0.026
1%
87%
Portland Cement Manufacturing
Y
0.020
1%
87%
Secondary Aluminum Production
Y
0.018
1%
88%
Lime Manufacturing
Y
0.017
1%
88%
Waste Solvent Recovery
Operations
Y
0.017
1%
89%
Wood Products/Manufacturing
Y
0.017
1%
90%
Miscellaneous Metal Parts &
Products (Surface Coating)
Y
0.016
1%
90%
Remaining 80 Other Potential
MACT source categories
Y
0.215
8%
98%
Remaining 25 Non-Potential
MACT source categories
N
0.066
2%
100%
Nationwide
2.865
Noncancer weightings can only be summed within a noncancer Target System, such as
neurological, respiratory, or developmental. There are 12 total Target Systems considered for NAT A
risk modeling. NATA05 establishes a priority for specific national and regional driver and
contributor pollutants and for noncancer risks, involve 6 of the 12 target systems
(http://www.epa.gov/ttn/atw/nata2005/05pdf/sum_results.pdf). Table 9 presents the NATA05
national and regional driver and contributor pollutants and their associated Target System, if
applicable.
Table 9. NATA05 National and Regional Pollutant
Drivers and Contributors
Pollutant
Target System
National Cancer Risk Driver
Formaldehyde
Respiratory
18
-------
Evaluate 2008 NEI to Identify Areas of Improvement that Benefit Use in Residual Risk Assessments
Table 9. NATA05 National and Regional Pollutant
Drivers and Contributors
Pollutant
Target System
Regional Cancer Risk Driver
Benzene
Immunological
PAHs
NA
Naphthalene
Respiratory
National Cancer
Risk Contributors
1,3-Butadiene
Reproductive
Arsenic compounds
Respiratory
Hexavalent chromium
Respiratory
Coke oven emissions
NA
Acetaldehyde
Respiratory
Acrylonitrile
Respiratory
Carbon tetrachloride
Liver
Ethylene Oxide
Neurological
T etrachloroethylene
Neurological
1,4-Dichlorobenzene
Liver
Ethylbenzene
Developmental
Regional Cancer
Risk Contributors
Nickel compounds
Respiratory/Immunological1
1,3 -dichloropropene
Respiratory
Methylene chloride
Liver
National Noncancer Hazard Driver
Acrolein
Respiratory
Regional Noncancer Hazard Drivers
2,4-Toluene Diisocyanate
Respiratory
Chlorine
Respiratory
Hexamethylene diisocyanate
Respiratory
Hydrochloric acid
Respiratory
Manganese compounds
Neurological
NA: Not applicable. Pollutant only has a cancer effect.
1 Nickel affects multiple target systems.
2
Diesel PM is listed as a Regional Noncancer Hazard Driver, PM2 5
from diesel engines is included the 2008 NEI.
The six Target Systems that guide the noncancer risk review priorities in NATA05 are
identified in Table 9 and are used here to help establish a category/ pollutant focus for this project. It
is appropriate to sum across all pollutants which affect the same Target System, and the results
presented in Table 10 include all available noncancer pollutants (i.e., those that have noncancer RfC
values) in the 2008 NEI for point sources. Table 10 presents 47 source categories which contributed
to the top 90% for the select noncancer target systems identified in Table 9. The percent contribution
values are shaded where equal to or greater than 10% of the national noncancer toxicity weighted
emissions for the noted target system.
19
-------
Evaluate 2008 NEI to Identify Areas of Improvement that Benefit Use in Residual Risk Assessments
3.3.3 Recommended Priority Source Category List
Using the above analyses, a compiled priority source category list is presented in Table 11.
There are 75 source categories and they are prioritized based on RTR analysis status and the toxicity
weight of HAP emissions. While these source categories are regulatory categories for the MACT
program, the emissions processes that contribute to these categories may be aggregated in different
ways for different sector descriptions. It is important to note that for the source categories presented
in Table 11, the RTR program does not have a definitive mapping of corresponding SCCs. Rather, it
is likely from the RTR perspective that the applicable SCCs will be developed after reviewing the
available data. Nevertheless, assessment of the 2008 NEI and beyond based on this priority list can
provide significant benefit in harmonizing efforts between the regulatory categories and the NEI.
The categories are grouped and prioritized for assessment in this project based on the following
criteria:
Priority
Group
Recent RTR Final
Date Since 2008 or
Soon to be Analyzed
Based on Consent
Decree Date
Contributes to Top 90%
of National Cancer
Toxicity Weighted
Emissions
Contributes to Top 90%
of the National
Noncancer Toxicity
Weighted Emissions
1
~
V
V
2
~
V
3
~
V
4
~
5
V
V
6
7
V
20
-------
Evaluate 2008 NEI to Identify Areas of Improvement that Benefit Use in Residual Risk Assessments REVISED DRAFT
Table 10. Listing of Source Categories Contributing to 90% of Noncancer Toxicity Weighting By Select Target System
Source Category
S3
13
fl
s
a
o
"oS
>•
a>
O
«
o
"So
o
"o
e
s
S
E
HH
Liver8
«
o
'So
o
"o
s
Z
Reproductive8
Respiratory8
Aircrafts
43%
10%
1%
30%
10%
Bulk Terminals/Bulk Plants
4%
32%
1%
Chemical Manufacturing
1%
3%
1%
2%
1%
Coke Batteries
1%
3%
1%
Electrical and Electronics Equipment: Finishing Operations
2%
2%
Electricity Generation via Combustion
9%
11%
2%
40%
15%
Ethanol Biorefineries
1%
1%
Ethylene Processes
1%
Ferroalloys Production
9%
Fiberglass Manufacturing
1%
Gasoline/Diesel Service Stations
3%
Glass Manufacturing
1%
Industrial/Commercial/Institutional Boilers & Process Heaters
1%
2%
1%
2%
Iron and Steel Foundries
3%
2%
4%
Landfill
1%
Lime Manufacturing
1%
Lumber/sawmill
1%
1%
Marine Vessel: Loading Rack
1%
Metal Can (Surface Coating)
1%
Military Base/National Security
2%
Miscellaneous Manufacturing Industries
2%
1%
4%
2%
2%
1%
Miscellaneous Metal Parts & Products (Surface Coating)
1%
1%
21
-------
Evaluate 2008 NEI to Identify Areas of Improvement that Benefit Use in Residual Risk Assessments REVISED DRAFT
Table 10. Listing of Source Categories Contributing to 90% of Noncancer Toxicity Weighting By Select Target System
S3
13
fl
s
a
o
"oS
>
C3
13
CJ
"5k
o
o
fl
3
s
Liver8
C3
13
CJ
"3d
o
o
u
s
«
o
S
73
O
a
C3
o
es
'a,
VI
Source Category
O
S
HH
z
a
Pi
Natural Gas Processing
1%
1%
Natural Gas Transmission and Storage
1%
3%
Oil and/or Gas Production
1%
13%
8%
25%
Organic Chemical Storage/Transportation
2%
3%
8%
9%
Petroleum Refineries
13%
i%
5%
Pharmaceutical Manufacturing
1%
Pipeline Compressor Stations
3%
3%
12%
Plastic Parts & Products (Surface Coating)
4%
Plywood, Particleboard, OSB Manufacturing
1%
1%
2%
3%
6%
Portland Cement Manufacturing
1%
5%
1%
2%
Primary Lead Smelting
3%
Primary Metal Production (Other) Production
2%
Printing, Coating & Dyeing of Fabrics
1%
Printing/Publishing Facilities
1%
2%
Pulp and Paper Production
3%
6%
20%
6%
8%
Secondary Metal (Other) Processing
5%
6%
Solid Waste Disposal - Industrial
1%
Steel Mills
8%
11%
Surface Coating Facilities (Other)
1%
1%
1%
2%
2%
1%
Synthetic Organic Chemical Manufacturing (HON)
4%
4%
19%
Tank Batteries
2%
2%
1%
1%
Waste Solvent Recovery Operations
1%
22
-------
Evaluate 2008 NEI to Identify Areas of Improvement that Benefit Use in Residual Risk Assessments REVISED DRAFT
Table 10. Listing of Source Categories Contributing to 90% of Noncancer Toxicity Weighting By Select Target System
Source Category
C3
13
c
s
a
o
"oS
>•
a>
O
C3
13
CJ
"5k
o
o
fl
3
s
s
HH
Liver8
C3
13
CJ
"3d
o
o
u
3
4)
z
Reproductive8
Respiratory8
Wastewater Treatment
1%
Wood Building Products (Surface Coating)
1%
1%
Wool Fiberglass Manufacturing
2%
Cumulative % Contribution
90%
90%
90%
90%
90%
90%
Remaining Source Categories
10%
10%
10%
10%
10%
10%
Total Noncancer Toxicity Weighting (unitless)
4,321,924
225,969
78,378
32,264,356
808,009
202,316,234
a Percent contribution value equal to or greater than 10% of the national noncancer toxicity weighted emissions for an analyzed target system are shaded.
23
-------
Evaluate 2008 NEI to Identify Areas of Improvement that Benefit Use in Residual Risk Assessments
Table 11. Recommended Assessment Priority for Source Categories Based on RTR
Analysis Schedule and Toxicity-weighting Results
Source Category
Priority
Grouping
RTR
Dates
Cancer
Toxicity
Noncancer
Toxicity
Oil and/or Gas Production
1
Petroleum Refineries
1
Sb
SA
Portland Cement Manufacturing
1
V
-/
Pulp and Paper Production
1
Se
Aerospace Industries
2
V
Secondary Aluminum Production
2
Ferroalloys Production
3
-/
Marine Vessel Loading
3
^a
s
Natural Gas Transmission and Storage
3
^a
s
Pharmaceuticals
3
^a
Primary Lead Smelting
3
^a
s
Wool Fiberglass Manufacturing
3
Acrylic/Modacrylic Fibers
4
Chromium Anodizing Tanks
4
^a
Decorative Chromium Electroplating
4
^a
Epichlorohydrin Elastomers Production,
P&R I
4
^a
Flexible Polyurethane Foam Production
4
Hard Chromium Electroplating
4
^a
Mineral Wool Production
4
Nitrile Butadiene Rubber Production, P&R I
4
^a
Off-Site Waste Recovery Operations
4
Pesticide Active Ingredient Production
4
Phosphate Fertilizers Manufacturing
4
Phosphoric Acid Manufacturing
4
Polybutadiene Rubber Production, P&R I
4
^a
Polycarbonates Production
4
Polyether Polyols Production
4
Polymers and Resins III
4
Polymers and Resins IV
4
Primary Aluminum Production
4
Printing and Publishing Production
4
~a
Secondary Lead Smelters
4
~a
Shipbuilding and Ship Repair
4
~a
Steel Pickling-HCL Process
4
~a
Styrene Butadiene Rubber and Latex
Production, P&R I
4
~a
24
-------
Evaluate 2008 NEI to Identify Areas of Improvement that Benefit Use in Residual Risk Assessments
Table 11. Recommended Assessment Priority for Source Categories Based on RTR
Analysis Schedule and Toxicity-weighting Results
Source Category
Priority
Grouping
RTR
Dates
Cancer
Toxicity
Noncancer
Toxicity
Wood Furniture Production
4
Aircrafts
5
Bulk Terminals/Bulk Plants
5
Chemical Manufacturing Facility
5
SA
V
Coke Batteries
5
SA
-/
Electrical and Electronics Equipment:
Finishing Operations
5
SA
-/
Electricity Generation via Combustion
5
~d
Industrial/Commercial/Institutional Boilers
& Process Heaters
5
~
~
Iron and Steel Foundries
5
~d
~
Lime Manufacturing
5
~
~
Miscellaneous Manufacturing Industries
5
~
Miscellaneous Metal Parts & Products
(Surface Coating)
5
s
V
Organic Chemical Storage/Transportation
5
s
V
Pipeline Compressor Stations
5
sd
ve
Plywood, Particleboard, OSB Manufacturing
5
V
V
Primary Metal Production (Other)
5
VA
s
Secondary Metal (Other) Processing
5
s
s
Steel Mills
5
Surface Coating Facilities (Other)
5
SA
y
Synthetic Organic Chemical Manufacturing
(HON)
5
V
Tank Battery Facilities
5
SA
V
Waste Solvent Recovery Operations
5
V
V
Wood Building Products Manufacturing
(Surface Coating)
5
s
V
Electroplating, Plating, Polishing, Anodizing,
and Coloring
6
SA
Rubber and Miscellaneous Plastics Products
Manufacturing
6
SA
Ethanol Biorefineries
7
V
Ethylene Processes
7
V
Fiberglass Manufacturing
7
V
Gasoline/Diesel Service Stations
7
V
Glass Manufacturing
7
V
Landfills
7
V
25
-------
Evaluate 2008 NEI to Identify Areas of Improvement that Benefit Use in Residual Risk Assessments
Table 11. Recommended Assessment Priority for Source Categories Based on RTR
Analysis Schedule and Toxicity-weighting Results
Source Category
Priority
Grouping
RTR
Dates
Cancer
Toxicity
Noncancer
Toxicity
Lumber/sawmills
7
~
Metal Can (Surface Coating)
7
-/
Military Base/National Security Facilities
7
-/
Natural Gas Processing
7
-/
Plastic Parts & Products (Surface Coating)
7
-/
Printing, Coating & Dyeing of Fabrics
7
-/
Printing/Publishing Facilities
7
-/
Solid Waste Disposal - Industrial
7
S
Wastewater Treatment
7
S
a Consent decree date for final within 2-6 years.
b No consent decree issued, but RTR being conducted through settlement agreement.
L RTR final date since 2008.
d Categories that are included in top 80% of national cancer toxicity weighted emissions.
e Categories with a percent contribution value equal to or greater than 10% of the national noncancer toxicity
weighted emissions for an analyzed target system.
26
-------
Evaluate 2008 NEI to Identify Areas of Improvement that Benefit Use in Residual Risk Assessments
4.0 Usefulness of the 2008 NEI for Recent RTR Work
In support of Task 3 of the WA, EPA staff from the EIAG and SPPD met to discuss the
usefulness of the 2008 NEI in recent RTR activities. In the past, RTR rulemaking relied on
emissions inventory data from previous NEIs (2002 and 2005), Toxics Release Inventory (TRI),
and/or industry data. With the release of 2008 NEI Version 1.5 in Spring 2011 and Version 2 in
Spring 2012, EIAG wanted to understand if these inventories were being considered in recent
RTR rulemaking. An evaluation of key data elements needed for RTR modeling is presented
above in Section 3.
On October 25, 2012, EPA staff met to discuss the usefulness of the 2008 NEI in recent
RTR activities. Specific attendees were:
• Ms. Lee Tooly, EPA EIAG/Work Assignment Manager, meeting facilitator
• Ms. Anne Pope, EPA SPPD
• Mr. Brian Shrager, EPA SPPD
• Ms. Andrea Siefers, EPA SPPD
• Ms. Darcy Wilson, ERG, EPA contractor
• Mr. Regi Oommen, ERG, EPA contractor
Anne Pope facilitates emissions data gathering for the SPPD staff conducting the RTR analysis.
Andrea Siefers worked on the previous chemical sector review which included as many as ten
MACT categories, and used the available emissions data from the 2005 National Air Toxics
Assessment (NATA) to develop an initial emissions information form for industry group(s) to
review. Brian Shrager is the SPPD RTR Coordinator, and was invited to describe the general
expectations of the NEI for the RTR work. His thoughts include the following:
• NEI is one tool among several to consider using as a starting point and especially if
no other data sources are readily available and if the expected emission magnitude for
the sector of interest is small, i.e., not likely to pose a large risk.
• NEI is particularly useful when conducting scoping exercises for one or multiple
categories to investigate the potential extent of risk due to hazardous air pollutant
(HAP) emissions. Other emission data sources consulted for such initial scoping
exercises include the EPA's NATA and the Toxic Release Inventory (TRI).
• It is expected that for some of the upcoming RTR work, the 2008 NEI will be
considered as a starting point to develop an initial dataset to invite industry groups to
review. A review mechanism includes referring the interested industry group to
available NEI point source data posted on the Web.
For discussion during the meeting, Ms. Tooly prepared and distributed a list of charge
questions. They are presented in Table 12.
27
-------
Evaluate 2008 NEI to Identify Areas of Improvement that Benefit Use in Residual Risk Assessments
Table 12. Charge Questions Evaluating Usefulness of 2008 NEI in RTR Activities
#
Question
1
What RTR sectors have recently or are presently using the 2008 NEI?
2
What version(s) of 2008 NEI is used?
3
How were the data accessed?
4
What data access method is working well?
5
What are you trying to do? The NEI is intended to help you do what?
6
What are your expectations for the NEI data?
7
How good or reliable do the data have to be?
8
What measures determine reliability
9
Based on what you were trying to do, specifically what 2008 NEI data were
helpful to satisfy your objective(s)?
10
What specific NEI data fell short of being useful for your purpose?
11
Overall, how useful was the 2008 NEI for your RTR work?
12
Are there specific data that needs to improve for future NEI data to be more useful
for RTR work?
The above questions were discussed and the responses are summarized below and
reflect additional review and comments provided by attendees. Additional information is
included by the meeting facilitator to support responses where considered helpful to clarify
current or anticipated operating conditions.
Question 1: What RTR sectors have recently or are presently using the 2008 NEI?
Response: 2008 vl.5 NEI data were used in the RTR Prioritization Task to help to inform
management in the selection of the next categories for analysis. That scoping exercise included
more than 50 MACT categories. 2008 vl.5 NEI data were also used to gap fill missing facility
HAP data for some of the facilities in the Flexible Polyurethane Foam Production (PUF) RTR
category. The 2008 v2 NEI data are currently being used to support some initial scoping work
for three chemical sector source categories (Ethylene Processes, Miscellaneous Organic
NESHAP (MON), and Organic Liquids Distribution). While these source categories are not
currently part of any RTR package, data are being reviewed in preparation for potential future
RTR or other rulemakings.
Question 2: What version(s) of the 2008 NEI is used?
Response: The RTR program started before 2008 NEI data were available, and initially used the
2002 NEI and then the 2005 NATA NEI. As 2008 became available, early category work
included review of 2008vl .5. For current RTR projects, the 2008 NEI v2 are being reviewed.
See the above question/ response for specific examples.
28
-------
Evaluate 2008 NEI to Identify Areas of Improvement that Benefit Use in Residual Risk Assessments
Question 3: How was/ is the (2008 NEI) data accessed? e.g., CHIEF website — large file
downloads, EIS reports, EIAG-assistedEIS data extractions
Response: Initially, SPPD requested data from EIAG on a facility-by-facility basis. SPPD staff
can also access the 2008 NEI data via the EIS Gateway, but the reports in the EIS Gateway do
not allow for large (batch) facility-specific extractions. RTR source categories may be applicable
to several hundred facilities, and query by the individual facility is not often feasible. Detailed
facility emissions data from the EIS Gateway are provided as relational data tables and are too
cumbersome for SPPD staff to use in a timely manner. A supporting contractor was engaged to
extract the entire 2008vl.5, and translate the data format to a "one record per line" (flat) format
which is commonly used for emissions modeling. The flat-file format is similar to the ORL
format utilized by the SMOKE emissions model to import HAP data. The flat file format used
for the RTR risk modeling is prepared by extracting key data elements from the EIS facility
relational tables (e.g., facility, unit, process, release point, etc.) and adding descriptions for coded
elements (e.g., pollutant codes, MACT codes, FIPS codes, etc.). Currently, the contractor is
working to download the entire 2008v2 to support RTR data requests.
Additional supporting information: EIAG with the assistance of the Information Transfer Group
(EIS administrators) delivered to SPPD the 2008 v2 NEI point sources data from EIS that
contains both the facility emissions and physical configuration information. The data were
provided in the relational table format. Facilities that are not of interest may be filtered out to
greatly reduce the number of records in the data set. EIS report enhancement requests for 2013
include attaching all of the facility emissions information to the current facility configuration
reports which would allow one extraction from the EIS Gateway for both facility emissions and
configuration information and result in a much reduced number of relational tables to manage.
Report enhancement requests also include allowing EIS Gateway users to build a large list of
specific facilities for data extraction.
Question 4: What data access method is working well?
Response: It is better for SPPD leads to be able to request data in large batches for specific
facilities, in an ORL (flat) format rather than relational tables. See above question/ response
regarding future EIS report enhancement to allow extraction at the Gateway for large list of
specific facilities.
Question 5: What are you trying to do? The NEI is intended to help you do what?
Response: The NEI data are one of several tools for initiating RTR category projects.
Specifically:
1. The NEI is the starting point for data review. SPPD in the past sent 2005 NATA
emissions data to industry for review and revision ("voluntary QA"); and
2. The NEI is used in completeness checks for HAPs emitted from facilities in a given
category, and to assess historical trends.
29
-------
Evaluate 2008 NEI to Identify Areas of Improvement that Benefit Use in Residual Risk Assessments
To accomplish this, it is desired that the NEI have complete and accurate data, and not
over-report HAP emissions.
Question 6: What are your expectations for the NEI data?
Response: Ideally, the NEI data would be used as-is for risk modeling with little or no revisions.
SPPD staff experience observes the need for more accurate and complete data for rulemaking
activities. This recognizes the complexity of preparing nationwide inventories, the "voluntary"
and potentially inconsistent or incomplete HAP reporting to the NEI, and the associated lag time
from the end of the base year and the release of data (in the past usually 2-3 years). Thus, SPPD
views the highest quality data to be those collected from facilities with Information Collection
Requests (ICRs), and quality assured by RTR leads. For some RTR categories, EPA implements
ICRs for ah facilities. In the absence of ICR, or prior to conducting an ICR, the NEI data are
reviewed, along with the most recent Toxics Release Inventory (TRI) data. However, for scoping
studies, the NEI emission levels for particular categories are reviewed - if they appear to be
representative and emissions are "low," the rule is given lower priority and effort for rule
development. In reality, the NEI data (any version) as well as TRI data are just the starting point
for developing an emission profile for specific facilities— the end point is often very different.
Question 7: How good or reliable do the data have to be?
Response: Litigation outcomes have indicated that ICR data provided by facilities are viewed as
accurate data and may be considered to have the most reliability. The data need to be complete
and model-ready, not just "representative." That being said, SPPD sees value in using the NEI
for scoping tasks to determine program priorities. SPPD quality assures the important parameters
needed for risk modeling, including the emission release point (ERP) coordinates. For example,
for risk modeling, facility-level coordinates are not accurate enough, and thus the coordinates
need to be accurate within 10 feet for risk assessments (e.g., emission release point coordinates
need to be reported to the fifth decimal point). Some facilities are near Environmental Justice
communities, and the difference in under-reporting of decimal points may inaccurately place
estimated concentrations in the wrong census block. Accurately representing emission releases
for fugitive sources, such as the dimensions of length, width, release height, and release angle, is
also critical.
Additional supporting information: A list of the important parameters for risk modeling that is
prioritized by SPPD for review and QA is in the Task 2 memo - "Prioritize the Type of
Improvements of Most Significant Benefit to RTR and Establish Assessment Focus". EIAG also
prioritizes review and QA of both the facility and the emission release point coordinates in EIS
with a goal to verify and improve the accuracy of the release point location. Possible errors in
release point coordinates are determined by investigating reasonable proximity to the facility
center. Possible errors in the facility center coordinates reported are determined based on relative
county boundaries and comparing to facility reported data in other EPA data systems - the
Facility Registry System (FRS) and TRI. Data corrections are made through geo-addressing
confirmations assisted by use of Google Maps. The typical outcome of EIAG's QA of facility
and emission release point coordinates is at least 4 digits after decimal, and is usually 5 or
30
-------
Evaluate 2008 NEI to Identify Areas of Improvement that Benefit Use in Residual Risk Assessments
6 digits. For emission release points where large deltas from the facility center are identified, and
it is confirmed that the facility is not actually that large, the release point coordinates are set
equal to facility center.
Question 8: What measures determine reliability?
Response: The most important measure is EPA and industry acceptance of the emissions data.
Facilities often don't recognize the data in the NEI that state and local agencies provide to EPA.
Detailed emissions data may be available, but not in the NEI. For example, SPPD data
verification steps for the Boiler MACT rule necessitated their review of available facility test
report from which data were used in calculating MACT floor emission limits. Other measures
include: the availability of expected pollutants in a consistent manner across applicable facilities
and processes; use of actual data instead of surrogate default data in stack parameters (i.e., when
stack parameter data are reported incomplete or determined in error, average default data may be
used); correct identification of processes subject to certain regulations, e.g., assignment of
regulatory codes; and use of the appropriate source classification codes (SCCs) to describe the
process. For example, emissions associated with Portland Cement Manufacturing MACT
processes should be coded as 30500606 (dry kilns), 30500706 (wet kilns), 30500622 (preheater
kiln), or precalciner kiln (30500623). In preparation of the emission inventory file for this
category, many processes were miscoded, and had to be corrected prior to emissions modeling.
Question 9: Based on what you were trying to do, specifically what 2008 NEI data were
helpful to satisfy your objective(s) ?
Response: The NEI was useful in prioritizing specific RTR source categories for program review
based on toxicity-weighting the emissions inventory.
Question 10: What specific NEI data fell short of being useful for your purpose?
Response: A few items were noted:
1. Timeliness has been an issue - between request and receipt of data from EIAG;
2. The data format of the relational data tables are not easy to work with;
3. Completeness issues in terms of facility and units, source category, and HAP coverage;
4. Lack of specificity in certain cases of source category (e.g., generic SCCs vs. specific
SCCs) and pollutant (e.g., strontium chromate is not an EIS applicable pollutant though it
may be noted in industry provided data);
5. Facility configuration sometimes not consistent between NEI and the industry-provided
ICR data (e.g., number of units/ processes, stack parameters and locational coordinates);
6. Lack of detailed and complete control information such as control devices, control
efficiencies, and capture efficiencies;
7. Lack of coordinates for individual release point and QA of individual emission release
point coordinates to insure coordinates are correctly located on stacks and fugitive
releases;
8. Fugitive releases are not often reported with fugitive length, width and angle;
31
-------
Evaluate 2008 NEI to Identify Areas of Improvement that Benefit Use in Residual Risk Assessments
9. Lack of information at process level on units/processes subject to regulation; and
10. Source of data not available for a number of data fields including the basis of emission
estimates, coordinates, permit IDs, and stack parameters.
Additional supporting information: See the supporting information included for Questions 7 and 12.
Question 11: Overall, how useful was the 2008 NEI for your RTR work?
Response: Useful as one tool among other data sources; especially useful in prioritizing review
for a number of RTR source categories.
Question 12: Are there specific data that need to improve for future NEI data to be more
useful for RTR work?
Response: Some data and coordination issues were noted. Issues relevant for the data quality
nature of the review and corrective action are listed first in items 1 -7 and coordination issues are
listed in 8-10.
1. Having the Facility Registry System (FRS) identifier populated and accurate;
2. SPPD needs to be able to identify stack parameters that are defaults;
3. SPPD needs to be able to identify the basis for the emissions estimate (i.e., calculated,
continuous measurement, engineering judgment); If calculated , SPPD needs to be able to
review emission factors, the source of the factors used, and the process activity data for
applicable year(s);
4. Need for more detailed control information;
5. Need to be able to assign regulatory codes based on specific SCCs (rather than generic
SCCs), unit descriptions, etc. For example, some rules such as Cooling Towers are
process- and HAP-specific;
6. In some cases, rolled-up HAPs are not accurate for use in risk assessments. The
Aerospace category is an example where coatings contain many different types of
chromium combinations. SPPD also needs dioxins reported in terms of congeners
preferably or as 2005 WHO TEQs;
7. SPPD needs to assign "Emission Process Group" to RTR source category processes to
better characterize risk modeling results and evaluate control technologies;
8. Having the data come in a more usable format;
9. Being able to get the data in a timely manner from EIS or EIAG; and
10. More of a coordinated effort between EIAG, the EPA Regional Office Emission
Inventory leads, and the SLT agencies about the need for high quality HAP emissions
and ancillary data needed for RTR activities.
32
-------
Evaluate 2008 NEI to Identify Areas of Improvement that Benefit Use in Residual Risk Assessments
Additional supporting information: Stack parameters reported by the agencies to EIS are checked
to determine if reasonable. If the QA check results in possible error, the data are rejected. The
state and local agencies are correcting a lot of stack parameter data. EIAG no longer defaults
stack parameters. Stack parameters that have not been verified by agencies may be legacy data
from initial load to the EIS. As of November 2012, automated data checks in EIS include more
stringent consistency requirements and data rejection rules for stack parameter data. The
emission calculation method is required when reporting emissions and is available in EIS
Gateway reports - Facility Emissions Snapshot, see Emissions table; and Facility Emissions
Summary report for process-level.
33
-------
Evaluate 2008 NEI to Identify Areas of Improvement that Benefit Use in Residual Risk Assessments
5.0 Assess 2008 NEI for Missing HAPs
In Task 4 of the WA, EPA assessed the 2008 NEI v2 for missing or under-reported
HAPs. The findings from this task were initially intended to focus on the priority source
categories, and their associated SCCs, identified in Table 11 in Section 3. However, the
majority of these priority source categories are scheduled for analysis within the next 2-6
years and will not typically be defined until the beginning of rulemaking activities. Thus,
assignment of SCCs for each of the priority source categories presented in Table 11 was
not available at the time of this project. Nevertheless, assessment of the 2008 NEI and
beyond based on this priority list can provide significant benefit in harmonizing efforts
between the regulatory categories and the NEI.
Prior to performing the analyses, a memorandum was prepared outlining the
proposed approach, the readily available references that would be consulted, external
support and data needs, timeline, and the expected outcomes. EPA conducted the
following analyses to accomplish this task:
• Develop a matrix of expected HAPs by SCC after reviewing selected data
references;
• Estimate potentially missing HAP-YOC and HAP-PM for nonpoint sectors; and
• Evaluate for point sources the TRI Data usage in RTR.
5.1 Matrix of Expected HAPs
Several data references were consulted to understand what HAPs were reasonable
to expect may be present for SCC process emissions. The SCCs listed in the matrix are
those that were referenced by the different data sources consulted. As a result, both point
and nonpoint SCCs are represented in the matrix. As noted above, the SCCs in the matrix
do not have a corresponding mapping at this time to the priority regulatory source
categories listed in Table 11, but are mapped to the EIS source sectors which will benefit
these regulatory source categories due to overlapping information. For example, while the
universe of applicable facilities and processes may not be defined yet for the Oil and Gas
RTR, the matrix presents expected HAPs for 114 point and nonpoint SCCs which may be
evaluated for Oil & Gas rulemaking activities.
As this task was unfolding, it became evident that extracting information was more
efficient from certain data references that are available via a large download file. For
example, while it is possible to search WebFIRE by entering SCCs online, the more
efficient option was to download the entire database that was available on the WebFIRE
website. This type of efficiency, where available, is described below for the data
references used. Additionally noted are some data references that were proposed in the
Work Plan but were not reviewed due to time and resource constraints. Some of those are
discussed in terms of pilot-level exercises to further explore and plan improved procedures
for future compilation cycles of the NEI.
34
-------
Evaluate 2008 NEI to Identify Areas of Improvement that Benefit Use in Residual Risk Assessments
5.1.1 Data References
Expected HAPs described by SCC will be the most useful to EPA's NEI data QC and
estimation efforts. The data resources listed below were reviewed to derive expected HAPs by
SCC or NAICS-level. For each data source below, the level of data detail is identified. The first
order priority for data gathering was placed on data sources with ready available SCC-level HAP
information.
Data Reference 1 - EPA Regulatory Websites
For this data reference, EPA reviewed the SPPD MACT, Area Source, Solid Waste, and
Special promulgated rules relating to air emissions, such as for Municipal Waste Combustors,
Electric Arc Furnaces, and Coke Ovens. The regulatory websites for these promulgated source
categories are presented in the Table 13 below, with the corresponding website address.
Table 13. SPPD MACT/Regulatory Websites
Type
Website
MACT Standards
http ://www. epa. gov/ttn/ atw/mactfnlalph. html
Area Source Standards
http://www.epa.gov/ttn/atw/area/arearules.html
Solid Waste Rules
(these rules cover 9
specific HAPs)
http ://www. epa. gov/ttn/ atw/129/ssi/ ssipg. html
http ://www. epa. gov/ttn/ atw/129/hmiwi/rihmiwi. html
http ://www. epa. gov/ttn/ atw/129/mwc/rimwc. html
http ://www. epa. gov/ttn/ atw/129/mwc/rimwc2. html
http://www.epa.gov/ttn/atw/129/ciwi/ciwipg.html
http://www.epa.gov/ttn/atw/129/oswi/oswipg.html
Mercury Air Toxics
Standards
http://www.epa.gov/airtoxics/utility/utilitypg.html
Lead NAAQS
http://www.epa.gov/airquality/lead/kitinventory.html
In total, over 130 regulatory webpages were reviewed to generate a matrix of regulatory
codes and pollutants. The regulatory codes were then translated to applicable SCCs based on the
2005 NATA NEI and RTR inventory.
35
-------
Evaluate 2008 NEI to Identify Areas of Improvement that Benefit Use in Residual Risk Assessments
Data Reference 2 - RTR Database for Modeling Files
The RTR is a combined effort to evaluate both risk and technology as required by the
Clean Air Act (CAA) after the application of maximum achievable control technology (MACT)
standards. Section 112(f) of the CAA requires EPA to complete a Report to Congress that
includes a discussion of methods the EPA would use to evaluate the risks remaining after the
application of MACT standards. These are known as residual risks. EPA published the Residual
Risk Report to Congress in March 1999. Section 112(f)(2) directs EPA to conduct risk
assessments on each source category subject to MACT standards, and to determine if additional
standards are needed to reduce residual risks. Section 112(d)(6) of the CAA requires EPA to
review and revise the MACT standards, as necessary, taking into account developments in
practices, processes and control technologies.
Although the RTR emissions modeling files are available with its associated rulemaking
documents (http://www.epa.gov/ttn/atw/rrisk/rtrpg.htmn. EPA utilized the master RTR
emissions database to extract the information needed. RTR emissions data are available at the
SCC-level by facility, and thus a matrix of SCCs and pollutants was generated.
Data Reference 3 - WebFIRE Database
WebFIRE is EPA's online emissions factor repository, retrieval, and development tool
(http://cfpub.epa.gov/webfire/). The WebFIRE database contains EPA's recommended emissions
factors for criteria and hazardous air pollutants (HAP) for industrial and non-industrial processes.
In addition, WebFIRE contains the individual data values used to develop the recommended
factors and other data submitted to EPA by federal, state, tribal, and local agencies; consultants;
and industries. For each recommended emissions factor and individual data value, WebFIRE
contains descriptive information such as industry and source category type, control device
information, the pollutants emitted and supporting documentation. Emission factors from EPA's
AP-42 are also housed in WebFIRE.
EPA is currently re-engineering WebFIRE to be more user-friendly. On-line searches of
the WebFIRE database can be done through querying; however, the database behind WebFIRE is
also available and that file was extracted and imported into a database for this analysis. Since
WebFIRE data are at the SCC-level, a matrix of SCCs and pollutants were generated.
Data Reference 4 - SPECIATE
SPECIATE is EPA's repository of volatile organic gas and particulate matter (PM)
speciation profiles of air pollution sources. These emissions source profiles can be used to
identify HAPs from specific source categories. EPA recently finalized SPECIATE Version 4.3,
located at: http://www.epa.gov/ttn/chief/software/speciate/index.html.
The SPECIATE 4.3 database includes a total of 5,592 PM, volatile organic gases, and
other gases profiles. Instead of searching through the on-line SPECIATE website the database
behind SPECIATE was extracted and imported into a database for this analysis. SPECIATE
36
-------
Evaluate 2008 NEI to Identify Areas of Improvement that Benefit Use in Residual Risk Assessments
profiles are available at the source category and pollutant level. As such, a crosswalk of
SPECIATE source categories to SCC was developed, and a matrix of SCCs and pollutants was
generated.
Data Reference 5 - Toxic Release Inventory (TRD
EPA's Toxics Release Inventory (TRI) is a database containing data on disposal or other
releases of over 650 toxic chemicals from thousands of U.S. facilities, and information about
how facilities manage those chemicals through recycling, energy recovery, and treatment. One of
TRI's primary purposes is to inform communities about toxic chemical releases to the
environment. TRI data are released annually, with the most recent year available being the 2011
preliminary dataset (http://www.epa.gov/tri/tridata/preliminarydataset/index.htmn. TRI data can
go back many years, and are typically based on threshold usage of chemicals. As such, it may be
possible that HAPs are reported in one TRI year, but not another. Thus, EPA reviewed multiple
years of TRI data 2008-2011 for this analysis.
Searches can be made on the TRI website (http://www.epa. gov/tri/). however, the
database behind TRI for the years of interest is available and those files were extracted and
imported into a database for this analysis. TRI data are available by facility name and the North
American Industrial Classification System (NAICS) code and pollutant level. As such, EPA
developed a matrix of NAICS codes and pollutants.
Data Reference 6 - Special Studies - CenSARA
EPA also reviewed two nonpoint inventories recently prepared by the Central States
Air Resource Agencies (CenSARA). CenSARA encompasses nine states: Arkansas, Iowa,
Kansas, Louisiana, Minnesota, Missouri, Nebraska, Oklahoma, and Texas. In 2012,
CenSARA developed two nonpoint source emissions inventories on behalf of its member
states. The first inventory focused on 2011 area source combustion emissions from the
institutional/commercial/industrial and residential sectors. The second inventory was for
2011 oil and gas emissions.
The pollutant focus for each inventory was different. While the combustion
emissions inventory developed emission estimates for all possible HAPs, the oil and natural
gas emissions inventory focused on 18 priority HAPs. Regardless, a matrix was developed
for each inventory by SCC and pollutant. These results are presented by nonpoint SCC.
Data Reference 7 - Special Studies - Barnett Shale Emissions Inventory
EPA also reviewed an emissions inventory developed by the Texas Commission on
Environmental Quality (TCEQ) for the Barnett Shale oil and natural gas activities
(http://www.tceq.texas.aov/airqualitv/point-source-ei/psei.html). The Barnett Shale is a
23-county geographical area located in North-Central Texas. As part of this inventory,
TCEQ prepared a calculator of expected pollutants for 14 source categories. After analysis
of this calculator, there were five priority pollutants that were estimated for each of the
37
-------
Evaluate 2008 NEI to Identify Areas of Improvement that Benefit Use in Residual Risk Assessments
14 source categories, and those same pollutants were also in the CenSARA emissions
inventory. However, the wellhead compressor engines contained additional HAPs not in the
CenSARA emissions inventory, and were included for this analysis. A matrix of wellhead
compressor types (2-stroke lean, 4-stroke lean, and 4-stroke rich burn) and pollutant were
generated. These results are presented by nonpoint SCC.
5.1.2 Master Matrix Results
All of the information from the above data references is compiled in a master matrix
of expected HAPs. The HAPs are grouped by HAP category name according to EPA's EIS
master pollutant dictionary. The compiled version is presented in Appendix A-l as a MS
Access database. Based on the data sources consulted, 180 HAPs are listed as expected for
3,529 specific SCCs. Those source category (represented by EIS Sector), SCC, pollutant
combinations are represented in the "Expected HAPs by SCC" table matrix. Over 500
NAICS codes for 159 pollutants are represented in the "Expected HAPs by NAICS" matrix.
Each data table matrix in the MS Access database is in the format presented in Table 14.
Summaries of the results are discussed and presented in the Appendix A-l.
Table 14. Example Format for Expected HAPs Matrix - NEI Point Sources
EIS Sector
SCC or
NAICS
Data Reference
HAP Reported
Acetaldehyde
Acetamide
Acetontrile
Antimony
Benzene
Vinylidene
Chloride
Xylenes
Industrial
Processes -
Petroleum
Refineries
30601001
MACT Rule
y
y
30601001
RTR Database
S
30601001
SPECIATE
S
30601101
MACT Rule
S
30601101
WebFIRE
30601101
RTR Database
S
30601101
SPECIATE
38
-------
Evaluate 2008 NEI to Identify Areas of Improvement that Benefit Use in Residual Risk Assessments
5.1.3 Other Data References Not Considered
Three data references that were previously identified in the Work Plan as possible
sources of information were not reviewed due to time constraints. They are presented here
as possible data references which can be evaluated in the future.
EPA Oil and Natural Gas Emission Inventory Tool
EPA's EI AG is currently developing a 2011 oil and natural gas emissions inventory
for the U.S. A consistent set of methods will be used to estimate pollutant emissions for
each oil and natural gas source category (SCC) evaluated. This emissions inventory can be
used to identify missing, under-reported, or incorrectly reported HAPs for this sector. For
example, in conversations with the State of Wyoming, it is believed that only a limited
number of HAPs are reported for their emission sources, and HAPs may be missing that can
otherwise be estimated. Expected level of detail from this emission inventory would include:
source category information, pollutants, and emissions/emission factors. Unfortunately, the oil
and gas emissions inventory has not been completed in time for this report.
Permitting Databases
Another data reference resource that may be useful to review is permit data for
facilities that are identified as sources of interest. At a minimum, a "pilot" approach could
be developed by comparing sources and pollutants within the permit to inventory data
submitted to the 2008 NEI v2. The results of the approach could identify additional HAPs to
be expected from emission sources. Additionally, the results can be shared with state/local
agencies in preparation for their 2011 emission inventory submittals. It is recommended that
priority be placed on reviewing Title Y permits that are located within each EPA Regional
Office's website, as presented in Table 15 below.
Table 15. EPA Regional Office Permitting Websites
EPA
Region
Website
1
http://www.epa.gov/regionl/eco/permits/title5/where.html
2
http://www.epa.gov/region2/air/permit/title v database.htm
3
http://www.epa.gov/reg3artd/permitting/petitions3.htm
4
http://www.epa.gov/region4/air/permits/index.htm
5
http://www.epa.gov/region5/air/permits/index.html
6
http://yosemite.epa.gov/r6/Apermit.nsf/AirP
7
http://www.epa.gov/region7/air/title5/titlevhp.htm
8
http://www.epa.gov/region8/air/permitting/index.html
9
http://www.epa.gov/region9/air/permit/title-v-permits.html
10
http://yosemite.epa.gov/R10/AIRPAGE.NSF/Permits/tvop/
39
-------
Evaluate 2008 NEI to Identify Areas of Improvement that Benefit Use in Residual Risk Assessments
Expected level of detail from permits would include: site-level information (e.g., facility
name, AFS ID, address, etc.), SCCs, pollutants, and actual/potential emissions. Due to time
constraints, it was not feasible to query permitting files from every state permitting program.
Conversely, the matrix of expected HAPs from this analysis could be distributed to EPA
regional offices as a resource to consider when reviewing Title 5 or New Source Review
permits.
Compare Ambient Monitoring Results to Emission Inventory Data
A final alternative approach to identify missing HAPs is to utilize ambient monitoring
data. EPA is currently developing Phase VII air toxics monitoring archives ("Archive"), which
houses HAP concentrations from 1973 to 2010. In theory, ambient concentrations at monitoring
sites near emission sources can serve as a validation check on emission inventories by comparing
pollutant species. For example, if ambient monitoring data near an emissions source is detecting
vinyl chloride, then the emission inventory for that source would be expected to emit vinyl
chloride. However, this type of analysis is not a perfect indicator of missing HAPs, as ambient
measurements can reflect the transport of air parcels that are not located near the source.
Additionally, pollutants that are ubiquitous, such as carbon tetrachloride, are not directly emitted
by stationary sources, but remain present in the atmosphere. Yet, this analysis may potentially
identify missing HAPs.
Recently, EPA's Office of Compliance Assurance (OECA) developed a database of
potential excess emitters by comparing ambient monitoring results, NATA 2005 results, and
emission inventory data from the 2008 NEI v2. If this approach is to be used, EPA would
propose to further synthesize the results by possibly targeting a subset of monitoring sites that
are more source-oriented, and/or examine HAPs that are non-mobile. To accomplish this,
geographic information system (GIS) technology would be used first to identify ambient
monitors in the OECA results that are within 1 to 2 miles of each point source. Next, EPA would
summarize the pollutants that are typically detected, including frequency of detection. For
facilities that are nearby, EPA would compare the monitored pollutant list to the pollutants
included in the emissions inventory, and note any differences.
The expected outcome from this may be a list of missing HAPs by targeted facility.
Additionally, an approach can be developed for application to the large OECA dataset. Expected
level of detail from this analysis would include: site-level information (e.g., facility name, EIS
ID, NEI ID, address, etc), SCCs, pollutants, and emissions. Due to time constraints, it was not
feasible to conduct this analysis.
5.2 Estimate Potentially Missing HAP-VOC and HAP-PM in 2008 NEI
For this analysis, EPA examined the 2011 Nonpoint Sources NEI vl augmentation
datasets (ftp://ftp.epa.aov/EmisInventorv/2011 nei/doc/ ). and compared those data to the
2008 Nonpoint Sources NEI Y2. The RTR program does not routinely evaluate nonpoint
sources at this time but this evaluation does provide useful insight on use of NEI data for
NATA. The 62 source category files prepared by EPA for the 2011 Nonpoint Sources NEI
40
-------
Evaluate 2008 NEI to Identify Areas of Improvement that Benefit Use in Residual Risk Assessments
vl, use consistent estimation methods, emission factors, and control information.
However, this type of pollutant consistency may not be the same for the final 2008
Nonpoint Sources NEI v2. For example, although a state/local/tribal (SLT) agency may
have submitted its own nonpoint sources inventory, it may have included only a limited
number of HAPs. Additionally, a SLT agency may have submitted YOC emissions, but
not corresponding HAP emissions. Thus, the ratios between the YOC emissions and the
YOC-HAP emissions are not necessarily consistent in the 2008 Nonpoint Sources NEI v2.
To examine potential missing HAPs in the 2008 Nonpoint Sources NEI v2, EPA
computed and applied the following ratios based on pollutant consistency between the:
• VOC HAP emissions and total YOC emissions;
• PMio HAPs and the total PMio emissions; and
• PM2, HAPs and the total PM2.5 emissions.
Table B-l presents a crosswalk of individual HAPs that are either VOCs or PM.
The pollutant codes listed in the crosswalk are valid codes in EPA's EIS master pollutant
dictionary, which can be downloaded as a MS Access database from EPA's CHIEF website
(http://www.epa.gov/ttn/chief/net/neip/appendix_6.mdb). It was decided for completeness
that both PMio-PRI and PM2.5-PRI be evaluated. EPA did the following to complete this
analysis:
1) Compiled all 2011 emission inventory files, i.e., the nonpoint augmentation data
sets, that have PM, VOC, and associated HAP data, into a 2011 master database;
2) From the 2011 master database, developed a list of targeted SCCs and related
emission factors for VOC and PM and associated VOC- and PM-HAPs;
3) From the 2011 Nonpoint Sources NEI vl master database, extracted the VOC, PM10-PRI,
and PM2.5-PRI emission factors; and the corresponding HAP emission factors associated with
VOC, PM10-PRI, and PM2 5-PRI;
4) Using the emission factors extracted from the 2011 Nonpoint Source NEI vl, calculated HAP
to VOC ratios; HAP to PMIO-PRI ratios; and HAP to PM2.5-PRI ratios -by county, SCC;
5) Matched the 2011 Nonpoint Sources NEI vl emission records by county, SCC,
and HAP to the 2008 Nonpoint Sources NEI v2 obtained from
http://www.epa.gov/ttn/chief/net/2008inventorv.html;
6) For the target SCCs above, extracted from the 2008 Nonpoint Sources NEI v2
the VOC, PM10-PRI, and PM2.5-PRI emissions; and the corresponding HAP
emissions associated with VOC, PM10-PRI, and PM2.5-PRI;
7) Applied the 2011 Nonpoint Sources NEI ratios, as follows:
a. 2011 HAP to VOC ratios to the 2008 VOC emissions
b. 2011 HAP to PMio ratios to the 2008 PM10-PRI emissions
c. 2011 HAP to PM2.5 ratios to the 2008 PM2.5-PRI emissions
8) Calculated the emissions difference between 2008 v2 HAP emissions and the
"ratio-applied" estimate for 2008 HAP emissions;
41
-------
Evaluate 2008 NEI to Identify Areas of Improvement that Benefit Use in Residual Risk Assessments
9) Calculated percent difference between the 2008 v2 HAP emissions and the
"ratio-applied" estimate for 2008 HAP emissions.
An example of the above approach is presented in Appendix B-2, with reference to
the MS Access database of analysis results. The MS Access database of results includes
the record matches between SCC and HAP, as well as records that are in 2008 NEI
inventory but not in the 2011 nonpoint NEI augmentation datasets, and records that are in
the 2011 nonpoint datasets that are not in the 2008 NEI. In total, over 750,000 county-
level, SCC, HAP records are directly matched records, i.e., found in both the 2011 dataset
and in the 2008 NEI. Of these, approximately 78% of the records have emission percent
differences of less than ±10%, while approximately 17% of the data records have
differences in emission amounts greater than 50%. Table 16 presents HAP nonpoint
emissions estimated by the ratio method for select SCCs that contribute to over 99% of the
missing emissions in the 2008 NEI, i.e., SCC /HAPs identified by the 2011 datasets that
are not present in the 2008 NEI. Table 16 shows that the 2008 NEI amount estimated as
potentially missing using the 2011 ratio method is quite small compared to the national
total amount for that HAP and select SCCs in the 2008 NEI. The potentially missing
amounts may also be viewed by county to understand degree of significance at the local
level. The database presents the results by state, county, and SCC. While Table 16
describes the amount of difference for select SCCs when target HAPs are missing in the
2008 NEI, county, county/SCC- level differences may also exist indicating values in the
2008 NEI much larger or much smaller than the emissions estimate derived from using the
2011 ratio method. Appendix B-2 which refers to the full database of results includes
summaries of such differences. The default nonpoint ratio may be different for a certain
state/county/SCC/HAP emissions value in 2008 for several reasons including because the
state-reported value may or may not relate to their YOC emissions or EPA completed the
HAP value and that HAP value was not related to the YOC emission factor value in the
2011 nonpoint augmentation dataset.
Table 16. Non-Point Stationary Sources - Top 5 HAP Emissions By 2011 Ratio Method
For Select SCCs Not Included in 2008 NEI
Emissions
%
Emissions in
Pollutant
Category Name
Estimate
Contribution
2008 NEI
EIS Sector(s)
via Ratio
Method
of Total
Missing
for Select
SCCs
(tpy)
Emissions
(tpy)
Solvent - Consumer &
Xylenes (Mixed
Isomers)
Commercial Solvent Use; Gas
Stations
58.70
56%
6,033.72
Solvent - Consumer &
Commercial Solvent Use; Gas
Toluene
Stations
30.79
29%
3,218.61
Solvent - Consumer &
Ethylbenzene
Commercial Solvent Use; Gas
11.07
11%
1,144.20
42
-------
Evaluate 2008 NEI to Identify Areas of Improvement that Benefit Use in Residual Risk Assessments
Table 16. Non-Point Stationary Sources - Top 5 HAP Emissions By 2011 Ratio Method
For Select SCCs Not Included in 2008 NEI
Pollutant
Category Name
EIS Sector(s)
Emissions
Estimate
via Ratio
Method
(tpy)
%
Contribution
of Total
Missing
Emissions
Emissions in
2008 NEI
for Select
SCCs
(tpy)
Stations
Methyl
Chloroform
Solvent -Industrial Surface
Coating & Solvent Use
2.51
2%
122.15
Methanol
Solvent -Industrial Surface
Coating & Solvent Use
0.98
1%
17.13
Remaining 21 HAPs
0.40
1%
651.80
Total
104.45
100%
11,187.61
5.3 Evaluate TRI Data Usage in RTR
For the 2008 NEI v2, limited TRI data were incorporated into the final inventory.
As a way to potentially add more data from TRI, EPA reviewed the RTR emissions
database to identify TRI data that were used for RTR rulemaking, and assessed this data
for potential usage for the 2011 NEI cycle. TRI data that were available for the 2008 NEI
v2, but were not used, is evaluated to determine if the data judged "sound" should be used
as a screening mechanism for the 2011 NEI cycle. That data judged sound which was
available in TRI and not used in the 2008 NEI will be identified as additional data to consider for
the 2011 NEI. For this analysis, EPA used the SMOKE-formatted version of the 2008 NEI,
which is publically available. SMOKE stands for Sparse-Matrix Operator Kernel Emissions,
which is a model used to convert emission inventories to the formats needed by air quality
modeling and which has its own input formats for the NEI.
EPA used the following steps to identify additional TRI data that could be
investigated for potential use in the 2011 NEI cycle:
1) From the RTR master database, pulled all facilities that had a MACT code
assignment.
2) Pulled all the emission inventory data for the facilities identified in Step 1.
3) Using the Data Source field, identified all TRI emission inventory records.
4) Mapped the TRI facilities to the 2008 NEI EIS ID using existing crosswalks
developed for the RTR inventory.
5) Compared the extracted TRI records for the RTR facilities with the 2008 NEI
v2 facility emissions available in the SMOKE file format. The comparison was
performed for the Facility and Pollutant Category Name, for both the RTR and
2008 NEI emissions. Use of the pollutant category name allows summary for
43
-------
Evaluate 2008 NEI to Identify Areas of Improvement that Benefit Use in Residual Risk Assessments
pollutant compounds groups where applicable, e.g., Chromium, Chromium VI,
and Chromium III are summarized as "Chromium Compounds."
6) Identified TRI emission records (by facility and pollutant category name) in the
RTR data set that were absent in the 2008 NEI v2. Prepared the
"Missing_RTR_TRI_IN_2008NEI" data table.
Some data considerations are presented below:
• The matching between TRI ID and EIS ID can be further examined. It is
possible that additional matches may be determined;
• It is possible that TRI data used for RTR source categories may not have reported
2008 emissions due to reporting threshold requirements;
• This analysis used the 2008 NEI v2 SMOKE emissions file dated June 2012. If
additional emissions data were entered into EIS after release date of the SMOKE file,
those data were not considered; and
• TRI dioxins were included in the TRI emissions, but not in the 2008 NEI v2 SMOKE
emissions file.
The results of this analysis are presented in Appendix C as referenced in a separate MS
Access database. The description of the data fields are included in a separate table of the
database and also listed here:
• REPORTING YEAR: year of TRI data used for the RTR source category
• NTI SITE ID: NEI Site identifier used to match facility in master RTR database
• TRIFID: Unique TRI facility identifier
• FACILITY NAME: TRI facility name
• FACILITY STREET: TRI facility street address
• FACILITY CITY: TRI facility city
• FACILITY STATE: TRI facility state postal code
• FACILITY ZIP CODE: TRI facility zip code
• PRIMARY SIC CODE: TRI facility Standard Industrial Classification (SIC)
• LATITUDE: TRI facility latitude (decimal degrees)
• LONGITUDE: TRI facility longitude (decimal degrees)
• CAS NUMBER: TRI chemical abstract services identifier
• CHEMICAL NAME: TRI pollutant name
• UNIT OF MEASURE: TRI unit of measure for the emissions
• ERP TYPE: TRI emission release point type (01 = fugitive; 02 = stack)
• EMISS_VALUE: TRI air emissions [convert to tons]
• Facilitv IDs 1-4: From FRS crosswalk, potential EIS IDs which the TRIFID can be
mapped to
• MACT _0101-1 to 1808-4: Flag to indicate the MACT Code in which the TRI data
was used for in the RTR master database
44
-------
Evaluate 2008 NEI to Identify Areas of Improvement that Benefit Use in Residual Risk Assessments
Table 17 presents the Top 10 emissions of HAPs that could have been potentially added
to the 2008 NEI, and which may warrant investigation during future NEI development cycles.
The Top 10 HAPs contribute 83% of the total TRI emissions not included in the 2008 NEI. The
resident county emissions in the 2008 NEI for the specific TRI facilities are also shown. For
chlorine, the negative difference between the TRI facility emissions and the total emissions for
the resident counties indicates more emissions in TRI for the facilities than are in the 2008 NEI.
The large positive differences in emissions indicate that the amount in TRI for the facilities are
small compared to the total sum in the counties where they reside. Comparisons to specific
county totals are available in the full database and may be investigated to understand significance
for individual counties.
Table 17. Top 10 TRI HAP Emissions Not Included in 2008 NEI Point Sources
Pollutant Category Name
TRI
Facility
Emissions
(tpy)
# of TRI
Facilities
2008 NEI County-
Level Emissions of
TRI Facilities (tpy)
Emissions
Difference
(County - TRI
facilities)
Chlorine
1,554.21
65
268.18
- 1,286
Hydrochloric Acid
1,514.59
70
12,807.62
11,293
Styrene
1,141.87
82
1,595.06
453
Xylenes (Mixed Isomers)
868.44
148
2,049.99
1,182
Methanol
823.33
160
4,473.67
3,651
Hexane
810.50
70
1,430.46
621
Toluene
332.17
104
1946.53
1,615
Methyl Isobutyl Ketone
268.15
74
360.63
93
Benzene
210.04
24
639.34
429
Ethylbenzene
209.61
108
390.09
180
Remaining 116 HAPs
1,702.89
1,696
4,408.88
2,706
Total
9,435.78
2,013
30,370.45
20,934
5.4 Missing HAP Results
The analyses conducted in Sections 5.1 through 5.3 highlight HAP emissions that are
potentially missing or under-reported from the 2008 NEI stationary sources inventories (point and
nonpoint sources). Specifically:
• Matrix of Expected HAPs: Appendix A-l presents a matrix of expected HAPs by SCC
which was compiled from multiple data references, and a matrix by NAICS code which
pertains solely to the TRI data reference. EPA also compiled a unique list of SCCs from
the matrix, and queried the 2008 NEI using those SCCs to identify missing HAPs, which
are presented in Appendix A-2. Table A-l presents a list of missing HAPs from the 2008
NEI to the Expected HAPs matrix by SCC. As noted in Section 5.0, the SCCs in the
matrix do not directly relate to the priority source categories presented in Table 11.
45
-------
Evaluate 2008 NEI to Identify Areas of Improvement that Benefit Use in Residual Risk Assessments
Finally, Appendix A-3 presents a comparison of expected HAPs by EIS Sector and SCC
to the actual number of HAPs that were reported to the 2008 v2 point sources NEI.
Table 18 summarizes these results. Of the 3,475 SCCs listed in this table, less than 9%
are reporting the same number of HAPs between expected and actual found in the 2008
NEI. Approximately 64% of the SCCs are showing under-reporting of HAPs (i.e., the
actual number of HAPs reported was less than the expected number of HAPs).
Table 18. Comparison of Expected HAPs to Actual HAPs in 2008
NEI
Expected HAPs...
# SCCs
% SCCs to Total
... equal Actual HAPs found in 2008
v2 point sources NEI
313
9%
... greater than Actual HAPs found in
2008 v2 point sources NEI
2,230
64%
... less than Actual HAP found in
2008 v2 point sources NEI
932
27%
Total
3,475
100%
• Missing HAPs via Ratios: For this analysis, an approach was developed to check for
pollutant consistency of nonpoint emissions for select SCCs between the 2008
Nonpoint NEI Y2 and the EPA's 2011 Nonpoint NEI Y1 augmentation data. The RTR
program does not routinely evaluate nonpoint sources but this evaluation does provide
useful insight on use of NEI data for NAT A. Table 16 shows that the 2008 NEI amount
estimated as potentially missing using the 2011 ratio method is quite small compared
to the national total amount for that HAP and select SCCs that are in the 2008 NEI.
The potentially missing amounts may also be viewed by county to understand degree
of significance at the local level. EPA also identified potentially under-reported or
over-reported HAP emissions for select SCCs in the 2008 Nonpoint NEI. In both cases,
the amount of potentially over or under-reported emissions could be significant,
especially in terms of NAT A risk modeling. Summaries of the Direct Matches and
Missing HAPs are also included in the Appendix B database as well as summaries for
the potentially under or over-reported HAPs in 2008 NEI (Tables B-2 and B-3).
• RTR TRI Evaluation: For RTR rulemaking, multiple sources of emissions information
are evaluated and compiled into developing an emissions modeling inventory. Some of
these data may include emissions data from TRI. For this analysis, EPA reviewed TRI
data that were used in RTR rulemaking and compiled a list of TRI facilities and their
reported HAPs. The TRI facilities were then mapped to EPA's EIS Facility ID. Once
matched, EPA compared the pollutant emissions from the TRI dataset to what was
reported in the 2008 Point Sources NEI V2. Summaries of these comparisons are in the
Appendix C database. In general, the amount in TRI for the facilities not included in the
2008 NEI are small compared to the total sum in the counties where they reside.
Comparisons to specific county totals are available in the full database and may be
46
-------
Evaluate 2008 NEI to Identify Areas of Improvement that Benefit Use in Residual Risk Assessments
investigated to understand significance for individual counties. For chlorine, the total
emissions for counties where the TRI facilities reside indicate more emissions in TRI for the
facilities than are in the 2008 NEI. These TRI facilities are being investigated for the 2011
NEI.
47
-------
Evaluate 2008 NEI to Identify Areas of Improvement that Benefit Use in Residual Risk Assessments
6.0 Recommended Improvement activities for future NEI development
USING THESE RESULTS
The Office of the Inspector General (OIG) recommended that EPA improve air
toxics emissions data needed to conduct residual risk assessments. Every development cycle
of the National Emissions Inventory (NEI) continues to improve the emissions information.
Many improvements for stationary sources that resulted in the 2008 NEI are expected to have
direct benefit to the RTR program. In an effort to improve the National Emissions Inventory
(NEI), EPA developed the Emissions Inventory System (EIS) to collect, compile, and store
emissions data. EIS contains numerous automated quality-assurance checks which can
provide greater confidence in the development of the NEI, and can also improve its use as a
data source for RTR rulemaking activities.
EPA evaluated the RTR program needs for emissions information and the most recent
inventory - the 2008 NEI, to gauge the degree of improvement in air toxics data needed to conduct
residual risk assessments. This evaluation indicates that the NEI is becoming increasingly more
effective in providing the emissions data needed for successful use by the RTR program. RTR
analysis to date has involved review and use of the 2005 NEI primarily, though the timing of some
more recent sector analysis allowed the use of early versions of the current 2008 NEI. The
development of the 2011 NEI is now underway. Many of the evaluation techniques used in this
project are operational as QA techniques for the 2011 NEI. For instance, TRI facility data is being
reviewed for more complete inclusion point source HAP data in the 2011 NEI. Also the EPA's use
of a consistent set of emission factor ratios for YOC and PM to HAP when adding nonpoint HAP
data where not reported by the SLT s, will improve the nonpoint data to be more consistent and
complete. Such improvements in completion of HAPs for the nonpoint sources will initially be
more beneficial to the NAT A 2011 as the RTR program continues to have predominant focus on
facility point source data.
As a result of this evaluation, the following recommended improvements are
presented for the current and/or future NEI development cycles:
• Continued coordination between EPA '.v RTR and NEI vrosrams will mutually
benefit data improvements and uses. As indicated in Section 4.0, Usefulness of
the 2008 NEI for Recent RTR Work - there are technical and procedural activities
that warrant continued coordination across these programs.
• Improve ease of access and use of the detailed NEI data needed for review by
the RTR prosram Section 4 of this report summarized information provided by
several RTR program staff regarding the use of the 2008 NEI. Given the complex
structure of the EIS, they found it difficult to use the 2008 NEI data. Difficulties
include: 1) too many data tables to link (i.e., a "flattened" data file containing all
the necessary elements for RTR emissions modeling); 2) inability to batch
download data (i.e., data can be searched one at a time through the EIS Gateway);
and 3) timeliness of the data (lag time from the end of a base year to release of
the data can be 3-4 year, which may be not be representative for rulemaking
48
-------
Evaluate 2008 NEI to Identify Areas of Improvement that Benefit Use in Residual Risk Assessments
activities. EPA has implemented access improvements since releasing earlier
versions of the 2008 NEI and should discuss those techniques with the RTR staff
regarding whether that has helped to alleviate their data access issues.
• Consolidating the RTR sector-specific emissions information in EIS can benefit
the NEI. In Section 3.3 of this report, EPA summarized RTR source categories
that were developed after the 2008 NEI cycle or are currently underway. As
discussed in Section 4.0, the RTR source sector inventories are generated from a
review of various data sources, including the facility specific information
requested and collected by EPA, i.e., the ICR data. The facility emissions and
physical configurations are verified by RTR program staff through the ICR
process and GIS technologies. This information can be a valuable resource in the
development of the NEI to alert the SLTs to review their facility configurations
and emissions. The EPA has initiated an effort to consolidate information
collected from industry by the RTR program into the EIS. That will facilitate
further consideration of that data for the NEI and specific SCCs.
• Future RTR source categories can use the 2011 NEI as a priority data tool.
Table 11 in Section 3.3 of this report, indicates a number of RTR source
categories scheduled to begin within the next 2 to 6 years. Therefore, a quality-
assured, complete 2011 NEI can be a valuable resource in developing and
compiling the baseline emissions inventory needed for the RTR program.
• The Priority Source Categories list (Table 11) can guide the development of
the 2011 NEI. In Section 3.3, EPA developed a list of priority source categories
based on RTR activity dates and cancer/noncancer toxicity weighting. This
approach further categorized source categories into "priority groups" which can
help EPA focus on more critical source categories if there are time and resource
constraints.
• Integrate matrix of expected HAPs into NEI OA activities and encourage
enhanced review by state/local/tribal (SIT) agencies to ensure complete
pollutant coverage. In Section 5.1, EPA reviewed numerous data sources (RTR,
WebFIRE, SPECIATE, etc.) to generate a master matrix of expected HAPs by
SCC. Approximately 64% of the SCCs are showing under-reporting of HAPs
(i.e., the actual number of HAPs reported was less than the expected number of
HAPs). Multiple years of TRI data were also reviewed to generate a similar
master matrix of expected HAPs by Facility ID and NAICS. This information can
be used by SLTs to ensure complete pollutant coverage prior to submission into
EIS and by EPA to facilitate completion of emissions data when not provided by
the SLTs.
49
-------
Evaluate 2008 NEI to Identify Areas of Improvement that Benefit Use in Residual Risk Assessments
• Investigate potentially missing HAP emissions for the 2011 NEI development
cycle. EPA computed YOC and PM to HAP emission factor ratios for many
nonpoint source categories that EPA will use for consistent data completion in the
2011 NEI when HAP data is not reported by S/L/Ts. Section 4.0 describes the
result of applying those ratios to the 2008 NEI data to estimate potentially
missing HAP emissions in the 2008 NEI. EPA can use those results as a QA
mechanism to engage the SLT agencies during the development of the 2011 NEI
to review for potentially missing or under-reported HAPs. Such improvements in
completion of HAPs for the nonpoint sources will initially be more beneficial to
the NATA 2011 as the RTR program continues to have predominant focus on
facility point source data. The 2008 NEI amount estimated as potentially
missing using the 2011 ratio method is quite small compared to the national
total amount for that HAP and select SCCs that are in the 2008 NEI. The
potentially missing amounts may also be viewed by county to understand degree
of significance at the local level. EPA also identified potentially under-reported
or over-reported HAP emissions for select SCCs in the 2008 Nonpoint NEI. In
both cases, the amount of potentially over or under-reported emissions could be
significant, especially in terms of NATA risk modeling.
• Review and consider additional TRI data for the 2011 NEI. As the 2011 point
sources inventory is developed in EIS, EPA may want to consider incorporating
TRI data used for recent RTR rulemaking activities, yet did not get integrated
into the 2008 point sources inventory. As presented in Section 5.3, the missing
emissions may be significant for specific facilities and can be investigated for
inclusion in the 2011 NEI.
• Continue examination of the EIS Facility Table for multiple and active facility
IDs and work with the SIT agencies to merge as appropriate and to discourage
multiple active IDs for one facility. In preparing the missing TRI data, it was
noted that for select facilities, the TRI facility identifier (TRIFID) could be
mapped to multiple EIS Facility IDs, based on crosswalks available on
Envirofacts. For example, a Monsanto plant in Iowa has a TRIFID
52761MNSNTWIGGI. According to Envirofacts, this TRI ID is mapped to EIS
Facility IDs 12807711, 12807611, 12807511, and 8121511. Based on additional
investigation, the State of Iowa requested that the original EIS Facility ID
(8121511)be made inactive and three EIS Facility IDs be generated to coincide
with the three Title V permits currently at that facility. The current NEI data
development process does include a seek and merge process for these situations
and involves the SLTs in problem identification and resolution.
50
-------
Evaluate 2008 NEI to Identify Areas of Improvement that Benefit Use in Residual Risk Assessments
7.0 References
U.S. EPA, 2012a. RTR Updated Schedule and Historical Dates. Email sent from Ms. Anne
Pope, U.S. EPA to Mr. Regi Oommen, Eastern Research Group, Inc. October 23,
2012.
U.S. EPA, 2012b. Development of RTR Modeling Files. Presentation made by Ms. Anne Pope to
EPA's SPPD. April 17, 2012.
U.S. EPA, 2007. 1990-2002 NEI HAP Trends: Success of CAA Air Toxic Programs in
Reducing HAP Emissions and Risk. Proceedings at the 16th Annual International
Emissions Inventory Conference. Raleigh, NC. May 2007. Internet address:
http ://www. epa. gov/ttn/ chief/ conference/ei 16/session6/a. pope, pdf
51
-------
Evaluate 2008 NEI to Identify Areas of Improvement that Benefit Use in Residual Risk Assessments
Appendix A - Expected HAPs Matrix (found in the "EXPECTED HAPs"
Microsoft Access database)
-------
Evaluate 2008 NEI to Identify Areas of Improvement that Benefit Use in Residual Risk Assessments
Appendix A-l. Expected HAPs Matrix (found in the "EXPECTEDHAPs by SCC" and
"EXPECTED HAPs BY NAICS" DATA TABLES IN THE MICROSOFT ACCESS DATABASE)
-------
Evaluate 2008 NEI to Identify Areas of Improvement that Benefit Use in Residual Risk Assessments
Appendix A-2. Summary of SCC and HAP Counts BY EIS Sector
EIS Sector
Number of SCCs
Number of Expected
HAPs
Agriculture - Livestock Waste
1
4
Bulk Gasoline Terminals
349
75
Dust - Construction Dust
5
13
Fuel Comb - Comm/Institutional - Biomass
31
60
Fuel Comb - Comm/Institutional - Coal
101
58
Fuel Comb - Comm/Institutional - Natural Gas
35
63
Fuel Comb - Comm/Institutional - Oil
52
62
Fuel Comb - Comm/Institutional - Other
73
75
Fuel Comb - Electric Generation - Biomass
21
54
Fuel Comb - Electric Generation - Coal
107
91
Fuel Comb - Electric Generation - Natural Gas
24
71
Fuel Comb - Electric Generation - Oil
52
67
Fuel Comb - Electric Generation - Other
65
77
Fuel Comb - Industrial Boilers, ICEs - Biomass
52
86
Fuel Comb - Industrial Boilers, ICEs - Coal
118
82
Fuel Comb - Industrial Boilers, ICEs - Natural Gas
52
109
Fuel Comb - Industrial Boilers, ICEs - Oil
111
78
Fuel Comb - Industrial Boilers, ICEs - Other
156
102
Fuel Comb - Residential - Natural Gas
1
5
Fuel Comb - Residential - Oil
3
15
Fuel Comb - Residential - Other
3
38
Gas Stations
51
25
Industrial Processes - Cement Manuf
4
11
Industrial Processes - Chemical Manuf
1198
145
Industrial Processes - Ferrous Metals
75
15
Industrial Processes - Mining
3
6
Industrial Processes - NEC
738
123
Industrial Processes - Non-ferrous Metals
167
43
Industrial Processes - Oil & Gas Production
245
64
Industrial Processes - Petroleum Refineries
215
87
Industrial Processes - Pulp & Paper
193
89
Industrial Processes - Storage and Transfer
1052
134
Miscellaneous Non-Industrial NEC
3
34
Mobile - Non-Road Equipment - Diesel
3
15
Mobile - Non-Road Equipment - Gasoline
1
16
Mobile - Non-Road Equipment - Other
9
19
Solvent - Degreasing
46
88
Solvent - Dry Cleaning
22
7
Solvent - Graphic Arts
125
70
Solvent - Industrial Surface Coating & Solvent Use
300
136
Waste Disposal
372
130
A-l
-------
Evaluate 2008 NEI to Identify Areas of Improvement that Benefit Use in Residual Risk Assessments
Appendix A-3. Comparison of Expected HAPs to HAPs Reported to the 2008 v2 Point
Sources NEI (found in the "Appendix A-
3 COMPARISON_OF EXPECTED HAPS_2008NEI HAPs" DATA TABLE IN THE MICROSOFT
Access database)
-------
Evaluate 2008 NEI to Identify Areas of Improvement that Benefit Use in Residual Risk Assessments
Appendix B.l - Crosswalk of VOC HAPs and PM HAPs Based on EPA's EIS (Emissions
Inventory System)
-------
Evaluate 2008 NEI to Identify Areas of Improvement that Benefit Use in Residual Risk Assessments
Table B-l. Crosswalk of VOC and PM HAPs to EPA EIS Pollutant Codes
EPA's EIS Pollutant Category
Name
EPA's EIS Pollutant Name
EPA's
EIS
Pollutant
Code
PM
VOC
1,1,2,2-Tetrachloroethane
1,1,2,2-Tetrachloroethane
79345
~
1,1,2-Trichloroethane
1,1,2-Trichloroethane
79005
~
1,2,4-Trichlorobenzene
1,2,4-Trichlorobenzene
120821
~
1,2-Epoxybutane
1,2-Epoxybutane
106887
~
1,3-Butadiene
1,3-Butadiene
106990
~
1,3 -Dichloropropene
1,3 -Dichloropropene
542756
~
1,4-Dichlorobenzene
1,4-Dichlorobenzene
106467
~
2,2,4-Trimethylpentane
2,2,4-Trimethylpentane
540841
~
2,4,6-T richlorophenol
2,4,6-Trichlorophenol
88062
~
2,4-Dichlorophenoxy Acetic Acid
2,4-Dichlorophenoxy Acetic Acid
94757
~
2,4-Dinitrophenol
2,4-Dinitrophenol
51285
y
2,4-Dinitrotoluene
2,4-Dinitrotoluene
121142
y
2,4-Toluene Diisocyanate
2,4-Toluene Diisocyanate
584849
y
2-Chloroacetophenone
2-Chloroacetophenone
532274
y
2-Nitropropane
2-Nitropropane
79469
y
4,4'-Methylenediphenyl
Diisocyanate
4,4'-Methylenediphenyl
Diisocyanate
101688
y
4-Nitrophenol
4-Nitrophenol
100027
y
Acetaldehyde
Acetaldehyde
75070
y
Acetamide
Acetamide
60355
y
Acetonitrile
Acetonitrile
75058
y
Acetophenone
Acetophenone
98862
y
Acrolein
Acrolein
107028
y
Acrylic Acid
Acrylic Acid
79107
y
Acrylonitrile
Acrylonitrile
107131
y
Allyl Chloride
Allyl Chloride
107051
y
Antimony Compounds
Antimony
7440360
Arsenic Compounds
Arsenic
7440382
Benzene
Benzene
71432
y
Benzyl Chloride
Benzyl Chloride
100447
y
Beryllium Compounds
Beryllium
7440417
Biphenyl
Biphenyl
92524
y
Bis(2-Ethylhexyl)Phthalate
Bis(2-Ethylhexyl)Phthalate
117817
y
Bromoform
Bromoform
75252
y
B-l
-------
Evaluate 2008 NEI to Identify Areas of Improvement that Benefit Use in Residual Risk Assessments
Table B-l. Crosswalk of VOC and PM HAPs to EPA EIS Pollutant Codes
EPA's EIS Pollutant Category
Name
EPA's EIS Pollutant Name
EPA's
EIS
Pollutant
Code
PM
VOC
Cadmium Compounds
Cadmium
7440439
Calcium Cyanamide
Calcium Cyanamide
156627
~
Captan
Captan
133062
~
Carbaryl
Carbaryl
63252
~
Carbon Tetrachloride
Carbon Tetrachloride
56235
~
Chlorobenzene
Chlorobenzene
108907
~
Chloroform
Chloroform
67663
~
Chloroprene
Chloroprene
126998
~
Chromium Compounds
Chromium
7440473
~
Chromium Compounds
Chromium (VI)
18540299
~
Chromium Compounds
Chromium III
16065831
~
Cobalt Compounds
Cobalt
7440484
~
Cresol/Cresylic Acid (Mixed
Isomers)
Cresol/Cresylic Acid (Mixed
Isomers)
1319773
y
Cresol/Cresylic Acid (Mixed
Isomers)
w-Cresol
108394
y
Cresol/Cresylic Acid (Mixed
Isomers)
o-Cresol
95487
y
Cresol/Cresylic Acid (Mixed
Isomers)
p-Cresol
106445
y
Cumene
Cumene
98828
y
Cyanide Compounds
Cyanide
57125
Cyanide Compounds
Hydrogen Cyanide
74908
Dibenzofuran
Dibenzofuran
132649
y
Dibutyl Phthalate
Dibutyl Phthalate
84742
y
Diethanolamine
Diethanolamine
111422
y
Dimethyl Phthalate
Dimethyl Phthalate
131113
y
Dimethyl Sulfate
Dimethyl Sulfate
77781
y
Epichlorohydrin
Epichlorohydrin
106898
y
Ethyl Acrylate
Ethyl Acrylate
140885
y
Ethyl Chloride
Ethyl Chloride
75003
y
Ethylbenzene
Ethyl Benzene
100414
y
Ethylene Dibromide
Ethylene Dibromide
106934
y
Ethylene Dichloride
Ethylene Dichloride
107062
y
Ethylene Glycol
Ethylene Glycol
107211
y
B-2
-------
Evaluate 2008 NEI to Identify Areas of Improvement that Benefit Use in Residual Risk Assessments
Table B-l. Crosswalk of VOC and PM HAPs to EPA EIS Pollutant Codes
EPA's EIS Pollutant Category
Name
EPA's EIS Pollutant Name
EPA's
EIS
Pollutant
Code
PM
VOC
Ethylene Oxide
Ethylene Oxide
75218
~
Ethylidene Dichloride
Ethylidene Dichloride
75343
~
Formaldehyde
Formaldehyde
50000
~
Glycol Ethers
2-Butoxyethyl Acetate
112072
~
Glycol Ethers
Butyl Carbitol Acetate
124174
~
Glycol Ethers
Carbitol Acetate
112152
~
Glycol Ethers
Cellosolve Acetate
111159
~
Glycol Ethers
Cellosolve Solvent
110805
~
Glycol Ethers
Diethylene Glycol Ethyl Methyl
Ether
1002671
~
Glycol Ethers
Diethylene Glycol Monobutyl
Ether
112345
y
Glycol Ethers
Diethylene Glycol Monoethyl
Ether
111900
y
Glycol Ethers
Diethylene Glycol Monomethyl
Ether
111773
y
Glycol Ethers
Glycol Ethers
171
y
Glycol Ethers
N-Hexyl Carbitol
112594
y
Glycol Ethers
Propyl Cellosolve
2807309
y
Hexachlorobenzene
Hexachlorobenzene
118741
y
Hexachlorobutadiene
Hexachlorobutadiene
87683
y
Hexachlorocyclopentadiene
Hexachlorocyclopentadiene
77474
y
Hexane
Hexane
110543
y
Hydrazine
Hydrazine
302012
y
Hydroquinone
Hydroquinone
123319
y
Isophorone
Isophorone
78591
y
Lead Compounds
Lead
7439921
Manganese Compounds
Manganese
7439965
Mercury Compounds
Mercury
7439976
Methanol
Methanol
67561
y
Methyl Bromide
Methyl Bromide
74839
y
Methyl Chloride
Methyl Chloride
74873
y
Methyl Chloroform
Methyl Chloroform
71556
y
Methyl Isobutyl Ketone
Methyl Isobutyl Ketone
108101
y
Methyl Methacrylate
Methyl Methacrylate
80626
y
B-3
-------
Evaluate 2008 NEI to Identify Areas of Improvement that Benefit Use in Residual Risk Assessments
Table B-l. Crosswalk of VOC and PM HAPs to EPA EIS Pollutant Codes
EPA's EIS Pollutant Category
Name
EPA's EIS Pollutant Name
EPA's
EIS
Pollutant
Code
PM
VOC
Methyl Tert-Butyl Ether
Methyl tert-Butyl Ether
1634044
~
Methylene Chloride
Methylene Chloride
75092
~
Methylhydrazine
Methylhydrazine
60344
~
N,N-Dimethylaniline
N,N-Dimethylaniline
121697
~
N,N-Dimethylformamide
N,N-Dimethylformamide
68122
~
Naphthalene
Naphthalene
91203
~
Nickel Compounds
Nickel
7440020
~
Nitrobenzene
Nitrobenzene
98953
~
o-Toluidine
o-Toluidine
95534
~
PAH/POM - Unspecified
PAH/POM - Unspecified
250
~
p-Dioxane
/;-Dioxane
123911
y
Pentachlorophenol
Pentachlorophenol
87865
y
Phenol
Phenol
108952
y
Phosphorus
Phosphorus
7723140
~
Phthalic Anhydride
Phthalic Anhydride
85449
y
Polychlorinated Biphenyls
2,4,4'-Trichlorobiphenyl (PCB-28)
7012375
y
Polychlorinated Biphenyls
2-Chlorobiphenyl (PCB-1)
2051607
y
Polychlorinated Biphenyls
4,4'-Dichlorobiphenyl (PCB-15)
2050682
y
Polychlorinated Biphenyls
Decachlorobiphenyl (PCB-209)
2051243
y
Polychlorinated Biphenyls
Heptachlorobiphenyl
28655712
y
Polychlorinated Biphenyls
Hexachlorobiphenyl
26601649
y
Polychlorinated Biphenyls
Pentachlorobiphenyl
25429292
y
Polychlorinated Biphenyls
Polychlorinated Biphenyls
1336363
y
Polychlorinated Biphenyls
T etrachlorobiphenyl
26914330
y
Polycyclic Organic Matter
12-Methylbenz(a) Anthracene
2422799
y
Polycyclic Organic Matter
1 -Methylphenanthrene
832699
y
Polycyclic Organic Matter
2-Chloronaphthalene
91587
y
Polycyclic Organic Matter
2-Methylnaphthalene
91576
y
Polycyclic Organic Matter
3 -Methylcholanthrene
56495
y
Polycyclic Organic Matter
5-Methylchrysene
3697243
y
Polycyclic Organic Matter
7,12-Dimethylbenz[a] Anthracene
57976
y
Polycyclic Organic Matter
Acenaphthene
83329
y
Polycyclic Organic Matter
Acenaphthylene
208968
y
Polycyclic Organic Matter
Anthracene
120127
y
B-4
-------
Evaluate 2008 NEI to Identify Areas of Improvement that Benefit Use in Residual Risk Assessments
Table B-l. Crosswalk of VOC and PM HAPs to EPA EIS Pollutant Codes
EPA's EIS Pollutant Category
Name
EPA's EIS Pollutant Name
EPA's
EIS
Pollutant
Code
PM
VOC
Polycyclic Organic Matter
B enz [a] Anthracene
56553
Polycyclic Organic Matter
Benzo(g,h,i)Fluoranthene
203123
~
Polycyclic Organic Matter
Benzo[a]Pyrene
50328
~
Polycyclic Organic Matter
Benzo[b]Fluoranthene
205992
~
Polycyclic Organic Matter
Benzo[e]Pyrene
192972
~
Polycyclic Organic Matter
Benzo[g,h,i,]Perylene
191242
~
Polycyclic Organic Matter
Benzo[k]Fluoranthene
207089
~
Polycyclic Organic Matter
B enzofluoranthene s
56832736
~
Polycyclic Organic Matter
Carbazole
86748
~
Polycyclic Organic Matter
Chrysene
218019
~
Polycyclic Organic Matter
Dibenzo[a,h] Anthracene
53703
~
Polycyclic Organic Matter
Fluoranthene
206440
~
Polycyclic Organic Matter
Fluorene
86737
~
Polycyclic Organic Matter
Indeno[ 1,2,3 -c,d]Pyrene
193395
~
Polycyclic Organic Matter
Methylanthracene
26914181
~
Polycyclic Organic Matter
PAH, total
130498292
y
Polycyclic Organic Matter
Perylene
198550
y
Polycyclic Organic Matter
Phenanthrene
85018
y
Polycyclic Organic Matter
Pyrene
129000
y
Propionaldehyde
Propionaldehyde
123386
Propylene Dichloride
Propylene Dichloride
78875
Propylene Oxide
Propylene Oxide
75569
Quinoline
Quinoline
91225
Selenium Compounds
Selenium
7782492
Styrene
Styrene
100425
T etrachloroethylene
T etrachloroethylene
127184
Titanium Tetrachloride
Titanium Tetrachloride
7550450
Toluene
Toluene
108883
T richloroethylene
T richloroethylene
79016
Triethylamine
Triethylamine
121448
Trifluralin
Trifluralin
1582098
Vinyl Acetate
Vinyl Acetate
108054
Vinyl Chloride
Vinyl Chloride
75014
Vinylidene Chloride
Vinylidene Chloride
75354
B-5
-------
Evaluate 2008 NEI to Identify Areas of Improvement that Benefit Use in Residual Risk Assessments
Table B-l. Crosswalk of VOC and PM HAPs to EPA EIS Pollutant Codes
EPA's EIS Pollutant Category
Name
EPA's EIS Pollutant Name
EPA's
EIS
Pollutant
Code
PM
VOC
Xylenes (Mixed Isomers)
/??-Xylene
108383
~
Xylenes (Mixed Isomers)
o-Xylene
95476
~
Xylenes (Mixed Isomers)
/^-Xylene
106423
~
Xylenes (Mixed Isomers)
Xylenes (Mixed Isomers)
1330207
~
B-6
-------
Evaluate 2008 NEI to Identify Areas of Improvement that Benefit Use in Residual Risk Assessments
Appendix B.l - 2008 to 2011 HAP-VOC and HAP-PM Ratios (found in the
"2008_2011_Ratios" Microsoft Access database)
B.l-1
-------
Evaluate 2008 NEI to Identify Areas of Improvement that Benefit Use in Residual Risk Assessments
Appendix B-2. Example Ratio Calculations
Example calculations are provided below as follows:
• Computation of the emission ratios based on the 2011 NEI vl Nonpoint Sources augmentation
datasets
• Application of the computed 2011 ratios to the 2008 NEI v2 Nonpoint Sources data to estimate
the amount of potentially missing HAPs.
The example is for specific HAP(s), SCC, and county - for Chromium (VI) and Formaldehyde
emissions, for SCC 2102011000 (Stationary Fuel Comb /Industrial /Kerosene /Total: All Boiler Types)
in Greenlee County, AZ (FIPS = 04011). The pertinent data fields are indicated from the MS Access
database of results.
To complete this analysis, EPA performed the following computations and comparisons:
1) Compiled all 2011 emission inventory files, i.e., the nonpoint augmentation datasets that have
PM, VOC, and associated HAP data, into a 2011 master database;
2) From the 2011 master database, developed a list of target SCCs and related emission factors for
VOC and PM and associated VOC- and PM-HAPs;
3) From the 2011 Nonpoint Sources NEI vl, extracted the 2011 VOC, PMi0-PRI, and PM2.5-PRI
emission factors; and the corresponding HAP emission factors associated with VOC, PM10-PRI,
and PM2.5-PRI:
4) Using the emission factors extracted from the 2011 Nonpoint Source NEI vl, calculated HAP to
VOC ratios; HAP to PM10-PRI ratios; and HAP to PM2.5-PRI ratios -by county, SCC:
2011
2011
FIPS
Ratio
Ratio
State
2011
HAP-
2011
Target
Target
and
2011
2011
Ref
2011
2011
VOC
2011
Target
HAP-
HAP-
County
Target
Ref
Poll
Ref Poll
Target HAP
EIS Poll
or
Target
HAP EF
VOC/ref
PM/ref
Code
SCC
Poll
EF
EF UOM
Name
Code
HAP-PM
HAP EF
UOM
VOC
PM
Chromium
04011
2102011000
PM10
2.21
LB/E3GAL
(VI)
18540299
HAP-PM
0.0000729
LB/E3GAL
3.299E-05
Chromium
04011
2102011000
PM2.5
1.49
LB/E3GAL
(VI)
18540299
HAP-PM
0.0000729
LB/E3GAL
4.893E-05
04011
2102011000
VOC
0.19
LB/E3GAL
Formaldehyde
50000
HAP-VOC
0.0324
LB/E3GAL
0.1705
Calculation of ratios, examples:
2011 HAP to VOC Ratio = 2011 HAP Emission Factor Numerator 2011 VOC Emission
Factor Numerator
= 0.0324 (LB E3GAL) 0.19 (LB/E3GAL)
= 0.1705
B.l-1
-------
Evaluate 2008 NEI to Identify Areas of Improvement that Benefit Use in Residual Risk Assessments
2011 HAP to PMio-PRIRatio = 2011 HAP Emission Factor Numerator 2011 PMio-PRI
Emission Factor Numerator
= 0.0000729 (LB E3GAL)/2.21 (LB/E3GAL)
= 3.299E-05
2011 HAP to PM2.5-PRIRatio = 2011 HAP Emission Factor Numerator72011 PM2.5-PRI
Emission Factor Numerator
= 0.0000729 (LB E3GAL) 1.49 (LB E3GAL)
= 4.893E-05
5) For the target SCCs, extracted from the 2008 Nonpoint Sources NEI v2, the VOC, PMIO-PRI,
PM2.5-PRI emissions, and the available target HAP emissions;
FIPS
State
and
County
Code
2011
Target SCC
2011
Ref
Poll
2011
Target HAP
Name
HAP-VOC
or
HAP-PM
2008 v2
VOC
Emissions
Tons
2008 v2
PMIO-
PRI
Emissions
Tons
2008 v2
PM25-
PRI
Emissions
Tons
2008 v2
Target
HAP
Emissions
Tons
2008
estimate
Target
HAP
emiss
per 2011
ratio calc
Tons
(2008v2
emiss) -
(2008 est
by 2011
ratio)
% Diff
wrt
2008v2
NEI
04011
2102011000
PM10
Chromium
(VI)
HAP-PM
3.127E-06
1.028E-10
1.032E-10
-3.662E-13
-0.35
04011
2102011000
PM2.5
Chromium
(VI)
HAP-PM
2.108E-06
1.028E-10
1.031E-10
-3.229E-13
-0.31
04011
2102011000
VOC
Formaldehyde
HAP-VOC
2.719E-07
4.568E-08
4.637E-08
-6.867E-10
-1.48
6) Applied the 2011 Nonpoint Sources NEI ratios to 2008 v2 emissions as follows:
a. 2011 HAP to VOC ratios to the 2008 VOC emissions:
0.171 * 2.719E-07 = 4.637E-08 TPY estimated Formaldehyde emissions
b. 2011 HAP to PM10 ratios to the 2008 PMio-PRI emissions:
3.299E-05 * 3.127E-06 = 1.032E-10 TPY estimated Chromium (VI) emissions
c. 2011 HAP to PM25 ratios to the 2008 PM25-PRI emissions:
4.893E-05 * 2.108E-06 = 1.03IE-10 TPY estimated Chromium (VI) emissions
B.l-2
-------
Evaluate 2008 NEI to Identify Areas of Improvement that Benefit Use in Residual Risk Assessments
7) Calculated the emissions difference between 2008 v2 HAP emissions and the "ratio-applied"
estimate of 2008 HAP emissions:
Formaldehyde Emissions Difference = 2008 v2 Emissions - Ratio-Applied 2008 Emissions
= 4.568E-08 TPY- 4.637E-08 TPY
= -6.867E-10 TPY
Chromium (VI) Emissions Difference = 2008 v2 Emissions - PMw Ratio-Applied 2008
Emissions
= 1.028E-10 TPY- 1.032E-10 TPY
=-3.662E-13
Chromium (VI) Emissions Difference = 2008 v2 Emissions - PM2.5 Ratio-Applied 2008
Emissions
= 1.028E-10 TPY- 1.03IE-10 TPY
=-3.229E-13
8) Calculated percent difference between the 2008 v2 HAP emissions and the "ratio-applied" 2008
HAP emissions:
Formaldehyde Emissions Percent Difference = ((2008 v2 Emissions TPY - Ratio-Applied
2008 Emissions TPY)/2008 v2 Emissions TPY) *100
= 4.568E-08 TPY- 4.637E-08 TPY * 100
4.568E-08 TPY
= -1.48%
Chromium (VI) Emissions Percent Difference = ((2008 v2 Emissions TPY - PM)o Ratio-
Applied 2008 Emissions TPY)/2008 v2 Emissions TPY) *100
= 1.028E-10 TPY- 1.032E-10 TPY * 100
1.028E-10 TPY
= -0.35%
Chromium (VI) Emissions Percent Difference = ((2008 v2 Emissions TPY - PM2.5 Ratio-
Applied 2008 Emissions TPY)/2008 v2 Emissions TPY) *100
= 1.028E-10 TPY- 1.03IE-10 TPY * 100
1.028E-10 TPY
= -0.31%
B.l-3
-------
Evaluate 2008 NEI to Identify Areas of Improvement that Benefit Use in Residual Risk Assessments
Table B-2. Summary of Emissions By EIS Sectors and Pollutant Category
Name for Negative Ratios (potential over-estimate in 2008 NEI)
EIS Sector
Pollutant Category Name
SumOf(2008v2 emiss) -
(2008 est by 2011 ratio)
Hexane
-76,617.07
Toluene
-62,247.79
Benzene
-43,096.87
2,2,4-Trimethylpentane
-38,308.53
Gas Stations
Xylenes (Mixed Isomers)
-23,942.83
Ethylbenzene
-4,787.41
Naphthalene
-2,394.21
Cumene
-478.84
Ethylene Dichloride
-0.55
Total
-251,874.11
Toluene
-1,233.38
Xylenes (Mixed Isomers)
-1,122.50
Methyl Isobutyl Ketone
-873.58
Methyl Chloroform
-579.63
Solvent - Industrial Surface
Coating & Solvent Use
Glycol Ethers
-125.10
Ethylene Glycol
-41.33
Ethylbenzene
-36.76
Methanol
-9.08
Cumene
-0.31
Naphthalene
-0.01
Total
-4,021.68
Xylenes (Mixed Isomers)
-1,150.42
Solvent - Consumer &
Toluene
-219.72
Commercial Solvent Use
Ethylbenzene
-79.07
Total
-1,449.20
Benzene
-38.23
Formaldehyde
-30.33
Acetaldehyde
-21.74
Styrene
-14.02
Commercial Cooking
Polycyclic Organic Matter
-11.56
Toluene
-10.63
Propionaldehyde
-5.52
Ethylbenzene
-2.62
Xylenes (Mixed Isomers)
-2.12
Naphthalene
-1.71
B.l-4
-------
Evaluate 2008 NEI to Identify Areas of Improvement that Benefit Use in Residual Risk Assessments
Table B-2. Summary of Emissions By EIS Sectors and Pollutant Category
Name for Negative Ratios (potential over-estimate in 2008 NEI)
EIS Sector
Pollutant Category Name
SumOf(2008v2 emiss) -
(2008 est by 2011 ratio)
Phenol
-1.49
Ethylene Dichloride
-1.02
4-Nitrophenol
-0.32
Cresol/Cresylic Acid (Mixed
Isomers)
-0.31
Biphenyl
-0.17
Acetophenone
-0.16
Dibutyl Phthalate
-0.12
Total
-142.08
Fuel Comb -
Comm/Institutional - Natural
Gas
Formaldehyde
-15.09
Lead Compounds
-0.94
Polycyclic Organic Matter
-0.06
Naphthalene
-1.12E-03
Benzene
-4.33E-05
Acetaldehyde
-1.27E-06
Total
-16.10
Fuel Comb - Industrial
Boilers, ICEs - Natural Gas
Formaldehyde
-12.58
Lead Compounds
-0.56
Polycyclic Organic Matter
-0.03
Benzene
-6.77E-03
Naphthalene
-4.57E-04
Acetaldehyde
-7.62E-07
Total
-13.18
Fuel Comb - Residential -
Natural Gas
Formaldehyde
-7.52
Polycyclic Organic Matter
-0.23
Benzene
-0.19
Naphthalene
-0.06
Acetaldehyde
-1.37E-03
Total
-8.00
Fuel Comb - Industrial
Boilers, ICEs - Oil
Formaldehyde
-2.89
Acetaldehyde
-0.22
Nickel Compounds
-0.15
Naphthalene
-0.09
Benzene
-7.84E-03
Selenium Compounds
-7.34E-03
B.l-5
-------
Evaluate 2008 NEI to Identify Areas of Improvement that Benefit Use in Residual Risk Assessments
Table B-2. Summary of Emissions By EIS Sectors and Pollutant Category
Name for Negative Ratios (potential over-estimate in 2008 NEI)
EIS Sector
Pollutant Category Name
SumOf(2008v2 emiss) -
(2008 est by 2011 ratio)
Manganese Compounds
-4.95E-03
Lead Compounds
-3.83E-03
Chromium Compounds
-2.98E-03
Arsenic Compounds
-2.62E-03
Mercury Compounds
-1.33E-03
Cadmium Compounds
-7.59E-04
Beryllium Compounds
-3.45E-04
Polycyclic Organic Matter
-1.76E-04
Total
-3.39
Lead Compounds
-0.69
Nickel Compounds
-0.37
Formaldehyde
-0.08
Acetaldehyde
-0.07
Selenium Compounds
-0.03
Manganese Compounds
-2.03E-03
Fuel Comb -
Comm/Institutional - Oil
Chromium Compounds
-1.76E-03
Beryllium Compounds
-1.68E-03
Arsenic Compounds
-1.52E-03
Cadmium Compounds
-9.39E-04
Mercury Compounds
-9.18E-04
Naphthalene
-8.12E-04
Benzene
-6.17E-05
Polycyclic Organic Matter
-4.50E-05
Total
-1.25
Formaldehyde
-0.74
Naphthalene
-0.18
Arsenic Compounds
-0.06
Biphenyl
-0.02
Fuel Comb -
Comm/Institutional - Coal
Polycyclic Organic Matter
-0.02
Cyanide Compounds
-0.01
Chromium Compounds
-1.73E-03
Benzyl Chloride
-1.11E-03
Cadmium Compounds
-2.83E-04
Methylhydrazine
-2.69E-04
Chloroform
-9.34E-05
B.l-6
-------
Evaluate 2008 NEI to Identify Areas of Improvement that Benefit Use in Residual Risk Assessments
Table B-2. Summary of Emissions By EIS Sectors and Pollutant Category
Name for Negative Ratios (potential over-estimate in 2008 NEI)
EIS Sector
Pollutant Category Name
SumOf(2008v2 emiss) -
(2008 est by 2011 ratio)
Dimethyl Sulfate
-7.60E-05
Ethyl Chloride
-6.65E-05
Bromoform
-6.17E-05
Xylenes (Mixed Isomers)
-5.86E-05
Methyl Tert-Butyl Ether
-5.54E-05
Methyl Chloroform
-3.17E-05
Methyl Methacrylate
-3.17E-05
Vinyl Acetate
-1.21E-05
2-Chloroacetophenone
-1.11E-05
Cumene
-8.46E-06
Ethylene Dibromide
-1.93E-06
2,4-Dinitrotoluene
-1.58E-09
Lead Compounds
-4.97E-10
Total
-1.04
Formaldehyde
-0.81
Benzene
-0.04
Fuel Comb - Residential -
Polycyclic Organic Matter
-0.03
Other
Naphthalene
-6.68E-03
Acetaldehyde
-1.40E-04
Total
-0.88
Selenium Compounds
-0.19
Lead Compounds
-0.13
Manganese Compounds
-0.08
Formaldehyde
-0.07
Arsenic Compounds
-0.05
Chromium Compounds
-0.04
Fuel Comb - Residential -
Oil
Cadmium Compounds
-0.04
Mercury Compounds
-0.04
Nickel Compounds
-0.04
Beryllium Compounds
-0.04
Acetaldehyde
-0.01
Polycyclic Organic Matter
-5.39E-03
Naphthalene
-3.20E-03
Benzene
-4.12E-04
Total
-0.74
B.l-7
-------
Evaluate 2008 NEI to Identify Areas of Improvement that Benefit Use in Residual Risk Assessments
Table B-2. Summary of Emissions By EIS Sectors and Pollutant Category
Name for Negative Ratios (potential over-estimate in 2008 NEI)
EIS Sector
Pollutant Category Name
SumOf(2008v2 emiss) -
(2008 est by 2011 ratio)
Cyanide Compounds
-0.19
Arsenic Compounds
-0.09
Lead Compounds
-0.03
Chromium Compounds
-0.03
Cadmium Compounds
-0.02
Formaldehyde
-5.52E-04
Benzyl Chloride
-5.98E-05
Methylhydrazine
-4.82E-05
Naphthalene
-4.54E-05
Ethyl Chloride
-3.93E-05
Methyl Chloroform
-5.85E-06
Fuel Comb - Industrial
Boilers, ICEs - Coal
Methyl Methacrylate
-5.85E-06
Chloroform
-2.83E-06
Methyl Tert-Butyl Ether
-3.44E-07
Bromoform
-3.03E-07
Xylenes (Mixed Isomers)
-2.89E-07
Dimethyl Sulfate
-2.87E-07
Biphenyl
-1.84E-07
Cumene
-1.66E-07
2-Chloroacetophenone
-1.48E-07
Polycyclic Organic Matter
-1.20E-07
Vinyl Acetate
-1.13E-07
Ethylene Dibromide
-9.06E-08
2,4-Dinitrotoluene
-6.39E-08
Total
-0.35
Chromium Compounds
-0.17
Mercury Compounds
-0.04
Lead Compounds
-0.03
Miscellaneous Non-
Cadmium Compounds
-0.02
Industrial NEC
Nickel Compounds
-1.53E-03
Arsenic Compounds
-1.20E-03
Beryllium Compounds
-5.48E-05
Total
-0.26
Fuel Comb -
Lead Compounds
-0.04
Comm/Institutional - Other
Formaldehyde
-0.02
B.l-8
-------
Evaluate 2008 NEI to Identify Areas of Improvement that Benefit Use in Residual Risk Assessments
Table B-2. Summary of Emissions By EIS Sectors and Pollutant Category
Name for Negative Ratios (potential over-estimate in 2008 NEI)
EIS Sector
Pollutant Category Name
SumOf(2008v2 emiss) -
(2008 est by 2011 ratio)
Polycyclic Organic Matter
-3.14E-03
Benzene
-8.29E-05
Naphthalene
-2.92E-05
Acetaldehyde
-6.50E-07
Total
-0.07
B.l-9
-------
Evaluate 2008 NEI to Identify Areas of Improvement that Benefit Use in Residual Risk Assessments
Table B-3. Summary of Emissions By EIS Sectors and Pollutant Category
Name for Positive Ratios (Potential under-estimate in 2008 NEI)
EIS Sector
Pollutant Category Name
SumOf(2008v2 emiss) -
(2008 est by 2011 ratio)
Solvent - Industrial Surface
Coating & Solvent Use
Toluene
885.45
Methyl Isobutyl Ketone
426.76
Ethylene Glycol
88.44
Xylenes (Mixed Isomers)
30.03
Glycol Ethers
19.56
Acetaldehyde
8.16
Methyl Chloroform
1.33
Dibutyl Phthalate
0.99
Naphthalene
0.23
Cumene
0.02
Total
1,460.97
Solvent - Consumer &
Commercial Solvent Use
Xylenes (Mixed Isomers)
877.58
Toluene
224.98
Ethylbenzene
80.99
Total
1,183.55
Fuel Comb -
Comm/Institutional - Natural
Gas
Formaldehyde
147.64
Benzene
87.38
Acetaldehyde
0.94
Naphthalene
2.39E-04
Lead Compounds
1.65E-04
Polycyclic Organic Matter
9.16E-07
Total
235.95
Fuel Comb - Residential -
Natural Gas
Formaldehyde
112.36
Benzene
59.41
Naphthalene
0.10
Acetaldehyde
0.05
Polycyclic Organic Matter
2.55E-03
Total
171.93
Fuel Comb - Industrial
Boilers, ICEs - Natural Gas
Formaldehyde
39.97
Benzene
23.18
Acetaldehyde
0.05
Lead Compounds
5.00E-03
Polycyclic Organic Matter
8.21E-05
Naphthalene
1.43E-05
B.l-10
-------
Evaluate 2008 NEI to Identify Areas of Improvement that Benefit Use in Residual Risk Assessments
Table B-3. Summary of Emissions By EIS Sectors and Pollutant Category
Name for Positive Ratios (Potential under-estimate in 2008 NEI)
EIS Sector
Pollutant Category Name
SumOf(2008v2 emiss) -
(2008 est by 2011 ratio)
Total
63.21
Polycyclic Organic Matter
6.73
Xylenes (Mixed Isomers)
2.65
Benzene
1.92
Formaldehyde
1.53
Acetaldehyde
1.11
Styrene
0.72
Toluene
0.66
Propionaldehyde
0.29
Ethylbenzene
0.14
Commercial Cooking
Naphthalene
0.13
Phenol
0.08
Ethylene Dichloride
0.06
4-Nitrophenol
0.02
Cresol/Cresylic Acid (Mixed
Isomers)
0.02
Biphenyl
9.27E-03
Acetophenone
8.60E-03
Dibutyl Phthalate
6.47E-03
Total
16.08
Ethylene Dichloride
5.56
Hexane
1.29
Toluene
1.05
Benzene
0.72
Gas Stations
2,2,4-Trimethylpentane
0.64
Xylenes (Mixed Isomers)
0.40
Ethylbenzene
0.08
Naphthalene
0.04
Cumene
0.01
Total
9.79
Chromium Compounds
3.34
Fuel Comb - Industrial
Boilers, ICEs - Coal
Lead Compounds
3.29
Arsenic Compounds
1.04
Formaldehyde
0.28
Cadmium Compounds
0.08
B.l-11
-------
Evaluate 2008 NEI to Identify Areas of Improvement that Benefit Use in Residual Risk Assessments
Table B-3. Summary of Emissions By EIS Sectors and Pollutant Category
Name for Positive Ratios (Potential under-estimate in 2008 NEI)
EIS Sector
Pollutant Category Name
SumOf(2008v2 emiss) -
(2008 est by 2011 ratio)
Cyanide Compounds
0.02
Polycyclic Organic Matter
7.85E-03
Ethylene Dibromide
4.20E-03
Beryllium Compounds
2.25E-03
Selenium Compounds
1.49E-03
Manganese Compounds
5.62E-04
Nickel Compounds
3.21E-04
Mercury Compounds
9.52E-05
Cumene
6.32E-05
Methyl Tert-Butyl Ether
5.93E-05
2-Chloroacetophenone
3.94E-05
Xylenes (Mixed Isomers)
3.16E-05
Vinyl Acetate
3.10E-05
Dimethyl Sulfate
1.47E-05
Benzyl Chloride
7.79E-06
Bromoform
3.53E-06
Naphthalene
1.22E-06
Methylhydrazine
1.04E-06
Chloroform
6.24E-07
Ethyl Chloride
5.09E-07
Biphenyl
2.55E-07
Methyl Methacrylate
2.34E-07
Methyl Chloroform
2.34E-07
2,4-Dinitrotoluene
8.04E-08
Total
8.07
Formaldehyde
6.72
Benzene
0.04
Fuel Comb - Residential -
Naphthalene
0.01
Other
Polycyclic Organic Matter
4.08E-04
Acetaldehyde
2.20E-04
Total
6.76
Formaldehyde
2.39
Fuel Comb - Residential - Oil
Chromium Compounds
2.25
Acetaldehyde
0.24
Selenium Compounds
0.10
B.l-12
-------
Evaluate 2008 NEI to Identify Areas of Improvement that Benefit Use in Residual Risk Assessments
Table B-3. Summary of Emissions By EIS Sectors and Pollutant Category
Name for Positive Ratios (Potential under-estimate in 2008 NEI)
EIS Sector
Pollutant Category Name
SumOf(2008v2 emiss) -
(2008 est by 2011 ratio)
Naphthalene
0.08
Lead Compounds
0.05
Manganese Compounds
0.04
Arsenic Compounds
0.02
Cadmium Compounds
0.02
Beryllium Compounds
0.02
Mercury Compounds
0.02
Nickel Compounds
0.02
Benzene
0.02
Polycyclic Organic Matter
4.50E-03
Total
5.26
Formaldehyde
3.95
Acetaldehyde
0.41
Fuel Comb -
Comm/Institutional - Other
Benzene
0.12
Naphthalene
7.07E-04
Lead Compounds
9.22E-05
Polycyclic Organic Matter
1.04E-05
Total
4.48
Formaldehyde
0.76
Chromium Compounds
0.46
Benzene
0.06
Selenium Compounds
0.02
Nickel Compounds
0.02
Manganese Compounds
0.02
Fuel Comb -
Comm/Institutional - Oil
Arsenic Compounds
0.01
Naphthalene
8.11E-03
Acetaldehyde
5.02E-04
Lead Compounds
2.30E-04
Cadmium Compounds
5.97E-05
Mercury Compounds
3.67E-05
Polycyclic Organic Matter
3.30E-05
Beryllium Compounds
2.96E-05
Total
1.36
Fuel Comb - Industrial
Benzene
0.37
Boilers, ICEs - Oil
Chromium Compounds
0.31
B.l-13
-------
Evaluate 2008 NEI to Identify Areas of Improvement that Benefit Use in Residual Risk Assessments
Table B-3. Summary of Emissions By EIS Sectors and Pollutant Category
Name for Positive Ratios (Potential under-estimate in 2008 NEI)
EIS Sector
Pollutant Category Name
SumOf(2008v2 emiss) -
(2008 est by 2011 ratio)
Nickel Compounds
0.08
Formaldehyde
0.07
Selenium Compounds
0.02
Manganese Compounds
0.02
Arsenic Compounds
0.02
Lead Compounds
0.01
Naphthalene
5.49E-04
Acetaldehyde
2.12E-04
Polycyclic Organic Matter
5.36E-05
Cadmium Compounds
2.91E-07
Mercury Compounds
1.93E-07
Beryllium Compounds
1.66E-07
Total
0.90
Lead Compounds
0.75
Chromium Compounds
0.01
Arsenic Compounds
5.73E-03
Cyanide Compounds
1.54E-03
Formaldehyde
7.66E-04
Cadmium Compounds
5.82E-04
Polycyclic Organic Matter
1.11E-04
Selenium Compounds
2.05E-05
Manganese Compounds
7.74E-06
Nickel Compounds
4.42E-06
Fuel Comb -
Mercury Compounds
1.31E-06
Comm/Institutional - Coal
Beryllium Compounds
3.32E-07
Benzyl Chloride
1.45E-07
Methyl Tert-Butyl Ether
1.23E-07
2-Chloroacetophenone
1.19E-07
Methylhydrazine
1.16E-07
Naphthalene
1.15E-07
Cumene
1.14E-07
Bromoform
1.12E-07
Vinyl Acetate
1.09E-07
Methyl Methacrylate
1.04E-07
Methyl Chloroform
1.04E-07
B.l-14
-------
Evaluate 2008 NEI to Identify Areas of Improvement that Benefit Use in Residual Risk Assessments
Table B-3. Summary of Emissions By EIS Sectors and Pollutant Category
Name for Positive Ratios (Potential under-estimate in 2008 NEI)
EIS Sector
Pollutant Category Name
SumOf(2008v2 emiss) -
(2008 est by 2011 ratio)
Ethyl Chloride
1.04E-07
Xylenes (Mixed Isomers)
9.75E-08
Chloroform
9.57E-08
Dimethyl Sulfate
8.56E-08
Ethylene Dibromide
7.58E-08
Biphenyl
5.52E-08
2,4-Dinitrotoluene
3.11E-08
Total
0.77
Miscellaneous Non-Industrial
NEC
Mercury Compounds
0.12
Lead Compounds
0.03
Nickel Compounds
5.57E-04
Arsenic Compounds
4.38E-04
Cadmium Compounds
2.87E-04
Beryllium Compounds
2.05E-05
Chromium Compounds
6.59E-06
Total
0.16
B.l-15
-------
Evaluate 2008 NEI to Identify Areas of Improvement that Benefit Use in Residual Risk Assessments
Appendix C - RTR TRI Data Not Used in 2008 NEI (found in the
"MISSING RTR_TRI_IN_2008NEI" Microsoft Access database)
-------
United States Office of Air Quality Planning and Standards Publication No. EPA-454/R-19-013
Environmental Protection Air Quality Assessment Division December 2012
Agency Research Triangle Park, NC
------- |