I A \
PKO^
Emissions Inventory Guidance for
Implementation of Ozone and Particulate Matter
National Ambient Air Quality Standards
(NAAQS) and Regional Haze Regulations
-------
-------
EPA-454/B-17-003
July 2017
Emissions Inventory Guidance for Implementation of Ozone and Particulate Matter National
Ambient Air Quality Standards (NAAQS) and Regional Haze Regulations
U.S. Environmental Protection Agency
Office of Air Quality Planning and Standards
Air Quality Assessment Division
Research Triangle Park, NC
-------
Table of Contents
List of Tables v
Acronyms and Abbreviations Used in this Guidance vi
1 Introduction 1
1.1 Purpose 1
1.2 Other Inventory Guidance and Resources 2
1.3 Summary of Relevant Statutory Provisions 4
1.4 Summary of Guidance Contents 7
2 SIP Inventory Overview 8
2.1 Recommended Steps for Building the Inventory Parts of a SIP 8
2.2 Discuss the Inventory Approval Process with EPA 10
2.3 Planning and Modeling Inventory Components 12
2.3.1 Ozone SIP Inventory Components 13
2.3.2 Annual and 24-Hour PM2.5 SIP Inventory Components 14
2.3.3 Regional Haze SIP Inventory Components 15
2.4 Other Relevant Federal Emissions Reporting Requirements 17
2.5 Emissions Inventory Definitions 18
2.5.1 Source Category/Data Category 19
2.5.2 Types of Emissions: Annual, Seasonal, Actual, Potential, and Allowable 19
2.5.3 Duration and Timing of Emissions 21
2.5.4 Data Elements 21
2.5.5 Emissions Reporting Data Elements 21
2.5.6 Throughput and Activity 23
2.5.7 Emission Factors 23
2.5.8 Point Sources 24
2.5.9 Nonpoint Sources 26
2.5.10 On-road Mobile Sources 26
2.5.11 Nonroad Mobile Sources 26
2.5.12 Events 27
2.5.13 Biogenic Sources 27
2.5.14 Inventory Codes 28
i
-------
2.5.15 Point Source Facility Details 29
2.5.16 Alternative IDs 31
2.5.17 Location Definitions 31
2.5.18 Control Data Elements 32
2.5.19 EIS Sectors, Tiers, and Facility Type 33
2.5.20 Other Terms 34
3 SIP Inventory Requirements and Recommendations 35
3.1 Inventory Component Timing 35
3.2 Inventory Preparation Plan 36
3.3 Base Year Inventory 38
3.4 Rate of Progress and Reasonable Further Progress Baseline NAA Inventory 39
3.4.1 Ozone ROP/RFP Requirements and Timing 40
3.4.2 Ozone ROP/RFP Emissions Guidance 40
3.4.3 PM2.5 RFP Requirements and Timing 41
3.4.4 PM2.5 RFP Emissions Guidance 41
3.5 Periodic Inventories 42
3.5.1 Ozone Periodic Inventory 42
3.5.2 Regional Haze Recent Year Inventory 44
3.6 Emissions Statement Requirement for SIPs 45
3.7 Maintenance Inventories 45
3.8 Projected Inventories 46
3.8.1 Ozone NAAQS Implementation 47
3.8.2 PM2.5 NAAQS Implementation 48
3.8.3 Regional Haze Implementation 48
3.8.4 Transportation Conformity Program and Motor Vehicle Emission Budgets 49
3.8.5 Additional Relevant Provisions 49
3.9 Public Hearings 50
3.10 Submit Inventory Parts of SIP to the EPA 52
3.10.1 Regulatory Requirements in Addition to Implementation Rules 52
3.10.2 How to Submit Inventory Parts of a SIP 53
3.10.3 What to Submit 54
4 Developing Current Emissions inventories 59
ii
-------
4.1 Inventory Years 59
4.1.1 Ozone Baseline Inventory Year 59
4.1.2 PM2.5 Baseline Inventory Year 60
4.1.3 Regional Haze Inventory Years 61
4.1.4 Considerations for Choosing an Alternative Baseline Year 62
4.2 Pollutants and Pollutant Precursors to Include in Inventories 62
4.2.1 Condensable PM Emissions 63
4.2.2 Regulatory Definition of VOC 65
4.2.3 Treatment of Methane 65
4.3 Identification of Priority Sources 66
4.4 Spatial Extent 69
4.4.1 Nonattainment Areas with Partial Counties 69
4.4.2 Point Source Locations Needed for Modeling of Traditional Nonpoint Sources 71
4.4.3 Shapefile-based Inventories 71
4.4.4 Link-based On-road Mobile Emissions and Travel Demand Models 71
4.5 Temporal Basis 72
4.5.1 Temporal Considerations for Ozone SIPs 72
4.5.2 Temporal Considerations for PM2.5 SIPs 73
4.6 Emission Factors for SIP Emissions Inventories 74
4.7 Inventory Data Codes 76
4.7.1 Codes to Use for Nonattainment Area Inventories 76
4.7.2 How to Request New Codes or Code Retirements 77
4.8 Estimation of Base Year Emissions 77
4.8.1 Stationary Point 77
4.8.2 Stationary Nonpoint 82
4.8.3 On-road Mobile 86
4.8.4 Nonroad Mobile Equipment 86
4.8.5 Other Nonroad Mobile 87
4.8.6 Biogenic and Geogenic Sources 96
4.8.7 Wildland Fires 99
4.9 Quality Assurance 101
4.9.1 QA Reference Materials and Planning 102
iii
-------
4.9.2 Types of Problems in Emissions Inventories 103
4.9.3 Checks Available through the EPA's Emissions Inventory System 103
4.9.4 Inventory Comparisons 104
4.9.5 Checking Data Codes 105
4.9.6 Point Versus Nonpoint and Reconciliation 105
4.9.7 Ranking Techniques 106
4.9.8 Outlier Analyses 108
4.9.9 Using Maps to QA Nonpoint Emissions 108
4.9.10 Expected Pollutants 108
4.9.11 Facility and Release Point Locations 109
4.9.12 Release Point Parameters 110
4.9.13 Biogenic QA/QC Ill
4.9.14 Timing of QA and Follow-up 112
5 Developing Projected Emissions Inventories 113
5.1 Projected Emissions Overview 113
5.1.1 Readily Available Projections Information 113
5.1.2 Background Materials 113
5.2 Identify and Prioritize Key Sectors and Regions for Projections 115
5.3 Collect Available Data and Models for Each Sector Needing Projections 115
5.3.1 Electric Generating Units 116
5.3.2 Non-EGU Stationary Sources 122
5.3.3 On-Road Mobile Sources 123
5.3.4 Nonroad Mobile Equipment 123
5.3.5 Other Nonroad Mobile Sources 124
5.3.6 Other Projection Resources 125
5.4 Evaluate and Refine Data for Projections in Inventory Region 127
5.5 QA for Projected Inventories 128
6 References 130
iv
-------
List of Tables
Table 1: List of EPA estimation resources for emissions inventories 2
Table 2: Steps for building the inventory parts of a SIP 9
Table 3: Recommended items to include in inventory information provided to the EPA 12
Table 4: References for Ozone SIP emissions inventory components 13
Table 5: References for PM2.5 SIP emissions inventory components 14
Table 6: References for regional haze SIP emissions inventory components 17
Table 7: Point source triennial reporting thresholds (potential to emit) for criteria pollutants in the Air
Emissions Reporting Rule 25
Table 8: Timing of emissions inventory components 35
Table 9: Checklist of elements to consider for including in Inventory Preparation Plans 37
Table 10: Conditions for AERR submittal to meet ozone periodic inventory requirement 43
Table 11: Suggested elements of an emissions inventory SIP document 55
Table 12: Suggested summaries to include in the background 56
Table 13: Summaries to include in the point sources section 57
Table 14: Summaries to include in the nonpoint and mobile sources sections 58
Table 15: Source types expected to include condensable PM 63
Table 16: Hierarchy of emission factors for use in point source emissions inventories 75
Table 17: List of nonpoint categories, pollutants, and available approaches and tools 83
Table 18: Aircraft SCCs and data categories in EPA estimates 89
Table 19: Commercial marine SCCs and emission types in EPA emissions 91
Table 20: Commercial marine vessel recent regulations and impacted pollutants 92
Table 21: Current EPA methods for commercial marine vessel emissions 92
Table 22: Locomotive SCCs 95
v
-------
Acronyms and Abbreviations Used in this Guidance
AEO
Annual Energy Outlook
AERR
Air Emissions Reporting Rule
APU
Auxiliary power unit
BEIS
Biogenic Emissions Inventory System
BELD3
Biogenic Emissions Land Use Database, version 3
AT
Air Taxis
CI
Category 1 (marine vessel)
C2
Category 2 (marine vessel)
C3
Category 3 (marine vessel)
CAA
Clean Air Act
CAMD
Clean Air Markets Division
CAP
Criteria air pollutant
CAS
Chemical Abstracts Service
CEMS
Continuous Emissions Monitoring System(s)
CMV
Commercial marine vessel
CNG
Compressed natural gas
CO
Carbon monoxide
CO 2
Carbon dioxide
DARS
Data Attribute Rating System
DOE
Department of Energy
ECA-IMO
Emissions Control Area-International Maritime Organization
EDD
Effective date of designation
EDMS
Emissions and Dispersion Modeling System
EE
Energy efficiency
EGU
Electric Generating Unit
EIA
Energy Information Administration
El IP
Emissions Inventory Improvement Program
EIS
Emissions Inventory System
EMCH
Emissions Modeling Clearinghouse
EPA
Environmental Protection Agency
EPMM
Electric Power Market Model
ERTAC
Eastern Regional Technical Advisory Committee
FAA
Federal Aviation Administration
FHWA
Federal Highway Administration
FIPS
Federal Information Processing Standards
FRS
Federal Registry Service
GA
General Aviation
GHG
Greenhouse gas
GIS
Geographic Information Systems
GSE
Ground support equipment
vi
-------
HAP Hazardous air pollutant
HEDD High electricity demand day
Hg Mercury
HCI Hydrochloric acid
IPM Integrated Planning Model
IPP Inventory Preparation Plan
LPG Liquified petroleum gas
LTO Landing and takeoff
MARKAL Market Allocation (model)
MEGAN Model of Emissions of Gases and Aerosols from Nature
MIMS Multimedia Integrated Modeling System
MOVES Motor Vehicle Emissions System
NAAQS National Ambient Air Quality Standards
NAA Nonattainment area
NAD North American Datum
NAICS North American Industry Classification System
NATA National Air Toxics Assessment
NEEDS National Electric Energy Data
NEI National Emissions Inventory
NEMS National Energy Modeling System
NESHAP National Emission Standard for Hazardous Air Pollutants
NH3 Ammonia
NLDN National Lightning Detection Network
NMIM National Mobile Inventory Model
NONROAD Nonroad emissions model
NMHC Non-methane hydrocarbon
NOx Nitrogen Oxides
03 Ozone
OAP Office of Atmospheric Programs
OAQPS Office of Air Quality Planning and Standards
OAR Office of Air and Radiation
OECA Office of Enforcement and Compliance Assurance
ORIS Office of Regulatory Information Systems
OTAQ Office of Transportation Air Quality
OTR Ozone Transport Region
PM Particulate matter
PM2.5 Particulate matter with diameter 2.5 microns and less
PM 10 Particulate matter with diameter 10 microns and less
QAPP Quality Assurance Project Plan
RE Renewable energy
RICE Reciprocating Internal Combustion Engines
RFP Reasonable Further Progress
vii
-------
ROP
Rate of Progress
see
Source classification code
SIP
State implementation plan
SMOKE
Sparse Matrix Operator Kernel Emissions
S02
Sulfur Dioxide
SSM
Startup, Shutdown, and Malfunction
STEEM
Ship Traffic, Energy and Environmental Model
TAF
Terminal Area Forecast
TCA
1,1,1-trichloroethane
TDM
Travel demand model
TIP
Tribal Implementation Plan
TOG
Total organic gases
TRI
Toxics Release Inventory
U.S.
United States
USDA
U.S. Department of Agriculture
VMT
Vehicle miles traveled
WFU
Wildland fire use
WGS
World Geodetic System
WLF
Wildland fire
WRF
Weather Research and Forecasting
VOC
Volatile organic compounds
viii
-------
1 Introduction
ose
This document provides guidance on how to develop emissions inventories to meet state
implementation plan (SIP) requirements for complying with the 8-hour ozone national ambient air
quality standard (NAAQS), the 24-hour and annual fine particulate matter (PM2.5) NAAQS, and the
regional haze regulations. It is intended for use by the United States (U.S.) Environmental Protection
Agency (EPA) Regional offices; state, local, and tribal air quality management authorities; and the
general public. This guidance is designed to implement national policy on these emissions inventories as
embodied in the Clean Air Act (CAA), the Ozone SIP Requirements Rule for the 2008 ozone NAAQS (also
called the Ozone Implementation Rule)1, the PM2.5 SIP Requirements Rule2, and the Regional Haze
Regulations (RHR).3
This document does not substitute for provisions or regulations of the CAA enumerated above, nor is it a
regulation itself. As the term "guidance" suggests, it provides recommendations on how to implement
the inventory requirements. Thus, it does not impose binding, enforceable requirements on any party,
nor does it assure that the EPA will approve all instances of its application, as the guidance may not
apply to a particular situation based upon the specific circumstances.
The EPA and state, local, and tribal decision makers retain the discretion to adopt approaches on a case-
by-case basis that differ from this guidance where appropriate. Final decisions by the EPA regarding a
particular SIP demonstration will only be made based on the statute and regulations, and will only be
made following a final submission and after notice and opportunity for public review and comment.
Interested parties are free to raise questions and objections about the appropriateness of the
application of this guidance to a particular situation; the EPA will, and state, local, and tribal air agencies
should, consider whether or not the recommendations in this guidance are appropriate in that situation.
Through the provisions of the Tribal Authority Rule, tribes are not required to develop emissions
inventories. If tribes choose to develop an emissions inventory that may become part of a SIP or Tribal
Implementation Plan (TIP) inventory covered by this guidance, the tribes are encouraged to follow the
provisions of this guidance. Notwithstanding the possible inventory development by local and tribal air
agencies, the remainder of this document will refer to inventories prepared for "SIPs" for compliance
with NAAQS.
1 Implementation of the 2008 National Ambient Air Quality Standards for Ozone: State Implementation Plan Requirements;
(40 CFR part 51 Subpart AA; see also https://www.epa.gov/ozone-pollution/implementation-2008-national-ambient-air-
qualitv-standards-naaqs-ozone-state).
2 Fine Particulate Matter National Ambient Air Quality Standards: State Implementation Plan Requirements, (40 CFR part 51
Subpart Z; see also https://www.epa.gov/pm-pollution/implementation-national-ambient-air-qualitv-standards-naaqs-fine-
particulate-matter).
3 Regional Haze Regulations (40 CFR part 51 Subpart P) See also https://www.epa.gov/visibilitv/visibilitv-regulatorv-actions.
1
-------
1.2 Other Inventory GiiMaii.ce and Resources
This document is a guide for state, local, and tribal air agencies (called "air agencies" in this guidance)
for preparing and submitting their emissions inventories for the 8-hour ozone NAAQS, 24-hour and
annual PM2.5 NAAQS, and for the regional haze program. It is not a procedures document that
comprehensively covers the methods for compiling and documenting emissions inventories.
Furthermore, the predecessors to this guidance have historically relied on separate guidance documents
to add information for various specific purposes. In Section 2.3, we list the various inventory
components needed for 8-hour ozone, PM2.5, and regional haze implementation and identify the
associated sections of this guidance and other relevant guidance materials.
In addition to this guidance, the "Guidance on the Use of Models and Other Analyses for Demonstrating
Attainment for Air Quality Goals for Ozone, PM2.5, and Regional Haze" (http://www3.epa.gov/ttn/
scram/guidance sip.htm), or "Modeling Guidance," covers the creation of emissions inventories for use
in air quality modeling. These modeling inventories differ from the inventories covered in this document
in that they usually have a larger spatial extent (i.e., a multi-state area as opposed to just the
nonattainment area or individual state) and require extra processing steps to obtain the spatial and
temporal resolution of the air quality model input species. To the extent the Modeling Guidance covers
future-year emissions, efforts have been made to harmonize the future-year guidance in that Guidance
with the information included in Section 5 of this document.
In accordance with 40 CFR 51.5(b), states are urged to use applicable, state-of-the art techniques for
estimating emissions. Types of emissions inventories are discussed in Section 3 and include base year
and projected inventories. A number of inventory development resources are available from the EPA for
use by air agencies for estimating base year emissions inventories. Table 1 provides these references
and a brief description of how they can be used. More information on how these resources can be
applied is provided in Sections 4 and 5.
Table 1: List of EPA estimation resources for emissions inventories
Resource and website
Description
National Emissions Inventory (NEI)
Summaries of the latest NEI data and documentation
are available from links provided at this website. More
detailed data are available to air agencies from the
Emissions Inventory System (EIS).
Emission Factors used for the National
Emissions Inventory (NEI)A
The National Emissions Inventory (NEI) program
provides emissions estimation tools (that include
emission factors) and point-source emission factors
based on EPA rule development efforts. In many cases,
these factors are more recent than those in WebFire
and AP-42, and agencies are encouraged to rely on
these where available.
Emissions Inventory System
The EIS allows for summaries and extraction of
detailed data for use by air agencies. The EIS allows
both state-submitted as well as the composite EPA-
2
-------
Resource and website
Description
state data used for the NEI to be queried and
downloaded. Training on reports is also available, and
Appendix A provides an EIS quick reference for county
and facility reports.
Local-scale emissions inventory
development
In 2010, the EPA completed a project to assess the
building of multi-pollutant inventories at the scale of
nonattainment areas. A project description and the
final report are available; the latter documents the
recommended actions for state and local agencies that
want to develop local-scale emissions inventories,
including SIP planning inventories.
Electric Generating Unit emissions data
from Continuous Emissions Monitoring
Svstems (CEMS)fl
Carbon dioxide (C02), nitrogen oxides (NOx), sulfur
dioxide (S02), and heat input data from Electric
Generating Units (EGUs) collected by the EPA. Heat
input data can be used to estimate hourly emissions of
other pollutants using emission factors.
Emissions Inventory Improvement Program
(EIIP) Documentation, Volumes 1-9
Emissions estimation procedures for point, area
(nonpoint) and mobile sources. While much of the
information included in these reports is still relevant,
the EIIP documentation for point and nonpoint sources
has some outdated parts, and the sections below
indicate some of those limitations and alternative
resources. The EIIP documentation for mobile sources
(volume 4) is out of date because it refers to previous
versions of EPA emissions models. Use more current
resources, listed below for mobile sources and fires.
While EIIP is also out of date for biogenic sources,
these sources are relevant only for modeling
inventories and not for the inventories covered by this
guidance.
Emission Factors - WebFire and AP-42
Emission factors are used to estimate emissions,
generally by multiplying by some activity value for that
emissions activity, where no better method of
emissions quantification or estimation is available. AP-
42 is not an appropriate source of emission factors for
mobile sources other than road dust
General Principles for the 5-Year Regional
Haze Progress Reports for the Initial
Regional Haze State Implementation
Plans
This document has been developed by the EPA for the
EPA Regional offices and states in preparing and
reviewing the 5-year progress reports for the initial
Regional Haze SIPs. This document summarizes and
clarifies the requirements for the first 5-year reports,
including emissions inventory aspects.
3
-------
Resource and website
Description
Motor Vehicle Emissions System (MOVES)
The model, technical documentation, and guidance on
the use and application of the MOVES model for on-
road and nonroad mobile sources.
State and Local Transportation Policy and
Guidance
Webpage that provides state and local agencies with
direction on how to implement strategies for
controlling emissions from mobile sources. Includes
links to guidance documents that describe process for
securing SIP air quality credits for voluntary and
innovative emission reduction programs. See also link
to nonroad documentation now available from the
MOVES page.
Transportation Conformity
General information, policy and technical guidance
about transportation conformity.
X From this main web page, obtain factors using the link to the latest available NEI or NEI Plan.
A Select "Data Sets and Published Reports".
Many other resources from other organizations exist. Where specific methods or tools from non-EPA
organizations are included in this documentation, this document includes relevant references for those
materials to help readers find these additional sources of information.
In addition to base year emissions, emissions projections are needed for a variety of reasons, including
redesignation maintenance plans, the attainment projected inventory for a nonattainment area (NAA),
air quality modeling for attainment plans, and projecting reasonable progress goals for visibility
improvement as part of Regional Haze SIPs. In Section 5, this guidance provides information on methods
for emissions projections. The EPA provides emissions projections data and documentation as part of its
modeling platforms that are available at the Emissions Modeling Clearinghouse
(https://www.epa.gov/air-emissions-modeling). On-road and nonroad mobile source projections are
covered in the mobile regulatory documentation associated with MOVES. Additional nonroad mobile
projections information (for aircraft, commercial marine vessels, and locomotives) are included as part
of this documentation in the nonroad mobile sources part of Section 5.
1.3 Summary of Relevant Statutory Provisions
Numerous sections of the CAA apply to the implementation of emissions inventory requirements. This
section provides a brief synopsis of some of those sections that may be relevant for reference in later
sections of this guidance. All of the sections below are a part of CAA Title I - Programs and Activities.
• Section 110(a) describes the basic requirements for SIPs, which provide for
implementation, maintenance, and enforcement of the NAAQS.
o Section 110(a)(1) specifies parameters for "infrastructure SIPs."
o Section 110(a)(2) specifies a "notice and public hearing"4 process is needed.
4 Public hearings are explained at 40 CFR 51.102(a).
General
General
4
-------
o Section 110(a)(2)(F) specifies that infrastructure SIPs shall require, as may
be prescribed by the EPA Administrator, "periodic reports on the nature and
amounts of emissions and emissions-related data" from stationary sources,
o Section 110(a)(2)(K) requires infrastructure SIPs to provide for air quality
modeling as determined necessary by the Administrator and allows the EPA
to require the submission upon request of data related to that air quality
modeling.
o Section 110(a)(2)(J) addresses requirements related to Part C of the CAA,
related to prevention of significant deterioration and visibility protection.
Readers should refer to the "Guidance on Infrastructure State Implementation Plan
(SIP) Elements under Clean Air Act Sections 110(a)(1) and 110(a)(2)"; September
2013.
• Section 110(p) requires states to submit "reports" according to the Administrator's
schedule, "relating to emissions reductions, vehicle miles traveled, congestion
levels, or any other information the Administrator may deem necessary to assess
the development effectiveness, need for revision, or implementation of any plan or
plan revision" for a SIP.
• Sections 169A and 169B focus on visibility protection for Federal class I areas and
require the EPA to promulgate regulations establishing requirements for
implementation plans to assure reasonable progress towards improving visibility. To
do this, the EPA has codified the regional haze program requirements in 40 CFR
51.308 and 51.309, with associated definitions in 51.301.
• Section 172(c)(3) in Part D, Subpart 1 describes NAA SIP requirements for all criteria
air pollutants (CAPs) - carbon monoxide (CO), lead (Pb), nitrogen dioxide (N02),
ozone, particulate matter (PM), and S02-that, unless superseded by subsequent
NAAQS-specific subparts, apply to SIPs for NAAs submitted to the EPA. This section
includes a requirement that such SIPs include a current inventory of actual
emissions in a NAA. It allows the EPA Administrator to require periodic updates
needed to determine if the general requirements of nonattainment SIPs (i.e.,
Subpart 1) are met.
• Section 175A describes maintenance plans, which are SIP revisions that provide for
the maintenance for 10 years of any primary NAAQS for which an area has been
redesignated to attainment and may include additional control measures necessary
to ensure this maintenance. Updated emissions inventories are to be included in a
maintenance plan.
General
Regional
Haze
General
General
5
-------
• Section 182(a)-(f) provide SIP inventory requirements for all areas designated as
nonattainment for the ozone NAAQS.
o Section 182(a) includes SIP requirements for ozone NAAs classified as
Marginal and above.
¦ Section 182(a)(1) requires a current and comprehensive inventory of
actual emissions of all sources of ozone precursors to be included in a
SIP.
¦ Section 182(a)(3)(A) requires periodic inventories to be provided by a
state to EPA, until an area is redesignated to attainment, no later than
the end of each 3-year period after the submission of the inventory for
182(a)(1).
¦ Section 182(a)(3)(B) requires areas to include in their SIP a requirement
for stationary sources of NOx and volatile organic compounds (VOC) to
provide the state with an annual statement showing actual emissions of
NOx and VOC.
o Section 182(b) includes SIP requirements for ozone NAAs classified as
Moderate and above.
¦ Section 182(b)(1)(A) requires a rate of progress (ROP) demonstration,
and describes requirements for annual reductions in VOC and NOx
emissions as necessary to attain the ozone NAAQS.
¦ Section 182(b)(1)(B) requires a baseline emissions inventory for NOx
and VOC from all anthropogenic sources in the NAA to be included in
ozone SIPs for purposes of rate of progress/reasonable further progress
(RFP) demonstrations.
o Section 182(c) includes SIP requirements for ozone NAAs classified as
Serious and above. This section defines major sources for areas classified as
Serious as sources that emit or have the potential to emit at least 50 tons of
VOC per year.
¦ Section 182(c)(2)(A) requires an attainment demonstration based on
photochemical modeling. A modeling inventory is needed for the
attainment demonstration analysis.
¦ Section 182(c)(2)(B) requires a RFP demonstration, and describes
requirements for a timeline of VOC emissions reductions. RFP
demonstrations must include baseline emissions for comparison against
the emissions reductions.
o Section 182(d) defines major sources for areas classified as Severe as
sources that emit or have the potential to emit at least 25 tons of VOC per
year.
o Section 182(e) defines major sources for areas classified as Extreme as
sources that emit or have the potential to emit at least 10 tons of VOC per
year.
Ozone
Ozone
6
-------
o Section 182(f) provides that all major source CAA requirements for VOC also
apply for NOx unless a waiver is received.
• Section 182(j)(l) requires states in a multi-state ozone NAA to take all reasonable
steps to coordinate, substantively and procedurally, the revisions and
implementation of SIPs for a multi-state ozone NAA.
• Section 184(b)(2) defines major sources for areas within the ozone transport region
(OTR) as any stationary source that emits, or has the potential to emit, 50 tons per
year of VOC. Per the requirement for Section 182(f), this definition also applies to
sources of NOx.
• Section 189(a)(2) in Part D, Subpart 4 provides a different and earlier schedule (18
months rather than 3 years) for PM SIPs than provided for in the general provision
in Part D, Subpart 1.
• Section 189(b)(3) (also Part D, Subpart 4) defines major sources in Serious PM NAAs
as sources that emit 70 tons per year of PMi0. Inventories for PM SIPs for Serious
NAAs should establish the inventory of major point sources using the lower
threshold. This section of the CAA has been determined through the courts to be
the appropriate governing section for PM2.5 implementation as well as PM10.
PM2.5
PM2.5
It is important to note the distinction between a requirement in the CAA that an inventory be a SIP
submission or included in a SIP submission, a requirement that a state's SIP provide for the preparation
and submission to EPA of an inventory or other emissions information, and a requirement that a SIP
demonstrate some fact (such that a NAAQS will be attained by a future date) that can only be
established by using sound emission inventory information in an appropriate manner. This distinction is
important because if an inventory is a required SIP submission, then certain procedural requirements
apply. Inventories that are SIP submissions, or included in SIP submissions, must comply with all
procedural requirements that are applicable to any other SIP submission including but not necessarily
limited to those specified CAA section 110 and 40 CFR 51.102, 103 and Appendix V. This document
clarifies which emission inventories are required SIP submissions and which are not.
1.4 Summary of Guidance Contents
This guidance document will define parts of and considerations for emissions inventories to meet SIP
requirements for complying with the 8-hour ozone NAAQS, the PM2.5 NAAQS (24-hour and annual), and
the regional haze regulations. Section 2 provides an overview of the creation, documentation, and
submission of the planning inventories. Section 3 has subsections for each of the separate inventory
requirements or recommendations, with details provided for all requirements except the base year and
projected emissions inventories. Sections 4 and 5 provide the detailed guidance on developing base year
emissions inventories and projected emissions inventories, respectively.
7
-------
2 SIP Inventory Overview
There are two categories of emissions inventories associated with SIP development: planning
inventories and modeling inventories. Modeling inventories are used for modeled attainment
demonstrations and setting reasonable progress goals for visibility protection. Planning inventories are
needed for ozone and PM2.5 SIPs to quantify nonattainment area emissions, and they are used as a part
of RFP analysis and transportation conformity. The primary focus of this guidance is the planning
inventories, whereas the modeling inventories are primarily addressed in the modeling guidance that is
provided separately in the Modeling Guidance document (see Section 1.2). Unless otherwise noted, all
inventories discussed in this guidance are planning inventories.
Generally, the EPA recommends that the planning inventories and modeling inventories be consistent,
though there are exceptions. The exceptions are noted where they arise throughout this guidance and
the Modeling Guidance; however, the EPA generally encourages air agencies to minimize inconsistencies
where possible.
The many aspects of planning inventories are covered more thoroughly in Section 3. These aspects
include the timing for submitting the various inventory components, the inventory preparation plan
(IPP), the base year NAA inventory, the RFP baseline inventory, the ozone periodic inventory, the ozone
emissions statement requirement for SIPs, the maintenance inventories, the attainment projected
inventory for the NAA, public hearings, and inventory approval by the EPA.
In this section, we provide an overview of the creation, documentation, and submission of the planning
inventories. Section 2.1 lists the various steps associated with building the planning inventory parts of a
SIP and identifies the relevant sections of this document that address those steps. The discussions
between air agencies and the EPA regarding the inventory approval process is described in Section 2.2.
In Section 2.3, we list all of the required inventory components, including modeling inventories that are
needed for ozone, PM2.5, or regional haze, but otherwise defer to the Modeling Guidance for the factors
for air agencies to consider in building those inventories. In Sections 2.4 and 2.5, we describe other
federal emissions reporting requirements that are relevant to NAAQS implementation and provide
definitions of key terms needed to understand, develop, and submit emissions inventories for SIPs.
2.1 Recommended Steps for Building the Inven 3
Throughout this document, we describe many technical components that could be included in
inventories in a SIP for ozone, PM2.5, or regional haze implementation. Before air agencies start to create
those components, it is important to recognize that there is a process associated with building them.
This section describes the process agencies should undertake for a successful inventory development
and submittal effort.
The inventory development process involves both required and optional steps. Some optional steps may
be taken by the air agency if it believes it is in its best interest to do so, and optional steps can also
become critical to a particular SIP development process and identified as such by an EPA Regional office.
Thus, a fundamental part of the process is establishing and maintaining open lines of communication
between the air agency and the respective Regional office. The Regional office should be viewed by the
8
-------
air agency as the typical starting point for questions related to planning and modeling inventory
development, and the Regional office staff will direct questions needing EPA headquarters input as they
arise.
A key aspect of this guidance is identifying opportunities for states to leverage existing work to help
them meet the implementation requirements. This starts in Section 2.4 with a description of how work
done in support of the NEI can be helpful for the SIP inventory work.
Table 2 provides a list of steps for completing the planning inventory and related parts of a SIP. The list is
provided in groups of steps that occur in phases presented roughly in chronological order. The second
column of the table indicates the section(s) of this document that give more information on the
particular step. States and Regional offices should look for opportunities to be efficient where possible
to reduce workload. For example:
• an Inventory Preparation Plan for an air agency that has prepared emission inventory SIPs for a
given NAAQS may not be needed, could reference a past plan, or could need fewer details;
• combining documentation from separate steps into a single document where that makes sense;
and
• referencing externally provided inventory documentation where externally prepared emissions
data have been used in the SIP inventory is acceptable (in this case, the air agency would just
explain why the externally prepared data are applicable for the SIP inventory).
Table 2: Steps for building the inventory parts of a SIP
Inventory-related SIP development step
Document
section
Planning
• Review this guidance.
All
• Become familiar with inventory development resources.
1.1
1.2
• During development of the Inventory Preparation Plan, clarify and discuss the
inventory approval process that your Regional office will use, including all
requirements, reporting formats and elements to be included, timing, and
technical documentation contents.
2.2
(general)
3.1 (timing)
• Understand required and potentially required inventory components for the SIP.
Determine which inventories must be created for your SIP.
2.3
• Understand inventory definitions and data requirements.
2.5
• Understand timing requirements of submitted components.
3.1
• Prepare and finalize the Inventory Preparation Plan.
3.2
• Prepare a quality assurance project plan (QAPP)
3.2
• Understand requirements and choose key inventory features, including inventory
year(s), the pollutants and precursors to be included, and priority emissions
sources.
4
Creating Inventories
• Assess what inventories already exist that can be used as a starting point.
2.4
3.3
9
-------
Inventory-related SIP development step
Document
section
• Create base-year (planning) inventories.
4
• Quality assure and refine base year inventory data in accordance with the IPP
and QAPP.
0
• Create projected (planning) emission inventories.
5
• Quality assure and refine projected planning emission inventories in accordance
with the IPP and QAPP.
5.5
Preparing Documentation
• Review requirements with respective Regional office in light of the particular
challenges associated with a given SIP development process, such a first
nonattainment SIP development for a state, area, or air agency staff.
2.2
• Prepare documentation.
3.10
Public Hearings
3.9
Submit to the EPA for Approval
3.10
/ " •. scuss the Inventory Appro . -:sss wl.':
Early in the SIP planning process, air agencies should discuss the emissions inventory approval process
for that specific SIP with their respective EPA Regional office. These discussions should address the
tangible items (e.g., reports, data files, summary spreadsheets) that will be delivered to the EPA as
necessary to support the EPA's review and approval.
A common source of confusion is whether inventories must undergo public review. Inventories that are
SIP submissions, or are included in SIP submissions, must comply with all procedural requirements that
are applicable to any other SIP submission including but not necessarily limited to those specified in CAA
section 110 and 40 CFR 51.102, 103 and Appendix V. Details on the public review issue are addressed in
more detail in Section 3.9.
As new issues arise during the SIP development process, air agencies or the EPA Regional offices may
need to reassess the previously discussed inventory approval process. Both agencies should be aware of
possible changes that are needed as the inventory development process occurs, and either party should
raise concerns about the need for discussion and possible revision of the process. Details of the process
and documentation can be modified through additional discussions as the need arises. As the inventory
development process works far better when lines of communication remain open, the EPA encourages
periodic coordination meetings or check-ins during the inventory development periods.
Table 3 provides a list of tangible reports, summaries, and data files that could be included as part of the
inventory efforts for SIPs. Some of these items are needed to meet statutory and/or regulatory
requirements, and these are listed in Tables 4 through 6 below. The form of reports, summaries, and
data files to be included should be clarified with the Regional office at the start of the inventory
development process for SIPs. States should discuss and confirm with the EPA Regional office the items
to be included in the SIP to avoid inadvertent omission of items required for approval of the inventory
parts of the SIP. While certain recommended items in Table 3 are not specifically required by the
10
-------
regulations, the EPA review of the required elements should include supporting documentation so that
the EPA can assess whether the approaches used for developing inventories are technically sufficient
and meet the requirements to use appropriate methods and current, accurate information.
One notable and important difference between regional haze and the other two programs (ozone
attainment SIPs and PM2.5 attainment SIPs) is that there is no requirement to submit detailed emissions
inventory data as an element of a required Regional Haze SIP. While the regional haze program expects
that inventories will be used to accomplish the required elements of the Regional Haze SIP and the
regional haze regulations require the SIP to provide for the preparation of emission inventories for
particular timeframes,5 the detailed emissions inventory data are not submitted as a SIP element (or SIP
revision) like the data must be for ozone and PM2.5. The recommended items in Table 3, however, will
allow the EPA to make a determination whether the emissions information used in Regional Haze
analysis is sufficient for the purposes of the SIP.
Following receipt of an emissions inventory from air agencies, the EPA will determine if it is part of a SIP
submission or not. If an emissions inventory is provided to the EPA as a component of a SIP submittal,
then EPA will process that SIP submittal, including the emissions inventory component, in accordance
with applicable laws and regulations (FN: CAA section 110 and 40 CFR 51.102, 103 and Appendix V). As
an example, emissions inventories required by CAA section 182(a)(1) are to be provided to EPA as a SIP
submission (FN: See 80 FR 12264,12291). Conversely, if an emissions inventory is provided to the EPA
outside of a SIP submittal, then EPA will review and process that inventory as necessary.
5 See 40 CFR 51.308(d)(4)(v) Regional Haze SIPs must "provide for" statewide emissions inventories of pollutants that
are reasonably anticipated to cause or contribute to visibility impairment in any mandatory Class I Federal area. The
Regional Haze Rule similarly requires state's Regional Haze SIPs for the second and subsequent implementation
periods to provide for such emissions inventories (40 CFR 51.308(f)(6)(v)).
11
-------
Table 3: Recommended items to include in inventory information provided to the EPA
Inventory Item
Ozone
PM2.5
Regional
Haze
Planning Inventories
Inventory Preparation Plan
X
X
X
Emissions Inventory Report, with explanations of methods, quality
assurance and emissions summaries. All inventories can be
included in such a report, including base year and projected
inventories. The inventories to be included depend on the
NAAQS, the nonattainment classification, and the stage of
implementation (initial SIP, redesignation, etc.).
X
X
Emissions Inventory Report, with explanations of methods, quality
assurance and emissions summaries. All inventories can be
included in such a report, including base year (applicable
only for the SIP due on July 31, 2007), most recent year, and
projected inventories.
X
Tables or spreadsheets of emissions summaries included in the
inventory report
X
X
X
Detailed emissions inventory files, databases, or spreadsheets
X
X
Modeling Inventories
Modeling Demonstration Report-The emissions information can
be a section in a larger report or could have its own report.
The content should include methods, quality assurance, and
emissions summaries for the modeling inventories,
including projected inventories. A statewide emissions
summary is needed, and this can be included in the
planning inventory SIP for convenience.
X
X
X
Tables or spreadsheet summaries of emissions inventory used for
modeling
X
X
X
Detailed emissions inventory files, databases, or spreadsheets
X
X
Detailed data about emissions controls used to model
X
X
Other
Emission Statement SIP showing the state's program to require
sources to submit emissions statements annually
X
2,3 Planning and Modeling Inventory Components
The ozone, PM2.5, or regional haze implementation rules have specific inventory requirements for SIPs to
support attainment, maintenance, and/or reasonable progress demonstrations. Additionally, there are
possible inventory details that could be required or advisable for various reasons. One major distinction
between the inventory requirements of the ozone and PM2.5 rules versus the RHR is that the former two
rules require inventories for NAAs, while the latter expects that states develop statewide inventories.
This section provides a list for each of the three rules for the required and possible inventory
components of SIPs, with additional details for the types of planning inventories.
12
-------
2.3.1 Ozone SIP Inventory Components
The Ozone Implementation Rule (40 CFR part 51, Subpart AA) provides the regulatory requirements for
inventories of actual emissions that support ozone NAAQS implementation by air agencies. Table 4
below is a list of components that could be part of preparing an ozone SIP. The governing portion of the
CAA, and/or rule citation and deadline are listed where applicable, as well as sections in this guidance
document or other guidance where they are addressed. The deadlines are expressed in months after the
effective date of designation (EDD) of the area. Note that not all components are required in every case.
Individual SIP requirements may be different depending on attainment designation, year of designation,
and other factors specified in the rules. These details should be discussed with the relevant EPA Regional
office.
States and other air agencies that perform modeled attainment demonstrations should also refer to the
Modeling Guidance (http://www3.epa.gov/ttn/scram/guidance sip.htm), which includes a section on
emissions inventories for modeling. As part of that guidance, emissions inventory summaries are
described for inclusion in technical documentation about the modeled attainment demonstration. To
assist the public when it comments on the SIP submittal and to facilitate EPA review of the SIP
submissions, these summaries of modeling inventories can be submitted as part of the documentation
provided with the emissions inventory SIP, but are not required SIP elements.
Table 4: References for Ozone SIP emissions inventory components
Inventory Component
Required by CAA?
(CAA Section)
Regulatory Citation
(Due Date)
Guidance Sections and/or
Document(s)
Inventory Preparation Plan
Section 3.2
Base year inventory for the
NAA
182(a)(1)
40 CFR 51.1115(a)
(24 months after EDD)
Section 3.3 and Section 4
ROP/RFP baseline NAA
inventory (part of the
ROP/RFP plan(s); for
Moderate and above
areas)
Moderate &
above:
182(b)(1)(B)
Serious & above:
182(c)(2)(B)
40 CFR 51.915
40 CFR 51.1110(b)
(36 months after EDD)
Section 3.4 and Section 4
Periodic NAA inventory
182(a)(3)
40 CFR 51.915
40 CFR 51.1115(b)
(No later than the end
of each 3-year period
after required base
year submission)
Section 3.5 and Section 4
Attainment projected
inventory for the NAA+
Summary required by
40 CFR 51.114(b)
(with plan)
Section 3.8 and Section 5
Maintenance NAA inventory
175A(a)*
Section 3.7 and Section 4
Source Emission Statements
requirement for SIPs
182(a)(3)(B)
40 CFR 51.915
(24 months after EDD)
Section 3.6
Inventories for modeled
attainment
demonstrations (base and
182(c)(2)(A)
40 CFR 51.908
40 CFR 51.1108
(Timing depends on
Seethe Modeling Guidance.
Summaries of inventories
(base and projected
13
-------
Inventory Component
Required by CAA?
(CAA Section)
Regulatory Citation
(Due Date)
Guidance Sections and/or
Document(s)
projected attainment year
modeling inventories)
nonattainment
classification)
attainment year) are
encouraged to be included
in technical documentation.
Public hearing requirement
for SIP inventories
110(a)(2)
40 CFR 51.102
Section 3.9
Planning inventory
documentation
Section 3.10
+ If this inventory is the mechanism by which the ROP/RFP plan requirements will show the emissions reductions required for
ROP/RFP, then this inventory or its summary would necessarily become part of the ROP/RFP plan(s). Motor vehicle emissions
budgets for purposes of transportation conformity are the on-road portion of the projected emissions associated with the
ROP/RFP plan.
* Maintenance NAA inventories are only associated with maintenance SIPs rather than attainment demonstration SIPs.
While modeling inventories are not an explicit requirement of the CAA, it is presumed here that states would not be able to
perform credible modeled attainment demonstrations without modeling inventories.
2,3,2 Annual and 24-Hour PM2.s SIP Inventory Components
The PM2.5 SIP Requirements Rule (40 CFR part 51, Subpart Z) provides the regulatory requirements for
inventories of actual emissions that support PM2.5 NAAQS implementation by state, local, and tribal
agencies. In Table 5 below, we list the inventory components that could be part of a PM2.5 SIP and
provide a map to the governing portion of the CAA, regulations, and the applicable guidance.
For areas that perform modeled attainment demonstrations, please also refer to the Modeling
Guidance, which includes a section on emissions inventories for modeling. As part of that guidance,
summaries are described for statewide inventories. For convenience, these summaries can be included
in the emissions inventory SIP. Areas that do not perform modeled attainment demonstrations would
not need to provide the statewide inventories for PM2.5.
Table 5: References for PM2.5 SIP emissions inventory components
Inventory Component
Required by
CAA?
(CAA Section)
Regulatory Citation
(Due Date)
Guidance Sections
and/or Document(s)
Inventory Preparation Plan
Section 3.2
Base year inventory for the NAA
172(c)(3)
Moderate
40 CFR 51.1008(a)(1)01
(18 months after EDD)
Section 3.3 and Section 4
Attainment projected inventory
for the NAA
Moderate:
40 CFR 51.1008(a)(2)01
(18 months after EDD)
Section 3.8 and Section 5
Maintenance NAA inventory
175A(a) +
This document, Section
3.7 and Section 4
Inventories for modeled
attainment demonstrations
(base year and projected
189(a)(1) for
Moderate
areas and
40 CFR 51.1001 and
Appendix W
(18 months after EDD)
Modeling Guidance.
Summaries of statewide
inventories (base and
projected attainment
14
-------
Inventory Component
Required by
CAA?
(CAA Section)
Regulatory Citation
(Due Date)
Guidance Sections
and/or Document(s)
attainment year modeling
inventories)
189(b)(1) for
Serious areasfl
year) are encouraged to
be included in technical
documentation.
Public hearing requirement for
SIP inventories
110(a)(2)
40 CFR 51.102
Section 3.9
Planning inventory
documentation
Section 3.10
a If an area is reclassified as Serious, the applicable requirements refer back to this subsection from 40 CFR 51.1008(b). If a
Serious area has failed to meet the PM2.5 NAAQS by the Serious area attainment date, then see 40 CFR 51.1008(c), which still
refers back to the Moderate area requirements and has a 12 month deadline.
+ Maintenance NAA inventories are only associated with maintenance SIPs rather than attainment demonstration SIPs.
While modeling inventories are not an explicit requirement of the CAA, it is presumed here that states would not be able to
perform credible modeling attainment demonstrations without modeling inventories.
2,3,3 Regional Haze SIP Inventory Components
The RHR (40 CFR part 51, Subpart P) has significantly different inventory requirements from ozone and
PM2.5 implementation rules. One reason for this difference is the time horizon of the regional haze
program. The program consists of implementation periods that are referred to as the first planning
period, second planning period and so forth. There are parts of regional haze implementation plans that
have been completed via the SIP for the first implementation period (ending in 2018) and may not need
to be revisited in SIPs for later implementation periods, but for completeness we also explain these
here. Another difference is that the RHR requires the use of statewide planning inventories rather than
inventories for only a smaller NAA. Finally, the most important difference is that there is no requirement
that emission inventories be submitted as a SIP revision. While the inventories are needed to support
the SIP, and the SIP must provide for the specified emission inventory information, the detailed
statewide inventories need not be submitted. This section briefly discusses the existing requirements
related to emission inventories for the first planning period, and then goes on to address the same
requirements as proposed for the second and subsequent planning periods.
For the first planning period, the RHR in 40 CFR 51.308(d)(3)(iii) required a SIP submission to include the
technical basis on which the State was relying to determine its apportionment of emission reduction
obligations necessary to achieve reasonable progress towards natural visibility conditions, including
identification of the "baseline emissions inventory" used for such technical analysis. The rule indicated
that the baseline emissions inventory year for the first planning period is presumed to be the most
recent year of the consolidated periodic emissions inventory (as defined in Subpart A and commonly
called the NEI). As a result, this "baseline" inventory used in the SIPs reflects a year from the 2000-2004
period, for which the 2002 was the applicable NEI year. For the first planning period, most states also
used a year from this period as the base year for their regional air quality modeling (states that were
members of the Lake Michigan Air Directors Consortium (LADCO) also used 2005 in their air quality
modeling).
15
-------
In addition, 40 CFR 51.308(d)(4)(v) required that the plan provide for a statewide inventory of emissions
for a "baseline" year, for the most recent year in which data are available ("recent year"), and estimates
of future projected emissions. This section made it clear that the inventories associated with the
planning effort needed to be "statewide." This guidance document does not explain in detail the
emission inventory requirements for SIPs for the first implementation period, most of which have
already been submitted and acted upon by the EPA.
The regional haze rule required a first progress report 5 years after submission of the first periodic
comprehensive SIP revisions in 40 CFR 51.308(g). Because many of the first of these required 5-year
progress report SIP revisions have already been submitted and the remaining progress reports should be
far along in preparation as of the date of this document, this guidance document does not address the
first-required progress reports. For guidance on these first required 5-year progress reports, readers
should use the separate guidance document "General Principles for the 5-Year Regional Haze Progress
Reports for the Initial Regional Haze State Implementation Plans."
Periodic SIP revisions for the second and subsequent planning periods are due on or before July 31,
2021, July 31, 2028, and every 10 years thereafter. 40 CFR 51.308(f) of the RHR outlines the
requirements for these comprehensive, 10-year SIP revisions. Section 51.308(f)(2) specifies that the SIP
revision include a long-term strategy, and section 51.308(f)(2)(iii) requires that the state document the
technical basis for its determinations of the emission reductions from anthropogenic sources that are
necessary for achieving reasonable progress towards natural visibility conditions. The wording in this
section requires that the state's emission information must, at a minimum, include "information on
emissions in a year at least as recent as the most recent year for which the State has submitted emission
inventory information to the Administrator in compliance with the triennial reporting requirements of
subpart A of this part." Additionally, the rule provides a 12-month grace period after the submission of
emissions inventory information before states are required to use that information in a regional haze SIP
submission.
Periodic progress reports for the second and subsequent planning periods are due on January 31, 2025,
July 31, 2033, and every 10 years thereafter (40 CFR 51.308(g)). Progress reports are scheduled to be
submitted midway between the most recent SIP revision and the subsequent "10-year" SIP revisions
(section 51.308(g)). Section 51.308(g)(4) requires states to analyze the change in emissions over the
period since the previous plan revision (i.e., the SIP revisions due in 2021, 2028, and every 10 years
thereafter). Per the rule, emissions changes should be identified by type of source or activity, and this
guidance provides recommendations for what emissions summaries these reports should include to
accomplish that objective. In addition, the rule specifies that the analysis of historical emissions must
extend at least through the most recent year for which the state has submitted emission inventory
information to the EPA in compliance with the triennial reporting requirement of the Air Emissions
Reporting Rule (AERR), 40 CFR part 51, Subpart A. For data (such as EGU data) that are submitted
directly via a centralized EPA emissions data system and then made available by EPA, the data used
must be the most recent year for which the EPA has provided state-level summary data or an internet-
based tool by which the State may obtain such a summary as of a date 6 months before the required
date of the progress report. Further, section 51.308(g)(5) requires an assessment of any significant
16
-------
changes in anthropogenic emissions within or outside the state that have occurred since the previous
SIP revision.
In the RHR, section 51.308(f)(6)(v) mirrors a part of section 51.308(d)(4)(v) by specifying that SIP
revisions for the second and subsequent planning periods must provide for a statewide inventory of
emissions for the most recent year in which data are available ("recent year") and estimates of
projected emissions. Therefore, in its SIP revision for the second and subsequent implementation
periods, a state will need to describe its process for preparing, or providing essential information for the
collective preparation of, a statewide emission inventory for the latest "recent year" and a projection
inventory for the last year of the implementation period. Table 6 summarizes a list of the Regional Haze
SIP emissions inventory components, including the appropriate references to the CAA, the regulations,
and the relevant sections of this guidance.
Table 6: References for regional haze SIP emissions inventory components
Inventory Component
Required?
(CAA Section)
Regulatory Citation
(Due Date)
Guidance Sections
and/or Document(s)
Inventory Preparation Plan
Section 3.2
Baseline statewide inventory
40 CFR
51.308(d)(4)(v)
(completed)
Previously completed as
Recent year statewide inventory
Projected statewide inventory
(For the first planning period)
169A(b)+
part of initial Regional
Haze SIP for the first
planning period.
40 CFR 51.308(g)(4)
Periodic progress report, recent
year statewide inventory
169A(b)+
and (5)
(see guidance noted
above)
Section 3.5.2 and Section
3.10.3
10-year SIP revisions,
40 CFR
Section 3.5.2 and Section
A
recent year statewide
169A(b)+
51.308(f)(6)(v)
inventory^
(varied dates)
10-year SIP revisions,
projected statewide inventory^
169A(b)+
40 CFR
51.308(f)(6)(v)
(varied dates)
Section 3.8 and Section 5
Documentation
Section 3.10
+ CAA 169A(b) provides the authority to provide guidelines to the states on appropriate techniques and methods to implement
the visibility protection goals in the CAA. This has been done through regulations as codified in 40 CFR 51.308 and 51.309.
If modeling is done to support the 10-year SIP revision, these are modeling inventories.
2,4 Other Relevant Federal Emissions Reporting Requirements
The Air Emissions Reporting Rule (AERR), 40 CFR part 51, Subpart A, requires states to provide to the
EPA annual statewide emissions inventory data, regardless of whether that state has nonattainment
areas. This regulation requires reporting of actual emissions of CO, NOx, VOC, S02, PM2.5, particulate
matter with diameter ten microns and less (PM10), and ammonia (NH3). For PM2.5, the AERR requires
states to report primary (i.e., "direct") PM2.5, which is defined as the sum of filterable PM2.5 and
condensable PM. The agencies are required to report the filterable PM2.5 and condensable PM
separately where they are present from the source (see Section 4.2.1). States may delegate the
reporting requirements to local air agencies. Tribes can also submit these data to the NEI, but are not
17
-------
required to do so. These data are used by the EPA's Office of Air and Radiation's (OAR) Office of Air
Quality Planning and Standards (OAQPS) to build the NEl, and a complete version that includes all
sources of emissions (with minor exceptions) is released every 3 years.
The AERR includes both triennial and annual statewide reporting requirements, with more extensive
reporting requirements for triennial inventory years. Triennial years for the NEI include 2011, 2014, and
every 3rd year thereafter. For the interim annual inventories, reporting is limited to emissions data from
only the larger point sources (Type "A" sources), as defined by Appendix A of 40 CFR part 51. Subpart A.
For the triennial inventories, lower point source thresholds are given in Appendix A, consistent with the
definition of major sources in 40 CFR part 70. All other sources of emissions must be reported as
nonpoint or mobile sources on a county basis in the triennial submissions. Additionally, lower point
source thresholds may apply to ozone, PM2.5, PM10, and CO NAAs, depending on the area's
nonattainment classification.
The AERR requirements are relevant for SIP planning inventory purposes in several ways. First, the AERR
and its associated EIS (data system) provide the appropriate data elements (fields to be reported) for
inventories for electronic transfer to the EPA. Second, AERR-based inventories are often used as a
starting point for SIP inventories. Third, in some cases, AERR submissions can meet some of the
inventory requirements described in this document. Finally, the resources provided by the EPA for states
to meet their AERR reporting obligations, such as emissions inventory tools and models, associated
documentation, emission factors, new emissions methods, and quality assurance approaches, can assist
inventory developers when making inventories for SIP planning and modeling purposes.
As a point of clarification, air agencies must meet all relevant planning inventory SIP requirements
regardless of the availability or use of the EPA data associated with the AERR/NEI process. While the EPA
resources can be helpful to states in meeting those requirements, the SIP requirements and the
associated public review and Regional office approval process are independent requirements that must
be met. It would not be sufficient to merely point to the EPA's publication of the NEI as having met the
SIP emissions inventory requirements. However, it may be sufficient for a state to submit as a SIP
revision an inventory that contains some or all of the same data as in the NEI, with sufficient justification
and using the applicable public comment process.
2 nissions Inventory Definitions
Emissions inventory work requires knowledge of many terms and codes. In this section, we define the
terms relevant to inventories that are used throughout this document. This section also references
external materials needed for obtaining definitions of all valid fields that can be used for building
inventories, which are based on the tables used to submit data to the EPA for the AERR. Many
definitions relevant for emissions inventories are defined in the CAA or codified in the AERR and the
implementation rules. These are included here for a convenient reference in a single location since
these terms are used throughout this document. Without these definitions, the meaning of the
subsequent sections would be less clear. The sample below is not intended to represent a
comprehensive list of inventory related definitions.
18
-------
2.5.1 Source Category/Data Category
Traditionally, the term "source category" has been used to identify the major types of emissions
inventory groupings: stationary point sources, stationary area (or nonpoint) sources6, on-road mobile
sources, nonroad mobile sources. In addition, fires and biogenic sources are sometimes included as
additional source categories.
In practice, some inventories have started including large moveable facilities, such as asphalt plants, in
the point sources inventory without associating them with a specific county. Additionally, some mobile
sources such as locomotives and commercial marine vessels (CMVs) are treated as nonpoint data (i.e.,
the formatting of it) for the purpose of reporting these data to the EPA via the EIS (see also Section 4.8.5
as well as the EIS information referenced in Table 1). Similarly, because of the data format and their
physical nature (i.e., a discrete geographic location), rail yards and airports are treated as point data in
the EIS.
With the advent of the EIS, the preferred term for indicating subsets of inventory data has become "data
category" rather than "source category" because the groupings have much more to do with how the
data are created and the data structure than the features of the actual emissions sources included in the
category. For example, locomotives and commercial marine vessels are included in the nonpoint data
category for inventories in the EIS, but in fact are nonroad mobile sources. Additionally, the EIS ushered
in a new data category called "events," which are typically wildfires or prescribed fires (agricultural fires
are part of the nonpoint data category), but in principle could include other significant, short-duration,
emissions events.7
2.5.2 Types of Emissions; Annual, Seasonal, Actual, Potential, and Allowable
Various terms are used in this and other documents to reference various types of emissions. Here, we
provide a brief list of definitions associated with various emissions types.
Actual emissions means the emissions of a pollutant from an affected source determined by taking into
account actual emission rates associated with normal source operation and actual or representative
production rates (i.e., capacity utilization and hours of operation) (40 CFR 51.491). This is in contrast
with potential emissions or allowable emissions. These actual emissions should include emissions of a
pollutant that occur during periods of startup, shutdown, and malfunction (see Section 4.8.1). In
addition, please see Section 4.1 for a description of selecting a year for creating actual emissions
estimates.
Allowable emissions means the emissions of a pollutant from an affected source determined by taking
into account the most stringent of all applicable SIP emissions limits and the level of emissions
6 Over time, stationary area sources have been renamed "nonpoint" sources to avoid confusion with the air toxics
regulatory term "area sources," which are often sources that meet the criteria pollutant definition of point
sources. Likewise, the term "stationary" is often ignored in common use for both the point source and nonpoint
labels.
7 While the EIS is not currently configured to accept event data other than for fires, this means that the idea of an
"event" could be expanded to other types of events that are not currently inventoried because of the difficult
nature and non-routine nature of these types of emissions.
19
-------
consistent with source compliance with all federal requirements related to attainment and maintenance
of the NAAQS and the production rate associated with the maximum rated capacity and hours of
operation (unless the source is subject to federally enforceable limits that restrict the operating rate, or
hours of operation, or both) (40 CFR 51.491).
Annual emissions are defined in the AERR (40 CFR 51.50) as actual emissions for a plant, point, or
process that are measured or calculated to represent a calendar year. This means that annual emissions
should include all emissions of the relevant sources at a facility or from a nonpoint or mobile process
occurring during a calendar year. In general, unless otherwise specified, annual emissions means actual
emissions, as defined above.
Average season day emissions means the sum of all emissions during the applicable season divided by
the number of days in that season. This definition is used in the context of inventories developed for
SIPs associated with the 24-hour PM2.5 NAAQS.
Potential to emit means the maximum capacity of a stationary source to emit a pollutant under its
physical and operational design (40 CFR 51.301). Also called "potential emissions," 40 CFR 51.301 also
defines this term to specify that any physical or operational limitation on the capacity of the source to
emit a pollutant including air pollution control equipment and restrictions on hours of operation or on
the type or amount of material combusted, stored, or processed, shall be treated as part of its design if
the limitation or the effect it would have on emissions is federally enforceable. Secondary emissions do
not count in determining the potential to emit of a stationary source.
As described in "Potential to Emit: A Guide for Small Businesses" (EPA-456/B-98-003), potential to emit
considers the equipment design and can also consider certain controls and limitations on the operation
of the equipment. Potential emissions can be expressed as potential controlled emissions, which are the
pollutant emissions while a source is operating at the maximum design capacity, a schedule of 8,760
hours per year, and considering the efficiency of the control equipment. In contrast, potential
uncontrolled emissions are simply the pollutant emissions while the source is operating at the maximum
design capacity and a schedule of 8,760 hours per year without factoring in controls.
Startup. Shutdown, and Malfunction (SSM) emissions are emissions that occur during startup,
shutdown, or malfunctions at industrial facilities. As a factual matter, the emissions rate during periods
of SSM can be much larger than the emissions rates during periods of normal operating procedure (that
is, not during periods of SSM). However, applicable emissions limitations apply at all times, including
during periods of SSM.8 More information about the startup and shutdown emissions in the SIP
inventory context is described in Sections 4.3 and 4.8.1.
Ozone season day emissions means an average day's emissions for a typical ozone season work
weekday. The state shall select, subject to EPA approval, the particular month(s) in the ozone season
8 On May 22, 2015, the EPA issued a final action to ensure that states have plans in place that are fully consistent
with CAA and court decisions concerning SSM operations. This action included a "SIP call" that directs 36 states
to correct specific SSM provisions in their SIPs. The SIP submission deadline for the states subject to this action is
November 22, 2016. See the rule website for more information.
20
-------
and the day(s) in the work week to be represented, considering the conditions assumed in the
development of RFP plans and/or emissions budgets for transportation conformity.
2.5.3 Duration and Timing of Emissions
Inventory start date means the first day of the inventory period. This can be used for compiling and
reporting monthly or other shorter period emissions, including events.
Inventory end date means the last day of the inventory period. This can be used for compiling and
reporting monthly or other shorter period emissions, including events.
Inventory year or Emissions year means the year for which emissions estimates are reported (40 CFR
51.50).
See also "Emission Type" in codes Section 2.5.14.
2.5.4 Data Elements
The term "data elements" means parts of a data file that contain individual items of data. These are also
called fields or variables. Examples of key data elements associated with emissions inventories include
emissions; activity; emission factors; state/county identifiers; facility, unit, and process identifiers; plant
names; control measure codes; and process codes. This list is just a sample and is not intended to
represent a comprehensive list of required codes (see also Section 2.5.5).
Note that emissions data categories (Section 2.5.1) are not a data element per se, but rather emissions
inventory submissions usually are broken out into separate datasets for each data category.
2.5.5 Emissions Reporting Data Elements
The Ozone Implementation Rule and PM2.5 SIP Requirements Rule refer to the AERR and this guidance
for information on reporting data elements. Since the AERR is focused on emissions submitted for the
annual NEI inventory, it only includes definitions for those data and fields that are required by the NEI
program. For ozone SIP inventories, ozone season day emissions are required, but are optional for
reporting in accordance with the AERR. For PM SIP inventories, either or both annual or average season
day emissions can be included (depending on the circumstances), but the average season day emissions
are optional for reporting for AERR purposes.
The many required and optional data elements for emissions inventories are described in Excel files
called Staging Requirements. These files are comprehensive for all of the required and optional fields
that can be submitted to the EPA for SIP inventories, not only the submissions for the NEI. The shared
field list for NEI and SIP inventories helps air agencies by promoting clarity and familiarity through
common data element definitions across multiple inventory requirements. The descriptions provided in
those Excel files are very limited, so this document provides additional information for the most
important of these fields. These spreadsheets are available at the Emissions Inventory Training website
at https://www.epa.gov/air-emissions-inventories/air-emissions-inventory-training using the links for
"Training Materials" listed for each training on the page. Within the Zip file provided for the "Training
Materials" links, an Excel file is available that includes "staging requirements" and lists the data
elements and their definitions.
21
-------
Although many fields in the Staging Requirements are optional for AERR reporting purposes, some of
these fields can be critical for inventory development for SIPs and even the NEI. By "optional," we mean
those fields that are not explicitly required by the AERR or that are not enforced during electronic
reporting as needing to be provided to successfully submit the data to the EPA. The list below identifies
fields that are "optional" but that are strongly recommended to be included in all emissions inventories,
documentation, and reporting as applicable for the sources in the state or NAA. For example, control
parameters are optional only because emissions controls do not exist on every unit and process in an
inventory, thus the EPA cannot enforce this during electronic reporting as a required field. However, the
control information is required by the AERR when it exists at the emissions source and should be
provided for SIP inventories.
Unit Description: This field describes the unit in words. It can be invaluable in pairing the unit and
corresponding processes to correct source classification codes (SCCs).
Unit design capacity: Unit design capacity means a measure of the size of a point source, based on the
reported maximum continuous throughput or output capacity of the unit (40 CFR 51.50). Useful for
boilers, this field is needed when trying to understand the cost per ton of pollutant emissions reduced
associated with controls on boilers. The cost per ton is generally higher for smaller boilers. The design
capacity can be critical in SIP planning for ensuring that good decisions are made about controls on
stationary boilers and other units. While required by the AERR, the EIS does not enforce this field as
many states do not have this information available. By including these data in SIP inventory
development, states can help to further improve the data available to them for decision making as well
as for consideration by the EPA.
Control measure: This field is a unique code for the type of control device or operational measure (e.g.,
wet scrubber, flaring, process change, and ban) used to reduce emissions (see 40 CFR 51.50). When
assessing whether an emissions source can be further controlled to contribute emissions reductions
needed for attainment, it is imperative to know whether there is a control measure included on a unit or
process. The type of measure is also needed to understand whether other controls can be added. The
associated data elements such as control efficiency, capture efficiency, rule effectiveness, and rule
penetration are all useful as well for the same reasons. These latter control parameters are also needed
when calculating potential controlled emissions.
Alternative facility names or identifiers: This field can be very useful in matching the facility with other
databases, such as the Toxics Release Inventory (TRI), the Federal Registry System, or the National
Electric Energy Data System (NEEDS) database. If air agencies change their agency IDs (facility, unit,
process, and release point) from one year to the next, this table can help prevent creating duplicates in
the EIS and other databases.
EIS IDs: While the EIS Facility ID, Unit ID, Process ID, and Release Point ID (see Section 2.5.15) are not
required by the AERR or for SIP inventories, it is the standard by which the EPA compiles and reports all
point source inventories. The EPA reports and modeling all use these EIS IDs, and so to the extent air
22
-------
agency databases can use these IDs in their data systems, it helps facilitate matching across the many
purposes for inventories.
Release Point Geographic Coordinates: As described in Section 2.5.17, the EIS and inventories generally
can have location coordinates (i.e., latitude/longitude) for a facility or at individual release point
locations.
2.5.6 Throughput and Activity
Throughput means a measurable factor or parameter that relates directly or indirectly to the emissions
of an air pollution source during the period for which emissions are reported (40 CFR 51.50). Throughput
is also called activity. Depending on the source category, activity information may refer to the amount of
fuel combusted, raw material processed, product manufactured, or material handled or processed. It
may also refer to population, employment, miles travelled, or number of existing units. Activity
throughput is typically the value that is multiplied against an emission factor to generate an emissions
estimate. Consequently, it is important to coordinate activity used for a given emissions source with the
available emission factors for that source.9
There are some special types of activities always used for certain categories. Examples for mobile
sources include:
Vehicle Miles Traveled (VMT): The number of miles travelled within a year or as an average daily value
for all vehicles within a group (e.g., vehicle type, fuel type, and county). See also MOVES Technical
Guidance (http://www.epa.gOv/otaq/models/moves/index.htm#sip).
Vehicle population: The number of vehicles of a certain type and fuel usage within a group (such as a
county).
Equipment population: The number of nonroad equipment units of a certain type and fuel usage within
a group (such as a county).
2.5.7 Emission Factors
An emission factor is a representative value that attempts to relate the quantity of a pollutant released
to the atmosphere with an activity associated with the release of that pollutant. These factors are
usually expressed as the weight of pollutant divided by a unit weight, volume, distance, or duration of
the activity emitting the pollutant (e.g., kilograms of particulate emitted per megagram of coal burned).
Such factors facilitate estimation of emissions from various sources of air pollution. In most cases, these
factors are simply averages of all available data of acceptable quality, and are generally assumed to be
representative of long-term averages for all facilities in the source category (i.e., a population average).
Emission factors have long been the fundamental key to developing emissions inventories for air quality
9 Some air agencies have raised the concern that throughput at industrial facilities is considered confidential
business information (CBI), however, this field is required by the AERR. Since this field is "emission data"
required by the AERR, it cannot be protected as CBI under CAA section 114(c).
23
-------
NAAQS implementation. More information on emission factors is available from the Clearinghouse for
Inventories and Emissions Factors (https://www.epa.gov/chief/) on the EPA website.
The general equation for emissions estimation is:
E = Ax EF X (l-ER/100)
where:
E = emissions,
A = activity total,
EF= emission factor, and
ER = overall emission reduction efficiency percent.
The ER term is the combination of the relevant percentages related to emissions controls and rules that
reduce emissions, as listed in Section 2.5.18.
Emission factors are not limited to factors that are only represent broad national industry averages
published by the EPA. Emission factors can also be stack-, process-, unit-, or facility-specific, depending
on the basis of the source test information. Whether for a single facility or a group of facilities of the
same type, it is still considered an emission factor for the purposes of this guidance.
2,5,8 Point Sources
As defined by the AERR in 40 CFR 51.50, point sources are large, stationary (non-mobile), identifiable
sources of emissions that release pollutants into the atmosphere. A point source is a facility that is a
major source under 40 CFR part 70 for one or more of the pollutants for which reporting is required by
40 CFR 51.15 (a)(1). This does not include the emissions of hazardous air pollutants, which are not
considered in determining whether a source is a point source for ozone, PM2.5, and regional haze
emissions inventory development and reporting. These point sources can be associated with a single
point or group of points in space. Examples of important point source emissions categories include
power plants, industrial boilers, petroleum refineries, cement plants, and other industrial plants.
According to the ozone and PM2.5 Implementation Rules, the AERR governs the emissions level above
which sources are inventoried as point sources. While the AERR also requires data from state agencies
for use in the NEI, the AERR emissions point source thresholds have been coordinated with the various
SIP implementation rules and definitions of major sources and, therefore, they apply to all inventories
provided to the EPA. The AERR bases the emissions level on a potential to emit emissions size threshold.
Table 7 provides the potential to emit point source reporting thresholds included in the AERR. For
evaluating the potential to emit threshold, it is appropriate to use potential controlled emissions for
controlled sources. As shown in Table 7, special requirements apply to NAA sources, where even lower
thresholds may apply depending upon the severity of the NAA classification.
Air agencies may include point sources at even lower levels in their NEI or SIP planning and modeling
inventories. The AERR thresholds simply provide a minimum requirement. Depending on the nature of
the nonattainment problem and the need to characterize smaller sources as point sources, a state may
24
-------
want to consider and the EPA Regional offices may encourage lower thresholds so air agencies can
produce effective SIPs.
Type A source means a large point source with potential annual emissions of any CAP or CAP precursor
greater than or equal to the corresponding emission threshold listed in the AERR. These sources must be
reported by states every year and have higher reporting thresholds. The thresholds are not repeated in
the table below because they are not relevant for the definition of point sources for SIP inventory
purposes, which follow the Type B thresholds for triennial reporting as shown below. See Appendix A of
40 CFR part 51. Subpart A for the Type A source thresholds.
Type B source means a point source smaller than a Type A source with potential annual emissions of any
CAP or CAP precursor greater than or equal to the relevant Type B emission threshold listed in the AERR
and in this section. Type B point sources are required by the AERR to be included in a state or local
inventory submission for point sources in the triennial years and are applicable to SIP planning and
modeling inventories as well.
Table 7: Point source triennial reporting thresholds (potential to emit) for criteria
pollutants in the Air Emissions Reporting Rule
Pollutant
Triennial NEI thresholds: potential to emit (tons/yr)
Type B sources (default)
NAA sources
1
SO 2
> 100
> 100
2
SO 2
PM2.5 (Serious) > 70
3
VOCor NOx....
> 100
> 100
4
VOCor NOx....
03 (Moderate within OTR) > 50+
5
VOCor NOx....
03 (Serious) > 50
6
VOCor NOx....
03(Severe)> 25
7
VOCor NOx....
03 (Extreme) > 10
8
VOCor NOx....
PM2.5 (Serious) > 70
9
VOC
03 (within OTR) > 50+
10
CO
> 1000
>1000+
11
CO
CO (all areas) > 100
12
Pb
>0.5*
>0.5
13
PMio
> 100
> 100
14
PM10 (Serious) > 70
15
PM2.5
> 100
> 100
16
PM25
PM2.5 (Serious) > 70
17
NH3
> 100
> 100
18
nh3
PM2.5 (Serious) > 70
+ This threshold for ozone nonattainment areas is incorrectly listed in the AERR as 100 tpy.
This is not consistent with other regulations. The EPA is working to make this correction in
the AERR.
* The lead threshold is in actual emissions rather than potential to emit to be consistent with
the lead monitoring rule requirements.
25
-------
2.5.9 Nonpoint Sources
Nonpoint sources are also called "area sources".10 These sources collectively represent individual
sources of emissions that have not been inventoried as specific point or mobile sources. These individual
sources treated collectively as nonpoint sources are typically too small, numerous, or difficult to
inventory using the methods for the other classes of sources. Examples of key nonpoint sources include
residential combustion, institutional and commercial boilers, agricultural burning, industrial and non-
industrial solvents and graphic arts, degreasing, most agricultural sources, consumer products,
architectural coatings, and sources of dust including roads and construction.
2.5.10 On-road Mobile Sources
On-road mobile sources are also called "highway mobile sources." These sources are the motor vehicles
(e.g., automobiles, buses, trucks) traveling on local and highway roads. A motor vehicle is any self-
propelled vehicle used to carry people or property on a street or highway (40 CFR 51.50). On-road
mobile sources should be estimated by the latest recommended on-road mobile source models.
Currently, that means the MOVES model for all states but California, and the EMFAC11 model for
California.
2.5.11 Nonroad Mobile Sources
Nonroad mobile sources are also called "off-highway" mobile sources. These are defined as a nonroad
engine or nonroad vehicle. As per 40 CFR 51.50, a nonroad engine is an internal combustion engine
(including the fuel system) that is not used in a [on-road] motor vehicle or a vehicle used solely for
competition, or that is not affected by sections 111 or 202 of the CAA. Also defined by 40 CFR 51.50, a
nonroad vehicle (rather than engine) is a vehicle that is run by a nonroad engine and that is not a [on-
road] motor vehicle or a vehicle used solely for competition. Examples of nonroad mobile sources
include aircraft, airport ground support equipment, locomotives, CMVs, construction equipment
powered by an internal combustion engine, agricultural equipment powered by an internal combustion
engine, and lawn and garden engines and equipment.
There are a variety of tools for estimating emissions from nonroad mobile sources. The EPA develops
and maintains approaches for aircraft, locomotives, and CMVs, which are documented as part of the NEI
program (for the latest approaches, refer to the latest NEI documentation.
Starting with the MOVES2014a version of MOVES, the MOVES model should be used for nonroad source
emissions. MOVES2014a includes features not previously available in MOVES that simplify processing of
emissions output and includes updated fuel input files that result in small changes in emission results. In
addition, the prior tools available for nonroad emissions estimations (NONROAD2008 and NMIM2008)
may no longer work with current operating systems, and the EPA cannot continue to provide technical
support for these models. Therefore, the EPA recommends that for modeling nonroad emissions, latest
version of MOVES be used for all new SIP development, although state and local agencies that have
10 This data category type is different from the air toxics definitional of "area sources" as defined by CAA section
112(a)(2).
11 The EMFAC model name is not an acronym. The documentation and website for EMFAC do not include a further
explanation of the name.
26
-------
already completed significant work with MOVES2014, NONROAD2008, or NMIM2008 can continue to do
so to allow for timely submission of the SIP. As a general matter, air agencies should review and possibly
update and customize the MOVES model inputs to give better emissions estimates for their state.
Mobile source inventory modelers should use the latest version of MOVES as posted at
https://www.epa.gov/moves. They should also subscribe to the MOVES email list
https://www.epa.gov/moves/forms/epa-mobilenews-listserv to be notified of any bug fixes or updates
to the MOVES model.
Unlike models for onroad emissions, the EPA does not specifically "approve" nonroad models. However,
use of alternative models can be justified as part of a SIP submission. For example, California has
previously developed and used a state-specific approach for estimating its nonroad source emissions for
SIP purposes, called OFFROAD2007
(http://www.arb.ca.gov/msei/offroad/downloads/models/offroad2007 1215 exe.zip). This model has
now been replaced by category-specific methods for many categories (see
http://www.arb.ca.gOv/msei/categories.htm#offroad motor vehicles). As described by the website, the
OFFROAD2007 model is still the default approach for categories not listed with newer methods on the
website. Any use of such an alternative approach is subject to review by the EPA as part of the review of
the SIP, and thus states are encouraged to coordinate with Regional offices on any use of alternative
models.
2.5.12 Events
Starting with the 2008 NEI, the EIS introduced a new data category called "Events," which are discrete
and large emissions sources that individually occur over a short period of time, but which can
collectively result in a significant amount of emissions. Example events captured by this data category
include naturally occurring wildfire and a managed prescribed burns. In EIS, these data are stored as
day-specific events that can span many days for a given fire/event. More information on events data
and its unique nature is available in Section 4.8.7.
2.5.13 Biogenic Sources
For the purposes of creating emissions inventories, biogenic emissions come from natural sources. They
need to be accounted for in photochemical grid models, as many types of biogenic emissions are
widespread and ubiquitous contributors to atmospheric chemistry leading to ozone and PM formation.
In the NEI, only the emissions from vegetation and soils are included, but other relevant sources include
volcanic emissions, lightning, and sea salt spray.
Biogenic emissions from vegetation and soils are computed using models that utilize spatial information
on vegetation and land use. The models also use environmental conditions of temperature and solar
radiation to apply emission factors. The model inputs are typically horizontally allocated (gridded), and
the outputs are gridded biogenic emissions, which can then be summed for use in planning inventories,
and speciated for use as input to photochemical grid models. As described in Section 4.8.6, a variety of
models are available to estimate the emissions from vegetation and soils.
27
-------
2,5,14 Inventory Codes
The list provided below is a sample of the codes that can be used in building inventories; it is not a
comprehensive list of codes that should be used in building inventories. The EIS is the repository for the
most updated valid codes used for building inventories.
Source classification codes: An SCC means a process-level code that describes the equipment, operation,
or practice that is emitting pollutants (40 CFR 51.50).12 These codes are 8 digits for point sources and 10
digits for nonpoint and mobile sources. The codes have multi-level descriptions, which are included in
the EIS codes table. The EPA provides the most current list of codes within EIS, in an SCC search page (at
https://ofmpub.epa.gov/sccsearch/), and web service (available at
https://ofmpub.epa.gov/sccwebservices/). These codes are also used in the emission factors online
database WebFire (https://www.epa.gov/electronic-reporting-air-emissions/webfire).
North American Industry Classification System (NAICS) codes: The NAICS codes are U.S. Department of
Commerce's codes for businesses by products or services and have replaced Standard Industrial
Classification (SIC) codes. They are assigned to facilities in inventories to identify the primary industrial
activity of the facility. They can be searched at: https://www.census.gov/eos/www/naics/.
Federal Information Processing Standards (FIPS) state and county codes: Federal Information Processing
Standards (FIPS) codes means the system of unique identifiers maintained by the National Institute of
Standards and Technology, which are used to identify counties and county equivalents (e.g., parishes,
boroughs) of the U.S. and its territories (e.g., American Samoa, Guam, Northern Mariana Islands, Puerto
Rico, and US Virgin Islands). In the EIS Codes Tables, additional codes have been added to support
offshore regions and sources, such as asphalt plants, that do not remain in a single county during the
entire year. More information can be found at: https://www.census.gov/geo/reference/codes/cou.html.
Tribal codes: A three-digit code that represents the American Indian Tribe or Alaskan Native entity. The
EIS Table also identifies which EPA region is associated with the tribe. These codes are used instead of
FIPS state and county codes when reporting emissions from tribal areas.
Pollutant codes: Pollutant code means a unique code for each reported pollutant assigned by the
reporting format specified by the EPA for each inventory year (40 CFR 51.50). All of the criteria
pollutants, except for lead, have codes that describe them: CO, NOX, VOC, S02, NH3, PM10-PRI, PM10-
FIL, PM25-PRI, PM25-FIL, and PM-CON. Lead and the hazardous air pollutants (HAPs) use numerical
codes13 to represent them, for example elemental lead is 7439921 and elemental mercury is 7439976.
Control measure codes: Control measure means the name of the type of control device (e.g., wet
scrubber, flare, baghouse) or practice (e.g., product substitution). The EIS Code Tables use a three-digit
12 Several air agencies have raised the concern that SCC at industrial facilities is considered CBI. However, this field is "emission
data" required by the AERR and thus cannot be provided as CBI under CAA section 114(c).
13 These numerical codes are the unique numerical identifiers assigned and maintained by the Chemical Abstracts Service (CAS)
and are often referred to as CAS numbers or CAS registry numbers. The CAS numbers often appear with dashes embedded in
the string of numbers, but these are dropped for most inventory reporting purposes. See:
https://www.cas.org/content/chemical-substances/faqs
28
-------
numeric control measure code to represent each of the different types of controls, and identifies each
as a device or a practice. The control device type codes were updated with additional codes in 2013.
These codes are required when control devices are present because they are critical to understanding
the nature of the emissions source and the possibility of additional emissions reductions to achieve
attainment.
Emission calculation method: This field is the code describing how the emissions for a pollutant were
calculated (e.g., by stack test, continuous emissions monitor, EPA emission factor, etc.) (40 CFR 51.50).
This is a required field for inventories because it allows the EPA to get some sense of the quality of the
emissions estimate. Comment fields can also be used to further explain the origin of the emissions
estimate.
Reporting period type: This field is the code describing the time period covered by the emissions
reported (i.e., annual, 5-month ozone season (previously used for NOx SIP call), ozone season day, or
average season day) (40 CFR 51.50). In concert with the release of the final PM SIP Requirements Rule, a
new code ("ASD") has been updated in EIS to support submissions to EIS of average season day
emissions used for SIPs addressing the 24-hour PM2.5 standard. This field is also called the emission type
code in some references.
2,5,15 Point Source Facility Details
See also Section 2.5.5, which points to the spreadsheets available that list all required and optional
emissions inventory data elements.
Facility Inventory means the data tables of all facility information included in the EIS, but could also be
used to refer to similar data tables for any inventory storage system. The facility inventory is the
complete list of facilities, units, processes, and stacks associated with all point sources. It is designed to
be the latest repository of facility information and needs updating only when changes occur to facilities.
The facility inventory concept was intended to allow facility data to be constant except where real
changes occur, allowing easier building and submitting of inventories and encouraging the use of
consistent facility IDs, unit ID, process ID, and stack IDs from one inventory year to the next.
Physical address means the location address (street address or other physical location description),
locality name, state, and a postal zip code of the facility. This is the physical location where the
emissions occur, not the corporate headquarters or a mailing address (40 CFR 51.50).
Site name (also called Facility Site Name) means the name of the facility emitting emissions that are
included in the inventory (40 CFR 51.50).
EIS ID means any facility, unit, process, or release point (stack) ID generated by the EIS and intended to
remain constant for the life of that facility, unit, process, or stack across inventory years. The proper
names of the EIS IDs in the EIS system are EIS Facility Site Identifier, EIS Emission Unit Identifier, EIS
Emission Process Identifier, and EIS Release Point Identifier.
29
-------
Agency ID means any facility, unit, process, or release point (stack) ID generated by an agency reporting
emissions to the EPA. These IDs are used in the air agency data systems and may not match the EIS IDs
described above. The EPA tracks both the EIS IDs and agency IDs within the EIS. Using consistent agency
IDs from year to year is highly encouraged, as it allows the EPA to more readily see which facilities are
being reported for the AERR requirements. Using inconsistent agency IDs can result in duplicative
facilities, units, processes, and release points within the EPA data systems.
Facility ID code (also called "Facility site identifier)" means the unique codes for a plant or facility
treated as a point source, containing one or more pollutant-emitting units (40 CFR 51.50). The EPA's
reporting format for a given inventory year may require several facility ID codes to ensure proper
matching between databases (e.g., the state's own current and most recent facility ID codes, the EPA-
assigned facility ID codes, and the ORIS (Department of Energy) ID code if applicable).
Unit ID code means a unique code for the unit that generates emissions, typically a physical piece of
equipment or a closely related set of such equipment (40 CFR 51.50). An example of a closely related set
of equipment is a boiler and its control equipment. The EPA's reporting format for a given inventory year
may require multiple unit ID codes to ensure proper matching between databases (e.g., the state's own
current and most recent unit ID codes, the EPA-assigned unit ID codes if any, and the ORIS (Department
of Energy) ID code if applicable).
Process ID code is also called the "emission process identifier" and it means a unique code for the
process generating the emissions, typically a description of a process (40 CFR 51.50).
Release Point ID code means a unique code for the point where emissions from one or more processes
release into the atmosphere (40 CFR 51.50). Traditionally called a Stack ID, but this was changed with EIS
to acknowledge that many emissions releases do not come from stacks, such as fugitive emissions.
Release point type code means the code for physical configuration of the release point (40 CFR 51.50).
The physical configurations include fugitive, vertical, horizontal, goose neck, vertical with rain cap, and
downward-facing vent. The first two types are the most common. The release point type is particularly
important for dispersion modeling needed for permitting, risk modeling, and other purposes.
Release Point Diameter Measure (or Stack diameter) means the inner physical diameter of a stack (40
CFR 51.50). This should be measured at the exit release point of the emissions, and the expected
reported units are feet.
Release Point Height Measure (or Stack height) means physical height of a stack above the surrounding
terrain (40 CFR 51.50). Expected reported units are feet.
Release Point Exit Gas Flow Rate Measure (or Exit gas flow rate) means the numeric value of the flow
rate of a stack gas (40 CFR 51.50). Various units are permitted by the EIS.
Release Point Exit Gas Temperature Measure (or Exit gas temperature) means the numeric value of the
temperature of an exit gas stream in degrees Fahrenheit (40 CFR 51.50).
30
-------
Release Point Exit Gas Velocity Measure (or Exit gas velocity) means the numeric value of the velocity of
an exit gas stream (40 CFR 51.50). Various units are permitted by the EIS.
2.5.16 Alternative IDs
In addition to the primary facility, unit, process, and release point IDs, alternative IDs can be very useful
for matching information in a given inventory to other databases. In the case of the EIS, it stores not
only the EIS unique IDs, but also the air agency IDs. Other IDs are stored as well, including the following:
EIA Plant Code (or ORIS Plant Code). This plant ID is defined by the U.S. Department of Energy (DOE)
Energy Information Administration (EIA) and used in its EIA-767 form. It has previously been defined as
the Office of Regulatory Information System (ORIS) code, though that term is no long used by DOE. The
EPA's Office of Atmospheric Programs' (OAP) Clean Air Markets Division (CAMD) uses whatever code
was originally assigned to the facility as the tracking code for compiling data for its emissions markets in
support of the acid rain program and NOx trading, in particular the CEMS data.
CAMD Unit ID. In addition to the EIA facility IDs, CAMD uses unit IDs for storing the emissions markets
data as well. The EIS stores the unit codes used by CAMD as an alternative ID called "EPACAMD," and
the values are usually based on the CAMD hourly CEMS data reports. These have historically also been
referred to as ORIS Boiler or ORIS Unit IDs, however, those references are no longer valid.
IPM/NEEDS ID. The EPA Integrated Planning Model (IPM) is used to forecast future-year electricity
demand and emissions. Underlying this model is the National Electric Energy Data System (NEEDS)
database that contains the generation unit records used to construct the "model" plants that represent
existing and planned/committed units in EPA modeling applications of IPM. This database mostly uses
IDs that are often the same as the EIA Plant Codes and the CAMD Unit IDs, but this database has more
units than are included in the hourly CEMS data. In the EIS, the EPA stores these alternative IDs as
"EPAIPM" IDs.
Toxics Release Inventory (TRI) ID is used by the EPA's Office of Environmental Information to build the
TR|from facility-submitted facility-total emissions to air, water, and land. This database includes NH3,
which is why it can be relevant to the NAAQS for PM2.5 implementation. In the EIS, the EPA stores these
alternative IDs as "EPATRI" IDs.
Federal Registry Service (FRS) ID is used by the EPA's Office of Environmental Information to link many
EPA databases with facility information (see https://www.epa.gov/enviro/facilitv-registry-service-frs).
2.5.17 Location Definitions
Geographic locations are critical in point source inventories for identifying the location of the emissions.
In the EIS, locations can be provided for both the facility center and for individual release points. Fugitive
release points have additional parameters that can be specified, such as the orientation of the release,
and these parameters are listed in the spreadsheets available that list all required and optional
emissions inventory data elements, referenced in Section 2.5.5.
31
-------
Latitude Measure (or just latitude) is the measure of the angular distance from the center of the earth to
a point north or south of the equator. Reported as decimal degrees. The geographic coordinate system
will have to be specified for example: North American Datum of 1927 (NAD 1927), North American
Datum of 1983 (NAD 1983), World Geodetic System of 1984 (WGS 1984). See the Horizontal Reference
Datum code below.
Longitude Measure (or just longitude) is the measure of the angular distance from the center of the
earth to a point on a meridian east or west of the prime meridian. Reported as decimal degrees. The
GCS should be specified as with latitude.
Horizontal Accuracy Measure is the measure of accuracy (in meters) of the latitude/longitude
coordinates.
In addition, refer to https://www.epa.gov/geospatial/latitudelongitude-data-standard for more
information on the proper usage of the following code set.
Horizontal Collection Method Code is the method used to determine the latitude/longitude coordinates
for a point on the earth. For example, 001 is address matching-house number, and 004 is address
matching nearest intersection.
Horizontal Reference Datum Code is part of the geographic coordinate system and represents the
position of the spheroid with respect to the center of the earth (for a geocentric datum like WGS 1984)
or with respect to a specific point on the surface of the earth (for local datums such as NAD 1927 and
NAD 1983). Options are: 001 (NAD 1927), 002 (NAD of 1983), and 003 (WGS 1984).
Geographic Reference Point Code represents the place for which geographic coordinates were
established. Code value should be 106 (e.g., point where substance is released).
2,5,18 Control Data Elements
Control efficiency means the efficiency by which a control device or measure reduces emissions for a
particular pollutant. For emissions reporting in accordance with the AERR, this data element applies for
both point and nonpoint sources and is called the "Percent Control Measures Reduction Efficiency." For
point sources, the data element is included in the "Control Pollutant Table" of the facility staging tables.
For nonpoint sources, the data element is included in the "Control Pollutant Table" of the nonpoint
staging tables. In the EIS, it is a percent and should be reported as a value greater than or equal to 0.1
and less than or equal to 99.9 (the EIS allows for one decimal place only).
Control Capture efficiency means the percentage of an exhaust gas stream actually collected for routing
to a set of control devices (40 CFR 51.50). It is sometimes combined within the value for the control
efficiency, but unlike control efficiency, it is not pollutant-specific. For emissions reporting in accordance
with the AERR, this data element is for point sources only, and it is called "Percent Control Capture
Efficiency" and is included in the "Control Approach Table" of the facility staging tables. It is a percent
and should be reported as a value greater than or equal to 0.1 and less than or equal to 99.9 (the EIS
allows for one decimal place only).
32
-------
Control Effectiveness means the percentage of time or activity throughput that a control approach is
operating as designed, including the capture and reduction devices. This percentage accounts for the
fact that controls typically are not 100 percent effective because of equipment downtime, upsets and
decreases in control efficiencies (40 CFR 51.50).
Control effectiveness is an estimate of the portion of the reporting period's activity for which the overall
control system or measure (including both capture and control measures) were operating as designed
(regardless of whether the control measure is due to rule or voluntary). It applies to both point and
nonpoint sources and is not pollutant-specific. For point source emissions reporting in accordance with
the AERR, this data element is called "Percent Control Approach Effectiveness" and is included in the
"Control Approach Table" of the facility staging tables. For nonpoint source emissions reporting in
accordance with the AERR, this data element is also called "Percent Control Approach Effectiveness" and
is included in the "Control Approach Table" of the nonpoint staging tables. In the EIS, it is a percent and
should be reported as a value greater than or equal to 0.1 and less than or equal to 99.9 (the EIS allows
for one decimal place only). For nonpoint, see also "rule effectiveness," below.
Rule penetration means the percentage of a nonpoint source category activity that is covered by the
reported control measures (40 CFR 51.50). For emissions reporting in accordance with the AERR, this
data element is called "Percent Control Approach Penetration" and is included in the "Control Approach
Table" of the nonpoint staging tables. In the EIS, it is a percent and should be reported as a value greater
than or equal to 0.1 and less than or equal to 99.9 (the EIS allows for one decimal place only).
Rule effectiveness means a rating of how well a regulatory program achieves all possible emissions
reductions. It is analogous to the "control effectiveness" above and, for emissions reporting in
accordance with the AERR, this information is reported using the EIS Control Effectiveness field for
nonpoint (see above). Like control effectiveness, it should reflect the assumption that controls typically
are not 100 percent effective because of equipment downtime, upsets, decreases in control efficiencies,
noncompliance, and other deficiencies in emission control. Rule effectiveness adjusts the control
efficiency from what could be realized under ideal conditions to what is actually emitted in practice due
to less than ideal conditions.
2,5,19 EIS Sectors, Tiers, and Facility Type
Because emissions inventories are large and detailed, most reports about emissions inventories include
summaries of the contents of those emissions. Some summaries use the data categories described
above (point, nonpoint, on-road, nonroad, events, and biogenic). For other summaries, that level of
detail is not enough. The EIS SCC tables provide three other mechanisms for summaries, all of which
depend on SCC.
First, the SCCs have a four-part description, and the first three levels of this description could be used as
a summary category. Second, the SCCs are assigned to "Tiers" using the fields "Tier 1 Description," "Tier
2 Description," and "Tier 3 Description." These Tiers have been used traditionally as part of many EPA
reports and trends for criteria emissions. While wildfires are included in the "Miscellaneous" Tier 1,
wildfires are sometimes separately broken out from Miscellaneous because of their natural origin and
33
-------
because they can be a large contributor for some pollutants. Finally, the EIS Sectors can be used to
summarize emissions (this is the field "Sector" in the EIS SCC code table). The EIS Sector was added with
the creation of the EIS and allocates emissions to about 60 categories with multi-pollutant inventories in
mind. Whereas the Tiers are more useful for criteria pollutants, the EIS Sectors work better for the
combination of HAPs and CAPs. The EIS Sectors can also be further aggregated to provide custom
summaries; one example of this is presented in the report "2008 National Emissions Inventory: Review,
Analysis and Highlights" (EPA-454/R-13-005, May 2013).
2,5,20 Other Terms
High electricity demand day (HEDD) means an ozone season day when demand for electricity is high. A
study of the Northeast Corridor14 air quality has indicated that NOx emissions from EGUs during HEDD
periods may be significantly greater than emissions that occur on an average ozone season day.
Distributed generation means the use of small power generators sited at or near the point of energy
consumption. This usually occurs during periods of peak energy demand or emergency loss of the
normal EGU generating capacity. Distributed generation sources of electricity can include rooftop solar
or wind sources, biomass combustion, as well as backup fossil fueled generators or other engines. The
combustion sources of distributed generation can cause emissions that could be of interest for SIP
planning purposes.
RFP projected emissions means the estimated emissions for direct PM2.5 and PM2.5 plan precursors by
sector for the years in which quantitative milestones are due for a nonattainment area.
14 "High Electric Demand Day and Air Quality in the Northeast." White Paper Prepared by the Northeast States for
Coordinated Air Use Management. June 5, 2006. Available at: http://www.nescaum.org/documents/high-
electric-demand-day-and-air-quality-in-the-northeast/final-white-paper-hi-electric-demand-day-06052006.pdf.
34
-------
3 SIP Inventory Requirements and Recommendations
As described in Section 2, there are many required and potentially required inventory efforts needed as
part of a SIP development effort. This section covers the fundamental attributes of planning inventories
and their preparation. The topics covered include the timing of the various inventory components
(Section 3.1), the inventory preparation plan (Section 3.2), the base year inventories (Section 3.3), the
ROP/RFP baseline NAA inventory (Section 3.4), the ozone periodic inventory (Section 3.5), the ozone
emissions statement requirement for SIPs (Section 3.6), maintenance inventories (Section 3.7), the
projected inventories (Section 3.8), public hearings (Section 3.9), and submitting the inventory to the
EPA for approval (Section 3.10). The details of inventory preparation for base year inventories and
projected inventories are left for Sections 4 and 5, respectively.
ventory Component Timing
Table 8 provides the timing requirements and recommendations (for optional items) for the inventory
components of SIPs. Where this table is inconsistent with the ozone, PM2.5, or regional haze regulatory
text that is applicable at the time a SIP is prepared and reviewed, the regulations have precedence over
the information provided here.
Table 8: Timing of emissions inventory components
Inventory Component
Required:
Timing
Inventory Preparation Plan
Prior to preparation of emissions inventories covered by
the SIP.
Ozone
Within 2 years of the effective date of designations
Base year NAA inventory
PM2.5
Within 18 months from the date of a nonattainment
designation.
ROP/RFP baseline NAA
inventory
Ozone
Moderate & above: Within 3 years of the effective date
of designations.
Serious & above: Additional plan within 4 years.
PM2.5
Within 18 months from the date of a nonattainment
designation.
Periodic NAA inventory Ozone
The end of each 3-year period after SIP submittal until
the NAA is redesignated to attainment.
Source Emission Statements
requirement for SIPs
Ozone
Within 2 years of the effective date of designations.
Base case statewide inventory
Ozone and
PM2.5+
With modeled attainment demonstration (if required,
for convenience, can be included with the inventory SIP).
Attainment projected
inventory for the NAA
Ozone:
summary
only
For Ozone with attainment demonstration or ROP/RFP
plan. On-road portion can become the motor vehicle
emissions budget.
PM2.5
Within 18 months from the date of a nonattainment
designation.
Projected statewide inventory
(summaries)
Ozone and
PM2.5+
With modeled attainment demonstration (for
convenience, can be included with the inventory SIP).
35
-------
Inventory Component
Required:
Timing
Baseline statewide inventory
(summaries)
Haze
Previously completed for initial Regional Haze SIPs.
Subsequent progress reports and SIP revisions may refer
to this inventory.
Recent statewide inventory
(summaries)
Haze
Needed for the SIPs due July 31, 2021, July 31, 2028 and
every 10 years thereafter. Needed for progress reports
due January 31, 2025, July 31, 2033 and every 10 years
thereafter.
Projected statewide inventory
for 2018 (summaries)
Haze
Previously completed to support first Regional Haze SIP
(due by December 17, 2007)
Projected statewide inventory
for 2028, 2038, etc.
(summaries)
Haze
To support Regional Haze SIPs for second and
subsequent implementation periods, due every 10 years
beginning on July 31, 2021.
Public hearing
Ozone and
PM2.5
Prior to submittal of inventory for approval by the EPA.
Inventory documentation
All
With associated SIP.
Maintenance NAA inventory
Ozone and
PM2.5
After attainment, with a redesignation SIP.
+ Only required in association with modeled attainment demonstration. If the inventory needed for the modeled attainment
demonstration is not the entire state, then a "statewide" inventory would not need to be provided, but rather the emissions
from the part of the state included in the modeling domain could be provided instead.
3,2 Inventory Preparation Plan.
Air agencies can use IPPs as a planning tool to guide inventory preparation and to help ensure that
emission estimates are of a high quality and consistent with CAA requirements. In accordance with 40
CFR 51.5(b), states are urged to use applicable, state-of-the art techniques for estimating emissions. In
some cases, the EPA Regional offices will have a preset list of expectations for air agencies to include in
their SIP and will provide this information to the agency at the start of the SIP planning process. In these
cases, an IPP may not be needed, especially for air agencies with prior SIP development experience for a
particular NAAQS.
The IPPs are an opportunity for state and local agencies to tell their EPA Regional office how they plan to
compile their emissions inventories and allow the EPA to provide feedback prior to the agency spending
resources in preparing their inventory. They can also allow the EPA Regional offices to coordinate to
maintain consistency across attainment plans where that is advisable. The IPPs can be especially
valuable for agencies with new nonattainment problems, less experienced inventory staff or for
agencies otherwise needing additional feedback on inventory development.
If used, IPPs precede the development of inventories; therefore, the timing of IPPs is early in the SIP
planning process. The EPA recommends that air planning agencies coordinate with their Regional office
on the need for a plan, the timing of delivery of the plan, the response time back from the EPA on the
plan, and any other IPP approval process details. Agencies should recognize that time may also be
needed in their planning schedule to address the EPA comments on the plan, while the EPA Regional
offices will attempt to provide as timely feedback as possible.
36
-------
A high-quality IPP could be a strong foundation for the emissions inventories used in planning and/or
attainment modeling, which are themselves critical pieces of SIP development; therefore, the IPP should
include enough details for the EPA to properly assess what the agency plans to do. The content can
include a number of key elements, for which a checklist is provided in Table 9. The actual elements to
include in a specific IPP can depend on a number of factors, depending on the needs of the state or local
agency as well as the EPA for understanding the inventory plans.
Table 9: Checklist of elements to consider for including in Inventory Preparation Plans
IPP Element
Important for
a High Quality
IPP?
Inventory objectives - the purpose and scope of the inventory
Y
The inventory year(s) and rationale for selecting them
Y
The pollutants to be included, including the expected definition of VOC (for
inventories needing VOC)
Y
For PM2.5, the temporal basis expected to be used for the SIP emissions, which can
include annual, average season day, or both (see Section 4.5)
Y
A list of inventories to be included with the SIP, including the types (e.g., reports,
data summaries) and formats of information (i.e., electronic data files and
their expected formats, electronic media used for transfer), that the agency
intends to include in their SIP
Y
Plans for (or results of) identifying the most important source sectors to the
nonattainment problem
Y
The source categories to be included, the emissions level(s) for defining point
sources, and if available, the expected priority for these
Y
The plans for compiling the inventory, including any other known inventories and
how the agency plans to assess and improve upon what is already available
A summary of expected approach differences among different inventories that could
be the same (e.g., between a base year NAA inventory and a base year
modeling inventory)
For projected inventories:
The drivers for future-year emissions growth and reductions
Y+
The source categories to be forecast
Y+
The expected approaches for forecasting each source category
Y+
A plan for public review
Important sources of emissions or other inventory uncertainties
Quality assurance steps to be taken
Y
A schedule for inventory preparation, review, and submission to the EPA, including
any plans to submit pieces of the inventory separately and in accordance with
regulatory timeline requirements
Y
+ If a projected inventory is required.
In addition to Table 9, other inventory planning resources may be of use for agency planning purposes.
During the preparation of their IPPs, state and local agencies are referred to Volume I of the El IP
guidance, available at https://www.epa.gov/air-emissions-inventories/volume-l-introduction, which
discusses emissions inventory planning and development. While somewhat out of date, the El IP volumes
37
-------
still can help guide new inventory developers, especially with basic information about creating
emissions. Users of that document should defer to this guidance where newer information is provided,
such as updated versions of mobile models. For stationary point sources, a good resource is El IP Volume
II, Chapter 1, Section 3. For nonpoint sources, EIIP Volume III, Chapter 1, Section 2 gives nonpoint-
specific planning information. The mobile source volumes for EIIP are more out of date and the EPA
Office of Transportation and Air Quality (OTAQ) mobile source guidance is preferable (see Table 1).
A quality assurance plan should always be a part of the IPP. In cases where a more formal QAPP is
developed, the QAPP will satisfy the need for an IPP quality assurance plan. In some cases, the EPA
Regional offices will include a QAPP or other quality assurance requirement as part of their CAA section
105 grant to a state that typically includes funding for emissions inventory development. Each Regional
office manages its own quality assurance plan. Information about EPA quality assurance specifications
and tools as well as a map that links to each Regional office's quality assurance project are available at
https://www.epa.gov/qualitv.
For additional information on quality assurance planning, Chapter 2 of Volume VI, "Quality Assurance
Procedures and DARS software" (https://www.epa.gov/air-emissions-inventories/volume-6-qualitv-
assurance-procedures-and-dars-software) provides some general guidance for quality assurance and
some specific approaches to uncertainty evaluation. Other quality assurance resources include Section 0
of this guidance, and quality assurance steps described in the latest NEI Technical Support
Documentation available at the latest NEI documentation page link on the main NEI web page. Where
air agencies are relying on their NEI submission as a starting point for meeting any SIP inventory
requirements, a QAPP developed for that air agency's NEI submission may offer a starting point for the
SIP inventory QAPP.
3,3 Base Year Inventory
The purpose of the base year inventory is to provide a current and comprehensive data source for
emissions contributing to the air quality problem addressed by the SIP. In the case of ozone and PM2.5,
the inventory requirement is for the NAA, whereas for regional haze, the inventory requirement is for a
statewide inventory. As listed in Table 4 and Table 5, the ozone or PM2.5 SIPs can include statewide
inventories as part of the modeled attainment demonstration, and these can be included as separate
inventory submissions included with the NAA inventory.
The general SIP provisions in CAA section 172(c)(3) explains that base year inventories should include a
"comprehensive, accurate, current inventory of actual emissions from all sources" in the applicable area.
Details associated with preparing a base year inventory are provided in Section 4, including selection of
year(s), temporal extent, spatial extent, pollutants to include, approaches for developing emissions, data
codes, and quality assurance. Additionally, CAA section 182(a)(1) provides a specific base year inventory
requirement for ozone SIPs. PM2.5SIPS generally rely on the general SIP provisions in CAA 172(c)(3),
though some PM-specific elements of CAA section 189 impact the timing and other aspects of PM2.5 SIP
inventories. Other requirements for SIP inventories are found in the SIP implementation rules as listed in
Table 4.
38
-------
The base year inventory is the starting point from which the other SIP-related inventories are derived.
These other inventories should relate to the base year inventory with regard to the emissions sources
included, pollutants, temporal aspect, and other key aspects that allow the SIP to be a coherent and
connected plan across the various inventories. For planning purposes, the base year inventory provides
a way for decision makers to consider the sources of emissions that contribute relevant pollutants and
consider emissions reductions strategies needed for the SIP.
Section 4.1 covers the choice of inventory years for the base year inventory. The choice of the inventory
year for the planning is a critical one that significantly impacts the workload of inventory staff and
greatly depends on availability of data and timing of other inventory efforts required by the EPA.
Section 4.2 covers the pollutants to be included in the base year inventory. These pollutants differ
depending on the nature of the air quality problem for a given SIP. Where feasible, the EPA encourages
planning agencies to include precursors for all ongoing nonattainment problems in a state to promote
efficiency in preparing multiple base year inventories, and to allow for the consideration of emissions
reduction strategies that improve air quality across multiple air quality concerns and nonattainment
issues. In addition, states may wish to include emissions of HAPs, which can have co-benefit reductions
associated with strategies aimed at reducing ozone, PM2.5, and/or regional haze.
The spatial and temporal extent of the base year inventories are also an important inventory
characteristic. These topics are covered fully in Sections 4.4 and 4.5. Each of the implementation rules
for ozone, PM2.5, or regional haze has different considerations and requirements associated with the
planning inventories. For example, ozone NAA inventories must include ozone season day emissions, the
PM2.5 NAA inventories can be either annual or season day (depending on the nature of the PM2.5
nonattainment problem), and the regional haze inventories are only annual.
The Ozone Implementation Rule and PM2.5 SIP Requirements Rule require inventories for the NAA.
These inventories are called the base year inventory for the NAA. These inventories may need to be
updated through multiple iterations of a SIP when they are used as baseline inventories for assessing
progress for ROP/RFP. For regional haze, the inventory for a year in the 2000-2004 period is commonly
called the baseline inventory. For clarity here, we also call the baseline regional haze inventory the
baseline statewide inventory to distinguish it from a NAA inventory.
.< "tgress and Reasonable Further Progress Baseline NAA
Inventory
RFP is a concept included in the CAA under Part D, Title I to assure that states take steps to ensure that
the air quality in NAAs makes steady and incremental progress toward attaining air quality standards.
The ROP/RFP baseline NAA inventory is necessary for ozone NAAs designated as Moderate and above.
This baseline inventory is part of a plan or plans that meet the RFP requirements as described by the
CAA and the Ozone Implementation Rule, and summarized here. The PM2.5 SIP Requirements Rule also
includes requirements for RFP as described below. Though there are specific considerations related to
ROP and RFP, the anthropogenic portion of the base year inventory for the NAA also serves as the
base ROP/RFP baseline NAA inventory for both ozone and PM2.5 SIP development.
39
-------
3.4.1 Ozone ROP/RFP Requirements and Timing
CAA section 182(b)(1)(A) requires Moderate and above NAAs to submit a plan to ensure a 15 percent
VOC reduction (accounting for growth) by 6 years after the NAA designation, which is the Moderate area
maximum attainment date. The 15 percent reduction requirement is commonly called "rate of progress"
or ROP. According to the Ozone Implementation Rule, air agencies must submit this plan within 3 years
after the area's nonattainment designation. This timing aligns with the Moderate NAA deadline for
submitting attainment demonstration SIPs.
In addition, CAA section 182(c)(2)(B) requires each Serious and above ozone NAA to submit a plan to
achieve at least a further annual 3 percent VOC reduction. Additionally, CAA section 182(c)(2)(C) allows
for substitution of NOx for VOC emissions reductions in the plan under certain conditions. This
requirement is also called "annual RFP." As described by the Ozone Implementation Rule, the annual
RFP requirement applies for any remaining 3-year periods out to the attainment date beyond the first 6
years. According to the Ozone Implementation Rule, air agencies with NAAs classified as Serious and
higher must submit this plan within 4 years after the area's nonattainment designation. This timing
aligns with the Serious and higher areas' deadline for submitting attainment demonstration SIPs;
however, these NAAs still must submit the plan for ROP within 3 years.
States maintain the flexibility to submit plans early to provide more time for implementation of their SIP
control measures. The EPA recommends that states complete their ROP/RFP plans as expeditiously as
practicable after designation to provide as much time as possible for sources to implement the emission
reductions.
The associated terminology for the various plans and inventories can be confusing, and so the
terminology used in this guidance is described here. In this guidance, we use "ROP/RFP plan" to refer
generally to all plans (i.e., those addressing just the 15 percent ROP requirement for Moderate areas as
well as plans additionally addressing the annual RFP requirement). For Serious and above areas, since
the timing allows that two separate plans for ROP and annual RFP could be submitted (i.e., at 3 years
and 4 years after designations, respectively), it may also be necessary to refer to such plans separately.
Consequently, this guidance uses "ROP/RFP plan(s)" to allow for the idea that multiple plans could be
submitted. Lastly, in the case where the individual plans must be referenced, they will be called "ROP
plan" and "RFP plan."
3.4.2 Ozone ROP/RFP Emissions Guidance
The same ROP/RFP baseline NAA inventory is used for all ROP/RFP plan(s), regardless of whether they
address the ROP requirement (for Moderate areas) or both ROP and annual RFP (for Serious and above).
As implied by the name, these inventories include emissions data just for the NAA region and thus,
excludes other areas in a state. The purpose of the ROP/RFP baseline NAA inventory is to provide a
reference point to determine the emissions reductions that must be achieved for the 15 percent VOC (or
NOx) reduction and the subsequent 3 percent reductions for Serious areas and above out to the
attainment date. This requirement stems from CAA requirements.
40
-------
For ozone, both VOC and NOx should be included in the ROP/RFP baseline NAA inventory. Further, the
EPA expects that the year of the ROP/RFP baseline NAA inventory will be the same year used for the
base year inventory for the NAA. In most cases, the ROP/RFP baseline inventory should also have the
same emissions values as are in the base year inventory for the NAA. One key exception is that biogenic
emissions should not be included in the ROP/RFP baseline.
The EPA also expects that the spatial extent of the ROP/RFP baseline NAA inventory will also match the
base year NAA inventory so that they are both either "whole county" inventories or both consistently
"partial county" inventories. Section 4.4.1 provides more information on partial county inventories.
In the event that updated ROP/RFP plan(s) are needed after initial submittal to meet the requirements
of the Ozone Implementation Rule, the ROP/RFP baseline NAA inventory may also need to be updated.
This can occur with significant methods changes, such as updated mobile models. For example, after an
initial ROP/RFP plan has been submitted as part of a SIP, a new mobile model may become available that
alters the emissions estimates. It would be inconsistent to compare estimates from a new mobile model
at the end of a 6-year period to estimates using an old model at the beginning of that 6-year period, and
so the EPA could be unable to determine whether a 15 percent reduction (or subsequent 3 percent
reductions) can be (or has been) achieved by an ROP/RFP plan. Thus, an updated ROP/RFP baseline NAA
inventory may need to be created using the updated mobile model to appropriately meet the
requirements of the Ozone Implementation Rule.
Additional ROP/RFP emissions reductions guidance is provided in Section 3.8.2.
3.4.3 PM2.5 RFP Requirements and Timing
Section 172 of the CAA addresses the PM2.5 attainment plan provisions in general. In addition, Section
189(c) of the CAA requires that PM2.5 SIPs contain quantitative milestones, which are to be achieved
every 3 years until the area is redesignated attainment and which demonstrate reasonable further
progress. For Moderate areas, these quantitative milestone years would be no later than 4.5 and 7.5
years after designation of the area. For Serious areas, these quantitative milestone years are no later
than 7.5 and 10.5 years after designation of the area. Most (if not all) Serious areas will have first been
Moderate areas, and thus could also be subject to the quantitative milestone at 4.5 years as well.
In the PM2.5 SIP Requirements Rule, one required element of an RFP plan is RFP projected emissions by
sector and precursor pollutant for each applicable quantitative milestone year. In addition, the
emissions reductions must show that the implementation schedule (of reductions) achieves either (1)
generally linear progress toward the projected attainment date; or (2) stepwise progress toward the
projected attainment date. The rule also holds out the possibility that a state could develop an
approvable RFP plan even if emissions of one or more PM2.5 plan precursors were not decreasing;
however, additional analysis would need to be done (as described in the rule).
3.4.4 PM2.s RFP Emissions Guidance
For PM2.5 SIPs, all plan precursors must be included in the RFP emissions provided as part of the RFP
plan. Further, the EPA expects that the year of the RFP baseline emissions will be the same year used for
the base year inventory for the NAA.
41
-------
Unlike other emissions inventory requirements, the PM2.5 RFP requirement does not specify that
detailed projected emissions inventory data must be developed and submitted to the EPA. The
emissions requirement for most sectors is for "RFP projected emissions for direct PM2.5 and all PM2.5 plan
precursors for each applicable milestone year," which does not indicate a detailed emissions inventory.15
In addition, the PM2.5 SIP Requirements Rule preamble specifically indicates emissions provided by
sector and pollutant are expected. It would be sufficient to include a table that has rows for each sector
and columns for the baseline and projected emissions of all relevant pollutants.
States may be able to develop their RFP projected emissions in some cases with less effort than it would
take to do a detailed emissions inventory projection. One approach for the calculating sector/pollutant
emissions for the milestone years would be to rely on the projected attainment inventory for the NAA
and use linear interpolation from the baseline year to the milestone year(s). To take this approach,
states should explain why the linear assumption makes sense for a particular sector. For example,
residential wood combustion emissions are projected to the future in part through phase-in rate
assumptions for newer and lower emitting woodstoves. Using this sort of "% adoption per year"
approach would support using a linear rate of change assumption such that the whole residential wood
combustion sector could be interpolated between the baseline year and the projected attainment year.
trio die Inventories
The Ozone Implementation Rule requires a periodic inventory for the NAA, and the Regional Haze Rule
requires a "recent year" statewide inventory. The ozone periodic inventories should be consistent with
the base year inventory in the temporal extent (seasonal, annual, or both), spatial extent (statewide,
NAA counties, partial counties), and sources inventoried. In some cases, for ozone, it may also be
advisable to have the emissions inventory methods of the periodic inventory consistent with the base
year inventory, but this decision should be made in consultation with the relevant EPA Regional office.
Regional offices may consider additional relaxation of the consistency recommendation above in specific
instances. For regional haze, the "recent year" inventory must always be statewide. There is no
expectation that the recent year inventories for regional haze will use consistent methodologies as were
used for the baseline inventory for a year in the range of 2002-2004 submitted as part of the first
Regional Haze SIP revision. Rather, the most recent emission estimation methods would be expected to
be used.
3.5.1 Ozone Periodic Inventory
For ozone, the requirement in 182(a)(3)(A) is that the periodic inventory be submitted no later than
each 3-year period after the submission of the base year NAA inventory. This language provides the
flexibility to air agencies to submit sooner than 3 years, but not later than 3 years. If any other current
NAA inventory is being submitted to the EPA for any reason, including ROP/RFP plans, then states
should consider also using that data to meet the periodic inventory requirement. Also, if the air agency
15 For the purposes of motor vehicle emissions budgets for transportation conformity (as required by 40 CFR part
93), an inventory of onroad mobile source emissions in the NAA is required (see 40 CFR 51.1012(a)(2)).
42
-------
comes into attainment and receives a Clean Data Determination prior to the 3-year period, or if a
maintenance plan (with a maintenance inventory) is being submitted prior to or in sync with this 3-year
timetable, then these actions are very likely to make it unnecessary for the air agency to submit a
separate ozone periodic inventory.
Since the periodic inventory requirement in the CAA does not specify the degree of detail required for
the periodic inventory, the EPA has defined the necessary components through the AERR. Thus, whether
the inventory is submitted as a separate submission or through the AERR, it would need to be consistent
with requirements of the AERR in order to meet the statutory periodic inventory requirements.
The periodic inventory is a check-in with the air agency on how the emissions are trending, and had
previously been considered to not have "regulatory significance" or been treated as a SIP submission.
This is no longer EPA's recommended approach. In accordance with the CAA, periodic inventories are a
revision to the inventory portion of the ozone SIP and are thus subject to CAA section 110 and 40 CFR
51.102, 103, and Appendix V. For discussion of public hearings, please see Section 3.9 of this guidance
document.
To meet the periodic inventory requirement for ozone, the periodic inventories must include ozone
season day emissions values. Since this emissions type is allowed to be submitted to the EPA through EIS
for the AERR, it is possible that the AERR submission could meet the periodic inventory requirement if
the conditions listed in Table 10 apply. As with all SIP requirements, air agencies should coordinate with
the Regional office regarding what is appropriate for a particular NAA.
Table 10: Conditions for AERR submittal to meet ozone periodic inventory requirement
Condition
Comment
Public hearing requirement
A public hearing has been offered and either no party requested a
public hearing or a public hearing has been held. Evidence should
be provided that this requirement has been met.
Ozone season day
emissions must be
submitted for the NAA
counties
If ozone season day emissions cannot be extracted from the EIS (as
is the case in 2014), using an AERR submission to meet the periodic
inventory requirement is only acceptable if approved by the
Regional office.
Emissions are submitted for
mobile sources.
Ongoing review of the AERR may result in requiring mobile model
inputs and making emissions data optional. If this occurs, the SIP
periodic inventory must still include emissions. Thus, emissions
would need to be submitted for the NAA counties regardless of
any AERR requirements to be able to meet the periodic ozone
inventory requirement.
Complete - all sectors
Using the AERR to meet the periodic inventory requirement is only
possible during triennial years (see Section 2.4) because the EIS
only accepts nonpoint and mobile data during these years, and
these data categories must be included to be complete.
Submitted within 3 years
since the base year
inventory.
It is acceptable to be less than 3 years, after which time the
triennial submissions would be every three years.
43
-------
Condition
Comment
Partial county approach
approved by Regional office
The EIS does not support partial counties. In some cases, it may be
acceptable to use the entire county emissions for the periodic
submission.
Notification
If the AERR approach is approved, the Regional office should be
notified in writing when the submission has been completed.
Documentation
Must be created. Submission is at the request of the EPA Regional
office.
3,5,2 Regional Haze Recent Year Inventory
For the Regional Haze Rule, a recent year inventory is needed to support the analyses required with
each) periodic progress report or 10-year SIP revision. The inventory is not submitted with the SIP or
progress report, but emissions summaries are expected as part of the analysis provided with the SIP.
The inventory to be used is a complete statewide annual inventory as was the initial baseline inventory
(included in the original Regional Haze SIP). See Table 1 above for the latest reference materials with
guidance on the Regional Haze Progress Reports. Since the regional haze requirement is for inventories
to be statewide and cover emissions over an annual period, the spatial and temporal aspects better
match the AERR reporting requirements as compared to the NAA inventories required for ozone and
PM2.5 implementation. In addition, since the recent year inventory is not an element of the SIP that must
be submitted itself, there is not a public hearing requirement for the inventories used for regional haze.
Thus, the AERR inventories are an appropriate starting point for meeting the requirement for the recent
year inventory for regional haze. If better information becomes available for particular source sectors at
a time when it can be incorporated into a state's long-term strategy development, it should be.
The inventory approach for the recent year inventories needed for periodic progress reports and SIP
revisions is relatively straightforward. The following list provides considerations for states in developing
their inventories:
• The rule prescribes that the inventory year is the same as triennial AERR year that has most
recently been submitted by the state to the EPA, with additional special considerations
o For the recent year emissions inventory for a 10-year SIP revision, states have a 12-
month grace period after submission of inventory information before the regional haze
rule requires the state to use that information in a SIP. (see 40 CFR 51.308(f)(2)(iii)). One
example of using another year would be to update the EGU portion of the inventory to
use more current data that is available because of CEMs. There may also be better
information on a particular source sector in a special study that collected more refined
source-specific information but for a different year.
o For the periodic progress reports, 40 CFR 51.308(g)(4) specifically requires that the
emissions information in the progress report must extend through the most recent year
for which the Administrator has provided a State-level summary via a centralized
44
-------
emissions data system operated by EPA (like the CEM data) if summaries of the data are
available from the EPA 6 months before the state's progress report due date.
• The inventory can be the same as the NEI or can be revised from the NEI. The RHR does not say
that the inventory must be the NEI inventory, rather, it says that the year must be the same
year. The same applies with respect to an EPA summary of data for a source sector reported
directly to the EPA.
• States must identify the recent inventory in their plan, but the inventory does not need to be
submitted to the EPA.
• States may rely on EPA-provided parts of the NEI to support the recent year inventory data
needed for their plan.
nissions Statement Requirement for SIPs
For ozone NAAs, states should develop emission reporting programs for VOC and NOx sources in
accordance with CAA section 182(a)(3)(B). The required state program and associated regulation defines
how states obtain emissions data directly from facilities and report it to the EPA as part of SIP
inventories and for the AERR. This is called an Emission Statement regulation, and it outlines how
facilities must report emissions and facility activity data to an air agency. Once the air agency develops
the Emission Statement regulation, the regulation should be sent to the EPA as a SIP for approval by the
deadline as specified in Table 8. As an element of a SIP, the public hearing requirements apply to the
Emission Statement SIP. Air agencies must ensure that their emissions reporting program, promulgated
in accordance with CAA section 182(a)(3)(B), complies with all applicable statutory and regulatory
requirements on a continuing basis.
To properly implement the emissions reporting requirements, Emission Statement regulations should be
coordinated carefully with the AERR and any associated guidance and specifications, the Modeling
Guidance, and this guidance. The required and optional data elements needed for inventories
(described in Section 2.5.5) should be included as fields for reporting to the EPA. Where the AERR/EIS
has required fields, those fields should be required by the Emission Statement regulations. In addition,
the Emission Statement regulations should require ozone season emissions in accordance with the final
Ozone Implementation Rule.
Additional details on developing emission statement regulations can be found in a dated, but still
relevant guidance document titled, "Guidance on the Implementation of an Emission Statement
Program (DRAFT)," available at https://www.epa.gov/air-emissions-inventories/implementation-
emission-statement-program. Where data elements to be collected are mentioned, users of that
guidance should defer to the paragraph above, explaining that data elements should be coordinated
with AERR/EIS requirements and the recommendations in this guidance for optional fields.
Lance Inventories
Maintenance inventories reflect the emissions conditions associated with a NAAQS redesignation to
attainment following an initial area designation of nonattainment. Thus, the attainment year inventory
45
-------
for the maintenance demonstration should reflect emissions during one of the three years for which
monitoring data showed compliance with the standard. The EPA recommends that the inventory year to
be used for the maintenance inventory should be a year for which all air quality monitors in the
nonattainment area had ozone concentrations below the level of the standard (e.g., for a nonattainment
area for the 2008 ozone NAAQS, all monitors in the area should have 4th high values below 0.075 ppm in
order for the year to be used for the maintenance inventory, also see Calcagni 1992 memo, page 8).
Otherwise, all characteristics of the attainment year inventory are the same as the base year inventory
submitted with the implementation plan prior to attainment. An attainment year maintenance
inventory is a component of a maintenance plan and is, therefore, submitted to the EPA as a SIP
revision. Thus all guidance in this document for completeness, temporal extent, spatial extent, public
hearings, documentation, and summaries applies to these inventories.
The on-road portion of the inventory for the relevant pollutants and precursors for the last year of the
maintenance plan would be the motor vehicle emissions budget(s) required for maintenance areas by 40
CFR 93.118(b). States may elect to also establish motor vehicle emissions budgets for the interim year or
years for which they have submitted inventories or for the attainment year of the maintenance plan.
Those decisions should be made through the interagency consultation process that is required by CAA
section 176(c)(4)(D) and 40 CFR 93.105.
Specific additional guidance for maintenance plan SIPs is also available in the memorandum titled,
"Maintenance Plan Guidance Document for Certain 8-hour Ozone Areas Under Section 110(a)(1) of
Clean Air Act", which can be found at
(https://www3.epa.gov/ttn/naaqs/aqmguide/collection/cp2/2005052Q wegman maintenance 8-
hr ozone section 110(a)(l).pdf).
jcted Inventories
The purpose of a projected inventory is to provide an emissions estimate for a year beyond the base
year. These inventories are also called future-year inventories. Projected inventories will reflect the net
effect of emissions resulting from anticipated changes in future activity (such as emissions increases due
to growth in VMT; emission decreases due to implementation of existing requirements for stationary
source and car/truck emissions); and the effects of new control measures needed to reduce emissions
and bring nonattainment areas and states into compliance with a particular NAAQS or regional haze
program requirement.
Projected inventories are used for a number of purposes in implementing the ozone, PM2.5, and
regional haze programs. As listed in Table 4, there are two possible attainment year projection
inventories. The first is the attainment projected inventory for the NAA. The second is the projected
inventory for the modeled attainment demonstration, which typically reflects emissions for the entire
modeling domain (commonly an area larger than the nonattainment area). Note that the projected
inventory for the modeled attainment demonstration represents the future year emissions throughout
the entire modeling domain (which will be a larger geographic area than the NAA).
46
-------
Projected inventories for certain years prior to the attainment year are also used in addressing the rate
of progress/reasonable further progress requirements in ozone and PM2.5 attainment plans. Projected
inventories are also needed for ozone or PM2.5 maintenance plans for areas to be redesignated to
attainment.
Creating projected inventories includes a large number of data sources and requires estimates for
stationary point, nonpoint, on-road mobile, and nonroad mobile sources. In addition, some areas may
choose to use average fires to represent these highly variable (from year to year) and large contributors
to emissions. Biogenic sources are not generally forecast into the future because they are so dependent
on meteorology. Section 5 covers additional details on projected inventories.
3,8,1 Ozone NAAQS Implementation
For ozone NAAQS implementation, projected inventories are used for three main purposes. They are
used for modeled attainment demonstrations, which are required for Moderate and above areas.; for
ROP/RFP plans, which are also required for Moderate and above areas; and in maintenance plans, to
show that NOx and VOC emissions will not exceed attainment year emissions levels at any time after
attainment.
Ozone ROP/RFP plans must describe how the required percent reductions of relevant pollutants will be
achieved on specific timetables. As described in Section 3.8.2, one possible use of the attainment
projected inventory for the NAA is for showing emissions reductions associated with an ROP/RFP plan.
While it is conceivably possible to estimate only the emissions reductions (rather than future year
emissions inventories) required by the ROP/RFP plan, the most common way to estimate the reductions
is by creating future-year inventories and then calculating the difference between the ROP/RFP baseline
and the attainment projected inventories for the NAA.
As described in Section 3.4 and the Ozone Implementation Rule, Moderate and above NAAs must have a
plan for meeting a 15 percent VOC emissions reduction requirement (ROP plan) within 6 years after the
NAA designation. Serious and above NAAs must also have a plan (RFP plan) to meet a requirement for
an additional 3 percent per year VOC reduction (annual RFP) for each 3-year period after the initial
6 years (after the NAA designation) until the attainment date. In both cases, the VOC reductions can be
substituted with NOx reductions when specific conditions apply. Serious and above areas may submit
two separate plans since the annual RFP plan is due 1 year later. Collectively, this guidance calls these
plans the ROP/RFP plan(s).
For the ROP requirement, an attainment year projected inventory for the NAA can also meet the ROP
requirement by providing a projected inventory that shows the 15 percent VOC (or NOx, as applicable)
reductions when compared to the ROP/RFP baseline NAA inventory. The EPA recommends that
summaries of these projected NAA inventories are included in the ROP/RFP plan(s) to communicate the
emissions reductions by comparing against the ROP/RFP baseline NAA inventory.
For Serious and above areas, an additional projected inventory may be needed to meet the ROP
requirement. This is because the ROP requirement at 6 years can be earlier than the expected
47
-------
attainment date. In the case for Serious and above areas, ROP plans cannot use the projected
attainment inventory for the NAA, but rather should use an inventory projected at the 6-year interval.
To estimate the emissions reductions associated with ROP/RFP plans, air agencies should rely on the
same techniques as for other projection purposes, which are described in Section 5. When these
reductions are calculated by creating projected inventories, the on-road portion of the inventories may
also need to follow transportation conformity guidance
(http:/www.epa.gov/otaq/stateresources/transconf/) so that it can be used to set the motor vehicle
emissions budgets.
3.8.2 PM2.5 NAAQS Implementation
For PM2.5 NAAQS implementation, the projected inventories are used for modeled attainment
demonstrations; for RFP plans; and in maintenance plans, to show that PM2.5 and PM2.5 plan precursor
emissions will not exceed attainment year emissions levels at any time after attainment.
For PM2.5 SIPs, the attainment projected inventory for the NAA is a required element of the SIP. The
detailed requirements for various aspects of this inventory are included in the regulatory text at 40 CFR
51.1008(a)(2) for Moderate areas and 40 CFR 51.1008(b)(2) for Serious areas. For the attainment
projected inventory for the NAA, a key difficulty can be calculating projected emissions for partial
counties, which would not necessarily be available from the modeling inventory. It may be possible to
use the gridding aspect of modeled inventories to estimate emissions in partial counties for nonpoint
and mobile sources. This issue is addressed more completely in Section 4.4.1.
The projected inventory for the modeled attainment demonstration represents the future year
emissions throughout the entire modeling domain (which will be a larger geographic area than the
NAA). This projected inventory represents estimated lower emission levels due to implementation of
ongoing and any new emission reduction strategies that are needed to attain the relevant PM2.5 NAAQS
in a future year, as reflected by air quality modeling.
PM2.5 RFP plans must include "RFP projected emissions" for each applicable quantitative milestone year,
based on the anticipated control measure implementation schedule. These emissions must be described
by source sector, and on a pollutant-by-pollutant basis (see section IV.F. of the PM2.5 NAAQS SIP
Requirements rule).
3.8.3 Regional Haze Implementation
For regional haze program implementation, section 308(f)(6)(v) includes the requirement for SIPs to
provide for a statewide inventory of estimates of future projected emissions. The purpose of the
projected inventory is to address the SIP requirement to focus on emission reduction measures that are
designed to achieve reasonable progress by the end of each implementation period (2018, 2028, and
every 10 years thereafter) by reducing emissions that lead to regional haze.
In addition, section 308(g)(4) of the regional haze rule addresses the emissions-related requirements for
the periodic reports. This section requires emissions changes to be identified by type of source or
activity. However, there is no requirement for a projected emissions inventory for progress reports. The
48
-------
tracking analysis would encompass emissions up to the year of the most recent NEI submission for all
sources, and for relevant sources up to the most recent year for which the EPA has provided a summary
of emission information submitted by sources directly to the EPA as of a date 6 months prior to the due
date of the progress report. More information about this comparison is given in EPA's "General
Principles for the 5-Year Regional Haze Progress Reports" document.
As described in 308(f)(2), the state's long-term strategy to address regional haze must include
"enforceable emission limitations. The EPA expects that the estimates of future projected emissions to
meet the requirements of 308(f)(6)(v) would therefore reflect the projected impacts of these
enforceable emission limitations.
The RHR does not assume that all sources in a state must be subject to an analysis of the four factors
laid out in 308(f)(2)(i), but instead requires the state to identify the sources considered by the state in
developing its long-term strategy and the criteria used to select the sources considered. However, once
the emissions reduction measures for these sources are identified pursuant to the four-factor analysis,
the emission reductions from those measures would be reflected in the projection of end-of-
implementation period emissions needed to set the reasonable progress goals 308(f)(3). The emissions
projection techniques provided in Section 5 below can provide the necessary technical guidance to
support states in projecting their future emissions.
3.8.4 Transportation Conformity Program and Motor Vehicle Emission Budgets
The on-road mobile portion of the attainment projected inventory for the NAA can serve as the motor
vehicle emissions budgets for transportation conformity purposes for ozone and PM2.5 NAAs. Specific
requirements for motor vehicle emissions budgets for ozone (NOx and VOCs) and PM2.5 (direct PM2.5,
NOx and possibly VOCs, S02 and NH3) can be found in 40 CFR 93.102(b). Since these budgets are
required for transportation conformity purposes by 40 CFR 93.118(b), such inventories provide a
convenient and consistent way to meet that requirement. In this case, the on-road inventory should be
created in accordance with transportation conformity guidance
(http:/www.epa.gov/otaq/stateresources/transconf/).
3.8.5 Additional Relevant Provisions
Additional relevant provisions of 40 CFR part 51 must be considered for ozone, PM2.5, and regional haze
implementation purposes. Subpart G of 40 CFR part 51 provides some foundational regulatory aspects
of the control strategy development needed for creating projected inventories, including what SIPs must
include with regard to control strategies and their associated inventories. In particular, 40 CFR 51.114
"Emissions data and projections" reads:
(a) Except for lead, each plan must contain a detailed inventory of emissions from point and area
sources. Lead requirements are specified in §51.117. The inventory must be based upon
measured emissions or, where measured emissions are not available, documented emission
factors.
(b) Each plan must contain a summary of emission levels projected to result from application of the
new control strategy.
49
-------
(c) Each plan must identify the sources of the data used in the projection of emissions.
In addition, 40 CFR 51.116 "Data availability" reads:
(a) The State must retain all detailed data and calculations used in the preparation of each plan or
each plan revision, and make them available for public inspection and submit them to the
Administrator at his request.
(b) The detailed data and calculations used in the preparation of plan revisions are not considered a
part of the plan.
(c) Each plan must provide for public availability of emission data reported by source owners or
operators or otherwise obtained by a State or local agency. Such emission data must be
correlated with applicable emission limitations or other measures. As used in this paragraph,
correlated means presented in such a manner as to show the relationship between measured or
estimated amounts of emissions and the amounts of such emissions allowable under the
applicable emission limitations or other measures.
Thus, states should keep these requirements in mind as they develop their projected inventories so that
these required SIP elements can be included.
ic Hearings
For ozone and PM2.5 implementation, some emissions inventories are considered a part of the SIP under
the CAA (see sections 172(c)(3), 182(a)(1)(A) and 182(a)(3)(A)), and in this guidance we have called these
planning inventories. Thus, for ozone and PM2.5 SIPs, detailed emissions inventory data are required to
be included in the public participation associated with SIPs as specified in CAA section 110(a)(2). For
regional haze, the detailed emissions inventory data are not expected to be a part of the information
provided for a public hearing on the regional haze SIP. If emissions inventory information (but not the
detailed data) is included as part of a SIP submission, the emissions information in the SIP submission
still needs to be included as part of the information provided as specified in CAA section 110(a)(2).
In addition, 40 CFR 51.102(a) further explains that a public hearing requirements means "The State must
hold a public hearing or provide the public the opportunity to request a public hearing." Thus, the
phrase "or provide the opportunity to request" means that it is possible that public hearings on
inventories will have no one request a public hearing16.
There are policy reasons that support the need for detailed emissions inventories data to be approved
with the SIP. Some emissions inventories must be used as a basis for ozone RFP plans for areas classified
16 A public hearing must be offered for every SIP submittal, but a hearing does not need to actually occur if no one requests a
hearing. The detailed requirements for public notice and hearings are given in 40 CFR 51.102. See also Attachment B of
"Regional Consistency for the Administrative Requirements of State Implementation Plan Submittals and the Use of Letter
Notices," EPA memorandum dated April 6, 2011, from Janet McCabe, Deputy Assistant Administrator for the Office of Air &
Radiation to the Regional Administrators,
www3.epa.eov/airaualitv/urbanair/sipstatus/docs/FINALSIPGuidelinesSubLtrsPN.pdf.
50
-------
as Moderate or above. Additionally, all NAAs seeking redesignation to attainment will rely on an
emissions inventory for purposes of submitting a maintenance plan under CAA section 175(A). Because
the emissions inventories and projections play such a fundamental role in determining the reductions
necessary to attain and maintain the standard, they need to receive full public scrutiny and be formally
incorporated into the SIP.
As explained in the final Ozone SIP Requirements Rule, the EPA is no longer advising states to follow our
existing September 29, 1992, guidance titled, "Public Hearing Requirements for 1990 Base-Year
Emissions Inventories for Ozone and Carbon Monoxide Nonattainment Areas" in implementing certain
SIP adoption and submission procedures for the emissions inventory requirements under CAA sections
182(a)(1) and 182(a)(3)(A) for purposes of the 2008 ozone NAAQS. In that guidance, the EPA indicated it
could provide states with a time-limited "de minimus" deferral of the CAA's state public hearing
requirement for the emissions inventory SIP revision until the inventory needs had "regulatory
significance." As there is nothing in the CAA provisions that provides for waiver or delay of the public
notification and hearing requirements specified in CAA section 110(a) de minimus or otherwise, we no
longer believe it is appropriate to advise states to follow the 1992 guidance. We remind states that the
EPA's implementing regulations at 40 CFR part 51 (Requirements for Preparation, Adoption, and
Submittal of Implementation Plans) provide flexibility for states to streamline SIP-related public
notification and hearing procedures (for example, only holding a public hearing if one is requested, per
40 CFR 51.102), and we encourage states to take advantage of those provisions in meeting the
emissions inventory requirements under CAA sections 182(a)(1) and 182(a)(3)(A).
For areas classified as Moderate or above, the state would need to provide the EPA with verification
that a public hearing was offered for the emissions inventory for any emissions inventory SIP revision at
the time the state submits the emissions inventory SIP.
In instances where a public hearing is held for the emissions inventory SIP revision, the EPA clarifies here
what is required by EPA regulations to be published and present for public review. Because emissions
inventory data can be voluminous, it is common to summarize that data for public hearing purposes.
The EPA does not interpret 40 CFR 51.102 to require that at a public hearing, held in accordance with
that provision, every database entry related to the emissions inventory must be summarized, presented,
and/or made available. Rather, the EPA expects that, in general, emissions inventory summaries would
be the primary source of information for public review, so long as the detailed emissions inventory data
are available to the public in accordance with the CAA and EPA regulations. Air agencies must comply
with 40 CFR 51.102(d)(2), and 40 CFR part 51, Appendix V.
At a minimum, the state should include summaries of the actual emissions. The EPA recommends using
the same tables as those recommended in the El Guidance for inclusion in the SIP documentation for
the purpose of the public review (see Section 3.10.3 of this guidance "What to Submit").
51
-------
-¦ ¦ vibmlt Inventory Parts of SIP to tl •"
40 CFR part 51 contains the requirements for SIPs, and many general provisions in Part 51 impact the
requirements for emissions inventory SIPs. Section 3.10.1 summarizes the applicable regulations. In
addition, the two key concerns for submitting information on inventories are: (1) how to submit,
including the format and (2) what to submit. These concerns are addressed in Sections 3.10.2 and
3.10.3, respectively.
3,10,1 Regulatory Requirements in Addition to Implementation Rules
The general regulatory requirements for SIP submissions are available in several subparts of 40 CFR part
51 that have an impact on the emissions inventory SIPs.
• Subpart F- Procedural Requirements
o §51.102 Public hearings (see Section 3.9 above)
o §51.103 Submission of plans, preliminary review of plans:
Cites Appendix V as the source of the requirements that constitutes an official
plan submission.
• Subpart G - Control Strategy
o §51.112 Demonstration of adequacy
This section requires a presentation of emissions levels expected to result from
implementation of each measure of a control strategy (in the modeled
attainment demonstration).
o §51.114 Emissions data and projections
This section has three subsections:
(a) requires a detailed inventory of emissions from point and area sources that
are based on measure emissions or, where measured emissions are not
available, documented emission factors;
(b) requires that the plan contain a summary of emission levels projected to
result from application of the new control strategy; and
(c) requires that each plan must identify the sources of the data used in the
projection of emissions.
o §51.116 Data availability
Subsection (c) requires that each plan must provide for public availability of
emission data reported by source owners or operators or otherwise obtained by
a state or local agency. It also requires correlation of emission data with
applicable emission limitations or other measures.
• Subpart Q - Reports
o §51.322 - Sources subject to emissions reporting
This section points to Subpart A, which is the AERR.
o §51.323 - Reportable emissions data and information
This section also points to Subpart A.
• Appendix V - Criteria for Determining the Completeness of Plan Submissions
52
-------
o 2.2 Technical Support. This subsection of Appendix V lists various parts of the
SIP that must be included that can be met as part of the El SIP, including:
(b) Identification of the location of affected sources included the EPA
attainment/nonattainment designations of the locations and the status of the
attainment plan for the affected area(s).
(c) Quantification of the changes in plan allowable emissions from the affected
sources; estimates of changes in current actual emissions from affected sources
or, where appropriate, quantification of changes in actual emissions from
affected sources through calculations of the differences between certain
baseline levels and allowable emissions anticipated as a result of the revision.
(e) Modeling information required to support the proposed revision, including
input data (which includes emissions).
o 3.2 Paper Plan Submissions. This section addresses the EPA's voluntary
guidelines for which states are requested to adhere for paper and electronically
formatted submissions.
3.10,2 flow to Submit Inventory Parts of a SIP
The EPA generally encourages air agencies to provide all parts of the SIP related to emission inventories
using digital media. For the Ozone SIP Requirements Rule, the preamble specifically indicates that
"states must make available the inventory data to the EPA as electronic files in ... electronic media, such
as FTP, zip drives, or DVDs." In most cases, it's simply best practice to provide this information digitally.
Because of the challenges that can arise with the use of formats that are supported by one agency and
not the other, air agencies should provide documentation at least in formats that are readily
transferrable, such as PDF for documentation files. Other formats can also be considered and agreed
upon in advance, and such details would be ideal to be included in an IPP.
When paper plans are submitted, some guidelines provided in 40 CFR part 51, Appendix V apply. In
accordance with section 3.2(a) of Appendix V, the EPA requires that when submitting one paper plan, it
must be accompanied by an electronic duplicate of the entire paper submission, preferably as a word
searchable document format (PDF).
For the format, the above information is provided as guidance. There are no specific formatting
requirements added by this guidance for the inventory data, emissions summaries, or documentation
submitted as part of the SIP. Requirements are included in implementation rules, and the ozone, PM2.5,
and regional haze implementation rules do not specify required formats for submitting inventory
information. For ozone and PM2.5, however, the data elements required for submitted emissions
inventories are to use the elements required by the AERR.
53
-------
Currently, the AERR data format does not directly support the use of partial counties. Thus, inventories
with partial counties are not currently able to use the EIS formats and states would need to use some
identifier in the format to denote partial counties.
The following provides a list of best practices that the EPA encourages air agencies to use when
submitting the Inventory Parts of a SIP:
• In accordance with Subpart A, detailed inventory data are required as part of emission inventory
SIPs. When providing full detailed inventory data (i.e., facility-SCC-pollutant data for point
sources or county-SCC-pollutant data for nonpoint sources):
o Use the data elements required by the AERR. This is required by the ozone and PM2.5
implementation rules.
o Coordinate with your Regional office about what format is going to be used. If EIS
format is used, consider submitting this to the EIS, but first contact the EPA NEI program
to make sure any such submission will not conflict with other inventories already
submitted to the EIS.
o If not using the EIS format, use Excel®, Access® or a rigorous delimited text format. For a
delimited text format, make sure that the text fields have double quotes and that the
delimiter chosen (e.g., a tab) does NOT appear in any of the text fields, since this will
prevent the delimited dataset from being imported easily into other data systems.
• For documents, use a readily available word processor format (e.g., Microsoft Word®) or convert
the document to PDF format.
• At times, it can be very time consuming to convert tables from a spreadsheet and insert them
into a word processor, especially if the information changes and has to be updated. The
guidance below recommends a lot of summary tables. Rather than cut/pasting into a text file,
states may consider creating a spreadsheet with individual worksheets (tabs) labeled as each
table included in the document. Then, just reference those tables/tabs from document. This can
be done with words only (e.g., "see Table 1 in Excel file") or using a more sophisticated data link
between the word processor and spreadsheet software.
3,10,3 What to Submit
With regard to exactly what to submit, the information provided here provides guidance on what would
be acceptable in most cases. The list of tangible data items and documentation forms the basis of what
the EPA will review and ultimately approve for a SIP. Table 11 provides the suggested inventory
elements of an Emissions Inventory SIP and indicates whether they are relevant for ozone, PM2.5, and/or
Regional Haze SIPs. Refer to the tables in Section 2.3 for more information on which elements are
required and which are optional.
This section (and Table 11) does not include RFP inventories since those are submitted with a separate
ROP/RFP plan. Similarly, it does not include information needed for a modeling attainment
demonstration submittal to the EPA, with the exception of the inclusion of statewide inventory
summaries for ozone and PM2.5 SIPs for areas that have modeled attainment demonstrations as part of
their SIP requirements.
54
-------
Table 11: Suggested elements of an emissions inventory SIP document
SIP Planning Inventory Documentation Element
Ozone PM2.5 Haze
Background: Includes key elements explained in this guidance, such as
the nature of the air quality problem (or arrival at attainment
status for maintenance SIPs), priority source categories and point
sources, pollutants covered, spatial extent, and other key
inventory aspects.
• Overall base year emissions summaries: All emissions are for
relevant pollutants only. Data categories would minimally be
point, nonpoint, nonroad mobile, and on-road mobile, and fires,
though more detailed categories are acceptable. Biogenic
sources and other sources can optionally be included as well,
and may be required by Regional offices for areas doing
modeled attainment demonstrations.
See Table 12
• Overall projected attainment year emissions summaries: The
same as the base year summaries except using the attainment
projected inventory for the NAA. Required for PM2.5 SIPs. These
summaries are optional for the ozone Emissions Inventory SIP.
For ozone and when developed in accordance with
transportation conformity guidance, the on-road portion also
can set the motor vehicle emission budgets for VOC and NOx. If
not provided in the Emission Inventory SIP, the summaries would
be needed in the Modeled Attainment Demonstration.
See Table 12
n/a
• Overall recent year emissions summaries compared to
projected emissions: All emissions are for relevant pollutants
only. Data categories would minimally be point, nonpoint,
nonroad mobile, on-road mobile, and fires.
See
Table
12
Stationary Point Sources:
Introduction with more detailed documentation of the
emissions from stationary point sources, including the point
source emissions threshold, how the point sources were
identified, how emission estimates were made including
emission factors, activity data, and control assumptions.
• This section includes additional point source emissions
summaries.
See Table 13
Stationary Area (Nonpoint) Sources:
Introduction with more detailed documentation of the
emissions from stationary nonpoint sources, including how the
nonpoint sources were identified, the model version used,
which sources were not addressed and why, how partial county
emissions were estimated, and how emission estimates were
made including models or tools, emission factors, activity data,
and control assumptions.
• This section includes additional nonpoint source emissions
summaries.
See Table 14
55
-------
SIP Planning Inventory Documentation Element
Ozone
PM2.5
Haze
Nonroad Mobile Sources
• Introduction with more detailed documentation of the
emissions from nonroad mobile sources, including how the
nonroad mobile sources were identified, which sources were
not addressed and why, how partial county emissions were
estimated, and how emissions estimates were made including
model version, source of input data, input and output data files,
and any post-processing methods used to develop the
inventory.
Y
• This section includes additional nonroad source emissions
summaries.
See Table 14
On-road Mobile Sources
• Introduction with more detailed documentation of the
emissions from on-road mobile sources, the model and version
used, how partial county emissions were estimated, and how
emissions estimates were made including sources of input data,
input and output data files, and any post-processing methods
used to develop the inventory.
Y
• This section includes additional on-road source emissions
summaries.
See Table 14
Other Sources
• If applicable, a description of other sources in the emissions
inventory not included and reasons why they are not included.
Y
• If applicable, this section includes summaries of other sources of
emissions.
See Table 14
Demonstration of Public Hearing: A brief description of the public
hearing process, a summary of the results, and any other relevant
materials. The purpose of this section is to provide evidence for the
public hearing and what the air agency learned through the process.
Y
Y
N
Table 12 provides the summaries to include in the background section. In the table, "data category" is
consistent with the definition provided in Table 11 and Section 2.5.1. If both the base year inventory and
the attainment projected inventory are included in the background, then the same summaries should
generally be provided for both.
Table 12: Suggested summaries to include in the background
Emissions Summaries
Ozone
PMz.5
Haze
Annual baseline, recent year emissions, projected emissions, and emissions
difference for the state by data category. The baseline emissions from
the original SIP may be used.
Y
Annual base year emissions and/or average season day emissions for the
NAA by data category
Y+
Annual base year emissions and/or average season day emissions for the
NAA by data category and county (or partial county)A
Y+
Ozone season day base year emissions for the NAA by data category
Y
56
-------
Emissions Summaries
Ozone
PMz.5
Haze
Ozone season day base year emissions for the NAA by data category and
county (or partial county)
Y
+ Either annual or season-day emissions could be reported (but not necessarily both), depending on the temporal basis used
for the emissions in the SIP. For areas that have nonattainment for both the annual standard and the 24-hour standard and
are using average-season-day inventory for their 24-hour SIP, then summaries of both temporal resolutions would be
needed.
t For SIPs that include air quality modeling, this section can include both summaries of the planning and the modeling
inventories. Since modeling domains are usually much larger, multi-state areas, the resolution of the modeling inventory
summaries depends on the modeling guidance and the preferences of the Regional office. One option is to include county-
level detail for the state(s) included in the NAA, but state-level detail for surrounding states.
K Partial county emissions can be included rather than, or in addition to, full county, where applicable.
A Providing both the county and partial county could provide more flexibility for states when submitting periodic inventories.
Table 13 provides the recommended minimal contents for summaries to include in the point sources
section. If both the base year inventory and attainment projected inventory are included in the
background, then the same summaries should generally be provided for both.
Table 13: Summaries to include in the point sources section
Emissions Summaries
Ozone
PM2.5
Haze
Annual recent year emissions, projected emissions, and emissions
difference for each AERR point source as a facility total, within the
state.
Y
Annual base year emissions and/or PM2.5 average season day base year
emissions for each point source as a facility total, within the NAA.
Y
Ozone season day base year emissions for each point source as a facility
total, within the NAA.
Y
These summaries should include: the county, Agency Facility ID, plant name, physical address, total
control efficiency, key control devices, NAICS code, and actual and/or projected emissions. Other
summaries may be included as well, such as by SCC or other fields that are necessary to describe the
inventory effectively. The Agency Facility ID or some other ID should be included that can easily map
to the EPA facility records in EIS. This can be more difficult when air agencies change their Agency
Facility IDs.
Table 14 provides the recommended minimal contents for summaries to include in the nonpoint and
mobile sources sections. If both the base year inventory and the attainment projected inventory are
included in the background, then the same summaries should generally be provided for both.
57
-------
Table 14: Summaries to include in the nonpoint and mobile sources sections
Emissions Summaries
Ozone
PM2.5
Haze
Annual recent year emissions, projected emissions, and emissions
difference by county and SCC (or SCC group*), within the state.
Y
Annual base year emissions and/or average season day emissions by county
and SCC (or SCC group*), within the NAA.
Y
Ozone season day base year emissions by county and SCC (or SCC group*),
within the NAA.
Y
For county and emissions processes reported that have reductions associatec
summaries should include any control efficiencies or rule effectiveness assum
projected emissions. Projected attainment year emissions summaries may als
performing modeled attainment demonstrations.
with rules, these
ed and actual and/or
o be required for areas
t See additional information below on SCC groups.
SCC groups for on-road and nonroad mobile sources can be useful in limiting the size of summaries
without much loss of information. For example, SCC groups can be formed by vehicle type and fuel,
dropping the road classifications. Vehicle types can be further grouped into light duty and heavy duty.
The amount of grouping appropriate depends on what a particular summary and SIP are trying to
communicate. SCC groups can be suggested as appropriate by Regional offices or air agencies as part of
the consultation for what will be included in the emissions inventory SIP.
58
-------
4 Developing Current Emissions inventories
This section covers the decisions that states will need to make regarding any "current" emissions
inventory, including a base year NAA or state inventory, RFP baseline inventory, periodic inventory (for
ozone) or most recent year inventory (for regional haze), or maintenance inventory. As a reminder, all
inventories described in this section are inventories of actual emissions; these inventories do not
contain emissions based on potential to emit nor allowable emissions. In accordance with 40 CFR
51.5(b), states are urged to use applicable, state-of-the art techniques for estimating emissions.
ventory Years
A fundamental step in creating a base year inventory is the choice of inventory year. The Ozone, PM2.5
and Regional Haze Implementation Rules include specific language regarding the choice of inventory
years, and those regulatory requirements supersede any guidance provided in this section. Since the
purpose of the base year inventory is to represent relatively current emissions conditions, there is a
general preference to use as recent a year as practical. This means that the year of emissions estimates
is usually several years earlier than the year in which the inventory work is done, given the time that it
takes to compile an emissions inventory (for example, an inventory with 2014 emissions may not be
compiled until 2016).
The primary considerations in choosing a baseline inventory year as covered in this guidance are: (1) the
implementation rule requirements, (2) transportation conformity requirements, and (3) the availability
of data. This is in contrast to inventories used for air quality modeling, for which meteorology, and
availability of ambient data are the primary considerations, with inventory availability secondary. This is
because the meteorological conditions selected should be associated with ozone formation that is
representative of either the ambient data relied upon for designation to nonattainment or recent
ambient data, as appropriate. Choosing a year based solely on inventory availability could result in
choosing an inappropriate year (considering ambient measurements and the meteorology that led to
those measured concentrations). As a result, the base year(s) used in the planning inventories can be
different than the modeling base year and associated modeling inventory.
4,1,1 Ozone Baseline Inventory Year
For ozone, states should become familiar with the specific rule requirements. The 2008 Ozone
Implementation Rule in 40 CFR 51.1115(a) specifies that the "inventory year shall be selected consistent
with the baseline year for the RFP plan as required by Section 51.1110(b), which reads as follows:
(b) Baseline emissions inventory for RFP plans. For the RFP plans required under this section, at
the time of the designation for the 2008 ozone NAAQS, the baseline emissions inventory shall be
the emissions inventory for the most recent calendar year for which a complete triennial
inventory is required to be submitted to EPA under the provisions of subpart A of this part. States
may use an alternative baseline emissions inventory provided the state demonstrates why it is
appropriate to use the alternative baseline year, and provided that the year selected is between
the years 2008 to 2012. All states associated with a multi-state nonattainment area must consult
and agree on a single alternative baseline year.
59
-------
Thus, for the 2008 ozone standard, the 2011 inventory year is the presumed inventory year. Guidance
on the considerations for using an alternative baseline year is provided below in Section 0.
The consideration of transportation conformity requirements is important because the rule has its own
language about selecting a base year. The applicable section of the rule is 40 CFR 93.119(e), which reads:
(e) Baseline year for various NAAQS. The baseline year is defined as follows:
(4) The most recent year for which EPA's Air Emissions Reporting Rule (40 CFR Part 51, Subpart
A) requires submission of on-road mobile source emissions inventories as of the effective date
of designations, in areas designated nonattainment for a NAAQS that is promulgated after
1997.
This requirement impacts the 2008 and 2015 ozone standards. For convenience, OTAQ provides an
online table that lists the appropriate base year for use for transportation conformity purposes (see
https://www.epa.gov/state-and-local-transportation/baseline-vear-baseline-vear-test-40-cfr-93119).
Unlike the ozone implementation rule, the transportation conformity rule does not include a provision
for an alternative baseline year; thus, states that wish to select an alternative year are advised to consult
with their Regional office as part of their SIP development.
Lastly, the consideration of "availability" is simply what data are (or will be) available to the air agency
during the time alloted for SIP preparation. Often, the triennial NEI year inventories (as described above)
are useful starting points for meeting the various "current" inventory requirements associated with SIPs.
For these triennial NEI years, states are already required to have prepared and provided to the EPA
inventories of criteria pollutants and precursors as well as model inputs for on-road and nonroad mobile
sources. In some cases, the states have submitted their data to the EPA, but the complete NEI is not yet
available because there is a 6- to 12-month lag between when the emissions have been submitted and
when the inventory is released by the EPA. In this case, the state should make every effort to use the
most currently available information and incorporate any new information that is made available by the
NEI release at the earliest practical time.
Other benefits of using an NEI year is that NEI data are available for all states, and information from
more than one state may be needed for attainment demonstration modeling purposes. Additionally,
other data needed for modeling (e.g., meteorology and SMOKE inputs) are more likely to be available
for the triennial NEI years.
4,1,2 PMis Baseline Inventory Year
For PM2.5, states should become familiar with the specific rule requirements. The PM2.5 SIP
Requirements Rule at 40 CFR 51.1008(a)(l)(i) states:
(i) The inventory year shall be one of the 3 years for which monitored data were used for the
designations or another technically appropriate year if justified by the state in the plan
submission.
60
-------
For the 2012 PM2.5 NAAQS, the EPA used ambient data from 2011 through 2013 to designate PM2.5
NAAs; thus any of these 3 years can be used as the baseline year. The idea behind this connection is
that the inventory should represent the emissions conditions in the local area associated with the
nonattainment problem. Given this purpose, situations may arise where the most applicable
available inventory year is not one of these 3 ambient-data years.
Two examples of such situations are offered here. In the first, a newer inventory may be advisable.
Because of the time that it takes for designations to occur and the period allowed for SIP submittal, a
significant increase in emissions could occur between the designation period and the time SIPs are
submitted. This has occurred in some areas due to oil and gas development, for example. In this case, it
may be more appropriate to use an inventory year that is newer than one of the 3 years of data used for
designations, to represent the higher emissions or different source mix associated with a more recent
year. Another example is that in an area where mobile sources are a small part of the emissions
inventory, emissions may not have changed much from a year prior to the 3-year period of ambient
data. In this case, an available inventory prior to one of the 3 years may sufficiently represent the
inventory. In both cases, the air agency preparing the SIP should explain these issues as part of its
justification for not using one of the 3 years.
The inventory baseline year is the same year used for the RFP baseline year. In addition, states must
meet quantitative milestone requirements as described in 40 CFR 51.1013(a). Within the 3 years
available for choosing a base year, a large source of emissions may be reduced. If a state chooses a year
after those reductions have occurred, then the state would not get the benefit of those reductions in
meeting their RFP and quantitative milestone requirements. In addition, if those emissions were a key
part of the nonattainment problem, then the choice of the later year would not follow the guidance that
the year selected be representative of the conditions leading to nonattainment.
The considerations for availability of emissions for PM2.5 planning inventories are the same as for ozone
as described in the previous subsection.
Regional Haze Inventory Years
As described in Section 2.3.3, the year of the baseline inventory has previously been set. For the initial
Regional Haze SIPs, the year used should have fallen between and be inclusive of the 2000 to 2004
baseline period. Most areas used 2002 as their baseline inventory year in their initial SIPs based on an
OAQPS memo about this issue.
The RHR addresses what states must do for subsequent planning periods, and requires that states create
and use a recent year inventory as part of their SIP analysis. As described in Section 2.3.3 above, the 10-
year SIP revisions and the progress reports in between SIP revisions include specific language regarding
the inventory year to use for developing those plans, which we address here in more detail.
For the progress reports, an inventory is used to compare back to the most recent previous inventory
from the previous SIP revision. The year for this comparison is the most recent year for which the state
has submitted emission inventory information to the EPA in compliance with the triennial reporting
requirements of the AERR. In addition, the rule 308(g)(4) states that: "With respect to sources that
61
-------
report directly to a centralized emissions data system operated by the Administrator, the analysis must
extend through the most recent year for which the Administrator has provided a State-level summary of
such reported data or an internet-based tool by which the state may obtain such a summary as of a date
6 months preceding the required date of the progress report." This latter requirement refers (at least) to
the direct reporting of the CEM data for EGU sources. Thus, the emissions year(s) for the progress report
tracking analysis would include the most recent NEI year that had been reported by the state and any
newer EGU (or other) data available through direct facility reporting.
For the 10-year SIP revisions, an inventory is used as a starting point to examine future emissions
reductions and assess expected progress on visibility improvements. Similar to the progress report
requirement, the rule at 40 CFR 51.308(f)(2)(iii) would require this starting year to be "at least as recent
as the year for which the state has submitted emission inventory information to the EPA in compliance
with the triennial reporting requirements of the AERR. However, if a State has made a submission for a
new inventory year to meet the requirements of subpart A in the period 12 months prior to submission
of the SIP, the State may use the inventory year of its prior submission." Section 4.1.4 provides more
information on considerations for choosing an alternative year.
4.1.4 Considerations for Choosing ;rnative Baseline Year
For both ozone and PM2.5 SIPs, the governing implementation rules allow states to choose a different
year than one of the presumptive years as described in the subsections above. In general, states seeking
to use newer inventory years will be easily able to justify that choice on the basis of improvement in
meeting the CAA requirement for a "current" inventory (see CAA section 172(c)(3) for PM2.5 and section
182(a)(1) for ozone). It is presumed that because of ongoing control programs for mobile and stationary
sources, which get additional reductions each year, more recent year inventories will reflect lower
emissions levels.
Another common reason for choosing a different base year includes a desire to have the base year for
planning inventories consistent with the base year used for modeling inventories. The modeling base
year is determined in part by meteorology that is conducive to formation of ambient levels of a NAAQS
pollutant that are above the NAAQS. Other reasons may also exist for choosing a different year, and
these should be provided as part of SIP documentation. Ideally, any such decisions would be made in the
planning phase of SIP development, in consultation with the Regional office.
4,2 Pollutants and Poll cursors to Include in Inventories
This section describes the pollutants and precursors that should be included in the planning inventories
for the ozone, PM2.5, and regional haze rules. Pollutants for ROP/RFP inventories are also described in
Section 3.4.2. Because many sources emit more than one of the precursor pollutants, and because the
precursor pollutants have the potential to be transported across state boundaries, the EPA encourages
air agencies to develop inventories for all pollutants, where possible, to support integrated, regional-
scale modeling and control strategy development for ozone, PM2.5, and regional haze.
Section 172(c)(3) of the CAA requires "the relevant pollutant or pollutants" to be included in inventories
for SIPs. The ozone, PM2.5 and regional haze implementation rules include specific language regarding
62
-------
the pollutants and precursors to be included, and those regulatory requirements supersede any
guidance provided in this section.
For the 8-hour ozone NAAQS, the pollutants to be inventoried are VOC and NOx. The ROP/RFP baseline
NAA inventory for ozone should also include VOC and NOx. For the PM2.5 NAAQS, the pollutants to be
inventoried are primary emissions (including both the filterable and condensable portions) of PM10 and
PM2.5, and emissions of S02, NH3, VOC, and NOx. The EPA has finalized the requirement that PM10
emissions are to be reported with the PM2.5 inventories because PM10 emissions are often used as the
basis for calculating PM2.5 emissions. For regional haze, the pollutants to be inventoried include all of the
pollutants and precursor pollutants identified for ozone and PM2.5.
4,2,1 Condensable PM Emissions
As mentioned previously, the AERR identifies reporting requirements for the implementation rules. In
the past, a source of confusion has been the AERR requirement for reporting primary PM10 and primary
PM2.5 emissions, which states, "As applicable, also report filterable and condensable components." The
AERR also defines condensable PM as:
Material that is vapor phase at stack conditions but which condenses and/or reacts upon
cooling and dilution in the ambient air to form solid or liquid PM immediately after
discharge from the stack. Note that all condensable PM, if present from a source, is
typically in the PM2.5 size fraction and, therefore, all of it is a component of both primary
PM2.5 and primary PMw.
While the wording in the AERR is clear that the EPA expects reporting of all components of primary PM
including condensable portion (and as individual components), the rule does not clarify "as applicable"
by identifying those sources that are expected to have condensable PM. To clarify this issue, Table 15
provides a list of source types for which condensable PM is expected by the AERR.
Table 15: Source types expected to include condensable PM
EIS Sectors (and subsectors)
Subsectors and
Degree of
Condensable
Data
Categories
Commercial Cooking
High
Nonpoint
Fuel Combustion - Electric Generating Utilities - all fuel types
Low, except Gas
Turbines are
negligible
Point
Fuel Combustion - Industrial Boilers - all fuel types
Point
Nonpoint
Fuel Combustion - Commercial/Institutional - all fuel types
Moderate and
depends on
control technology
Point
Nonpoint
Fuel Combustion - Residential - Oil, Natural Gas, Other
Significant
Nonpoint
Industrial Processes - Cement Manufacturing
(Kilns)
Low
Point
Industrial Processes - Chemical Manufacturing
(Fuel Fired Process Heaters, Carbon Black Fuel Fired Furnaces
and Dryers, Charcoal Manufacturing, Fuel Fired Ore Calciners,
Low - Moderate,
depends on
industry
Point
63
-------
Subsectors and
Degree of
Data
EIS Sectors (and subsectors)
Condensable
Categories
Ammonia Production Fuel Fired Reformers, Fuel Fired
Incinerators, other fuel fired equipment)
Industrial Processes - Ferrous Metals
(Silicon Metal Electric Smelting Furnaces, Electric Arc
Primary metals:
Furnaces, Electric Induction Furnaces, Basic Oxygen Furnaces,
High, others Low -
Point
Iron & Steel Pouring/Casting, Foundry Cupolas, fuel fired
Moderate
equipment)
Industrial Processes - NEC
(Brick Manufacturing Kilns, Lime Manufacturing Kilns, Glass
Coke Ovens: High,
others Low
Manufacturing Furnaces, Mineral Wool Blow Chambers, Fuel
Point
Fired Process Heaters, Coke Manufacturing Ovens, other fuel
fired equipment)
Industrial Processes - Non-ferrous Metals
Primary metals:
High, Asphalt
electrode
facilities: High,
Drying: High;
Others: Low
(Aluminum Ore Electro-reduction, Smelting Furnaces, Zinc
Electrothermal Furnaces, Pouring/Casting,
Point
Rolling/Drawing/Extruding, Fuel Fired Process Heaters, other
fuel fired equipment)
Industrial Processes - Oil & Gas Production
(Gas Well Dehydrators, Gas & Oil Well Heaters, Fuel Fired
High
Point
Compressor Engines, Heater Treaters, Process Heaters,
Nonpoint
Steam Generators)
Industrial Processes - Petroleum Refineries
(Process Heaters, Catalytic Cracking Units, Coke Calciners,
Low
Point
Flares, Process Gas Incinerators, other fuel combustion)
Industrial Processes - Pulp & Paper
(Recovery Furnaces, Lime Kilns, Paper Machine/Pulp Dryer,
Sawmill Sanding: Cyclone Exhaust, Particleboard Cooler or
Moderate - High
Point
Dryer, Smelt Dissolving Tank, Particleboard Presses)
Waste Disposal
(Landfill Flares, Other Landfill Gas Waste Gas Combustion,
Low - Moderate
Point
Incinerators)
For stationary point and nonpoint sources, the EPA recognizes that emission factors for condensable PM
are limited, as well as emission factors for many industrial sources of filterable PM2.5- Resource
constraints for many years have made it infeasible for the EPA to conduct sufficient testing programs to
populate all of the necessary emission factors. Nevertheless, state and local air agencies and their
facilities are responsible for estimating emissions from sources required by the AERR.
In response to this issue, the EPA continues to work to help state and local air agencies meet their
responsibilities. The EPA has devised a mechanism to more readily collect source test data for use in
developing improved emission factors. More information on this approach can be found at
https://www.epa.gov/electronic-reporting-air-emissions/electronic-reporting-tool-ert. The EPA invites
64
-------
industry, state, local and tribal agencies to provide source test data they may collect during the
preparation of PM inventories to support the development of emission factors.
Additionally, the EPA has used available existing emission factors to develop a "PM augmentation tool"
(see https://www.epa.gov/air-emissions-inventories/pm-augmentation). The tool uses ratios of the
most applicable available emission factors to calculate one PM component from another. In the absence
of improved PM condensable emission factors or facility-specific source data, inventory developers can
use this tool to calculate PM condensable emissions from filterable PM2.5 or total PM2.5 emissions.
Condensable PM emissions are not an issue associated with mobile sources. The MOVES model used for
these sectors produces primary emission estimates for PM, including particle components (e.g.,
elemental carbon and organic carbon) and gaseous hydrocarbons (e.g., VOC, non-methane organic
gases). These pollutant metrics would include both "filterable" and "condensable" PM. These are
currently not separable in MOVES, and for SIP inventory purposes, they can be reported as they are
output from MOVES without additional modification.
4.2.2 Regulatory Definition of VOC
The EPA's current regulatory definition of VOC (40 CFR 51.100(s)) excludes constituents considered to be
negligibly reactive in atmospheric photochemistry. These include methane, ethane, methylene chloride,
1,1,1-trichloroethane (TCA), several Freon compounds, acetone, perchloroethylene, and others. States
should use the CFR reference above to find the complete list. Over time, additional compounds may be
exempted from this VOC definition. The exempt compounds are considered negligibly reactive, although
some can influence the formation of ozone when present in sufficient amounts; for example, methane
impacts on ozone have been observed in areas of high methane emissions associated with oil and gas
development. The emission factors used to estimate organic emissions generally account for
nonmethane hydrocarbons (NMHCs).
For SIP purposes, state, local and tribal agencies should report VOC as defined by the EPA (40 CFR
51.100(s)). This is the case regardless of the treatment of methane for modeling inventories, which is
addressed in more detail below.
4.2.3 Treatment of Methane
The treatment of methane in emissions inventories may be relevant for emissions used in modeled
attainment demonstrations. Methane is not expected in other planning inventories associated with SIPs.
This section is included to explain the context for which methane emissions can be important and to
specifically note its possible relationship to the VOC emissions that are associated with SIP planning
inventories. In some cases, air agencies interested in considering non-VOC-based estimates of methane
in their modeling inventories may optionally choose to consider methane as part of non-modeling
inventories as well.
The air quality models used for modeled attainment demonstrations use methane as one of the
modeled pollutants. However, these models have typically used a fixed concentration of methane that
does not change. Thus, typical applications have not used methane emissions from emissions
inventories.
65
-------
Unlike the air quality models, the emissions processing steps to prepare emissions inventories for use in
air quality models do compute methane emissions. Methane has traditionally been calculated by
applying a factor to the VOC emissions to calculate total organic gases (TOG). These TOG emissions are
then split during emissions processing into "model species" that include methane. With the low
reactivity of methane and the absence of its use in air quality modeling for ozone, the VOC-to-TOG
factors and the methane part of the speciation factors were not given much scrutiny. Previous versions
of this guidance encouraged inventory developers to simply use the default VOC-to-TOG factors that
were associated with the speciation profiles without further analysis, since the methane emissions were
not used. This approach may still be appropriate for most or all NAAs for their modeled attainment
demonstration.
In limited situations, air agencies may want to investigate the extent that methane emissions play a role
in ozone episodes. Hence, additional efforts going beyond the use of the default VOC-to-TOG profiles
and default methane fractions from the speciation profiles may develop over time. In this case, it may
be of interest to include methane emissions inputs to air quality models to achieve sufficient model
performance for a modeled attainment demonstration. Approaches can include incorporating methane
emissions from other sources into emissions inventories used for modeled air quality. For example, the
MOVES model estimates methane emissions directly, so there would be no need to use VOC speciation
to estimate those emissions. Air agencies considering methane as a contributing factor should consult
with their EPA Regional office to discuss the best sources for such data as this inventory development
area is undergoing many changes at this time. Furthermore, more specific VOC-to-TOG factors and
speciation profiles may be useful, particularly for analysis related to areas of very high methane
emissions. These issues are all associated with modeled attainment demonstrations, and do not affect
the VOC emissions reported as part of the emissions inventory SIP.
4,3 Identification of Priority Sources
The Ozone Implementation Rule and PM2.5 SIP Requirements Rule refer to the AERR requirements for
creating the inventories used for SIPs. This includes the AERR requirement that inventories must include
all sources of emissions. Thus, inventories used for SIPs are expected to be complete, with emissions
from all sources characterized in some way. In addition, to better understand routine emissions sources,
emissions from known or quantifiable fugitive sources such as leaking tanks or valves should be included
in inventories when these emissions can be quantified in both the base and future years and because
their reductions could help in achieving attainment. These emissions should not be excluded based on
an invalid assumption that they are "malfunctions." In addition, startup/shutdown emissions should be
included in inventories (see Section 4.8.1).
While completeness is required, there is no requirement that all parts of an emissions inventory be
completed with the same degree of rigor. Because emissions inventory development is time consuming
and expensive, it is wise for inventory developers first to focus their resources on the sources of
emissions making up the bulk of the pollutants relevant for the NAAQS in question. To the extent that an
inventory is to be used for multiple NAAQS or for multi-pollutant analysis, then the larger sources
important for all relevant pollutants should be characterized with as much technical rigor as is feasible.
66
-------
Fortunately, existing inventories can help states guide their initial understanding of the sources of
pollutants in a particular area. Air agencies can rely on the latest NEI or previously compiled SIP
inventories to guide further development. The NEI is commonly used as a starting point for state and
regional planning purposes for NAAQS implementation and regional haze planning. The EIS can help by
providing emissions summaries for counties or groups of counties and for emissions sectors or SCCs (See
Appendix A).
For additional information, the Particulate Emissions volume of the El IP technical report (see
https://www.epa.gov/air-emissions-inventories/volume-9-particulate-emissions) provides thorough
recommendations for identifying sources of PM emissions. These recommendations can be useful for
other pollutants as well.
Starting with an existing inventory poses some risks and challenges. Here we try to address the most
common issues associated with SIP inventories when starting with existing inventories:
1. How do I know which sources in an existing inventory should get further review?
The EPA suggests the following considerations in deciding which sources to be reviewed:
a. The largest contributors to the pollutants of interest (i.e., the pollutants contributing to
ozone and/or PM2.5 formation) using some coarse groupings, such as the EIS Sectors,
Tier 2, or Tier 3 groups.
b. Sources within the NAA (or contributing to it in the case of modeling attainment
demonstration inventories). Special considerations to sources closest to and upwind of
violating monitors may deserve special attention.
c. Knowledge about the growth or retraction of a particular sector of the economy. Oil and
gas production is a recent example of a quickly growing industrial source of emissions.
d. Considering whether the proportion of emissions associated with a particular source is
likely to have an impact on the outcome of decisions. Emissions methods/changes for
sectors that are likely to have a large impact on the planning process should get higher
priority.
e. Assessing whether the approaches used in the preparation of SIP emissions are updated
with current approaches, including appropriate emission factors. This is particularly
important for source categories for which there is large uncertainty and those with
ongoing research.
2. What if the original inventory does not use good approaches?
The EPA recommends starting with inventories that have available and clear documentation. If
an inventory developer cannot figure out how the emissions were estimated, then it will be
difficult to build upon those results.
With documentation, inventory developers should be able to evaluate the quality of approaches
used. As part of the NEI program, the EPA publishes updated methods for many nonpoint
emissions categories. The highest quality point source emissions estimates are based on facility-
specific testing, and where not available, current emission factors. Mobile emissions should be
67
-------
based on the latest mobile models, using as many location-specific inputs as possible.
If it is determined that a starting inventory does not use the latest methods, it is advisable to
reassess those emissions with newer methods (or other available inventories) prior to deciding if
that category requires even further work.
3. If I need to make updates, where are the latest approaches available?
The EPA's NEI program is a good source of the latest approaches and approaches in
development. In many cases, multi-jurisdictional organizations also have updated approaches
for various emissions sectors. The EIIP volumes still have a lot of valuable information on
emissions, though some sections are outdated (see https://www.epa.gov/air-emissions-
inventories/emission-inventory-improvement-program-eiip). Approaches have historically
included AP-42 and WebFire emission factors, though these resources are outdated for some
sectors. For more information on updated emission factors, see Section 4.6.
4. What do I do if the NEI approach has added PM or other emissions that I did not report, and I
think that they are invalid?
The EPA commonly augments PM emissions provided by states in the NEI because many states
do not report total PM2.5 and neglect the condensable portion of PM emissions (see also Section
4.2.1). Consistent with the requirements for complete SIP inventories, the EPA works to create a
complete and accurate NEI. If emissions have been added erroneously, states should make
those improvements, but should also let the NEI program staff know that there may be a flaw in
their methods and provide a better source of information. Just because a source has not
reported condensable PM emissions does not mean that they do not exist or should not have
been characterized. In some cases, states may need to go back to sources to clarify that
condensable PM emissions must be provided to the state agencies.
Another source for added EPA criteria pollutant and precursor emissions is NH3 for point sources
from the TRI. Since these are emissions that the facility submits to the EPA through the TRI
program, they should potentially be considered as a source of information when identifying
sources of NH3.
Only the facility total emissions are available from the TRI program (with stack and fugitive
emissions denoted separately). This is not generally enough detail for a SIP inventory. For SIP
inventories for PM2.5, air agencies would need to follow up with their Regional office to develop
more refined information about the facility NH3 so that NH3 emissions could be included in the
SIP inventories. If the source is a "point source" based on the AERR thresholds for other
pollutants, then including NH3 emissions as part of the point source emissions is required by the
AERR and, therefore, in PM2.5 SIP inventories in accordance with the PM2.5 SIP Requirements
Rule. If the source is not a point source based on thresholds of other pollutants, then the AERR
NH3 threshold of 100 tpy potential to emit would apply in determining if the facility should be
reported as a point source. Otherwise, the emissions should be included as a nonpoint source in
68
-------
the SIP inventory. If a state air agency believes that these emissions are erroneous or
unimportant, some discussion of these sources should be included in their SIP as to why the NH3
emissions as identified by TRI are not being included in a SIP inventory in some form. For
incorrect TRI emissions, states should consider coordinating with the facility and the EPA's TRI
program to improve the emission estimates.
4,4 Spatial Extent
The spatial extent of an inventory refers to counties, states, or other geographic regions covered by the
emissions data. As described in previous sections, inventories included in or supporting SIPs are either
for the NAA area only (for ozone and PM2.5) or are for the entire state (for regional haze and ozone or
PM2.5 when modeled attainment demonstrations are needed). Modeled attainment demonstrations are
likely to use emissions from a regional multi-state area. While it is not a SIP requirement to submit an
inventory for other states, the emissions are needed to do the modeling.
A number of complicating issues related to spatial extent arise with inventories. Below, more
information is provided on partial counties, when point-source locations are needed for emissions that
have traditionally been nonpoint sources, shapefile-based inventories, and link-based mobile emissions.
lonattainment Areas with Partial Counties
In some cases, NAAs do not follow county boundaries. Since most inventories and many underlying
datasets are available at the county level, this poses a challenge for emissions calculations, in particular
for nonroad and mobile sources.
Air agencies may wish to use full-county emissions in meeting some SIP requirements, despite the
designation of partial-county NAAs. The EPA has considered the advisability of using whole-county
emissions to represent partial county emissions and has determined that this is not possible. In
particular, using whole-county emissions would pose a challenge for using these inventories for
ROP/RFP analyses, which can only include reductions within the NAA. Thus, areas designated as partial
counties should estimate those partial county emissions using one of the approaches described here.
For point sources, Geographic Information System (GIS) technology enables straightforward selection of
point sources into a partial-county NAA by intersecting a dataset of point source latitude/longitude
coordinates with a shapefile of the NAA boundaries.
For nonpoint and mobile sources, available inventories and methods are often designed to create
county totals, making partial counties much more difficult. A variety of acceptable approaches exist for
calculating partial county emissions, though the particular approaches chosen should be agreed to as
part of developing the plan for the SIP planning inventory as described in Section 2.2.
1. Custom calculations: Air agencies could develop a custom approach to estimate emissions using
activity data estimated for the partial county area. Since many source categories need to be
estimated, this could be a very labor intensive approach, or could be used for just key categories
where other approaches are insufficient.
69
-------
2. Shapefile-based inventories: See Section 4.4.3.
3. Use of spatial allocation surrogates: Some sources of data exist at sub-county resolution,
including road networks, travel demand model results (see also Section 4.4.4), and census data
that are available at census tracts or blocks. When the emissions activity can be reasonably
assumed to follow an allocation method based on one such surrogate dataset, GIS techniques
can be followed to apportion a county emissions estimate to a partial county NAA based on the
patterns of these sub-county data.
4. Use of gridded modeling data: Modeled attainment demonstrations rely on air quality modeling
that uses gridded emissions data, which could be resolved at a sub-county resolution. Gridding
resolutions of 12-km, 4-km, or even 1-km have been used. Similar to what was described above
in the previous approach, emissions processing relies on spatial surrogates to apportion county-
level data to grid cells needed for modeling. Rather than take a custom spatial surrogate
approach, these gridded data may be sufficient to apportion county-level data to a partial-
county NAA using GIS overlay techniques. Such an approach would sum the parts of the gridded
data intersecting the partial county area to provide the partial county area inventory. The
primary difference in this approach compared to the previous approach is the intermediate step
of using the gridded data. This could be appealing since those gridded data may already be
available from modeling done for attainment demonstrations.
5. Expert judgment: When there is no appropriate sub-county data surrogate available, when air
quality modeling is not part of the SIP process, or when a category is relatively unimportant to
the SIP planning process, other methods may be advisable. In this case, an air agency should
explain their rationale to the respective EPA Regional office and come to agreement on a
technically reasonable approach. Such approaches could include using the entire county
estimated emissions, assuming some fraction based on understanding of county activities, or
taking an area-proportional (uniform distribution) approach. In this latter approach, the partial
county emissions are assumed to be the total county emissions times the area of the partial
county divided by the area of the total county.
Since these partial county emissions calculations could need to be done for a variety of inventories and
for SIP revisions, considerations should be made for the repeatability of the approach so that it does not
become overwhelming to the analysis. In collaboration with the EPA Regional office, a balance needs to
be found among level of effort, technical credibility, and repeatability.
The latest spatial approaches and data are available with the latest emissions modeling platform data on
the Emissions Modeling Clearinghouse (EMCH). The El IP Volume III, Chapter 1, Section 4.3 provides
some additional information on spatial allocation; however, the examples used for spatial allocation
approaches are much too old for most applications, and more sophisticated spatial surrogate
approaches are available than those given as examples. In particular, except in rare circumstances, the
use of population data as an allocation method is outdated and not recommended.
70
-------
Point Source Locations Needed for Modeling of Traditional Nonpoint Sources
In some cases, the details needed in a modeled attainment demonstration will push inventory
developers to use point-source locations for source categories that have traditionally been nonpoint
sources. Some examples include sources of agricultural ammonia, dry cleaners, and gas stations. Rather
than creating a much larger point source inventory, one approach is to keep a source as nonpoint, but
use an emissions processing technique for modeling known as area-to-point allocation.
While this topic is primarily one for consideration in modeling inventories, it is included here because
the planning inventory often provides a starting point for the modeling inventory (for the NAA counties).
To the extent that modeling needs can drive additional detail in the planning inventories, this area-to-
point feature can allow small sources to be included as nonpoint sources in a planning inventory but still
allow point source locations to be used for modeling.
In area-to-point allocation, the emissions processor for modeling starts with county-wide estimates and
a user-provided dataset that allocates county wide emissions to discrete latitude/longitude points
(within the county). The inventory developer still must obtain the point source locations and proportion
of countywide emissions, but the point source inventory remains more focused on the largest sources.
Using an area-to-point allocation approach also avoids having to define stack parameters for inclusion of
small sources in the point source inventory, which is useful because the stack parameters are not
necessarily relevant to or available for these source categories.
Shapefile-based Inventories
The EPA collects inventories for the NEI for commercial marine and locomotives using GIS shapefile-
based inventories. Rather than including emissions as a county total, the emissions are allocated to
individual sub-county areas or lines, each with its own GIS polygon shape. This approach was developed
primarily to support fine scale and risk modeling that are done at a scale finer than most modeled
attainment demonstrations.
These inventories can be useful for computing partial county emissions for those categories using this
approach. Otherwise, it is not expected that states would need to start with the shapefile-based
inventories for SIP planning. In some cases, states may wish to use such inventories in circumstances
where the CMV and locomotive emissions spatial distribution is critical to the modeled attainment
demonstration in some way. This may be important in NAAs that contain large ports and/or are near
major coastal or inland waterways. Even in this case, use of county total inventories with appropriate
spatial allocation using other techniques will still be able to allocate emissions sufficiently for both
nonattainment area inventories (if partial counties are needed) and for modeled attainment
demonstrations.
-based On-road Mobile Emissions and Travel Demand Models
SIP planning and modeling inventories often put great focus on mobile sources through the use of
detailed emissions by roadway segments (or links) and the use of travel demand models (TDMs). These
models require their own sets of inputs, which depend on the specific TDM used. The MOVES Technical
Guidance provides general guidance on the development of MOVES inputs. Details on using TDMs for
71
-------
link-based on-road mobile emissions are available from the EIIP document "Use of Locality-Specific
Transportation Data for the Development of Source Emission Inventories"
(https://www.epa.gov/sites/production/files/2015-08/documents/iv02.pdf). An example of the use of
TDMs can be found in "Use of Travel Demand Model Data to Improve Inventories in Philadelphia"
(http://www.epa.gov/ttnchiel/conference/eil5/sessionl/cook.pdf). This possibility is simply mentioned
here for completeness and is not necessarily recommended, since these efforts are often substantial.
The approach can also help with partial county emissions estimates.
4,5 Temporal Basis
Various temporal bases may be needed for meeting different inventory requirements for SIPs. These can
be different for ozone, PM2.5, and regional haze. The emissions that should be used in planning
inventories for ozone SIPs are ozone season day emissions. The emissions that should be used in
planning inventories for PM2.5 SIPs are annual emissions for SIPs addressing the annual or 24-hour PM2.5
NAAQS or seasonal emissions for SIPs addressing the 24-hour PM2.5 NAAQS. At a state's option, an
inventory may contain both annual and average season data emissions for SIPs addressing the 24-hour
PM2.5 NAAQS. The temporal basis for emissions summaries in Regional Haze SIPs is always annual actual
emissions, but air quality modeling for regional haze should use more temporally disaggregated
emission inputs.
4,5,1 Temporal Considerations for Ozone SIPs
For ozone SIPs, ozone season day emissions (as defined in Section 2.5.2) are used for the base year
inventory for the NAA. Since other planning inventories must be consistent with the base year inventory,
inventories such as the projected attainment year inventory for the NAA and the periodic inventories
must also include ozone season day emissions. Additionally, for states that include modeled attainment
demonstrations in their SIP, statewide (or modeling domain-wide) ozone season day emissions and
statewide annual emissions may also be important for Regional office evaluation and approval of the
modeled attainment demonstration.
Ozone season day emissions are defined in Section 2.5.2. As explained in the definition, states must
select the representative months and work week days to include in the calculation of the ozone season
day emissions. The temporal basis for these emissions should be representative of the conditions
leading to nonattainment, as recommended by the state and approved by the EPA prior to finalizing the
emissions approach to be used. The EPA acknowledges that there could be instances where ozone
values are elevated above the level of the NAAQS only in non-summer months or on weekend days,
including when winter emissions or weekend emissions dominate the episode. In recognition of this
situation, the terminology used has changed from "summer day emissions" to "ozone season
emissions." The general expectation of high emissions activity on weekdays is the basis for the weekday
specification in the definition of ozone season day emissions. However, in the event that weekend
emissions were to be higher than weekday emissions, the state should consider whether weekday
emissions are sufficiently representative of the nonattainment problem. Since the goal of the definition
of ozone season day emissions is representativeness of the emissions contributing to the ozone
nonattainment problem, if including the weekend emissions resulted in a better representation of
72
-------
emissions, then states may be able to justify including weekend emissions in coordination with the EPA
Regional office.
In general, to calculate ozone season day emissions for any inventory, reporting agencies should
estimate emissions for a specific period (month, week, selected days, or a day) where emissions (and
other factors) resulted in ozone values above the level of the NAAQS leading to nonattainment. The
preference should be given for weekdays as previously described, unless there is a reason to use other
days. An ozone season day emissions value should be calculated by summing the emissions on those
days and dividing by the number of days included in the sum.
For mobile source emissions, mobile source emissions models can produce output for an average
weekday or weekend day. Consult the EPA's MOVES Technical Guidance. This guidance provides
numerous recommendations including how to determine temperatures to use in calculating emissions
for base year and attainment year NAA inventories. The most current version of this document is
available on the MOVES website (https://www.epa.gov/moves) at the link that directs to the current
version of MOVES.
As noted in the preamble for the Ozone Implementation Rule, HEDD periods could have higher
emissions than during other periods, which could contribute to ozone formation in a particular
nonattainment area. Additional contributing factors to HEDD-period impacts include location and timing
of the HEDD-related emissions, amounts of emissions from other sources during the HEDD period, and
meteorology. If an air agency determines that emissions increases (within a NAA) associated with HEDD
periods significantly contribute to ozone nonattainment, then the air agency should include those higher
emissions in the planning inventories: the base year inventory for the NAA (which also serves as the
ROP/RFP baseline), and the attainment projected inventory for the NAA. As described above, in
consultation with the EPA Regional office, air agencies should select periods to use for defining the
appropriate days to include in the calculation of ozone season day emissions that are representative of
the conditions leading to nonattainment. Thus, air agencies that have determined HEDD periods are
important should select such periods as part of their ozone season day emissions estimation approach.
As a result, air agencies could estimate HEDD emissions in a way that represents an appropriate portion
of their NAA emissions inventories. More information on how to estimate HEDD-related emissions is
provided in Section 4.8.1.
4,5,2 Temporal Considerations for PM2.5 SIPs
The emissions must be reported as annual total emissions, average-season-day emissions, or both, as
appropriate for the relevant PM2.5 NAAQS. The rationale for the type(s) of emissions provided must be
included as part of the attainment plan. When seasonal emissions are included, the rationale for the
seasonal period must also be included as part of the attainment plan. A discussion of the EPA's rationale
for including the option of seasonal or annual inventories is provided in Section IV.B.2.d of the PM2.5
Implementation Rule preamble. The regulatory text for this requirement is also provided at 40 CFR
51.1008(a)(l)(iii).
73
-------
In the case of the 24-hour NAAQS, the form of the NAAQS is based upon monitored values on particular
days with high levels of ambient PM2.5, and in some nonattainment areas, those days may occur only
during a distinct and definable season of the year. For the PM2.5 NAAQS, states can meet the inventory
requirement with different combinations of temporal resolutions for the emissions. For the annual
standard, annual emissions must be submitted. For the 24-hour standard, states must submit either an
annual or an average-season-day inventory and optionally may submit both. For a nonattainment area
for both the annual and 24-hour standard, states can meet the inventory requirement with only an
annual inventory or with both an annual and average-season-day inventory.
In contrast with the annual PM2.5 NAAQS, the 24-hour PM2.5 NAAQS is designed to protect against peak
exposures. Thus, for the 24-hour PM2.5 NAAQS, there are circumstances in which the EPA believes that a
seasonal emissions inventory alone may be sufficient for attainment planning purposes. 40 CFR
51.1008(a)(l)(iii) allows states to use seasonal inventories for attainment plan development for attaining
the 24-hour PM2.5 standard in areas that are designated nonattainment for only the 24-hour standard.
Use of a seasonal emissions inventory will also be appropriate only if the monitored violations of the
24-hour PM2.5 NAAQS in the area occur during an identifiable season. In the event that it is appropriate
to rely solely on a seasonal emissions inventory, the state should confer with the EPA concerning the
exact length of the season and the start and stop dates of the season. The duration and start and stop
dates of the season will be an important component of the attainment plan and must be approved by
the EPA along with other elements of the attainment plan for a given nonattainment area. Further,
seasonal inventories must use average-season-day emissions values for this purpose, as defined by 40
CFR 51.1000. The nature of some seasonal PM2.5 emissions sources (e.g., residential wood combustion)
does not allow for only weekday emissions to be included in the inventory; therefore, all days must be
included. The state would need to explain the rationale for the duration of the season used for the
inventory as part of the attainment plan submission. To justify the use of a seasonal emissions
inventory, the state must demonstrate why a seasonal emissions inventory is appropriate for the
particular PM2.5 nonattainment area in question.
nission Factors for SIP Emissions Inventories
As expressed in 40 CFR 51.114(a), inventories "must be based on measured emissions or, where
measured emissions are not available, documented emission factors." Given that the preference is
measured data, the intent of the regulations is for emissions to be specific to the source of the
emissions. To ensure that emissions estimates that use factors are high quality, the emission factors for
SIP development should come from credible sources. The possible sources necessarily differ for
stationary, on-road mobile, and nonroad sources. The mobile sources with EPA models are the most
straightforward. For all states but California, on-road mobile emissions should be estimated with the
latest EPA on-road model, MOVES. For California, the most recent EPA-approved version of EMFAC
should be used. For stationary sources and remaining nonroad mobile categories, there are several
approaches that can be acceptable, with the guiding principle of using credible sources of data. Where
emissions inventory SIPs are to be approved by the EPA, inventory developers should work with the
appropriate EPA Regional office to make sure the emission factors are acceptable.
74
-------
Air agencies should pay close attention to the methods that they are using to calculate emissions to
ensure updated approaches are used. It would be difficult for an EPA Regional office to approve an
inventory that used an outdated method for an important part of the inventory if an updated method
was available at the start of the SIP planning inventory development effort. Since the EPA could have
multiple approaches posted for the same sector and some of those approaches could be outdated, air
agencies should not assume that all EPA-posted methods are equivalent and acceptable. For example,
some emission factors from AP-42 for nonpoint sources are outdated when compared to nonpoint
methods and factors used for the NEI. The EPA recommends that air agencies refer to the latest NEI
development web pages and contact Regional offices to confirm that their planned approaches are up
to date. The IPP process described above can be helpful in this process.
For most stationary point sources, air agencies have rules in place that require the submission of point
source data to the state for air planning purposes. Much of these data are then submitted to the EPA as
part of the NEI and used as the basis for point source inventories in SIPs. These state rules may have
their own requirements about emission factors and source tests to be used; where possible, these rules
should follow this guidance.
The EPA recommends selecting emission factors based on the hierarchy of emission factor sources
shown in Table 16, which is consistent with the requirement in 40 CFR 51.114(a) that emission
inventories for SIPs use measured data or emissions factors if measured data are not available. The
higher up on this hierarchy the emission factor chosen, the better the quality of the emissions estimate.
Based on the emission factor approach used, the states should also set the required "emission
calculation method code" (see Section 2.5.14) for inclusion in their inventory data.
Table 16: Hierarchy of emission factors for use in point source emissions inventories
Rank
Emission Factor Approach
Emission factor
method code to use
1
Continuous emissions monitoring system
1
2
Facility/process-specific emission factor. Average of
representative stack tests (on one stack) downstream of controls
(or if no controls).
4
3
Single representative facility/process-specific stack test
downstream of controls (or if no controls).
4
4
Facility/process-specific uncontrolled emission factor, plus control
efficiency. Average of representative stack tests (on one stack)
upstream of controls, adjusted based on expected control
efficiency.
24
5
Single representative facility/process-specific stack test upstream
of controls, adjusted based on expected control efficiency.
24
6
Site-specific emission factor. Average of representative stack tests
across multiple stacks or processes, downstream of controls (or if
no controls).
10
7
Site-specific uncontrolled emission factor, plus control efficiency.
Average of representative stack tests across multiple stacks or
processes, upstream of controls, adjusted based on expected
30
75
-------
Emission factor
Rank
Emission Factor Approach
method code to use
control efficiency. Only makes sense when all processes
measured are fed to the same type/combination of control
devices.
8
Material balance
3
9
Manufacturer specification
7
Best available other emission factor for uncontrolled or including
controls, which is one of:
EPA emission factor
8
10
S/L/T emission factor
9
Vendor emission factor
11
Trade group emission factor
12
Other emission factor
13
Best available pre-control emission factor, plus control efficiency,
which is one of:
EPA emission factor, plus control efficiency
28
11
S/L/T emission factor, plus control efficiency
29
Vendor emission factor, plus control efficiency
31
Trade group emission factor, plus control efficiency
32
Other emission factor, plus control efficiency
33
12
Engineering judgement
2
Emission factor with best available speciation profile, one of:
13
EPA speciation profile
5
S/L/T speciation profile - e.g., to calculate metals emissions
C.
from filterable PM2.5 or volatile HAP from VOC.
ventory Data Codes
Building inventories includes the use of various codes, such as SCCs, state/county FIPS, tribal, control
measure, NAICS, pollutant and emission method codes. A full accounting of all possible codes is
available as part of the EIS "Reporting Code Tables" available to EIS users using the "Reporting Code
Tables" link at the left-hand side. There are more than 60 separate lists of codes that define all of the
required and optional coded data elements based on the AERR. As we have noted above, the AERR is
relevant as the rule that codifies the correct development of all inventories submitted to the EPA. The
SCCs are also searchable at: https://ofmpub.epa.gov/sccsearch/.
:odes to Use for Nonattainment Area Inventories
Air agencies developing inventories for SIP planning should use valid EIS codes in all of their inventories.
Since codes change over time, the EPA has adopted the approach in the EIS to "retire" codes when no
longer needed and to add codes when needed. Codes are generally retired in coordination with NEI
development cycles, ideally retiring codes only when plenty of notice has been provided to states in
advance of an NEI inventory submission period. As such, codes that are retired may have already been
used as part of a SIP development process, and that is okay. This is the advantage to retiring codes
rather than removing them. Conversely, the EPA may not be able to retire a code until an appropriate
76
-------
point in the triennial NEI cycle. If an air agency wishes to submit a SlP-related inventory to the EIS for the
NEI or another purpose, it will need to use valid (and not retired) codes for the year being submitted.
How to Request New Codes or Code Retirements
In general, if a new code is needed for compiling an inventory, then an air agency staff member should
contact the EPA to request a new code. The recommended contact method is to use the call or email
using the contact information listed for "Facility Inventory and Point Emissions" on the EIS contacts page
at https://www.epa.gov/air-emissions-inventories. Requests for codes should include which code
table(s) are affected, why existing codes are not sufficient, and a recommendation for a code and its
definition and purpose. While the EPA anticipates that code retirement requests will usually come from
the EPA rather than from external groups, a code retirement request should be made using the same
mechanisms.
The EPA most often gets requests for new SCC codes since new emissions processes arise in industries,
or processes are newly identified as having emissions associated with them. If an air agency inventory
developer cannot find an appropriate SCC code for an inventory process, a choice needs to be made to
either use a generic ("miscellaneous") SCC or to request a new SCC. For cases in which the contribution
of the emissions process to any one precursor pollutant is minimal, inventory developers may choose to
report using one of the miscellaneous SCCs. If the process is a significant source of one or more
pollutants or that process might be a part of a control program, then an SCC request is probably
warranted. The EPA staff will evaluate the SCC request to make sure that a similar code does not already
exist, consider codes that have been created but not yet activated for use in submitted inventories17,
assess the need for the SCC, and either indicate an existing appropriate SCC to use or assign a new
appropriate code.
lation of Base Year Emissions
This section describes the estimation of base year (or "recent year" in the case of regional haze)
emissions for use in planning inventories for ozone, PM2.5 and regional haze implementation.
Subsections cover point, nonpoint, on-road mobile, nonroad mobile, biogenic sources, and fires,
respectively.
4,8,1 Stationary Point
Stationary point sources can be estimated through continuous measurements and through more
traditional methods. El IP Volume II. Chapter 1. Sections 1 through 7 provide still-relevant information on
compilation and calculation of point source inventories. Readers developing point source inventories
should also use that report as a resource.
17 Some SCCs are set up for use in collecting source test data and developing emission factors. Sometime after that,
the SCCs can be used for creating and submitting inventories, depending on the availability of emission factors
and/or the interest for using those SCCs by industry and air agencies.
77
-------
Sources estimated with CEMS
The best estimate of stationary point source emissions is obtained through the use of CEMS. Several
pollutants, including NOx, S02, C02, and mercury (Hg) can be measured with CEMS. Fuel heat input on an
hourly basis is usually also be collected as part of the EPA's CEMS programs, and heat content can be
used to estimate emissions for other pollutants using emission factors (with specific test-based factors
preferred). More information and guidance on CEMS can be found at
http://www.epa.gov/airmarkets/emissions/continuous-factsheet.html.
The vast majority of CEMS-equipped sources use these measurements as part of reporting programs to
the EPA. In particular, the Acid Rain Program and N0X SIP call and other trading programs collect data as
part of the EPA's Clean Air Markets . Through that program, CEMS-based emissions are available
through the "Data and Maps" resource, including emissions summaries and files ready to be used in the
SMOKE system used for emissions modeling.
For a few sources, emissions must be reported only for part of the year. In these cases, users of the data
should be careful not to override the annual emissions estimates for such units with the partial-year
data included in the CEMS databases. This can be a particular challenge for modeling inventories, where
the more temporal resolution is useful for the time periods covered, but the other time periods must
use a different approach to apportion the remaining emissions to the rest of the year.
Other (non-CEMS) sources
El IP Volume II. Chapter 1. Section 4 provides general emission estimation procedure information. It
provides a table similar to Table 16, but also focuses on the cost associated with the different
approaches. It acknowledges that the cost incurred with collecting the more specific source test data
should be commensurate with the need for a higher data quality. Like the El IP figure, Table 16 also lists
the costlier (and better) approaches higher on the list. The EIIP volume provides guidance on each of the
various methods of computing emissions.
The remainder of EIIP Volume II includes chapters that address various combustion, manufacturing, and
production activities that are point sources. Information in these chapters can be used to estimate
ozone, PM2.5, and regional haze precursor emissions where they address the source categories of
interest. In some cases, the EIIP chapters are somewhat dated and so may not cover newer emissions
processes. In these cases, inventory developers should adapt the techniques or develop new techniques
to ensure their estimates are relevant to the sources. The EIIP point source chapters within Volume II
cover topics as follows (numbers correspond to chapter numbers):
2. Boilers
3. Hot mix asphalt plants
4. Fugitive emissions from equipment leaks
5. Wastewater collection and treatment
6. Semiconductor manufacturing
7. Surface coating operations
78
-------
8. Paint and ink manufacturing
9. Secondary metal production
10. Oil and gas production
11. Plastic products manufacturing
12. Effects of control device efficiencies and malfunctions
13. Stone mining and quarrying
14. Uncontrolled emission factors for criteria pollutants
15. Packaging and graphic arts
16. Chemical manufacturing
Each industry- or source-specific document contains a brief description; identification of emission
points; an overview of methods available for estimating emissions; example calculations for each
technique presented; a brief discussion on quality assurance (QA) and quality control (QC); and the SCCs
needed for entry of the data into a database management system. The SCCs included in each volume
apply to the process emission points, in-process fuel use, storage tank emissions, fugitive emissions, and
control device fuel (if applicable).
For NH3 sources, a report is also available in the "nonpoint" section of EIIP entitled "Estimating
Ammonia Emissions from Anthropogenic Nonagricultural Sources - Draft Final Report" (April 2004) and
covers many point source processes as well as nonpoint ones. Although a lot of research has been done
on possible new approaches to agricultural NH3 estimation as point sources, these "farm-based"
approaches have not been completed yet. At this time, the EPA augments agricultural NH3 sources only
as nonpoint sources, rather than as point sources. Readers should see Section 4.8.2 for more
information on emissions estimation from nonpoint sources.
A final resource is to use the facility-total NH3 from the TRI. For SIP planning inventory purposes, these
emissions should be apportioned to individual processes, but at least can be a starting point in the
absence of other information. TRI provides emissions by stack and fugitive, with all stack emissions
summed together and all fugitive emissions together. This level of detail would not be sufficient for a
significant NH3 source and so, while TRI can be a starting point, additional information would need to be
gathered from the facility to apportion the emissions to the processes at the facility.
Startup/shutdown emissions
The regulatory expectations for SSM emissions are included as part of 40 CFR part 52 in the EPA's "State
Implementation Plans: Response to Petition for Rulemaking; Restatement and Update of EPA's SSM
Policy Applicable to SIPs; Findings of Substantial Inadequacy; and SIP Calls to Amend Provisions Applying
Excess Emissions During Periods of Startup, Shutdown, and Malfunction" final rule (SSM rule). A fact
sheet and list of affected states is available on the website for this action.
The requirements of the Ozone and PM2.5 Implementation Rules and the Regional Haze Rule require the
reporting of historical annual-total emissions only (and in some areas "typical" seasonal and/or daily
emissions for certain pollutants), not day-to-day emissions. As noted in the preamble of the final SSM
rule:
79
-------
"Actual emissions during SSM events should be included in these annual emissions. While
data formats are available from the EPA to allow a state to segregate the total annual
emissions during SSM events from annual emissions during other types of operation,
segregation is not a requirement and few states do so" (80 FR 113, page 33950).
The key takeaway message from this guidance is that since startup/shutdown emissions are planned and
predictable, they should be included in SIP planning inventories. While the final SSM rule generalizes
that "SSM" emissions should be included, this guidance clarifies that this means startup/shutdown
rather than malfunction as further described below. In addition, states should pay particular attention to
estimating startup/shutdown emissions if these types of emissions are believed by the air agency to
have a significant contribution to the NAAQS design values.
Since malfunctions are, by nature, unpredictable and given the myriad different types of malfunctions
that can occur, malfunction emissions would be difficult to estimate and future-year malfunction events
cannot be readily predicted. Thus, states are not obligated to include malfunction emissions in the base
inventory for the NAA, ROP/RFP plans, or attainment projected inventory for the NAA. However, to the
extent that malfunctions become a regular and predictable event, then such emissions should be
quantified with regular and predictable emissions and included in emission inventories for planning
purposes. The elimination of high emissions during routinely reoccurring malfunctions could potentially
help achieve significant emissions reductions needed by a state in attaining the relevant NAAQS.
For sources with CEMs that capture emission rates at all times, the CEM will capture the
startup/shutdown emissions without taking any of the other steps described below. Without CEMs,
estimating startup/shutdown emissions becomes more complicated.
Startup/shutdown emissions can be quantified by assessing the emissions during startup or shutdown
periods using traditional emissions estimation approaches of throughput and emission factors. However,
the throughput and emission factors must be customized for the activity during these periods.
Generally, emissions rates during these periods are higher and less well characterized than a typical
operating emissions rate, and so some degree of expert judgement may be needed in assessing these
emissions. For example, if control devices are not effective or active during these periods, the emissions
during these types would not be able to reflect controlled emissions rates. If device emissions rates are
associated with an operating temperature and the emissions are higher during a warm-up period,
adjustments to the emissions during the startup period could be made.
In addition to estimating the actual emissions during startup/shutdown periods, another approach to
estimate startup/shutdown emissions is to adjust control parameters via the emissions calculation
parameters of rule effectiveness or primary capture efficiency. Using these parameters for
startup/shutdown adjustments is not their original purpose, but can be a simple way to account for the
emissions and still have a record of the routine versus startup/shutdown portions of the emissions. For
example, a lower capture efficiency or rule effectiveness can adjust the estimate of emissions to be
higher, accounting for extra emissions from startup/shutdown activities. In the absence of
measurements of throughput and/or emissions rates during startup/shutdown periods, expert judgment
80
-------
may need to be used to define the adjustment factors to use. This approach, however, can complicate
the use of the control parameters during development and documentation of control strategies since
the control parameters reflect things other than controls - thus, care should be taken when using this
approach.
Emissions associated with HEDD periods
As described in Section 4.5, when air agencies have determined that emissions associated with HEDD
periods are important for addressing their nonattainment problem, those air agencies should include
HEDD-related emissions in their planning inventories. This subsection first describes what information is
available to help assess when HEDD periods occur and to help provide activity data that could be used as
part of estimating the associated emissions. Second, this subsection describes how to consider HEDD-
related emissions in planning inventories (e.g., the base year inventory for the NAA) as compared to
those in modeling inventories.
Emissions from HEDD periods can include additional emissions from more typical EGU boilers, turbines,
and other units (collectively called peaking units), and also from combustion-based sources of
distributed generation such as large back-up generators and other engines. A key challenge in
developing emissions estimates for HEDD-periods is determining when the peaking units and distributed
generation units were operating and consequently produced emissions. In the case of peaking units,
many of these sources have CEMS that record activity and emissions levels, and these data are reported
to EPA. In the case of distributed generation units, little or no information may be readily available in
part because the energy produced by the backup generators and other engines is not measured by the
electric utilities and independent system operators (ISOs), but rather may be experienced by the utilities
and ISOs as an unmeasured demand reduction.
During HEDD periods, energy customers may use backup generators to participate in demand response
programs. Though some useful information may be available from the utilities and ISOs running those
programs, there may also be no information available. The specific rules, reporting, and frequency of
use will vary by the utility or ISO that is offering the demand response program. Demand response
program requirements vary and can include: (a) completely excluding the use of backup generators, (b)
allowing the use of backup generators and requiring some reporting, and (c) not requiring customers to
report anything about actions taken to provide the demand response. The lack of data on demand
response backup activity has been noted by others based on past analyses.1819 Where emissions from
demand response programs may be important, air agencies should first collaborate with their state
energy and public utility commission counterparts who may have established mechanisms for
information collection and sharing. Air agencies may also need to assess the demand response programs
in their area by working directly with utilities and/or ISOs.
18 White paper: "Air Quality, Electricity, and Back-up Stationary Diesel Engines in the Northeast," Northeast States
for Coordinated Air Use Management (NESCAUM), August 1, 2012.
19 See also the Mid-Atlantic Distributed Resources Initiative (MADRI) working group meeting information available
at http://sites.energetics.com/madri/meetings_2012.html.
81
-------
Starting in 2015, the National Emission Standard for Hazardous Air Pollutants (NESHAP) and New Source
Performance Standards for Reciprocating Internal Combustion Engines (RICE) (40 CFR part 60, Subparts
Mil and JJJJ, and 40 CFR part 63, Subpart ZZZZ), also called the RICE rules, now require owners/operators
of emergency engines participating in emergency demand response programs to report the dates and
times that the engines operate for emergency demand response as well as engine information such as
horsepower and model year. While the RICE rules do not require the reporting of NOx emissions or
information about non-emergency programs, they will start to help provide information useful for
understanding the activity of these back-up generators and engines, which may also be helpful for ozone
planning purposes.
While it may continue to be difficult to find data needed for assessing HEDD-related emissions, air
agencies may still include these emissions in both their modeling inventories and planning inventories.
The HEDD-related emissions included by air agencies in planning inventories should have some logical
relationship with the emissions in modeling inventories. The planning inventories will of course be
limited spatially to the NAA, while the modeling inventories will cover a larger area. Air agencies should
extract the appropriate NAAs to include in the planning inventories. Temporally, the modeling
inventories will include day- and hour-specific emissions for seasonal or episodic periods, which can
include HEDD periods, while the planning inventories will be an ozone season day emissions value. As
explained in Section 4.5, the days used to calculate the ozone season day emissions are chosen for the
particular SIP in consultation with the Regional office. Some examples of possible relationships between
the modeling inventories (episode-specific) and the planning inventories (ozone season day) are as
follows:
• Ozone season day emissions could be an average across all days of the episode-specific
emissions used in the modeled attainment demonstration.
• Ozone season day emissions could be an average of a longer episode or series of episodes
that overlap with the episodic period used for the modeled attainment demonstration.
• Ozone season day emissions could be a representative selection of the episodic-specific
emissions, such as a median emissions value or using a single day emissions value that
somehow is representative of other days.
These are not the only possible examples. Any approach chosen to determine appropriate HEDD-related
ozone season day emissions should be developed in consultation with the Regional office and should
seek to meet the overall goal for the base year inventory. Namely, the NAA inventory should be
representative of the emissions conditions within the NAA that typically lead to nonattainment
problems.
4,8,2 Stationary Nonpoint
Nonpoint sources collectively represent individual sources that have not been inventoried as specific
point or mobile sources. These individual sources treated collectively as nonpoint sources are typically
too small, numerous, or difficult to inventory using the methods for the other classes of sources.
82
-------
Stationary nonpoint sources can be estimated through numerous different approaches. Though some
approaches are outdated, the EllP Volume III. Chapter 1. Sections 1 through 6 provide relevant
information on compilation and calculation of nonpoint source inventories. Also, the Volume III
webpage contains numerous supplemental reports for specific source categories.
For many categories, more updated approaches are included in the EPA tools developed for use by air
agencies in developing inventories. The latest tools are provided as part of the materials for the latest
NEI - either as developmental materials for an ongoing NEI effort, or as supporting materials to
document a completed NEI effort.
Table 17 provides a list of emissions categories, pollutants, and available approaches documentation
and/or tools. Tools are constantly being updated so that the reader should verify if a newer version of a
tool exists by consulting the Air Emissions Inventories website and looking at the most current NEI page
for the latest tools and documentation. In addition, the EPA maintains a Sharepoint site with
developmental tools for in-progress inventories. SLT agencies may request access to the EPA Sharepoint
site by using the Contact Us link on the Air Emissions Inventories website.
Table 17: List of nonpoint categories, pollutants, and available approaches and tools
EPA-estimated
emissions source
Possible
Overlap
with
Point
Available Nonpoint Resources (Posted year)
description
Agricultural Fertilizer
2014 Fertilizer Application vl.O 22apr2016.zip (2016)
Application
Old: ag fertilizer application 2011.zip (2013)
Agricultural Fires
2014 Draft Agricultural Burning and Grass/Pasture Estimated Emissions (2016)
Agricultural Tilling
2014 Agricultural Tilling v3.1 10mar2016.zip (2016)
Animal Husbandry
2014 Ag Livestock vl.O 20mav2016.zip (2016)
Old: animal livestock emissions 2011.zip (2009)
Architectural Coatings
Solvent Tool vl 5.zip (2016)
EIIP, Volume III. Chapter 3 (1995)
Asphalt Paving
Y
2014 NPt Asphalt 18nov2015 edit03302016.zip (2016)
EIIP. Volume III. Chapter 17 (2001)
Asphalt Roofing Kettles
Y
See abstract available with the EIIP Volume III website (2000)
Autobody Refinishing
Y
Solvent Tool vl 5.zip (2016)
Aviation Gasoline
2014 Av Gas Stage 1 15nov2015.zip (2015)
2014 Av Gas Stage 2 15nov2015.zip (2015)
Backyard Barbecues
2014 Backvard BBQs 15nov2015.zip (2016)
Commercial Cooking
2014 Commercial Cooking vl.2 08mar2016.zip (2016)
Consumer & Commercial
-All personal care
products
Solvent Tool vl 5.zip (2016)
EIIP Volume III. Chapter 5 (1996)
Consumer & Commercial
Solvent Tool vl 5.zip (2016)
-All household products
EIIP Volume III. Chapter 5 (1996)
Consumer & Commercial
-All coatings and
Solvent Tool vl 5.zip (2016)
related products
83
-------
EPA-estimated
emissions source
Possible
Overlap
with
Point
Available Nonpoint Resources (Posted year)
description
Consumer & Commercial
-All adhesives and
sealants
Solvent Tool vl 5.zip (2016)
EIIP Volume III, Chapter 5 (1996)
Consumer & Commercial
-All FIFRArelated
Solvent Tool vl 5.zip (2016)
products
Degreasing
Y
Solvent Tool vl 5.zip (2016)
Dental Preparation and
Use
epa mercurv dataset 62714(2).zip (2015)
Documentation for the 1999 Base Year Nonpoint area source National
Emissions inventory for HAPs, page A-30 (2000)
Dry Cleaning
Y
Solvent Tool vl 5.zip (2016)
EIIP, Volume III, Chapter 4 (1996)
Dust - Paved Roads
2014 Road Dust v2.1 09mar2016.zip (2016)
Dust-Unpaved Roads
2014 Road Dust v2.1 09mar2016.zip (2016)
Dust - Construction
2014 Construction Dust v3.0 18feb2016.zip (2016)
Gasoline
2014 Gasoline Distribution vl.O with PT subtraction 01apr2016.zip (2016)
Distribution/Marketing
EIIP, Volume III, Chapter 11 (2001)
General Laboratory
Activities
epa mercurv dataset 62714(2).zip (2015)
Graphic Arts
Y
Solvent Tool vl 5.zip (2016)
EIIP Volume III, Chapter 7 (1996)
Human Cremation
Y
cremation 2011.zip (2013)
Industrial,
ICI Tool vl 4.zip (2015)
Commercial/Institu-
Y
Estimating Ammonia Emissions From Anthropogenic Nonagricultural Sources -
tional Fuel Combustion
Draft Final Report (April 2004)
Industrial Surface
Solvent Tool vl 5.zip (2016)
Coating
EIIP, Volume III, Chapter 8 (1997)
Lamp Breakage (Landfill
emissions)
epa mercurv dataset 62714(2).zip (2015)
Lamp (Fluorescent)
Recycling
epa mercurv dataset 62714(2).zip (2015)
Landfills
Y
EIIP, Volume III, Chapter 15 (2001)*
Mining and Quarrying
2014 Mining and Quarrving v2.3 09mar2016.zip (2016)
OIL GAS TOOL 2014 NEI EXPLORATION VI 5.zip (2016)
Oil and Gas Production
Y
OIL GAS TOOL 2014 NEI PRODUCTION VI 5 Access2007.zip (2016)
OIL GAS TOOL 2014 NEI PRODUCTION VI 5 Access2013.zip (2016)
Open Burning (Leaves,
Brush, Residential
Household Waste, Land
2014 Open Burning vl.l 03mar2016.zip (2016)
Clearing)
Other Special Purpose
Coatings
Y
Solvent Tool vl 5.zip (2016)
Pesticide Application
2014 Agricultural Pesticides v2.0 18feb2016.zip (2016)
EIIP Volume III, Chapter 9 (2001)
Portable Gas Cans
(Residential and
2014 Portable Fuel Containers 25nov2015.zip (2016)
Commercial)
84
-------
EPA-estimated
emissions source
Possible
Overlap
with
Point
Available Nonpoint Resources (Posted year)
description
Publicly Owned
Treatment Works
Y
2014 POTW nonpoint emissions 23march2016.zip (2016)
Residential Heating -
Fossil Fuels
2014 Residential Heating non-Wood vl.2 09mar2016.zip (2016)
See also abstracts available for coal, fuel oil & kerosene, gas & liquified
petroleum gas (LPG) available with the EIIP Volume III website (1999).
Residential Heating -
2014 RWC v3.0 28apr2016.zip (2016)
Wood
EIIP, Volume III, Chapter 2 (2001)
Structure Fires
EIIP, Volume III, Chapter 18 (2001)
Traffic Markings
Solvent Tool vl 5.zip (2016)
EIIP, Volume III, Chapter 14 (1997)
Vehicle Fires
See abstract available with the EIIP Volume III website (2000)
Waste Disposal
Estimating Ammonia Emissions From Anthropogenic Nonagricultural Sources -
(Incineration)
Draft Final Report (April 2004)
* Landfill emissions estimates are available in the NEI as point sources, based on the activity information submitted to the EPA
as part of the Greenhouse Gas Reporting Rule.
As indicated in Table 17 by the column "Possible Overlap with Point/' some source types can have
emissions in both the point source data category and the nonpoint one. Some categories that may have
both types of sources are denoted, and this list is based on what has been seen in the NEI. The list of
categories in this situation depend on an individual state's policies, such as what the emissions size
threshold is for a point source and whether any categories are treated as point sources despite having
low individual emissions.
To prevent overestimation of nonpoint emissions for sectors that also have point sources, it is important
to perform point/nonpoint reconciliation. This reconciliation starts by subtracting the activity associated
with the point sources from the total activity data (prior to computing the nonpoint emissions). Since
point sources often have emissions controls, it is usually not sufficient to just subtract point source
emissions from total activity-based emissions to prevent double counting. Subtracting controlled point
source emissions from uncontrolled total activity-based emissions would result in too much nonpoint
emissions. Rather, the point source activity must be determined and subtracted from total activity to get
the nonpoint activity estimate. Alternatively, uncontrolled point source emissions can be calculated and
subtracted from the total activity-based emissions estimates to derive a nonpoint emissions estimate.
Part of developing refined nonpoint emissions estimates for a SIP could involve the use of surveys to
collect data. Inventory preparers often use survey questionnaires to gather point source emissions
inventory data. The El IP Volume III. Chapter 24. written in 2000, provides guidance on developing and
conducting surveys for area source inventories. Since this document predates some of the newer online
survey techniques, the sections describing the use of the internet are outdated. As part of the
documentation for inventories, survey design and results should be kept and included in any
documentation reports or appendices.
85
-------
Oii-road Mobile
Emissions from on-road vehicles are the result of several emission processes, including the combustion
of fuel while vehicles are starting, idling, and moving, evaporation of fuel from the fuel system and
during refueling, and from brake wear and tire wear.
For all states but California, on-road mobile emissions should be estimated with the latest EPA on-road
mobile model, MOVES. MOVES estimates emissions for many combinations of vehicle classifications and
fuel types. Readers should follow the latest MOVES guidance. For California, the most recent EPA-
approved version of EMFAC should be used.
The available guidance covers using MOVES for inventory development in SIPs in states other than
California. The guidance focuses on determining what the appropriate inputs are and how MOVES
should be run to develop emissions estimates for on-road vehicles. It describes when default MOVES
inputs are appropriate for SIPs and regional conformity analysis and what factors to consider for
providing locality-specific MOVES inputs.
MOVES estimates emissions for multiple source use types (the terminology that MOVES uses to describe
vehicle type) and fuel types. MOVES also estimates emissions for different road types that represent
rural and urban unrestricted and restricted (accessed only by an on or off ramp) access roads, as well as
off-network emissions such as vehicle starts, evaporative emissions while parked, and hoteling
emissions when long-haul combination trucks are parked for long periods with the engine running under
load.
One of the key elements for use of MOVES is creating a Run Specification (RunSpec) file, which defines
the place and time period of the analysis as well as the vehicle types, road types, fuel types, and the
emission-producing processes and pollutants that will be included in the MOVES run. The MOVES
Technical Guidance describes using the MOVES graphical user interface GUI to set up a MOVES run for a
particular scale (e.g., county), time period, set of pollutants, and the vehicle types, road types, and fuel
types applicable to a particular area and time.
Another key element for use of MOVES is entering local meteorology, fleet, activity, fuel, and control
measure information using the County Data Manager. The MOVES Technical Guidance discusses when
local data should be used in place of model default data, as well as potential sources of local data.
For some input parameters, there is overlap with the inputs needed for other models, such as the
nonroad mobile models (e.g., meteorology and fuels). Efforts should be made to use the same source of
data across multiple categories when the inputs are shared. Not doing so calls into question the validity
of one or both approaches. Air agencies should explain the use of shared inputs, or reasons for not using
them, as part of the documentation provided with their inventories.
lonroad Mobile Equipment
Nonroad mobile equipment emissions result from the use of fuel in a diverse collection of vehicles and
equipment, including:
86
-------
• recreational vehicles, such as all-terrain vehicles and off-road motorcycles;
• logging equipment, such as chain saws;
• agricultural equipment, such as tractors;
• construction equipment, such as graders and back hoes;
• industrial equipment, such as forklifts and sweepers;
• residential and commercial lawn and garden equipment, such as leaf and snow blowers;
• recreational marine vessels, such as power boats.
Starting with MOVES 2014, MOVES incorporates and expands upon the capabilities of the older
NONROAD model. States are required to use MOVES for any nonroad source emissions estimates for SIP
development that is started since the October 2014 release of MOVES 2014.
MOVES nonroad capabilities include predicting emissions for all of these nonroad equipment categories
listed above. Emissions from CMVs, locomotives, and aircraft are estimated in other ways and described
in Section 4.8.5. The MOVES nonroad capabilities include more than 80 basic and 260 specific types of
nonroad equipment and further stratifies equipment types by horsepower rating. Fuel types include
gasoline, diesel, compressed natural gas (CNG), and Liquified Petroleum Gas (LPG).
The nonroad capabilities of the MOVES model estimates emissions for all criteria pollutant and
precursors from both exhaust and non-exhaust processes (diurnal, refueling spillage, vapor
displacement, hot soak, running loss, take permeation, hose permeation, and crankcase emissions).
NONROAD supports many SIP-related inventory development needs, including support for partial
counties and seasonal emissions.
Air agencies have the option of replacing default model inputs with more representative data. Common
input adjustments include equipment population, geographic allocations, and local growth rates. If
agencies make changes to default model values, the agency should submit the input files to the EPA as
well as a description of why the defaults were changed. Additional guidance on improving nonroad
estimates for lawn and garden equipment can be found in El I P. Volume IV. Chapter 3.
For some input parameters, there is overlap with the inputs needed for other models, such as the on-
road mobile models (e.g., meteorology and fuels). Efforts should be made to use the same source of
data across multiple categories when the inputs are shared. Not doing so calls into question the validity
of one or both approaches. Air agencies should explain the use of shared inputs, or reasons for not using
them, as part of the documentation provided with their inventories.
4,8,5 Other Nonroad Mobile
The MOVES model does not provide air agencies with a tool for estimating emissions for CMVs, aircraft,
and locomotives. Thus, separate methods are used for each of these emissions sectors.
For purposes of reporting data to the EPA, aircraft emissions are often reported as point sources, with
this being a requirement for reporting data to the EIS for use in the NEI. Likewise, the CMV and
locomotive data are included in the nonpoint data category in the EIS. While these sources are still
87
-------
mobile sources from the traditional standpoint of emissions categories, for data formatting and
reporting purposes in the EIS, they are treated along with these non-mobile data categories.
4,8.5.1 Aircraft
The aircraft sector includes all aircraft types used for public, private, and military purposes. This includes
four types of aircraft: commercial, air taxis (AT), general aviation (GA), and military. A critical detail
about the aircraft is whether each aircraft is turbine- or piston-driven, which allows the emissions
estimation model to assign the fuel used, jet fuel or aviation gas, respectively. The fraction of turbine-
and piston-driven aircraft is either collected or assumed for all aircraft types. The NEI data are available
for this sector for use as a starting point in SIPs, though the NEI includes military aircraft only where
state or local air agencies have submitted military aircraft emissions data.
Commercial aircraft include those used for transporting passengers, freight, or both. Commercial aircraft
tend to be larger aircraft powered with jet engines. Air taxis carry passengers, freight, or both, but
usually are smaller aircraft and operate on a more limited basis than the commercial aircraft. General
aviation includes most other aircraft used for recreational flying and personal transportation. Finally,
military aircraft are associated with military purposes, and they sometimes have activity at non-military
airports.
The national AT and GA fleet includes both jet- and piston-powered aircraft. Most of the AT and GA fleet
are made up of larger piston-powered aircraft, though smaller business jets can also be found in these
categories. Military aircraft cover a wide range of aircraft types such as training aircraft, fighter jets,
helicopters, and jet-powered and piston-powered planes of varying sizes.
Emissions at airports also include other sources that are a part of the nonroad equipment emissions
described in Section 4.8.4. These emissions come from 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. In the NEI point source
inventory, these APUs and GSE are located at the airport facilities as point sources along with the
aircraft exhaust emissions (though not as part of the aircraft sector). Emissions from APUs and from GSE
can be calculated using the EPA's NONROAD model and by the Federal Aviation Administration (FAA)
Emissions and Dispersion Modeling System (EDMS). If the air agency chooses to calculate these
emissions with the EDMS, care should be taken to remove exclude those emissions from emissions
associated with the NONROAD model.
The EPA collaborates with the FAA to develop and maintain methods for estimating aircraft emissions
using EDMS, which also estimates emissions from aircraft in addition to the APUs and GSE as noted
above. The emissions developed in this way are associated with aircrafts' landing and takeoff (LTO)
cycle. The EPA encourages state and local air agencies to collect aircraft LTO activity data that can be
used in improving estimates using the EPA EDMS-based methods. More details on the EPA methods are
available in the documentation and methods associated with the latest NEI year found at
https://www.epa.gov/air-emissions-inventories/national-emissions-inventory.
88
-------
The aircraft sector includes numerous SCCs as shown in Table 18, and all but two of these are associated
with the NEI point source inventory. Historically, airport emissions have been calculated as a nonpoint
source, but with the latest approaches having all airport locations readily available, such a simplification
is no longer needed. The result of this change from nonpoint to point sources is that most of the SCCs
are shown as 10-digit SCCs (usually used just for nonpoint sources) but are now used as point SCCs.
Table 18: Aircraft SCCs and data categories in EPA estimates
see
Data Category
SCC Description
EPA estimates
2275001000
Point
Mobile Sources; Aircraft; Military Aircraft; Total
X
2275020000
Point
Mobile Sources; Aircraft; Commercial Aircraft-
Total: All Types
X
2275050011
Point
Mobile Sources; Aircraft; General Aviation; Piston
X
2275050012
Point
Mobile Sources; Aircraft; General Aviation; Turbine
X
2275060011
Point
Mobile Sources; Aircraft; Air Taxi; Piston
X
2275060012
Point
Mobile Sources; Aircraft; Air Taxi; Turbine
X
2260008005
Point
Mobile Sources; Off-highway Vehicle Gasoline 2-
Stroke; Aircraft Ground Support Equipment
X
2265008005
Point
Mobile Sources; Off-highway Vehicle Gasoline 4-
Stroke; Aircraft Ground Support Equipment
X
2267008005
Point
Mobile Sources; LPG; Aircraft Ground Support
Equipment
X
2268008005
Point
Mobile Sources; CNG; Aircraft Ground Support
Equipment
X
2270008005
Point
Mobile Sources; Off-highway Vehicle Diesel-
Aircraft Ground Support Equipment
X
2275070000
Point
Mobile Sources; Aircraft; Aircraft Auxiliary Power
Total
X
2275085000
Nonpoint
Mobile Sources; Aircraft; Unpaved Airstrips; Total
2275087000
Nonpoint
Mobile Sources; Aircraft; In-flight (non-Landing-
Takeoff cycle)
X
In June 2012, the EPA promulgated standards for NOx emissions from commercial aircraft engines.
Information about this rule including fact sheets, test procedures, and emissions documentation is
available on the EPA website for this purpose. In that rule, the EPA adopted Tier 6 and Tier 8 standards
for NOx on new aircraft engines, with compliance dates in 2013 and 2014. Prior rules published in 2005
and 1997 also reduce aircraft NOx emissions with engine standards (documented on the same website).
When estimating emissions from aircraft, it is best practice to consider the impacts of these rules for
improving the accuracy of emissions inventories. This is why the EPA recommends using the EPA EDMS-
based approach with locality-specific data.
State or local air agencies may determine that the EPA default method is not appropriate for SIP
purposes, though the EPA recommends considering the approach but using more locality-specific
activity information. For areas that have provided the EPA locality-specific airport activity data, the NEI
estimates may very well provide emissions needed to meet SIP requirements. The EPA method is more
89
-------
robust than older methods and includes national control programs as noted below. As shown in the
table above, the EPA's methods do not estimate emissions from unpaved airstrips (SCC 2275085000).
Furthermore, the "in-flight" SCC (2275087000) is covered by the EPA's method for only lead (Pb)
emissions and is related to general aviation aircraft that still use leaded fuel. Air agencies may want to
consider unpaved airstrips and/or in-flight emissions for other pollutants as part of their inventories for
improved model performance or other reasons, though doing so may not be necessary depending on
the significance of those emissions in a particular area. Any locality-specific data or custom emissions
associated with aircraft sources are expected to be included with supporting documentation for the
emissions inventory used for a SIP.
4,8,5,2 Commercial Marine
The CMV emissions category includes boats and ships used either directly or indirectly in the conduct of
commerce or military activity. The majority of vessels in this category are powered by diesel engines
that are either fueled with distillate or residual fuel oil blends. For the purpose of this inventory, we
assume that Category 3 (C3) vessels primarily use residual blends while Category 1 and 2 (CI and C2)
vessels typically use distillate fuels. Another consideration for emissions estimation of CMV sources is
auxiliary engines. All three types of vessels may have auxiliary engines that are used for onboard power
generation or other needs.
The C3 inventory includes vessels which use C3 engines for propulsion. C3 engines are defined as having
displacement above 30 liters per cylinder. C3 inventories typically include emissions from both
propulsion and auxiliary engines used on these vessels, as well as those on gas and steam turbine
vessels. Geographically, the inventories include port and interport emissions that occur within the area
that extends 200 nautical miles (nm) from the official U.S. shoreline, which is roughly equivalent to the
border of the U.S. Exclusive Economic Zone. For national inventories created by the EPA, only some of
these emissions are allocated to states based on official state boundaries that typically extend 3 nm
offshore. The EPA recommends that state water boundaries also be used when determining which
emissions should be included with a particular state or county.
The CI and C2 vessels tend to be smaller ships that operate closer to shore, and along inland and
intercoastal waterways, with engines greater than or equal to 37kW. Naval vessels are not included in
CI and C2 vessel inventories, though Coast Guard vessels are included. While these C1/C2 vessels are
most often associated with near-land applications and thus included as within state waters, sometimes
these vessels are used outside of state waters, particularly in the Gulf of Mexico. In these cases, as with
C3 vessels, the EPA recommends that state water boundaries be used when calculating emissions
associated with a particular state or county.
The CMV source category does not include recreational marine vessels, which are generally less than
100 feet in length, most being less than 30 feet and powered by either inboard or outboard engines. The
calculation of these emissions is included in the NONROAD model, and these emissions are, therefore,
considered "nonroad equipment."
90
-------
The EPA develops and maintains methods for estimating CMVs, and these methods are used as
"fallback" methods for use in the NEI. They are also available as a fallback for states without other data
for this sector. At this time, the EPA's methods are top-down national methods, which apportion
nationally estimated emissions values to the states and counties. In the latest such approaches (used for
the 2008 and 2011 NEI), this apportionment includes allocation to GIS shapefiles for ports and underway
emissions. These shapefiles are available with the latest NEI documentation and include separate files
for ports and underway. Key differences associated with various EPA methods from the 2002 NEI
through the 2011 NEI have focused on allocation of emissions to the shapefiles, which has been done in
collaboration with state and local air agencies.
The current EPA methods also use emissions types. Each of the CMV SCCs requires an appropriate
emissions type (M=maneuvering, H=hoteling, C=cruise, Z=reduced speed zone), because emission
factors vary by emission type. For the NEI, each SCC and emissions type combination was allocated to a
shapefile identifier in the nonpoint inventory. The allowed combinations of SCCs and emission types are
shown in Table 19. The default values are those assumed when the actual emission type may be
unknown; for example, emissions that occur in shipping lanes are assumed to be 'cruising' and cannot
be 'hoteling,' which only occurs at ports.
Table 19: Commercial marine SCCs and emission types in EPA emissions
SCC
SCC Description
Allowed
Default
2280002100
Marine Vessels, Commercial Diesel Port
M
M
2280002200
Marine Vessels, Commercial Diesel Underway
C
C
2280003100
Marine Vessels, Commercial Residual Port
H
H
2280003100
Marine Vessels, Commercial Residual Port
M
H
2280003200
Marine Vessels, Commercial Residual Underway
C
C
2280003200
Marine Vessels, Commercial Residual Underway
Z
C
State or local agencies may determine that the EPA default method is not appropriate for SIP purposes
or may need further refinement by air agencies in developing SIPs. For example, the general calculation
method may be a reasonable approach, but the activity and spatial allocation achieved from a top-down
approach could be improved by using the same methods with locally resolved input data to calculate
emissions for specific areas.
When estimating CMV emissions, reporting agencies should consider national regulations that decrease
emissions from this sector, such as increasingly stringent emissions requirements for new marine diesel
engines and sulfur requirements for marine diesel fuel and adjust their estimates accordingly. Emissions
reductions resulting from requirements for new marine engines are generally not immediately reflected
in inventories as newer engines are gradually incorporated into the vessel fleet and as older
engines/vessels are retired. Emissions reductions resulting from the use of lower sulfur fuels are
generally seen more quickly, though estimating impacts can be complicated by vessels coming from
other parts of the world with different fuel standards. Table 20 provides a list of more recent relevant
regulations and associated references for the CMV category. Older Tier 1 and 2 regulations on marine
diesel engines, dating back to 1999, are also listed on the websites shown in the table.
91
-------
Table 20: Commercial marine vessel recent regulations and impacted pollutants
Regulation
Pollutants
impacted
Reference
Great Lakes Steamship Repower Incentive Program
(January, 2012)
S02, PM2.5
Regulations for
Emissions from
Marine Vessels
Tier 2 and Tier 3 Emissions Standards for Category 3
Marine Diesel (April, 2010). Includes both engine
standards and fuel sulfur limits.
S02, NOx, PM2.5
Tier 3 and Tier 4 Emission Standards for Category 1 and
2 Marine Diesel Engines (June, 2008).
PM2.5, NOx
Nonroad Diesel Tier 4 Rule (May 2004). Decreased
allowable levels of sulfur in marine diesel fuel by 99%,
taking effect in 2007.
S02
For states needing to refine their emissions estimates beyond the values produced by the EPA's default
methods, Table 21 provides a list of the latest resources for states in better understanding how to
calculate emissions for the CMV sources. The newer references are included higher in the table. The
EPA's recent method updates have occurred in conjunction with rule development efforts, and so some
of these documents are associated with inventories developed for the rules listed above. Even newer
methods may become available over time and can be considered for use in SIP planning inventory
development. Much of this information is also available on the EPA nonroad mobile website under
"Other Nonroad Reference Material". Any locality-specific data or custom emissions associated with
CMV emissions are expected to be included with supporting documentation for the emissions inventory
used for a SIP.
Table 21: Current EPA methods for commercial marine vessel emissions.
Inventory approach resource
CMV Approaches For.,
Reference
Current Methodologies in
Preparing Mobile Source
Port-Related Emissions
Inventories (April, 2009)
Ocean going (generally C3)
vessels, harbor craft, cargo
handling,
https://archive.epa.gov/sectors/web
/pdf/ports-emission-inv-april09.pdf
Regulatory Impact Analysis:
Control of Emissions of Air
Pollution from Category 3
Marine Diesel Engines (June,
2009)
Bottom-up port C3 emissions for
89 deep water and 28 Great Lake
ports in the U.S. and interport
emissions from the Waterway
Network Ship Traffic, Energy and
Environmental Model (STEEM).
https://nepis.epa.gov/Exe/ZvPDF.cgi
/P1005ZGH.PDF?Dockev=P1005ZGH.
PDF
See Chapter 3.
Category 2 Vessel Census,
and Spatial Allocation
Assessment and Category 1
and Category 2 in-Port/At-
Sea Splits (EPA-HQ-OAR-
2003-0190-0423)
CI and C2 vessels
https://www.regulations.gov/docum
ent?D=EPA-HQ-OAR-2003-019Q-
0423
Final Regulatory Impact
Analysis: Control of
CI and C2 propulsion, CI marine
auxiliary, <37kW commercial
https://nepis.epa.gov/Exe/ZvPDF.cgi
/P10024CN.PDF?Dockev=P10024CN.
92
-------
Inventory approach resource
CMV Approaches For...
Reference
Emissions of Air Pollution
from Locomotive Engines
and Marine Compression
Ignition Engines Less than 30
Liters per Cylinder (May
2008)
propulsion, and <37kW marine
auxiliary.
PDF. Chapter 1 provides more in-
depth description of engine types,
while Chapter 3 provides emissions
calculation approach.
Emissions Inventory
Improvement Program
Evaporative VOC emissions from
marine transport of petroleum
liquids, including the loading and
unloading of the petroleum
liquids.*
El I P. Volume III. Chapter 12 "Marine
Vessel Loading, Ballasting, and
Transit."
t These emissions are often included in as part of the nonpoint inventory rather than associated with the CMV
sector.
Older approaches are also available. In 1999, the EPA sponsored two studies, "Commercial Marine
Activity for Great Lake and Inland River Ports in the United States," and "Commercial Marine Activity for
Deep Sea Ports in the United States." Both of these reports are available at the website for Reports
related to Emissions Control Areas for Marine Vessels. These studies provide activity profiles for
approximately 150 ocean, lake, and river ports, and present a method for an inventory preparer to
allocate time-in-mode activity data from one of four typical ports to another port that has similar
characteristics. Activity profiles for a typical port include: 1) number of vessels in each category; 2)
vessel characterization, including propulsion size (horsepower), capacity tonnage, and engine age; and
3) number of hours at each time-in-mode associated with cruising, reduced speed, maneuvering, and
hoteling. These methods provide alternative approaches for state and local air agencies in preparing SIP
inventories, but care should be taken to incorporate the impact of national rules incorporated into the
newer approaches as documented in Table 21.
4,8,5,3 Locomotives
The locomotive emissions category includes diesel-electric engines, which use 2- or 4-stroke diesel
engines and an alternator or a generator to produce the electricity required to power its traction
motors. The EPA default approach does not include locomotives powered by electricity or steam. It is
believed that the number of wood or coal driven steam locomotives in the U.S. is currently very small;
therefore, these types of locomotives are not generally included. In specific instances, if such sources
are relevant to a NAA, then they may need to be included. The locomotive sources are further divided
into categories: Class I line haul; Class ll/lll line haul; Passenger; Commuter; and Yard.
In the NEI, rail yard emissions are treated as point sources. Whether rail yards should be treated as a
point source for SIP planning purposes depends on the importance of these emissions to the NAA and
the analyses that need to be done for any modelled attainment demonstration. Emissions at these
sources can include both stationary sources (e.g., switching engines) and mobile sources (locomotives).
Because of the mobile source component, a rail yard doesn't necessary meet the technical definition of
a major source. Nevertheless, for NAAs where rail yards are a significant source, the EPA recommends
that states consider treating these as point sources so that the best spatial allocation can be used for
modeled attainment demonstration.
93
-------
Several definitions come into play in determining the locomotive category, which impacts the SCC that
should be used.
• Class I railroads are those that operate in many different states over thousands of miles of track.
They are defined by the U.S. Surface Transportation Board as having annual carrier operating
revenues of $250 million or more. As of 2014, there were 8 such railroads in the U.S.: Amtrak
and seven freight railroads, designated Class I based on 2011 measurements released in 2013
("Class I Railroad Statistics". Association of American Railroads. April 17, 2013.)
• Class ll/lll railroads are also known as short line and regional railroads, and they range in size
from very small (a few carloads of freight per month) to multi-state railroads just under Class I
size. Class II railroads are defined as having revenues greater than $20.5 million but less than
$277.7 million for a least 3 consecutive years. Switching and terminal railroads are excluded
from Class II status. A Class III railroad is defined as a rail company with annual operating
revenue of less than $20 million.
• Line haul refers to higher power usage of locomotive engines during extended mileage. It is also
associated with a test cycle used for estimating emission factors intended to represent
emissions during line-haul activity.
The EPA develops and maintains methods for estimating locomotive emissions, and these methods are
used in the NEI in the absence of more updated data from air agencies. They are also available as a
fallback for states without other data for this sector. The EPA's 2011 national rail estimates were
developed by applying growth factors to the 2008 NEI values based on railroad freight traffic data from
the 2008 and 2011 "R-l" reports submitted by all Class I rail lines to the Surface Transportation Board
and employment statistics from the American Short Lines and Regional Railroad Association for Class II
and III. In the latest such approaches (used for the 2008 and 2011 NEI), this apportionment includes
allocation to GIS shapefiles for line-haul and rail yard emissions. The EPA project report Development of
2011 Railroad Component for National Emissions Inventory (September 2012) provides the EPA's latest
estimation approach and references the 2008 approach as well. The report for the 2008 NEI was another
project report, called Documentation for Locomotive Component of the National Emissions Inventory
methodology (May 2011) and available in the zip file posted with the 2008 NEI documentation. Ongoing
work for the 2014 NEI and future inventories will further improve upon past methods.
The 2008 and 2011 EPA default method was developed with a consortium of the EPA, state, and
industry, led by the Eastern Regional Technical Advisory Committee (ERTAC). The ERTAC group provided
data and reviewed methods as is described in the 2008 NEI Technical Support Document and associated
documentation. Table 22 lists the locomotive SCCs used in the EPA default methodology. Other SCCs
exist as valid SCCs, but are not used in the EPA method, so states should consider this when comparing
the EPA estimates to locality-specific estimates and when merging emissions inventories that include
both state/local components and the EPA estimates.
94
-------
Table 22: Locomotive SCCs
Data
see
Description
EPA Estimated?
Category
2285002006
Mobile Sources Railroad Equipment Diesel Line Haul
Locomotives: Class 1 Operations
Yes - in shape
files
Nonpoint
2285002007
Mobile Sources Railroad Equipment Diesel Line Haul
Yes-in shape
Nonpoint
Locomotives: Class II / III Operations
files
2285002008
Mobile Sources Railroad Equipment Diesel Line Haul
no
Nonpoint
Locomotives: Passenger Trains (Amtrak)
2285002009
Mobile Sources Railroad Equipment Diesel Line Haul
Locomotives: Commuter Lines
no
Nonpoint
2285002010
Railroad Equipment Diesel Yard Locomotives
no
Nonpoint
28500201
Internal Combustion Engines Railroad Equipment
Diesel Yard
Yes - as point
sources
Point
State or local agencies may determine that the EPA default method is not appropriate for SIP purposes
or may need further refinement by air agencies in developing SIPs. A number of ways to improve the
EPA estimates were noted in the original 2011 report (see
ftp://ftp.epa.gov/Emislnventorv/2011/doc/2008nei locomotive report.pdf) that documented the
approach used for the 2008 NEI. For the Class I line-haul emissions, the report identifies the most
important considerations such as the rail link-level freight tonnage. Other variables were not addressed
in the methodology, including track grade and track speed. While the method calculated railroad-
specific fleet- averaged emission factors from industry fleet data, it is noted that emissions from
individual engines are highly variable even for the regulated engine categories depending on variables
such as the specific locomotive model, operation cycle, and conditions of operation. Furthermore, the
EPA approach does not address the seasonality of the emissions, while activity levels of different
engines do vary seasonally and could impact emissions for seasonal inventories. Any locality-specific
data or custom emissions associated with locomotive sources are expected to be included with
supporting documentation for the emissions inventory used for a SIP.
For the rail yard emissions, we have already noted above the consideration of rail yards as point sources.
In addition, the 2011 report mentioned in the last paragraph notes additional considerations for
improvements. For example, using the tonnage hauled as an indicator of the amount of switching
activity is an assumption that could reduce uncertainty. The variability associated with actual switching
duty-cycles and in the number of switchers operating at some rail yards seasonally was not considered
in the method.
Locomotive idling emissions is another issue that is not clearly considered in the default EPA
methodology and could have an impact on emissions estimates. During normal operations, locomotives
have numerous reasons why they may remain idling including the switching operations noted above.
The EPA has included regulations on idle emissions to help address this source. More details on idling
emissions from locomotives is available in "Control of Emissions from Idling Locomotives" (EPA, March
2008).
95
-------
The EPA website on locomotive emissions regulations provides a resources for understanding the
regulations that apply to locomotives and how those may impact emissions. Only one recent regulation
impacts emissions from locomotives, while several older regulations also apply. In June 2008, the EPA
published the Control of Emissions of Air Pollution for Locomotive Engines and Marine Compression
Ignition Engines Less than 30 Liters Per Cylinder. This rule includes standards for existing and
remanufactured locomotives, Tier 3 and Tier 4 standards for newly built locomotives, and idle reduction
requirements for new and remanufactured locomotives. The rule reduces emissions of PM2.5 and NOx.
The emissions estimation approach used for this rule is available in Chapter 3 of the Regulatory Impact
Assessment for the rule, and this approach differs from the EPA method used for the NEI. Associated
with the rule, the EPA developed a fact sheet on locomotive emission factors (April, 2009) that can be
used in estimating emissions from locomotives.
4,8,6 Biogenic ant! Geogenic Sources
Biogenic sources are a subset of natural emissions sources that may contribute significantly to an
emissions inventory. Vegetation (i.e., forests and agriculture) is the predominant biogenic source of VOC
and is typically the only source that is included in a biogenic VOC emissions inventory. Microbial activity
in the soil contributes to natural biogenic NOx and CO emissions.
The EPA makes available biogenic emissions estimates as part of the NEI. Starting with the 2008 NEI, the
biogenic vegetation and soil emissions have been included in the official NEI releases. In some cases,
these EPA-generated emissions may be sufficient for use in SIP inventories other than modeling
inventories.
Biogenic emissions from vegetation and soils are computed using a model which utilizes spatial
information on vegetation and land use and environmental conditions of temperature and solar
radiation.20 The model inputs are typically horizontally allocated (gridded) data, and the outputs are
gridded biogenic emissions that can then be speciated and utilized as input to photochemical grid
models. Several models exist, as described in the following subsections.
Modeled biogenic vegetation emissions are significant for understanding ozone formation, particularly
in the Eastern U.S., and the dependence on meteorology makes using case-specific meteorology very
important. In addition, the impact and model characterization of biogenic terpene and sequiterpene on
secondary aerosol formation for PM modeling is thought to be important. Emissions data developers
should consider the latest available information on this issue where biogenic sources are key sources in
their modeling region.
Biogenic Emissions Inventory System (BEIS)
The EPA has developed the Biogenic Emissions Inventory System, version 3.6 (BEIS3.6) for use in
estimating biogenic emissions. This model can be run both as part of a CMAQ model run and
20 In most cases, ozone, PM, and regional haze modeling includes the emissions from vegetation and soils, but not other
potentially relevant sources such as volcanic emissions, lightning, and sea salt.
96
-------
alternatively as a module of the SMOKE system. The CMAQ approach is described by the CMAQ wiki at
https://www.airqualitymodeling.Org/index.php/CMAQv5.l Biogenic Emissions (BEIS) Update. The
latest documentation on the SMOKE approach to run BEIS is available at
http://www.cmascenter.org/smoke/.
The key inputs to BEIS include:
• Temperature data at 2 meters, which can be obtained from modeled meteorological data, such
as data from the Weather Research and Forecasting (WRF) Model. Such data are also needed for
input to air quality models used for modeled attainment demonstrations sometimes needed as
part of SIPs. The BEIS3.14 documentation has more information on the meteorological inputs.
• Land-use data from the Biogenic Emissions Land Use Database, version 4.1 (BELD4). BELD4 data
provides data on the 230 vegetation classes at 1-km resolution over most of North America.
These data are available at https://www.epa.gov/air-emissions-modeling/biogenic-emission-
sources. These land use data can be created with the Multimedia Integrated Modeling System
(MIMS) Spatial Allocator (https://www.cmascenter.org/sa-tools/). Air agencies can alternatively
rely on the EPA's 12-km gridded data for modeling domains that overlap with the EPA national
domain, provided with the EPA's latest modeling platform on the Emissions Modeling
Clearinghouse-
Other Biogenic Models
Other biogenic models such as Model of Emissions of Gases and Aerosols from Nature (MEGAN)
(http://wiki.seas.harvard.edu/geos-
chem/index.php/MEGAN v2.1 plus Guenther 2012 biogenic emissions) are acceptable as well,
provided that the users have demonstrated their suitability for the modeling application through model
performance evaluation (see Modeling Guidance for more information). A version of MEGAN has also
been incorporated into an air quality model called the WRF model coupled with Chemistry (WRF-CHEM)
(http://www.pnl.gov/atmospheric/research/wrf-chem/). California has developed a GIS-based biogenic
model called BEIGIS (http://www.arb.ca.gov/ei/biogenicei.htm). The EPA previously made a comparison
between BEIS and MEGAN (available at
http://www.cmascenter.org/conference/2008/slides/pouliot tale two cmas08.ppt).
BEIS, MEGAN, and other models continue to evolve and so agencies using these models should evaluate
the suitability of one model over the other as part of a modeled attainment demonstration. Once a
choice has been made for modeling purposes, emissions should also be summed across the NAA for
inclusion in the base year inventory for the NAA and/or the attainment projected inventory for the NAA.
Geogenic and Other Natural Sources
Geogenic emissions are primarily the result of oil or natural gas seeps and soil wind erosion. In addition,
lightning may also be a significant contributor to natural N0X emissions in an inventory area. Volcanoes
and fumaroles (i.e., vapor or gas vents in a volcanic region) can be additional sources of geogenic
emissions.
97
-------
As a source of ozone precursor emissions, geogenic sources are usually less significant than biogenic
sources. However, geogenic wind erosion may contribute substantially to PM emissions in an area. In
2012, the EPA released CMAQ v5.0 that includes a module that dynamically estimates hourly natural
emissions of fine and coarse dust particles due to wind action over arid and agricultural land. This allows
the contribution of this source to be included in a modeled attainment demonstration without the need
to prepare an emissions inventory first. The windblown dust approach is based on the work of J.S. Fu et
al. (2016). For the emissions inventory SIP, air agencies using this feature should evaluate the results of
the CMAQ-based approach to determine if it provides a reasonable representation of dust emissions in
their modeling domain. Often times, CMAQ overestimates ambient crustal matter concentrations
without the addition of windblown dust and so adding emissions may not be advisable.
If windblown dust emissions are included in SIP planning, such emissions should be included in the base
year and attainment projected inventories to meet the requirements for a complete emissions
inventory. In the case of using CMAQ, this would require that the CMAQ-based emissions be summed
for the NAA. Since CMAQ produces gridded emissions estimates, air agencies would need to sum
emissions from the relevant grid cells or portions of cells associated with the NAA and/or other
geopolitical regions as appropriate. One relevant detail about the AERR regarding windblown dust is that
they are specifically exempted from the reporting requirements in 40 CFR 51.20(d)
State, local and tribal agencies should also prepare an inventory of all other significant geogenic sources
in the inventory area. Methods for estimating VOC emissions from oil and gas seeps, as well as NOx
emissions from lightning, are described in the El IP document, Volume V, Biogenic Sources Preferred
Methods available at https://www.epa.gov/air-emissions-inventories/volume-5-biogenic-sources. For oil
and gas seeps, the preferred method is to develop a local emission factor based on the study of oil or
gas seeps in the inventory area.
Lightning produces NO, which is oxidized to N02 in the presence of ozone or in a photochemically
reactive atmosphere. This N0X is formed in the mid-upper troposphere and can impact the distribution
of reactive nitrogen as well as ozone. NO emissions from lightning can be estimated by collecting activity
data on the cloud-to-ground lightning flashes, assuming a frequency of intra-cloud flashes based on the
value for cloud-to-ground lightning flashes, and applying appropriate emission factors (in molecules NO
per flash) to these activity levels. Because lightning is not a direct source of N02, accounting for this
source category is more important for air quality modeling purposes than for SIP planning inventory
purposes. Given the altitude of these emissions, it is not usually important to include these emissions in
the base year inventory for the NAA or in the attainment projected inventory for the NAA.
CMAQ v5.0 includes a parameterized approach to estimate NOx emissions and its vertical distribution
that is generated from lightning. Over the Continental U.S., the approach can use lightning flash totals
from the National Lightning Detection Network (NLDN). For cases where such data are not available, an
alternate algorithm can be used that estimates the number of lightning flashes based on the modeled
convective precipitation. The peer reviewed literature contains both a description of the approach (Allen
et al., 2011) and a model evaluation (Appel et al., 2010). More information is available from the CMAQ
v5.0 wiki (see
98
-------
https://www.airqualitvmodeling.org/index.php/CMAQ version 5.0 (February 2012 release) Technical
Documentation#Lightning NOx emissions).
4,8,7 WIMlancl Fires
Because the CAA specifies that planning inventories (e.g., the base year inventory for the NAA) must
include aN sources of emissions, the inventories should, therefore, include fires. Thus, wildfires and
prescribed fires, including wildland fire use, should be included in the NAA inventories when they
overlap with NAA boundaries. Generally, areas will find that the estimates available in the NEI will
suffice for providing a reasonable estimate of these emissions, with the purpose of providing a general
idea of the level of contribution to the overall emissions within the NAA.
Wildland fires (WLFs) are generally defined as any non-structure fire that occurs in the wild land (an area
in which human activity and development are essentially non-existent, except for roads, railroads,
power lines, and similar transportation facilities). The National Wildfire Coordinating Group21 has
identified the following two types of WLFs.22
• Prescribed fire is any fire intentionally ignited by management actions in accordance with
applicable laws, policies, and regulations to meet specific land or resource management
objectives.
• Wildfire is any fire started by an unplanned ignition caused by lightning; volcanoes; other acts of
nature; unauthorized activity; or accidental, human-caused actions, or a prescribed fire that has
developed into a wildfire.
Wildland fires are a unique source of emissions with respect to NAAQS attainment planning. For ozone,
PM2.5, and regional haze planning purposes, these fires are major sources of primary PM2.5 and VOC and
a moderate source of NOx. They can consequently impact the ambient PM2.5 and ozone; however, they
have a number of unique properties:
• Depending on the fuel source and acreage burned, both prescribed and wildfires can emit
significant emissions, but emissions from prescribed fires are generally lower due to the burning
characteristics and the relative amounts of flaming and smoldering emissions. In addition,
prescribed fires can be used to prevent future wildfires, which are generally larger, hotter, more
dangerous, and more polluting.
• Wildfires are naturally occurring events for which the emissions are generally not reasonable to
prevent or control. As such, air agencies may request that the EPA exclude from regulatory
determinations ambient data influenced by the wildfire if the request meets the criteria,
schedule, and procedural requirements for an exceptional event set forth in 40 CFR 50.14.23
21 The National Wildfire Coordinating Group was formed under a cooperative agreement between the
Departments of Agriculture and the Interior and consists of representatives from each of the Federal Land
Management agencies and the National Association of State Foresters.
22 Definitions are derived from the Glossary of Wildland Fire Terminology, National Wildfire Coordinating Group,
PMS 205, July 2012. Available at http://www.nwcg.gov/glossarv-of-wildland-fire-terminology.
23 'Treatment of Data Influenced by Exceptional Events; Final Rule" (81 FR 68216, October 3, 2016).
99
-------
• Fire emissions can rise very high in the atmosphere on plumes, and the pollutants can be
transported beyond state boundaries. Fires occurring outside NAAs can have impacts within
NAAs on days that may not be determined to be exceptional events.
• Since wildfires and prescribed fires generally occur in forests and rangelands, they are less likely
to be within the boundaries of NAAs, which are usually urbanized areas.
• Fires can also reduce ozone formation because their smoke can block sunlight that otherwise
would result in more ozone.
SmartFire emissions
The EPA has adopted for use in the NEI a satellite-based approach to estimate fire emissions at specific
locations and as daily emissions. The SmartFire approach (see http://www.airfire.org/smartfire/) was
developed by researchers with support from the U.S. Forest Service (USFS). The EPA and USFS have
worked together to implement approaches for collecting activity data from states to feed SmartFire for
use in calculating emissions. The SmartFire approach uses the following general equation to estimate
wildfires and prescribed fires. Accurate estimates of fire emissions rely on accurate estimates of the
terms in the equation below.
Emissions = Area burned * Fuel Load Available * Fuel Consumed (Burn Efficiency) * Emission Factors
The SmartFire2 (SF2) approach has been used most recently for the 2011 and 2014 NEIs.
SF2 estimates the "area burned" term in the above equation, in conjunction with the BlueSky
framework model that estimates the last three terms in the above equation. The "fuel load available"
term is estimated using the Fuel Characteristic Classification System (FCCS) maps in the BlueSky model.
The "fuel consumed" term is estimated from BlueSky using the CONSUME3 model, which predicts the
fraction of fuel that burns based on many parameters including fuel moisture. Finally, the "emission
factors" term is estimated in BlueSky using the Fire Emissions Prediction Simulator which relies on EFs
from the literature apportioned by flaming and smoldering combustion. Since SF2 was recently
developed, direct references to its development in conjunction with updated BlueSky methods are not
yet available; however, the following reference can be used in general for past applications of these
process models in the SF/BlueSky process: https://www.airfire.org/smartfire/. Reid (2013) provides
more exacting details on the specific procedures used in developing the NEI prescribed and wildfires. As
new NEI years are developed and resources are available, additional improvements to SF2 are expected.
Some states have determined that SF2 is not sufficient for their needs because the satellite-based
methods can miss detection of prescribed burning that occurs under thick tree canopies. As a general
matter, the EPA encourages all users of SmartFire to work with their forestry agencies to obtain and use
local-specific information about area burned, fuel types, fuel load, burn efficiency, and emission factors.
Exceptional Events and Inventories
When developing design values for a particular NAA, certain ambient monitoring data may be excluded
if an exceptional events submittal is approved by the EPA (see 40 CFR 50.1, 40 CFR 50.14, and 40 CFR
100
-------
51.930). It has been suggested during the course of developing this guidance that emissions associated
with exceptional events should be removed from the inventories. The purpose of this recommendation
was the improvement of model performance evaluations and projected attainment modeling.
In spite of potential impacts on modeled concentrations, in most cases, the EPA does not recommend
removing fire emissions from inventories for the following reasons. First, it is not necessary to remove
fires from use in the model performance evaluation because the EPA does not expect ambient data to
be removed from the model performance evaluation (www3.epa.gov/ttn/scram/guidance sip.htm).
Second, the EPA recommends holding the emissions of fires constant between the base and future years
for the purpose of future year inventories (such as are required for PM2.5 NAAQS) and for calculating a
modeled relative response as part of a modeled attainment demonstration. This approach ensures that
the modeled fires have a minimal influence on estimated future year concentrations that are calculated
using the relative change in modeled concentrations. Third, even though in some cases there may be a
direct causal link between fire emissions and monitored air quality data, resulting in exceptional events,
removing fire emissions directly associated with an exceptional event from an inventory requires
knowledge of exactly which fire(s) at specific locations on specific days and hours contributed to the
exceptional event. This may be especially difficult when multiple fires over multiple days lead to
identified exceptional events. Therefore, from an inventory perspective, it may not be straightforward
to remove emissions specifically associated with any particular exceptional event. Determining which
emissions to remove from the inventory could be a significant effort with limited benefit.
The EPA, therefore, encourages states to use a representative mix of prescribed fire and wildfire in their
inventories. In the past, some plans under previous PM2.5 NAAQS have estimated the actual fire
emissions and temporal and spatial patterns from a given year and used this estimate as the assumed
future baseline for planning, while others have used average emissions over multiple years. Other
approaches may be appropriate as well. Moreover, regardless of the approach used, the EPA still
encourages states to submit actual wildfire and prescribed fire activity data that are critical to
developing emissions estimates to the NEI as suggested in the AERR. Additionally, the EPA encourages
states to consult with their EPA Regional office about the fire approach to be used. While other
approaches may be more suitable in particular situations, the recommendations provided here provide
a starting point for discussions about the most appropriate approach for a particular area.
4,9 Quality Assurance
As part of the 8-hour ozone NAAQS, PM2.5 NAAQS rules, state, local and tribal agencies are strongly
encouraged to perform QA checks and procedures on their inventories. State, local and tribal agencies
can develop and select the QA procedures they will perform, and should include the details of their QA
program (including specific procedures) in their IPPs.
The purpose of QA is to ensure the development of a complete, accurate, and consistent emissions
inventory. A well-developed and well-implemented QA program fosters confidence in the inventory and
in any resulting regulatory and/or control program.
101
-------
When air agencies are using data from the NEI, additional QA is warranted considering the purpose for
SIPs is not the purpose for which the NEI was developed. Since a great deal of QA is done on the data
used for the NEI, most of the QA work has been done and so starting with the NEI can often save time.
The overall QA program provides routine and consistent checks and documentation points in the
inventory development process to verify data integrity, correctness, and completeness; identifies and
reduces errors and omissions; maximizes consistency within the inventory preparation and
documentation process; and facilitates internal and external inventory review processes. Quality
assurance activities include technical reviews, accuracy checks, and the use of approved, standardized
procedures for emission calculations, and should be included in inventory development planning, data
collection and analysis, emission calculations, and reporting.
Reference Materials and Planning
The EPA has developed several guidance documents designed to assist state, local and tribal agencies in
designing and implementing their QA programs. The El IP Volume VI. addresses QA issues, including the
following:
Chapter 1: Introduction - The Value ofQA/QC
Chapter 2: Planning and Documentation
Chapter 3: General QA/QC Methods
Chapter 4: Evaluating the Uncertainty of Emission Estimates
Chapter 5: Model QA Plan
Appendix A: Example Audit Report
Appendix B: Technical Systems Audit Checklist Example
Appendix C: Example 1 of Data Quality Audit Checklist
Appendix D: Example 2 of Data Quality Audit Checklist
Appendix E: Performance Evaluation Checklist Example
Appendix F: EIIP Recommended Approach to Using the Data Attribute Rating System (DARS)
On October 20, 2008, the EPA issued the latest version of the EPA Quality Program Policy (CIO 2106.0).
This policy supersedes the previous CIO 2105.0 (formerly EPA Order 5360.1 A2). This policy established
among other things, the requirement for QAPPs for all EPA-supported programs that generate
environmental data. As described by the "EPA Requirements for QA Project Plans" guidance:
The purpose of a QA Project Plan is to document planning results for environment data operations and to
provide a project-specific blueprint for obtaining the type and quality of environmental data needed for a
specific decision or use.
The orders describing the QA requirements are available at the EPA's website for Agency-wide Quality
System Documents (https://www.epa.gov/qualitv/agencv-wide-qualitv-svstem-documents). That
website also includes specifications for non-EPA organizations: "EPA Requirements for Quality
Management Plans" (EPA/240/B-01/001; March 2001) and "EPA Requirements for QA Project Plans"
(EPA/240/B-01/003; March 2001). Additional documentation is also available on the same website
under the subheading of "General Guidance" that air agencies may find helpful in developing their own
QA plans for their inventory development efforts.
102
-------
States should also consult with their EPA Regional office to determine how to meet the QA planning
requirements for the SIP planning inventory development.
Types of Problems in Emissions Inventories
Emissions inventories are compiled from a large number of disparate data sources, which can lead to
many types of data problems. These problems can be divided into value errors and structure errors.
Value errors are errors in the values of emissions inventory data fields, such as the emissions amounts,
incorrect codes, missing values that are required, or incorrect data such as facility names, stack heights,
or latitude/longitude locations. Correct QA can identify these errors and allow data developers to
correct them before the data are used for their intended purpose.
Structure errors refers to those data problems that cause databases to not work, such as missing records
in relational database tables (e.g., "widows" and "orphans"), duplicate records (i.e., multiple records
that represent the same emissions, causing double counting), or incorrect formats (e.g., the wrong
number of fields, missing fields that are required, mismatched data types, such as integer versus text).
Data with no structure errors are described as having good "data integrity." Generally, these types of
problems should be automatically prevented by using robust data systems to manage the data in ways
that match the expected downstream data use, but this is not always the case. For example, data
systems managing emissions inventory data should be developed with the AERR requirements in mind,
since these are the governing requirements for SIP-related inventories.
Checks Available through the EPA's Emissions Inventory System
While many inventories used for SIPs are not also submitted to the EPA through its EIS (the system the
EPA uses to build the NEI), the EIS does have a large number of documented QA checks that the system
performs as it is loading data submitted by air agencies. This automated QA provides a large number of
data integrity checks that air agencies can consider including in their own data systems or checking as
part of their inventory QA. To access a list of these checks, EIS users must use the EIS Gateway and use
the "QA Checks" link under the "Reference Data" heading on the left of the Gateway interface. The
checks are not otherwise published on the EPA's website and can be requested by non-EIS users by
using the "Contact Us" link provided at https://www.epa.gov/chief. The list has more than 600 data
checks that are performed automatically by the EIS. While many of these checks are only relevant for
getting data into the EIS, they promote standardization of inventories and proper formatting in
accordance with the AERR, which (as noted previously) provides the EPA's standards for inventory
preparation including for SIP inventories.
Since the timing of development of SIP inventories is rarely, if ever, coincident with the timing required
for submitting emissions for the NEI, it is generally not feasible for air agencies to submit SIP inventories
to EIS. However, the same types of checks can and should be done on SIP inventories to insure
appropriate inventory quality.
The EIS checks include both critical errors and warnings. Critical errors prevent data getting into the EPA
data system until they are corrected, whereas warnings allow the data to go through, but may need to
be resolved to ensure the data transmitted to the EPA are correct. Examples of critical errors include:
103
-------
• Checks that data codes for required fields use valid values (for the given inventory year; see also
Section 4.7).
• Fields required by the AERR and for data integrity are present.
• Certain fields are within a critical range check (e.g., a year is between 1900 and 2050, stack
height is between 1 and 1300 meters).
• Certain fields have appropriate lengths (e.g., ZIP codes are between 5 and 10 characters).
• Percentages are provided as percentages (1 to 100) and not as a fraction.
• Data element types (e.g., integer, text, numerical) of required data fields are provided
consistent with their required type.
• MOVES county database submission contains the correct files.
• NMIM county database submission contains the correct files.
• Conflict with pollutants reported for grouped pollutants versus individual members of the
group.
• If stack diameter information is reported, the release point exit gas flow rate or velocity should
be reported. Both may also be reported.
Examples of warnings include:
• Checks that data codes for optional fields use valid values (for the given inventory year).
• Data value is longer than the allowable width (some text fields will be truncated if longer than
what the EIS allows).
• Emissions value is outside of a presumed valid range.
• Exit Gas Stack temperature is outside an appropriate range.
• When submitting monthly data, all months are not submitted together.
• File names of mobile data files use incorrect file extensions.
• Data element types (e.g., integer, text, numerical) of optional data fields are inconsistent with
their required type.
nventory Comparisons
Inventory comparisons can be used in multiple ways to effectively identify problems in new inventories.
This is most useful when a high-quality prior inventory is already available. In states with a long history
of developing SIPs, past inventories will be readily available for such comparisons. In states with new or
emerging attainment problems, the EPA suggests comparing SIP inventories to the NEI. Even when the
comparisons reveal problems not with the new inventory but with the one being compared, this can
increase confidence in the inventory of interest.
Emissions are one of the obvious things to compare, and these comparisons are usually done on a
pollutant-by-pollutant basis for pollutants important to a given analysis. As noted in the section on
ranking techniques (Section 4.9.7), ranked emissions (e.g., facility, unit, county, SCC group) can be
compared to the same ranked emissions from other inventories. Such comparisons will often identify
differences worth further exploration when one large source appears on one list and not the other.
104
-------
In addition to ranking, all records across an inventory (or inventory summary) can be compared and
percent differences can be computed. This is usually done by computing "new emissions - old
emissions" divided by "old emissions." Large or unexpected percent differences can be investigated
further to make sure that the differences have some reasonable explanation, such as newer emissions
being controlled, a downturn in activity for one inventory, process changes, fuel changes, or a partial or
complete facility closure. In some cases, methods changes may be the reason for such differences,
including updated emission factors or test data.
For nonpoint emissions, graphical inventory comparisons can be useful. This can be done by not only
using calculated percent differences or ranking, but also using GIS techniques to map county level
emissions and their differences. It can be very useful to map the old emissions, new emissions, and the
computed difference or percent difference side-by-side. This can reveal areas of large emissions that
have significantly changed from one inventory to the next, raising questions for further exploration.
When comparing to an EPA inventory that used EPA methods for a particular sector, this could reveal
that the SIP planning inventory is a better inventory with updated methods, or it could reveal that
outdated methods are being used. In any case, making the comparison will help inventory developers
understand the potential improvements that could be made or that have been made, which will help in
understanding reviewers' expected reactions to the data.
For multi-state inventories needed for SIP development, state-to-state comparisons may be useful in
addition to county maps. This is especially true if emissions for one state were developed differently
from emissions for another state. The EPA encourages states to work together to promote consistent
approaches across a multi-state NAA, but a final QA check of the consistency can be done using the
comparison techniques described here.
Checking Data Codes
As described in Section 4.7, air agencies should check the inventory data codes they are using against
valid codes. Valid codes are defined by the EPA within the EIS. Air agencies that start with the NEI may
significantly minimize the burden of this recommendation.
"oint Versus Nonpoint ant! Reconciliation
Several QA issues arise related to point versus nonpoint sources and inventories: (1) correct
identification of point sources, (2) prevention of double counting across point and nonpoint, and (3)
prevention of missing nonpoint emissions.
Minimum thresholds are specified in the AERR for which sources with facility total potential to emit
emissions above those thresholds must be treated as a point source in SIP planning inventories. This is
more fully explained in Section 2.5.8, with lower thresholds relevant for ozone and PM NAAs. Air
agencies should implement a QA check (or otherwise have a data collection process) that ensures all
facilities with potential emissions (potential controlled emissions, if applicable) that are greater than
these thresholds are included in their inventory as point sources. See also Section 2.5.2 for more
information on calculating potential emissions.
105
-------
When a particular type of source includes contributions from both point and nonpoint sources, double
counting can occur. Quality assurance steps should be taken to ensure that double counted emissions
have been prevented. A good example of this is smaller boilers. Both point source facilities (based on
the potential to emit thresholds) and smaller facilities are likely to have emissions from non-industrial
boilers. For instance, some large institutional facilities such as hospitals may be included as a point
source with their associated boiler emissions, whereas smaller institutional facilities are not included as
point sources and would thus be included as a nonpoint source under an "institutional/commercial"
boiler SCC.
The EPA provides calculation approaches for many nonpoint emissions that allow for subtracting out the
activity of the point source parts prior to estimating nonpoint emissions, including that for the
institutional/commercial boiler example. This is called "point/nonpoint reconciliation." Whether
nonpoint emissions are estimated by the EPA or by air agencies, an assessment of quality should be
done to ensure that sectors associated with both point and nonpoint sources do not double count
emissions. See also Section 4.8.2 for additional information on preventing double counting. A list of
source sectors with potential point/nonpoint overlap is provided in Table 17 above and in the latest NEI
Technical Support Document, available at https://www.epa.gov/air-emissions-inventories/national-
emissions-inventory.
A third type of QA should be done for categories that have both point and nonpoint contributions. It
involves an assessment of potentially missing nonpoint emissions. Because the requirement for SIP
inventories is to include emissions from all sources, it is not sufficient to exclude nonpoint emissions
when part of a sector is covered by point sources, unless it can be shown that the nonpoint portion is
negligible. At a minimum, air agencies should consider the tools available from the EPA to estimate
nonpoint sectors (see Section 4.8.2) and evaluate whether or not their point source inventory covers all
of the activity associated with the nonpoint categories. For inventory years not covered by EPA tools,
agencies should compare point total activity as collected from point sources versus the same type of
activity data used in the EPA tools. For example, U.S. DOE fuel consumption data can be used for
industrial, commercial, and institutional boilers. The source of the activity data used for the latest EPA
nonpoint methods is included in the descriptions of those methods in the nonpoint tools described in
Section 4.8.2 and with the latest NEI documentation.
banking Techniques
Ranking emissions records can be a very useful tool to identifying value errors in emissions values or
other numeric inventory data fields, or at least in finding records that need further investigation.
Ranking involves first defining a set of sources and a level of aggregation of emissions and then ordering
those emissions values from highest to lowest or lowest to highest. This analysis must be done one
pollutant at a time. An analyst then considers whether those sources at the top of the list make sense.
Point sources, facility level emissions ranking
For point sources, ranking can be done at the facility total, unit total, or process level (i.e., at the SCC
level), and each level has its place for reviewing inventories. A facility level ranking can often be followed
106
-------
by subsequent investigation to understand the reasons (at the unit and/or process level) for very high or
very low values. An analyst using a ranking procedure would assess the facilities with the largest
emissions and seek to understand whether they make sense. For example, for combustion products
such as NOx or S02, one would expect to see large EGUs and other industrial sources. Ranking of
emissions values is usually done for one pollutant at a time, reviewing most carefully those pollutants
that are most critical to the particular purpose of the inventory. For example, an inventory supporting
ozone analysis would want to review NOx and VOC most carefully.
It can be useful to compare the list of top facilities for a given pollutant from one inventory to a previous
or alternative inventory for a given state or NAA. This can raise questions regarding large changes from
one inventory to another, which could be explained (e.g., a decrease due to controls) or could identify a
problem (e.g., a missing value or data entry problem). In addition, facility-level ranking can use
additional grouping such as by industry code (i.e., NAICS), to identify unexpectedly high or low emissions
for facilities within a particular industry.
Point sources, sub-facility emissions ranking
Unit- and process-level summaries can additionally be ranked for more specific analyses. For example,
all boilers could be ranked and particularly high or low emissions values investigated. Similarly, all
natural gas processes at particular unit types (e.g., boilers) or by industry (NAICS) could be ranked. These
rankings can show real variations, but they can also find incorrect values that stand out when ordered
and reviewed in this way. There are many variations that could produce useful results, and which ones
are useful for a particular inventory depends on many factors and the purpose of the inventory.
Nonpoint and mobile source ranking
Nonpoint and mobile emissions can also be ranked as a technique to help with QA. Since these
inventories are primarily compiled at the county/SCC resolution, the ranking can be done only at
combinations of county, SCC, and for specific pollutants. Total county nonpoint emissions could be
ranked to look for overly high or low values, but the same effect can be accomplished in a visual way
with mapping approaches (see Section 4.9.9). Specific SCCs can also be selected and the emissions by
county ranked, or individual counties selected and SCCs ranked and compared to other counties.
Since nonpoint emissions are based on various types of activity data that also has a county or other
spatial resolution (such as census tract), these activity data can be used to "normalize" the nonpoint
emissions prior to ranking. For example, emissions coming from households (e.g., residential heating,
residential wood combustion, consumer solvents) can be divided by population or housing data and
ranked for identify possibly strange results. Since some of the activity data used for these categories
starts with state or regional resolution and is subdivided to apportion to the county resolution, such
comparisons can reveal illogical outcomes of such approaches. For example, while it makes sense that
residential wood combustion occurs more per household in rural areas, it wouldn't make sense for a
very low population county to have a disproportionate amount of emissions and urban areas have none.
107
-------
tutlier Analyses
One result of the ranking approaches described above will be to identify outliers. However, such ranking
is a relatively unsophisticated way of identifying such outliers. A more robust approach can be used for
point sources where states have the underlying activity data reported by facilities on a per-process
basis. Dividing emissions by activity values would produce a set of implied emission factors, or rates. The
rates for a similar set of processes (based on SCC) and control devices in use might be expected to be
within an order of magnitude of each other. Emission rate values that are too far outside the expected
relationship could be further investigated to better understand the source of any unexpected values,
which might include an error in the units of measure or the presence of controls or an incorrect SCC, as
well as data entry errors for emission values or other related fields.
(sing Maps to QA Nonpoint Emissions
Nonpoint and mobile source emissions data are associated with states and counties and, therefore, are
ideal candidates for mapping with GIS software. By mapping emissions values for individual SCCs or
groups of SCCs, data analysts can identify unexpected patterns in the data that could signal an error in
creating the data. Though it is technically more challenging to use GIS software, this approach can often
be more valuable than ranking in identifying data problems.
Like the ranking techniques, GIS maps can also use emissions data normalized by some other underlying
or related spatial data, such as population, land use, urban/rural designations, or housing. This type of
information is increasingly available online from the U.S. Census Bureau and other sources. Data
analysts can use the other data to identify unexpected relationships that possibly indicate bad
assumptions made in the underlying methods.
Ixpected Pollutants
Air agencies can use their expertise with emissions sources to identify which pollutants are expected
from particular processes. In addition, for the 2014 NEI cycle, the EPA has provided a list of expected
pollutants for each SCC. While this list is not comprehensive for any pollutant that can be associated
with a process, it provides a bare minimum of essential pollutants associated with SCCs. For example,
NOx is always expected as a by-product of combustion processes. In addition to expected pollutants,
analysts can also identify pollutants that are not expected from a particular process, such as S02 from an
evaporative process.
Once identified, such lists can be used to cross-check inventories to identify either problems with
reported emissions, pollutant codes used, or process codes used. For example, if a combustion process
does not have NOx emissions, it could be that the wrong pollutant code was used in reporting the NOx
and its showing up as something else. Or, the process itself may be incorrectly using a combustion
process code.
Expected pollutants can also help with making sure the correct PM components are provided for PM
inventories, in particular that condensable PM emissions are included in the total PM emissions and that
both filterable and condensable PM emissions are reported. This issue was described in Section 4.2.1
along with a list of the types of sources expected to have such emissions.
108
-------
Some emissions data analysts may not be familiar with various sources enough to know which pollutants
should be expected. For point sources, agency staff can rely on the sum total of all reported data across
multiple sources for a particular process to get a good idea of what should be expected. For example, if
all but one industrial boiler burning natural gas reported VOC, then it would be worth further
investigation as to why VOC was not reported for that one. Since the AERR specifies that all criteria
pollutants must be reported if any one pollutant exceeds the potential to emit threshold, one should
expect that similar processes from multiple facilities would have many, if not all, the same pollutants.
There are other ways to identify expected pollutants. These include looking at past inventories created
by the same air agency or looking at the NEI. Another way is to see what emission factors are available
for the process in Webfire. While these are not foolproof ways of finding expected pollutants, they can
be very helpful. Comparing a draft inventory with such a list can also identify unexpected pollutants to
consider. Used with ranking techniques, knowing what pollutants are expected can also help when you
find zero or null values for a facility, unit, or process.
'acility and Release Point Locations
Facility and release point (e.g., stack) locations are key elements of inventories that are used primarily in
preparation of inventories for air quality modeling and also in permit modeling. Facility locations are
also used in mapping point sources for various public communications purposes or other reasons
needing less precision. While facility resolution is acceptable for many modeling applications, including
12-km gridded modeling, the need for release point locations increases for finer resolution modeling.
Thus, for4-km or 1-km modeling, release point location information should be considered.
With regard to the more detailed release point coordinates, the EPA recognizes the need for air agencies
to leverage SIP inventory development efforts for possible use for other inventory needs. Thus, air
agencies should be particularly attuned to release point geographic coordinates for sources that could
be involved in permitting or risk analysis. For regulatory dispersion modeling, release point location is
critical (rather than only facility location).
The EPA's EIS and other inventory data systems support two types of locations: facility and release
point. There are four key errors that can occur, and each can be identified with their own type of check,
as listed below.
• Facility is not consistent with the state/county FIPS code
This error can be found using GIS, SAS® or other calculations to compare the latitude/longitude
location with the county boundaries using some tolerance. Because county boundaries have
varying degrees of accuracy, a tolerance should be used that allows for locations near, but not
always 100% within, county boundaries.
• Facility location is not consistent with the physical location of the facility
Since some facilities can be on very large tracts of land, facility locations developed using
geocoded addresses or facility entrances are often not accurate with regard to the actual
location of the facility. For emissions inventory purposes, the EPA encourages that facility
locations be positioned on top of a part of the facility that emits one or more pollutants, and the
109
-------
facility centroid is often sufficient for doing so. Using freely available mapping software such as
Google Maps, latitude/longitude locations can be plotted on satellite images to ensure that
values are on top of an emitting source and not on the road, at the corporate headquarters, or
in some other invalid location.
• Release point location is not consistent with facility location
This check is only useful once the facility location is accurate. Such an error can be found using
GIS, SAS® or other calculations to determine the distance between the facility location and the
release point locations. Alternatively, a subtraction of the absolute values of the latitudes and
longitudes can be used to determine the difference in degrees. If that linear distance or the
distance in degrees is greater than some preset criteria, then the locations can be investigated
further. For some types of facilities such as refineries or major chemical complexes, the facilities
may be very spread out and therefore may have legitimately large distances between release
points.
• Release point location is not consistent with the physical location of the release points
For many applications, including some SIP modeling applications, all release points using the
same (usually facility) latitude/longitude is sufficient. However, with the availability of
inexpensive GPS technology, it is more realistic for facilities to collect and submit individual
release point location information. Furthermore, for some types of analysis, particular fine scale
modeling for permitting and risk analysis, correct release point locations can make the
difference between a good analysis and a useless one. It is very difficult to check the accuracy of
release point locations, but using mapping software overlaid on satellite images can provide
some sense of whether the release points are meaningful or not. Coordinates reported to one-
ten-thousandth of a degree (four digits after the decimal point) give a precision of
approximately 30 feet. Thus, a simple check to see if such values are useful is to examine the
number of decimal places included in a given database.
Since a facility's location is not likely to change, the EPA has implemented a procedure in the EIS that
allows latitude/longitudes to be locked, meaning that they cannot be changed accidently during a data
exchange with a state. We encourage air agencies to implement a similar approach in their data systems
so that the efforts of QA done on one inventory can be inherited by other inventories created by that
agency. Locking mechanisms prevent the data from being changed inadvertently through a new data
collection that relies on data from an unverified data source. Thus, a comparison of the locations used in
the SIP inventory to any locked and verified site coordinates in EIS would also highlight potential errors.
Release Point Parameters
Like release point locations, release point parameters (also called stack parameters) are key elements of
inventories that are used primarily in preparation of inventories for air quality modeling. The release
point parameters impact the modeling of emissions from facilities because they specify details of the
way the emissions are released (e.g., they specify how high off the ground the pollutants are emitted
and the exit velocity and temperature of those emissions). One basic and important distinction for
release points is whether the release point is a stack or a fugitive type of emissions, and different types
of parameters are specified for stacks than for fugitive emissions sources.
110
-------
Correct release point parameters for stacks are important for modeling done for ozone and PM2.5
modeled attainment demonstrations as well as modeling done for Regional Haze SIPs, whereas fugitive
release parameters are not as critical for that purpose. This is because stack emissions parameters
impact the modeling of plume rise and, therefore, determine how high the emissions rise in the
modeled layers of the air quality model. Fugitive emissions are emitted in the modeled surface layer
with no additional characterization. This is in contrast to more detailed plant-level modeling (i.e., done
for permit modeling or risk modeling purposes), where both stack and fugitive parameters are critical.
Because the focus of this document is on SIP purposes, we focus on stack parameters here.
Stack parameters included in the EIS and historically are stack height, stack diameter (at the exit point of
the emissions), exit gas velocity, exit gas flow rate, and exit gas temperature. Additionally, a release
point type is available in the EIS to specify a code for whether the "stack" has a particular orientation of
vertical, horizontal, goose neck, vertical with rain cap, or a downward facing vent. Valid release point
codes are available in the EIS reporting code tables in the EIS Gateway.
Quality assurance for stack parameters can be done using several of the QA methods already described.
Range checks can be applied; these can be the default EIS range checks (see Section 4.9.3) or range
checks developed by an air agency. A more robust range check could be developed for specific types of
processes associated with a release point or for specific NAICS codes. For example, one would expect
higher stacks for EGUs than for lumber mills. Additionally, ranking techniques can be applied to stack
parameters (see Section 4.9.7), including options for grouping ranked stack parameters by facility, unit,
process and/or specific types of facilities, units, or processes. This is somewhat complicated by the fact
that multiple units or processes can feed a given release point, and when this happens, the predominant
unit or process could be used for grouping purposes. Finally, more sophisticated outlier analysis could
additionally be used, by finding correlations between stack parameters such as height and diameter and
investigating those combinations that statistically fall outside an expected relationship or ratio.
liogenic QA/QC
Biogenic emissions are somewhat more challenging to QA than emissions for other source categories
because they may be generated differently (i.e., as part of a gridded, hourly modeling exercise rather
than more basic inventory development techniques). The following list provides some basic checks that
can be done as part of the generation of biogenic emissions:
1. Check the biogenic emissions tool logs for warnings/errors.
2. Check the logs to confirm that the expected inputs were used.
3. Compare state or county total emissions to another year or case's biogenic emission totals.
4. Ensure that summer emissions are greater than winter emissions. This can be done by summing
the emissions across the modeled summer days (e.g., May-October) and for the other models
and then comparing the totals on a state and/or county basis. As a quick (but less rigorous)
check, example summer hourly or daily emissions can be visually compared to example winter
hourly or daily emissions using software that displays the gridded emissions.
Ill
-------
5. Visually inspect sample days of the gridded, hourly emissions using software that displays the
gridded emissions. The VOC emissions should generally be higher in southern latitudes because
of warmer temperatures. This difference should be more pronounced in winter in cases where
the southern areas have more actively growing vegetation. The NO emission should generally be
higher in agricultural areas with significant fertilizer application.
• ./lining i "¦ nil Follow-up
As air agencies, as well as the EPA Regional offices, work quickly to meet relevant inventory and SIP
deadlines, data problems are exacerbated when they are discovered too late in the process. A key part
of making sure that a QA plan is effective is deciding (as part of the planning) when a given QA check is
to be done and ensuring (as part of implementing that plan) that it is actually done. To do this
effectively, QA steps should be timed in a QA plan at the earliest possible time that the step can be
performed so that errors are caught as early in the process as possible. Furthermore, assurances must
be put in place that someone is confirming that the checks have been done along the way, so that errors
are caught. At the end of the data development process, some accounting for the history of the QA
should be available along with the data, so that others can confirm that the checks have been done and
can be reassured. Where possible, evidence of completion of each QA step should be referenced, such
as a comparison report.
112
-------
5 Developing Projected Emissions Inventories
:cted Emissions Overview
Emissions estimates for future years24 are called "emissions projections" or "projected emissions."
These projections include emissions growth (due to increased or decreased activities) and emissions
controls due to regulations, settlement agreements (or consent decrees), fuel-switches, or other
programs that reduce emissions in specific ways in the future. Emissions controls can be either on-the-
books or considered as part of a control strategy to reduce emissions as part of SIP planning. For point
sources, emissions projections can also include facility and/or unit-level shutdowns due to settlement
agreements (or consent decrees) and/or ownership decisions.
The primary goal in making projections is to obtain a reasonable and technically credible estimate of
future-year emissions that accounts for key variables. The EPA encourages the air agencies to
incorporate in their analyses the variables that have historically been shown to drive their economy and
emissions, as well as the changes in growth patterns and regulations that are expected to take place
between the time of their base year and projected attainment year. For details on which future year(s)
should be modeled for attainment demonstrations, readers should refer to the Modeling Guidance
(https://www3.epa.gov/ttn/scram/guidance sip.htm).
5.1.1 Readily Available Projections Information
The EPA's latest projection approaches and data are available as a part of the EPA's "emissions modeling
platform" development. These platforms are released and updated periodically in accordance with the
EPA regulatory and analysis activity on the Emissions Modeling Clearinghouse at
https://www.epa.gov/air-emissions-modeling. While each platform comes with a single or limited set of
future years included, the data and thoroughly documented approaches can help air agencies to
develop and improve their own emissions projections.
Additionally, as part of SIP planning, various situations lead to requirements for states to implement
Reasonably Available Control Technology (RACT) and Best Available Control Technology (BACT) as part
of their attainment planning. The emissions reductions expected based on these programs would need
to be include in emissions projections used for modeled attainment demonstrations and other projected
emissions such as the projected attainment inventory for the nonattainment area (see also Section
3.8.1).
5.1.2 Background Materials
For point sources, Volume II "Point Sources" of the El IP series includes Chapter 12 "How to Incorporate
the Effects of Air Pollution Control Device Efficiencies and Malfunctions into Emission Inventory
Estimates" was published in 2000 and still provides the authoritative information related to point source
projections. In particular, Appendix A "Clearing Up the Rule Effectiveness Confusion" and Appendix B
24 Technically, projected emissions could be made from a base year to a year that has already occurred when data
are not yet available for a very recent year. However, for the purposes of the remainder of this guidance, we will
assume projected emissions refers to future year projected emissions.
113
-------
"Emission Inventories and Proper Uses of Rule Effectiveness" provide a thorough discussion of rule
effectiveness that is relevant for current emissions projections.
In addition, the 1999 El IP report entitled "Procedures for Preparing Emissions Projections"
(https://www.epa.gov/air-emissions-inventories/volume-10-emissions-proiections-final-report)
introduces the basic concepts behind emissions projections issues and approaches. The concepts of
control efficiency, capture efficiency, rule effectiveness, and rule penetration are introduced in Section
2.5.18. The El IP document provides information on how to use these factors to estimate projected
emissions. However, some key features of this document need to be updated, including the reference to
the Economic Growth and Analysis System, which is no longer supported by the EPA. Other outdated
concepts include the assumption that economic growth can be a good indicator of emissions growth.
While some correlation can exist, it cannot be assumed but rather must be demonstrated for a
particular industry or facility. The correlation has not been well observed historically in the NEI data
because of various issues including technology improvements and fuel-switching that often reduce
emissions without regulation.
In this section, we will address an overall approach for tackling emissions projection issues that
incorporates use of this document and points out where other approaches or new data are available to
amend the information included in that report.
Some general aspects of the EIIP projections document that are now out of date are as follows.
• The latest SCC codes are available as described in Section 2.5.14.
• The AP-42 emission factor website is now https://www.epa.gov/air-emissions-factors-and-
quantification/ap-42-compilation-air-emission-factors.
• The reference in the Table 2.1-1 web link is no longer applicable. Instead, the following
information can be used to help provide similar information:
o Energy consumption by industry report, 2002:
http://www.eia.doe.gov/oiaf/analysispaper/industrv/consumption.html
o Annual Energy outlook report, early release 2014:
http://www.eia.gov/forecasts/aeo/er/index.cfm
• The references in Table 2.1-2 are also out of date and should be updated as follows:
o U.S. EPA Emissions trends reports. These reports can be used to provide the historical
trends of emissions sectors based on the NEI: https://www.epa.gov/air-trends
o Integrated Planning Model: See reference provided above in Section 1.2
o Multiple Projections System - is no longer in use
o California Emission Forecasting System:
http://www.arb.ca.gov/ei/drei/maintain/database.htm
The EPA recommends that developers of projected emissions take the steps in the bulleted list below.
Each of these steps corresponds to a subsequent subsection.
114
-------
• Identify sectors of the inventory that require projections and sectors for which projections are
not advisable, and prioritize these sectors based on their expected impact on the modeling
region (Section 5.2).
• Collect the available data and models that can be used to project emissions for each of the
sectors (Section 5.3).
• For key sectors, determine what information will impact the projection results the most, and
ensure that the data values reflect conditions or expectations of the inventory region (Section
5.4).
• Quality assurance (Section 5.5)
5.2 Identify ami Prioritize Key Sectors and Regions for Projections
Emissions data developers should evaluate their inventories for those sectors and regions that are most
critical to their modeling results. The purpose of identifying these key sectors is to direct more resources
and efforts to estimating future-year emissions for these sectors in the subsequent steps.
Sectors can be subsets of the larger groups of stationary area, on-road mobile, nonroad mobile, and
point sources. For ozone modeling, it will be most important to ensure correct projections for sectors
with higher NOx emissions. In larger urban areas that contain VOC-limited regions, sectors with higher
VOC emissions will also be important. Mobile sources are typically large contributors to both VOC and
NOx; therefore, they are usually a priority in estimation of future-year emissions, particularly for urban
areas. For PM and regional haze modeling, the relative importance of primary PM2.5 emissions and its
precursors will depend on the area of the country. Primary PM2.5 emissions may be important in urban
areas or areas heavily influenced by local sources such as residential wood combustion or industrial
facilities. In the East, S02 emissions may be of primary importance. For PM2.5 planning, NOx, VOC, and
NH3 will vary in importance depending on the particular situation, which needs to be assessed as part of
the SIP planning process.
Emissions data developers should consider the proximity of the emissions and the expectation of long-
range transport when reviewing projection assumptions. Those emissions that have an impact on the
nonattainment status (whether transported from outside areas or locally generated) should be given
highest priority in evaluating the projection assumptions. For example, less priority can be given to
emissions from distant states and/or for pollutants that do not transport from those states to the NAA.
In some cases, there are sectors that are very difficult or highly uncertain to project into the future.
Wildfire emissions is a common example of such a sector, since models are not readily available that
estimate the magnitude and location of future-year emissions. In these cases, it is reasonable to create
an "average year" inventory and temporal distribution of the emissions. Other approaches to future-
year fire emissions include holding the emissions constant at the same levels as included in the base
year.
5.3 Colle , ; < '• ata and Models ¦ ^ch Sector Needing Projections
For each sector needing projections (and in particular for the priority sectors), air agency staff should
collect and review growth and control information. The sources of such information depend upon the
115
-------
sector of interest. Some data resources and approaches are available and provided here, and new data
and approaches will become available subsequent to the publication of this document. The information
provided here is a snapshot of information that will grow and adapt over time, and emissions modelers
should seek out such new information as part of this step.
Emissions modelers can rely on available default data for sectors and regions with lower priority for
consideration (as described in the previous section). For sectors and regions that are important to
modeling the NAA, emissions modelers should spend additional effort obtaining improved information
from other modeling applications (such as modeling done by RPOs or other states) or developing the
improved information themselves.
As mentioned previously, the EPA provides control assumptions with its modeling data used for
promulgating air quality rules and impact assessments. The EPA approaches typically include the
impacts of federal rules that have been finalized. Impacts of proposed rules are rarely included, as the
changes in emissions impacts can be very large between the proposed and final rules. These data are
posted at the Emissions Modeling Clearinghouse. Sometimes, more current information is available than
what is available from the EPA, for example, when reconsideration amendments are developed and
finalized after the EPA has publicly released its projected inventories. In the subsequent subsections, we
focus on sources of growth information.
5,3,1 Electric Generating Units
This section conveys the general principles and key criteria that inform development and review of EGU
emission projections in support of CAA-related state regulatory actions (such as SIPs) subject to EPA
approval. The section includes technical information related to EGU emission inventories, key
modeling/projection input assumptions, appropriate data sources, and general methodological criteria
to be used by states and EPA Regional offices.
This guidance does not recommend or restrict the use of any specific tool to generate credible
projections of EGU emissions. It explains a process for documenting credible data and input assumptions
with accompanying justifications; describes how to demonstrate a reasonable relationship between
inputs and outputs for a given application; and identifies relevant resources that should be taken into
consideration. Therefore, this section is not intended to be an exhaustive representation or study of
models and specific approaches. Rather, this document is intended to provide the user guidance in
choosing an appropriate approach. Many factors contribute to the selection of the adequate and
appropriate EGU projection approach to use in a specific application. The user is encouraged to consider
different approaches appropriate for each application, taking into consideration the type of input data
available, the level of detail needed to meet the end goal of the application.
Emission projections for EGUs do not tend to follow a simple growth path from historical emission data.
The composition and behavior of the generating fleet, and resulting power sector emission patterns
across facilities, states, and regions, vary substantially over time based on changing economic conditions
as well as changes in fuel markets and regulatory requirements. Therefore, evaluating the credibility of
power sector emission projections necessitates evaluating the reasonableness of several key projection
116
-------
elements or assumptions, while keeping in mind that appropriateness of the method also depends on
the span of future year(s) being projected (the "modeling horizon"). These projection elements include:
1. Electricity demand, taking into account the on-the-books energy efficiency programs and
policies directly influencing demand growth over the modeling horizon
2. The composition of the future generating fleet; if the future year(s) provides enough lead time,
consideration of economic entry (new facility construction) and exit (existing facility retirement)
that would change the composition of the fleet.
3. Generation pattern across facilities that is consistent with economic factors, power system
operability, reliability, and environmental regulatory constraints
4. Potential emissions at each generator, taking into account pollution content of available fuels as
well as the ability of control equipment or processes to capture that pollution
State, tribal and local agencies have the choice to independently develop power sector emission
projections or to use emissions projections from other entities (federal or regional bodies, including
RTOs and ISOs). If a state, tribal or local agency chooses to develop independent projections (or to
derive such projections from other available forecasts), data sources should be referenced and clear
justification should be stated for the projection methodology, including a clear demonstration of the
relationship between demand, cost, generation capacity and emissions. Documentation supporting any
set of power sector emission projections should include:
1. Inventory of all sources of generation accounted for in the projections, including but not limited
to all individual emitting EGUs.
2. Information regarding electricity demand for the future year:
a. The demand forecast for EGU projections should be referenced and documented
(potential sources include AEO, NERC, regional transmission organizations, independent
system operators, state energy offices or public utility commissions, and regional
councils that develop energy planning).
b. Identify the energy efficiency (EE) programs and/or policies explicitly included in the
demand forecast. This clarifies which EE programs or policies are influencing the growth
rate assumptions. (For example, AEO explicitly includes federal appliance standards,
lighting efficiency standards, building energy codes and other federal energy efficiency
programs).25
3. A detailed profile of future capacity, including:
a. Projected changes to the operable fleet, such as retired capacity, new capacity
(including renewable sources such as wind, solar, geothermal, and hydro), changes to
capacity type (e.g., refueling), and pollution control retrofits.
25 Refer to the EPA's Resource for States on estimating EE/RE relative to ElA's Annual Energy Outlook for SIPs
located in Section 5.3.1.2 of this document for more information.
117
-------
b.
Physical characteristics of the operable fleet, including capacity, fuel type, heat rate,
renewable energy capacity factor and pollution controls (i.e., list of emissions controls
and their associated removal efficiencies for pollutants of concern).
4. A detailed unit-level profile of projected power system operation, including generation, fuel
type and use, pollution control configuration, and emissions.
5. A detailed list of all constraints (with corresponding future compliance dates) that apply to
reduce emissions at one or more sources, such as:
a. Federal, state and local rules
b. NSR settlements
c. Consent decrees
d. Rate or permit requirements
e. All other on-the-books regulatory requirements
6. Representations of the following, or if not directly factored in, reasoning and methodology
explaining how these elements are accounted for indirectly:
a. Fuel supplies and/or prices
b. Generator availability constraints
c. Minimum generation requirements (such as qualifying renewable energy (RE)
generation to meet RPS targets)
d. Cost and performance characteristics of existing and potential new generation and
pollution control technology options
e. Transmission, reserve margin, and other system operating constraints
7. Any other information used in and/or relevant to the projections, such as: considerations for
appropriate time horizon, appropriate geographic coverage, internally consistent assumptions,
and preventing double-counting of emissions sources between EGUs and other categories26.
26 To prevent double-counting of emissions sources between EGUs and other categories, emissions developers
should be careful to match the sources of emissions from external models with the base year emissions sources.
This helps to ensure that the EGU part of the point source inventory is separated from the non-EGU part based on
the facilities included in the future-year EGU model. For any projection approaches that use the National Electric
Energy Data System (NEEDS) as an EGU source inventory, the NEEDS records should be compared to the base year
point inventory using the ORIS Plant ID field from both. In some cases (e.g., co-generation facilities), only some of
the units at the facilities are included in NEEDS; therefore, the separation of EGUs from non-EGUs should be done
at the unit level, not by facility. One resource for this information is also in the NEI, which contains a field that
indicates whether the units are NEEDS units or not.
118
-------
Electric generating unit projections for different purposes (such as attainment demonstrations,
maintenance plans, and reasonable further progress plans), as well as varying factors such as pollutants,
geographic areas, and/or how far the projection period is into the future, may prompt adaptation of
different approaches.
Some approaches use computational algorithmic models that determine dispatch based on economic
assumptions while obeying physical constraints of generators within the selected universe of EGUs.
Examples of typical constraints may include, but are not limited to, ramp rates, outage schedules,
transmission limits, system reserve margin limits, and emission limitations.
Other approaches follow a user-prescribed set of rules and relationships to simulate the behavior of
EGUs based on electricity demand trends, historical electricity production trends and data from other
forecasting models. Analysts may also consider combining elements of or outputs from different
projection approaches so long as the resulting set of emission projections across EGUs respects
reasonable systemic limitations, such as maintaining the expected net energy needed for load in a given
region.
States should provide information to EPA that documents the methodology and how the chosen
projection approach adequately addresses the analytic drivers described above, in the context of the
associated regulatory action and with the timeframe of the projections taken into account. For example,
the state should explain how its projections are consistent with expected changes in in economic and
regulatory drivers over the timeframe of the analysis that would affect fleet composition as well as
patterns of generation and associated emissions across the fleet composition as well as patterns of
generation and associated emissions across the fleet.
A state may also consider whether limited modifications to other EGU projections (such as those
provided by EPA, EIA, or ISO/RTOs) may reasonably address the analytic drivers described above. Such
an approach should include documentation of where and how the modified set of EGU projections
differs from the unmodified (original) set of projections, and why it is appropriate for the given
regulatory action or future scenario to make such modifications without conducting a more thorough or
complete analysis.
5.3.1.1 Examples of Existing ECU Projection Tools
There are specific tools and approaches available that can be used in ways that meet the general
principles and key criteria explained in Section 5.3.1. Mention of specific tools in this guidance should
not be taken as an endorsement or guarantee that any set of projections from any such tool will
necessarily meet the criteria described in this guidance that EPA will use to evaluate EGU emissions
projections, as any tool or model may be used appropriately or inappropriately for a given application.
Examples of EGU projection tools include.
The ERTAC EGU Growth Tool is an open-source tool developed by the ERTAC (Eastern Regional Technical
Advisory Committee) group of state and regional associations that benefits from an in-depth
collaboration among a large multi-state region. More information on this tool is available on the ERTAC
119
-------
EGU website (http://www.marama.org/2013-ertac-egu-forecasting-tool-documentation), including
sample files and documentation.
RFF's Haiku Model is a simulation of regional electricity markets and interregional electricity trade in the
continental United States. The model accounts for capacity planning, investment, and retirement over a
multi-year horizon and for system operation over seasons of the year and times of day. The model
identifies least-cost compliance strategies for compliance with various types of regulations of air
pollutants. Documentation of this model can be found at
http://www.rff.org/research/publications/haiku-documentation-rff-s-electricitv-market-model-version-
20
The Integrated Planning Model (IPM) is a multi-regional, dynamic, deterministic linear programming
model of the U.S. electric power sector. It provides forecasts of least cost capacity expansion, electricity
dispatch, and emission control strategies while meeting energy demand and environmental,
transmission, dispatch, and reliability constraints. The IPM can be used to evaluate the cost and
emissions impacts of proposed policies to limit emissions of S02, N0X, carbon dioxide (C02), mercury
(Hg), and HCI from the electric power sector. More information on IPM is available at the Power Sector
Modeling EPA website (https://www.epa.gov/airmarkets/power-sector-modeling).
The Market Allocation (MARKAL) model is a data-driven energy system model. NE-MARKAL is a state-
based and regionally specific version of MARKAL. NE-MARKAL represents each Northeast jurisdiction
and can examine the criteria air pollutant outcomes with plant-specific detail from various technology
options and overall system constraints. More information is available at:
http://www.nescaum.org/topics/ne-markal-model.
The National Energy Modeling System (NEMS) was developed by the EIA and it is used to produce
Annual Energy Outlook (AEO) projections of energy supply, demand and prices, as well as environmental
regulatory constraints and known future pollution control installations to create future year N0X, S02
and heat input estimates. More information on NEMS can be found at
https://www.eia.gov/forecasts/aeo/info nems archive.cfm.
The Regional Energy Deployment System (ReEDS). developed by National Renewable Energy Labs (NREL)
is a long-term capacity-expansion model for the deployment of electric power generation technologies
and transmission infrastructure throughout the contiguous United States. More information on ReEDS
can be found at http://www.nrel.gov/analysis/reeds .
The US Regional Economy. Greenhouse Gas, and Energy Model (US-REGEN) is a model developed by the
Electric Power Research Institute. It combines a detailed dispatch and capacity expansion model of the
United States electric sector with a high-level dynamic computable general equilibrium (CGE) model of
the United States economy, including sectoral detail in electric power production, energy demand, and
transportation. The US-REGEN is capable of modeling a wide range of environmental and energy policies
in both the electric and non-electric sectors. More information can be found at
http://eea.epri.com/models.html.
120
-------
A variety of proprietary load forecast optimization models are utilized by ISO's and system operators to
assess and control more limited and shorter term wholesale power flows across the grid. These tools
can provide very useful feedback to forecasts derived from more planning focused tools but may require
supplemental information to reflect the scale of geographies required for air quality inventory forecasts.
5,3,1,2 Existing Material for More Information
The resources listed in this section can provide additional information that may be helpful for EGU
emissions projections.
Guidance for 1-Hour S02 Nonattainment Area SIP Submissions. This document describes non-binding
recommendations on a wide range of issues that are likely to arise as state, local and tribal governments
develop nonattainment SIPs for the 1-hour S02 NAAQS. This document is available at
https://www.epa.gov/so2-pollution/guidance-l-hour-sulfur-dioxide-so2-nonattainment-area-state-
implementation-plans-sip.
Roadmap for Incorporating Energy Efficiency and Renewable Energy in SIPs/TIPs. This document
provides a roadmap to assist state, tribal and local agencies with accounting for and incorporating
energy efficiency (EE) and RE policies and programs in SIPs and TIPs. The roadmap accomplishes this task
by clarifying guidance the EPA issued in 2004 on incorporating EE/RE policies and programs into SIPs, as
well as related guidance EPA issued in 2004 and in 2005. This document is available at
https://www.epa.gov/energv-efficiencv-and-renewable-energv-sips-and-tips/energy-
efficiencvrenewable-energy-roadmap.
Resource for States on estimating EE/RE relative to ElA's Annual Energy Outlook for SIPs. This
document helps states identify their on-the-books energy efficiency and renewable energy policies and
estimate the incremental electricity savings of these policies; develop a methodology for projecting a
jurisdiction's energy demand both with and without the incremental electricity savings; and estimate
the change in power sector emissions attributable to the incremental electricity savings. For more
information visit: https://www.epa.gov/sites/production/files/2015-08/documents/
including ee and re policies in ed projections 03302015 final 508.pdf.
Assessing the Multiple Benefits of Clean Energy Programs: A Resource for States. This document helps
state energy, environmental, and economic policy makers identify the spectrum of EGU projection tools
and methodologies to quantify the energy, environmental (specifically greenhouse gas, criteria air
pollutants, and health), and economic benefits of clean energy. This document is available at
https://www.epa.gov/statelocalclimate/assessing-multiple-benefits-clean-energy-resource-states.
EPA's AVERT (Avoided Emissions geneRation Tool). This tool estimates the N0X, S02 and C02 EGU
emissions from implementing EE/RE programs and policies. This tool can estimate the EGU emission
impacts at the EGU, county, state and regional levels in the short-term future (1-5 years). Outputs from
AVERT are compatible with the SMOKE interface. For more information, visit:
https://www.epa.gov/avert.
121
-------
TSD for Incorporating RE and Demand-Side EE Impacts into State Plan Demonstrations for the Clean
Power Plan (CPP). This technical support document describes various quantitative approaches for
projecting the future year direct energy impacts (MWhs) of energy efficiency and renewable energy
programs and policies. This document is available at https://www.epa.gov/sites/production/files/2015"
11/documents/tsd-cpp-incorporating-re-ee.pdf
5,3,2 Non-EGU Stationary Sources
Non-EGU stationary sources include both the point and nonpoint sources, but the methods for
projecting them to the future can rely on many of the same sources of information.
Since 2006 (EPA, 2006a), the EPA has been assuming that emissions growth does not track with
economic growth for many stationary sources (both point and nonpoint). This "no-growth" assumption
is based on an examination of historical emissions and economic data. Emissions (as of 2005) had
declined for several years and those reductions could not be directly attributed to specific control
programs despite increasing economic-based growth factors for many metrics over the same time
period. While the EPA continues to work toward improving the projection approach in its own work, we
are still using this no-growth assumption for many emissions sectors. Of course, where air agencies have
better data about the expected growth in emissions or the possibility of sources reaching their permit
limits, that information should be included in developing more refined emissions projections.
The EPA creates and uses data that quantify assumed base year growth (projections), controls,
settlement agreements (or consent decrees), and other unit and facility-specific information on non-
EGU point and stationary area sources. These are included in the most recent emissions platform
Technical Support Document (TSD) available at the Emissions Modeling Clearinghouse. The TSDs include
the methodologies and background information for projecting sources such as Portland cement,
refineries, oil and gas, industrial/commercial/institutional boilers, residential wood combustion,
agricultural sources, fuel sulfur rules, corn and cellulosic ethanol and biodiesel plants, and several other
categories. The data sources include national sources with regional and state-specific information as
well as projection information provided by regional modeling centers and states and local air agencies.
In addition, the EPA encourages air agencies to consider the latest settlement agreements and consent
decrees that are updated periodically by the EPA's Office of Enforcement and Compliance Assurance
(OECA) at: https://www.epa.gov/enforcement/air-enforcement.
The additional resources described in Section 5.3.6 are also used by the EPA to develop projections used
in its modeling platforms. These same resources are available to state, local, and tribal agencies to
develop their own projection methods for non-EGU stationary sources.
Finally, for nonpoint sources, Section 3 of the EIIP projections document provides additional details
about projecting nonpoint-source. However, several of the references or statements in that document
are now out of date, as follows.
• The web sites for SCCs, AP-42 emission factors, and more information about area-source
inventories are out of date. These should be updated to the same sites as were listed in the
122
-------
point sources section above. Furthermore, this guidance includes more updated information
than the El IP report referenced.
• The ASEM model mentioned in Section 3.2 of the El IP projection document was not completed
and is not available.
• The references in Table 3.1-2 are out of date and should be updated as follows:
o U.S. EPA Emissions trends reports. These reports can be used to provide the historical
trends of emissions sectors based on the National Emissions inventory:
https://www.epa.gov/air-trends/.
o Multiple Projections System is no longer in use
o California Emission Forecasting System: this system is no longer available. Interested
parties should contact California Air Resources Board for more information on their
latest emissions projection approaches.
The tools and data sources described in Section 5.3.6 below should also be considered for use for
projection of point and nonpoint stationary source emissions.
5.3.3 Oil-load Mobile Sources
As described in Section 4.8.3, for all states but California, on-road mobile emissions should be estimated
with the latest EPA on-road mobile model, MOVES, following the EPA's latest guidance available at the
MOVES website. For California, the latest EPA-approved version of EMFAC should be used.
Emissions preparation with MOVES for projected emissions is much like preparation for a base year.
MOVES includes assumptions about emissions reductions associated with national emissions programs
including the effects of national vehicle and fuel standards. Some MOVES inputs, such as meteorology,
should be the same as for the base year inventory. Others, such as VMT and vehicle population should
be grown/reduced to reflect expected changes in population and activity. MOVES default estimates for
future VMT and vehicle population are based on national growth rates and should not be used for SIP
inventory purposes because growth rates vary substantially from county to county.
Because of the change to the using the MOVES model and the newer guidance provided on the OTAQ
websites, the El IP projections document is no longer relevant for on-road mobile sources.
5.3.4 Nonroad Mobile Equipment
Section 5 of the El IP projections document provides details about projecting nonroad mobile sources.
While some information is still valid, several of the references or statements in that document are now
out of date, as follows.
• The name "Office of Mobile Sources" has changed to "OTAQ" and it has a new primary website.
• The EPA's NONROAD model has been incorporated into the MOVES model and should no longer
be used for new SIP development.
• According to a California Air Resources Board website, category-specific inventory models and
methods have replaced most uses of California's OFFROAD model. The information on
California's nonroad emissions is available at a different website URL. A link to the
123
-------
OFFROAD2007 model is also available from this page to be used for categories not listed on that
page.
MOVES includes assumptions about emissions reductions associated with nonroad source national
emissions programs. MOVES also includes assumptions about national growth in nonroad activity and
equipment population. Local estimates of future activity and equipment populations should be
substituted for the national estimates when possible.
5,3,5 Other Nonroad Mobile Sources
Aircraft, locomotives, and CMV sources of emissions do not have a specific model designed for
projections, and so additional effort is needed for these three inventory components.
For aircraft emissions, FAA data can be used to estimate growth for aircraft emissions. The FAA provides
summary historical and forecast statistics on passenger demand and aviation activity at U.S. airports.
The summary level forecasts are based on individual airport projections. The Terminal Area Forecast
(TAF) includes forecasts for active airports in the National Plan of Integrated Airport Systems (NPIAS).
The TAF system can be accessed at the FAA's TAF query page. The TAF system allows users to create
their own forecast scenarios. The TAF database, which contains a query data application, allows the
public to access and print historical and forecast (up to 2040) aviation activity data by individual airport,
state or FAA region. Air agencies can use this information to adjust their base year emissions based on
projected airport activity.
For locomotives and marine vessel emissions, the best source of data on the impact of national rules
from OTAQ is the websites provided for each of the rules and their associated Federal Register notices.
As described for the base year emissions in Section 4.8.5, the locomotive marine rule information is
included at the EPA website for that regulation. For CMV sources, Table 20 (from the base year
emissions section) provides a good list of individual rules and references. Since each rule is different,
there is no single guidance that can be made about how these should be incorporated. However, the
EPA incorporates these data into their modeling platforms, and the most recent projection
methodologies are documented in the TSDs included with the modeling platforms on the Emissions
Modeling Clearinghouse at https/www.epa.gov/air-emissions-modeling .
Additionally for C3 vessels, emissions data are available from an Emissions Control Area-International
Maritime Organization (ECA-IMO) collaboration. These emissions are available at a fine-grid resolution
and are subject to various engine and fuel standards in the future. Like base year emissions, air agencies
should consider the state versus federal waters when deciding what parts of the ECA-IMO data to use
(see Section 4.8.5.2). Growth/control factors have been developed for this C3 CMV inventory, and the
EPA is using them for future year C3 CMV inventory development. New approaches are likely to develop
over time, so air agency staff should consult the web pages listed here for the latest available
information. The latest EPA modeling platform projection methodology for C3 marine emissions is also
available in the latest emissions modeling platform TSD at the Emissions Modeling Clearinghouse
mentioned previously.
124
-------
5,3,6 Other Projection Resources
In addition to sector-specific resources, there are other sources of information that projection
developers can use for more than a single sector as information to help determine the most applicable
growth rates for the modeling region. These include the Bureau of Labor Statistics "Employment
Outlook/' Census Bureau data, Trade Organizations and individual facilities, and the Chemical Economics
Handbook. The following paragraphs provide a brief description and additional references for each of
the sources of data just listed.
Sources for EPA nonpoint methods
As described in Section 4.8.2, the EPA develops methods for numerous nonpoint methods. Each of these
methods depends on some surrogate activity. Many of these activity sources come from USDA, DOE,
U.S. Census Bureau, and Department of Labor databases available online. Appendix B provides a list of
nonpoint sectors and associated activity reports and websites. These data do not include future-year
projections, but may include newer information than were used for estimating the base year
information. As such, these data can help with understanding the possible changes in activity (and
perhaps emissions) from the date of the underlying base year activity data to the present day. It should
be noted that emissions for 1 year (e.g., 2011) may be estimated using activity data from a prior year
(e.g., 2009) because of timing needed for preparing the data does not match with the timing of the
release of the underlying data. When emissions are estimated using data earlier than the emissions base
year, it further strengthens the benefit of newer activity data. Furthermore, these datasets often include
historical trends, which could be of use in understanding what may happen in the future.
Since the base year emissions for nonpoint methods are calculated based on direct activity indicators
such as population, fuel consumption, employment, and VMT, projections of these direct activity
indicators may also be used to estimate future emissions. Population is related to sectors such as
consumer products, architectural coatings, portable fuel containers, and commercial cooking. VMT also
plays a major role in calculation of paved road emissions. Growth projections for population and fuel
consumption are readily available from the U.S. Census and the DOE's Annual Energy Outlook (see
below). VMT projections are provided periodically as part of the EPA emissions modeling platforms and
may be available in some states from other sources. Employment projections may also be available in
some states and could be used for industrial categories where the state believes that employment is an
appropriate surrogate for facility activity and therefore emissions growth or retraction.
MARKAL
The MARKAL model is a public sector, data-driven, energy-system optimization model. A user inputs the
structure of the energy system to be modeled and data to characterize each of the technologies and
resources included in the model run. MARKAL calculates the least cost set of technologies over time to
satisfy specific demands, using linear and mixed-integer linear programming techniques. Outputs of the
model include the technological mix at intervals in the future, energy demand, and estimates of criteria
and GHG emissions, among others. The DOE uses MARKAL for its System for the Analysis of Global
Energy Markets (SAGE) model. SAGE is used to produce ElA's annual International Energy Outlook.
125
-------
The EPA has applied MARKAL for technology and emissions evaluations, which required the
development of the EPA MARKAL technology database (EPANMD). This technology-rich database
represents the major sectors in the U.S. energy system, including the commercial, industrial, residential,
transportation, and electricity generation sectors. The DOE's AEO and the NEMS were used to construct
the energy supply, demand, and technology characterizations. Data for technologies not represented in
the AEO and NEMS were derived from other widely recognized authoritative sources (e.g., Electric
Power Research Institute's Technical Assessment Guide and DOE's Office of Transportation Technology's
Quality Metrics report). More information on the EPA's use of MARKAL and the EPANMD can be found
on the website for "EPAUS9R - An Energy Systems Database for use with the Market Allocation
(MARKAL) Model" and EPA report number NRMRL-RTP-375 (EPA, 2006b).
Control Strategy Tool (CoST)
The purpose of CoST is to estimate the emission reductions and engineering costs associated with
control strategies applied to point, nonpoint, and mobile sources of air pollutant emissions to support
the analyses of air pollution policies and regulations. Control strategy results can be exported to CSV
files or viewed in a graphical table that supports sorting, filtering, and plotting. The results can also be
merged with the original inventory to create controlled emissions inventories that can be exported to
files that can be input to the emissions model SMOKE. CoST allows for user-supplied control measures
data, provides tracking of analyses and outputs, and supports built-in QA.
The Control Measure dataset focuses on criteria pollutants and their precursors, and it includes primary
control efficiencies, co-pollutant control efficiencies, and control costs. The data also includes an
extensive set of mobile source control measures. More information on the Control Measure dataset and
CoST is available at https://www.epa.gov/economic-and-cost-analvsis-air-pollution-regulations/cost-
analvsis-modelstools-air-pollution.
Annual Energy Outlook
The ElA's AEO forecast data was a primary input for the final version of EGAS. However, with EGAS no
longer being supported by the EPA, the latest versions of the AEO are the best available source of
socioeconomic forecast data for many industrial sectors. The latest AEO data are available at:
http://www.eia.gov/forecasts/aeo/er/index.cfm.
Census Bureau
The U.S. Census Bureau provides data on total employees, total payroll, number of establishments, and
value of shipments by NAICS code (2 digit - 6 digit) for specific years. This database is ideal for examining
changes in the number of establishments between the two latest available years and also for historical
changes. Therefore, the actual census data are useful primarily to evaluate emissions changes from the
base year to the present (or most recently available year) and comparing to the data included in the
AEO. The census data are available at http://www.census.gov/2010census/data/.
126
-------
In addition, the Census Bureau projects future population, which is available at
http://www.census.gov/population/proiections/. This information can be used to help project nonpoint
emissions categories that include population as an activity input for the base year emissions.
Trade Organizations and Individual Facilities
Trade organizations can also be a helpful resource because they often have projections for future-year
growth of their industry and may be aware of pending facility closures or new facility construction. This
resource is most relevant for large industrial sources that usually are included in point source
inventories. In addition, emissions modelers should consider contacting large sources of emissions in
their modeling region for their expectations of emissions growth or reduction. In many cases, large
industries may be willing to provide such information when presented with what will be used if no
additional information can be provided. Additional information may be found in permit applications for
future development or controls at individual facilities. Except in the case of permit applications,
emissions projections developers should be aware of possible conflicts of interest for this source of
information. For example, industry groups or facilities may have an incentive to under-represent future-
year emissions in hopes of avoiding additional requirements for reducing emissions. However, many
industrial representations have been valuable stakeholders who are interested in providing accurate
information.
Chemical Economics Handbook
The Chemical Economics Handbook, produced by SRI Consulting (a part of IHS, Inc.27), is a series of
reports on prices, production and consumption of hundreds of chemical industry products and
commodities. Past and current information on chemical products and commodities is available, and
projections of future prices, production and consumption are often available. Reports on specific
industries are also available. Reports at an industry level can often be used to verify the efficacy of
future industry modeling results. Each report is updated every 3 years. Projections, some up to 5 years
from the current day, are often prepared using proprietary methods. Reports are available by
subscription, and can be obtained as hard copy, CD, or at
http://www.sriconsulting.com/CEH/Public/index.html.
5,4 Evaluate and Refine Data for Projections In Inventory Region
For key sectors, emissions projections developers should determine what information will impact the
projection results the most, and ensure that the data values reflect conditions or expectations of the
modeling region. The key information is identified based on a combination of base-year emissions data
and growth and control rates. An iterative process is helpful to complete this step, as follows:
1. Estimate future-year emissions by combining the base-year emissions with the projection
information (as described in the El IP document referenced in Section 5.1).
27 A number of other industry reports appears to be available from the IHS home page from the "Industries" tab.
Some of these websites appear to have databases available for a fee, but they have yet not been assessed by EPA
for applicability to emissions projections.
127
-------
2. Review future-year emissions data to identify large sectors of key pollutants (e.g. NOx, S02,
PM2.5, NH3 and VOC). Emissions summaries by state/SCC or state/NAICS for point sources can be
helpful for this step to consolidate the data to review.
3. For the largest future-year emissions sectors, review the growth and control assumptions,
compare among all available data sources, and/or collect additional information to refine the
assumptions. A representative growth rate should be identified from the available data sources
and all information known about the sources and sectors. Stakeholder review of the data can be
helpful during this step; for example, an industrial facility with large projected emissions may be
able to review the data and provide additional information (e.g., expected capacities, fuels or
controls) for a more informed future-year emissions estimate.
4. Additionally, data developers should also compare the future-year emissions to base-year
emissions, and if economic-based growth is used, to identify overall growth rates for industrial
sectors and scrutinize excessively high growth rates, especially when associated with significant
emissions.
5. Using the new information, repeat step 1 and continue these steps until a satisfactory future-
year projection across the entire inventory has been completed.
Emissions models (e.g., CoST, SMOKE, MOVES, IPM, etc.) provide the capability to create future-year
inventories using base year inventories and projection information. Emissions modelers will need to
convert the projected emissions data into specific formats for input to these models. Prior to starting
this step, data developers should determine which emissions model will be used to perform the
calculations and make sure that the type of information needed by the model is being collected.
5,5 QA for Projected Inventories
As mentioned in the previous subsection, final review of the inventories is an iterative process. During
each iteration, adaptations of the methods described in Section 4.9 can be used for QA of projected
emissions inventories. The following list gives examples of some of the ways of the methods for base
year inventories can be adapted for projected emissions inventories.
• The same checks used for finding structure errors for base year emissions should be used for
ensuring the projected inventories use the correct structures. The EIS checks are less relevant
since projected emissions are more likely to be in a modeling format (since they are used for
modeling) than are base year inventories. Even so, data developers should make sure that data
structure or formatting problems do not lead to incorrect emissions data being used in planning
inventories or modeling.
• Like base year inventories, data codes should be compared to lists of valid codes. It is just as
important to have future-year SCCs be valid as it is for base year SCCs.
• Ranking techniques can be used to identify facilities, units, and processes that have been grown
to unrealistic values. Ranking sources by emissions values can identify any that rise to the top of
the list of largest sources without a good reason why. Ranking techniques can also be used for
projected nonpoint emissions data.
128
-------
• Inventory comparisons can be made between projected and base year inventories to help make
sure:
o Location information is consistent between base and future year (needed for modeling
only).
o Release point parameters are consistent between base and future year. Where controls
have been added, one may expect that the release point parameters could change, for
example, with higher stacks or lower exit gas temperatures (needed primarily for
modeling).
o Emissions increases and decreases at the facility, unit, and process level are consistent
with what is expected.
o Emissions increases and decreases at the county and county/SCC levels are consistent
with what is expected. If certain sectors are being held constant (i.e., no better
data/approaches are available), then data developers can confirm that these sectors did
not change emissions.
129
-------
6 References
Allen, D. J., Pickering, K. E., Pinder, R. W., Henderson, B. H., Appel, K. W., and Prados, A.: Impact of
lightning-NO on Eastern United States photochemistry during the summer of 2006 as
determined using the CMAQ model, Atmos. Chem. Phys. Discuss., 11, 17699-17757,
doi:10.5194/acpd-ll-17699-2011, 2011.
Appel, K. W., Foley, K. M., Bash, J. O., Pinder, R. W., Dennis, R. L., Allen, D. J., and Pickering, K.: A multi-
resolution assessment of the Community Multiscale Air Quality (CMAQ) Model v4.7 wet
deposition estimates for 2002-2006, Geosci. Model Dev. Discuss., 3, 2315-2360,
doi:10.5194/gmdd-3-2315-2010, 2010.
EPA, 2006a. Regulatory Impact Analyses, 2006 National Ambient Air Quality Standards for Particle
Pollution. U.S. Environmental Protection Agency, Office of Air Quality Planning and Standards,
October, 2006. Docket # EPA-HQ-OAR-2001-0017, # EPAHQ-OAR-2006-0834. Appendix D:
http://www.epa.gov/ttn/ecas/regdata/RIAs/Appendix%20D--lnventorv.pdf.
EPA, 2006b. Shay, C. L., S. Yeh, J. Decarolis, D. H. Loughlin, C. L. Gage, And E. Wright. EPA U.S. National
MARKAL Database: Database Documentation. U.S. Environmental Protection Agency,
Washington, DC, EPA/600/R-06/057, 2006.
Fu, J.S., Dong, X., Huang, K., Tong, D., and Zhuang, G.: Model development of dust emission and
heterogeneous chemistry within the Community Multiscale Air Quality modeling system and
its application over East Asia, Atmospheric Chemistry and Physics, 16, 8157-8180, 2016.
Reid, S.B., Technical Memorandum, Sonoma Technology, Inc., Preparation of Version 2 of the Wildland
Fire Emissions Inventory for 2011, April 26, 2013.
130
-------
Appendix A
Emissions Inventory System - Quick Reference
How to Create County and Facility NOx and VOC Summaries for Select Counties
For access to the Emissions Inventory System (EIS), refer to
https://www.epa.gov/air-emissions-inventories/emission-inventorv-SYstem-eis-gatewaY.
See also Online EIS training "Reports - May 2012" at
https://www.epa.gov/air-emissions-inventories/emission-inventorv-training
Create a summary of NOx and VOC emissions BY SECTOR and COUNTY for a group of counties
1. Select "Request Reports" from the left hand sidebar on the EIS Gateway.
2. Under the Emissions Summaries heading, select "By Geography."
3. Select Report Type "County by Sector."
4. Click the Data Set or Selection bar. This brings you to a "filter" that allows you to select an
inventory on which to run a report.
To find a version of the NEl, type "NEl V" in the search box under the Data Set or Selection.
Then, click the appropriate row for the year and version of the inventory for your needs.
5. Click on the Pollutants bar to access the pollutant filter. Use the search box provided and type in
"VOC" and select that resulting row for VOC (click on it with the mouse). Repeat this for NOx.
Typing both filters at the same time will not work.
6. Because we are looking at counties, go directly to the County filter. DO NOT SELECT A STATE
FIRST from the State filter (this will select all counties in selected states, even if you then select
specific counties in the county filter). Use the search box in the County filter to find the counties
of interest either by name or State/County FIPS code. Select the counties of interest by clicking
on those rows. If no counties are selected, all counties will be returned.
7. Lastly, select your Sectors using the Sectors search box. If no sectors are selected, the report will
be created with all sectors. The EIS Sectors are explained in more detail in Section of the 2011
NEI technical documentation.
8. Click the "Request Download" button at the bottom of the screen. These reports are run by EIS
every 15 minutes and are available under the Report Download section on the left hand sidebar
of the Gateway.
Note: Using the "Run Report" button will give you a faster response and display the summary on
the screen, but may not include all records for your report. A warning is given at the top of the
screen if the "Run Report" result is incomplete.
Create a summary of NOx and VOC emissions BY FACILITY for a group of counties
1. Select "Request Reports" from the left hand sidebar on the EIS Gateway.
2. Under the Emissions Summaries heading, select "By Facility."
A-l
-------
3. Select Report Type "Facility". For more detailed (sub-facility) data, you may select "Emissions
Unit" or "Emissions Process."
4. Follow Step 4 in the list above.
5. Follow Step 5 in the list above.
6. Follow Step 6 in the list above.
7. Lastly, you may refine your search even more using the optional filters of Facility Types, NAICS,
Regulations, or Source Classifications. If you do not use these filters, all facilities for the selected
counties that include NOx and VOC emissions will be returned in the report.
8. Follow Step 8 in the list above.
A-2
-------
Appendix B
EPA Nonpoint Methods and Associated Activity Data and
ites
B-l
-------
Category
Data Type
Last
Release
Date
Data Lag
(years)
Report
Source
Link
Agriculture Production - Livestock
Livestock
Population
May 2014
(2012 data)
2
Census of Agriculture
USDA
https://www.agcensus.usda.gov/
Agriculture Production - Livestock
Livestock
Population
Continuous
Updates
USDA NASS Quick Stats
USDA
https://www.nass.usda.gov/Quick_Stats/
Asphalt Paving - Cutback
Cutback
Asphalt Use
2008 Asphalt Usage Survey for the
United States and Canada
Asphalt Institute
http://www.asphaltinstitute.org/
Asphalt Paving - Cutback
VMT
2014
2
Highway Statistics 2014
FHWA
http://www.fhwa.dot.gov/policyinformation/quickfinddata/qftravel.cfm
Asphalt Paving - Emulsified
Emulsified
Asphalt Use
2008 Asphalt Usage Survey for the
United States and Canada
Asphalt Institute
http://www.asphaltinstitute.org/
Asphalt Paving - Emulsified
VMT
2014
2
Highway Statistics 2014
FHWA
http://www.fhwa.dot.gov/policyinformation/quickfinddata/qftravel.cfm
Aviation Gasoline Distribution -
Stage 1
AvGas
Consumption
Sept 2015
(2014 data)
1.5
Petroleum Supply Annual
EIA
http://www.eia.gov/petroleum/supply/annual/volume1/
Aviation Gasoline Distribution -
Stage 2
AvGas
Consumption
Sept 2015
(2014 data)
1.5
Petroleum Supply Annual
EIA
http://www.eia.gov/petroleum/supply/annual/volume1/
Commercial Cooking
Population
June 2016
(2015 data)
1.5
Population Estimates Program
GCT-T1: Population Estimates
U.S. Census
Bureau
http://www.census.gov/popest/
Construction Dust -
NonResidential Construction
Value of
Construction
April 2016
0
Annual Value of Construction Put in
Place
U.S. Census
Bureau
http://www.census.gov/construction/c30/c30index.html
Construction Dust -
NonResidential Construction
Business
Patterns
2014
2
County Business Patterns (e.g., no.
establishments, employees)
U.S. Census
Bureau
http://www.census.gov/econ/cbp/index.html
Construction Dust -
NonResidential Construction
Construction
Costs
June 2016
1 mo.
Producer Price Index
U.S. Dept. of
Labor
http://www.bls.goV/ppi/#tables
Construction Dust - Residential
Construction
Housing Units
(2014 data)
2
Annual Housing Units Authorized by
Building Permits CO201 OA
U.S. Census
Bureau
https://www.census.gov/popest/data/housing/totals/2014/
Construction Dust - Residential
Construction
Housing Starts
June 2016
2 mo.
New Privately Owned Housing Units
Started for previous quarter
U.S. Census
Bureau
http://www.census.gov/construction/nrc/index.html
Construction Dust - Residential
Construction
Ratio for 5 or
more units
May 2016
(2015 data)
5 mo
Table 2au. New Privately Owned
Housing Units Authorized
Unadjusted Units for Regions,
Divisions, and States 2009
U.S. Census
Bureau
http://www.census.gov/construction/bps/stateannual.html
Construction Dust - Residential
Construction
Types of
Foundation
(2015 data)
Type of Foundation in New One-
Family Houses Completed
U.S. Census
Bureau
http://www.census.gov/construction/chars/completed.html
(see row for "Foundation").
Construction Dust - Road
Construction
VMT
2014
2
Highway Statistics 2014
FHWA
http://www.fhwa.dot.gov/policyinformation/quickfinddata/qftravel.cfm
B-l
-------
Category
Data Type
Last
Release
Date
Data Lag
(years)
Report
Source
Link
Construction Dust - Road
Construction
Housing Starts
June 2016
2 mo.
New Privately Owned Housing Units
Started for previous quarter
U.S. Census
Bureau
http://www.census.gov/construction/nrc/index.html
Fertilizer Application
Commercial
Fertilizer Usage
2012
Commercial Fertilizers
Association of
American Plant
Food Control
Officials
http://www.aapfco.Org/publications.html#comm (must order)
Fuel Combustion - Commercial
Fuel Sales
2014
2
Fuel Oil and Kerosene Sales
EIA
http://www.eia.gov/dnav/pet/pet_cons_821use_dcu_nus_a.htm
Fuel Combustion - Commercial
Business
Patterns
2014
2
County Business Patterns (e.g., no.
establishments, employees)
U.S. Census
Bureau
http://www.census.gov/econ/cbp/index.html
Fuel Combustion - Commercial
Gov't Payrolls
2016
(2012 data)
4
Census of Governments
U.S. Census
Bureau
https://www.census.gov/govs/cog/
Fuel Combustion - Commercial
Fuel
Consumption
2014
2
State Energy Data System (SEDS)
EIA
http://www.eia.gov/state/seds/
Fuel Combustion - Industrial
Fuel Sales
2014
2
Fuel Oil and Kerosene Sales
EIA
http://www.eia.gov/dnav/pet/pet_cons_821use_dcu_nus_a.htm
Fuel Combustion - Industrial
Business
Patterns
2014
2
County Business Patterns (e.g., no.
establishments, employees)
U.S. Census
Bureau
http://www.census.gov/econ/cbp/index.html
Fuel Combustion - Industrial
Energy
Consumption
March 2012
(2010 data)
3.5
Energy consumption as a fuel, by
industry and region
EIA
http://www.eia.gov/consumption/manufacturing/
Fuel Combustion - Industrial
Fuel
Consumption
2014
2
State Energy Data System (SEDS)
EIA
http://www.eia.gov/state/seds/
Gasoline Distribution - PFCs
Gasoline Distribution - Stage 1
Business
Patterns
2014
2
County Business Patterns (e.g., no.
establishments, employees)
U.S. Census
Bureau
http://www.census.gov/econ/cbp/index.html
Gasoline Distribution - Stage 1
Fuel
Consumption
Sept 2015
(2014 data)
1.5
Petroleum Supply Annual
EIA
http://www.eia.gov/petroleum/supply/annual/volume1/
Gasoline Distribution - Stage 2
Industrial Fuel Combustion
Business
Patterns
2014
2
County Business Patterns (e.g., no.
establishments, employees)
U.S. Census
Bureau
http://www.census.gov/econ/cbp/index.html
Industrial Fuel Combustion
Fuel Sales
2011
2
Fuel Oil and Kerosene Sales
EIA
http://www.eia.gov/dnav/pet/pet_cons_821use_dcu_nus_a.htm
Industrial Fuel Combustion
Energy
Consumption
June 2009
3.5
2006 Manufacturing Energy
Consumption Survey
EIA
http://www.eia.gov/emeu/mecs/mecs2006/2006tables.html
Industrial Fuel Combustion
Fuel
Consumption
2014
2
State Energy Data System (SEDS)
EIA
http://www.eia.gov/state/seds/
Open Burning - Household Waste
Solid Waste
Generation
2011
2
Municipal Solid Waste in the United
States: 2009 Facts and Figures
EPA
https://nepis.epa.gov/Exe/ZyPDF.cgi/P100GMT6.PDF?Dockey=P10
0GMT6.PDF
B-2
-------
Category
Data Type
Last
Release
Date
Data Lag
(years)
Report
Source
Link
Open Burning - Household Waste
Population
June 2016
(2015 data)
1.5
Population Estimates Program
GCT-T1: Population Estimates
U.S. Census
Bureau
http://www.census.gov/popest/
Open Burning - Land Clearing
Business
Patterns
2014
2
County Business Patterns (e.g., no.
establishments, employees)
U.S. Census
Bureau
http://www.census.gov/econ/cbp/index.html
Open Burning - Land Clearing
Housing Units
(2014 data)
2
Annual Housing Units Authorized by
Building Permits CO201 OA
U.S. Census
Bureau
https://www.census.gov/popest/data/housing/totals/2014/
Open Burning - Land Clearing
Value of
Construction
2016
0
Annual Value of Construction Put in
Place
U.S. Census
Bureau
https://www.census.gov/construction/c30/c30index.html
Open Burning - Land Clearing
Housing Starts
June 2016
2 mo.
New Privately Owned Housing Units
Started for previous quarter
U.S. Census
Bureau
http://www.census.gov/construction/nrc/index.html
Open Burning - Land Clearing
Ratio for 5 or
more units
May 2016
(2015 data)
5 mo
Table 2au. New Privately Owned
Housing Units Authorized
Unadjusted Units for Regions,
Divisions, and States 2009
U.S. Census
Bureau
http://www.census.gov/construction/bps/stateannual.html
Open Burning - Leaves and Brush
Solid Waste
Generation
2013
3
Municipal Solid Waste in the United
States: 2013 Facts and Figures
EPA
https://www.epa.gov/smm/advancing-sustainable-materials-
management-facts-and-figures-report
Open Burning - Leaves and Brush
Population
June 2016
(2015 data)
1.5
Population Estimates Program
GCT-T1: Population Estimates
U.S. Census
Bureau
http://www.census.gov/popest/
Paved Road Dust
VMT
2014
2
Highway Statistics 2014
FHWA
http://www.fhwa.dot.gov/policyinformation/quickfinddata/qftravel.cfm
Publicly Owned Treatment Works
Nationwide
Flow rate
January
2016
4
Clean Watersheds Needs Survey
2012 Data and Reports, Detail
Report
EPA
https://www.epa.gov/cwns/clean-watersheds-needs-survey-cwns-
2012-report-and-data#access
Publicly Owned Treatment Works
Population
June 2016
(2015 data)
1.5
Population Estimates Program
GCT-T1: Population Estimates
U.S. Census
Bureau
http://www.census.gov/popest/
Residential Coal
Coal
Distribution
Dec 2016
(2014 data)
2
Annual Coal Distribution
EIA
http://www.eia.gov/cneaf/coal/page/coaldistrib/a_distributions.html
Residential Coal
Fuel
Consumption
2014
2
State Energy Data System (SEDS)
EIA
http://www.eia.gov/state/seds/
Residential Coal
House Heating
Fuel Type
2014
2
House Heating Fuel (online report
from "Advanced Search")
U.S. Census
Bureau
http://factfinder.census.gov/faces/nav/jsf/pages/searchresults.xhtml?
refresh=t#none (Enter: "heating fuel" in search field)
Residential Distillate Fuel
Fuel
Consumption
2014
2
State Energy Data System (SEDS)
EIA
http://www.eia.gov/state/seds/
Residential Distillate Fuel
House Heating
Fuel Type
2014
2
House Heating Fuel (online report
from "Advanced Search")
U.S. Census
Bureau
http://factfinder.census.gov/faces/nav/jsf/pages/searchresults.xhtml?
refresh=t#none (Enter: "heating fuel" in search field)
B-3
-------
Category
Data Type
Last
Release
Date
Data Lag
(years)
Report
Source
Link
Residential Kerosene
Fuel
Consumption
2014
2
State Energy Data System (SEDS)
EIA
http://www.eia.gov/state/seds/
Residential Kerosene
House Heating
Fuel Type
2014
2
House Heating Fuel (online report
from "Advanced Search")
U.S. Census
Bureau
http://factfinder.census.aov/faces/nav/isf/paaes/searchresults.xhtml?
refresh=t#none (Enter: "heatina fuel" in search field)
Residential LPG
Fuel
Consumption
2014
2
State Energy Data System (SEDS)
EIA
http://www.eia.gov/state/seds/
Residential LPG
House Heating
Fuel Type
2014
2
House Heating Fuel (online report
from "Advanced Search")
U.S. Census
Bureau
http://factfinder.census.aov/faces/nav/isf/paaes/searchresults.xhtml?
refresh=t#none (Enter: "heatina fuel" in search field)
Residential Natural Gas
Fuel
Consumption
2014
2
State Energy Data System (SEDS)
EIA
http://www.eia.gov/state/seds/
Residential Natural Gas
House Heating
Fuel Type
2014
2
House Heating Fuel (online report
from "Advanced Search")
U.S. Census
Bureau
htto://factfinder.census.aov/faces/nav/isf/Daaes/searchresults.xhtml?
refresh=t#none (Enter: "heatina fuel" in search field)
Residential Wood
Timber Output
2012
4
Timber Products Output Reports
USDA
http://srsfia2.fs.fed.us/php/tpo_2009/tpo_rpa_int1.php
Residential Wood
House Heating
Fuel Type
2014
2
House Heating Fuel (online report
from "Advanced Search")
U.S. Census
Bureau
http://factfinder.census.aov/faces/nav/isf/paaes/searchresults.xhtml?
refresh=t#none (Enter: "heatina fuel" in search field)
Residential Wood
Heating
Equipment
Type
2013
3
American Housing Survey for the
United States: 2013 Summary
Tables, Table C-03-AO: Heating, Air
Conditioning, and Appliances - All
Occupied Units
U.S. Census
Bureau
http://www.census.gov/programs-surveys/ahs/data.html
Residential Wood
Metropolitan
Characteristics
2013
3
American Housing Survey
Metropolitan Data
U.S. Census
Bureau
http://www.census.gov/programs-surveys/ahs/data/2013/ahs-2013-
summary-tables/metropolitan-summary-tables—ahs-2013.html
Solvent Usage - Other
Population
June 2016
(2015 data)
1.5
Population Estimates Program
GCT-T1: Population Estimates
U.S. Census
Bureau
http://www.census.gov/popest/
Solvent Usage - Other
Business
Patterns
2014
2
County Business Patterns (e.g., no.
establishments, employees)
U.S. Census
Bureau
http://www.census.gov/econ/cbp/index.html
Solvent Usage - Surface Coatings
Population
June 2016
(2015 data)
1.5
Population Estimates Program
GCT-T1: Population Estimates
U.S. Census
Bureau
http://www.census.gov/popest/
Solvent Usage - Surface Coatings
Business
Patterns
2014
2
County Business Patterns (e.g., no.
establishments, employees)
U.S. Census
Bureau
http://www.census.gov/econ/cbp/index.html
Unpaved Road Dust
VMT
2014
2
Highway Statistics 2014
FHWA
http://www.fhwa.dot.gov/policyinformation/quickfinddata/qftravel.cfm
B-4
-------
United States Office of Air Quality Planning and Standards Publication No. EPA-454/B-17-003
Environmental Protection Air Quality Assessment Division July, 2017
Agency Research Triangle Park, NC
------- |