Technical Support Document (TSD)

Preparation of Emissions Inventories for the 2016vl North American

Emissions Modeling Platform

March 2021

Contacts:

Alison Eyth, Jeff Vukovich, Caroline Farkas

U.S. Environmental Protection Agency
Office of Air Quality Planning and Standards
Air Quality Assessment Division
Emissions Inventory and Analysis Group
Research Triangle Park, North Carolina


-------
TABLE OF CONTENTS

LIST OF TABLES	IV

LIST OF FIGURES	VII

LIST OF APPENDICES	VIII

ACRONYMS	IX

1	INTRODUCTION	12

2	EMISSIONS INVENTORIES AND APPROACHES	15

2.1	2016 POINT SOURCES (PTEGU, PT_OILGAS, PTNONIPM, AIRPORTS)	19

2.1.1	EGUsector (ptegu)	21

2.1.2	Point source oil and gas sector (pt oilgas)	22

2.1.3	Non-IPM sector (ptnonipm)	25

2.1.4	A ircraft and ground support equipment (airports)	28

2.2	2016 NONPOINT SOURCES (AFDUST, AG, NP_OILGAS, RWC, NONPT)	29

2.2.1	Area fugitive dust sector (afdust)	29

2.2.2	Agriculture Sector (ag)	36

2.2.2.1	Livestock Waste Emissions	37

2.2.2.2	Fertilizer Emissions	38

2.2.3	Nonpoint Oil and Gas Sector (np oilgas)	41

2.2.4	Residential Wood Combustion (rwc)	43

2.2.5	Nonpoint (nonpt)	44

2.3	2016 Onroad Mobile sources (onroad)	48

2.4	2016 Nonroad Mobile sources (cmv, rail, nonroad)	61

2.4.1	Category 1, Category 2 Commercial Marine Vessels (cmv_clc2)	61

2.4.2	Category 3 Commercial Marine Vessels (cmv_c3)	64

2.4.3	Rail Sources (rail)	68

2.4.4	Nonroad Mobile Equipment Sources (nonroad)	77

2.5	2016 Fires (ptfire, ptagfire)	83

2.5.1	Wild and Prescribed Fires (ptfire)	83

2.5.2	Point source Agriculture Fires (ptagfire)	90

2.6	2016 Biogenic Sources (beis)	93

2.7	Sources Outside of the United States	95

2.7.1	Point Sources in Canada and Mexico (othpt)	95

2.7.2	Fugitive Dust Sources in Canada (othafdust, othptdust)	95

2.7.3	Nonpoint and Nonroad Sources in Canada and Mexico (othar)	96

2.7.4	Onroad Sources in Canada and Mexico (onroad can, onroadjnex)	96

2.7.5	Fires in Canada and Mexico (ptfire othna)	96

2.7.6	Ocean Chlorine	96

3	EMISSIONS MODELING	97

3.1	Emissions modeling Overview	97

3.2	Chemical Speciation	101

3.2.1	VOC speciation	104

3.2.1.1	County specific profile combinations	107

3.2.1.2	Additional sector specific considerations for integrating HAP emissions from inventories into speciation	108

3.2.1.3	Oil and gas related speciation profiles	Ill

3.2.1.4	Mobile source related VOC speciation profiles	112

3.2.2	PM speciation	117

3.2.2.1 Mobile source related PM2.5 speciation profiles	118

3.2.3	NO x speciation	120

3.2.4	Creation of Sulfuric Acid Vapor (SULF)	120

3.3	Temporal Allocation	122

3.3.1	Use ofFFlO format for finer than annual emissions	123

3.3.2	Electric Generating Utility temporal allocation (ptegu)	124

3.3.2.1 Base year temporal allocation of EGUs	124

ii


-------
3.3.2.2 Future year temporal allocation of EGUs	128

3.3.3	Airport Temporal allocation (airports)	134

3.3.4	Residential Wood Combustion Temporal allocation (rwc)	136

3.3.5	Agricultural Ammonia Temporal Profiles (ag)	140

3.3.6	Oil and gas temporal allocation (npoilgas)	141

3.3.7	Onroad mobile temporal allocation (onroad)	141

3.3.8	Nonroad mobile temporal allocation(nonroad)	146

3.3.9	Additional sector specific details (afdust, beis, cmv, rail, nonpt, ptnonipm, ptfire)	147

3.4	Spatial Allocation	149

3.4.1	Spatial Surrogates for U.S. emissions	149

3.4.2	Allocation method for airport-related sources in the U.S.	155

3.4.3	Surrogates for Canada and Mexico emission inventories	155

3.5	Preparation of Emissions for the CAMx model	159

3.5.1	Development of CAMx Emissions for Standard CAMx Runs	159

3.5.2	Development of CAMx Emissions for Source Apportionment CAMx Runs	161

4	DEVELOPMENT OF FUTURE YEAR EMISSIONS	165

4.1	EGU Point Source Projections (ptegu)	169

4.2	Non-EGU Point and Nonpoint Sector Projections	172

4.2.1	Background on the Control Strategy Tool (CoST)	173

4.2.2	CoST Plant CLOSURE Packet (ptnonipm, ptoilgas)	177

4.2.3	CoST PROJECTION Packets (afdust, ag, cmv, rail, nonpt, np oilgas, ptnonipm, pt oilgas, rwc)	177

4.2.3.1	Fugitive dust growth (afdust)	178

4.2.3.2	Livestock population growth (ag)	179

4.2.3.3	Category 1, Category 2 Commercial Marine Vessels (cmv_clc2)	179

4.2.3.4	Category 3 Commercial Marine Vessels (cmv_c3)	180

4.2.3.5	Oil and Gas Sources (pt oilgas, np oilgas)	182

4.2.3.6	Non-EGU point sources (ptnonipm)	184

4.2.3.7	Nonpoint Sources (nonpt)	186

4.2.3.8	Airport sources (airports)	187

4.2.4	CoST CONTROL Packets (nonpt, np oilgas, ptnonipm, pt oilgas)	187

4.2.4.1	Residential Wood Combustion (rwc)	189

4.2.4.2	Oil and Gas NSPS (np oilgas, pt oilgas)	191

4.2.4.3	LUCE NSPS (nonpt, ptnonipm, np oilgas, pt oilgas)	194

4.2.4.4	Fuel Sulfur Rules (nonpt, ptnonipm)	197

4.2.4.5	Natural Gas Turbines NOx NSPS (ptnonipm, pt oilgas)	198

4.2.4.6	Process Heaters NOx NSPS (ptnonipm, pt oilgas)	200

4.2.4.7	CLSWL (ptnonipm)	203

4.2.4.8	Petroleum Refineries NSPS Subpart JA (ptnonipm)	204

4.2.4.9	Ozone Transport Commission Rules (nonpt)	204

4.2.4.10	State-Specific Controls (ptnonipm)	205

4.3	Projections Computed Outside of CoST	206

4.3.1	Nonroad Mobile Equipment Sources (nonroad)	206

4.3.2	Onroad Mobile Sources (onroad)	207

4.3.3	Locomotives (rail)	209

4.3.1	Sources A dded in the 2021 fi Case	210

4.3.2	Sources Outside of the United States (onroadcan, onroadmex, othpt, ptfire othna, othar, othafdust, othptdust)
211

4.3.2.1	Canadian fugitive dust sources (othafdust, othptdust)	211

4.3.2.2	Point Sources in Canada and Mexico (othpt)	212

4.3.2.3	Nonpoint sources in Canada and Mexico (othar)	213

4.3.2.1	Onroad sources in Canada and Mexico (onroad can, onroad mex)	214

5	EMISSION SUMMARIES	215

6	REFERENCES	226

in


-------
List of Tables

Table 2-1. Platform sectors for the 2016 emissions modeling case	16

Table 2-2. Point source oil and gas sector NAICS Codes	22

Table 2-3. 2014NEIv2-to-2016 projection factors for pt_oilgas sector for 2016vl inventory	23

Table 2-4. 2016fh pt oilgas national emissions (excluding offshore) before and after 2014-to-2016

projections (tons/year)	24

Table 2-5. Pennsylvania emissions changes for natural gas transmission sources (tons/year)	24

Table 2-6. SCCs for Census-based growth from 2014 to 2016	25

Table 2-7. 2016vl platform SCCs for the airports sector	28

Table 2-8. Afdust sector SCCs	29

Table 2-9. Total impact of fugitive dust adjustments to unadjusted 2016 vl inventory	33

Table 2-10. 2016vl platform SCCs for the ag sector	36

Table 2-11. National back-projection factors for livestock: 2017 to 2016	37

Table 2-12. Source of input variables for EPIC	40

Table 2-13. 2014NEIv2-to-2016 oil and gas projection factors for CO and OK	42

Table 2-14. 2016 vl platform SCCs for RWC sector	43

Table 2-15. Projection factors for RWC by SCC	44

Table 2-16. 2016vl platform SCCs for Census-based growth	46

Table 2-17. MOVES vehicle (source) types	48

Table 2-18. Submitted data used to prepare onroad activity data	49

Table 2-19. Factors applied to project VMT from 2014 to 2016 to prepare default activity data	50

Table 2-20. Older Vehicle Adjustments Showing the Fraction of IHS Vehicle Populations to Retain for

2016\ 1 and 2017 Mil	58

Table 2-21. 2016vl platform SCCs for cmv_clc2 sector	61

Table 2-22. Vessel groups in the cmv_clc2 sector	63

Table 2-23. 2016vl platform SCCs for cmv_c3 sector	65

Table 2-24. 2017 to 2016 projection factors for C3 CMV	68

Table 2-25. 2016vl SCCs for the Rail Sector	69

Table 2-26. Class I Railroad Reported Locomotive Fuel Use Statistics for 2016	69

Table 2-27. 2016 Line-haul Locomotive Emission Factors by Tier, AAR Fleet Mix (grams/gal)	71

Table 2-28. Surface Transportation Board R-l Fuel Use Data - 2016	72

Table 2-29. 2016 Yard Switcher Emission Factors by Tier, AAR Fleet Mix (grams/gal)4	72

Table 2-30. Expenditures and fuel use for commuter rail	75

Table 2-31. Submitted nonroad input tables by agency	81

Table 2-32. Alaska counties/census areas for which nonroad equipment sector-specific emissions are

removed in 2016vl	82

Table 2-33. SCCs included in the ptfire sector for the 2016vl inventory	83

Table 2-34. National fire information databases used in 2016vl ptfire inventory	84

Table 2-35. List of S/L/T agencies that submitted fire data for 2016vl with types and formats	86

Table 2-36. Brief description of fire information submitted for 2016vl inventory use	86

Table 2-37. SCCs included in the ptagfire sector for the 2016vl inventory	90

Table 2-38. Assumed field size of agricultural fires per state(acres)	92

Table 2-39. Hourly Meteorological variables required by BEIS 3.61	94

Table 3-1. Key emissions modeling steps by sector	98

Table 3-2. Descriptions of the platform grids	100

Table 3-3. Emission model species produced for CB6 for CMAQ	102

Table 3-4. Integration status of naphthalene, benzene, acetaldehyde, formaldehyde and methanol (NBAFM)

for each platform sector	106

Table 3-5. Ethanol percentages by volume by Canadian province	108

iv


-------
Table 3-6. MOVES integrated species in M-profiles	109

Table 3-7. Basin/Region-specific profiles for oil and gas	Ill

Table 3-8. TOG MOVES-SMOKE Speciation for nonroad emissions in MOVES2014a used for the 2016

Platform	112

Table 3-9. Select mobile-related VOC profiles 2016	113

Table 3-10. Onroad M-profiles	114

Table 3-11. MOVES process IDs	115

Table 3-12. MOVES Fuel subtype IDs	116

Table 3-13. MOVES regclass IDs	116

Table 3-14. SPECIATE4.5 brake and tire profiles compared to those used in the 201 lv6.3 Platform	119

Table 3-15. Nonroad PM2.5 profiles	120

Table 3-16. NOx speciation profiles	120

Table 3-17. Sulfate split factor computation	121

Table 3-18. SO2 speciation profiles	121

Table 3-19. Temporal settings used for the platform sectors in SMOKE	122

Table 3-20. U.S. Surrogates available for the 2016vl modeling platforms	150

Table 3-21. Off-Network Mobile Source Surrogates	152

Table 3-22. Spatial Surrogates for Oil and Gas Sources	152

Table 3-23. Selected 2016 CAP emissions by sector for U.S. Surrogates (short tons in 12US1)	153

Table 3-24. Canadian Spatial Surrogates	156

Table 3-25. CAPs Allocated to Mexican and Canadian Spatial Surrogates (short tons in 36US3)	157

Table 3-26. Emission model species mappings for CMAQ and CAMx	160

Table 3-27. State tags for 2023llil. 2028fhl USSA modeling	162

Table 4-1. Overview of projection methods for the 2023 and 2028 regional cases	165

Table 4-2. EGU sector NOx emissions by State for the 2023 and 2028 regional cases	171

Table 4-3. Subset of CoST Packet Matching Hierarchy	174

Table 4-4. Summary of non-EGU stationary projections subsections	175

Table 4-5. Reductions from all facility/unit/stack-level closures in 2016vl	177

Table 4-6. Increase in total afdust PM2.5 emissions from projections in 2016vl	178

Table 4-7. National projection factors for livestock: 2016 to 2023 and 2028	179

Table 4-8. National projection factors for cmv_clc2	180

Table 4-9. California projection factors for cmv_clc2	180

Table 4-10. 2016-to-2023 and 2016-2028 CMV C3 projection factors outside of California	181

Table 4-11. 2016-to-2023 and 2016-2028 CMV C3 projection factors for California	181

Table 4-12. Year 2014-2017 high-level summary of national oil and gas exploration activity	184

Table 4-13. EIA's 2019 Annual Energy Outlook (AEO) tables used to project industrial sources	185

Table 4-14. Assumed retirement rates and new source emission factor ratios for NSPS rules	188

Table 4-15. Projection factors for RWC	190

Table 4-16. Non-point (npoilgas) SCCs in 2016vl modeling platform where Oil and Gas NSPS controls

applied	191

Table 4-17. Emissions reductions for np oilgas sector due to application of Oil and Gas NSPS	193

Table 4-18. Point source SCCs in ptoilgas sector where Oil and Gas NSPS controls were applied	193

Table 4-19. VOC reductions (tons/year) for the pt oilgas sector after application of the Oil and Gas NSPS

CONTROL packet for both future years 2023 and 2028	194

Table 4-20. SCCs and Engine Type in 2016vl modeling platform where RICE NSPS controls applied for

nonpt and ptnonipm sectors	194

Table 4-21. Non-point Oil and Gas SCCs in 2016vl modeling platform where RICE NSPS controls applied

	195

Table 4-22. Nonpoint Emissions reductions after the application of the RICE NSPS	196

Table 4-23. Ptnonipm Emissions reductions after the application of the RICE NSPS	196

v


-------
Table 4-24. Oil and Gas Emissions reductions for np oilgas sector due to application of RICE NSPS	196

Table 4-25. Point source SCCs in ptoilgas sector where RICE NSPS controls applied	196

Table 4-26. Emissions reductions (tons/year) in pt oilgas sector after the application of the RICE NSPS

CONTROL packet for future years 2023 and 2028	197

Table 4-27. Summary of fuel sulfur rule impacts on nonpoint S02 emissions for 2023 and 2028	197

Table 4-28. Summary of fuel sulfur rule impacts on ptnonipm S02 emissions for 2023 and 2028	198

Table 4-29. Stationary gas turbines NSPS analysis and resulting emission rates used to compute controls. 198
Table 4-30. Ptnonipm SCCs in 2016vl modeling platform where Natural Gas Turbines NSPS controls

applied	199

Table 4-31. Ptnonipm emissions reductions after the application of the Natural Gas Turbines NSPS	199

Table 4-32. Point source SCCs in pt oilgas sector where Natural Gas Turbines NSPS control applied	200

Table 4-33. Emissions reductions (tons/year) for pt oilgas after the application of the Natural Gas Turbines

NSPS CONTROL packet for future years 2023 and 2028	200

Table 4-34. Process Heaters NSPS analysis and 2016vl new emission rates used to estimate controls	201

Table 4-35. Ptnonipm SCCs in 2016vl modeling platform where Process Heaters NSPS controls applied. 201

Table 4-36. Ptnonipm emissions reductions after the application of the Process Heaters NSPS	202

Table 4-37. Point source SCCs in pt oilgas sector where Process Heaters NSPS controls were applied	202

Table 4-38. NOx emissions reductions (tons/year) in ptoilgas sector after the application of the Process

Heaters NSPS CONTROL packet for futures years 2023 and 2028	203

Table 4-39. Summary of CISWI rule impacts on ptnonipm emissions for 2023 and 2028	203

Table 4-40. Summary of NSPS Subpart JA rule impacts on ptnonipm emissions for 2023 and 2028	204

Table 4-41. Factors used to Project 2016 VMT to 2023 and 2028 	208

Table 4-42. Class I Line-haul Fuel Projections based on 2018 AEO Data	209

Table 4-43. Class I Line-haul Historic and Future Year Projected Emissions	210

Table 4-44. AEO growth rates for rail sub-groups	210

Table 4-45. Sources Added in the 2021fi Case	211

Table 5-1. National by-sector CAP emissions summaries for the 2016fh case, 12US1 grid (tons/yr)	216

Table 5-2. National by-sector CAP emissions summaries for the 2023fhl case, 12US1 grid (tons/yr)	217

Table 5-3. National by-sector CAP emissions summaries for the 2028fhl case, 12US1 grid (tons/yr)	218

Table 5-4. National by-sector CAP emissions summaries for the 2016fh case, 36US3 grid (tons/yr)	219

Table 5-5. National by-sector CAP emissions summaries for the 2023fhl case, 36US3 grid (tons/yr)	220

Table 5-6. National by-sector CAP emissions summaries for the 2028fhl case, 36US3 grid (tons/yr)	221

Table 5-7. National by-sector CAP emissions summaries for the 2016fi case, 12US1 grid (tons/yr)	222

Table 5-8. National by-sector CAP emissions summaries for the 2021fi case, 12US1 grid (tons/yr)	223

Table 5-9. National by-sector Ozone Season NOx emissions summaries 12US1 grid (tons/o.s.)	224

Table 5-10. National by-sector Ozone Season VOC emissions summaries 12US1 grid (tons/o.s.)	225

vi


-------
List of Figures

Figure 2-1. Impact of adjustments to fugitive dust emissions due to transport fraction, precipitation, and

cumulative	35

Figure 2-2. "Bidi" modeling system used to compute 2016 Fertilizer Application emissions	39

Figure 2-3. Representative Counties in 2016vl	60

Figure 2-4. 2017NEI/2016 platform geographical extent (solid) and U.S. ECA (dashed)	62

Figure 2-5. 2016 US Railroad Traffic Density in Millions of Gross Tons per Route Mile (MGT)	70

Figure 2-6. Class I Railroads in the United States5	70

Figure 2-7. 2016-2017 Active Rail Yard Locations in the United States	73

Figure 2-8. Class II and III Railroads in the United States5	74

Figure 2-9. Amtrak Routes with Diesel-powered Passenger Trains	76

Figure 2-10. Processing flow for fire emission estimates in the 2016vl inventory	88

Figure 2-11. Default fire type assignment by state and month in cases where a satellite detect is only source

of fire information	89

Figure 2-12. Blue Sky Modeling Framework	89

Figure 2-13. Normbeis3 data flows	94

Figure 2-14. Tmpbeis3 data flow diagram	95

Figure 3-1. Air quality modeling domains	100

Figure 3-2. Process of integrating NBAFM with VOC for use in VOC Speciation	106

Figure 3-3. Profiles composited for the new PM gas combustion related sources	117

Figure 3-4. Comparison of PM profiles used for Natural gas combustion related sources	118

Figure 3-5. Eliminating unmeasured spikes in CEMS data	124

Figure 3-6. Temporal Profile Input Unit Counts by Fuel and Peaking Unit Classification	126

Figure 3-7. Example Daily Temporal Profiles for the LADCO Region and the Gas Fuel Type	127

Figure 3-8. Example Diurnal Temporal Profiles for the MANE-VU Region and the Coal Fuel Type	127

Figure 3-9. Non-CEMS EGU Temporal Profile Application Counts	128

Figure 3-10. Future Year Emissions Follow the Pattern of Base Year Emissions	131

Figure 3-11. Excess Emissions Apportioned to Hours Less than the Historic Maximum	132

Figure 3-12. Regional Profile Applied due to not being able to Adjust below Historic Maximum	133

Figure 3-13. Regional Profile Applied, but Exceeds Historic Maximum in Some Hours	133

Figure 3-14. Diurnal Profile for all Airport SCCs	134

Figure 3-15. Weekly profile for all Airport SCCs	135

Figure 3-16. Monthly Profile for all Airport SCCs	135

Figure 3-17. Alaska Seaplane Profile	136

Figure 3-18. Example of RWC temporal allocation in 2007 using a 50 versus 60 °F threshold	137

Figure 3-19. RWC diurnal temporal profile	138

Figure 3-20. Data used to produce a diurnal profile for OHH, based on heat load (BTU/hr)	139

Figure 3-21. Day-of-week temporal profiles for OHH and Recreational RWC	139

Figure 3-22. Annual-to-month temporal profiles for OHH and recreational RWC	140

Figure 3-23. Example of animal NH3 emissions temporal allocation approach, summed to daily emissions

	141

Figure 3-24. Example of temporal variability of NOx emissions	142

Figure 3-25. Sample onroad diurnal profiles for Fulton County, GA	143

Figure 3-26. Methods to Populate Onroad Speeds and Temporal Profiles by Road Type	144

Figure 3-27. Regions for computing Region Average Speeds and Temporal Profiles	144

Figure 3-28. Example of Temporal Profiles for Combination Trucks	145

Figure 3-29. Example Nonroad Day-of-week Temporal Profiles	146

Figure 3-30. Example Nonroad Diurnal Temporal Profiles	147

Figure 3-31. Agricultural burning diurnal temporal profile	148

vii


-------
Figure 3-32. Prescribed and Wildfire diurnal temporal profiles	149

Figure 4-1. EIA Oil and Gas Supply Regions as of AEO2019	183

List of Appendices

Appendix A: CB6 Assignment for New Species

Appendix B: Profiles (other than onroad) that are new or revised in SPECIATE4.5 that were used in the

2014 v7.2 Platform
Appendix C: Mapping of Fuel Distribution SCCs to BTP, BPS and RBT

viii


-------
Acronyms

AADT

Annual average daily traffic

AE6

CMAQ Aerosol Module, version 6, introduced in CMAQ v5.0

AEO

Annual Energy Outlook

AERMOD

American Meteorological Society/Environmental Protection Agency



Regulatory Model

AIS

Automated Identification System

APU

Auxiliary power unit

BEIS

Biogenic Emissions Inventory System

BELD

Biogenic Emissions Land use Database

BenMAP

Benefits Mapping and Analysis Program

BPS

Bulk Plant Storage

BTP

Bulk Terminal (Plant) to Pump

C1C2

Category 1 and 2 commercial marine vessels

C3

Category 3 (commercial marine vessels)

CAMD

EPA's Clean Air Markets Division

CAMx

Comprehensive Air Quality Model with Extensions

CAP

Criteria Air Pollutant

CARB

California Air Resources Board

CB05

Carbon Bond 2005 chemical mechanism

CB6

Version 6 of the Carbon Bond mechanism

CBM

Coal-bed methane

CDB

County database (input to MOVES model)

CEMS

Continuous Emissions Monitoring System

CISWI

Commercial and Industrial Solid Waste Incinerators

CMAQ

Community Multiscale Air Quality

CMV

Commercial Marine Vessel

CNG

Compressed natural gas

CO

Carbon monoxide

CONUS

Continental United States

CoST

Control Strategy Tool

CRC

Coordinating Research Council

CSAPR

Cross-State Air Pollution Rule

EO, E10, E85

0%, 10% and 85% Ethanol blend gasoline, respectively

ECA

Emissions Control Area

ECCC

Environment and Climate Change Canada

EF

Emission Factor

EGU

Electric Generating Units

EIA

Energy Information Administration

EIS

Emissions Inventory System

EPA

Environmental Protection Agency

EMFAC

EMission FACtor (California's onroad mobile model)

EPIC

Environmental Policy Integrated Climate modeling system

FAA

Federal Aviation Administration

FCCS

Fuel Characteristic Classification System

FEST-C

Fertilizer Emission Scenario Tool for CMAQ

FF10

Flat File 2010

FINN

Fire Inventory from the National Center for Atmospheric Research

FIPS

Federal Information Processing Standards

IX


-------
FHWA

Federal Highway Administration

HAP

Hazardous Air Pollutant

HMS

Hazard Mapping System

HPMS

Highway Performance Monitoring System

ICI

Industrial/Commercial/Institutional (boilers and process heaters)

I/M

Inspection and Maintenance

IMO

International Marine Organization

IPM

Integrated Planning Model

LADCO

Lake Michigan Air Directors Consortium

LDV

Light-Duty Vehicle

LPG

Liquified Petroleum Gas

MACT

Maximum Achievable Control Technology

MARAMA

Mid-Atlantic Regional Air Management Association

MATS

Mercury and Air Toxics Standards

MCIP

Meteorology-Chemistry Interface Processor

MMS

Minerals Management Service (now known as the Bureau of Energy



Management, Regulation and Enforcement (BOEMRE)

MOVES

Motor Vehicle Emissions Simulator

MSA

Metropolitan Statistical Area

MTBE

Methyl tert-butyl ether

MWC

Municipal waste combustor

MY

Model year

NAAQS

National Ambient Air Quality Standards

NAICS

North American Industry Classification System

NBAFM

Naphthalene, Benzene, Acetaldehyde, Formaldehyde and Methanol

NCAR

National Center for Atmospheric Research

NEEDS

National Electric Energy Database System

NEI

National Emission Inventory

NESCAUM

Northeast States for Coordinated Air Use Management

NH3

Ammonia

NLCD

National Land Cover Database

NO A A

National Oceanic and Atmospheric Administration

NONROAD

OTAQ's model for estimation of nonroad mobile emissions

NOx

Nitrogen oxides

NSPS

New Source Performance Standards

OHH

Outdoor Hydronic Heater

OTAQ

EPA's Office of Transportation and Air Quality

ORIS

Office of Regulatory Information System

OKI)

EPA's Office of Research and Development

OSAT

Ozone Source Apportionment Technology

PFC

Portable Fuel Container

PM2.5

Particulate matter less than or equal to 2.5 microns

PM10

Particulate matter less than or equal to 10 microns

PPm

arts per million

ppmv

Parts per million by volume

PSAT

Particulate Matter Source Apportionment Technology

RACT

Reasonably Available Control Technology

RBT

Refinery to Bulk Terminal

RIA

Regulatory Impact Analysis

RICE

Reciprocating Internal Combustion Engine

X


-------
RWC

Residential Wood Combustion

RPD

Rate-per-vehicle (emission mode used in SMOKE-MOVES)

RPH

Rate-per-hour (emission mode used in SMOKE-MOVES)

RPP

Rate-per-profile (emission mode used in SMOKE-MOVES)

RPV

Rate-per-vehicle (emission mode used in SMOKE-MOVES)

RVP

Reid Vapor Pressure

see

Source Classification Code

SMARTFIRE2

Satellite Mapping Automated Reanalysis Tool for Fire Incident Reconciliation



version 2

SMOKE

Sparse Matrix Operator Kernel Emissions

SOi

Sulfur dioxide

SOA

Secondary Organic Aerosol

SIP

State Implementation Plan

SPDPRO

Hourly Speed Profiles for weekday versus weekend

S/L/T

state, local, and tribal

TAF

Terminal Area Forecast

TCEQ

Texas Commission on Environmental Quality

TOG

Total Organic Gas

TSD

Technical support document

USD A

United States Department of Agriculture

VIIRS

Visible Infrared Imaging Radiometer Suite

VOC

Volatile organic compounds

VMT

Vehicle miles traveled

VPOP

Vehicle Population

WRAP

Western Regional Air Partnership

WRF

Weather Research and Forecasting Model

2014NEIv2

2014 National Emissions Inventory (NEI), version 2

XI


-------
1 Introduction

The U.S. Environmental Protection Agency (EPA), working in conjunction with the National Emissions
Inventory Collaborative, developed an air quality modeling platform for criteria air pollutants to represent
the years of 2016, 2023 and 2028. The starting point for the 2016 inventory was the 2014 National
Emissions Inventory (NEI), version 2 (2014NEIv2), although many inventory sectors were updated to
represent the year 2016 through the incorporation of 2016-specific state and local data along with
nationally-applied adjustment methods. The year 2023 and year 2028 inventories were developed starting
with the 2016 inventory using sector-specific methods as described below. The inventories support
several applications, including modeling in support of the Revised Cross State Air Pollution Rule
(CSAPR) Update for the 2008 Ozone National Ambient Air Quality Standards (NAAQS).

The air quality modeling platform consists of all the emissions inventories and ancillary data files used for
emissions modeling, as well as the meteorological, initial condition, and boundary condition files needed
to run the air quality model. This document focuses on the emissions modeling data and techniques
including the emission inventories, the ancillary data files, and the approaches used to transform
inventories for use in air quality modeling.

The National Emissions Inventory Collaborative is a partnership between state emissions inventory staff,
multi-jurisdictional organizations (MJOs), federal land managers (FLMs), EPA, and others to develop a
North American air pollution emissions modeling platform with a base year of 2016 for use in air quality
planning. The Collaborative planned for three versions of the 2016 platform: alpha, beta, and Version 1.0.
This numbering format for this platform is different from previous EPA platforms which had the first
number based on the version of the NEI, and the second number as a platform iteration for that NEI year
(e.g., 7.3 where 7 represents 2014 NEI-based platforms, and 3 means the third iteration of the platform).
For the emissions modeling documented in this technical support document (TSD), the emissions values
for most sectors are the same as those in the Inventory Collaborative 2016vl Emissions Modeling
Platform, available from http://views.cira.colostate.edu/wiki/wiki/10202. In the file packages for this
platform, the platform may sometimes be known as the 2016v7.3 platform. The specification sheets
posted on the 2016vl platform release page on the Wiki provide many details regarding the inventories
and emissions modeling techniques in addition to those addressed in this TSD.

Some updates were made to the 2016vl platform after the fall 2019 release that were included in the
Revised CSAPR Update modeling, including some minor revisions to commercial marine vessel (CMV)
emissions, and electric generating unit (EGU) emissions developed in January 2020. Updates to 2016vl
to correct airport emissions and 2016 EGU processing made in June and July of 2020 were not included
in the CSAPR Update modeling because the modeling was already complete by that time. The updated
data and a description of them are available on the EPA FTP site

ftp://newftp.epa.gov/air/emismod/2016/vl/postvl updates/. If you cannot access the FTP site through the
provided link, this link points to the same data:
https://gaftp.epa.gov/Air/emismod/2016/vl/postvl updates.

This 2016 emissions modeling platform includes all criteria air pollutants (CAPs) and precursors, and a
group of hazardous air pollutants (HAPs). The group of HAPs are those explicitly used by the chemical
mechanism in the Community Multiscale Air Quality (CMAQ) model (Appel et al., 2018) for
ozone/particulate matter (PM): chlorine (CI), hydrogen chloride (HC1), benzene, acetaldehyde,
formaldehyde, methanol, naphthalene. The modeling domain includes the lower 48 states and parts of
Canada and Mexico. The modeling cases for this platform were developed for the Comprehensive Air

12


-------
Quality Model with Extensions (CAMx). However, the emissions modeling process first prepares outputs
in the format used by CMAQ, after which those emissions data are converted to the formats needed by
CAMx.

The 2016 platform used in this study consists of a 2016 base case, a 2023 case, and a 2028 case with the
abbreviations 2016fh_16j, 2023fhl_16j, and 2028fhl_16j, respectively. Additional cases that included
source apportionment by state and in some cases inventory sector were also developed. This platform
accounts for atmospheric chemistry and transport within a state-of-the-art photochemical grid model. In
the case abbreviation 2016fh_16j, 2016 is the year represented by the emissions; the "f' represents the
base year emissions modeling platform iteration, which here shows that it is 2014 NEI-based (whereas for
2011 NEI-based platforms, this letter was "e"); and the "h" stands for the eighth configuration of
emissions modeled for a 2014-NEI based modeling platform. The cases named 2023fhl_16j and
2028fhl_16j are the same as the original 2023fh and 2028fh future year cases, except that they include
EGU emissions that were developed in January 2020 and slightly updated commercial marine vessel
emissions. The case 2016fi was developed after some issues were identified with the 2016fh airport
emissions inventory and with the processing of EGU emissions at specific units when multiple units in the
NEI are mapped to multiple Continuous Emissions Modeling System (CEMS) units. The case 2021fi was
developed to provide a representation of emissions in 2021.

The 2016vl emissions modeling platform includes point sources, nonpoint sources, commercial marine
vessels (CMV), onroad and nonroad mobile sources, and fires for the U.S., Canada, and Mexico. Some
platform categories use more disaggregated data than are made available in the NEI. For example, in the
platform, onroad mobile source emissions are represented as hourly emissions by vehicle type, fuel type
process and road type while the NEI emissions are aggregated to vehicle type/fuel type totals and annual
temporal resolution. Temporal, spatial and other changes in emissions between the NEI and the emissions
input into the platform are described primarily in the platform specification sheets, although a full NEI
was not developed for the year 2016 because only point sources above a certain potential to emit must be
submitted for years between the full triennial NEI years (e.g., 2014, 2017, 2020). Emissions from Canada
and Mexico are used for the modeling platform but are not part of the NEI.

The primary emissions modeling tool used to create the air quality model-ready emissions was the Sparse
Matrix Operator Kernel Emissions (SMOKE) modeling system (http://www.smoke-model.org/), version
4.7 (SMOKE 4.7) with some updates. Emissions files were created for a 36-km national grid and for a
12-km national grid, both of which include the contiguous states and parts of Canada and Mexico as
shown in Figure 3-1.

The gridded meteorological model used to provide input data for the emissions modeling was developed
using the Weather Research and Forecasting Model (WRF,

https://ral.ucar.edu/solutions/products/weather-research-and-forecasting-model-wrf) version 3.8,
Advanced Research WRF core (Skamarock, et al., 2008). The WRF Model is a mesoscale numerical
weather prediction system developed for both operational forecasting and atmospheric research
applications. The WRF was run for 2016 over a domain covering the continental U.S. at a 12km
resolution with 35 vertical layers. The run for this platform included high resolution sea surface
temperature data from the Group for High Resolution Sea Surface Temperature (GHRSST) (see
https://www.ghrsst.org/) and is given the EPA meteorological case label "16j." The full case name
includes this abbreviation following the emissions portion of the case name to fully specify the name of
the case as "2016fh_16j."

13


-------
This document contains five sections and several appendices. Section 2 describes the 2016 inventories
input to SMOKE. Section 3 describes the emissions modeling and the ancillary files used with the
emission inventories. Methods to develop future year emissions are described in Section 4. Data
summaries are provided in Section 5. Section 6 provides references. The Appendices provide additional
details about specific technical methods or data.

14


-------
2 Emissions Inventories and Approaches

This section summarizes the emissions data that make up the 2016vl platform. This section provides
details about the data contained in each of the platform sectors for the base year and the future year.
The original starting point for the emission inventories was the 2014NEIv2 although emissions for most
sectors have been updated to better represent the year 2016. Documentation for the 2014NEIv2, including
a TSD, is available at https://www.epa.gov/air-emissions-inventories/2014-national-emissions-inventory-
nei-technical-support-document-tsd. Documentation for each 2016vl emissions sector in the form of
specification sheets is available on the 2016vl page of Inventory Collaborative Wiki
(http://views.cira.colostate.edu/wiki/wiki/10202). In addition to the NEI-based data for the broad
categories of point, nonpoint, onroad, nonroad, and events (i.e., fires), emissions from the Canadian and
Mexican inventories and several other non-NEI data sources are included in the 2016 platform.

The triennial NEI data for CAPs are largely compiled from data submitted by state, local and tribal
(S/L/T) air agencies. HAP emissions data are also from the S/L/T agencies, but, are often augmented by
the EPA because they are voluntarily submitted. The EPA uses the Emissions Inventory System (EIS) to
compile the NEI. The EIS includes hundreds of automated quality assurance checks to help improve data
quality, and also supports tracking release point (e.g., stack) coordinates separately from facility
coordinates. The EPA collaborates extensively with S/L/T agencies to ensure a high quality of data in the
NEI. Using the 2014NEIv2 as a starting point, the National Inventory Collaborative worked to develop a
modeling platform that more closely represents the year 2016. All emissions modeling sectors were
modified in some way to better represent the year 2016 for the 2016vl platform.

The point source emission inventories for the platform include partially updated emissions to represent
2016 based on state-submitted data and adjustments to much of the remaining 2014 data to better
represent 2016. Agricultural and wildland fire emissions represent the year 2016. Most nonpoint source
sectors started with 2014NEIv2 emissions and were adjusted to better represent the year 2016. Fertilizer
emissions, nonpoint oil and gas emissions, and onroad and nonroad mobile source emissions represent the
year 2016. For CMV emissions, emissions were developed based on 2017 NEI CMV emissions and the
sulfur dioxide (SO2) emissions reflect rules that reduced sulfur emissions for CMV that took effect in the
year 2015. For fertilizer ammonia emissions, a 2016-specific emissions inventory is used in this platform.
Nonpoint oil and gas emissions were developed using 2016-specific data for oil and gas wells and their
2016 production levels.

Onroad and nonroad mobile source emissions were developed using the Motor Vehicle Emission
Simulator (MOVES). Onroad emissions for the platform were developed based on emissions factors
output from MOVES2014b for the year 2016, run with inputs derived from the 2014NEIv2 including
activity data (e.g., vehicle miles traveled and vehicle populations) provided by state and local agencies or
otherwise projected to the year 2016. MOVES2014b was also used to generate nonroad emissions
because it included important updates related to nonroad engine population growth rates and spatial
allocation factors.

For the purposes of preparing the air quality model-ready emissions, emissions from the five NEI data
categories are split into finer-grained sectors used for emissions modeling. The significance of an
emissions modeling or "platform sector" is that the data are run through the SMOKE programs
independently from the other sectors except for the final merge (Mrggrid). The final merge program
combines the sector-specific gridded, speciated, hourly emissions together to create CMAQ-ready
emission inputs. For studies that use CAMx, these CMAQ-ready emissions inputs are converted into the
file formats needed by CAMx.

15


-------
Table 2-1 presents an overview the sectors in the 2016 platform and how they generally relate to the
2014NEIv2 as their starting point. The platform sector abbreviations are provided in italics. These
abbreviations are used in the SMOKE modeling scripts, inventory file names, and throughout the
remainder of this document. Through the Collaborative workgroups, state and local agencies provided
data used in the development of most sectors.

Table 2-1. Platform sectors for the 2016 emissions modeling case

Platform Sector:

abbreviation

NEI Data
Category

Description and resolution of the data input to SMOKE

EGU units:

ptegu

Point

Point source electric generating units (EGUs) for 2016 from the
Emissions Inventory System (EIS), based on 2014NEIv2 with most
sources updated to 2016. Includes some specific S/L/T updates. The
inventory emissions are replaced with hourly 2016 Continuous
Emissions Monitoring System (CEMS) values for nitrogen oxides
(NOx) and SO2 for any units that are matched to the NEI, and other
pollutants for matched units are scaled from the 2016 point inventory
using CEMS heat input. Emissions for all sources not matched to
CEMS data come from the raw inventory. Annual resolution for
sources not matched to CEMS data, hourly for CEMS sources.

Point source oil and
gas:

ptoilgas

Point

Point sources for 2016 including S/L/T updates for oil and gas
production and related processes based on facilities with the following
NAICS: 2111,21111,211111,211112 (Oil and Gas Extraction);
213111 (Drilling Oil and Gas Wells); 213112 (Support Activities for
Oil and Gas Operations); 2212, 22121, 221210 (Natural Gas
Distribution); 48611, 486110 (Pipeline Transportation of Crude Oil);
4862, 48621, 486210 (Pipeline Transportation ofNatural Gas).
Includes offshore oil and gas platforms in the Gulf of Mexico
(FIPS=85). Oil and gas point sources that were not already updated to
year 2016 in the baseline inventory were projected from 2014 to 2016.
Annual resolution.

Aircraft and ground
support equipment:

airports

Point

Emissions from aircraft up to 3,000 ft elevation and emissions from
ground support equipment based on 2017 NEI data. Note that these
emissions were found to be overestimated in June 2020.

Remaining non-
EGU point:

ptnonipm

Point

All 2016 point source inventory records not matched to the ptegu,
airports, or pt_oilgas sectors, including updates submitted by state and
local agencies. Year 2016 rail yard emissions were developed by the
rail workgroup. Annual resolution.

Agricultural:

ag

Nonpoint

Nonpoint livestock and fertilizer application emissions. Livestock
includes ammonia and other pollutants (except PM2 5) and was
backcasted from a draft version of 2017NEI based on animal
population data from the U.S. Department of Agriculture (USDA)
National Agriculture Statistics Service Quick Stats, where available.
Fertilizer includes only ammonia and is estimated for 2016 using the
FEST-C model. County and monthly resolution.

Agricultural fires
with point
resolution: ptagfire

Nonpoint

2016 agricultural fire sources based on EPA-developed data with state
updates, represented as point source day-specific emissions. They are
in the nonpoint NEI data category, but in the platform, they are treated
as point sources. Mostly at daily resolution with some state-submitted
data at monthly resolution.

16


-------
Platform Sector:

abbreviation

NEI Data
Category

Description and resolution of the data input to SMOKE

Area fugitive dust:

afdust

Nonpoint

PMio and PM2 5 fugitive dust sources from the 2014NEIv2 nonpoint
inventory with paved road dust grown to 2016 levels; including
building construction, road construction, agricultural dust, and road
dust. The NEI emissions are reduced during modeling according to a
transport fraction (newly computed for the 2016 beta platform) and a
meteorology-based (precipitation and snow/ice cover) zero-out.
Afdust emissions from the portion of Southeast Alaska inside the
36US3 domain are processed in a separate sector called 'afdust_ak'.
County and annual resolution.

Biogenic:

beis

Nonpoint

Year 2016, hour-specific, grid cell-specific emissions generated from
the BEIS3.61 model within SMOKE, including emissions in Canada
and Mexico using BELD v4.1 "water fix" land use data (including
improved treatment of water grid cells).

Category 1, 2 CMV:

cmv_clc2

Nonpoint

Category 1 and category 2 (C1C2) commercial marine vessel (CMV)
emissions sources backcast to 2016 from the 2017NEI using a
multiplier of 0.98.emissions. Includes C1C2 emissions in U.S. state
and Federal waters, and also all non-U.S. C1C2 emissions including
those in Canadian waters. Gridded and hourly resolution.

Category 3 CMV:

cmv_c3

Nonpoint

Category 3 (C3) CMV emissions converted to point sources based on
the center of the grid cells. Includes C3 emissions in U.S. state and
Federal waters, and also all non-U.S. C3 emissions including those in
Canadian waters. Emissions are backcast to 2016 from 2017NEI
emissions based on factors derived from U.S. Army Corps of
Engineers Entrance and Clearance data and information about the
ships entering the ports. Gridded and hourly resolution.

Locomotives :
rail

Nonpoint

Line haul rail locomotives emissions developed by the rail workgroup
based on 2016 activity and emission factors. Includes freight and
commuter rail emissions and incorporates state and local feedback.
County and annual resolution.

Remaining
nonpoint:

nonpt

Nonpoint

2014NEIv2 nonpoint sources not included in other platform sectors
with sources proportional to human population activity data grown to
year 2016; incorporates state and local feedback. County and annual
resolution.

Nonpoint source oil
and gas:
np oilgas

Nonpoint

2016 nonpoint oil and gas emissions output from the NEI oil and gas
tool along with state and local feedback. County and annual resolution.

Residential Wood
Combustion:

rwc

Nonpoint

2014NEIv2 nonpoint sources from residential wood combustion
(RWC) processes projected to the year 2016. County and annual
resolution.

Nonroad:

nonroad

Nonroad

2016 nonroad equipment emissions developed with the MOVES2014b
model which incorporates updated equipment growth rates. MOVES
was used for all states except California and Texas, which submitted
emissions. County and monthly resolution.

17


-------
Platform Sector:

abbreviation

NEI Data
Category

Description and resolution of the data input to SMOKE

Onroad:

onroad

Onroad

2016 onroad mobile source gasoline and diesel vehicles from moving
and non-moving vehicles that drive on roads, along with vehicle
refueling. Includes the following modes: exhaust, extended idle,
auxiliary power units, evaporative, permeation, refueling, and brake
and tire wear. For all states except California, developed using winter
and summer MOVES emissions tables produced by MOVES2014b
coupled with activity data projected to year 2016 or provided by S/L/T
agencies. SMOKE-MOVES was used to compute emissions from the
emission factors and activity data. Onroad emissions for Alaska,
Hawaii, Puerto Rico and the Virgin Islands were computed using the
same method as the continental U.S.,but are part of the
onroad nonconus sector.

Onroad California:

onroadcaadj

Onroad

2016 California-provided CAP onroad mobile source gasoline and
diesel vehicles based on the EMFAC model, which ere gridded and
temporalized using MOVES2014b results. Volatile organic compound
(VOC) HAP emissions derived from California-provided VOC
emissions and MOVES-based speciation.

Point source fires-
ptjire

Events

Point source day-specific wildfires and prescribed fires for 2016
computed using Satellite Mapping Automated Reanalysis Tool for Fire
Incident Reconciliation version 2 (SMARTFIRE2) and BlueSky
Framework (Sullivan, 2008 and Raffuse, 2007) for both flaming and
smoldering processes (i.e., SCCs 281XXXX002). Smoldering is
forced into layer 1 (by adjusting heat flux). Incorporates state inputs.
Daily resolution.

Non-US. Fires:
ptfireothna

N/A

Point source day-specific wildfires and prescribed fires for 2016
provided by Environment Canada with data for missing months, and
for Mexico and Central America, filled in using fires from the Fire
Inventory (FINN) from National Center for Atmospheric Research
(NCAR) fires (NCAR, 2016 and Wiedinmyer, C., 2011). Daily
resolution.

Other Area Fugitive
dust sources not
from the NEI:
othafdust

N/A

Fugitive dust sources of particulate matter emissions excluding land
tilling from agricultural activities, from Environment and Climate
Change Canada (ECCC) 2015 emission inventory, except that
construction dust emissions were reduced to levels compatible with
their 2010 inventory. A transport fraction adjustment is applied along
with a meteorology-based (precipitation and snow/ice cover) zero-out.
County and annual resolution.

Other Point Fugitive
dust sources not
from the NEI:
othptdust

N/A

Fugitive dust sources of particulate matter emissions from land tilling
from agricultural activities, ECCC 2015 emission inventory, but wind
erosion emissions were removed. A transport fraction adjustment is
applied along with a meteorology-based (precipitation and snow/ice
cover) zero-out. Data were originally provided on a rotated 10-km grid
for beta, but were smoothed so as to avoid the artifact of grid lines in
the processed emissions. Monthly resolution.

18


-------
Platform Sector:

abbreviation

NEI Data
Category

Description and resolution of the data input to SMOKE

Other point sources
not from the NEI:
othpt

N/A

Point sources from the ECCC 2015 emission inventory, including
agricultural ammonia, along with emissions from Mexico's 2008
inventory projected to 2014 and 2018 and then interpolated to 2016.
Agricultural data were originally provided on a rotated 10-km grid for
beta, but were smoothed so as to avoid the artifact of grid lines in the
processed emissions. Monthly resolution for Canada agricultural and
airport emissions, annual resolution for the remainder of Canada and
all of Mexico.

Other non-NEI
nonpoint and
nonroad:

othar

N/A

Year 2015 Canada (province or sub-province resolution) emissions
from the ECCC inventory: monthly for nonroad sources; annual for
rail and other nonpoint Canada sectors. Year 2016 Mexico (municipio
resolution) emissions, interpolated from 2014 and 2018 inventories
that were projected from their 2008 inventory: annual nonpoint and
nonroad mobile inventories.

Other non-NEI
onroad sources:

onroad can

N/A

Monthly year 2015 Canada (province resolution or sub-province
resolution, depending on the province) from the ECCC onroad mobile
inventory.

Other non-NEI
onroad sources:

onroad mex

N/A

Monthly year 2016 Mexico (municipio resolution) onroad mobile
inventory based on MOVES-Mexico runs for 2014 and 2018 then
interpolated to 2016.

Other natural emissions are also merged in with the above sectors: ocean chlorine and sea salt. The ocean
chlorine gas emission estimates are based on the build-up of molecular chlorine (Cb) concentrations in
oceanic air masses (Bullock and Brehme, 2002). In CMAQ, the species name is "CL2". The sea salt
emissions were developed with version 4.1 of the OCEANIC pre-processor that comes with the CAMx
model. The preprocessor estimates time/space-varying emissions of aerosol sodium, chloride and sulfate;
gas-phase chlorine and bromine associated with sea salt; gaseous halo-methanes; and dimethyl sulfide
(DMS). These additional oceanic emissions are incorporated into the final model-ready emissions files for
CAMx.

The emission inventories in SMOKE input formats for the platform are available from EPA's Air
Emissions Modeling website: https://www.epa.gov/air-emissions-modeling/2014-2016-version-7-air-
emissions-modelirig-platforms, under the section entitled "2016vl Platform". The platform "README"
file indicates the particular zipped files associated with each platform sector. A number of reports (i.e.,
summaries) are available with the data files for the 2016 platform. The types of reports include state
summaries of inventory pollutants and model species by modeling platform sector and county annual
totals by modeling platform sector. Additional types of data including outputs from SMOKE and inputs to
CAMx are available from the Intermountain West Data Warehouse.

2.1 2016 point sources (ptegu, pt_oilgas, ptnonipm, airports)

Point sources are sources of emissions for which specific geographic coordinates (e.g., latitude/longitude)
are specified, as in the case of an individual facility. A facility may have multiple emission release points
that may be characterized as units such as boilers, reactors, spray booths, kilns, etc. A unit may have
multiple processes (e.g., a boiler that sometimes burns residual oil and sometimes burns natural gas).

This section describes NEI point sources within the contiguous U.S. and the offshore oil platforms which
are processed by SMOKE as point source inventories. A full NEI is compiled every three years including
2011, 2014 and 2017. In the intervening years, emissions information about point sources that exceed
certain potential to emit threshold are required to be submitted to the EIS that is used to compile the NEI.

19


-------
A comprehensive description of how EGU emissions were characterized and estimated in the 2014 NEI is
located in Section 3.4 in the 2014NEIv2 TSD. The methods for emissions estimation are similar for the
interim year of 2016, but there is no TSD available specific to the 2016 point source submissions to EIS.
Additional information on state submissions through the collaborative process are available in the
collaborative specification sheets.

The point source file used for the modeling platform is exported from EIS into the Flat File 2010 (FF10)
format that is compatible with SMOKE (see

https://www.cmascenter.Org/smoke/documentation/4.7/html/ch08s02s08.htmn.

For the 2016vl platform, the export of point source emissions, including stack parameters and locations
from EIS, was done on June 12, 2018. The flat file was modified to remove sources without specific
locations (i.e., their FIPS code ends in 777). Then the point source FF10 was divided into four NEI-based
platform point source sectors: the EGU sector (ptegu), point source oil and gas extraction-related
emissions (pt oilgas), airport emissions were put into the airports sector, and the remaining non-EGU
sector also called the non-IPM (ptnonipm) sector. The split was done at the unit level for ptegu and
facility level for pt oilgas such that a facility may have units and processes in both ptnonipm and ptegu,
but cannot be in both pt oilgas and any other point sector. Additional information on updates made
through the collaborative process is available in the collaborative specification sheets.

The EGU emissions are split out from the other sources to facilitate the use of distinct SMOKE temporal
processing and future-year projection techniques. The oil and gas sector emissions (pt oilgas) were
processed separately for summary tracking purposes and distinct future-year projection techniques from
the remaining non-EGU emissions (ptnonipm).

The inventory pollutants processed through SMOKE for all point source sectors were: carbon monoxide
(CO), NOx, VOC, SO2, ammonia (NH3), particles less than 10 microns in diameter (PM10), and particles
less than 2.5 microns in diameter (PM2.5), and all of the air toxics listed in Table 3-3. The Naphthalene,
Benzene, Acetaldehyde, Formaldehyde, and Methanol (NBAFM) species are explicit in the CB6-CMAQ
chemical mechanism and are taken from the HAP emissions in the flat file (if present for a source) as
opposed to using emissions generated through VOC speciation, as is normally done for non-toxics
modeling applications. To prevent double counting of mass, NBAFM species are removed from VOC
speciation profiles, thus resulting in speciation profiles that may sum to less than 1. This is called the
"no-integrate" VOC speciation case and is discussed in detail in Section 3.2.1.1. The resulting VOC in
the modeling system may be higher or lower than the VOC emissions in the NEI; they would only be the
same if the HAP inventory and speciation profiles were exactly consistent. For HAPs other than those in
NBAFM, there is no concern for double-counting since CMAQ handles these outside the CB6
mechanism.

The ptnonipm and pt oilgas sector emissions were provided to SMOKE as annual emissions. For those
ptegu sources with CEMS data that could be matched to the point inventory from EIS, hourly CEMS NOx
and SO2 emissions were used rather than the annual total NEI emissions. For all other pollutants at
matched units, the annual emissions were used as-is from the NEI, but were allocated to hourly values
using heat input from the CEMS data. For the sources in the ptegu sector not matched to CEMS data,
daily emissions were created using an approach described in Section 2.1.1. For non-CEMS units other
than municipal waste combustors and cogeneration units, IPM region- and pollutant-specific diurnal
profiles were applied to create hourly emissions.

20


-------
2.1.1 EGU sector (ptegu)

The ptegu sector contains emissions from EGUs in the 2016 NEI point inventory that could be
matched to units found in the National Electric Energy Data System (NEEDS) v6 database
(https://www.epa.gov/airmarkets/national-electric-energy-data-svstem-needs-v6). The matching was
prioritized according to the amount of the emissions produced by the source. In the SMOKE point flat
file, emission records for sources that have been matched to the NEEDS database have a value filled into
the IPMYN column based on the matches stored within EIS. The 2016 NEI point inventory consists of
data submitted by S/L/T agencies and EPA to the EIS for Type A (i.e., large) point sources. Those EGU
sources in the 2014 NEIv2 inventory that were not submitted or updated for 2016 and not identified as
retired were retained. The retained 2014 NEIv2 EGUs in CT, DE, DC, ME, MD, MA, NH, NJ, NY, NC,
PA, RI, VT, VA, and WV were projected from 2014 to 2016 values using factors provided by the Mid-
Atlantic Regional Air Management Association (MARAMA).

Higher generation capacity units in the ptegu sector are matched to 2016 CEMS data from EPA's Clean
Air Markets Division (CAMD) via ORIS facility codes and boiler ID. For the matched units, SMOKE
replaces the 2016 emissions of NOx and SO2 with the CEMS emissions, thereby ignoring the annual
values specified in the NEI. For other pollutants at matched units, the hourly CEMS heat input data are
used to allocate the NEI annual emissions to hourly values. All stack parameters, stack locations, and
Source Classification Codes (SCC) for these sources come from the NEI or updates provided by data
submitters outside of EIS. Because these attributes are obtained from the NEI, the chemical speciation of
VOC and PM2.5 for the sources is selected based on the SCC or in some cases, based on unit-specific data.
If CEMS data exists for a unit, but the unit is not matched to the NEI, the CEMS data for that unit is not
used in the modeling platform. However, if the source exists in the NEI and is not matched to a CEMS
unit, the emissions from that source are still modeled using the annual emission value in the NEI
temporally allocated to hourly values. The EGU flat file inventory is split into a flat file with CEMS
matches and a flat file without CEMS matches to support analysis and temporalization.

In the SMOKE point flat file, emission records for point sources matched to CEMS data have values
filled into the ORIS FACILITY CODE and ORIS BOILER ID columns. The CEMS data in SMOKE-
ready format is available at http://ampd.epa.gov/ampd/ near the bottom of the "Prepackaged Data" tab.
Many smaller emitters in the CEMS program are not identified with ORIS facility or boiler IDs that can
be matched to the NEI due to inconsistencies in the way a unit is defined between the NEI and CEMS
datasets, or due to uncertainties in source identification such as inconsistent plant names in the two data
systems. Also, the NEEDS database of units modeled by IPM includes many smaller emitting EGUs that
do not have CEMS. Therefore, there will be more units in the NEEDS database than have CEMS data.
The temporal allocation of EGU units matched to CEMS is based on the CEMS data, whereas regional
profiles are used for most of the remaining units. More detail can be found in Section 3.3.2.

Some EIS units match to multiple CAMD units based on cross-reference information in the EIS alternate
identifier table. The multiple matches are used to take advantage of hourly CEMS data when a CAMD
unit specific entry is not available in the inventory. Where a multiple match is made the EIS unit is split
and the ORIS facility and boiler IDs are replaced with the individual CAMD unit IDs. The split EIS unit
NOX and S02 emissions annual emissions are replaced with the sum of CEMS values for that respective
unit. All other pollutants are scaled from the EIS unit into the split CAMD unit using the fraction of
annual heat input from the CAMD unit as part of the entire EIS unit. The NEEDS ID in the "ipm_yn"
column of the flat file is updated with a "_M_" between the facility and boiler identifiers to signify that
the EIS unit had multiple CEMS matches. The inventory records with multiple matches had the EIS unit
identifiers appended with the ORIS boiler identifier to distinguish each CEMS record in SMOKE.

21


-------
For sources not matched to CEMS data, except for municipal waste combustors (MWCs) waste-to-energy
and cogeneration units, daily emissions were computed from the NEI annual emissions using average
CEMS data profiles specific to fuel type, pollutant,1 and IPM region. To allocate emissions to each hour
of the day, diurnal profiles were created using average CEMS data for heat input specific to fuel type and
IPM region. See Section 3.3.2 for more details on the temporal allocation approach for ptegu sources.
MWC and cogeneration units were specified to use uniform temporal allocation such that the emissions
are allocated to constant levels for every hour of the year. These sources do not use hourly CEMs, and
instead use a PTDAY file with the same emissions for each day, combined with a uniform hourly
temporal profile applied by SMOKE.

The ptegu inventory for the 2016fi case includes an update that allows SMOKE to properly process
CEMS emissions when there are multiple CEMS units mapped to the same NEI unit. This caused NOx
and S02 emissions in 2016fi to be higher at some units.

2.1.2 Point source oil and gas sector (pt_oilgas)

The ptoilgas sector consists of point source oil and gas emissions in United States, primarily pipeline-
transportation and some upstream exploration and production. Sources in the pt oilgas sector consist of
sources which are not electricity generating units (EGUs) and which have a North American Industry
Classification System (NAICS) code corresponding to oil and gas exploration, production, pipeline-
transportation or distribution. The pt oilgas sector was separated from the ptnonipm sector by selecting
sources with specific NAICS codes shown in Table 2-2. The use of NAICS to separate out the point oil
and gas emissions forces all sources within a facility to be in this sector, as opposed to ptegu where
sources within a facility can be split between ptnonipm and ptegu sectors.

Table 2-2. Point source oil and gas sector NAICS Codes

NAICS

Type of point
source

NAICS description

2111, 21111

Production

Oil and Gas Extraction

211111

Production

Crude Petroleum and Natural Gas Extraction

211112

Production

Natural Gas Liquid Extraction

213111

Production

Drilling Oil and Gas Wells

213112

Support

Support Activities for Oil and Gas Operations

2212, 22121, 221210

Distribution

Natural Gas Distribution

4862, 48621, 486210

Transmission

Pipeline Transportation of Natural Gas

48611, 486110

Transmission

Pipeline Transportation of Crude Oil

The starting point for the 2016vl emissions platform pt oilgas inventory was the 2016 point source NEI.
The 2016 NEI includes data submitted by S/L/T agencies and EPA to the EIS for Type A (i.e., large)
point sources. Point sources in the 2014 NEIv2 not submitted for 2016 were pulled forward from the 2014
NEIv2 unless they had been marked as shut down. For the federally-owned offshore point inventory of
oil and gas platforms, a 2014 inventory was developed by the U.S. Department of the Interior, Bureau of
Ocean and Energy Management, Regulation, and Enforcement (BOEM).

1 The year to day profiles use NOx and SO2 CEMS for NOx and SO2, respectively. For all other pollutants, they use heat input
CEMS data.

22


-------
The 2016 ptoilgas inventory includes sources with updated data for 2016 and sources carried forward
from the 2014NEIv2 point inventory. Each type of source can be identified based on the calc_year field in
the flat file 2010 (FF10) formatted inventory files, which is set to either 2016 or 2014. The pt oilgas
inventory was split into two components: one for 2016 sources, and one for 2014 sources. The 2016
sources were used in 2016vl platform without further modification. Updates were made to selected West
Virginia Type B facilities based on comments from the state.

For pt oilgas emissions that were carried forward from the 2014NEIv2, the emissions were projected to
represent the year 2016. Each state/ SCC/NAICS combination in the inventory was classified as either an
oil source, a natural gas source, a combination of oil and gas, or designated as a "no growth" source.
Growth factors were based on historical state production data from the Energy Information
Administration (EIA) and are listed in Table 2. National 2016 pt oilgas emissions before and after
application of 2014-to-2016 projections are shown in Table 3. The historical production data for years
2014 and 2016 for oil and natural gas were taken from the following websites:

•	https://www.eia.gov/dnav/pet/pet crd crpdn adc mbbl a.htm (Crude production)

•	http://www.eia.gov/dnav/ng/ng sum lsum a epgO fgw mmcf a.htm (Natural gas production)

The "no growth" sources include all offshore and tribal land emissions, and all emissions with aNAICS
code associated with distribution, transportation, or support activities. As there were no 2015 production
data in the EIA for Idaho, no growth was assumed for this state; the only pt oilgas sources in Idaho were
pipeline transportation related. Maryland and Oregon had no oil production data on the EIA website. The
factors provided in Table 2-8 were applied to sources with NAICS = 2111,21111,211111,211112, and
213111 and with production-related SCC processes. Table 2-3 provides a national summary of emissions
before and after this 2 year projection for these sources in the pt oilgas sector. Table 2-4 shows the
national emissions for pt oilgas following the projection to 2016.

Table 2-3. 2014NEIv2-to-2016 projection factors for pt oilgas sector for 2016vl inventory

State

Natural Gas
growth

Oil growth

Combination gas/oil growth

Alabama

-9.0%

-17.5%

-13.2%

Alaska

1.9%

-1.1%

0.4%

Arizona

-55.7%

-85.7%

-70.7%

Arkansas

-26.7%

13.6%

-6.6%

California

-14.2%

-9.1%

-11.7%

Colorado

3.5%

22.0%

12.8%

Florida

8.0%

-13.2%

-2.6%

Idaho

0.0%

0.0%

0.0%

Illinois

13.2%

-9.5%

1.8%

Indiana

-6.2%

-27.5%

-16.9%

Kansas

-15.0%

-23.4%

-19.2%

Kentucky

-1.6%

-23.1%

-12.4%

Louisiana

-11.0%

-17.4%

-14.2%

Maryland

70.0%

N/A

N/A

Michigan

-12.6%

-23.4%

-18.0%

Mississippi

-10.9%

-16.3%

-13.6%

Missouri

-66.7%

-37.2%

-52.0%

Montana

-11.9%

-22.5%

-17.2%

23


-------
State

Natural Gas
growth

Oil growth

Combination gas/oil growth

Nebraska

27.3%

-25.0%

1.2%

Nevada

0.0%

-12.3%

-6.2%

New Mexico

1.4%

17.4%

9.4%

New York

-33.4%

-36.8%

-35.1%

North Dakota

31.4%

-4.3%

13.6%

Ohio

181.0%

44.4%

112.7%

Oklahoma

5.9%

6.9%

6.4%

Oregon

-18.0%

N/A

N/A

Pennsylvania

24.8%

-7.9%

8.5%

South Dakota

-33.9%

-21.7%

-27.8%

Tennessee

-31.9%

-22.1%

-27.0%

Texas

-6.1%

1.0%

-2.6%

Utah

-19.8%

-25.4%

-22.6%

Virginia

-10.0%

-50.0%

-30.0%

West Virginia

28.9%

0.7%

14.8%

Wyoming

-7.5%

-4.7%

-6.1%

Table 2-4. 2016fh ptoilgas national emissions (excluding offshore) before and after 2014-to-2016

projections (tons/year)

Pollutant

Before
projections

After projections

% change 2014 to 2016

CO

175,929

177,690

1.0%

NH3

4,347

4,338

-0.2%

NOX

377,517

379,866

0.6%

PM10-PRI

12,630

12,397

-1.8%

PM25-PRI

11,545

11,286

-2.2%

S02

35,236

34,881

-1.0%

VOC

127,242

129,253

1.6%

The state of Pennsylvania provided new emissions data for natural gas transmission sources for year
2016. The PA point source data replaced the emissions used in 2016beta. Table 2-5 illustrates the change
in emissions with this update.

Table 2-5. Pennsylvania emissions changes for natural gas transmission sources (tons/year).



State





2016



2016 vl -

State

FIPS

NAICS

Pollutant

beta

2016 vl

beta

Pennsylvania

42

486210

CO

2,787

2,385

403

Pennsylvania

42

486210

NOX

5,737

5,577

160

Pennsylvania

42

486210

PM10-PRI

400

227

173

Pennsylvania

42

486210

PM25-PRI

399

209

191

Pennsylvania

42

486210

S02

30

33

-3

Pennsylvania

42

486210

VOC

1,221

1,149

71

24


-------
2.1.3 Non-IPM sector (ptnonipm)

With minor exceptions, the ptnonipm sector contains point sources that are not in the airport, ptegu or
pt oilgas sectors. For the most part, the ptnonipm sector reflects the non-EGU sources of the NEI point
inventory; however, it is likely that some small low-emitting EGUs not matched to the NEEDS database
or to CEMS data are present in the ptnonipm sector. The ptnonipm emissions in the 2016vl platform have
been updated from the 2016 NEI point inventory with the following changes.

Non-IPM Projection from 2014 to 2016 inside MARAMA resion

2014-to-2016 projection packets for all nonpoint sources were provided by MARAMA for the following
states: CT, DE, DC, ME, MD, MA, NH, NJ, NY, NC, PA, RI, VT, VA, and WV.

New Jersey provided their own projection factors for projection from 2014 to 2016 which were mostly the
same as those provided by MARAMA, except for three SCCs with differences (SCCs: 2302070005,
2401030000, 2401070000). For those three SCCs, the projection factors provided by New Jersey were
used instead of the MARAMA factors.

Non-IPM Projection from 2014 to 2016 outside MARAMA resion

In areas outside of the MARAMA states, historical census population, sometimes by county and
sometimes by state, was used to project select nonpt sources from the 2014NEIv2 to 2016vl platform.
The population data was downloaded from the US Census Bureau. Specifically, the "Population,
Population Change, and Estimated Components of Population Change: April 1, 2010 to July 1, 2017" file
(https://www2.census.gov/programs-survevs/popest/datasets/2010-2017/counties/totals/co-est2Q17-
alldata.csv). A ratio of 2016 population to 2014 population was used to create a growth factor that was
applied to the 2014NEIv2 emissions with SCCs matching the population-based SCCs listed in Table 2-6
Positive growth factors (from increasing population) were not capped, but negative growth factors (from
decreasing population) were flatlined for no growth.

Table 2-6. SCCs for Census-based growth from 2014 to 2016

S( (

Tier 1

Tier 2 Description

Tier 3

Tier 4



Description



Description

Description

23020

Industrial

Food and Kindred Products:

Commercial Charbroiling

Conveyorized

02100

Processes

SIC 20



Charbroiling

23020

Industrial

Food and Kindred Products:

Commercial Charbroiling

Under-fired

02200

Processes

SIC 20



Charbroiling

23020

Industrial

Food and Kindred Products:

Commercial Deep Fat

Total

03000

Processes

SIC 20

Frying



23020

Industrial

Food and Kindred Products:

Commercial Deep Fat

Flat Griddle Frying

03100

Processes

SIC 20

Frying



23020

Industrial

Food and Kindred Products:

Commercial Deep Fat

Clamshell Griddle

03200

Processes

SIC 20

Frying

Frying

24010

Solvent

Surface Coating

Architectural Coatings

Total: All Solvent

01000

Utilization





Types

24010

Solvent

Surface Coating

Architectural Coatings -

Total: All Solvent

02000

Utilization



Solvent-based

Types

24010

Solvent

Surface Coating

Architectural Coatings -

Total: All Solvent

03000

Utilization



Water-based

Types

24011

Solvent

Surface Coating

Industrial Maintenance

Total: All Solvent

00000

Utilization



Coatings

Types

25


-------
S( (

Tier 1

Tier 2 Description

Tier 3

Tier 4



Description



Description

Description

24012

Solvent

Surface Coating

Other Special Purpose

Total: All Solvent

00000

Utilization



Coatings

Types

24250

Solvent

Graphic Arts

All Processes

Total: All Solvent

00000

Utilization





Types

24250

Solvent

Graphic Arts

Lithography

Total: All Solvent

10000

Utilization





Types

24250

Solvent

Graphic Arts

Letterpress

Total: All Solvent

20000

Utilization





Types

24250

Solvent

Graphic Arts

Rotogravure

Total: All Solvent

30000

Utilization





Types

24250

Solvent

Graphic Arts

Flexography

Total: All Solvent

40000

Utilization





Types

24400

Solvent

Miscellaneous Industrial

Adhesive (Industrial)

Total: All Solvent

20000

Utilization



Application

Types

24600

Solvent

Miscellaneous Non-industrial:

All Processes

Total: All Solvent

00000

Utilization

Consumer and Commercial



Types

24601

Solvent

Miscellaneous Non-industrial:

All Personal Care

Total: All Solvent

00000

Utilization

Consumer and Commercial

Products

Types

24602

Solvent

Miscellaneous Non-industrial:

All Household Products

Total: All Solvent

00000

Utilization

Consumer and Commercial



Types

24604

Solvent

Miscellaneous Non-industrial:

All Automotive

Total: All Solvent

00000

Utilization

Consumer and Commercial

Aftermarket Products

Types

24605

Solvent

Miscellaneous Non-industrial:

All Coatings and Related

Total: All Solvent

00000

Utilization

Consumer and Commercial

Products

Types

24606

Solvent

Miscellaneous Non-industrial:

All Adhesives and

Total: All Solvent

00000

Utilization

Consumer and Commercial

Sealants

Types

24608

Solvent

Miscellaneous Non-industrial:

All FIFRA Related

Total: All Solvent

00000

Utilization

Consumer and Commercial

Products

Types

24609

Solvent

Miscellaneous Non-industrial:

Miscellaneous Products

Total: All Solvent

00000

Utilization

Consumer and Commercial

(Not Otherwise Covered)

Types

24618

Solvent

Miscellaneous Non-industrial:

Pesticide Application: All

Total: All Solvent

00000

Utilization

Commercial

Processes

Types

24618

Solvent

Miscellaneous Non-industrial:

Pesticide Application: All

Surface Application

00001

Utilization

Commercial

Processes



24618

Solvent

Miscellaneous Non-industrial:

Pesticide Application: All

Soil Incorporation

00002

Utilization

Commercial

Processes



24618

Solvent

Miscellaneous Non-industrial:

Pesticide Application:

Not Elsewhere

70999

Utilization

Commercial

Non-Agricultural

Classified

24658

Solvent

Miscellaneous Non-industrial:

Pesticide Application

Total: All Solvent

00000

Utilization

Consumer



Types

25010

Storage and

Petroleum and Petroleum

Residential Portable Gas

Permeation

11011

Transport

Product Storage

Cans



25010

Storage and

Petroleum and Petroleum

Residential Portable Gas

Evaporation (includes

11012

Transport

Product Storage

Cans

Diurnal losses)

25010

Storage and

Petroleum and Petroleum

Residential Portable Gas

Spillage During

11013

Transport

Product Storage

Cans

Transport

25010

Storage and

Petroleum and Petroleum

Residential Portable Gas

Refilling at the Pump -

11014

Transport

Product Storage

Cans

Vapor Displacement

25010

Storage and

Petroleum and Petroleum

Residential Portable Gas

Refilling at the Pump -

11015

Transport

Product Storage

Cans

Spillage

26


-------
S( (

Tier 1

Tier 2 Description

Tier 3

Tier 4



Description



Description

Description

25010

Storage and

Petroleum and Petroleum

Commercial Portable Gas

Permeation

12011

Transport

Product Storage

Cans



25010

Storage and

Petroleum and Petroleum

Commercial Portable Gas

Evaporation (includes

12012

Transport

Product Storage

Cans

Diurnal losses)

25010

Storage and

Petroleum and Petroleum

Commercial Portable Gas

Spillage During

12013

Transport

Product Storage

Cans

Transport

25010

Storage and

Petroleum and Petroleum

Commercial Portable Gas

Refilling at the Pump -

12014

Transport

Product Storage

Cans

Vapor Displacement

25010

Storage and

Petroleum and Petroleum

Commercial Portable Gas

Refilling at the Pump -

12015

Transport

Product Storage

Cans

Spillage

26300

Waste Disposal

Treatment and Recovery

Wastewater Treatment,

Total Processed

20000





Public Owned



26400

Waste Disposal

Treatment and Recovery

TSDFs, All TSDF Types

Total: All Processes

00000









28100

Miscellane-ous

Other Combustion

Residential Grilling

Total

25000

Area Sources







28100

Miscellane-ous

Other Combustion

Cremation

Humans

60100

Area Sources







Other non-IPM updates in 2016vl

In New Jersey, emissions for SCCs for Industrial (2102004000) and Commercial/Institutional
(2103004000) Distillate Oil, Total: Boilers and Internal Combustion (IC) Engines were removed at that
state's request. These emissions were derived from EPA estimates, and double counted emissions that
were provided by New Jersey and assigned to other SCCs.

The state of New Jersey also requested that animal waste NH3 emissions from the following SCCs be
removed: 2806010000 - Cats, 2806015000 - Dogs, 2807020001 - Black Bears, 2807020002 - Grizzly
Bears, 2807025000 - Elk, 2807030000 - Deer, and 2810010000 - Human Perspiration and Respiration.
These emissions existed in CA, DE, ME, NJ, and UT, and were removed from all states.

The state of Alaska reported several nonpoint sources that were missing in 2014NEIv2. Some of the
sources reported by Alaska were identified in our EGU inventory and removed from the new nonpoint
inventory. The rest of the stationary sources were converted to an FFlO-formatted nonpoint inventory and
included in 2016vl platform in the nonpt sector.

The state of Alabama requested that their Industrial, Commercial, Institutional (ICI) Wood emissions
(2102008000), which totaled more than 32,000 tons/year of PM2.5 emissions in the beta version of this
emissions modeling platform and were significantly higher than other states' ICI Wood emissions, be
removed from 2016vl platform.

The state of New York provided a new set of non-residential wood combustion emissions for inclusion in
2016vl platform. These new combustion emissions replace the emissions derived from the MARAMA
projection.

The 2016fi case includes updates to a few specific ptnonipm units including the closure of the Guardian
Corp facility (#2989611) which closed in 2015, and adjusted the emissions at AV RANCHOS WATER -
WELL #4 to match those at WELL #9 because the emissions were determined to be unrealistically high.

27


-------
2.1.4 Aircraft and ground support equipment (airports)

The airport sector contains emissions of all pollutants from aircraft, categorized by their itinerant class
(i.e., commercial, air taxi, military, or general), as well as emissions from ground support equipment. The
starting point for the 2016 version 1 (vl) platform airport inventory is the airport emissions from the 2017
National Emissions Inventory (NEI). The SCCs included in the airport sector are shown in Table 2-7.

Table 2-7. 2016vl platform SCCs for the airports sector

S( (

Tier 1 description

Tier 2 (k'scriplion

Tier 3 (k'scriplion

Tier 4 (k'scriplion

2265008005

Mobile Sources

Off-highway Vehicle
Gasoline, 4-stroke

Airport Ground

Support

Equipment

Airport Ground
Support Equipment

2267008005

Mobile Sources

LPG

Airport Ground

Support

Equipment

Airport Ground
Support Equipment

2268008005

Mobile Sources

compressed natural gas
(CNG)

Airport Ground

Support

Equipment

Airport Ground
Support Equipment

2270008005

Mobile Sources

Off-highway Vehicle
Diesel

Airport Ground

Support

Equipment

Airport Ground
Support Equipment

2275001000

Mobile Sources

Aircraft

Military Aircraft

Total

2275020000

Mobile Sources

Aircraft

Commercial
Aircraft

Total: All Types

2275050011

Mobile Sources

Aircraft

General Aviation

Piston

2275050012

Mobile Sources

Aircraft

General Aviation

Turbine

2275060011

Mobile Sources

Aircraft

Air Taxi

Piston

2275060012

Mobile Sources

Aircraft

Air Taxi

Turbine

2275070000

Mobile Sources

Aircraft

Aircraft Auxiliary
Power Units

Total

40600307

Chemical
Evaporation

Transportation and
Marketing of Petroleum
Products

Gasoline Retail
Operations -
Stage I

Underground Tank
Breathing and
Emptying



Internal



Distillate Oil
(Diesel)



20200102

Combustion
Engines

Industrial

Reciprocating

The 2016vl airport emissions inventory was created from the 2017NEI airport emissions that were
estimated using the Federal Aviation Administration's (FAA's) Aviation Environmental Design Tool
(AEDT). Additional information about the 2017NEI airport inventory and the AEDT can be found in the
2017 National Emissions Inventory Technical Support Document (https://www.epa.gov/air-emissions-
inventories/2017-national-emissions-inventory-nei-technical-support-document-tsd). The 2017NEI
emissions were adjusted from 2017 to represent year 2016 emissions using FAA data. Adjustment factors
were created using airport-specific numbers, where available, or the state default by itinerant class
(commercial, air taxi, and general) where there were not airport-specific values in the FAA data.
Emissions growth for facilities is capped at 500% and the state default growth is capped at 200%. Military
state default values were kept flat to reflect uncertainly in the data regarding these sources.

28


-------
After the release of the April 2020 version of the 2017NEI, an error in the computation of the airport
emissions was identified and it was determined that they were overestimated. The error impacted
commercial aircraft emissions. The airport emission in the 2016fi case were recomputed based on
corrected 2017NEI emissions that were incorporated into the January 2021 release of 2017 NEI. The
corrected inventories and outputs from SMOKE were posted on the 2016vl FTP site
(ftp://newftp.epa.gov/air/emismod/2016/vl/postvl updates/ also available at
https://gaftp.epa.gov/Air/emismod/2016/vl/postvl updates).

2.2 2016 Nonpoint sources (afdust, ag, npoilgas, rwc, nonpt)

This section describes the stationary nonpoint sources in the NEI nonpoint data category. Locomotives,
CI and C2 CMV, and C3 CMV are included in the NEI nonpoint data category, but are mobile sources
that are described in Section 2.4.

The nonpoint tribal-submitted emissions are dropped during spatial processing with SMOKE due to the
configuration of the spatial surrogates. Part of the reason for this is to prevent possible double-counting
with county-level emissions and also because spatial surrogates for tribal data are not currently available.
These omissions are not expected to have an impact on the results of the air quality modeling at the 12-km
resolution used for this platform.

The following subsections describe how the sources in the NEI nonpoint inventory were separated into
modeling platform sectors, along with any data that were updated replaced with non-NEI data.

2.2.1 Area fugitive dust sector (afdust)

The area-source fugitive dust (afdust) sector contains PMio and PM2.5 emission estimates for nonpoint
SCCs identified by EPA as dust sources. Categories included in the afdust sector are paved roads,
unpaved roads and airstrips, construction (residential, industrial, road and total), agriculture production,
and mining and quarrying. It does not include fugitive dust from grain elevators, coal handling at coal
mines, or vehicular traffic on paved or unpaved roads at industrial facilities because these are treated as
point sources so they are properly located. Table 2-8 is a listing of the Source Classification Codes
(SCCs) in the afdust sector.

Table 2-8. Afdust sector SCCs

sec

Tier 1
description

Tier 2
description

Tier 3 description

Tier 4 description

2275085000

Mobile Sources

Aircraft

Unpaved Airstrips

Total

2294000000

Mobile Sources

Paved Roads

All Paved Roads

Total: Fugitives

2294000002

Mobile Sources

Paved Roads

All Paved Roads

Total: Sanding/Salting -
Fugitives

2296000000

Mobile Sources

Unpaved Roads

All Unpaved Roads

Total: Fugitives

2311000000

Industrial
Processes

Construction: SIC
15 -17

All Processes

Total

2311010000

Industrial
Processes

Construction: SIC
15 -17

Residential

Total

2311010070

Industrial
Processes

Construction: SIC
15 -17

Residential

Vehicle Traffic

2311020000

Industrial
Processes

Construction: SIC
15 -17

Industrial/Commercial/
Institutional

Total

2311030000

Industrial
Processes

Construction: SIC
15 -17

Road Construction

Total

29


-------
sec

Tier 1
description

Tier 2
description

Tier 3 description

Tier 4 description

2325000000

Industrial
Processes

Mining and
Quarrying: SIC 14

All Processes

Total

2325060000

Industrial
Processes

Mining and
Quarrying: SIC 10

Lead Ore Mining and Milling

Total

2801000000

Miscellaneous
Area Sources

Ag. Production -
Crops

Agriculture - Crops

Total

2801000003

Miscellaneous
Area Sources

Ag. Production -
Crops

Agriculture - Crops

Tilling

2801000005

Miscellaneous
Area Sources

Ag. Production -
Crops

Agriculture - Crops

Harvesting

2801000007

Miscellaneous
Area Sources

Ag. Production -
Crops

Agriculture - Crops

Loading

2801000008

Miscellaneous
Area Sources

Ag. Production -
Crops

Agriculture - Crops

Transport

2805001000

Miscellaneous
Area Sources

Ag. Production -
Livestock

Beef cattle - finishing operations
on feedlots (drylots)

Dust Kicked-up by Hooves
(use 28-05-020, -001,-002,
or -003 for Waste

2805001100

Miscellaneous
Area Sources

Ag. Production -
Livestock

Beef cattle - finishing operations
on feedlots (drylots)

Confinement

2805001200

Miscellaneous
Area Sources

Agriculture
Production -
Livestock

Beef cattle - finishing operations
on feedlots (drylots)

Manure handling and storage

2805001300

Miscellaneous
Area Sources

Agriculture
Production -
Livestock

Beef cattle - finishing operations
on feedlots (drylots)

Land application of manure

2805002000

Miscellaneous
Area Sources

Ag. Production -
Livestock

Beef cattle production composite

Not Elsewhere Classified

2805003100

Miscellaneous
Area Sources

Ag. Production -
Livestock

Beef cattle - finishing operations
on pasture/range

Confinement

2805007100

Miscellaneous
Area Sources

Ag. Production -
Livestock

Poultry production - layers with
dry manure management systems

Confinement

2805007300

Miscellaneous
Area Sources

Ag. Production -
Livestock

Poultry production - layers with
dry manure management systems

Land application of manure

2805008100

Miscellaneous
Area Sources

Ag. Production -
Livestock

Poultry production - layers with
wet manure management systems

Confinement

2805008200

Miscellaneous
Area Sources

Ag. Production -
Livestock

Poultry production - layers with
wet manure management systems

Manure handling and storage

2805008300

Miscellaneous
Area Sources

Ag. Production -
Livestock

Poultry production - layers with
wet manure management systems

Land application of manure

2805009100

Miscellaneous
Area Sources

Ag. Production -
Livestock

Poultry production - broilers

Confinement

2805009200

Miscellaneous
Area Sources

Ag. Production -
Livestock

Poultry production - broilers

Manure handling and storage

2805009300

Miscellaneous
Area Sources

Ag. Production -
Livestock

Poultry production - broilers

Land application of manure

2805010100

Miscellaneous
Area Sources

Ag. Production -
Livestock

Poultry production - turkeys

Confinement

2805010200

Miscellaneous
Area Sources

Ag. Production -
Livestock

Poultry production - turkeys

Manure handling and storage

2805010300

Miscellaneous
Area Sources

Ag. Production -
Livestock

Poultry production - turkeys

Land application of manure

2805018000

Miscellaneous
Area Sources

Ag. Production -
Livestock

Dairy cattle composite

Not Elsewhere Classified

2805019100

Miscellaneous
Area Sources

Ag. Production -
Livestock

Dairy cattle - flush dairy

Confinement

30


-------
sec

Tier 1
description

Tier 2
description

Tier 3 description

Tier 4 description

2805019200

Miscellaneous
Area Sources

Ag. Production -
Livestock

Dairy cattle - flush dairy

Manure handling and storage

2805019300

Miscellaneous
Area Sources

Ag. Production -
Livestock

Dairy cattle - flush dairy

Land application of manure

2805020002

Miscellaneous
Area Sources

Ag. Production -
Livestock

Cattle and Calves Waste
Emissions

Beef Cows

2805021100

Miscellaneous
Area Sources

Ag. Production -
Livestock

Dairy cattle - scrape dairy

Confinement

2805021200

Miscellaneous
Area Sources

Ag. Production -
Livestock

Dairy cattle - scrape dairy

Manure handling and storage

2805021300

Miscellaneous
Area Sources

Ag. Production -
Livestock

Dairy cattle - scrape dairy

Land application of manure

2805022100

Miscellaneous
Area Sources

Ag. Production -
Livestock

Dairy cattle - deep pit dairy

Confinement

2805022200

Miscellaneous
Area Sources

Ag. Production -
Livestock

Dairy cattle - deep pit dairy

Manure handling and storage

2805022300

Miscellaneous
Area Sources

Ag. Production -
Livestock

Dairy cattle - deep pit dairy

Land application of manure

2805023100

Miscellaneous
Area Sources

Ag. Production -
Livestock

Dairy cattle - drylot/pasture dairy

Confinement

2805023200

Miscellaneous
Area Sources

Ag. Production -
Livestock

Dairy cattle - drylot/pasture dairy

Manure handling and storage

2805023300

Miscellaneous
Area Sources

Ag. Production -
Livestock

Dairy cattle - drylot/pasture dairy

Land application of manure

2805025000

Miscellaneous
Area Sources

Ag. Production -
Livestock

Swine production composite

Not Elsewhere Classified
(see also 28-05-039, -047, -
053)

2805030000

Miscellaneous
Area Sources

Ag. Production -
Livestock

Poultry Waste Emissions

Not Elsewhere Classified
(see also 28-05-007, -008, -
009)

2805030007

Miscellaneous
Area Sources

Ag. Production -
Livestock

Poultry Waste Emissions

Ducks

2805030008

Miscellaneous
Area Sources

Ag. Production -
Livestock

Poultry Waste Emissions

Geese

2805035000

Miscellaneous
Area Sources

Ag. Production -
Livestock

Horses and Ponies Waste
Emissions

Not Elsewhere Classified

2805039100

Miscellaneous
Area Sources

Ag. Production -
Livestock

Swine production - operations
with lagoons (unspecified animal
age)

Confinement

2805039200

Miscellaneous
Area Sources

Ag. Production -
Livestock

Swine production - operations
with lagoons (unspecified animal
age)

Manure handling and storage

2805039300

Miscellaneous
Area Sources

Ag. Production -
Livestock

Swine production - operations
with lagoons (unspecified animal
age)

Land application of manure

2805040000

Miscellaneous
Area Sources

Ag. Production -
Livestock

Sheep and Lambs Waste
Emissions

Total

2805045000

Miscellaneous
Area Sources

Ag. Production -
Livestock

Goats Waste Emissions

Not Elsewhere Classified

2805047100

Miscellaneous
Area Sources

Ag. Production -
Livestock

Swine production - deep-pit house
operations (unspecified animal
age)

Confinement

2805047300

Miscellaneous
Area Sources

Ag. Production -
Livestock

Swine production - deep-pit house
operations (unspecified animal
age)

Land application of manure

31


-------
SCC

Tier 1
description

Tier 2
description

Tier 3 description

Tier 4 description

2805053100

Miscellaneous
Area Sources

Ag. Production -
Livestock

Swine production - outdoor
operations (unspecified animal
age)

Confinement

The starting point for the afdust emissions is the 2014 National Emissions Inventory version 2. The
methodologies to estimate emissions for each SCC in the preceding table are described in the 2014 NEI
version 2 Technical Support Document.2 The 2014 emissions were adjusted to better represent 2016 as
described below.

MARAMA States area fugitive dust emissions

The MARAMA states include Connecticut, Delaware, the District of Columbia (DC), Maine, Maryland,
Massachusetts, New Hampshire, New Jersey, New York, North Carolina, Pennsylvania, Rhode Island,
Vermont, Virginia, and West Virginia. MARAMA submitted county-specific projection factors for their
states to project afdust emissions from the 2014NEI2 to 2016 for paved roads (SCC 2294000000),
residential construction dust (SCC 2311010000), industrial/commercial/institutional construction dust
(SCC 2311020000), road construction dust (SCC 2311030000), dust from mining and quarrying (SCC
2325000000), agricultural crop tilling dust (SCC 2801000003), and agricultural dust kick-up from beef
cattle hooves (SCC 2805001000). Other afdust emissions, including unpaved road dust emissions, were
held constant at 2014NEIv2 values.

Non-MARAMA States area fugitive dust emissions

For paved roads (SCC 2294000000) in non-MARAMA states, the 2014NEIv2 paved road emissions in
afdust were projected to year 2016 based on differences in county total vehicle miles traveled (VMT)
between 2014 and 2016:

2016 afdust paved roads = 2014 afdust paved roads * (2016 county total VMT) / (2014 county total VMT)

The development of the 2016 VMT is described in the onroad documentation. All emissions other than
those for paved roads are held constant in the 2016vl inventory, including unpaved roads for these states.

Area Fugitive Dust Transport Fraction

The afdust sector is separated from other nonpoint sectors to allow for the application of a "transport
fraction," and meteorological/precipitation reductions. These adjustments are applied using a script that
applies land use-based gridded transport fractions based on landscape roughness, followed by another
script that zeroes out emissions for days on which at least 0.01 inches of precipitation occurs or there is
snow cover on the ground. The land use data used to reduce the NEI emissions determines the amount of
emissions that are subject to transport. This methodology is discussed in Pouliot, et al., 2010, and in
"Fugitive Dust Modeling for the 2008 Emissions Modeling Platform" (Adelman, 2012). Both the
transport fraction and meteorological adjustments are based on the gridded resolution of the platform (i.e.,
12km grid cells); therefore, different emissions will result if the process were applied to different grid
resolutions. A limitation of the transport fraction approach is the lack of monthly variability that would
be expected with seasonal changes in vegetative cover. While wind speed and direction are not accounted
for in the emissions processing, the hourly variability due to soil moisture, snow cover and precipitation is
accounted for in the subsequent meteorological adjustment.

2 https://www.epa.gov/air-emissions-inventories/2014-national-emissions-inventorv-nei-technical-support-document-tsd

32


-------
For the data compiled into the 2014NEIv2, meteorological adjustments are applied to paved and unpaved
road SCCs but not transport adjustments. For the 2014NEIvl, the meteorological adjustments were
inadvertently not applied. This created a large difference between the 2014NEIvl and 2014NEIv2 dust
emissions which did not impact the modeling platform because the modeling platform applies
meteorological adjustments and transport adjustments based on unadjusted NEI values (for both vl and
v2). Thus, for the 2014NEIv2, the meteorological adjustments that were applied (to paved and unpaved
road SCCs) had to be backed out so that the entire sector could be processed consistently in SMOKE and
the same grid-specific transport fractions and meteorological adjustments could be applied sector-wide.
Because it was determined that some counties in 2014NEIv2 did not have the adjustment applied, their
emissions were used as-is. Thus, the FF10 that is run through SMOKE consists of 100% unadjusted
emissions, and after SMOKE all afdust sources have both transport and meteorological adjustments
applied. The total impacts of the transport fraction and meteorological adjustments for 2016vl are shown
in Table 2-9. Note that while totals from AK, HI, PR, and VI are included at the bottom of the table, they
are from non-continental U.S. (non-CONUS) modeling domains.

Table 2-9. Total impact of fugitive dust adjustments to unadjusted 2016 vl inventory

SliUc

I iiiidjiislcd

PMio

I iiiidjiislcd
I'M:.?

< liiiniio in
PMio

< liiiniio in
I'M;?

PMio
Reduction

I'M:.?
Reduction

Alabama

535,218

63,682

-372,853

-44,336

70%

70%

Arizona

264,628

32,808

-96,814

-11,809

37%

36%

Arkansas

321,488

49,397

-211,050

-31,802

66%

64%

California

314,917

41,395

-134,347

-17,059

43%

41%

Colorado

242,327

36,848

-121,263

-17,718

50%

48%

Connecticut

23,740

3,385

-17,548

-2,510

74%

74%

Delaware

14,566

2,502

-8,843

-1,533

61%

61%

District of
Columbia

2,619

378

-1,627

-236

62%

62%

Florida

721,379

82,397

-412,621

-46,899

57%

57%

Georgia

557,354

66,609

-389,482

-46,272

70%

69%

Idaho

454,301

55,978

-241,373

-28,363

53%

51%

Illinois

997,748

143,992

-619,594

-88,735

62%

62%

Indiana

718,027

84,663

-498,442

-58,430

69%

69%

Iowa

387,029

60,253

-222,941

-34,557

58%

57%

Kansas

613,183

99,486

-277,007

-44,234

45%

44%

Kentucky

312,872

42,952

-233,163

-31,762

75%

74%

Louisiana

266,812

35,788

-172,875

-22,923

65%

64%

Maine

38,345

5,963

-31,893

-4,978

83%

83%

Maryland

105,892

16,672

-68,246

-10,824

64%

65%

Massachusetts

148,284

18,297

-112,998

-13,852

76%

76%

Michigan

390,994

48,838

-286,999

-35,560

73%

73%

Minnesota

405,052

61,723

-250,646

-37,609

62%

61%

Mississippi

434,575

53,546

-299,888

-36,494

69%

68%

Missouri

1,604,501

185,103

-1,084,830

-124,078

68%

67%

33


-------
Sialo

I nari.jiiMcri
PMu.

I nari.jiiMcri

I'M:..*

( haniie in

PIMio

Change in
PM:*

PMio
Reduction

PM: ?
Reduction

Montana

432,844

62,062

-236,341

-32,695

55%

53%

Nebraska

349,373

55,303

-165,083

-25,739

47%

47%

Nevada

161,820

23,360

-54,899

-7,953

34%

34%

New Hampshire

22,330

4,607

-18,436

-3,803

83%

83%

New Jersey

40,336

9,118

-26,776

-6,035

66%

66%

New Mexico

490,617

54,236

-200,695

-22,038

41%

41%

New York

264,041

44,137

-196,162

-32,785

74%

74%

North Carolina

206,465

30,017

-141,501

-20,610

69%

69%

North Dakota

473,241

82,478

-249,646

-43,138

53%

52%

Ohio

931,847

116,560

-638,127

-79,098

68%

68%

Oklahoma

450,904

67,915

-232,046

-33,983

51%

50%

Oregon

659,099

73,832

-456,949

-49,830

69%

67%

Pennsylvania

242,608

37,707

-179,647

-27,959

74%

74%

Rhode Island

4,935

785

-3,503

-556

71%

71%

South Carolina

164,477

22,016

-110,278

-14,795

67%

67%

South Dakota

339,195

63,248

-169,300

-31,302

50%

49%

Tennessee

295,092

43,414

-204,746

-29,995

69%

69%

Texas

1,264,131

180,314

-636,591

-87,931

50%

49%

Utah

209,800

26,453

-111,587

-13,771

53%

52%

Vermont

22,437

3,275

-18,644

-2,699

83%

82%

Virginia

286,237

37,007

-211,882

-27,348

74%

74%

Washington

242,907

41,851

-135,713

-23,281

56%

56%

West Virginia

123,003

15,127

-105,093

-12,911

85%

85%

Wisconsin

690,830

89,899

-486,508

-62,683

70%

70%

Wyoming

240,156

29,140

-123,388

-14,561

51%

50%

Domain Total
(12km CONUS)

18,484,575

2,506,516

11,280,883

-1,500,070

61%

60%

Alaska

112,025

11,562

-101,822

-10,508

91%

91%

Hawaii

109,120

11,438

-73,612

-7,673

67%

67%

Puerto Rico

5,889

1,313

-4,355

-984

74%

75%

Virgin Islands

3,493

467

-1,477

-195

42%

42%

Figure 2-1 illustrates the impact of each step of the adjustment. The reductions due to the transport
fraction adjustments alone are shown at the top of the figure. The reductions due to the precipitation
adjustments alone are shown in the middle of the figure. The cumulative emission reductions after both
transport fraction and meteorological adjustments are shown at the bottom of the figure. The top plot
shows how the transport fraction has a larger reduction effect in the east, where forested areas are more
effective at reducing PM transport than in many western areas. The middle plot shows how the
meteorological impacts of precipitation, along with snow cover in the north, further reduce the dust
emissions.

34


-------
Figure 2-1. Impact of adjustments to fugitive dust emissions due to transport fraction,

precipitation, and cumulative
2016fh (vl) afdust annual : PM2 5, xportfrac - unadjusted

%

\ It

m-

i &



r

¦ A

\J4

Y-v'v/^

fSKf v

\

* i

/¦ ¦ .'"~i ¦*¦?"*S \ V I ' V

—tv	/ %

	^

#



: < /

./ I

it

I

t

A/



$4 >'

/ i \ /tz" /

) vy r

L )

\ \ \
, vtX
wl\

w \

M

7 x



/ \



a r

Max: 0.0 Mjrv -B20.4

•A\v

i . V J

M %y£® /'

1|

2016fh (vl) afdust annual : PM2 5, precip adjusted - xportfrac adjusted

m



/ [ft
F-

i

i

£3

l A

/r

L\

*,

W-^v'

A



\,

	--JU _

\l

\ * "7?

\ KN;

V0 i

Jy, i
\

".. "j i .4^ .

—¦ I-.. O , v'-S	, .

^:ey;S^/7"W

	¦%

-J /¦" * 'V\ \y

V—¦ I ->

-jwy

ylr

\

v

Max: 0.0001373 Min: -405.6

Y« v, ^

U, \ c

X
/
/

) *-



\-i

\



J \ ^-3

irX -

% N -x -
y ¦

35


-------
2.2.2 Agriculture Sector (ag)

The ag sector includes NH3 emissions from fertilizer and emissions of all pollutants other than PM2.5
from livestock in the nonpoint (county-level) data category of the 2017NEI. PM2.5 from livestock are in
the Area Fugitive Dust (afdust) sector. Combustion emissions from agricultural equipment, such as
tractors, are in the Nonroad sector. The sector now includes VOC and HAP VOC in addition to NH3.
The 2016 version 1 (vl) platform uses a 2016-specific fertilizer inventory from the USDA's
Environmental Policy Integrated Climate (EPIC) model combined with a 2016 USDA-based county-level
back-projection of 2017NEI livestock emissions. The SCCs included in the ag sector are shown in Table
2-10.

Table 2-10. 2016vl platform SCCs for the ag sector

see

Tier 1 description

Tier 2 description

Tier 3 description

Tier 4 description

2801700099

Miscellaneous
Area Sources

Ag. Production
- Crops

Fertilizer Application

Miscellaneous
Fertilizers

2805002000

Miscellaneous
Area Sources

Ag. Production
- Livestock

Beef cattle production
composite

Not Elsewhere
Classified

2805007100

Miscellaneous
Area Sources

Ag. Production
- Livestock

Poultry production -
layers with dry manure
management systems

Confinement

2805009100

Miscellaneous
Area Sources

Ag. Production
- Livestock

Poultry production -
broilers

Confinement

2805010100

Miscellaneous
Area Sources

Ag. Production
- Livestock

Poultry production -
turkeys

Confinement

2805018000

Miscellaneous
Area Sources

Ag. Production
- Livestock

Dairy cattle composite

Not Elsewhere
Classified

36


-------
SCC

Tier 1 description

Tier 2 (k'scriplion

Tier 3 (k'scriplion

Tier 4 (k'scriplion

2805025000

Miscellaneous
Area Sources

Ag. Production
- Livestock

Swine production
composite

Not Elsewhere
Classified (see also
28-05-039,-047, -053)

2805035000

Miscellaneous
Area Sources

Ag. Production
- Livestock

Horses and Ponies
Waste Emissions

Not Elsewhere
Classified

2805040000

Miscellaneous
Area Sources

Ag. Production
- Livestock

Sheep and Lambs Waste
Emissions

Total

2805045000

Miscellaneous
Area Sources

Ag. Production
- Livestock

Goats Waste Emissions

Not Elsewhere
Classified

2.2.2.1 Livestock Waste Emissions

The 2016vl platform livestock emissions consist of a back-projection of 2017NEI livestock emissions to
the year 2016 and include NH3 and VOC. The livestock waste emissions from 2017NEI contain
emissions for beef cattle, dairy cattle, goats, horses, poultry, sheep, and swine. The data come from both
state-submitted emissions and EPA-calculated emission estimates. Further information about the 2017NEI
emissions can be found in the 2017 National Emissions Inventory Technical Support Document
(https://www.epa.gov/air-emissions-inventories/2017-national-emissions-inventory-nei-technical-support-
document-tsd). Back-projection factors for 2016 emission estimates are based on animal population data
from the USDA National Agriculture Statistics Service Quick Stats

(https://www.nass.usda.gov/Quick_Stats/). These estimates are developed by data collected from annual
agriculture surveys and the Census of Agriculture that is completed every five years. These data include
estimates for beef, layers, broilers, turkeys, dairy, swine, and sheep. Each SCC in the 2017NEI livestock
inventory, except for 2805035000 (horses and ponies) and 2805045000 (goats), was mapped to one of
these USDA categories. Then, back-projection factors were calculated based on USDA animal
populations for 2016 and 2017. Emissions for animal categories for which population data were not
available (e.g. horses, goats) were held constant in the projection.

Back-projection factors were calculated at the county level, but only where county-level data was
available for a specific animal category. County-level factors were limited to a range of 0.8 to 1.2. Data
were not available for every animal category in every county. State-wide back-projection factors based on
state total animal populations were calculated and applied to counties where county-specific data was not
available for a given animal category. However, data were often not available for every animal category
in every state. For categories other than beef and dairy, data are not available for most states. In cases of
missing state-level data, a national back-projection factor was applied. Back-projection factors were not
pollutant-specific and were applied to all pollutants. The national back-projection factors, which were
only used when county or state data were not available, are shown in Table 2-11. The national factors
were created using a ratio between animal inventory counts for 2017 and 2016 from the USDA National
livestock inventory projections published in February 2018
(https://www.ers.usda.gov/webdocs/outlooks/87459/oce-2018-l.pdf?v=7587.1).

Table 2-11. National back-projection factors for livestock: 2017 to 2016

beef

-1.8%

swine

-3.6%

broilers

-2.0%

turkeys

-0.3%

layers

-2.3%

37


-------
dairy

-0.4%

sheep

+0.4%

2.2.2.2 Fertilizer Emissions

Fertilizer emissions for 2016 are based on the Fertilizer Emission Scenario Tool for CMAQ (FEST-C)
model (https://www.cmascenter.org/fest-c/). The bidirectional version of CMAQ (v5.3) and the Fertilizer
Emissions Scenario Tool for CMAQ FEST-C (vl.3) were used to estimate ammonia (NH3) emissions
from agricultural soils. The approach to estimate year-specific fertilizer emissions consists of these steps:

•	Run FEST-C to produce nitrate (N03), Ammonium (NH4+, including Urea), and organic
(manure) nitrogen (N) fertilizer usage estimates

•	Use USDA Economic Research Services crop specific fertilizer use data and state submitted data
to adjust the FEST-C fertilizer totals to match the USDA and State submitted.

•	Run the CMAQ model with bidirectional ("bidi") NH3 exchange to generate gaseous ammonia
NH3 emission estimates.

•	Calculate county-level emission factors as the ratio of bidirectional CMAQ NH3 fertilizer
emissions to FEST-C total N fertilizer application.

•	Assign the NH3 emissions to one SCC: ".. .Miscellaneous Fertilizers" (2801700099).

FEST-C is the software program that processes land use and agricultural activity data to develop inputs
for the CMAQ model when run with bidirectional exchange. FEST-C reads land use data from the
Biogenic Emissions Landuse Dataset (BELD), meteorological variables from the Weather Research and
Forecasting model, and nitrogen deposition data from a previous or historical average CMAQ simulation.
FEST-C, then uses the EPIC modeling system (https://epicapex.tamu.edu/epic/) to simulate the
agricultural practices and soil biogeochemistry and provides information regarding fertilizer timing,
composition, application method and amount.

An iterative calculation was applied to estimate fertilizer emissions for the 2016 platform. We first
estimate fertilizer application by crop type using FEST-C modeled data. After receipt and addressing of
comments to the extent possible, we then adjusted the fertilizer application estimates using state submitted
data, (currently only Iowa), and USDA Economic Research Service state and crop specific survey data.
The USDA and state submitted annual fertilizer data was used to estimate the ratio of UDSA/state
fertilizer use to FEST-C annual total fertilizer estimates for each state and crop with USDA or state data.
This ratio is then applied to the FEST-C fertilizer application rates for each state and crop with data. A
maximum annual fertilization rate was estimated from the FEST-C simulation and annual adjusted totals
were limited to this rate to prevent unrealistically higher fertilization rates. Then we ran the CMAQ v5.3
model with the Surface Tiled Aerosol and Gaseous Exchange (STAGE) deposition option with
bidirectional exchange to estimate fertilizer and biogenic NH3 emissions. We use this approach for three
reasons: (1) FEST-C estimates fertilizer applications based on crop nutrient needs which is typically lower
than real world fertilization rates; (2) FEST-C fertilizer timing and application methods are assumed to be
correct; and (3) We desired a method to incorporate state submitted and USDA reported data into the final
fertilization emission estimates.

Example Calculation:

Adjustment of FEST-C fertilizer rates using state or USDA data:

38


-------
/

l'Crta justed,crop TTICIX

Fert submitted,crop ,,

I	; e/ r FEST-C,crop'' : max,crop

\ncrop

Z Per^FEST -C.crop

Where Fertadjusted,cropis the FEST-C 12km grid cell adjusted fertilization rate, Fertsubmittedxropis the USDA
or State submitted state mean annual application data for the specified crop, in kg ha"1, FERTfest-c.ctop is
the initial FEST-C 12km grid cell fertilization rate for the state being considered, ncrop is the number of
grid cells with fertilization use for the specified crop in the state, and Fertmax,crop is the maximum
fertilization rate estimated from EPIC for the crop.

Figure 2-2. "Bidi" modeling system used to compute 2016 Fertilizer Application emissions

The Fertilizer Emission Scenario Tool for CMAQ

(FEST-C)

Fertilizer Activity Data

The following activity parameters were input into the EPIC model:

•	Grid cell meteorological variables from WRF (see Table 3)

•	Initial soil profiles/soil selection

•	Presence of 21 major crops: irrigated and rain fed hay, alfalfa, grass, barley, beans, grain corn,
silage corn, cotton, oats, peanuts, potatoes, rice, rye, grain sorghum, silage sorghum, soybeans,
spring wheat, winter wheat, canola, and other crops (e.g.. lettuce, tomatoes, etc.)

•	Fertilizer sales to establish the type/composition of nutrients applied

39


-------
• Management scenarios for the 10 USDA production regions. These include irrigation, tile

drainage, intervals between forage harvest, fertilizer application method (injected versus surface
applied), and equipment commonly used in these production regions.

The WRF meteorological model was used to provide grid cell meteorological parameters for year 2016
using a national 12-km rectangular grid covering the continental U.S. The meteorological parameters in
Table 2-12 were used as EPIC model inputs.

Table 2-12. Source of input variables for EPIC

EPIC input variable

Variable Source

Daily Total Radiation (MJ m2)

WRF

Daily Maximum 2-m Temperature (C)

WRF

Daily minimum 2-m temperature (C)

WRF

Daily Total Precipitation (mm)

WRF

Daily Average Relative Humidity (unitless)

WRF

Daily Average 10-m Wind Speed (m s"1 )

WRF

Daily Total Wet Deposition Oxidized N (g/ha)

CMAQ

Daily Total Wet Deposition Reduced N (g/ha)

CMAQ

Daily Total Dry Deposition Oxidized N (g/ha)

CMAQ

Daily Total Dry Deposition Reduced N (g/ha)

CMAQ

Daily Total Wet Deposition Organic N (g/ha)

CMAQ

Initial soil nutrient and pH conditions in EPIC were based on the 1992 USDA Soil Conservation Service
(CSC) Soils-5 survey. The EPIC model then was run for 25 years using current fertilization and
agricultural cropping techniques to estimate soil nutrient content and pH for the 2016 EPIC/WRF/CMAQ
simulation.

The presence of crops in each model grid cell was determined through the use of USDA Census of
Agriculture data (2012) and USGS National Land Cover data (2011). These two data sources were used to
compute the fraction of agricultural land in a model grid cell and the mix of crops grown on that land.

Fertilizer sales data and the 6-month period in which they were sold were extracted from the 2014
Association of American Plant Food Control Officials (AAPFCO,

http://www.aapfco.org/publications.htmn. AAPFCO data were used to identify the composition (e.g.,
urea, nitrate, organic) of the fertilizer used, and the amount applied is estimated using the modeled crop
demand. These data were useful in making a reasonable assignment of what kind of fertilizer is being
applied to which crops.

Management activity data refers to data used to estimate representative crop management schemes. The
USDA Agricultural Resource Management Survey (ARMS,

https://www.nass.usda.gov/Surveys/Guide to NASS Surveys/Ag Resource Management/) was used to
provide management activity data. These data cover 10 USDA production regions and provide

40


-------
management schemes for irrigated and rain fed hay, alfalfa, grass, barley, beans, grain corn, silage corn,
cottonoats, peanuts, potatoes, rice, rye, grain sorghum, silage sorghum, soybeans, spring wheat, winter
wheat, canola, and other crops (e.g. lettuce, tomatoes, etc.).

Fertilizer Emission Factors

The emission factors were derived from the 2016 CMAQ FEST-C outputs adjusted using USDA
Economic Research Service (ERS) state and crop specific reported annual fertilizer rates. Total fertilizer
emission factors for each month and county were computed by taking the ratio of total fertilizer NH3
emissions (short tons) to total nitrogen fertilizer application (short tons).

12 km by 12 km gridded NH3 emissions were mapped to a county shape file polygon. The cell was
assigned to a county if the grid centroid fell within the county boundary.

2.2.3 Nonpoint Oil and Gas Sector (np_oilgas)

While the major emissions sources associated with oil and gas collection, processing, and distribution
have traditionally been included in the National Emissions Inventory (NEI) as point sources (e.g., gas
processing plants, pipeline compressor stations, and refineries), the activities occurring "upstream" of
these types of facilities have not been as well characterized in the NEI. Here, upstream activities refer to
emission units and processes associated with the exploration and drilling of oil and gas wells, and the
equipment used at the well site to then extract the product from the well and deliver it to a central
collection point or processing facility. The types of unit processes found at upstream sites include
separators, dehydrators, storage tanks, and compressor engines.

The nonpoint oil and gas (np oilgas) sector, which consists of oil and gas exploration and production
sources, both onshore and offshore (state-owned only). In the 2016vl platform, these emissions are
mostly based on the EPA Oil and Gas Tool run with data specific to the year 2016, with some states
submitting their own inventory data. Because of the growing importance of these emissions, special
consideration is given to the speciation, spatial allocation, and monthly temporalization of nonpoint oil
and gas emissions, instead of relying on older, more generalized profiles.

EPA Oil and Gas Tool

EPA developed the 2016 Nonpoint Oil and Gas Emission Estimation Tool (the "Tool") to estimate the
non-point oil and gas inventory for the 2016vl platform. The Tool was previously used to estimate
emissions for the 2014 NEI. Year 2016 oil and gas activity data were supplied to EPA by some state air
agencies, and where state data were not supplied to EPA, EPA populated the 2016vl inventory with the
best available data. The Tool is an Access database that utilizes county-level activity data (e.g. oil
production and well counts), operational characteristics (types and sizes of equipment), and emission
factors to estimate emissions. The Tool creates a CSV-formatted emissions dataset covering all national
nonpoint oil and gas emissions. This dataset is then converted to FF10 format for use in SMOKE
modeling. A separate report named "2016 Nonpoint Oil and Gas Emission Estimation Tool V1_0
December_2018.docx" was generated that provides technical details of how the tool was applied for the
2016vl platform (ERG, 2018).

In the 2016beta platform, it was found that the number of active wells in the state of Illinois was too high
(-48,000 total wells). After various discussions and other communications with the Illinois
Environmental Protection Agency (IEPA), a more accurate number of active of wells (-20,000 total
wells) was obtained and the new data were used in a rerun of the Oil and Gas Tool to produce new
emissions for the state of Illinois. These new emissions estimates for Illinois are in the 2016vl modeling

41


-------
platform. The reduction in total number of active wells resulted in NOX and VOC emissions being
reduced by about 14,000 tons and 48,000 tons, respectively, in 2016vl when compared to 2016beta
emissions.

Nonpoint Oil and Gas Alternative Datasets

Some states provided, or recommended use of, a separate emissions inventory for use in 2016vl platform
instead of emissions derived from the EPA Oil and Gas Tool. For example, the California Air Resources
Board (CARB) developed their own npoilgas emissions inventory for 2016 for California that were used
for the 2016vl platform.

In Pennsylvania for the 2016vl modeling platform, the emissions associated with unconventional wells
for year 2016 were supplied by the Pennsylvania Department of Environmental Protection (PA DEP). The
Oil and Gas Tool was used to produce the conventional well emissions for 2016. Together these
unconventional and conventional well emissions represent the total non-point oil and gas emissions for
Pennsylvania. The resulting NOX emissions for Pennsylvania were increased by about 16,000 tons in
2016vl when compared to the 2016beta emissions. The VOC emissions were reduced by about 56,000
tons in 2016vl due to these emissions changes in Pennsylvania.

Colorado Department of Public Health and Environment (CDPHE) requested that the 2014NEIv2 be
projected to 2016 instead of using data from the EPA Oil and Gas Tool. For Colorado projections were
applied to CO, NOX, PM, and S02, but not VOC. VOC emissions for year 2016 were assumed to equal
year 2014 levels for Colorado. Projection factors for Colorado are listed in Table 2-13 and are based on
historical production trends.

Oklahoma Department of Environmental Quality requested that np oilgas emissions from 2014NEIv2 be
projected to 2016 for all source except lateral compressors. Projection factors for Oklahoma np oilgas
production, based on historical production data, are listed in Table 2-13. For lateral compressor emissions
in Oklahoma, the EPA Oil and Gas Tool inventory for 2016 was used, except with a 72% cut applied to
all emissions. Exploration np oilgas emissions in Oklahoma are based on the EPA Oil and Gas Tool
inventory for 2016, without modification.

Table 2-13. 2014NEIv2-to-2016 oil and gas projection factors for CO and OK.

State/region

Emissions type

Factor

Pollutant(s)

Colorado

Oil

+22.0%

CO, NOX, S02

Colorado

Natural Gas

+3.5%

CO, NOX, PM, S02

Colorado

Combination Oil + NG

+12.8%

CO, NOX, PM, S02

Oklahoma

Oil Production

+6.9%

All

Oklahoma

Natural Gas Production

+5.9%

All

Oklahoma

Combination Oil + NG Production

+6.4%

All

Oklahoma

Coal Bed Methane Production

-30.0%

All

42


-------
2.2.4 Residential Wood Combustion (rwc)

The RWC sector includes residential wood burning devices such as fireplaces, fireplaces with inserts, free
standing woodstoves, pellet stoves, outdoor hydronic heaters (also known as outdoor wood boilers),
indoor furnaces, and outdoor burning in firepits and chimneys. Free standing woodstoves and inserts are
further differentiated into three categories: 1) conventional (not EPA certified); 2) EPA certified,
catalytic; and 3) EPA certified, noncatalytic. Generally, the conventional units were constructed prior to
1988. Units constructed after 1988 had to meet EPA emission standards and they are either catalytic or
non-catalytic. The source classification codes (SCCs) in the RWC sector are listed in Table 2-14.

Table 2-14. 2016 vl platform SCCs for RWC sector

SCC

Tier 1 Description

Tier 2
Description

Tier 3
Description

Tier 4 Description

2104008100

Stationary Source
Fuel Combustion

Residential

Wood

Fireplace: general

2104008210

Stationary Source
Fuel Combustion

Residential

Wood

Woodstove: fireplace
inserts; non-EPA certified

2104008220

Stationary Source
Fuel Combustion

Residential

Wood

Woodstove: fireplace
inserts; EPA certified; non-
catalytic

2104008230

Stationary Source
Fuel Combustion

Residential

Wood

Woodstove: fireplace
inserts; EPA certified;
catalytic

2104008310

Stationary Source
Fuel Combustion

Residential

Wood

Woodstove: freestanding,
non-EPA certified

2104008320

Stationary Source
Fuel Combustion

Residential

Wood

Woodstove: freestanding,
EPA certified, non-catalytic

2104008330

Stationary Source
Fuel Combustion

Residential

Wood

Woodstove: freestanding,
EPA certified, catalytic

2104008400

Stationary Source
Fuel Combustion

Residential

Wood

Woodstove: pellet-fired,
general (freestanding or FP
insert)

2104008510

Stationary Source
Fuel Combustion

Residential

Wood

Furnace: Indoor,
cordwood-fired, non-EPA
certified

2104008610

Stationary Source
Fuel Combustion

Residential

Wood

Hydronic heater: outdoor

2104008700

Stationary Source
Fuel Combustion

Residential

Wood

Outdoor wood burning
device, NEC (fire-pits,
chimeas, etc)

2104009000

Stationary Source
Fuel Combustion

Residential

Firelog

Total: All Combustor
Types

For all states other than California, Washington, and Oregon RWC emissions from the NEI2014v2 were
projected to 2016 using projection factors derived by MARAMA based on implementing the projection
methodology from EPA's 2011 platform into a spreadsheet tool. Projection factors are by SCC and SCC-
pollutant; SCC-only factors (i.e., factors that do not specify a pollutant) are applied to all pollutants
without an SCC-pollutant factor. Table 2-15 lists the SCC-based projection factors applied to RWC
sources.

43


-------
Table 2-15. Projection factors for RWC by SCC

SCC

SCC description

I'olllllillll

2014-l<>-2016

2104008100

Fireplace: general



2.00%

2104008210

Woodstove

fireplace inserts; non-EPA certified



-3.40%

2104008220

Woodstove

fireplace inserts; EPA certified; non-catalytic

PM10-PRI

2.29%

2104008220

Woodstove

fireplace inserts; EPA certified; non-catalvtic

PM25-PRI

2.29%

2104008220

Woodstove

fireplace inserts; EPA certified; non-catalytic



5.25%

2104008230

Woodstove

fireplace inserts; EPA certified; catalytic

PM10-PRI

2.44%

2104008230

Woodstove

fireplace inserts; EPA certified; catalytic

PM25-PRI

2.44%

2104008230

Woodstove

fireplace inserts; EPA certified; catalytic



5.25%

2104008310

Woodstove

freestanding, non-EPA certified

CO

-2.35%

2104008310

Woodstove

freestanding, non-EPA certified

PM10-PRI

-2.17%

2104008310

Woodstove

freestanding, non-EPA certified

PM25-PRI

-2.17%

2104008310

Woodstove

freestanding, non-EPA certified

VOC

-2.06%

2104008310

Woodstove

freestanding, non-EPA certified



-2.35%

2104008320

Woodstove

freestanding, EPA certified, non-catalvtic

PM10-PRI

2.29%

2104008320

Woodstove

freestanding, EPA certified, non-catalytic

PM25-PRI

2.29%

2104008320

Woodstove

freestanding, EPA certified, non-catalytic



5.25%

2104008330

Woodstove

freestanding, EPA certified, catalytic

PM10-PRI

2.47%

2104008330

Woodstove

freestanding, EPA certified, catalytic

PM25-PRI

2.47%

2104008330

Woodstove

freestanding, EPA certified, catalytic



5.25%

2104008400

Woodstove

pellet-fired, general (freestanding or FP insert)

PM10-PRI

14.40%

2104008400

Woodstove

pellet-fired, general (freestanding or FP insert)

PM25-PRI

14.40%

2104008400

Woodstove

pellet-fired, general (freestanding or FP insert)



14.38%

2104008510

Furnace: Indoor, cordwood-fired, non-EPA certified

CO

-9.70%

2104008510

Furnace: Indoor, cordwood-fired, non-EPA certified

PM10-PRI

-6.15%

2104008510

Furnace: Indoor, cordwood-fired, non-EPA certified

PM25-PRI

-6.15%

2104008510

Furnace: Indoor, cordwood-fired, non-EPA certified

VOC

-9.74%

2104008510

Furnace: Indoor, cordwood-fired, non-EPA certified



-9.70%

2104008610

Hvdronic heater: outdoor

PM10-PRI

2.99%

2104008610

Hydronic heater: outdoor

PM25-PRI

2.99%

2104008610

Hydronic heater: outdoor



2.00%

2104008700

Outdoor wood burning device, NEC (fire-pits, chimineas, etc)



2.00%

2104009000

Fire log total



2.00%

For California, Oregon, and Washington, the RWC emissions were held constant atNEI2014v2 levels for
2016. This approach is consistent with the RWC projections used in the EPA's 2011 emissions modeling
platform.

After the 2014NEIv2 was published, it was determined that the 2014NEIv2 RWC inventory was missing
woodstove emissions for certain pollutants in Idaho. The missing emissions for woodstove SCCs
2104008210, 2104008230, 2104008310, 2104008330 were added to the inventory prior to projecting it to
2016 for the vl platform.

2.2.5 Nonpoint (nonpt)

The starting point for the 2016vl platform nonpt inventory is the 2014NEIv2, including all nonpoint
sources that are not included in the afdust, ag, cmv_clc2, cmv_c3, np oilgas, rail, or rwc sectors. The
types of sources in the nonpt sector include, but are not limited to:

44


-------
•	stationary source fuel combustion, including industrial, commercial, and residential and orchard
heaters;

•	commercial sources such as commercial cooking;

•	industrial processes such as chemical manufacturing, metal production, mineral processes,
petroleum refining, wood products, fabricated metals, and refrigeration;

•	solvent utilization for surface coatings such as architectural coatings, auto refinishing, traffic
marking, textile production, furniture finishing, and coating of paper, plastic, metal, appliances,
and motor vehicles;

•	solvent utilization for degreasing of furniture, metals, auto repair, electronics, and manufacturing;

•	solvent utilization for dry cleaning, graphic arts, plastics, industrial processes, personal care
products, household products, adhesives and sealants;

•	solvent utilization for asphalt application and roofing, and pesticide application;

•	storage and transport of petroleum for uses such as portable gas cans, bulk terminals, gasoline
service stations, aviation, and marine vessels;

•	storage and transport of chemicals;

•	waste disposal, treatment, and recovery via incineration, open burning, landfills, and composting;

•	cellulosic biorefining;

•	miscellaneous area sources such as cremation, hospitals, lamp breakage, and automotive repair
shops.

The nonpoint emissions in 2016vl platform are equivalent to those in the 2014NEIv2 except for the
following changes:

Nonyoint projection to 2016 inside MARAMA region

2014-to-2016 projection packets for all nonpoint sources were provided by MARAMA for the following
states: CT, DE, DC, ME, MD, MA, NH, NJ, NY, NC, PA, RI, VT, VA, and WV.

New Jersey provided their own projection factors for projection from 2014 to 2016 which were mostly the
same as those provided by MARAMA, except for three SCCs with differences (SCCs: 2302070005,
2401030000, 2401070000). For those three SCCs, the projection factors provided by New Jersey were
used instead of the MARAMA factors.

Nonyoint projection to 2016 outside MARAMA region

In areas outside of the MARAMA states, historical census population, sometimes by county and
sometimes by state, was used to project select nonpt sources from the 2014NEIv2 to 2016vl platform.
The population data was downloaded from the US Census Bureau. Specifically, the "Population,
Population Change, and Estimated Components of Population Change: April 1, 2010 to July 1, 2017" file
(https://www2.census.gov/programs-survevs/popest/datasets/2010-2017/counties/totals/co-est2Q17-
alldata.csv). A ratio of 2016 population to 2014 population was used to create a growth factor that was
applied to the 2014NEIv2 emissions with SCCs matching the population-based SCCs listed in Table 2-16.
Positive growth factors (from increasing population) were not capped, but negative growth factors (from
decreasing population) were flatlined for no growth.

45


-------
Table 2-16. 2016vl platform SCCs for Census-based growth

S( (

Tier 1

Description

Tier 2 Description

Tier 3
Description

Tier 4
Description

2302002100

Industrial
Processes

Food and Kindred
Products: SIC 20

Commercial Charbroiling

Conveyorized Charbroiling

2302002200

Industrial
Processes

Food and Kindred
Products: SIC 20

Commercial Charbroiling

Under-fired Charbroiling

2302003000

Industrial
Processes

Food and Kindred
Products: SIC 20

Commercial Deep Fat
Frying

Total

2302003100

Industrial
Processes

Food and Kindred
Products: SIC 20

Commercial Deep Fat
Frying

Flat Griddle Frying

2302003200

Industrial
Processes

Food and Kindred
Products: SIC 20

Commercial Deep Fat
Frying

Clamshell Griddle Frying

2401001000

Solvent
Utilization

Surface Coating

Architectural Coatings

Total: All Solvent Types

2401002000

Solvent
Utilization

Surface Coating

Architectural Coatings -
Solvent-based

Total: All Solvent Types

2401003000

Solvent
Utilization

Surface Coating

Architectural Coatings -
Water-based

Total: All Solvent Types

2401100000

Solvent
Utilization

Surface Coating

Industrial Maintenance
Coatings

Total: All Solvent Types

2401200000

Solvent
Utilization

Surface Coating

Other Special Purpose
Coatings

Total: All Solvent Types

2425000000

Solvent
Utilization

Graphic Arts

All Processes

Total: All Solvent Types

2425010000

Solvent
Utilization

Graphic Arts

Lithography

Total: All Solvent Types

2425020000

Solvent
Utilization

Graphic Arts

Letterpress

Total: All Solvent Types

2425030000

Solvent
Utilization

Graphic Arts

Rotogravure

Total: All Solvent Types

2425040000

Solvent
Utilization

Graphic Arts

Flexography

Total: All Solvent Types

2440020000

Solvent
Utilization

Miscellaneous
Industrial

Adhesive (Industrial)
Application

Total: All Solvent Types

2460000000

Solvent
Utilization

Miscellaneous Non-
industrial: Consumer
and Commercial

All Processes

Total: All Solvent Types

2460100000

Solvent
Utilization

Miscellaneous Non-
industrial: Consumer
and Commercial

All Personal Care Products

Total: All Solvent Types

2460200000

Solvent
Utilization

Miscellaneous Non-
industrial: Consumer
and Commercial

All Household Products

Total: All Solvent Types

2460400000

Solvent
Utilization

Miscellaneous Non-
industrial: Consumer
and Commercial

All Automotive Aftermarket
Products

Total: All Solvent Types

2460500000

Solvent
Utilization

Miscellaneous Non-
industrial: Consumer
and Commercial

All Coatings and Related
Products

Total: All Solvent Types

2460600000

Solvent
Utilization

Miscellaneous Non-
industrial: Consumer
and Commercial

All Adhesives and Sealants

Total: All Solvent Types

2460800000

Solvent
Utilization

Miscellaneous Non-
industrial: Consumer
and Commercial

All FIFRA Related Products

Total: All Solvent Types

46


-------
S( (

Tier 1

Tier 2 Description

Tier 3

Tier 4



Description



Description

Description

2460900000

Solvent

Miscellaneous Non-

Miscellaneous Products

Total: All Solvent Types



Utilization

industrial: Consumer
and Commercial

(Not Otherwise Covered)



2461800000

Solvent

Miscellaneous Non-

Pesticide Application: All

Total: All Solvent Types



Utilization

industrial:
Commercial

Processes



2461800001

Solvent

Miscellaneous Non-

Pesticide Application: All

Surface Application



Utilization

industrial:
Commercial

Processes



2461800002

Solvent

Miscellaneous Non-

Pesticide Application: All

Soil Incorporation



Utilization

industrial:
Commercial

Processes



2461870999

Solvent

Miscellaneous Non-

Pesticide Application: Non-

Not Elsewhere Classified



Utilization

industrial:
Commercial

Agricultural



2465800000

Solvent
Utilization

Miscellaneous Non-
industrial: Consumer

Pesticide Application

Total: All Solvent Types

2501011011

Storage and

Petroleum and

Residential Portable Gas

Permeation



Transport

Petroleum Product
Storage

Cans



2501011012

Storage and

Petroleum and

Residential Portable Gas

Evaporation (includes Diurnal



Transport

Petroleum Product
Storage

Cans

losses)

2501011013

Storage and

Petroleum and

Residential Portable Gas

Spillage During Transport



Transport

Petroleum Product
Storage

Cans



2501011014

Storage and

Petroleum and

Residential Portable Gas

Refilling at the Pump - Vapor



Transport

Petroleum Product
Storage

Cans

Displacement

2501011015

Storage and

Petroleum and

Residential Portable Gas

Refilling at the Pump -



Transport

Petroleum Product
Storage

Cans

Spillage

2501012011

Storage and

Petroleum and

Commercial Portable Gas

Permeation



Transport

Petroleum Product
Storage

Cans



2501012012

Storage and

Petroleum and

Commercial Portable Gas

Evaporation (includes Diurnal



Transport

Petroleum Product
Storage

Cans

losses)

2501012013

Storage and

Petroleum and

Commercial Portable Gas

Spillage During Transport



Transport

Petroleum Product
Storage

Cans



2501012014

Storage and

Petroleum and

Commercial Portable Gas

Refilling at the Pump - Vapor



Transport

Petroleum Product
Storage

Cans

Displacement

2501012015

Storage and

Petroleum and

Commercial Portable Gas

Refilling at the Pump -



Transport

Petroleum Product
Storage

Cans

Spillage

2630020000

Waste Disposal

Treatment and
Recovery

Wastewater Treatment,
Public Owned

Total Processed

2640000000

Waste Disposal

Treatment and
Recovery

TSDFs, All TSDF Types

Total: All Processes

2810025000

Miscellaneous
Area Sources

Other Combustion

Residential Grilling

Total

2810060100

Miscellaneous
Area Sources

Other Combustion

Cremation

Humans

47


-------
2.3

2016 Onroad Mobile sources (onroad)

Onroad mobile source include emissions from motorized vehicles operating on public roadways. These
include passenger cars, motorcycles, minivans, sport-utility vehicles, light-duty trucks, heavy-duty trucks,
and buses. The sources are further divided by the fuel they use, including diesel, gasoline, E-85, and
compressed natural gas (CNG) vehicles. The sector characterizes emissions from parked vehicle
processes (e.g., starts, hot soak, and extended idle) as well as from on-network processes (i.e., from
vehicles as they move along the roads). Except for California, all onroad emissions are generated using
the SMOKE-MOVES emissions modeling framework that leverages MOVES-generated emission factors,
county and SCC-specific activity data, and hourly meteorological data. The onroad source classification
codes (SCCs) in the modeling platform are more finely resolved than those in the National Emissions
Inventory (NEI). The NEI SCCs distinguish vehicles and fuels. The SCCs used in the model platform
also distinguish between emissions processes (i.e., off-network, on-network, and extended idle), and road
types.

Onroad emissions were computed with SMOKE-MOVES by multiplying specific types of vehicle activity
data by the appropriate emission factors. This section includes discussions of the activity data and the
emission factor development. The vehicles (aka source types) for which MOVES computes emissions are
shown in Table 2-17. SMOKE-MOVES was run for specific modeling grids. Emissions for the
contiguous U.S. states and Washington, D.C., were computed for a grid covering those areas. Emissions
for Alaska, Hawaii, Puerto Rico, and the U.S. Virgin Islands were computed by running SMOKE-
MOVES for distinct grids covering each of those regions and are included in the onroad nonconus sector.
In some summary reports these non-CONUS emissions are aggregated with emissions from the onroad
sector.

Table 2-17. MOVES vehicle (source) types

MOYKS vehicle Ivpe

Description

II P.MS vehicle Ivpe

11

Motorcycle

10

21

Passenger Car

25

31

Passenger Truck

25

32

Light Commercial Truck

25

41

Intercity Bus

40

42

Transit Bus

40

43

School Bus

40

51

Refuse Truck

50

52

Single Unit Short-haul Truck

50

53

Single Unit Long-haul Truck

50

54

Motor Home

50

61

Combination Short-haul Truck

60

62

Combination Long-haul Truck

60

Onroad Activity Data Development

SMOKE-MOVES uses vehicle miles traveled (VMT), vehicle population (VPOP), and hours of hoteling,
to calculate emissions. These datasets are collectively known as "activity data". For each of these activity
datasets, first a national dataset was developed; this national dataset is called the "EPA default" dataset.
Second, data submitted by state agencies were incorporated where available, in place of the EPA default
data. EPA default activity was used for California, but the emissions were scaled to California-supplied

48


-------
values during the emissions processing. The agencies for which submitted VMT and VPOP data were
used for 2016 platforms are shown in Table 2-18 along with the timing of the submission: 2014vl or 2016
beta or 2016 vl. Data submitted for the 2014 NEI were adjusted before they were used for 2016
platforms.

Table 2-18. Submitted data used to prepare onroad activity data

Agency

2016 VMT

2016 VPOP

Alaska

yes(2014vl)

yes (2014vl)

Arizona - Maricopa

yes(2014vl)

yes (2014vl)

Arizona - Pima

yes (vl)

yes (vl)

Colorado

yes (beta)

yes (vl)

Connecticut

yes (beta)

yes (2014vl)

Delaware

yes(2014vl)

yes (2014vl)

District of Columbia

yes(2014vl)

yes (2014vl)

Georgia

yes (beta)

yes (beta)

Idaho

yes(2014vl)

yes (2014vl)

Illinois - Chicago area

yes(vl)

yes (vl)

Illinois - rest of state

yes (beta)

yes (2014vl)

Indiana - Louisville area

yes (vl)



Kentucky - Jefferson

yes (vl)

yes (2014vl)

Kentucky - Louisville exurbs

yes (vl)



Maine

yes (2014v2)

yes (2014v2)

Maryland

yes (beta)

yes (beta)

Massachusetts

yes (vl)

yes (vl)

Michigan - Detroit area

yes (beta)

yes (2014vl)

Michigan - rest of state

yes (beta)

yes (2014vl)

Minnesota

yes (beta)

yes (2014vl)

Missouri

yes (2014vl)

yes (2014vl)

Nevada - Clark

yes (beta)

yes (beta)

Nevada - Washoe

yes(2014vl)

yes (2014vl)

New Hampshire

yes (beta)

yes (beta)

New Jersey

yes (beta)

yes (vl)

New Mexico - Bernalillo

yes(2014vl)

yes (2014vl)

New York

yes(2014vl)

yes (2014vl)

North Carolina

yes (beta)

yes (beta)

Ohio

yes(2014vl)

yes (2014vl)

Oregon

yes(2014vl)

yes (2014vl)

Pennsylvania

yes (beta)

yes (beta)

Rhode Island

yes (2014vl)

yes (2014vl)

South Carolina

yes (beta)

yes (beta)

Tennessee - Davidson

yes(2014vl)

yes (2014vl)

Tennessee - Knox

yes(2014vl)

yes (2014vl)

Tennessee - rest of state

yes(2014v2)

yes (2014v2)

Texas

yes (2014vl)

yes (2014vl)

Vermont

yes(2014v2)

yes (2014v2)

Virginia

yes (beta)

yes (2014v2)

49


-------
Agency

2016 VMT

2016 VPOP

Washington

yes (2014v2)

yes (2014v2)

West Virginia

yes (beta)

yes (beta)

Wisconsin

yes (beta)

yes (beta)

Vehicle Miles Traveled (VMT)

EPA calculated default 2016 state VMT by projecting the 2014NEIv2 platform VMT to 2016. The
2014NEIv2 Technical Support Document has details on the development of those VMT
(https://www.epa.gov/air-emissions-inventories/2014-national-emissions-inventorv-nei-technical-support-
document-tsd). The data projected to 2016 were used for states that did not submit 2016 VMT data.
Projection factors to grow state VMT from 2014 to 2016 were based on state-level VMT data from the
Federal Highway Administration (FHWA) VM-2 reports
(https://www.fhwa.dot.gov/policvinformation/statistics/2014/vm2.cfm and

https://www.fhwa.dot.gov/policvinformation/statistics/2016/vm2.cfm). For most states, separate factors
were calculated for urban VMT and rural VMT. Some states have a very different distribution of urban
activity versus rural activity between 2014NEIv2 and the FHWA data, due to inconsistencies in the
definition of urban versus rural. For those states, a single state-wide projection factor based on total
FHWA VMT across all road types was applied to all VMT independent of road type. The following states
used a single state-wide projection factor to adjust the VMT to 2016 levels: AK, GA, IN, ME, MA, NE,
NM, NY, ND, TN, and WV. Also, state-wide projection factors in Texas and Utah were developed from
alternative VMT datasets provided by their respective Departments of Transportation. The VMT
projection factors for all states are provided in Table 2-19.

Table 2-19. Factors applied to project VMT from 2014 to 2016 to prepare default activity data

State

Rural roads

I rhan roads

Projection Kaclor Source

Alabama

5.36%

5.47%

FHWA VM-2 urban/rural

Alaska

8.27%

8.27%

FHWA VM-2 total

Arizona

1.07%

6.35%

FHWA VM-2 urban/rural

Arkansas

4.80%

5.36%

FHWA VM-2 urban/rural

California

1.06%

2.39%

FHWA VM-2 urban/rural

Colorado

5.97%

6.67%

FHWA VM-2 urban/rural

Connecticut

1.33%

1.45%

FHWA VM-2 urban/rural

Delaware

4.42%

6.75%

FHWA VM-2 urban/rural

District of Columbia

0.00%

2.68%

FHWA VM-2 urban/rural

Florida

10.27%

6.64%

FHWA VM-2 urban/rural

Georgia

10.10%

10.10%

FHWA VM-2 total

Hawaii

6.14%

4.21%

FHWA VM-2 urban/rural

Idaho

5.51%

7.80%

FHWA VM-2 urban/rural

Illinois

3.40%

1.96%

FHWA VM-2 urban/rural

Indiana

5.02%

5.02%

FHWA VM-2 total

Iowa

6.17%

6.05%

FHWA VM-2 urban/rural

Kansas

2.42%

6.52%

FHWA VM-2 urban/rural

Kentucky

2.52%

3.26%

FHWA VM-2 urban/rural

50


-------
Slsilc

Uursil roiids

I rhnii roads

Projection Kador Source

Louisiana

-5.49%

7.10%

FHWA VM-2 urban/rural

Maine

3.75%

3.75%

FHWA VM-2 total

Maryland

4.98%

4.75%

FHWA VM-2 urban/rural

Massachusetts

7.42%

7.42%

FHWA VM-2 total

Michigan

5.62%

0.66%

FHWA VM-2 urban/rural

Minnesota

2.66%

2.97%

FHWA VM-2 urban/rural

Mississippi

1.83%

4.96%

FHWA VM-2 urban/rural

Missouri

4.70%

4.17%

FHWA VM-2 urban/rural

Montana

3.32%

4.34%

FHWA VM-2 urban/rural

Nebraska

5.54%

5.54%

FHWA VM-2 total

Nevada

8.30%

5.30%

FHWA VM-2 urban/rural

New Hampshire

5.00%

3.65%

FHWA VM-2 urban/rural

New Jersey

5.41%

2.83%

FHWA VM-2 urban/rural

New Mexico

10.01%

10.01%

FHWA VM-2 total

New York

-4.90%

-4.90%

FHWA VM-2 total

North Carolina

7.47%

8.41%

FHWA VM-2 urban/rural

North Dakota

-7.35%

-7.35%

FHWA VM-2 total

Ohio

4.61%

5.42%

FHWA VM-2 urban/rural

Oklahoma

4.72%

1.23%

FHWA VM-2 urban/rural

Oregon

8.05%

4.84%

FHWA VM-2 urban/rural

Pennsylvania

-4.30%

4.73%

FHWA VM-2 urban/rural

Rhode Island

3.26%

3.26%

FHWA VM-2 urban/rural

South Carolina

9.70%

8.89%

FHWA VM-2 urban/rural

South Dakota

3.23%

2.64%

FHWA VM-2 urban/rural

Tennessee

6.29%

6.29%

FHWA VM-2 total

Texas

7.82%

7.82%

TxDOT3

Utah

11.62%

11.62%

UDOT4

Vermont

5.55%

2.24%

FHWA VM-2 urban/rural

Virginia

-4.93%

9.78%

FHWA VM-2 urban/rural

Washington

6.86%

4.43%

FHWA VM-2 urban/rural

West Virginia

2.21%

2.21%

FHWA VM-2 total

Wisconsin

4.15%

9.32%

FHWA VM-2 urban/rural

Wyoming

-1.38%

-1.53%

FHWA VM-2 urban/rural

Puerto Rico

0.00%

0.00%

No FHWA VM-2 data

Virgin Islands

0.00%

0.00%

No FHWA VM-2 data

For the 2016vl platform, VMT data submitted by state and local agencies were incorporated and used in
place of EPA defaults, as described below. Note that VMT data need to be provided to SMOKE for each
county and SCC. The onroad SCCs characterize vehicles by MOVES fuel type, vehicle (aka source) type,

3	2014: https://ftp.dot.state.tx.us/pub/txdot-info/trf7crash statistics/2014/01 .pdf
2016: https://ftp.dot.state.tx.us/pub/txdot-info/trf/crash statistics/2016/01 .pdf

4	2014: https://www.udot.utah.gov/main/uconowner.gf?n=27035817009129993
2016: https://www.udot.utah.gov/main/uconowner.gf?n=36418522778889648

51


-------
emissions process, and road type. Any VMT provided at a different resolution than this were converted to
a full county-SCC resolution to prepare the data for processing by SMOKE.

Air agencies from CO, CT, GA, IL, MD, NJ, NC, VA, WI, and Pima County (AZ) provided 2016 VMT
data by county and Highway Performance Monitoring Systems (HPMS) vehicle type to be used for the
2016beta and 2016vl platforms. That level of detail is sufficient for MOVES, but SMOKE also needs
VMT broken out by MOVES vehicle type (which is more detailed than HPMS vehicle type), and by fuel
type, and road type. To get VMT at the resolution needed by SMOKE, the county-HPMS VMT data
provided by the states were loaded into the county databases (CDBs) that are used to run MOVES.
MOVES CDBs include fuel type splits, road type splits, and VPOP by MOVES vehicle type. Using those
tables, county-HPMS VMT data were converted into the county-SCC VMT data that are needed by
SMOKE. One exception to the use of local data in these states was for North Carolina, where EPA default
VMT for buses was used along with state-submitted VMT for other vehicle types.

South Carolina and Massachusetts submitted VMT by county-HPMS using the same HPMS splits in
every county in the state. Unlike Massachusetts, South Carolina did not provide county-specific road type
splits. Instead, a new set of county-specific HPMS splits was developed from the EPA default VMT. For
all HPMS types except 25 (light cars and trucks), county-HPMS ratios were calculated from the EPA
default VMT, and then scaled up or down so that the overall state-HPMS ratio would match South
Carolina's state-HPMS ratio. For HPMS type 25, the county-HPMS ratios were set equal to the remainder
within each county so that all ratios within each county sum to 1.0. The new VMT by county-HPMS
varies by county while respecting the state-wide HPMS splits in South Carolina's original VMT dataset.
The VMT was then split to full SCC level using a similar procedure as other states that submitted VMT at
the county-HPMS level.

Pennsylvania and New Hampshire submitted VMT for the 2016beta platform at the full county-SCC
level, already in the FF10 format needed by SMOKE. These data were used directly for the 2016vl
platform, except for the redistribution of light duty VMT (see last item in this subsection).

Michigan and Minnesota submitted 2016 VMT by county and by road type for the 2016beta platform.

Fuel type and vehicle type distributions from the EPA default VMT were used to convert these data to full
SCC.

West Virginia submitted county total VMT only for the 2016beta platform. Fuel, vehicle, and road type
distributions from the EPA default VMT were used to convert their data to full SCC.

For the 2016beta platform, Clark County, NV, submitted VMT by county and MOVES vehicle type,
which is more detailed than HPMS vehicle type, but nevertheless cannot be imported into MOVES CDBs
as easily to facilitate the creation of VMT at the full SCC detail. Fuel type and road type distributions
from the EPA default VMT were used to convert these data to full SCC.

For the 2016vl platform, VMT was provided by:

•	Massachusetts (by HPMS, to override what was provided for beta)

•	Chicago area (8 counties, by HPMS/road; excluded motorcycles)

•	Louisville area (5 counties, county totals restricted/unrestricted)

•	Pima County AZ (by HPMS)

52


-------
Some of the provided data were adjusted following quality assurance, as described below in the VPOP
section.

A final step was performed on all state-submitted VMT. The distinction between a "passenger car"
(MOVES vehicle type 21) versus a "passenger truck" (MOVES vehicle type 31) versus a "light
commercial truck" (MOVES vehicle type 32) is not always consistent between different datasets. This
distinction can have a noticeable effect on the resulting emissions, since MOVES emission factors for
passenger cars are quite different than those for passenger trucks and light commercial trucks.

To ensure consistency in the 21/31/32 splits across the country, all state-submitted VMT for MOVES
vehicle types 21,31, and 32 (all of which are part of HPMS vehicle type 25) was summed, and then re-
split using the 21/31/32 splits from the EPA default VMT. VMT for each source type as a percentage of
total 21/31/32 VMT was calculated by county from the EPA default VMT. Then, state-submitted VMT
for 21/31/32 was summed and then resplit according to those percentages.

This was done for all states and counties listed above which submitted VMT for 2016. Most of the states
listed above did not provide VMT down to the source type, so splitting the light-duty vehicle VMT does
not create an inconsistency with state-provided data in those states. Exceptions are New Hampshire and
Pennsylvania: those two states provided SCC-level VMT, but these were reallocated to 21/31/32 so that
the splits are performed in a consistent way across the country. The 21/31/32 splits in the EPA default
VMT can be traced back to the 2014NEIv2 VPOP data obtained from IHS-Polk.

Speed Activity (SPEED/SPDIST)

In SMOKE 4.7, SMOKE-MOVES was updated to use speed distributions similarly to how they are used
when running MOVES in inventory mode. This new speed distribution file, called SPDIST, specifies the
amount of time spent in each MOVES speed bin for each county, vehicle (aka source) type, road type,
weekday/weekend, and hour of day. This file contains the same information at the same resolution as the
Speed Distribution table used by MOVES but is reformatted for SMOKE. Using the SPDIST file results
in a SMOKE emissions calculation that is more consistent with MOVES than the old hourly speed profile
(SPDPRO) approach, because emission factors from all speed bins can be used, rather than interpolating
between the two bins surrounding the single average speed value for each hour as is done with the
SPDPRO approach.

As was the case with the previous SPDPRO approach, the SPEED inventory that includes a single overall
average speed for each county, SCC, and month, must still be read in by the SMOKE program Smkinven.
SMOKE requires the SPEED dataset to exist even when speed distribution data are available, even though
only the speed distribution data affects the selection of emission factors. The SPEED dataset is carried
over from 2014NEIv2, while the SPDIST dataset is new for the 2016vl platform. Both are based on a
combination of the Coordinating Research Council (CRC) A-100 data and MOVES CDBs.

Vehicle Population (VPOP)

The EPA default VPOP dataset was based on the EPA default VMT dataset described above. For each
county, fuel type, and vehicle type, a VMT/VPOP ratio (miles per vehicle per year) was calculated based
on the 2014NEIv2 VMT and VPOP datasets. That ratio was applied to the 2016 EPA default VMT, to
produce an EPA default VPOP projection.

53


-------
As with VMT, several state and local agencies submitted VPOP data for the beta and vl platforms, and
those data were used in place of the EPA default VPOP. The VPOP SCCs used by SMOKE are similar to
the VMT SCCs, except the emissions process is represented as "00" because it is not relevant to vehicle
population data.

For the 2016 beta platform, GA, MD, MA, NJ, NC, WI, and Pima County AZ provided VPOP data for
the year 2016 by county and MOVES vehicle type. That level of detail is sufficient for MOVES, but
SMOKE also needs VPOP broken out by fuel type. To get VPOP by full SCC, the county-vehicle VPOP
data provided by the states were loaded into the MOVES CDBs. Using fuel type tables in the CDBs, it is
possible to take county-vehicle VPOP data and create county-SCC VPOP data at the resolution needed by
SMOKE. For Massachusetts, based on quality assurance checks, modifications to their VPOP like those
done for their VMT were not needed. Wisconsin provided VPOP for 2016 by county and HPMS vehicle
type instead of by MOVES vehicle type, but the same procedure was applied as for other states in this
group. For North Carolina, EPA default VPOP data were used for buses along with the state-submitted
VPOP for other vehicle types, consistent with the VMT.

West Virginia and Clark County, Nevada also provided VPOP for the 2016 beta platform by county and
MOVES vehicle type. Because they did not provide VMT by county-HPMS, these data were not put into
MOVES databases for splitting. Instead, the VPOP data were split to full SCC using county-vehicle to
county-SCC ratios calculated from the 2016 beta VMT - not the EPA default VMT, but the final VMT
incorporating state data and split to full SCC within MOVES CDBs. So effectively, MOVES CDBs were
used to split their VPOP to full SCC, but only indirectly. West Virginia's VPOP dataset did not include
any intercity buses (MOVES vehicle type 41), thus intercity bus VPOP data were taken from the EPA
default VPOP.

The FFlO-formatted county-SCC VPOP data provided by Pennsylvania and New Hampshire for the 2016
beta platform were used for the 2016vl platform.

EPA default VPOP data were used for the states that submitted VMT but did not submit VPOP (CT, IL,
MI, MN, and VA). The new VMT that South Carolina provided, in addition to the recalculation of HPMS
splits between counties, introduced some issues with VMT/VPOP ratios when comparing the 2016beta
VMT with EPA default beta VPOP. The largest VMT/VPOP ratio issues were for HD vehicles. Because
the light-duty (LD) VPOP data are based on the IHS-Polk registration data, only the heavy-duty (HD)
VPOP data were modified for South Carolina using the EPA defaults. For HD VPOP in South Carolina:
new VPOP = EPA default VPOP * (SC-submitted VMT / EPA default VMT). In other words, the same
changes that were made to the VMT as a result of the new state data were also made to the VPOP on a
percentage basis. This preserves VMT/VPOP ratios for HD vehicles in South Carolina compared to the
EPA default data. This procedure resulted in some changes to the overall HD VPOP total in South
Carolina, both at the county level and state level.

VPOP by source type was not re-split among the LD types 21/31/32. This is consistent with the 2016beta
platform, in which all state-submitted VMT was re-split, but state-submitted VPOP at the source type
level or better was not.

For 2016vl, VPOP data were provided for:

•	Massachusetts (by HPMS)

•	Chicago area (8 counties, by source type)

54


-------
•	Colorado (by source type)

•	New Jersey (by source type)

•	Pima County, AZ (by source type)

The state-submitted VMT and VPOP data underwent several modifications based on quality assurance:
Colorado:

1.	There was a lot of inconsistency between the VMT and VPOP when it was broken down into
individual vehicle types. Colorado indicated that we shouldn't put too much stock into the HPMS-
>vehicle breakdowns in their VPOP data. So, we summed their VPOP to HPMS type and re-split to
vehicle type based on splits from beta VPOP.

2.	Due to concerns about VMT/VPOP ratios for long haul source types (41, 53, 62), we recalculated the
VPOP from VMT using average national VMT/VPOP ratios from 2014v2: 53,000 for 41s; 18,600 for
53s, and 68,000 for 62s. We also recalculated the 52 VPOP as old 52+53 VPOP minus new 53 VPOP.
In one county (08019), 52 VPOP ended up negative, so we increased the 53 VMT/VPOP ratio (which
decreased the VPOP) for that county only.

3.	There were also some VMT/VPOP ratios at the county level for HPMS vehicle types 42, 43, and 61
that were greater than 150,000 miles/year. For these, we increased the VPOP for these county-vehicle
combinations so that the VMT/VPOP ratio would never exceed 150,000. This affected 6 county-
vehicle combinations, mostly with small VPOP.

Chicago area:

1.	Chicago provided separate VMT for HPMS vehicle types 20 and 30, which were summed and re-
split based on 2016beta platform VMT to keep LD vehicle type distributions consistent.

2.	Motorcycles VMT and VPOP were taken from the 2016beta platform.

3.	Based on email communication and number comparison, the provided Chicago area bus VMT
(submitted as total buses), appear to include only data for bust types 41 and 42 only and not 43
(school). So, the bus VMT were allocated to the 4land 42 types and school bus VMT (43) were
carried forward from 2016beta.

4.	For bus VPOP, Chicago did not provide intercity buses, so those were carried forward from
2016beta, but their transit and school bus VPOP values were retained.

5.	The provided 50/60 VPOP appeared to be much too low, so we recalculated it based on their VMT
combined with average VMT/VPOP ratios: 24,000 for 51s; 10,000 for 52s; 18,600 for 53s; 4,000
for 54s; 57,000 for 61s and 68,000 for 62s.

6.	Counties 17063 and 17093 had VPOP for 41/42 but no VMT. We added VMT from the 2016beta
platform for these county-vehicle combinations. The VMT for 41 was carried forward from
2016beta to 2016vl. For 42, the 2016vl VMT = beta VMT * (vl VPOP / beta VPOP).

Pima County: The provided 50/60 VPOP was not based on vehicle registrations, so we recalculated
based on their VMT combined with average VMT/VPOP ratios (as was done for Chicago).

Hotelina Hours (HOTEUNG)

Hoteling hours activity is used to calculate emissions from extended idling and auxiliary power units
(APUs) for heavy duty diesel vehicles. Many states have commented that EPA estimates of hoteling
hours, and therefore emissions resulting from hoteling are higher than they could realistically be in reality

55


-------
given the available parking spaces. Therefore, recent hoteling activity datasets, including the 2014NEIv2,
2016 beta, and 2016vl platforms, incorporate reductions to hoteling activity data based on the availability
of truck stop parking spaces in each county, as described below. For 2016vl, hoteling hours were
recomputed using a new factor identified by EPA's Office of Transportation and Air Quality as more
appropriate based on recent studies.

The method used in 2016vl is the following:

1	Start with 2016vl VMT for 62 on restricted roads, by county.

2	Multiply that by 0.007248 hours/mile (Sonntag, 2018). This results in about 73.5% less
hoteling hours as compared to the 2014v2 approach.

3	Apply parking space reductions as has been done for 2016beta, except for states that
requested we not do that (CO, ME, NJ, NY).

Hoteling hours were adjusted down in counties for which there were more hoteling hours assigned to the
county than could be supported by the known parking spaces. To compute the adjustment, we started
with the hoteling hours for the county as computed by the above method, and then we applied reductions
directly to the 2016 hoteling hours based on known parking space availability so that there were not more
hours assigned to the county than the available parking spaces could support if they were full every hour
of every day.

A dataset of truck stop parking space availability with the total number of parking spaces per county was
used in the computation of the adjustment factors. This same dataset is used to develop the spatial
surrogate for hoteling emissions. For the 2016vl platform, the parking space dataset includes several
updates compared to 2016beta platform, based on information provided by some states (e.g., MD). Since
there are 8,784 hours in the year 2016; the maximum number of possible hoteling hours in a particular
county is equal to 8,784 * the number of parking spaces in that county. Hoteling hours for each county
were capped at that theoretical maximum value for 2016 in that county, with some exceptions as outlined
below.

Because the truck stop parking space dataset may be incomplete in some areas, and trucks may sometimes
idle in areas other than designated spaces, it was assumed that every county has at least 12 parking spaces,
even if fewer parking spaces are found in the parking space dataset. Therefore, hoteling hours are never
reduced below 105,408 hours for the year in any county. If the unreduced hoteling hours were already
below that maximum, the hours were left unchanged; in other words, hoteling activity are never increased
as a result of this analysis.

A handful of high activity counties that would otherwise be subject to a large reduction were analyzed
individually to see if their parking space count seemed unreasonably low. In the following counties, the
parking space count and/or the reduction factor was manually adjusted:

•	17043 / DuPage IL (instead of reducing hoteling by 89%, applied no adjustment)

•	39061 / Hamilton OH (parking spot count increased to 20 instead of the minimum 12)

•	47147 / Robertson TN (parking spot count increased to 52 instead of just 26)

•	51015 / Augusta VA (parking space count increased to 48 instead of the minimum 12)

•	51059 / Fairfax VA (parking spot count increased to 20 instead of the minimum 12)

56


-------
Georgia and New Jersey submitted hoteling activity for the 2016vl platform. For these states, the EPA
default projection was replaced with their state data. New Jersey provided their hoteling activity in a
series of HotellingHours MOVES-formatted tables, which include separate activity for weekdays and
weekends and for each month and which have units of hours-per-week. These data first needed to be
converted to annual totals by county.

For Georgia we were going to bring forward their beta HOTELING but found it was now much too large
compared to other states once the new hoteling factor was implemented. After discussion with Georgia
Department of Natural Resources staff, we agreed to recalculate from VMT for all counties except for
those where parking > 0 and restricted VMT = 0. In those counties, Georgia's 2016beta hoteling were
reduced by 73.5% (the same reduction factor applied to the rest of the country).

Alaska Department of Natural Resources staff requested that we zero out hoteling activity in several
counties due to the nature of driving patterns in their region. In addition, there are no hoteling hours or
other emissions from long-haul combination trucks in Hawaii, Puerto Rico, or the Virgin Islands.

All parking space counts are the same as 2016beta except Maryland, which submitted an update for
2016vl.

The states of Colorado, Maine, New Jersey, and New York requested that no reductions be applied to the
hoteling activity based on parking space availability. For these states, we did not apply any reductions
based on parking space availability and left the hours that were computed using the updated method for
2016vl; or in the case of New Jersey, their submitted activity; unchanged. Otherwise, the submitted data
from New Jersey would have been subject to reductions. The submitted data from Georgia did not exceed
the maximum value in any county, so their submitted data did not need to be reduced.

Finally, the county total hoteling must be split into separate values for extended idling (SCC 2202620153)
and APUs (SCC 2202620191). New Jersey's submittal of hoteling activity specified a 30% APU split,
and this was used for all New Jersey counties. For the rest of the country, a 12.4% APU split was used for
the year 2016, meaning that APUs are used for 12.4% of the hoteling hours.

Onroad Emission Factor Table Development

MOVES2014b was run in emission rate mode to create emission factor tables using CB6 speciation for
the years 2016, 2020, 2023, and 2028, for all representative counties and fuel months. MOVES was run
for all counties in Alaska, Hawaii, and Virgin Islands, and for a single representative county in Puerto
Rico. The county databases (CDBs) used to run MOVES to develop the emission factor tables were
updated from those used in the 2016beta platform.

Age distributions are a key input to MOVES in determining emission rates. The age distributions for
2016vl were updated based on vehicle registration data obtained from the CRC A-l 15 project, subject to
reductions for older vehicles determined according to CRC A-l 15 methods but using additional age
distribution data that became available as part of the 2017 NEI submitted input data. One of the findings
of CRC project A-l 15 is that IHS data contain higher vehicle populations than state agency analyses of
the same Department of Motor Vehicles data, and the discrepancies tend to increase with increasing
vehicle age (i.e., there are more older vehicles in the IHS data). The CRC project dealt with the
discrepancy by releasing datasets based on raw (unadjusted) information and adjusted sets of age

57


-------
distributions, where the adjustments reflected the differences in population by model year of 2014 IHS
data and 2014 submitted data from a single state.

For the 2016 platform and 2017 NEI, EPA repeated the CRC's assessment of IHS vs. state discrepancies
but with updated 2017 information and for more states. The 2017 light-duty vehicle (LDV) populations
from the CRC A-l 15 project were compared by model year to the populations submitted by state/local
(S/L) agencies for the 2017 NEI. The comparisons by model year were used to develop adjustment
factors that remove older age LDVs from the IHS dataset. Out of 31 S/L agencies that provided data, 16
provided LDV population and age distributions with snapshot dates of January 2017, July 2017, or 2018.
The other 15 had either unknown or older (back to 2013) data pull dates, so were not a fair comparison to
the 2017 IHS data.

We reviewed the population by model year comparisons for each of the 16 geographic areas vs. IHS
separately for source type 21 and for source type 31 plus 32 together. We reallocated the S/L agency
populations of cars (source type 21) and light trucks (source types 31 and 32) to match IHS car and light-
duty truck splits by county for consistent VIN decoding. We also removed the state of Georgia from the
pool of S/L agencies used to calculate the adjustment factors to avoid its influence on a pooled geographic
adjustment. Georgia already works closely with IHS on VIN decoding, and as a result, their submittal
matched IHS. The IHS data are higher than the pooled state data by 6.5 percent for cars and 5.9 percent
for light trucks.

We calculated the vehicle age distribution adjustment factors as one minus the fraction of vehicles to
remove from IHS to equal the state data, with two exceptions. The model year range 2006/2007 to 2017
receives no adjustment and the model year 1987 receives a capped adjustment that equals the adjustment
to 1988. Table 2-20 below shows the fraction of vehicles to keep by model year based on this analysis.
The adjustments were applied to the 2016 IHS-based age distributions from CRC project A-l 15 prior to
use in 2016vl. In addition, we removed the county-specific fractions of antique license plate vehicles
present in the registration summary from IHS. Nationally, the prevalence of antique plates is only 0.8
percent, but as high as 6 percent in some states (e.g., Mississippi).

Table 2-20. Older Vehicle Adjustments Showing the Fraction of IHS Vehicle Populations to Retain

for 2016vl and 2017 NEI

Model Year

(illS

l.iiihl

pre-1989

0.675

0.769

1989

0.730

0.801

1990

0.732

0.839

1991

0.740

0.868

1992

0.742

0.867

1993

0.763

0.867

1994

0.787

0.842

1995

0.776

0.865

1996

0.790

0.881

1997

0.808

0.871

1998

0.819

0.870

1999

0.840

0.874

2000

0.838

0.896

2001

0.839

0.925

2002

0.864

0.921

2003

0.887

0.942

58


-------
Model Year

(illS

l.ilihl

2004

0.926

0.953

2005

0.941

0.966

2006

1

0.987

2007-2017

1

1

In addition to removing the older and antique plate vehicles from the IHS data, we accounted for 25
counties that were outliers because their fleet age was significantly younger than typical. We limited our
outlier identification to LDV source types 21,31, and 32, because they're the most important. Many rural
counties also have outliers for low-population source types such as Transit Bus and Refuse Truck; these
do not have much of an impact on the inventory overall and reflect sparse data in low-population areas
and therefore do not require correction.

The most extreme examples of LDV outliers were Light Commercial Truck age distributions where over
50 percent of the population in the entire county is 0 and 1 years old. These sorts of young fleets can
happen if the headquarters of a leasing or rental company is the owner/entity of a relatively large number
of vehicles relative to the county-wide population. While the business owner of thousands of new
vehicles may reside in a single county, the vehicles likely operate in broader areas without being
registered where they drive. To avoid creating artificial low spots of LDV emissions in these outlier
counties, we flagged all counties above a 0.35 fraction of new vehicles and excluded their age distribution
from the final set of grouped age distributions that went into the 2016vl CDBs.

The 2016 age distributions were then grouped using a population-weighted average of the source type
populations of each county in the representative county group. The end-product was age distributions for
each of the 13 source types in each of the 315 representative counties for 2016vl. It should be noted that
the long-haul truck source types 53 (Single Unit) and 62 (Combination Unit) are a nationwide average due
to the long-haul nature of their operation.

Input data tables provided by states were reviewed before they were used. Some submitted data tables
were found to be from previous emissions modeling platforms, primarily NEI 2014v2, 2016 alpha, or
2016 beta, and these were not explicitly used as most were already incorporated into the CDBs. All
average speed distributions in 2016vl came from the CRC A-100 study, and most age distributions (other
than accepted submittals for New Jersey, Pima County, Arizona, and Wisconsin) came from methods
described above for 2016 vl. The following submitted MOVES input data (other than the activity data
discussed above) were incorporated into the 2016vl base year MOVES CDBs:

•	Chicago (IL) Metropolitan Agency for Planning: FF10 VMT, FF10 VPOP, Month/Day VMT
Fraction, Ramp Fractions

•	Georgia Department of Natural Resources: Fuel Supply (county assignments to fuel type groups)

•	Louisville (KY) Metro Air Pollution Control District: Road Type Distributions, Ramp Fractions

•	Maryland Department of the Environment: Truck Stop Locations (these affect the spatial
surrogate but not the MOVES run)

•	New Jersey Department of Environmental Protection: Age Distribution

•	Pima (AZ) Association of Governments: Age Distribution, I/M Coverage, Day VMT Fraction,
Road Type Distribution

•	Wisconsin Department of Natural Resources: Age Distribution, I/M Coverage

59


-------
Once the input data were incorporated into the CDBs, a new set of representative counties was developed.
Each county in the continental U.S. was classified according to its state, altitude (high or low), fuel
region, the presence of inspection and maintenance programs, the mean light-duty age, and the fracti on of
ramps. A binning algorithm was executed to identify "like counties", and then specific requests for
representative county groups by states were honored from the states of Maryland, New York, New J ersey,
Wisconsin, Michigan, and Georgia. The final result was 315 representative counties (up from 304 in
2016 beta) as shown in Figure 2-3. The representative counties themselves changed substantially; of the
315 representative counties, 145 were not representative counties in 2016 beta. The CDBs for these 145
counties were developed from the 2014NEIv2 counties and updated to represent the year 2016. For more
information on the development of the 2016 age distributions and representative counties and the review
of the input data, see the memoranda "Onroad 2016vl documentation_20191007" and
"RepCountiesFor2016v 1 -2017_13jun2019" (ERG, 2019).

Figure 2-3. Representative Counties in 2016vl

Reference County Groups 2016 V1

To create the 2016vl emission factors, MOVES was run separately for each representative county and
fuel month for each temperature bin needed for calendar year 2016. The CDBs used to run MOVES
include the state-specific control measures such as the California low emission vehicle (LEV) program,
except that fuels were updated to represent calendar year 2016. In addition, the range of temperatures rnn

60


-------
along with the average humidities used were specific to the year 2016. The MOVES results were post-
processed into CSV-formatted emission factor tables that can be read by SMOKE-MOVES.

Onroad California Inventory Development

The California Air Resources Board (CARB) provided their own onroad emissions inventories based on
their EMFAC2017 model. EMFAC2017 was run by CARB for model years 2016, 2023, 2028, and 2035.
Details on how SMOKE-MOVES emissions were adjusted to match the CARB-based 2016 inventory are
provided in the Emissions Processing Requirements section of this document.

2.4 2016 Nonroad Mobile sources (cmv, rail, nonroad)

The nonroad mobile source emission modeling sectors consist of nonroad equipment emissions (nonroad),
locomotive (rail) and CMV emissions.

2.4.1 Category 1, Category 2 Commercial Marine Vessels (cmv_c1c2)

The cmv_clc2 inventory sector contains small to medium-size engine CMV emissions. Category 1 and
Category 2 (C1C2) marine diesel engines typically range in size from about 700 to 11,000 hp. These
engines are used to provide propulsion power on many kinds of vessels including tugboats, towboats,
supply vessels, fishing vessels, and other commercial vessels in and around ports. They are also used as
stand-alone generators for auxiliary electrical power on many types of vessels. Category 1 represents
engines up to 7 liters per cylinder displacement. Category 2 includes engines from 7 to 30 liters per
cylinder.

The cmv_clc2 inventory sector contains sources that traverse state and federal waters that are in the
2017NEI along with emissions from surrounding areas of Canada, Mexico, and international waters. The
cmv_clc2 sources are modeled as point sources but using plume rise parameters that cause the emissions
to be released in the ground layer of the air quality model.

The cmv_clc2 sources within state waters are identified in the inventory with the Federal Information
Processing Standard (FIPS) county code for the state and county in which the vessel is registered. The
cmv_clc2 sources that operate outside of state waters but within the Emissions Control Area (ECA) are
encoded with a state FIPS code of 85. The ECA areas include parts of the Gulf of Mexico, and parts of
the Atlantic and Pacific coasts. The cmv_clc2 sources in the 2016vl inventory are categorized as
operating either in-port or underway and as main and auxiliary engines are encoded using the SCCs listed
in Table 2-21.

Table 2-21. 2016vl platform SCCs for cmv_clc2 sector

sec

Tier 1 Description

Tier 2 Description

Tier 3 Description

Tier 4 Description

2280002101

C1/C2

Diesel

Port

Main

2280002102

C1/C2

Diesel

Port

Auxiliary

2280002201

C1/C2

Diesel

Underway

Main

2280002202

C1/C2

Diesel

Underway

Auxiliary

61


-------
Category 1 and 2 CMV emissions were developed for the 2017 NEI,5 The 2017 NEI emissions were
developed based signals from Automated Identification System (AIS) transmitters. AIS is a tracking
system used by vessels to enhance navigation and avoid collision with other AIS transmitting vessels.
The USEPA Office of Transportation and Air Quality received AIS data from the U.S. Coast Guard
(USCG) in order to quantify all ship activity which occurred between January 1 and December 31, 2017.
The provided AIS data extends beyond 200 nautical miles from the U.S. coast (Figure 2-4). This
boundary is roughly equivalent to the border of the US Exclusive Economic Zone and the North
American EC A, although some non-EC A activity are captured as well

Figure 2-4. 2017NEl/2016 platform geographical extent (solid) and U.S. ECA (dashed)

The AIS data were compiled into five-minute intervals by the USCG, providing a reasonably refined
assessment of a vessel's movement. For example, using a five-minute average, a vessel traveling at 25
knots would be captured every two nautical miles that the vessel travels. For slower moving vessels, the
distance between transmissions would be less. The ability to track vessel movements through AIS data
and link them to attribute data, has allowed for the development of an inventory of very accurate emission
estimates. These AIS data were used to define the locations of indivi dual vessel movements, estimate
hours of operation, and quantify propulsion engine loads. The compiled AIS data also included the
vessel's International Marine Organization (IMO) number and Maritime Mobile Service Identifier
(MMSI); which allowed each vessel to be matched to their characteristics obtained from the Clarksons
ship registry (Clarksons, 2018).

USEPA used the engine bore and stroke data to calculate cylinder volume. Any vessel that had a
calculated cylinder volume greater than 30 liters was incorporated into the USEPA's new Category 3
Commercial Marine Vessel (C3CMV) model. The remaining records were assumed to represent Category
1 and 2 (C1C2) or non-ship activity. The C1C2 AIS data were quality assured including the removal of
duplicate messages, signals from pleasure craft, and signals that were not from CMV vessels (e.g., buoys,

5 Category 1 and 2 Commercial Marine Vessel 2017 Emissions Inventory (ERG, 2019).

62


-------
helicopters, and vessels that are not self-propelled). Following this, there were 422 million records
remaining.

The emissions were calculated for each time interval between consecutive AIS messages for each vessel
and allocated to the location of the message following to the interval. Emissions were calculated
according to Equation 2-1.

g

Emissionsintervai = Time (hr)intervai x Power(kW) x £"F( ) x LLAF	Equation 2-1

Power is calculated for the propulsive (main), auxiliary, and auxiliary boiler engines for each interval and
emission factor (EF) reflects the assigned emission factors for each engine, as described below. LLAF
represents the low load adjustment factor, a unitless factor which reflects increasing propulsive emissions
during low load operations. Time indicates the activity duration time between consecutive intervals.

Next, vessels were identified in order determine their vessel type, and thus their vessel group, power
rating, and engine tier information which are required for the emissions calculations. See the 2017 NEI
documentation for more details on this process. Following the identification, 108 different vessel types
were matched to the C1C2 vessels. Vessel attribute data was not available for all these vessel types, so the
vessel types were aggregated into 13 different vessel groups for which surrogate data were available as
shown in Table 2-22. 11,302 vessels were directly identified by their ship and cargo number. The
remaining group of miscellaneous ships represent 13 percent of the AIS vessels (excluding recreational
vessels) for which a specific vessel type could not be assigned.

Table 2-22. Vessel groups in the cmv_clc2 sector

Vessel Group

NEI Area Ship Count

Bulk Carrier

37

Commercial Fishing

1,147

Container Ship

7

Ferry Excursion

441

General Cargo

1,498

Government

1,338

Miscellaneous

1,475

Offshore support

1,149

Reefer

13

Ro Ro

26

Tanker

100

Tug

3,994

Work Boat

77

Total in Inventory:

11,302

As shown in Equation 2-1, power is an important component of the emissions computation. Vessel-
specific installed propulsive power ratings and service speeds were pulled from Clarkson's ship registry

63


-------
and adopted from the Global Fishing Watch (GFW) dataset when available. However, there is limited
vessel specific attribute data for most of the C1C2 fleet. This necessitated the use of surrogate engine
power and load factors, which were computed for each vessel group shown in Table 2. In addition to the
power required by propulsive engines, power needs for auxiliary engines were also computed for each
vessel group. Emissions from main and auxiliary engines are inventoried with different SCCs as shown
in Table 2-21.

The final components of the emissions computation equation are the emission factors and the low load
adjustment factor. The emission factors used in this inventory take into consideration the EPA's marine
vessel fuel regulations as well as exhaust standards that are based on the year that the vessel was
manufactured to determine the appropriate regulatory tier. Emission factors in g/kWhr by tier for NOx,
PMio, PM2.5, CO, CO2, SO2 and VOC were developed using Tables 3-7 through 3-10 in USEPA's (2008)
Regulatory Impact Analysis on engines less than 30 liters per cylinder. To compile these emissions
factors, population-weighted average emission factor were calculated per tier based on C1C2 population
distributions grouped by engine displacement. Boiler emission factors were obtained from an earlier Entec
study (Entec, 2004). If the year of manufacture was unknown then it was assumed that the vessel was
Tier 0, such that actual emissions may be less than those estimated in this inventory. Without more
specific data, the magnitude of this emissions difference cannot be estimated.

Propulsive emissions from low-load operations were adjusted to account for elevated emission rates
associated with activities outside the engines' optimal operating range. The emission factor adjustments
were applied by load and pollutant, based on the data compiled for the Port Everglades 2015 Emission
Inventory.6 Hazardous air pollutants and ammonia were added to the inventory according to
multiplicative factors applied either to VOC or PM2.5.

For more information on the emission computations for 2017, see the supporting documentation for the
2017 NEI C1C2 CMV emissions. The emissions from the 2017 NEI were adjusted to represent 2016 in
the cmv_clc2 sector using factors derived from U.S. Army Corps of Engineers national vessel Entrance
and Clearance data7 by applying a factor of 0.98 to all pollutants. For consistency, the same methods were
used for California, Canadian, and other non-U.S. emissions.

2.4.2 Category 3 Commercial Marine Vessels (cmv_c3)

The cmv_c3 inventory is brand new for the 2016vl platform. It was developed in conjunction with the
CMV inventory for the 2017 NEI. This sector contains large engine CMV emissions. Category 3 (C3)
marine diesel engines are those at or above 30 liters per cylinder, typically these are the largest engines
rated at 3,000 to 100,000 hp. C3 engines are typically used for propulsion on ocean-going vessels
including container ships, oil tankers, bulk carriers, and cruise ships. Emissions control technologies for

6	USEPA. EPA and Port Everglades Partnership: Emission Inventories and Reduction Strategies. US Environmental
Protection Agency, Office of Transportation and Air Quality, June 2018.
https://nepis.epa.gov/Exe/ZyPDF.cgi?Dockey=P100UKV8.pdf.

7	U.S. Army Corps of Engineers [USACE], Foreign Waterborne Transportation: Foreign Cargo Inbound and Outbound
Vessel Entrances and Clearances. US Army Corps of Engineers, 2018.

64


-------
C3 CMV sources are limited due to the nature of the residual fuel used by these vessels.8 The cmv_c3
sector contains sources that traverse state and federal waters; along with sources in waters not covered by
the NEI in surrounding areas of Canada, Mexico, and international waters.

The cmv_c3 sources that operate outside of state waters but within the federal Emissions Control Area
(ECA) are encoded with a FIPS state code of 85, with the "county code" digits representing broad regions
such as the Atlantic, Gulf of Mexico, and Pacific. The ECA areas include parts of the Gulf of Mexico,
and parts of the Atlantic and Pacific coasts. CMV C3 sources around Puerto Rico, Hawaii and Alaska,
which are outside the ECA areas, are included in the 2016vl inventory but are in separate files from the
emissions around the continental United States (CONUS). The cmv_c3 sources in the 2016vl inventory
are categorized as operating either in-port or underway and are encoded using the SCCs listed in Table
2-23 and distinguish between diesel and residual fuel, in port areas versus underway, and main and
auxiliary engines. In addition to C3 sources in state and federal waters, the cmv_c3 sector includes
emissions in waters not covered by the NEI (FIPS = 98) and taken from the "ECA-IMO-based" C3 CMV
inventory.9 The ECA-IMO inventory is also used for allocating the FlPS-level emissions to geographic
locations for regions within the domain not covered by the AIS selection boxes as described in the next
section.

Table 2-23. 2016vl platform SCCs for cmv_c3 sector

see

Tier 1 Description

l ii'i-2 Description

Tier 3 Dcscriplion

Tier 4 Dcscriplion

2280002103

C3

Diesel

Port

Main

2280002104

C3

Diesel

Port

Auxiliary

2280002203

C3

Diesel

Underway

Main

2280002204

C3

Diesel

Underway

Auxiliary

2280003103

C3

Residual

Port

Main

2280003104

C3

Residual

Port

Auxiliary

2280003203

C3

Residual

Underway

Main

2280003204

C3

Residual

Underway

Auxiliary

Prior to creation of the 2017 NEI, "The EPA received Automated Identification System (AIS) data from
United States Coast Guard (USCG) in order to quantify all ship activity which occurred between January
1 and December 31, 2017. The International Maritime Organization's (IMO's) International Convention
for the Safety of Life at Sea (SOLAS) requires AIS to be fitted aboard all international voyaging ships
with gross tonnage of 300 or more, and all passenger ships regardless of size (IMO, 2002). In addition,
the USCG has mandated that all commercial marine vessels continuously transmit AIS signals while
transiting U.S. navigable waters. As the vast majority of C3 vessels meet these requirements, any omitted
from the inventory due to lack of AIS adoption are deemed to have a negligible impact on national C3
emissions estimates. The activity described by this inventory reflects ship operations within 200 nautical
miles of the official U.S. baseline. This boundary is roughly equivalent to the border of the U.S Exclusive
Economic Zone and the North American ECA, although some non-ECA activity is captured as well
(Figure 2-4).

8	https://www.epa.gov/regulations-emissions-vehicles-and-engines/regulations-emissions-marine-vessels

9	https://www.epa.gOv/sites/production/files/2017-08/documents/2014v7.0 2014 emismod tsdvl.pdf

65


-------
The 2017 NEI data were computed based on the AIS data from the USGS for the year of 2017. The AIS
data were coupled with ship registry data that contained engine parameters, vessel power parameters, and
other factors such as tonnage and year of manufacture which helped to separate the C3 vessels from the
C1C2 vessels. Where specific ship parameters were not available, they were gap-filled. The types of
vessels that remain in the C3 data set include: bulk carrier, chemical tanker, liquified gas tanker, oil
tanker, other tanker, container ship, cruise, ferry, general cargo, fishing, refrigerated vessel, roll-on/roll-
off, tug, and yacht.

Prior to use, the AIS data were reviewed - data deemed to be erroneous were removed, and data found to
be at intervals greater than 5 minutes were interpolated to ensure that each ship had data every five
minutes. The five-minute average data provide a reasonably refined assessment of a vessel's movement.
For example, using a five-minute average, a vessel traveling at 25 knots would be captured every two
nautical miles that the vessel travels. For slower moving vessels, the distance between transmissions
would be less.

The emissions were calculated for each C3 vessel in the dataset for each 5-minute time range and
allocated to the location of the message following to the interval. Emissions were calculated according to
Equation 2-2.

g

Emissionsintervai = Time (hr)intervai x Power(kW) x EF{x LLAF	Equation 2-2

Power is calculated for the propulsive (main), auxiliary, and auxiliary boiler engines for each interval and
emission factor (EF) reflects the assigned emission factors for each engine, as described below. LLAF
represents the low load adjustment factor, a unitless factor which reflects increasing propulsive emissions
during low load operations. Time indicates the activity duration time between consecutive intervals.

Emissions were computed according to a computed power need (kW) multiplied by the time (hr) and by
an engine-specific emission factor (g/kWh) and finally by a low load adjustment factor that reflects
increasing propulsive emissions during low load operations.

The resulting emissions were available at 5-minute intervals. Code was developed to aggregate these
emissions to modeling grid cells and up to hourly levels so that the emissions data could be input to
SMOKE for emissions modeling with SMOKE. Within SMOKE, the data were speciated into the
pollutants needed by the air quality model,10 but since the data were already in the form of point sources
at the center of each grid cell, and they were already hourly, no other processing was needed within
SMOKE. SMOKE requires an annual inventory file to go along with the hourly data, so those files were
also generated for each year.

10 Ammonia (NH3) was also added by SMOKE in the speciation step.

66


-------
On January 1st, 2015, the EC A initiated a fuel sulfur standard which regulated large marine vessels to use
fuel with 1,000 ppm sulfur or less. These standards are reflected in the cmv_c3 inventories.

There were some areas needed for modeling that the AIS request boxes did not cover (see Figure 2-4).
These include a portion of the St. Lawrence Seaway transit to the Great Lakes, a small portion of the
Pacific Ocean far offshore of Washington State, portions of the southern Pacific Ocean around off the
coast of Mexico, and the southern portion of the Gulf of Mexico that is within the 36-km domain used for
air quality modeling. In addition, a determination had to be made regarding whether to use the existing
Canadian CMV inventory or the more detailed AlS-based inventory. In 2016vl, the AlS-based inventory
was used in the areas for which data were available, and the areas not covered were gap-filled with
inventory data from the 2016beta platform, which included data from Environment Canada and the 2011
ECA-IMO C3 inventory.

For the gap-filled areas not covered by AIS selections or the Environment Canada inventory, the 2016beta
nonpoint C3 inventory was converted to a point inventory to support plume rise calculations for C3
vessels. The nonpoint emissions were allocated to point sources using a multi-step allocation process
because not all of the inventory components had a complete set of county-SCC combinations. In the first
step, the county-SCC sources from the nonpoint file were matched to the county-SCC points in the 2011
ECA-IMO C3 inventory. The ECA-IMO inventory contains multiple point locations for each county-
SCC. The nonpoint emissions were allocated to those points using the PM2.5 emissions at each point as a
weighting factor.

Cmv_c3 underway emissions that did not have a matching FIPS in the ECA-IMO inventory were
allocated using the 12 km 2014 offshore shipping activity spatial surrogate (surrogate code 806). Each
county with underway emissions in the area inventory was allocated to the centroids of the cells
associated with the respective county in the surrogate. The emissions were allocated using the weighting
factors in the surrogate.

The resulting point emissions centered on each grid cell were converted to an annual point 2010 flat file
format (FF10). Pictures of the emissions are shown in Section 7 of this document. A set of standard stack
parameters were assigned to each release point in the cmv_c3 inventory. The assigned stack height was
65.62 ft, the stack diameter was 2.625 ft, the stack temperature was 539.6 °F, and the velocity was 82.02
ft/s. Emissions were computed for each grid cell needed for modeling.

Adjustment of the 2017 NEI CMV C3 to 2016

Because the NEI emissions data were for 2017, an analysis was performed of 2016 versus 2017 entrance
and clearance data (ERG, 2019a). Annual, monthly, and daily level data were reviewed. Annual ratios of
entrance and clearance activity were developed for each ship type as shown in Table 2-24. For vessel
types with low populations (C3 Yacht, tug, barge, and fishing vessels), an annual ratio of 0.98 was
applied.

67


-------
Table 2-24. 2017 to 2016 projection factors for C3 CMV

Ship Type

Annual Ratio"

Barge

1.551

Bulk Carrier

1.067

Chemical Tanker

1.031

Container Ship

1.0345

Cruise

1.008

Ferry Ro Pax

1.429

General Cargo

0.888

Liquified Gas Tanker

1.192

Miscellaneous Fishing

0.932

Miscellaneous Other

1.015

Offshore

0.860

Oil Tanker

1.101

Other Tanker

1.037

Reefer

0.868

Ro Ro

1.007

Service Tug

1.074

a Above ratios are applied to the 2017 emission values to estimate 2016 values

The cmv_c3 projection factors were pollutant-specific and region-specific. Most states are mapped to a
single region with a few exceptions. Pennsylvania and New York were split between the East Coast and
Great Lakes, Florida was split between the Gulf Coast and East Coast, and Alaska was split between
Alaska East and Alaska West. The non-federal factors listed in this table were applied to sources outside
of U.S. federal waters (FIPS 98). Volatile Organic Compound (VOC) Hazardous Air Pollutant (HAP)
emissions were projected using the VOC factors. NH3 emissions were held constant at 2014 levels.

2.4.3 Rail Sources (rail)

The rail sector includes all locomotives in the NEI nonpoint data category. The 2016vl inventory SCCs
are shown in Table 2-25. This sector excludes railway maintenance activities. Railway maintenance
emissions are included in the nonroad sector. The point source yard locomotives are included in the
ptnonipm sector. In 2014NEIv2, rail yard locomotive emissions were present in both the nonpoint (rail
sector) and point (ptnonipm sector) inventories. For the 2016vl platform, rail yard locomotive emissions
are only in the point inventory / ptnonipm sector. Therefore, SCC 2285002010 is not present in the
2016vl platform rail sector, except in three California counties. The California Air Resources Board
(CARB) submitted rail emissions, including rail yards, for 2016vl platform. In three counties, CARB's
rail yard emissions could not be mapped to point source rail yards, and so those counties' emissions were
included in the rail sector.

68


-------
Table 2-25. 2016vl SCCs for the Rail Sector

sec

Sector

Description: Mobile Sources prefix for all

2285002006

rail

Railroad Equipment; Diesel; Line Haul Locomotives: Class I Operations

2285002007

rail

Railroad Equipment; Diesel; Line Haul Locomotives: Class II / III Operations

2285002008

rail

Railroad Equipment; Diesel; Line Haul Locomotives: Passenger Trains
(Amtrak)

2285002009

rail

Railroad Equipment; Diesel; Line Haul Locomotives: Commuter Lines

2285002010

rail

Railroad Equipment; Diesel; Yard Locomotives (nonpoint)

28500201

rail

Railroad Equipment; Diesel; Yard Locomotives (point)

Class I Line-haul Methodology

In 2008 air quality planners in the eastern US formed the Eastern Technical Advisory Committee
(ERTAC) for solving persistent emissions inventory issues. This work is the fourth inventory created by
the ERTAC rail group. For the 2016 inventory, the Class I railroads granted ERTAC Rail permission to
use the confidential link-level line-haul activity GIS data layer maintained by the Federal Railroad
Administration (FRA). In addition, the Association of American Railroads (AAR) provided national
emission tier fleet mix information. This allowed ERTAC Rail to calculate weighted emission factors for
each pollutant based on the percentage of the Class I line-haul locomotives in each USEPA Tier level
category. These two datasets, along with 2016 Class I line-haul fuel use data reported to the Surface
Transportation Board (Table 2-26), were used to create a link-level Class I emissions inventory, based on
a methodology recommended by Sierra Research. Rail Fuel Consumption Index (RFCI) is a measure of
fuel use per ton mile of freight. This link-level inventory is nationwide in extent, but it can be aggregated
at either the state or county level.

Table 2-26. Class I Railroad Reported Locomotive Fuel Use Statistics for 2016

Class 1 Railroads

2016 U-l Reported Locomotive
Kuel I se («al/vcar)

UK 1
(ton-ill ilcs/gal)

Adjusted

urc i

(lon-niilcs/gal)

Line-Maul"

Switcher

BNSF

1,243,366,255

40,279,454

972

904

Canadian National

102,019,995

6,570,898

1,164

1,081

Canadian Pacific

56,163,697

1,311,135

1,123

1,445

CSX Transportation

404,147,932

39,364,896

1,072

1,044

Kansas City
Southern

60,634,689

3,211,538

989

995

Norfolk Southern

437,110,632

28,595,955

920

906

Union Pacific

900,151,933

85,057,080

1,042

1,095

Totals:

3,203,595,133

204,390,956

1,006

993

* Includes work trains; Adjusted RFCI values calculated from FRA gross ton-mile data as described on page 7. RFCI total is ton-mile weighted mean.

Annual default emission factors for locomotives based on operating patterns ("duty cycles") and the
estimated nationwide fleet mixes for both switcher and line-haul locomotives are available. However,
Tier level fleet mixes vary significantly between the Class I and Class II/III railroads. As can be seen in
Figure 2-5 and Figure 2-6, Class I railroad activity is highly regionalized in nature and is subject to

69


-------
variations in terrain across the country which can have a significant impact on fuel efficiency and overall
fuel consumption.

Figure 2-5. 2016 US Railroad Traffic Density in Millions of Gross Tons per Route Mile (MGT)

Traffic Density

0.02 - 4.99 MGT
5.00 - 9.99 MGT

	 10.00 - 19.99 MGT

20.00 - 39.99 MGT
40.00 - 59.99 MGT
60.00 -99.99 MGT
>= 100.00 MGT

Figure 2-6. Class I Railroads in the United States3

UP

Source: Federal Railroad Administration - December 2016

70


-------
For the 2016 inventory, the AAR provided a national line-haul Tier fleet mix profile representing the
entire Class I locomotive fleet. A locomotive's Tier level determines its allowable emission rates based
on the year when it was built and/or re-manufactured. The national fleet mix data was then used to
calculate weighted average in-use emissions factors for the line-haul locomotives operated by the Class I
railroads as shown in Table 2-27.

Table 2-27. 2016 Line-haul Locomotive Emission Factors by Tier, AAR Fleet Mix (grams/gal)



AAU









Tier Level

Meet Mix
Uatio

I'M it.

IK

NOx

(O

Uncontrolled (pre-1973)

0.047494

6.656

9.984

270.4

26.624

Tier 0(1973-2001)

0.188077

6.656

9.984

178.88

26.624

Tier 0+ (Tier 0 rebuilds)

0.141662

4.16

6.24

149.76

26.624

Tier 1 (2002-2004)

0.029376

6.656

9.776

139.36

26.624

Tier 1+ (Tier 1 rebuilds)

0.223147

4.16

6.032

139.36

26.624

Tier 2 (2005-2011)

0.124536

3.744

5.408

102.96

26.624

Tier 2+ (Tier 2 rebuilds)

0.093607

1.664

2.704

102.96

26.624

Tier 3 (2012-2014)

0.123113

1.664

2.704

102.96

26.624

Tier 4 (2015 and later)

0.028988

0.312

0.832

20.8

26.624

2016 Weighted EF's

1.000000

4.117

6.153

138.631

26.624

Based on values in EPA Technical Highlights: Emission Factors for Locomotives, EPA Office of Transportation and Air Quality, EPA-420-F-09-025, April
2009.

Weighted Emission Factors (EF) per pollutant for each gallon of fuel used (grams/gal or lbs/gal) were
calculated for the US Class I locomotive fleet based on the percentage of line-haul locomotives certified
at each regulated Tier level (Equation 2-3).

9

EFi = I EFiT x fr	Equation 2-3

7=1

where:

EFi = Weighted Emission Factor for pollutant i for Class I locomotive fleet (g/gal).

EFiT = Emission Factor for pollutant i for locomotives in Tier T (g/gal) (Table 4).
fr = Percentage of the Class I locomotive fleet in Tier T expressed as a ratio.

While actual engine emissions will vary within Tier level categories, the approach described above likely
provides reasonable emission estimates, as locomotive diesel engines are certified to meet the emission
standards for each Tier. It should be noted that actual emission rates may increase over time due to
engine wear and degradation of the emissions control systems. In addition, locomotives may be operated
in a manner that differs significantly from the conditions used to derive line-haul duty-cycle estimates.

Emission factors for other pollutants are not Tier-specific because these pollutants are not directly
regulated by USEPA's locomotive emission standards. PM2.5 was assumed to be 97% of PM104, the ratio
of volatile organic carbon (VOC) to (hydrocarbon) HC was assumed to be 1.053, and the emission factors

71


-------
used for sulfur dioxide (SO2) and ammonia (NH3)were 0.0939 g/gal4 and 83.3 mg/gal6, respectively. The
2016 SO2 emission factor is based on the nationwide adoption of 15 ppm ultra-low sulfur diesel (ULSD)
fuel by the rail industry.

The remaining steps to compute the Class 1 rail emissions involved calculating class I railroad-specific
rail fuel consumption index values and calculating emissions per link. The final
link-level emissions for each pollutant were then aggregated by state/county FIPS code and then
converted into an FF10 format used by SMOKE. More detail on these steps is described in the
specification sheet for the 2016vl rail sector emissions.

Rail yard Methodology

Rail yard emissions were computed based on fuel use and/or yard switcher locomotive counts for the class
I rail companies for all of the rail yards on their systems. Three railroads provided complete rail yard
datasets: BNSF, UP, and KCS. CSX provided switcher counts for its 14 largest rail yards. This reported
activity data was matched to existing yard locations and data stored in USEPA's Emissions Inventory
System (EIS) database. All existing EIS yards that had activity data assigned for prior years, but no
reported activity data for 2016 were zeroed out. New yard data records were generated for reported
locations that were not found in EIS. Special care was made to ensure that the new yards added to EIS
did not duplicate existing data records. Data for non-Class I yards was carried forward from the 2014
NEI.

Since the railroads only supplied switcher counts, average fuel use per switcher values were calculated for
each railroad. This was done by dividing each company's 2016 R-l yard fuel use total by the number of
switchers reported for each railroad. These values were then used to allocate fuel use to each yard based
on the number of switchers reported for that location. Table 2-28 summarizes the 2016 yard fuel use and
switcher data for each Class I railroad. The emission factors used for rail yard switcher engines are
shown in Table 2-29.

Table 2-28. Surface Transportation Board R-l Fuel Use Data - 2016

Kiiili'ttiid

2111(. IM Yard

l-licl I SO (liill)

i:rt.\( ciilculiiicd

I'lK'l I SO (liill)

1 (I011 li I'ied
S« ilchois

r.K 1 AC per Sm holier l-'uel
I SO l«!ll)

BiNSF

40,279,454

40,740,317

442

92,173

CSXT

39,364,896

43,054,795

455

94,626

CN

6,570,898

6,570,898

103

63,795

KCS

3,211,538

3,211,538

176

18,247

NS

28,595,955

28,658,528

458

62,573

CPRS

1,311,135

1,311,135

70

18,731

UP

85,057,080

85,057,080

1286

66,141

All Class I's

204,390,956

208,604,291

2,990

69,767

Table 2-29. 2016 Yard Switcher Emission Factors by Tier, AAR Fleet Mix (grams/gal)4

Tier Level

AAU l leel
Mix K;i1 io

I'M HI

IK

NOx

CO

Uncontrolled (pre-1973)

0.2601

6.688

15.352

264.48

27.816

Tier 0 (1973-2001)

0.2361

6.688

15.352

191.52

27.816

72


-------
Tier 0+ (Tier 0 rebuilds)

0.2599

3.496

8.664

161.12

27.816

Tier 1 (2002-2004)

0.0000

6.536

15.352

150.48

27.816

Tier 1+ (Tier 1 rebuilds)

0.0476

3.496

8.664

150.48

27.816

Tier 2 (2005-2011)

0.0233

2.888

7.752

110.96

27.816

Tier 2+ (Tier 2 rebuilds)

0.0464

1.672

3.952

110.96

27.816

Tier 3 (2012-2014)

0.1018

1.216

3.952

68.4

27.816

Tier 4 (2015 and later)

0.0247

0.228

1.216

15.2

27.816

2016 Weighted EF's

0.9999

4.668

11.078

178.1195

27.813

Based on values in EPA Technical Highlights: Emission Factors for Locomotives,EPA Office of Transportation and Air Quality, EPA-420-F-09-025, April
2009. AAR fleet mix ratios did not add up to 1.0000, which caused a small error for the CO weighted emission factor as shown above.

In addition to the Class I rail yards, Emission estimates were calculated for four large Class III railroad
hump yards which are among the largest classification facilities in the United States. These four yards are
located in Chicago (Belt Railway of Chicago-Clearing and Indiana Harbor Belt-Blue Island) and Metro-
East St. Louis (Alton & Southern-Gateway and Terminal Railroad Association of St. Louis-Madison).
Figure 2-7 shows the spatial distribution of active yards in the 2016vl and 2017 NEI inventories.

Figure 2-7. 2016-2017 Active Rail Yard Locations in the United States

2016-2017 Active Rallyaids-

~ RailyjiiI LocataOn*
	 Clas-s I Railroads.

Sewrfeft	Adnwiadrttf

Class II and III Methodology

There are approximately 560 Class II and III Railroads operating in the United States, most of which are
members of the American Short Line and Regional Railroad Association (ASLRRA). While there is a lot
of information about individual Class II and III railroads available online, a significant amount of effort

73


-------
would be required to convert this data into a usable format for the creation of emission inventories. In
addition, the Class II and III rail sector has been in a constant state of flux ever since the railroad industry
was deregulated under the Staggers Act in 1980. Some states have conducted independent surveys of
their Class II and III railroads and produced emission estimates, but no national level emissions inventory
existed for this sector of the railroad industry prior to ERTAC Rail's work for the 2008 NEI.

Class II and III railroad activities account for nearly 4 percent of the total locomotive fuel use in the
combined ERTAC Rail emission inventories and for approximately 35 percent of the industry's national
freight rail track mileage. These railroads are widely dispersed across the country and often utilize older,
higher emitting locomotives than their Class I counterparts Class II and III railroads provide
transportation services to a wide range of industries. Individual railroads in this sector range from small
switching operations serving a single industrial plant to large regional railroads that operate hundreds of
miles of track. Figure 2-8 shows the distribution of Class II and III railroads and commuter railroads
across the country. This inventory will be useful for regional and local modeling, helps identify where
Class II and III railroads may need to be better characterized, and provides a strong foundation for the
future development of a more accurate nationwide short line and regional railroad emissions inventory. A
picture of the locations of class II and III railroads is shown in Figure 2-8. The data sources, calculations,
and assumptions used to develop the Class II and III inventory are described in the 2016vl rail
specification sheet.

Figure 2-8. Class II and III Railroads in the United States5

Commuter Rail Methodology

Commuter rail emissions were calculated in the same way as the Class II and III railroads. The primary
difference is that the fuel use estimates were based on data collected by the Federal Transit

74


-------
Administration (FTA) for the National Transit Database. 2016 fuel use was then estimated for each of the
commuter railroads shown in Table 2-30 by multiplying the fuel and lube cost total by 0.95, then dividing
the result by Metra's average diesel fuel cost of $1.93/gallon. These fuel use estimates were replaced
with reported fuel use statistics for MARC (Maryland), MBTA (Massachusetts), Metra (Illinois), and NJT
(New Jersey). The commuter railroads were separated from the Class II and III railroads so that the
appropriate SCC codes could be entered into the emissions calculation sheet.

Table 2-30. Expenditures and fuel use for commuter rail

1 R\

(ode

S\s(i-m

( ilios Sencd

Propulsion
1 > pi-

DOT 1 ml A

1 -ii ho Cosls

Ki-porli-d/I'lsliniiili-d
I-iii-I I si-

ACEX

Altamont Corridor
Express

San Jose / Stockton

Diesel

$889,828

437,998.24

CMRX

Capital MetroRail

Austin

Diesel

No data

n/a

DART

A-Train

Denton

Diesel

$0

0.00

DRTD

Denver RTD: A&B
Lines

Denver

Electric

$0

0.00

JPBX

Caltrain

San Francisco / San Jose

Diesel

$7,002,612

3,446,881.55

LI

MTA Long Island Rail
Road

New York

Electric and
Diesel

$13,072,158

6,434,481.92

MARC

MARC Train

Baltimore / Washington, D.C.

Diesel and
Electric

$4,648,060

4.235.297.57

MBTA

MBTA Commuter Rail

Boston / Worcester / Providence

Diesel

$37,653,001

12.142.826.00

MNCW

MTA Metro-North
Railroad

New York / Yonkers / Stamford

Electric and
Diesel

$13,714,839

6,750,827.49

NICD

NICTD South Shore
Line

Chicago / South Bend

Electric

$181,264

0.00

NIRC

Metra

Chicago

Diesel and
Electric

$52,460,705

25.757.673.57

NJT

New Jersey Transit

New

York / Newark / Trenton / Philadelphia

Electric and
Diesel

$38,400,031

16.991.164.00

NMRX

New Mexico Rail
Runner

Albuquerque / Santa Fe

Diesel

$1,597,302

786,236.74

CFCR

SunRail

Orlando

Diesel

$856,202

421,446.58

MNRX

Northstar Line

Minneapolis

Diesel

$708,855

348,918.26

Not
Coded

SMART

San Rafael-Santa Rosa (Opened 2017)

Diesel

n/a

0.00

NRTX

Music City Star

Nashville

Diesel

$456,099

224,504.69

SCAX

Metrolink

Los Angeles / San Bernardino

Diesel

$19,245,255

9,473,052.98

SDNR

NCTD Coaster

San Diego / Oceanside

Diesel

$1,489,990

733,414.77

SDRX

Sounder Commuter
Rail

Seattle / Tacoma

Diesel

$1,868,019

919,491.22

SEPA

SEPTA Regional Rail

Philadelphia

Electric

$483,965

0.00

SLE

Shore Line East

New Haven

Diesel

No data

n/a

TCCX

Tri-Rail

Miami / Fort Lauderdale / West Palm
Beach

Diesel

$5,166,685

2,543,186.92

TREX

Trinity Railway
Express

Dallas / Fort Worth

Diesel

No data

n/a

UTF

UTA FrontRunner

Salt Lake City / Provo

Diesel

$4,044,265

1,990,700.39

VREX

Virginia Railway
Express

Washington, D.C.

Diesel

$3,125,912

1,538,661.35

WSTX

Westside Express
Service

Beaverton

Diesel

No data

n/a

*Reported fuel use values were used for MARC, MBTA, Metra, and New Jersey Transit.

75


-------
Intercity Passenger Methodology (Amtrak)

2016 marked the first time that a nationwide intercity passenger rail emissions inventory was created for
Amtrak. The calculation methodology mimics that used for the Class II and III and commuter railroads
with a few modifications. Since link-level activity data for Amtrak was unavailable, the default
assumption was made to evenly distribute Amtrak's 2016 reported fuel use across all of it diesel-powered
route-miles shown in Figure 2-9. Participating states were instructed that they could alter the fuel use
distribution within their jurisdictions by analyzing Amtrak's 2016 national timetable and calculating
passenger train-miles for each affected route. Illinois and Connecticut chose to do this and were able to
derive activity-based fuel use numbers for their states based on Amtrak's 2016 reported average fuel use
of 2.2 gallons per passenger train-mile. In addition, Connecticut provided supplemental data for selected
counties in Massachusetts, New Hampshire, and Vermont. Amtrak also submitted company-specific fleet
mix information and company-specific weighted emission factors were derived. Amtrak's emission rates
were 25% lower than the default Class II and III and commuter railroad emission rate. Details on the
computation of the Amtrak emissions are available in the rail specification sheet.

Figure 2-9. Amtrak Routes with Diesel-powered Passenger Trains

Seurc* Finktjl R*Aci*j Mnn*aj*n KI4

Other Data Sources

The California Air Resources Board (CARB) provided rail inventories for inclusion in the 2016vl
platform. C ARB's rail inventories were used in California, in place of the national dataset described
above. For rail yards, the national point source rail yard dataset was used to allocate CARB-submitted rail
yard emissions to point sources where possible. That is, for each California county with at least one rail
yard in the national dataset, the emissions in the national rail yard dataset were adjusted so that county

76


-------
total rail yard emissions matched the CARB dataset. In other words, 2016vl platform includes county
total rail yard emissions from CARB, but the locations of rail yards are based on the national
methodology. There are three counties with CARB-submitted rail yard emissions, but no rail yard
locations in the national dataset; for those counties, the rail yard emissions were included in the rail sector
using SCC 2285002010.

North Carolina separately provided passenger train (SCC 2285002008) emissions for use in the platform.
We used NC's passenger train emissions instead of the corresponding emissions from the Lake Michigan
Air Directors Consortium (LADCO) dataset.

None of these rail inventory sources included HAPs. For VOC speciation, the EPA preferred augmenting
the inventory with HAPs and using those HAPs for integration, rather than running the sector as a no-
integrate sector. So, Naphthalene, Benzene, Acetaldehyde, Formaldehyde, and Methanol (NBAFM)
emissions were added to all rail inventories, including the California inventory, using the same
augmentation factors as are used to augment HAPs in the NEI.

2.4.4 Nonroad Mobile Equipment Sources (nonroad)

The mobile nonroad equipment sector includes all mobile source emissions that do not operate on roads,
excluding commercial marine vehicles, railways, and aircraft. Types of nonroad equipment include
recreational vehicles, pleasure craft, and construction, agricultural, mining, and lawn and garden
equipment. Nonroad equipment emissions were computed by running the MOVES2014b model,11 which
incorporates the NONROAD2008 model. MOVES2014b replaced MOVES2014a in August 2018, and
incorporates updated nonroad engine population growth rates, nonroad Tier 4 engine emission rates, and
sulfur levels of nonroad diesel fuels. MOVES2014b provides a complete set of HAPs and incorporates
updated nonroad emission factors for HAPs. MOVES2014b was used for all states other than California
and Texas, which developed their own emissions using their own tools. VOC and PM speciation profile
assignments are determined by MOVES and applied by SMOKE.

MOVES2014b provides estimates of NONHAPTOG along with the speciation profile code for the
NONHAPTOG emission source. This was accomplished by using NHTOG#### as the pollutant code in
the Flat File 2010 (FF10) inventory file that can be read into SMOKE, where #### is a speciation profile
code. One of the speciation profile codes is '95335a' (lowercase 'a'); the corresponding inventory
pollutant is NONHAPTOG95335A (uppercase 'A') because SMOKE does not support inventory
pollutant names with lowercase letters. Since speciation profiles are applied by SCC and pollutant, no
changes to SMOKE were needed to use the inventory file with this profile information. This approach
was not used for California or Texas, because the datasets in those states included VOC.

MOVES2014b, unlike MOVES2014a, also provides estimates of PM2.5 by speciation profile code for the
PM2.5 emission source, using PM25_#### as the pollutant code in the FF10 inventory file, where #### is
a speciation profile code. To facilitate calculation of coarse particulate matter (PMC) within SMOKE, and
to help create emissions summaries, an additional pollutant representing total PM2.5 called PM25TOTAL
was added to the inventory. As with VOC / TOG, this approach is not used for California or Texas.

MOVES2014b outputs emissions data in county-specific databases, and a post-processing script converts
the data into FF10 format. Additional post-processing steps were performed as follows:

• County-specific FFlOs were combined into a single FF10 file.

11 https://www.epa. gov/moves

77


-------
•	Emissions were aggregated from the more detailed SCCs modeled in MOVES to the SCCs
modeled in SMOKE. A list of the aggregated SMOKE SCCs is in Appendix A of the 2016vl
nonroad specification sheet.

•	To reduce the size of the inventory, HAPs that are not needed for air quality modeling, such as
dioxins and furans, were removed from the inventory.

•	To reduce the size of the inventory further, all emissions for sources (identified by county/SCC)
for which total CAP emissions are less than 1*10"10 were removed from the inventory. The
MOVES model attributes a very tiny amount of emissions to sources that are actually zero, for
example, snowmobile emissions in Florida. Removing these sources from the inventory reduces
the total size of the inventory by about 7%.

•	Gas and particulate components of HAPs that come out of MOVES separately, such as
naphthalene, were combined.

•	VOC was renamed VOC INV so that SMOKE does not speciate both VOC and NONHAPTOG,
which would result in a double count.

•	PM25TOTAL, referenced above, was also created at this stage of the process.

•	California and Texas emissions from MOVES were deleted and replaced with the CARB- and
TCEQ-supplied emissions, respectively.

Emissions for airport ground support vehicles (SCCs ending in -8005), and oil field equipment (SCCs
ending in -10010), were removed from the mobile nonroad inventory, to prevent a double count with the
ptnonipm and np oilgas sectors, respectively.

National Updates: Agricultural and Construction Equipment Allocation

The methodology for developing Agricultural equipment allocation data for the 2016vl platform was
developed by the North Carolina Department of Environmental Quality (NCDEQ). EPA updated the
Construction equipment allocation data for the vl platform.

NCDEQ compiled regional and state-level Agricultural sector fuel expenditure data for 2016 from the US
Department of Agriculture, National Agricultural Statistics Service (NASS), August 2018 publication,
"Farm Production Expenditures 2017 Summary."12 This resource provides expenditures for each of 5
major regions that cover the Continental U.S., as well as state-level data for 15 major farm producing
states. Because of the limited coverage of the NASS source relative to that in MOVES, it was necessary to
identify a means for estimating the 2016 Agricultural sector allocation data for the following States and
Territories from a different source: Alaska, Hawaii, Puerto Rico, and U.S. Virgin Islands. The approach
for these areas is described below.

For the Continental U.S., NCDEQ first allocated the remainder of the regional fuel expenditures to states
in each region for which state-level data are not reported. For this allocation, NCDEQ relied on 2012 fuel
expenditure data from NASS' 2012 Census of Agriculture (note that 2017 data were not yet available at
the time of this effort).13 The next step to developing county-level allocation data for agricultural

12	Accessed from htto://usda.mannlib.Cornell.edu/MannUsda/viewDocumentlnfo.do?documentID= 1066. November 2018.

13	Accessed from https://www.nass.usda.gov/Publications/AgCensus/2012/. November 2018.

78


-------
equipment was to multiply the state-level fuel expenditure estimates by county-level allocation ratios.
These allocation ratios were computed from county-level fuel expenditure data from the NASS' 2012
Census of Agriculture. There were 17 counties for which fuel expenditure data were withheld in the
Census of Agriculture. For these counties, NDEQ allocated the fuel expenditures that were not accounted
for in the applicable state via a surrogate indicator of fuel expenditures. For most states, the 2012 Census
of Agriculture's total machinery asset value was the surrogate indicator used to perform the allocation.
This indicator was found to have the strongest correlation to agricultural sector fuel expenditures based on
analysis of 2012 state-level Census of Agriculture values for variables analyzed (correlation coefficient of
0.87).14 Because the analyzed surrogate variables were not available for the two counties in New York
without fuel expenditure data, farm sales data from the 2012 Census of Agriculture were used in the
allocation procedure for these counties.

For Alaska and Hawaii, NCDEQ estimated 2016 state-level fuel production expenditures by first applying
the national change in fuel expenditures between 2012 and 2016 from NASS' "Farm Production
Expenditures" summary publications to 2012 state expenditure data from the 2012 Census of Agriculture.
Next, NCDEQ applied an adjustment factor to account for the relationship between national 2012 fuel
expenditures as reported by the Census of Agriculture and those reported in the Farm Production
Expenditures Summary. Hawaii's state-level fuel expenditures were allocated to counties using the same
approach as the states in the Continental U.S. (i.e., county-level fuel expenditure data from the NASS'
2012 Census of Agriculture). Alaska's fuel expenditures total was allocated to counties using a different
approach because the 2012 Census of Agriculture reports fuel expenditures data for a different list of
counties than the one included in MOVES. To ensure consistency with MOVES, NCDEQ allocated
Alaska's fuel expenditures based on the current allocation data in MOVES, which reflect 2002 harvested
acreage data from the Census of Agriculture.

Because NCDEQ did not identify any source of fuel expenditures data for Puerto Rico or the U.S. Virgin
Islands, the county allocation percentages that are represented by the 2002 MOVES allocation data were
used for these territories.15

For the Construction sector, MOVES2014b uses estimates of 2003 total dollar value of construction by
county to allocate national Construction equipment populations to the state and local levels.16 However,
the 2016 Nonroad Collaborative Work Group sought to update the surrogate data used to geographically
allocate Construction equipment with a more recent data source thought to be more reflective of
emissions-generating Construction equipment activity at the county level: acres disturbed by residential,
non-residential, and road construction activity.

The nonpoint sector of the 2014 National Emissions Inventory (NEI) includes estimates of Construction
Dust (PM2.5), for which acreage disturbed by residential, non-residential, and road construction activity is
a function.17 The 2014 NEI Technical Support Document18 includes a description of the methods used to
estimate acreage disturbed at the county level by residential, non-residential, and road construction
activity, for the 50 states.

Acreage disturbed by residential, non-residential, and road construction were summed together to arrive at
a single value of acreage disturbed by Construction activities at the county level. County-level acreage

14	Other variables analyzed were inventory of tractors and inventory of trucks.

15	For reference, these allocations were 0.0639 percent for Puerto Rico and 0.0002 percent for the U.S. Virgin Islands.

16	https://nepis.epa.gov/Exe/ZvPDF.cgi?Dockev=P1004LDX.pdf

17	https://www.epa.gov/air-emissions-inventories/2014-national-emissions-inventorv-nei-data

18	https://www.epa.gov/sites/production/files/2018-07/documents/nei2014v2 tsd 05iul2018.pdf

79


-------
disturbed were then summed together to arrive at acreage disturbed at the state level. State totals were
then summed to arrive at a national total of acreage disturbed by Construction activities.

Puerto Rico and the U.S. Virgin Islands are not included in the Construction equipment geographic
allocation update, so their relative share of the national population of Construction equipment remains the
same as MOVES2014b defaults.

For both the Agricultural and Construction equipment sectors, the surrogatequant and surrogateyearlD
fields in the model's nr state surrogate table, which allocates equipment from the state- to the county-level,
were populated with the county-level surrogates described above (fuel expenditures in 2016 for
Agricultural equipment; acreage disturbed by construction activity in 2014 for Construction equipment).
In addition, the nrbaseyearequippopulation table, which apportions the model's national equipment
populations to the state level, was adjusted so that each state's share of the MOVE-S2014b base-year
national populations of Agricultural and Construction equipment is proportional to each state's share of
national acreage disturbed by construction activity (Construction equipment) and agricultural fuel
expenditures (Agricultural equipment). Additionally, the model's nr surrogate table, which defines the
surrogate data used in the nrstatesurrogate table, was updated to reflect the 2016vl changes to the
Agricultural and Construction equipment sectors.

Updated nrsurrogate, nrstatesurrogate, and nrbaseyearequippopulation tables, along with instructions for
utilizing these tables in MOVES runs, are available for download from EPA's ftp site:
ftp://newftp.epa.gov/air/emismod/2016/vl/reports/nonroad/ or at
https://gaftp.epa.gov/air/emismod/2016/vl/reports/nonroad/).

State-Supplied Nonroad Data

As shown Table 2-31 several state and local agencies provided nonroad inputs for use in the 2016vl
platform. Additionally, per the table footnotes, EPA reviewed data submitted by state and local agencies
for the 2014 and 2017 National Emissions Inventories and utilized that information where appropriate
(data specific to calendar years 2014 and 2017 were not used in 2016vl).

80


-------
Table 2-31. Submitted nonroad input tables by agency

slsili'id

Sisilc or
( oiiniMii's) in
llii' Aiii'iio

=

— /.
= =

11

— 'j

5 j*

C' o

2 w

z

Q.

r —•

3 s

2

2

= _2

~

5T <—i
=

& zz
~ 2
s —
-

s _2

* —

.5 lb
.= -

i 2

Tj. "=
- a.
= s

a.

o

r E.

'v *z

rE -5
•- s
s —

-E 2

= TZ
'J

_2

s -
.= 2

:Z :j
jz

s =
c s

= k

— o
s *

^ _
.«*.

Tt "

x —

= 'j
i* —

= h

i.

zl 5
*_ 'J
•I s
= —

V5 

Submitted data with modification: deleted records that were not snowmobile source types 1002-1010.
c NEI 2014v2 data used for 2016vl platform.

D Submitted data.

17

Spreadsheet "ladco_nei2017_nrmonthallocation.xlsx."

17

Submitted data with modification: deleted records that were not the snowmobile surrogate ID 14.

Emissions Inside California and Texas

California nonroad emissions were provided by CARB for the years 2016, 2023, and 2028.

All California nonroad inventories are annual, with monthly temporalization applied in SMOKE.
Emissions for oil field equipment (SCCs ending in -10010) were removed from the California inventory
in order to prevent a double count with the np oilgas sector.

Texas nonroad emissions were provided by the Texas Commission on Environmental Quality for the
years 2016, 2023, and 2028, using TCEQ's TexN2 tool.19 This tool facilitates the use of detailed Texas-
specific nonroad equipment population, activity, fuels, and related data as inputs for MOVES2014b, and
accounts for Texas-specific emission adjustments such as the Texas Low Emission Diesel (TxLED)
program.

19 For more information on the TexN2 tool please see: ftp://amdaftp.tcea.texas.gov/EI/nonroad/TexN2/

81


-------
Nonroad Updates from State Comments

The 2016 Nonroad Collaborative Work group received a small number of comments on the 2016beta
inventory, all of which were addressed and implemented in the 2016vl nonroad inventory:

•	Georgia Department of Natural Resources: incorporate updated fuel supply {nrfuelsupply
table) for 45 Georgia counties, to reflect the removal of summer Reid Vapor Pressure restrictions
in 2016; utilize updated geographic allocation factors (nr state surrogate table) for the
Commercial, Lawn & Garden (commercial, public, and residential), Logging, Manufacturing,
Golf Carts, Recreational, Railroad Maintenance Equipment and A/C/Refrigeration sectors, using
data from the U.S. Census Bureau and U.S. Forest Service.

•	Lake Michigan Air Directors Consortium (LADCO): update seasonal allocation of agricultural
equipment activity (nrmonthallocation table) for Illinois, Indiana, Iowa, Michigan, Minnesota,
Missouri, Ohio, and Wisconsin.

•	Texas Commission on Environmental Quality: replace MOVES2014b nonroad emissions for
Texas with emissions calculated with TCEQ's TexN2 model.

•	Alaska Department of Environmental Conservation: remove emissions as calculated by
MOVES2014b for several equipment sector-county/census areas combinations in Alaska, due to
an absence of nonroad activity (see Table 2-32).

Table 2-32. Alaska counties/census areas for which nonroad equipment sector-specific emissions

are removed in 2016vl

Nonroad Kqiiipmcnl Sector

Counties/Census Areas (I IPS) lor which equipment
sector emissions are rcmo\ed in 20I6\ 1

Agricultural

Aleutians East (02013), Aleutians West (02016), Bethel
Census Area (02050), Bristol Bay Borough (02060),
Dillingham Census Area (02070), Haines Borough (02100),
Hoonah-Angoon Census Area (02105), Ketchikan Gateway
(02130), Kodiak Island Borough (02150), Lake and
Peninsula (02164), Nome (02180), North Slope Borough
(02185), Northwest Arctic (02188), Petersburg Borough
(02195), Pr of Wales-Hyder Census Area (02198), Sitka
Borough (02220), Skagway Borough (02230), Valdez-
Cordova Census Area (02261), Wade Hampton Census Area
(02270), Wrangell City + Borough (02275), Yakutat City +
Borough (02282), Yukon-Koyukuk Census Area (02290)

Logging

Aleutians East (02013), Aleutians West (02016), Nome
(02180), North Slope Borough (02185), Northwest Arctic
(02188), Wade Hampton Census Area (02270)

Railway Maintenance

Aleutians East (02013), Aleutians West (02016), Bethel
Census Area (02050), Bristol Bay Borough (02060),
Dillingham Census Area (02070), Haines Borough (02100),
Hoonah-Angoon Census Area (02105), Juneau City +
Borough (02110), Ketchikan Gateway (02130), Kodiak
Island Borough (02150), Lake and Peninsula (02164), Nome

82


-------
Nonroiul Kqiiipmenl Sector

Counties/Census A rests (MI'S) lor which equipment
sector emissions ;irc remo\oil in 20I6\ 1



((PI SO), ). North Slope Borough (<) "* 1S5) Norllnwsl Arctic
(02188), Petersburg Borough (02195), Pr ofWales-Hyder
Census Area (02198), Sitka Borough (02220), Southeast
Fairbanks (02240), Wade Hampton Census Area (02270),
Wrangell City + Borough (02275), Yakutat City + Borough
(02282), Yukon-Koyukuk Census Area (02290)

2.5 2016 Fires (ptfire, ptagfire)

Multiple types of fires are represented in the modeling platform. These include wild and prescribed fires
that are grouped into the ptfire sector, and agricultural fires that comprise the ptagfire sector. All ptfire
and ptagfire fires are in the United States. Fires outside of the United States are described in the
ptfire othna sector later in this document.

2.5.1 Wild and Prescribed Fires (ptfire)

Wildfires and prescribed burns that occurred during the inventory year are included in the year 2016
version 1 (2016vl) inventory as event and point sources. The point agricultural fires inventory (ptagfire)
is described in a separate section. For purposes of emission inventory preparation, wildland fire (WLF) is
defined as any non-structure fire that occurs in the wildland. The wildland is defined an area in which
human activity and development are essentially non-existent, except for roads, railroads, power lines, and
similar transportation facilities. Wildland fire activity is categorized by the conditions under which the
fire occurs. These conditions influence important aspects of fire behavior, including smoke emissions. In
the 2016vl inventory, data processing was conducted differently depending on the fire type, as defined
below:

•	Wildfire (WF): 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.

•	Prescribed (Rx) fire: any fire intentionally ignited by management actions in accordance with
applicable laws, policies, and regulations to meet specific land or resource management
objectives. Prescribed fire is one type of fire fuels treatment. Fire fuels treatments are vegetation
management activities intended to modify or reduce hazardous fuels. Fuels treatments include
prescribed fires, wildland fire use, and mechanical treatment.

The SCCs used for the ptfire sources are shown in Table 2-33. The ptfire inventory includes separate
SCCs for the flaming and smoldering combustion phases for wildfire and prescribed burns. Note that
prescribed grassland fires or Flint Hills, Kansas have their own SCC in the 2016vl inventory. The year
2016 fire season also included some major wild grassland fires. These wild grassland fires were assigned
the standard wildfire SCCs shown in Table 2-33.

Table 2-33. SCCs included in the ptfire sector for the 2016vl inventory

SCC

Description

2801500170

Grassland fires; prescribed

2810001001

Forest Wildfires; Smoldering; Residual smoldering only (includes grassland
wildfires)

83


-------
see

Description

2810001002

Forest Wildfires; Flaming (includes grassland wildfires)

2811015001

Prescribed Forest Burning; Smoldering; Residual smoldering only

2811015002

Prescribed Forest Burning; Flaming

National Fire Information Data

Numerous fire information databases are available from U.S. national government agencies. Some of the
databases are available via the internet while others must be obtained directly from agency staff. Table
2-34 provides the national fire information databases that were used for the 2016vl ptfire inventory,
including the website where the 2016 data were downloaded.

Table 2-34. National fire information databases used in 2016vl ptfire inventory

Dataset Name

Fire
Types

Form
at

Agenc

y

Coverage

Source

Hazard Mapping
System (HMS)

WF/R
X

CSV

NOA
A

North
America

httDs://www.osDo.noaa.gov/Products/land/h
ms.html

Geospatial Multi-
Agency

Coordination(GeoM
AC)











WF

SHP

USGS

Entire US

https://www.geomac.gov/GeoMACTransiti
on.shtml

Incident Command
System Form 209:
Incident Status
Summary (ICS-209)

WF/R
X

CSV

Multi

Entire US

httDs://fam. nwcg.gov/fam-web/

National

Association of State
Foresters (NASF)

WF

CSV

Multi

Participati
ng US
states

https://fam.nwcg.gov/fam-web/ (see Public
Access Reports, Free Data Extract, then NASF State
Data Extract)

Monitoring Trends
in Burn Severity
(MTBS)

WF/R
X

SHP

USGS,
USFS

Entire US

httDs://www.mtbs.gov/direct-download

Forest Service
Activity Tracking
System (FACTS)

RX

SHP

USFS

Entire US

Hazardous Fuel Treatment Reduction: Polygon
at https://data.fs.usda.aov/aeodata/edw/
datasets.DhD

US Fish and
Wildland Service
(USFWS) fire
database

WF/R
X

CSV

USFW

s

Entire US

Direct communication with USFWS

The Hazard Mapping System (HMS) was developed in 2001 by the National Oceanic and
Atmospheric Administration's (NOAA) National Environmental Satellite and Data Information
Service (NESDIS) as a tool to identify fires over North America in an operational environment. The
system utilizes geostationary and polar orbiting environmental satellites. Automated fire detection
algorithms are employed for each of the sensors. When possible, HMS data analysts apply quality control
procedures for the automated fire detections by eliminating those that are deemed to be false and adding
hotspots that the algorithms have not detected via a thorough examination of the satellite imagery.

84


-------
The HMS product used for the 2016vl inventory consisted of daily comma-delimited files containing fire
detect information including latitude-longitude, satellite used, time detected, and other information. The
Visible Infrared Imaging Radiometer Suite (VIIRS) satellite fire detects were introduced into the HMS in
late 2016. Since it was only available for a small portion of the year, the VIIRS fire detects were removed
for the entire year for consistency. In the 2016alpha inventory, the grassland fire detects were put in the
point agricultural fire sector (ptagfire). As there were a few significant grassland wildfires in Kansas and
Oklahoma in year 2016, all grassland fire detects were included in the ptfire sector for the 2016vl
inventory. These grassland fires were processed through Satellite Mapping Automated Reanalysis Tool
for Fire Incident Reconciliation version 2 (SMARTFIRE2) and BlueSky Framework.

GeoMAC (Geospatial Multi-Agency Coordination) is an online wildfire mapping application designed for
fire managers to access maps of current U.S. fire locations and perimeters. The wildfire perimeter data is
based upon input from incident intelligence sources from multiple agencies, GPS data, and infrared (IR)
imagery from fixed wing and satellite platforms.

The Incident Status Summary, also known as the "ICS-209" is used for reporting specific information on
significant fire incidents. The ICS-209 report is a critical interagency incident reporting tool giving daily
'snapshots' of the wildland fire management situation and individual incident information which include
fire behavior, size, location, cost, and other information. Data from two tables in the ICS-209 database
were merged and used for the 2016vl ptfire inventory: the

SIT209_HISTORY_INCIDENT_209_REPORTS table contained daily 209 data records for large fires,
and the SIT209 HISTORY INCIDENTS table contained summary data for additional smaller fires.

The National Association of State Foresters (NASF) is a non-profit organization composed of the
directors of forestry agencies in the states, U.S. territories, and District of Columbia to manage and protect
state and private forests, which encompass nearly two-thirds of the nation's forests. The NASF compiles
fire incident reports from agencies in the organization and makes them publicly available. The NASF fire
information includes dates of fire activity, acres burned, and fire location information.

Monitoring Trends in Burn Severity (MTBS) is an interagency program whose goal is to consistently map
the burn severity and extent of large fires across the U.S. from 1984 to present. The MTBS data includes
all fires 1,000 acres or greater in the western United States and 500 acres or greater in the eastern United
States. The extent of coverage includes the continental U.S., Alaska, Hawaii, and Puerto Rico. Fire
occurrence and satellite data from various sources are compiled to create numerous MTBS fire products.
The MTBS Burned Areas Boundaries Dataset shapefiles include year 2016 fires and that are classified as
either wildfires, prescribed burns or unknown fire types. The unknown fire type shapes were omitted in
the 2016vl inventory development due to temporal and spatial problems found when trying to use these
data.

The US Forest Service (USFS) compiles a variety of fire information every year. Year 2016 data from the
USFS Natural Resource Manager (NRM) Forest Activity Tracking System (FACTS) were acquired and
used for 2016vl emissions inventory development. This database includes information about activities
related to fire/fuels, silviculture, and invasive species. The FACTS database consists of shapefiles for
prescribed burns that provide acres burned,and start and ending time information.

The US Fish and Wildland Service (USFWS) also compiles wildfire and prescribed burn activity on their
federal lands every year. Year 2016 data were acquired from USFWS through direct communication with
USFWS staff and were used for 2016vl emissions inventory development. The USFWS fire information
provided fire type, acres burned, latitude-longitude, and start and ending times.

85


-------
State/Local/Tribal Fire Information

During the 2016 emissions modeling platform development process, S/L/T agencies were invited by EPA
and 2016 Inventory Collaborative Fire Workgroup to submit all fire occurrence data for use in developing
the 2016vl fire inventory. A template form containing the desired format for data submittals was
provided to S/L/T air agencies. The list of S/L/T agencies that submitted fire data is provided in Table 2-
35. Data from nine individual states and one Indian Tribe were used for the 2016vl ptfire inventory.

Table 2-35. List of S/L/T agencies that submitted fire data for 2016vl with types and formats.



Fire



S/L/T agency name

Types

Format

NCDEQ

WF/RX

CSV

KDHE

RX/AG

CSV

CO Smoke Mgmt





Program

RX

CSV

Idaho DEQ

AG

CSV

Nez Perce Tribe

AG

CSV

GADNR

ALL

EIS

MN

RX/AG

CSV

WA ECY

AG

CSV

NJDEP

WF/RX

CSV

Alaska DEC

WF/RX

CSV

The data provided by S/L/T agencies were evaluated by EPA and further feedback on the data submitted
by the state was requested at times. Table 2-36 provides a summary of the type of data submitted by each
S/L/T agency and includes spatial, temporal, acres burned and other information provided by the
agencies.

Table 2-36. Brief description of fire information submitted for 2016vl inventory use.

S/L/T

agency

name

Fire
Types

Description

NCDEQ

WF/RX

Fire type, period-specific, latitude-longitude and acres burned
information. Technical direction was to remove all fire detects
that were not reconciled with any other national or state
agency database.

Kansas
DHE

RX/AG

Day-specific, county-centroid located, acres burned for Flint
Hills prescribed burns for Feb 27-May 4 time period.
Reclassified fuels for some agricultural burns. A grassland
gridding surrogate was used to spatially allocate the day-
specific grassland fire emissions.

Colorado
Smoke
Mgmt
Program

RX

Day-specific, latitude-longitude, and acres burned for
prescribed burns

86


-------
S/L/T

agency

name

Fire
Types

Description

Idaho DEQ

AG

Day-specific, latitude-longitude, acres burned for agricultural
burns. Total replacement of 2016 alpha fire inventory for
Idaho.

Nez Perce
Tribe

AG

Day-specific, latitude-longitude, acres burned for agricultural
burns. Total replacement of 2016 alpha fire inventory within
the tribal area boundary.

Georgia
DNR

ALL

Data submitted included all fires types via EIS. The wildfire
and prescribed burn data were provided as daily, point
emissions sources. The agricultural burns were provided as
day-specific point emissions sources.

Minnesota

RX/AG

Corrected latitude-longitude, day-specific and acres burned
for some prescribed and agricultural burns.

Washington
ECY

AG

Month-specific, latitude-longitude, acres burned, fuel loading
and emissions for agricultural burns. Not day-specific so
allocation to daily implemented by EPA. WA state direction
included to continue to use the 2014NEIv2 pile burns that
were included in the non-point sector for 2016vl.

New Jersey
DEP

WF/RX

Day-specific, latitude-longitude, and acres burned for wildfire
and prescribed burns.

Alaska DEC

WF/RX

Day-specific, latitude-longitude, and acres burned for wildfire
and prescribed burns.

Fire Emissions Estimation Methodology

The national and S/L/T data mentioned earlier were used to estimate daily wildfire and prescribed burn
emissions from flaming combustion and smoldering combustion phases for the 2016vl inventory.
Flaming combustion is more complete combustion than smoldering and is more prevalent with fuels that
have a high surface-to-volume ratio, a low bulk density, and low moisture content. Smoldering
combustion occurs without a flame, is a less complete burn, and produces some pollutants, such as
PM2.5, VOCs, and CO, at higher rates than flaming combustion. Smoldering combustion is more
prevalent with fuels that have low surface-to-volume ratios, high bulk density, and high moisture content.
Models sometimes differentiate between smoldering emissions that are lofted with a smoke plume and
those that remain near the ground (residual emissions), but for the purposes of the 2016vl inventory the
residual smoldering emissions were allocated to the smoldering SCCs listed in Table 2-33. SCCs
included in the ptfire sector for the 2016vl inventoryTable 2-33. The lofted smoldering emissions
were assigned to the flaming emissions SCCs in Table 2-33.

Figure 2-10 is a schematic of the data processing stream for the 2016vl inventory for wildfire and
prescribe burn sources. The ptfire inventory sources were estimated using Satellite Mapping Automated
Reanalysis Tool for Fire Incident Reconciliation version 2 (SMARTFIRE2) and Blue Sky Framework.
SMARTFIRE2 is an algorithm and database system that operate within a geographic information system
(GIS). SMARTFIRE2 combines multiple sources of fire information and reconciles them into a unified
GIS database. It reconciles fire data from space-borne sensors and ground-based reports, thus drawing on
the strengths of both data types while avoiding double-counting of fire events. At its core, SMARTFIRE2

87


-------
is an association engine that links reports covering the same fire in any number of multiple databases. In
this process, all input information is preserved, and no attempt is made to reconcile conflicting or
potentially contradictory information (for example, the existence of a fire in one database but not
another).

For the 2016vl inventory, the national and S/L/T fire information was input into SMARTFIRE2 and then
merged and associated based on user-defined weights for each fire information dataset. The output from
SMARTFIRE2 was daily acres burned by fire type, and latitude-longitude coordinates for each fire. The
fire type assignments were made using the fire information datasets. If the only information for a fire was
a satellite detect for fire activity, then the flow described in Figure 2-11 was used to make fire type
assignment by state and by month.

Figure 2-10. Processing flow for fire emission estimates in the 2016vl inventory

Input Data Sets
(state/local/tribal and national data sets)

O

Data Preparation



Data Aggregation and Reconciliation
(SmartFire2)



Daily fire locations
with fire size and type





Fuel Moisture and
Fuel Loading Data

Smoke Modeling (BlueSky Framework)

Daily smoke emissions
for each fire

Emissions Post-Processing



Final Wildland Fire Emissions Inventory

88


-------
Figure 2-11. Default fire type assignment by state and month in cases where a satellite detect is only

source of fire information.

4 -

Default Fire Type
Assignment

WF Months

~	5,6,7,8

~	5,6,7,8,9,10

~	6,7,8
I None

The BlueSky Modeling Framework version 3.5 (revision #38169) was used to calculate fuel loading and
consumption, and emissions using various models depending on the available inputs as well as the desired
results. The contiguous United States and Alaska, where Fuel Characteristic Classification System
(FCCS) fuel loading data are available, were processed using the modeling chain described in Figure
2-12. The Fire Emissions Production Simulator (FEPS) in the Bluesky Framework generated all of the
CAP emission factors for wildland fires used in the 2016vl inventory. The FIAPs were derived from
regional emissions factors from Urbanski (2014).

Figure 2-12. Blue Sky Modeling Framework

Location

Dates

Type

Size



Fuels

(FCCS v3;





~L

Consumption

Emission



LF v1.4)

(Consume v4)

*-»¦ Factors

(FEPS v2)

Emissions

Bluesky Framework v3.5.0

89


-------
For the 2016vl inventory, the FCCSv2 spatial vegetation cover was upgraded to the LANDFIRE vl.4
fuel vegetation cover (See: https://www.landfire.gov/fccs.php). The FCCSv3 fuel bed characteristics were
implemented along with LANDFIREvl.4 to provide better fuel classification for the BlueSky Framework.
The LANDFIREvl.4 raster data were aggregated from the native resolution and projection to 200 meter
resolution using a nearest-neighbor methodology. Aggregation and reprojection was required to allow
these data to work in the BlueSky Framework.

2.5.2 Point source Agriculture Fires (ptagfire)

The point source agricultural fire (ptagfire) inventory sector contains daily agricultural burning emissions.
Daily fire activity was derived from the NOAA Hazard Mapping System (HMS) fire activity data. The
agricultural fires sector includes SCCs starting with '28015'. The first three levels of descriptions for
these SCCs are: 1) Fires - Agricultural Field Burning; Miscellaneous Area Sources; 2) Agriculture
Production - Crops - as nonpoint; and 3) Agricultural Field Burning - whole field set on fire. The SCC
2801500000 does not specify the crop type or burn method, while the more specific SCCs specify field or
orchard crops and, in some cases, the specific crop being grown. The SCCs for this sector listed are in
Table 2-37.

Table 2-37. SCCs included in the ptagfire sector for the 2016vl inventory

SCC

Description

2801500000

Miscellaneous Area Sources;Agriculture Production - Crops - as nonpoint;Agricultural Field Burning - whole
field set on fire;Unspecified crop type and Burn Method

2801500100

Miscellaneous Area Sources;Agriculture Production - Crops - as nonpoint;Agricultural Field Burning - whole

field set on fire;Field Crops Unspecified

2801500112

Miscellaneous Area Sources;Agriculture Production - Crops - as nonpoint;Agricultural Field Burning - whole

field set on fire;Field Crop is Alfalfa: Backfire Burning

2801500130

Miscellaneous Area Sources;Agriculture Production - Crops - as nonpoint;Agricultural Field Burning - whole
field set on fire;Field Crop is Barley: Burning Techniques Not Significant

2801500141

Miscellaneous Area Sources;Agriculture Production - Crops - as nonpoint;Agricultural Field Burning - whole
field set on fire;Field Crop is Bean (red): Headfire Burning

2801500150

Miscellaneous Area Sources;Agriculture Production - Crops - as nonpoint;Agricultural Field Burning - whole
field set on fire;Field Crop is Corn: Burning Techniques Not Important

2801500151

Miscellaneous Area Sources;Agriculture Production - Crops - as nonpoint;Agricultural Field Burning - whole

field set on fire;Double Crop Winter Wheat and Corn

2801500152

Miscellaneous Area Sources;Agriculture Production - Crops - as nonpoint;Agricultural Field Burning - whole

field set on fire;DoubleCrop Corn and Soybeans

2801500160

Miscellaneous Area Sources;Agriculture Production - Crops - as nonpoint;Agricultural Field Burning - whole
field set on fire;Field Crop is Cotton: Burning Techniques Not Important

2801500170

Miscellaneous Area Sources;Agriculture Production - Crops - as nonpoint;Agricultural Field Burning - whole
field set on fire;Field Crop is Grasses: Burning Techniques Not Important

2801500171

Miscellaneous Area Sources;Agriculture Production - Crops - as nonpoint;Agricultural Field Burning - whole

field set on fire;Fallow

2801500182

Miscellaneous Area Sources;Agriculture Production - Crops - as nonpoint;Agricultural Field Burning - whole
field set on fire;Field Crop is Hay (wild): Backfire Burning

2801500202

Miscellaneous Area Sources;Agriculture Production - Crops - as nonpoint;Agricultural Field Burning - whole

field set on fire;Field Crop is Pea: Backfire Burning

90


-------
see

Description

2801500220

Miscellaneous Area Sources;Agriculture Production - Crops - as nonpoint;Agricultural Field Burning - whole
field set on fire;Field Crop is Rice: Burning Techniques Not Significant

2801500250

Miscellaneous Area Sources;Agriculture Production - Crops - as nonpoint;Agricultural Field Burning - whole
field set on fire;Field Crop is Sugar Cane: Burning Techniques Not Significant

2801500262

Miscellaneous Area Sources;Agriculture Production - Crops - as nonpoint;Agricultural Field Burning - whole

field set on fire;Field Crop is Wheat: Backfire Burning

2801500263

Miscellaneous Area Sources;Agriculture Production - Crops - as nonpoint;Agricultural Field Burning - whole

field set on fire;DoubleCrop Winter Wheat and Cotton

2801500264

Miscellaneous Area Sources;Agriculture Production - Crops - as nonpoint;Agricultural Field Burning - whole
field set on fire;DoubleCrop Winter Wheat and Soybeans

2801500300

Miscellaneous Area Sources;Agriculture Production - Crops - as nonpoint;Agricultural Field Burning - whole

field set on fire;Orchard Crop Unspecified

2801500320

Miscellaneous Area Sources;Agriculture Production - Crops - as nonpoint;Agricultural Field Burning - whole

field set on fire;Orchard Crop is Apple

2801500350

Miscellaneous Area Sources;Agriculture Production - Crops - as nonpoint;Agricultural Field Burning - whole

field set on fire;Orchard Crop is Cherry

2801500410

Miscellaneous Area Sources;Agriculture Production - Crops - as nonpoint;Agricultural Field Burning - whole

field set on fire;Orchard Crop is Peach

2801500420

Miscellaneous Area Sources;Agriculture Production - Crops - as nonpoint;Agricultural Field Burning - whole

field set on fire;Orchard Crop is Pear

2801500500

Miscellaneous Area Sources;Agriculture Production - Crops - as nonpoint;Agricultural Field Burning - whole

field set on fire; Vine Crop Unspecified

2801500600

Miscellaneous Area Sources;Agriculture Production - Crops - as nonpoint;Agricultural Field Burning - whole

field set on fire;Forest Residues Unspecified

The EPA estimated biomass burning emissions using remote sensing data. These estimates were then
reviewed by the states and revised as resources allowed. As many states did not have the resources to
estimate emissions for this sector, remote sensing was necessary to fill in the gaps for regions where there
was no other source of data. Crop residue emissions result from either pre-harvest or post-harvest burning
of agricultural fields. The crop residue emission inventory for 2016 is day-specific and includes
geolocation information by crop type. The method employed and described here is based on the same
methods employed in the 2014 NEI with a few minor updates. It should be noted that grassland fires were
moved from the agricultural burning inventory sector to the prescribed and wildland fire sector for
2016beta and 2016vl inventories. This was done to prevent double-counting of fires and because the
largest fire (acres burned) in 2016 was a wild grassland fire in Kansas.

Daily, year-specific agricultural burning emissions were derived from HMS fire activity data, which
contains the date and location of remote-sensed anomalies. As point source inventories, the locations of
the fires are identified with latitude-longitude coordinates for specific fire events. The HMS activity data
were filtered using 2016 USDA cropland data layer (CDL). Satellite fire detects over agricultural lands
were assumed to be agricultural burns and assigned a crop type. Detects that were not over agricultural
lands were output to a separate file for use in the point source wildfire (ptfire) inventory sector. Each
detect was assigned an average size of between 40 and 80 acres based on crop type. The assumed field
sizes are found in Table 2-38.

91


-------
Table 2-38. Assumed field size of agricultural fires per state(acres)

State

Field Size

Alabama

40

Arizona

80

Arkansas

40

California

120

Colorado

80

Connecticut

40

Delaware

40

Florida

60

Georgia

40

Idaho

120

Illinois

60

Indiana

60

Iowa

60

Kansas

80

Kentucky

40

Louisiana

40

Maine

40

Maryland

40

Massachusetts

40

Michigan

40

Minnesota

60

Mississippi

40

Missouri

60

Montana

120

Nebraska

60

Nevada

40

New Hampshire

40

New Jersey

40

New Mexico

80

New York

40

North Carolina

40

North Dakota

60

Ohio

40

Oklahoma

80

Oregon

120

Pennsylvania

40

Rhode Island

40

South Carolina

40

South Dakota

60

Tennessee

40

Texas

80

Utah

40

Vermont

40

Virginia

40

Washington

120

West Virginia

40

Wisconsin

40

Wyoming

80

92


-------
Another feature of the ptagfire database is that the satellite detections for 2016 were filtered out to
exclude areas covered by snow during the winter months. To do this, the daily snow cover fraction per
grid cell was extracted from a 2016 meteorological Weather Research Forecast (WRF) model simulation.
The locations of fire detections were then compared with this daily snow cover file. For any day in which
a grid cell had snow cover, the fire detections in that grid cell on that day were excluded from the
inventory. Due to the inconsistent reporting of fire detections for year 2016 from the Visible Infrared
Imaging Radiometer Suite (VIIRS) platform, any fire detections in the HMS dataset that were flagged as
VIIRS or Suomi National Polar-orbiting Partnership satellite were excluded. In addition, certain crop
types (corn and soybeans) were excluded from the following states: Iowa, Kansas, Indiana, Illinois,
Michigan, Missouri, Minnesota, Wisconsin, and Ohio. Kansas was not included in this list in the 2014NEI
but added for 2016. The reason for these crop types being excluded is because states have indicated that
these crop types are not burned.

Crop type-specific emissions factors were applied to each daily fire to calculate criteria and hazardous
pollutant emissions. In all prior NEIs for this sector, the HAP emission factors and the VOC emission
factors were known to be inconsistent. The HAP emission factors were copied from the HAP emission
factors for wildfires in the 2014 NEI and in the 2016 beta and version 1 modeling platforms. The VOC
emission factors were scaled from the CO emission factors in the 2014 NEI and the 2016 beta and version
1 modeling platforms. See Pouliot et al, 2017 for a complete table of emission factors and fuel loading by
crop type.

Heat flux values for computing fire plume rise were calculated using the size and assumed fuel loading of
each daily fire. Emission factors and fuel loading by crop type are available in Table 1 of Pouliot et al.
(2017). This information is needed for a plume rise calculation within a chemical transport modeling
system. In prior NEIs including the 2014 NEI, all the emissions were placed into layer 1 (i.e. ground
level).

The daily agricultural and open burning emissions were converted from a tabular format into the
SMOKE-ready daily point Flat File 2010 (FF10) format. The daily emissions were also aggregated into
annual values by location and converted into the annual point flat file format.

2.6 2016 Biogenic Sources (beis)

Biogenic emissions for the entire year 2016 were developed using the Biogenic Emission Inventory
System version 3.61 (BEIS3.61) within SMOKE. The landuse input into BEIS3.61 is the Biogenic
Emissions Landuse Dataset (BELD) version 4.1 which is based on an updated version of the USDA-
USFS Forest Inventory and Analysis (FIA) vegetation speciation-based data from 2001 to 2014 from the
FIA version 5.1.

BEIS3.61 has some important updates from BEIS 3.14. These include the incorporation of Version 4.1 of
the Biogenic Emissions Landuse Database (BELD4), and the incorporation of a canopy model to estimate
leaf-level temperatures (Pouliot and Bash, 2015). BEIS3.61 includes a two-layer canopy model. Layer
structure varies with light intensity and solar zenith angle. Both layers of the canopy model include
estimates of sunlit and shaded leaf area based on solar zenith angle and light intensity, direct and diffuse
solar radiation, and leaf temperature (Bash et al., 2016). The new algorithm requires additional
meteorological variables over previous versions of BEIS. The variables output from the Meteorology-
Chemistry Interface Processor (MCIP) that are used for BEIS3.61 processing are shown in Table 2-39.
The 2016 version 1 of the BEIS3 modeling for year 2016 included processing for both a 36km (36US3)

93


-------
and 12km domain (12US1) (see Figure 3-1). The 12US2 modeling domain can also be supported by
taking a subset or window of the 12US1 BEIS3 emissions dataset.

Table 2-39. Hourly Meteorological variables required by BEIS 3.61

Variable

Description

LAI

leaf-area index

PRSFC

surface pressure

Q2

mixing ratio at 2 m

RC

convective precipitation

RGRND

solar rad reaching sfc

RN

nonconvective precipitation

RSTOMI

inverse of bulk stomatal resistance

SLYTP

soil texture type by USD A category

SOIM1

volumetric soil moisture in top cm

SOIT1

soil temperature in top cm

TEMPG

skin temperature at ground

USTAR

cell averaged friction velocity

RADYNI

inverse of aerodynamic resistance

TEMP2

temperature at 2 m

SMOKE-BEIS3 modeling system consists of two programs named: 1) Normbeis3 and 2) Tmpbeis3.
Normbeis3 uses emissions factors and BELD4 landuse to compute gridded normalized emissions for
chosen model domain (see Figure 2-13). The emissions factor file (B360FAC) contains leaf-area-indices
(LAI), dry leaf biomass, winter biomass factor, indicator of specific leaf weight, and normalized emission
fluxes for 35 different species/compounds. The BELD4 file is the gridded landuse for 276 different
landuse types. The output gridded domain is the same as the input domain for the land use data. Output
emission fluxes (B3GRD) are normalized to 30 °C, and isoprene and methyl-butenol fluxes are also
normalized to a photosynthetic active radiation of 1000 |imol/m2s.

Figure 2-13. Normbeis3 data flows

The normalized emissions output from Normbeis3 (B3GRD) are input into Tmpbeis3 along with the
MCIP meteorological data, chemical speciation profile to use for desired chemical mechanism, and
BIOSEASON file used to indicate how each day in year 2016 should be treated, either as summer or
winter. Figure 2-14 illustrates the data flows for the Tmpbeis3 program. The output from Tmpbeis
includes gridded, speciated, hourly emissions both in moles/second (B3GTS L) and tons/hour
(B3GTSS).

94


-------
Figure 2-14. Tmpbeis3 data flow diagram.

^ Program ^

Shows inpul or outojl

QolionaJ

'J

Biogenic emissions do not use an emissions inventory and do not have SCCs. The gridded land use data,
gridded meteorology, an emissions factor file, and a speciation profile are further described in the next
section.

2.7

Sources Outside of the United States

The emissions from Canada and Mexico and other areas outside of the U.S. are included in these
emissions modeling sectors: othpt, othar, othafdust, othptdust, onroadcan, onroadmex, and
ptfire othna. The "oth" refers to the fact that these emissions are usually "other" than those in the NEI,
and the remaining characters provide the SMOKE source types: "pt" for point, "ar" for "area and nonroad
mobile," "afdust" for area fugitive dust (Canada only), and "ptdust" for point fugitive dust. Because
Canada and Mexico onroad mobile emissions are modeled differently from each other, they are separated
into two sectors: onroad can and onroad mex. Emissions for Mexico are based on the Inventario
Nacional de Emisiones de Mexico, 2008 projected to year 2016 (ERG, 2014a). Additional details for
these sectors can be found in the 2016vl platform specification sheets.

2.7.1 Point Sources in Canada and Mexico (othpt)

Canadian point sources were taken from the ECCC 2015 emission inventory, including upstream oil and
gas emissions, agricultural ammonia and VOC, along with point source emissions from Mexico's 2008
inventory projected to 2014 and 2018 and then interpolated to 2016. The Canadian point source inventory
is pre-speciated for the CB6 chemical mechanicsm. Also for Canada, agricultural data were originally
provided on a rotated 10-km grid for the 2016beta platform. These were smoothed out so as to avoid the
artifact of grid lines in the processed emissions. The data were monthly resolution for Canadian
agricultural and airport emissions, along with some Canadian point sources, and annual resolution for the
remainder of Canada and all of Mexico.

2.7.2 Fugitive Dust Sources in Canada (othafdust, othptdust)

Fugitive dust sources of particulate matter emissions excluding land tilling from agricultural activities,
were provided by Environment and Climate Change Canada (ECCC) as part of their 2015 emission
inventory. Different source categories were provided as gridded point sources and area (nonpoint) source

95


-------
inventories. Following consultation with ECCC, construction dust emissions in the othafdust inventory
were reduced to levels compatible with their 2010 inventory.

Gridded point source emissions resulting from land tilling due to agricultural activities were provided as
part of the ECCC 2015 emission inventory. The provided wind erosion emissions were removed. The
data were originally provided on a rotated 10-km grid for the 2016 beta platform, but these were
smoothed so as to avoid the artifact of grid lines appearing in the emissions output from SMOKE. The
othptdust emissions have a monthly resolution.

A transport fraction adjustment that reduces dust emissions based on land cover types was applied to both
point and nonpoint dust emissions, along with a meteorology-based (precipitation and snow/ice cover)
zero-out of emissions when the ground is snow covered or wet.

2.7.3	Nonpoint and Nonroad Sources in Canada and Mexico (othar)

ECCC provided year 2015 Canada province, and in some cases sub-province, resolution emissions from
for nonpoint and nonroad sources. The nonroad sources were monthly while the nonpoint and rail
emissions were annual. For Mexico, year 2016 Mexico nonpoint and nonroad inventories at the
municipio resolution were interpolated from 2014 and 2018 inventories that were projected from their
2008 inventory. All Mexico inventories were annual resolution. Canadian CMV inventories that had
been included in this sector in past modeling platforms are now included in the cmv_clc2 and cmv_c3
sectors as point sources.

2.7.4	Onroad Sources in Canada and Mexico (onroad_can, onroad_mex)

ECCC provided monthly year 2015 onroad emissions for Canada at the province resolution or sub-
province resolution depending on the province. For Mexico, monthly year 2016 onroad inventories at the
municipio resolution were used. The Mexico onroad emissions are based on MOVES-Mexico runs for
2014 and 2018 that were then interpolated to 2016

2.7.5	Fires in Canada and Mexico (ptfire_othna)

Annual point source 2016 day-specific wildland emissions for Mexico, Canada, Central America, and
Caribbean nations were developed from a combination of the Fire Inventory from NCAR (FINN) daily
fire emissions and fire data provided by Environment Canada when available. Environment Canada
emissions were used for Canada wildland fire emissions for April through November and FINN fire
emissions were used to fill in the annual gaps from January through March and December. Only CAP
emissions are provided in the ptfire othna sector inventories.

For FINN fires, listed vegetation type codes of 1 and 9 are defined as agricultural burning, all other fire
detections and assumed to be wildfires. All wildland fires that are not defined as agricultural are assumed
to be wild fires rather than prescribed. FINN fire detects less than 50 square meters (0.012 acres) are
removed from the inventory. The locations of FINN fires are geocoded from latitude and longitude to
FIPS code.

2.7.6	Ocean Chlorine

The ocean chlorine gas emission estimates are based on the build-up of molecular chlorine (Cb)
concentrations in oceanic air masses (Bullock and Brehme, 2002). Data at 36 km and 12 km resolution
were available and were not modified other than the model-species name "CHLORINE" was changed to
"CL2" to support CMAQ modeling.

96


-------
3 Emissions Modeling

The CMAQ and CAMx air quality models require hourly emissions of specific gas and particle species
for the horizontal and vertical grid cells contained within the modeled region (i.e., modeling domain). To
provide emissions in the form and format required by the model, it is necessary to "pre-process" the "raw"
emissions (i.e., emissions input to SMOKE) for the sectors described above in Section 2. In brief, the
process of emissions modeling transforms the emissions inventories from their original temporal
resolution, pollutant resolution, and spatial resolution into the hourly, speciated, gridded resolution
required by the air quality model. Emissions modeling includes temporal allocation, spatial allocation,
and pollutant speciation. Emissions modeling sometimes includes the vertical allocation of point sources,
but many air quality models also perform this task because it greatly reduces the size of the input
emissions files if the vertical layers of the sources are not included.

As seen in Section 2, the temporal resolutions of the emissions inventories input to SMOKE vary across
sectors and may be hourly, daily, monthly, or annual total emissions. The spatial resolution may be
individual point sources; totals by county (U.S.), province (Canada), or municipio (Mexico); or gridded
emissions. This section provides some basic information about the tools and data files used for emissions
modeling as part of the modeling platform. For additional details that may not be covered in this section,
see the specification sheets provided with the 2016vlplatform as many will contain additional sector-
specific information.

3.1 Emissions modeling Overview

SMOKE version 4.7 was used to process the raw emissions inventories into emissions inputs for each
modeling sector into a format compatible with CMAQ, which were then converted to CAMx. For sectors
that have plume rise, the in-line plume rise capability allows for the use of emissions files that are much
smaller than full three-dimensional gridded emissions files. For quality assurance of the emissions
modeling steps, emissions totals by specie for the entire model domain are output as reports that are then
compared to reports generated by SMOKE on the input inventories to ensure that mass is not lost or
gained during the emissions modeling process.

When preparing emissions for the air quality model, emissions for each sector are processed separately
through SMOKE, and then the final merge program (Mrggrid) is run to combine the model-ready, sector-
specific 2-D gridded emissions across sectors. The SMOKE settings in the run scripts and the data in the
SMOKE ancillary files control the approaches used by the individual SMOKE programs for each sector.
Table 3-1 summarizes the major processing steps of each platform sector with the columns as follows.

The "Spatial" column shows the spatial approach used: "point" indicates that SMOKE maps the source
from a point location (i.e., latitude and longitude) to a grid cell; "surrogates" indicates that some or all of
the sources use spatial surrogates to allocate county emissions to grid cells; and "area-to-point" indicates
that some of the sources use the SMOKE area-to-point feature to grid the emissions (further described in
Section 3.4.2).

The "Speciation" column indicates that all sectors use the SMOKE speciation step, though biogenics
speciation is done within the Tmpbeis3 program and not as a separate SMOKE step.

The "Inventory resolution" column shows the inventory temporal resolution from which SMOKE needs
to calculate hourly emissions. Note that for some sectors (e.g., onroad, beis), there is no input inventory;

97


-------
instead, activity data and emission factors are used in combination with meteorological data to compute
hourly emissions.

Finally, the "plume rise" column indicates the sectors for which the "in-line" approach is used. These
sectors are the only ones with emissions in aloft layers based on plume rise. The term "in-line" means
that the plume rise calculations are done inside of the air quality model instead of being computed by
SMOKE. The air quality model computes the plume rise using stack parameters, the Briggs algorithm,
and the hourly emissions in the SMOKE output files for each emissions sector. The height of the plume
rise determines the model layers into which the emissions are placed. The plume top and bottom are
computed, along with the plumes' distributions into the vertical layers that the plumes intersect. The
pressure difference across each layer divided by the pressure difference across the entire plume is used as
a weighting factor to assign the emissions to layers. This approach gives plume fractions by layer and
source. The othpt sector has only "in-line" emissions, meaning that all of the emissions are treated as
elevated sources and there are no emissions for those sectors in the two-dimensional, layer-1 files created
by SMOKE. Other inline-only sectors are: cmv_c3, ptegu, ptfire, ptfire othna, ptagfire. Day-specific
point fire emissions are treated differently in CMAQ. After plume rise is applied, there are emissions in
every layer from the ground up to the top of the plume. Note that SMOKE has the option of grouping
sources so that they are treated as a single stack when computing plume rise. For the modeling cases
discussed in this document, no grouping was performed because grouping combined with "in-line"
processing will not give identical results as "offline" processing (i.e., when SMOKE creates 3-
dimensional files). This occurs when stacks with different stack parameters or latitude and longitudes are
grouped, thereby changing the parameters of one or more sources. The most straightforward way to get
the same results between in-line and offline is to avoid the use of stack grouping.

Table 3-1. Key emissions modeling steps by sector.

Platform sector

Spatial

Speciation

Inventory
resolution

Plume rise

afdust adj

Surrogates

Yes

annual



afdust ak adj
(36US3 only)

Surrogates

Yes

annual



ag

Surrogates

Yes

monthly



airports

Point

Yes

annual

None

beis

Pre-gridded
land use

in BEIS3 .61

computed hourly



cmv clc2

Surrogates

Yes

annual



cmv c3

Point

Yes

annual

in-line

nonpt

Surrogates &
area-to-point

Yes

annual



nonroad

Surrogates &
area-to-point

Yes

monthly



np oilgas

Surrogates

Yes

annual



onroad

Surrogates

Yes

monthly activity,
computed hourly



onroadcaadj

Surrogates

Yes

monthly activity,
computed hourly



onroad nonconus
(36US3 only)

Surrogates

Yes

monthly activity,
computed hourly



onroad can

Surrogates

Yes

monthly



98


-------
Platform sector

Spatial

Speciation

Inventory
resolution

Plume rise

onroad mex

Surrogates

Yes

monthly



othafdust ad]

Surrogates

Yes

annual



othar

Surrogates

Yes

annual &
monthly



othpt

Point

Yes

annual &
monthly

in-line

othptdust ad]

Point

Yes

monthly

None

ptagfire

Point

Yes

daily

in-line

pt oilgas

Point

Yes

annual

in-line

ptegu

Point

Yes

daily & hourly

in-line

ptfire

Point

Yes

daily

in-line

ptfire othna

Point

Yes

daily

in-line

ptnonipm

Point

Yes

annual

in-line

rail

Surrogates

Yes

annual



rwc

Surrogates

Yes

annual



Biogenic emissions can be modeled two different ways in the CMAQ model. The BEIS model in SMOKE
can produce gridded biogenic emissions that are then included in the gridded CMAQ-ready emissions
inputs, or alternatively, CMAQ can be configured to create "in-line" biogenic emissions within CMAQ
itself. For this platform, biogenic emissions were processed in SMOKE and included in the gridded
CMAQ-ready emissions. When CAMx is the targeted air quality modeling, BEIS is run within SMOKE
and the resulting emissions are included with the ground-level emissions input to CAMx.

SMOKE has the option of grouping sources so that they are treated as a single stack when computing
plume rise. For this platform, no grouping was performed because grouping combined with "in-line"
processing will not give identical results as "offline" processing (i.e., when SMOKE creates 3-
dimensional files). This occurs when stacks with different stack parameters or latitudes/longitudes are
grouped, thereby changing the parameters of one or more sources. The most straightforward way to get
the same results between in-line and offline is to avoid the use of grouping.

SMOKE was run for two modeling domains: a 36-km resolution CONtinental United States "CONUS"
modeling domain (36US3), and the 12-km resolution domain. 12US2. More specifically, SMOKE was
run on the 12US1 domain and emissions were extracted from 12US1 data files to create 12US2 emission.
The domains are shown in Figure 3-1. All grids use a Lambert-Conformal projection, with Alpha = 33°,
Beta = 45° and Gamma = -97°, with a center of X = -97° and Y = 40°. Table 3-2 describes the grids for
the three domains.

99


-------
Table 3-2. Descriptions of the platform grids

Common
Name

Grid
Cell Size

Description
(see Figure 3-1)

Grid name

Parameters listed in SMOKE grid
description (GRIDDESC) file:
projection name, xorig, yorig, xcell,
ycell, ncols, nrows, nthik

Continental
36km grid

36 km

Entire conterminous
US, almost all of
Mexico, most of
Canada (south of

60°N)

36US3

'LAM 40N97W', -2952000, -2772000,
36.D3, 36.D3, 172, 148, 1

Continental
12km grid

12 km

Entire conterminous
US plus some of
Mexico/Canada

12US1_45 9X299

'LAM 40N97W', -2556000, -1728000,
12.D3, 12.D3, 459,299, 1

US 12 km or

"smaller"

CONUS-12

12 km

Smaller 12km
CONUS plus some of
Mexico/Canada

12US2

'LAM 40N97W', -2412000 , -
1620000, 12.D3, 12.D3, 396, 246, 1

Figure 3-1. Air quality modeling domains

100


-------
3.2 Chemical Speciation

The emissions modeling step for chemical speciation creates the "model species" needed by the air
quality model for a specific chemical mechanism. These model species are either individual chemical
compounds (i.e., "explicit species") or groups of species (i.e., "lumped species"). The chemical
mechanism used for the 2016 platform is the CB6 mechanism (Yarwood, 2010). We used a particular
version of CB6 that we refer to as "CMAQ CB6" that breaks out naphthalene from model species XYL,
resulting in explicit model species NAPH and XYLMN instead of XYL and uses SOAALK. This
platform generates the PM2.5 model species associated with the CMAQ Aerosol Module version 6 (AE6).
Table 3-3 lists the model species produced by SMOKE in the platform used for this study. Updates to
species assignments for CB05 and CB6 were made for the 2014v7.1 platform and are described in
Appendix A.

101


-------
Table 3-3. Emission model species produced for CB6 for CMAQ

Inventory Pollutant

Model Species

Model species description

Cl2

CL2

Atomic gas-phase chlorine

HC1

HCL

Hydrogen Chloride (hydrochloric acid) gas

CO

CO

Carbon monoxide

NOx

NO

Nitrogen oxide



N02

Nitrogen dioxide



HONO

Nitrous acid

S02

S02

Sulfur dioxide



SULF

Sulfuric acid vapor

nh,

NH3

Ammonia



NH3 FERT

Ammonia from fertilizer

voc

ACET

Acetone



ALD2

Acetaldehyde



ALDX

Propionaldehyde and higher aldehydes



BENZ

Benzene (not part of CB05)



CH4

Methane



ETH

Ethene



ETHA

Ethane



ETHY

Ethyne



ETOH

Ethanol



FORM

Formaldehyde



IOLE

Internal olefin carbon bond (R-C=C-R)



ISOP

Isoprene



KET

Ketone Groups



MEOH

Methanol



NAPH

Naphthalene



NVOL

Non-volatile compounds



OLE

Terminal olefin carbon bond (R-C=C)



PAR

Paraffin carbon bond



PRPA

Propane



SESQ

Sequiterpenes (from biogenics only)



SOAALK

Secondary Organic Aerosol (SOA) tracer



TERP

Terpenes (from biogenics only)



TOL

Toluene and other monoalkyl aromatics



UNR

Unreactive



XYLMN

Xylene and other polyalkyl aromatics, minus
naphthalene

Naphthalene

NAPH

Naphthalene from inventory

Benzene

BENZ

Benzene from the inventory

Acetaldehyde

ALD2

Acetaldehyde from inventory

Formaldehyde

FORM

Formaldehyde from inventory

Methanol

MEOH

Methanol from inventory

PM10

PMC

Coarse PM >2.5 microns and <10 microns

PM2.5

PEC

Particulate elemental carbon <2.5 microns



PN03

Particulate nitrate <2.5 microns



POC

Particulate organic carbon (carbon only) <2.5 microns



PS04

Particulate Sulfate <2.5 microns



PAL

Aluminum



PCA

Calcium

102


-------
Inventory Pollutant

Model Species

Model species description



PCL

Chloride



PFE

Iron



PK

Potassium



PH20

Water



PMG

Magnesium



PMN

Manganese



PMOTHR

PM2.5 not in other AE6 species



PNA

Sodium



PNCOM

Non-carbon organic matter



PNH4

Ammonium



PSI

Silica



PTI

Titanium

Sea-salt species (non -

PCL

Particulate chloride

anthropogenic)20

PNA

Particulate sodium

The TOG and PM2.5 speciation factors that are the basis of the chemical speciation approach were
developed from the SPECIATE 4.5 database (https://www.epa.gov/air-emissions-modeling/speciate-2).
which is the EPA's repository of TOG and PM speciation profiles of air pollution sources. The
SPECIATE database development and maintenance is a collaboration involving the EPA's Office of
Research and Development (ORD), Office of Transportation and Air Quality (OTAQ), and the Office of
Air Quality Planning and Standards (OAQPS), in cooperation with Environment Canada (EPA, 2016).
The SPECIATE database contains speciation profiles for TOG, speciated into individual chemical
compounds, VOC-to-TOG conversion factors associated with the TOG profiles, and speciation profiles
for PM2.5.

Some key features and recent updates to speciation from previous platforms include the following:

•	VOC speciation profile cross reference assignments for point and nonpoint oil and gas sources
were updated to (1) make corrections to the 201 lv6.3 cross references, (2) use new and revised
profiles that were added to SPECIATE4.5 and (3) account for the portion of VOC estimated to
come from flares, based on data from the Oil and Gas estimation tool used to estimate emissions
for the NEI. The new/revised profiles included oil and gas operations in specific regions of the
country and a national profile for natural gas flares;

•	the Western Regional Air Partnership (WRAP) speciation profiles used for the np oilgas sector
are the SPECIATE4.5 revised versions (profiles with "_R" in the profile code);

•	the VOC and PM speciation process for nonroad mobile has been updated - profiles are now
assigned within MOVES2014b which outputs the emissions with those assignments; also the
nonroad profiles themselves were updated;

•	VOC and PM speciation for onroad mobile sources occurs within MOVES2014a except for brake
and tirewear PM speciation which occurs in SMOKE;

•	speciation for onroad mobile sources in Mexico is done within MOVES and is more consistent
with that used in the United States;

20 These emissions are created outside of SMOKE.

103


-------
•	the PM speciation profile for C3 ships in the US and Canada was updated to a new profile,
5675AE6; and

•	As with previous platforms, some Canadian point source inventories are provided from
Environment Canada as pre-speciated emissions; however for the 2015 inventory, not all CB6-
CMAQ species were provided; missing species were supplemented by speciating VOC which was
provided separately.

Speciation profiles and cross-references for this study platform are available in the SMOKE input files for
the 2016 platform. Emissions of VOC and PM2.5 emissions by county, sector and profile for all sectors
other than onroad mobile can be found in the sector summaries for the case. Totals of each model species
by state and sector can be found in the state-sector totals workbook for this case.

3.2.1 VOC speciation

The speciation of VOC includes HAP emissions from the 2014NEIv2 in the speciation process. Instead
of speciating VOC to generate all of the species listed in Table 3-3, emissions of five specific HAPs:
naphthalene, benzene, acetaldehyde, formaldehyde and methanol (collectively known as "NBAFM") from
the NEI were "integrated" with the NEI VOC. The integration combines these HAPs with the VOC in a
way that does not double count emissions and uses the HAP inventory directly in the speciation process.
The basic process is to subtract the specified HAPs emissions mass from the VOC emissions mass, and to
then use a special "integrated" profile to speciate the remainder of VOC to the model species excluding
the specific HAPs. The EPA believes that the HAP emissions in the NEI are often more representative of
emissions than HAP emissions generated via VOC speciation, although this varies by sector.

The NBAFM HAPs were chosen for integration because they are the only explicit VOC HAPs in the
CMAQ version 5.2. Explicit means that they are not lumped chemical groups like PAR, IOLE and
several other CB6 model species. These "explicit VOC HAPs" are model species that participate in the
modeled chemistry using the CB6 chemical mechanism. The use of inventory HAP emissions along with
VOC is called "HAP-CAP integration."

The integration of HAP VOC with VOC is a feature available in SMOKE for all inventory formats,
including PTDAY (the format used for the ptfire and ptagfire sectors). The ability to use integration with
the PTDAY format was made available in the version of SMOKE used for the 2014v7.1 platform, but this
new feature is not used for the 2016 platform because the ptfire and ptagfire inventories for 2016 do not
include HAPs. SMOKE allows the user to specify the particular HAPs to integrate via the INVTABLE.
This is done by setting the "VOC or TOG component" field to "V" for all HAP pollutants chosen for
integration. SMOKE allows the user to also choose the particular sources to integrate via the
NHAPEXCLUDE file (which actually provides the sources to be excluded from integration21). For the
"integrated" sources, SMOKE subtracts the "integrated" HAPs from the VOC (at the source level) to
compute emissions for the new pollutant "NONHAPVOC." The user provides NONHAPVOC-to-
NONHAPTOG factors and NONHAPTOG speciation profiles.22 SMOKE computes NONHAPTOG and
then applies the speciation profiles to allocate the NONHAPTOG to the other air quality model VOC
species not including the integrated HAPs. After determining if a sector is to be integrated, if all sources

21	Since SMOKE version 3.7, the options to specify sources for integration are expanded so that a user can specify the
particular sources to include or exclude from integration, and there are settings to include or exclude all sources within a sector.
In addition, the error checking is significantly stricter for integrated sources. If a source is supposed to be integrated, but it is
missing NBAFM or VOC, SMOKE will now raise an error.

22	These ratios and profiles are typically generated from the Speciation Tool when it is run with integration of a specified list of
pollutants, for example NBAFM.

104


-------
have the appropriate HAP emissions, then the sector is considered fully integrated and does not need a
NHAPEXCLUDE file. If, on the other hand, certain sources do not have the necessary HAPs, then an
NHAPEXCLUDE file must be provided based on the evaluation of each source's pollutant mix. The
EPA considered CAP-HAP integration for all sectors in determining whether sectors would have full, no
or partial integration (see Figure 3-2. Process of integrating NBAFM with VOC for use in VOC
Speciation). For sectors with partial integration, all sources are integrated other than those that have
either the sum of NBAFM > VOC or the sum of NBAFM = 0.

In this platform, we create NBAFM species from the no-integrate source VOC emissions using speciation
profiles. Figure 3-2 illustrates the integrate and no-integrate processes for U.S. Sources. Since Canada
and Mexico inventories do not contain HAPs, we use the approach of generating the HAPs via speciation,
except for Mexico onroad mobile sources where emissions for integrate HAPs were available.

It should be noted that even though NBAFM were removed from the SPECIATE profiles used to create
the GSPRO for both the NONHAPTOG and no-integrate TOG profiles, there still may be small fractions
for "BENZ", "FORM", "ALD2", and "MEOH" present. This is because these model species may have
come from species in SPECIATE that are mixtures. The quantity of these model species is expected to be
very small compared to the BAFM in the NEI. There are no NONHAPTOG profiles that produce
"NAPH."

In SMOKE, the INVTABLE allows the user to specify the particular HAPs to integrate. Two different
INVTABLE files are used for different sectors of the platform. For sectors that had no integration across
the entire sector (see Table 3-4), EPA created a "no HAP use" INVTABLE in which the "KEEP" flag is
set to "N" for NBAFM pollutants. Thus, any NBAFM pollutants in the inventory input into SMOKE are
automatically dropped. This approach both avoids double-counting of these species and assumes that the
VOC speciation is the best available approach for these species for sectors using this approach. The
second INVTABLE, used for sectors in which one or more sources are integrated, causes SMOKE to keep
the inventory NBAFM pollutants and indicates that they are to be integrated with VOC. This is done by
setting the "VOC or TOG component" field to "V" for all five HAP pollutants. Note for the onroad
sector, "full integration" includes the integration of benzene, 1,3 butadiene, formaldehyde, acetaldehyde,
naphthalene, acrolein, ethyl benzene, 2,2,4-Trimethylpentane, hexane, propionaldehyde, styrene, toluene,
xylene, and methyl tert-butyl ether (MTBE).

105


-------
Figure 3-2. Process of integrating NBAFM with VOC for use in VOC Speciation

CMAQ-CB6 species

Table 3-4. Integration status of naphthalene, benzene, acetaldehyde, formaldehyde and methanol

(NBAFM) for each platform sector

Platform
Sector

Approach for Integrating NEI emissions of Naphthalene (N), Benzene (B),
Acetaldehyde (A), Formaldehyde (F) and Methanol (M)

ptegu

No integration, create NBAFM from VOC speciation

ptnonipm

No integration, create NBAFM from VOC speciation

ptfire

No integration, no NBAFM in inventory, create NBAFM from VOC speciation

ptfire othna

No integration, no NBAFM in inventory, create NBAFM from VOC speciation

ptagfire

No integration, no NBAFM in inventory, create NBAFM from VOC speciation

airport

No integration, create NBAFM from VOC speciation

ag

Partial integration (NBAFM)

afdust

N/A - sector contains no VOC

beis

N/A - sector contains no inventory pollutant "VOC"; but rather specific VOC species

cmv clc2

Full integration (NBAFM)

cmv c3

Full integration (NBAFM)

rail

Partial integration (NBAFM)

nonpt

Partial integration (NBAFM)

nonroad

Full integration (NBAFM in California, internal to MOVES elsewhere)

np oilgas

Partial integration (NBAFM)

othpt

No integration, no NBAFM in inventory, create NBAFM from VOC speciation

pt oilgas

No integration, create NBAFM from VOC speciation

rwc

Partial integration (NBAFM)

106


-------
Platform
Sector

Approach for Integrating NEI emissions of Naphthalene (N), Benzene (B),
Acetaldehyde (A), Formaldehyde (F) and Methanol (M)

onroad

Full integration (internal to MOVES); however, MOVES2014a speciation was CB6-
CAMx, not CB6-CMAQ, so post-SMOKE emissions were converted to CB6-CMAQ

onroad can

No integration, no NBAFM in inventory, create NBAFM from speciation

onroadmex

Full integration (internal to MOVES-Mexico); however, MOVES-MEXICO speciation
was CB6-CAMx, not CB6-CMAQ, so post-SMOKE emissions were converted to CB6-
CMAQ

othafdust

N/A - sector contains no VOC

othptdust

N/A - sector contains no VOC

othar

No integration, no NBAFM in inventory, create NBAFM from VOC speciation

Integration for the mobile sources estimated from MOVES (onroad and nonroad sectors, other than for
California) is done differently. Briefly there are three major differences: 1) for these sources integration
is done using more than just NBAFM, 2) all sources from the MOVES model are integrated, and 3)
integration is done fully or partially within MOVES. For onroad mobile, speciation is done fully within
MOVES2014a such that the MOVES model outputs emission factors for individual VOC model species
along with the HAPs. This requires MOVES to be run for a specific chemical mechanism. MOVES was
run for the CB6-CAMx mechanism rather than CB6-CMAQ, so post-SMOKE onroad emissions were
converted to CB6-CMAQ. More specifically, the CB6-CAMx mechanism excludes XYLMN, NAPH,
and SOAALK. After SMOKE processing, we converted the onroad and onroadmex emissions to CB6-
CMAQ as follows:

•	XYLMN = XYL[1]-0.966*NAPHTHALENE[1]

•	PAR = PAR[l]-0.00001 *NAPHTHALENE[1]

•	SOAALK = 0.108*PAR[1]

For nonroad mobile, speciation is partially done within MOVES such that it does not need to be run for a
specific chemical mechanism. For nonroad, MOVES outputs emissions of HAPs and NONHAPTOG are
+split by speciation profile. Taking into account that integrated species were subtracted out by MOVES
already, the appropriate speciation profiles are then applied in SMOKE to get the VOC model species.
HAP integration for nonroad uses the same additional HAPs and ethanol as for onroad.

3.2.1.1 County specific profile combinations

SMOKE can compute speciation profiles from mixtures of other profiles in user-specified proportions via
two different methods. The first method, which uses a GSPROCOMBO file, has been in use since the
2005 platform; the second method (GSPRO with fraction) was used for the first time in the 2014v7.0
platform. The GSPRO COMBO method uses profile combinations specified in the GSPRO COMBO
ancillary file by pollutant (which can include emissions mode, e.g., EXH VOC), state and county (i.e.,
state/county FIPS code) and time period (i.e., month). Different GSPRO COMBO files can be used by
sector, allowing for different combinations to be used for different sectors; but within a sector, different
profiles cannot be applied based on SCC. The GSREF file indicates that a specific source uses a
combination file with the profile code "COMBO." SMOKE computes the resultant profile using the
fraction of each specific profile assigned by county, month and pollutant.

In previous platforms, the GSPRO COMBO feature was used to speciate nonroad mobile and gasoline-
related stationary sources that use fuels with varying ethanol content. In these cases, the speciation
profiles require different combinations of gasoline profiles, e.g., 0% ethanol (E0) and 10% ethanol (E10)
profiles. Since the ethanol content varied spatially (e.g., by state or county), temporally (e.g., by month),
and by modeling year (future years have more ethanol), the GSPRO COMBO feature allowed

107


-------
combinations to be specified at various levels for different years. The GSPROCOMBO is no longer
needed for nonroad sources outside of California because nonroad emissions within MOVES have the
speciation profiles built into the results, so there is no need to assign them via the GSREF or
GSPRO COMBO feature. For the 2016 alpha platform, GSPRO COMBO is still used for nonroad
sources in California and for certain gasoline-related stationary sources nationwide. The fractions
combining the E0 and E10 profiles are based on year 2010 regional fuels and do not vary by month.
GSPRO COMBO is not needed for inventory years after 2016, because the vast majority of fuel is
projected to be E10 in future years.

Starting with the 2016v7.2 beta and regional haze platforms, a GSPRO COMBO is used to specify a mix
of E0 and E10 fuels in Canada. ECCC provided percentages of ethanol use by province, and these were
converted into E0 and E10 splits. For example, Alberta has 4.91% ethanol in its fuel, so we applied a mix
of 49.1% E10 profiles (4.91% times 10, since 10% ethanol would mean 100% E10), and 50.9% E0 fuel.
Ethanol splits for all provinces in Canada are listed in Table 3-5. The Canadian onroad inventory includes
four distinct FIPS codes in Ontario, allowing for application of different E0/E10 splits in Southern
Ontario versus Northern Ontario. In Mexico, only E0 profiles are used.

Table 3-5. Ethanol percentages by volume by Canadian province

Province

Ethanol % by volume (E10 = 10%)

Alberta

4.91%

British Columbia

5.57%

Manitoba

9.12%

New Brunswick

4.75%

Newfoundland & Labrador

0.00%

Nova Scotia

0.00%

NW Territories

0.00%

Nunavut

0.00%

Ontario (Northern)

0.00%

Ontari o ( S outhern)

7.93%

Prince Edward Island

0.00%

Quebec

3.36%

Saskatchewan

7.73%

Yukon

0.00%

A new method to combine multiple profiles became available in SMOKE4.5. It allows multiple profiles
to be combined by pollutant, state and county (i.e., state/county FIPS code) and SCC. This was used
specifically for the oil and gas sectors (pt oilgas and np oilgas) because SCCs include both controlled
and uncontrolled oil and gas operations which use different profiles.

3.2.1.2 Additional sector specific considerations for integrating HAP
emissions from inventories into speciation

The decision to integrate HAPs into the speciation was made on a sector by sector basis. For some
sectors, there is no integration and VOC is speciated directly; for some sectors, there is full integration
meaning all sources are integrated; and for other sectors, there is partial integration, meaning some
sources are not integrated and other sources are integrated. The integrated HAPs are either NBAFM or, in
the case of MOVES (onroad, nonroad, and MOVES-Mexico), a larger set of HAPs plus ethanol are
integrated. Table 3-4 above summarizes the integration method for each platform sector.

108


-------
For the rail sector, the EPA integrated NBAFM for most sources. Some SCCs had zero BAFM and,
therefore, they were not integrated. These were SCCs provided by states for which EPA did not do HAP
augmentation (2285002008, 2285002009 and 2285002010) because EPA does not create emissions for
these SCCs. The VOC for these sources sum to 272 tons, and most of the mass is in California (189 tons)
and Washington state (62 tons).

Speciation for the onroad sector is unique. First, SMOKE-MOVES is used to create emissions for these
sectors and both the MEPROC and INVTABLE files are involved in controlling which pollutants are
processed. Second, the speciation occurs within MOVES itself, not within SMOKE. The advantage of
using MOVES to speciate VOC is that during the internal calculation of MOVES, the model has complete
information on the characteristics of the fleet and fuels (e.g., model year, ethanol content, process, etc.),
thereby allowing it to more accurately make use of specific speciation profiles. This means that MOVES
produces emission factor tables that include inventory pollutants (e.g., TOG) and model-ready species
(e.g., PAR, OLE, etc).23 SMOKE essentially calculates the model-ready species by using the appropriate
emission factor without further speciation.24 Third, MOVES' internal speciation uses full integration of
an extended list of HAPs beyond NBAFM (called "M-profiles"). The M-profiles integration is very
similar to NBAFM integration explained above except that the integration calculation (see Figure 3-2.
Process of integrating NBAFM with VOC for use in VOC Speciation) is performed on emissions factors
instead of on emissions, and a much larger set of pollutants are integrated besides NBAFM. The list of
integrated pollutants is described in Table 3-6. An additional run of the Speciation Tool was necessary to
create the M-profiles that were then loaded into the MOVES default database. Fourth, for California, the
EPA applied adjustment factors to SMOKE-MOVES to produce California adjusted model-ready files.
By applying the ratios through SMOKE-MOVES, the CARB inventories are essentially speciated to
match EPA estimated speciation. This resulted in changes to the VOC HAPs from what CARB submitted
to the EPA. Finally, MOVES speciation used the CAMx version of CB6 which does not split out
naphthalene.

Table 3-6. MOVES integrated species in M-profiles

MOVES ID

Pollutant Name

5

Methane (CH4)

20

Benzene

21

Ethanol

22

MTBE

24

1,3-Butadiene

25

Formaldehyde

26

Acetaldehyde

27

Acrolein

40

2,2,4-Trimethylpentane

41

Ethyl Benzene

42

Hexane

43

Propionaldehyde

23	Because the EF table has the speciation "baked" into the factors, all counties that are in the county group (i.e., are mapped to
that representative county) will have the same speciation.

24	For more details on the use of model-ready EF, see the SMOKE 3.7 documentation:
https://www.cmascenter.Org/smoke/documentation/3.7/html/.

109


-------
MOVES ID

Pollutant Name

44

Styrene

45

Toluene

46

Xylene

185

Naphthalene gas

For the nonroad sector, all sources are integrated using the same list of integrated pollutants as shown in
Table 3-6. Outside of California, the integration calculations are performed within MOVES. For
California, integration calculations are handled by SMOKE. The CARB-based nonroad inventory
includes VOC HAP estimates for all sources, so every source in California was integrated as well. Some
sources in the original CARB inventory had lower VOC emissions compared to sum of all VOC HAPs.
For those sources, VOC was augmented to be equal to the VOC HAP sum, ensuring that every source in
California could be integrated. The CARB-based nonroad data includes exhaust and evaporative mode-
specific data for VOC, but, does not contain refueling.

MOVES-MEXICO for onroad used the same speciation approach as for the U.S. in that the larger list of
species shown in Table 3-6 was used. However, MOVES-MEXICO used CB6-CAMx, not CB6-CMAQ,
so post-SMOKE we converted the emissions to CB6-CMAQ as follows:

•	XYLMN = XYL[1]-0.966*NAPHTHALENE[1]

•	PAR = PAR[1]-0.00001*NAPHTHALENE[1]

•	SOAALK = 0.108*PAR[1]

For most sources in the rwc sector, the VOC emissions were greater than or equal to NBAFM, and
NBAFM was not zero, so those sources were integrated, although a few specific sources that did not meet
these criteria could not be integrated. In all cases, these sources have SCC= 2104008400 (pellet stoves),
and NBAFM > VOC, but not by a significant amount. This results from the sum of NBAFM emission
factors exceeding the VOC emission factor. In total, the no-integrate rwc sector sources sum to 4.4 tons
VOC and 66 tons of NBAFM. Since for the NATA case the NBAFM are used from the inventory, these
no-integrate NBAFM emissions were used in the speciation.

For the nonpt sector, sources for which VOC emissions were greater than or equal to NBAFM, and
NBAFM was not zero, were integrated. There is a substantial amount of mass in the nonpt sector that is
not integrated: 731,000 tons which is about 20% of the VOC in that sector. It is likely that there would be
sources in nonpt that are not integrated because the emission source is not expected to have NBAFM. In
fact, 390,000 tons of the no-integrate VOC have no NBAFM in the speciation profiles used for these no-
integrate sources. Of the portion of no-integrate VOC with NBAFM there is 3,900 tons NBAFM in the
profiles (that are dropped from the profiles per the procedure in Figure 3-2. Process of integrating
NBAFM with VOC for use in VOC Speciation) for these no-integrate sources.

For the biog sector, the speciation profiles used by BEIS are not included in SPECIATE. BEIS3.61
includes the species (SESQ) that is mapped to the BEIS model species SESQT (Sesquiterpenes). The
profile code associated with BEIS3.61 for use with CB05 is "B10C5," while the profile for use with CB6
is "B10C6." The main difference between the profiles is the explicit treatment of acetone emissions in
B10C6.

110


-------
3.2.1.3 Oil and gas related speciation profiles

Most of the recently added VOC profiles from SPECIATE4.5 (listed in Appendix B) are in the oil and gas
sector. A new national flare profile, FLR99, Natural Gas Flare Profile with DRE >98% was developed
from a Flare Test study and used in the v7.0 platform. For the oil and gas sources in the np oilgas and
pt oilgas sectors, several counties were assigned to newly available basin or area-specific profiles in
SPECIATE4.5 that account for measured or modeled, from measured compositions specific to a particular
region of the country. In the 2011 platform, the only county-specific profiles were for the WRAP, but in
the 2014 and 2016 platforms, several new profiles were added for other parts of the country. The 2016
platform uses the latest version of the WRAP profiles. These profiles are denoted with an _R suffix, and
reflect newer data and corrections to older WRAP profiles. All WRAP profile codes were renamed to
include an "_R" to distinguish between the previous set of profiles (even those that did not change). For
the Uintah basin and Denver-Julesburg Basin, Colorado, more updated profiles were used instead of the
WRAP profiles. Table 3-7 lists the region-specific profiles assigned to particular counties or groups of
counties. Although this platform increases the use of regional profiles, many counties still rely on the
national profiles. A minor change in 2016vl was to use county-specific profile assignments from SCC
2310121700 for the SCCs 2310021500, 2310421700 in Pennsylvania.

In addition to region-specific assignments, multiple profiles were assigned to particular county/SCC
combinations using the SMOKE feature discussed in 3.2.1.1. Oil and gas SCCs for associated gas,
condensate tanks, crude oil tanks, dehydrators, liquids unloading and well completions represent the total
VOC from the process, including the portions of process that may be flared or directed to a reboiler. For
example, SCC 2310021400 (gas well dehydrators) consists of process, reboiler, and/or flaring
emissions. There are not separate SCCs for the flared portion of the process or the reboiler. However, the
VOC associated with these three portions can have very different speciation profiles. Therefore, it is
necessary to have an estimate of the amount of VOC from each of the portions (process, flare, reboiler) so
that the appropriate speciation profiles can be applied to each portion. The Nonpoint Oil and Gas
Emission Estimation Tool generates an intermediate file which provides flare, non-flare (process), and
reboiler (for dehydrators) emissions for six source categories that have flare emissions: by county FIPS
and SCC code for the U.S. From these emissions we can compute the fraction of the emissions to assign
to each profile. These fractions can vary by county FIPS, because they depend on the level of controls,
which is an input to the Speciation Tool.

Table 3-7. Basin/Region-specific profiles for oil and gas

Profile
Code

Description

Region (if not in
the profile name)

DJVNT R

Denver-Julesburg Basin Produced Gas Composition from Non-
CBM Gas Wells



PNC01 R

Piceance Basin Produced Gas Composition from Non-CBM Gas
Wells



PNC02 R

Piceance Basin Produced Gas Composition from Oil Wells



PNC03 R

Piceance Basin Flash Gas Composition for Condensate Tank



PNCDH

Piceance Basin, Glycol Dehydrator



PRBCB R

Powder River Basin Produced Gas Composition from CBM Wells



PRBCO R

Powder River Basin Produced Gas Composition from Non-CBM
Wells



PRM01 R

Permian Basin Produced Gas Composition for Non-CBM Wells



SSJCB R

South San Juan Basin Produced Gas Composition from CBM
Wells



111


-------
Profile
Code

Description

Region (if not in
the profile name)

SSJCO R

South San Juan Basin Produced Gas Composition from Non-CBM
Gas Wells



SWFLA R

SW Wyoming Basin Flash Gas Composition for Condensate
Tanks



SWVNT R

SW Wyoming Basin Produced Gas Composition from Non-CBM
Wells



UNT01 R

Uinta Basin Produced Gas Composition from CBM Wells



WRBCO R

Wind River Basin Produced Gagres Composition from Non-CBM
Gas Wells



95087a

Oil and Gas - Composite - Oil Field - Oil Tank Battery Vent Gas

East Texas

95109a

Oil and Gas - Composite - Oil Field - Condensate Tank Battery
Vent Gas

East Texas

95417

Uinta Basin, Untreated Natural Gas



95418

Uinta Basin, Condensate Tank Natural Gas



95419

Uinta Basin, Oil Tank Natural Gas



95420

Uinta Basin, Glycol Dehydrator



95398

Composite Profile - Oil and Natural Gas Production - Condensate
Tanks

Denver-Jule sburg
Basin

95399

Composite Profile - Oil Field - Wells

State of California

95400

Composite Profile - Oil Field - Tanks

State of California

95403

Composite Profile - Gas Wells

San Joaquin Basin

3.2.1.4 Mobile source related VOC speciation profiles

The VOC speciation approach for mobile source and mobile source-related source categories is
customized to account for the impact of fuels and engine type and technologies. The impact of fuels also
affects the parts of the nonpt and ptnonipm sectors that are related to mobile sources such as portable fuel
containers and gasoline distribution.

The VOC speciation profiles for the nonroad sector other than for California are listed in Table 3-8. They
include new profiles (i.e., those that begin with "953") for 2-stroke and 4-stroke gasoline engines running
on EO and E10 and compression ignition engines with different technologies developed from recent EPA
test programs, which also supported the updated toxics emission factor in MOVES2014a (Reichle, 2015
and EPA, 2015b). California nonroad source profiles are presented in Table 3-9.

Table 3-8. TOG MOVES-SMOKE Speciation for nonroad emissions in MOVES2014a used for the

2016 Platform

Profile

Profile Description

Engine
Type

Engine
Technology

Engine
Size

Horse-
power
category

Fuel

Fuel
Sub-
type

Emission
Process

95327

SI 2-stroke E0

SI 2-stroke

all

All

all

Gasoline

E0

exhaust

95328

SI 2-stroke E10

SI 2-stroke

all

All

all

Gasoline

E10

exhaust

95329

SI 4-stroke E0

SI 4-stroke

all

All

all

Gasoline

E0

exhaust

95330

SI 4-stroke E10

SI 4-stroke

all

All

all

Gasoline

E10

exhaust

95331

CI Pre-Tier 1

CI

Pre-Tier 1

All

all

Diesel

all

exhaust

112


-------
Profile

Profile Description

Engine
Type

Engine
Technology

Engine
Size

Horse-
power
category

Fuel

Fuel
Sub-
type

Emission
Process

95332

CI Tier 1

CI

Tier 1

All

all

Diesel

all

exhaust

95333

CI Tier 2

CI

Tier 2 and 3

all

all

Diesel

all

exhaust

95333

CI Tier 2

CI

Tier 4

<56 kW
(75 hp)

S

Diesel

all

exhaust

8775

ACES Phase 1 Diesel
Onroad

CI Tier 4

Tier 4

>=56 kW
(75 hp)

L

Diesel

all

exhaust

8753

E0 Evap

SI

all

all

all

Gasoline

E0

evaporative

8754

E10 Evap

SI

all

all

all

Gasoline

E10

evaporative

8766

E0 evap permeation

SI

all

all

all

Gasoline

E0

permeation

8769

E10 evap permeation

SI

all

all

all

Gasoline

E10

permeation

8869

E0 Headspace

SI

all

all

all

Gasoline

E0

headspace

8870

E10 Headspace

SI

all

all

all

Gasoline

E10

headspace

1001

CNG Exhaust

All

all

all

all

CNG

all

exhaust

8860

LPG exhaust

All

all

all

all

LPG

all

exhaust

Speciation profiles for VOC in the nonroad sector account for the ethanol content of fuels across years. A
description of the actual fuel formulations for 2014 can be found in the 2014NEIv2 TSD. For previous
platforms, the EPA used "COMBO" profiles to model combinations of profiles for E0 and E10 fuel use,
but beginning with 2014v7.0 platform, the appropriate allocation of E0 and E10 fuels is done by MOVES.

Combination profiles reflecting a combination of E10 and E0 fuel use are still used for sources upstream
of mobile sources such as portable fuel containers (PFCs) and other fuel distribution operations associated
with the transfer of fuel from bulk terminals to pumps (BTP), which are in the nonpt sector. They are also
used for California nonroad sources. For these sources, ethanol may be mixed into the fuels, in which
case speciation would change across years. The speciation changes from fuels in the ptnonipm sector
include BTP distribution operations inventoried as point sources. Refinery-to-bulk terminal (RBT) fuel
distribution and bulk plant storage (BPS) speciation does not change across the modeling cases because
this is considered upstream from the introduction of ethanol into the fuel. The mapping of fuel
distribution SCCs to PFC, BTP, BPS, and RBT emissions categories can be found in Appendix C.

Table 3-9 summarizes the different profiles utilized for the fuel-related sources in each of the sectors for
2016. The term "COMBO" indicates that a combination of the profiles listed was used to speciate that
subcategory using the GSPRO COMBO file.

Table 3-9. Select mobile-related VOC profiles 2016

Sector

Sub-category

2014

Nonroad- California & non US

gasoline exhaust

COMBO

8750a Pre-Tier 2 E0 exhaust
8751a Pre-Tier 2 E10 exhaust

Nonroad-California

gasoline evaporative

COMBO

8753	E0 evap

8754	E10 evap

Nonroad-California

gasoline refueling

COMBO

8869 E0 Headspace

113


-------
Sector

Sub-category

2014





8870

E10 Headspace

Nonroad-California

diesel exhaust

8774

Pre-2007 MY HDD exhaust

Nonroad-California

diesel evap-
orative and diesel refueling

4547

Diesel Headspace

nonpt/
ptnonipm

PFC and BTP

COMBO

8869

8870

E0 Headspace
E10 Headspace

nonpt/
ptnonipm

Bulk plant storage (BPS)
and refine-to-bulk terminal
(RBT) sources

8869

E0 Headspace

The speciation of onroad VOC occurs completely within MOVES. MOVES accounts for fuel type and
properties, emission standards as they affect different vehicle types and model years, and specific
emission processes. Table 3-10 describes all of the M-profiles available to MOVES depending on the
model year range, MOVES process (processID), fuel sub-type (fuelSubTypelD), and regulatory class
(regClassID). Table 3-11 through Table 3-13 describe the meaning of these MOVES codes. For a
specific representative county and future year, there will be a different mix of these profiles. For
example, for HD diesel exhaust, the emissions will use a combination of profiles 8774M and 8775M
depending on the proportion of HD vehicles that are pre-2007 model years (MY) in that particular county.
As that county is projected farther into the future, the proportion of pre-2007 MY vehicles will decrease.
A second example, for gasoline exhaust (not including E-85), the emissions will use a combination of
profiles 8756M, 8757M, 8758M, 8750aM, and 875laM. Each representative county has a different mix
of these key properties and, therefore, has a unique combination of the specific M-profiles. More detailed
information on how MOVES speciates VOC and the profiles used is provided in the technical document,
"Speciation of Total Organic Gas and Particulate Matter Emissions from On-road Vehicles in
MOVES2014" (EPA, 2015c).

Table 3-10. Onroad M-profiles

Profile

Profile Description

Model Years

ProcessID

FuelSubTypelD

RegClassID

1001M

CNG Exhaust

1940-2050

1,2,15,16

30

48

4547M

Diesel Headspace

1940-2050

11

20,21,22

0

4547M

Diesel Headspace

1940-2050

12,13,18,19

20,21,22

10,20,30,40,41,
42,46,47,48

8753M

E0 Evap

1940-2050

12,13,19

10

10,20,30,40,41,42,
46,47,48

8754M

E10 Evap

1940-2050

12,13,19

12,13,14

10,20,30,40,41,
42,46,47,48

8756M

Tier 2 E0 Exhaust

2001-2050

1,2,15,16

10

20,30

8757M

Tier 2 E10 Exhaust

2001-2050

1,2,15,16

12,13,14

20,30

8758M

Tier 2 El5 Exhaust

1940-2050

1,2,15,16

15,18

10,20,30,40,41,
42,46,47,48

8766M

E0 evap permeation

1940-2050

11

10

0

8769M

E10 evap permeation

1940-2050

11

12,13,14

0

8770M

El5 evap permeation

1940-2050

11

15,18

0

8774M

Pre-2007 MY HDD
exhaust

1940-2006

1,2,15,16,17,90

20, 21, 22

40,41,42,46,47, 48

114


-------
Profile

Profile Description

Model Years

ProcessID

FuelSubTypelD

RegClassID

8774M

Pre-2007 MY HDD
exhaust

1940-2050

9125

20, 21, 22

46,47

8774M

Pre-2007 MY HDD
exhaust

1940-2006

1,2,15,16

20, 21, 22

20,30

8775M

2007+ MY HDD exhaust

2007-2050

1,2,15,16

20, 21, 22

20,30

8775M

2007+ MY HDD exhaust

2007-2050

1,2,15,16,17,90

20, 21, 22

40,41,42,46,47,48

8855M

Tier 2 E85 Exhaust

1940-2050

1,2,15,16

50, 51, 52

10,20,30,40,41,
42,46,47,48

8869M

E0 Headspace

1940-2050

18

10

10,20,30,40,41,
42,46,47,48

8870M

E10 Headspace

1940-2050

18

12,13,14

10,20,30,40,41,
42,46,47,48

8871M

El5 Headspace

1940-2050

18

15,18

10,20,30,40,41,
42,46,47,48

8872M

El5 Evap

1940-2050

12,13,19

15,18

10,20,30,40,41,
42,46,47,48

8934M

E85 Evap

1940-2050

11

50,51,52

0

8934M

E85 Evap

1940-2050

12,13,18,19

50,51,52

10,20,30,40,41,
42,46,47,48

8750aM

Pre-Tier 2 E0 exhaust

1940-2000

1,2,15,16

10

20,30

8750aM

Pre-Tier 2 E0 exhaust

1940-2050

1,2,15,16

10

10,40,41,42,46,47,48

875 laM

Pre-Tier 2 E10 exhaust

1940-2000

1,2,15,16

11,12,13,14

20,30

875 laM

Pre-Tier 2 E10 exhaust

1940-2050

1,2,15,16

11,12,13,14,15, 1826

10,40,41,42,46,47,48

Table 3-11. MOVES process IDs

Process ID

Process Name

1

Running Exhaust

2

Start Exhaust

9

Brakewear

10

Tire wear

11

Evap Permeation

12

Evap Fuel Vapor Venting

13

Evap Fuel Leaks

15

Crankcase Running Exhaust

16

Crankcase Start Exhaust

17

Crankcase Extended Idle Exhaust

18

Refueling Displacement Vapor Loss

19

Refueling Spillage Loss

20

Evap Tank Permeation

21

Evap Hose Permeation

22

Evap RecMar Neck Hose Permeation

25	91 is the processed for APUs which are diesel engines not covered by the 2007 Heavy-Duty Rule, so the older technology
applieds to all years.

26	The profile assingments for pre-2001 gasoline vehicles fueled on E15/E20 fuels (subtypes 15 and 18) were corrected for
MOVES2014a. This model year range, process, fuelsubtype regclass combinate is already assigned to profile 8758.

115


-------
23

Evap RecMar Supply/Ret Hose Permeation

24

Evap RecMar Vent Hose Permeation

30

Diurnal Fuel Vapor Venting

31

HotSoak Fuel Vapor Venting

32

RunningLoss Fuel Vapor Venting

40

Nonroad

90

Extended Idle Exhaust

91

Auxiliary Power Exhaust

Table 3-12. MOVES Fuel subtype IDs

Fuel Subtype ID

Fuel Subtype Descriptions

10

Conventional Gasoline

11

Reformulated Gasoline (RFG)

12

Gasohol (E10)

13

Gasohol (E8)

14

Gasohol (E5)

15

Gasohol (E15)

18

Ethanol (E20)

20

Conventional Diesel Fuel

21

Biodiesel (BD20)

22

Fischer-Tropsch Diesel (FTD100)

30

Compressed Natural Gas (CNG)

50

Ethanol

51

Ethanol (E85)

52

Ethanol (E70)

Table 3-13. MOVES regclass IDs

Reg. Class ID

Regulatory Class Description

0

Doesn't Matter

10

Motorcycles

20

Light Duty Vehicles

30

Light Duty Trucks

40

Class 2b Trucks with 2 Axles and 4 Tires (8,500 lbs < GVWR <= 10,000 lbs)

41

Class 2b Trucks with 2 Axles and at least 6 Tires or Class 3 Trucks (8,500 lbs < GVWR <= 14,000
lbs)

42

Class 4 and 5 Trucks (14,000 lbs < GVWR <= 19,500 lbs)

46

Class 6 and 7 Trucks (19,500 lbs < GVWR <= 33,000 lbs)

47

Class 8a and 8b Trucks (GVWR > 33,000 lbs)

48

Urban Bus (see CFR Sec 86.091 2)

For portable fuel containers (PFCs) and fuel distribution operations associated with the bulk-plant-to-
pump (BTP) distribution, ethanol may be mixed into the fuels; therefore, county- and month-specific
COMBO speciation was used (via the GSPROCOMBO file). Refinery to bulk terminal (RBT) fuel
distribution and bulk plant storage (BPS) speciation are considered upstream from the introduction of

116


-------
ethanol into the fuel; therefore, a single profile is sufficient for these sources. No refined information on
potential VOC speciation differences between cellulosic diesel and cellulosic ethanol sources was
available; therefore, cellulosic diesel and cellulosic ethanol sources used the same SCC (30125010:
Industrial Chemical Manufacturing, Ethanol by Fermentation production) for VOC speciation as was used
for corn ethanol plants.

3.2.2 PM speciation

In addition to VOC profiles, the SPECIATE database also contains profiles for speciating PM2.5. PM2.5
was speciated into the AE6 species associated with CMAQ 5.0.1 and later versions. Of particular note for
the 2016v7.2 beta and regional haze platforms, the nonroad PM2.5 speciation was updated as discussed
later in this section. Most of the PM profiles come from the 911XX series (Reff et. al, 2009), which
include updated AE6 speciation.27 Starting with the 2014v7.1 platform, we replaced profile 91112
(Natural Gas Combustion - Composite) with 95475 (Composite -Refinery Fuel Gas and Natural Gas
Combustion). This updated profile is an AE6-ready profile based on the median of 3 SPECIATE4.5
profiles from which AE6 versions were made (to be added to SPECIATE5.0): boilers (95125a), process
heaters (95126a) and internal combustion combined cycle/cogen plant exhaust (95127a). As with profile
91112, these profiles are based on tests using natural gas and refinery fuel gas (England et al., 2007).
Profile 91112 which is also based on refinery gas and natural gas is thought to overestimate EC.

Profile 95475 (Composite -Refinery Fuel Gas and Natural Gas Combustion) is shown along with the
underlying profiles composited in Figure 3-3. Figure 3-4 shows a comparison of the new profile as of the
2014v7.1 platform with the one that we had been using in the 2014v7.0 and earlier platforms.

Figure 3-3. Profiles composited for the new PM gas combustion related sources

Zinc
Sulfate
Silicon
Potassium !

Particulate Non-Carbon Organic Matter	_

Other Unspeciated PM2.5
Organic carbon
Nitrate
Nickel

Metal-bound Oxygen — 1
Iron P-
Elemental Carbon
Copper
Chloride ion
Calcium P"

Bromine Atom
Ammonium
Aluminum

0	10	20	30	40	50	60

Weight Percent

¦	Composite -Refinery Fuel Gas and Natural Gas Combustion 95475

¦	Gas-fired process heater exhaust 95126a

¦	Gas-fired internal combustion combined cycle/cogeneration plant exhaust 95127a

¦	Gas-fired boiler exhaust 95125a

27 The exceptions are 5675AE6 (Marine Vessel - Marine Engine - Heavy Fuel Oil) used for cmv_c3 and 92018 (Draft
Cigarette Smoke - Simplified) used in nonpt. 5675AE6 is an update of profile 5675 to support AE6 PM speciation.

117


-------
Figure 3-4. Comparison of PM profiles used for Natural gas combustion related sources

Zinc
Sulfate
Silicon
Potassium

Particulate Non-Carbon Organic Matter
Other Unspeciated PM2.5
Organic carbon
Nitrate
Nickel

Metal-bound Oxygen
Iron

Elemental Carbon
Copper
Chloride ion
Calcium
Bromine Atom
Ammonium
Aluminum

20	30	40

Weight Percent

I Composite -Refinery Fuel Gas and Natural Gas Combustion 95475
Natural Gas Combustion - Composite 91112

3.2.2.1 Mobile source related PM2.5 speciation profiles

For the onroad sector, for all processes except brake and tire wear, PM speciation occurs within MOVES
itself, not within SMOKE (similar to the VOC speciation described above). The advantage of using
MOVES to speciate PM is that during the internal calculation of MOVES, the model has complete
information on the characteristics of the fleet and fuels (e.g., model year, sulfur content, process, etc.) to
accurately match to specific profiles. This means that MOVES produces EF tables that include total PM
(e.g., PMio and PM2.5) and speciated PM (e.g., PEC, PFE, etc). SMOKE essentially calculates the PM
components by using the appropriate EF without further speciation.28 The specific profiles used within
MOVES include two CNG profiles, 45219 and 45220, which were added to SPECIATE4.5. A list of
profiles is provided in the technical document, "Speciation of Total Organic Gas and Particulate Matter
Emissions from On-road Vehicles in MOVES2014" (EPA, 2015c).

For onroad brake and tire wear, the PM is speciated in the moves2smk postprocessor that prepares the
emission factors for processing in SMOKE. The formulas for this are based on the standard speciation
factors from brake and tire wear profiles, which were updated from the v6.3 platform based on data from
a Health Effects Institute report (Schauer, 2006). Table 3-14 shows the differences in the v7.1 and v6.3
profiles.

28 Unlike previous platforms, the PM components (e.g., POC) are now consistently defined between MOVES2014 and CMAQ.
For more details on the use of model-ready EF, see the SMOKE 3.7 documentation:
https://www.cmascenter.0rg/smoke/documentation/3.7/l1tml/.

118


-------
Table 3-14. SPECIATE4.5 brake and tire profiles compared to those used in the 2011v6.3 Platform

Inventory
Pollutant

Model
Species

V6.3 platform
brakewear profile:
91134

SPECIATE4.5 brakewear
profile: 95462 from
Schauer (2006)

V6.3 platform

tirewear
profile: 91150

SPECIATE4.5 tirewear
profile: 95460 from
Schauer (2006)

PM2 5

PAL

0.00124

0.000793208

6.05E-04

3.32401E-05

PM2 5

PCA

0.01

0.001692177

0.00112



PM2 5

PCL

0.001475



0.0078



PM2 5

PEC

0.0261

0.012797085

0.22

0.003585907

PM2 5

PFE

0.115

0.213901692

0.0046

0.00024779

PM2 5

PH20

0.0080232



0.007506



PM2 5

PK

1.90E-04

0.000687447

3.80E-04

4.33129E-05

PM2 5

PMG

0.1105

0.002961309

3.75E-04

0.000018131

PM2 5

PMN

0.001065

0.001373836

1.00E-04

1.41E-06

PM2 5

PMOTHR

0.4498

0.691704999

0.0625

0.100663209

PM2 5

PNA

1.60E-04

0.002749787

6.10E-04

7.35312E-05

PM2 5

PNCOM

0.0428

0.020115749

0.1886

0.255808124

PM2 5

PNH4

3.00E-05



1.90E-04



PM2 5

PN03

0.0016



0.0015



PM2 5

POC

0.107

0.050289372

0.4715

0.639520309

PM2 5

PSI

0.088



0.00115



PM2 5

PS04

0.0334



0.0311



PM2 5

PTI

0.0036

0.000933341

3.60E-04

5.04E-06

The formulas used based on brake wear profile 95462 and tire wear profile 95460 are as follows:

POC = 0.6395 * PM25TIRE + 0.0503 * PM25BRAKE
PEC = 0.0036 * PM25TIRE + 0.0128 * PM25BRAKE
PN03 = 0.000 * PM25TIRE + 0.000 * PM25BRAKE
PS04 = 0.0 * PM25TIRE + 0.0 * PM25BRAKE
PNH4 = 0.000 * PM25TIRE + 0.0000 * PM25BRAKE
PNCOM = 0.2558 * PM25TIRE + 0.0201 * PM25BRAKE

For California onroad emissions, adjustment factors were applied to SMOKE-MOVES to produce
California adjusted model-ready files. California did not supply speciated PM, therefore, the adjustment
factors applied to PM2.5 were also applied to the speciated PM components. By applying the ratios
through SMOKE-MOVES, the CARB inventories are essentially speciated to match EPA estimated
speciation.

For nonroad PM2.5, speciation is partially done within MOVES such that it does not need to be run for a
specific chemical mechanism. For nonroad, MOVES outputs emissions of PM2.5 split by speciation
profile. Similar to how VOC and NONHAPTOG are speciated, PM2.5 is now also speciated this way
starting with MOVES2014b. California nonroad emissions, which are not from MOVES, continue to be
speciated the traditional way with speciation profiles assigned by SMOKE using the GSREF cross-
reference. The PM2.5 profiles assigned to nonroad sources are listed in Table 3-15.

119


-------
Table 3-15. Nonroad PM2.5 profiles

SPECIATE4.5
Profile Code

SPECIATE4.5 Profile Name

Assigned to Nonroad
sources based on Fuel
Type

8996

Diesel Exhaust - Heavy-heavy duty truck - 2007
model year with NCOM

Diesel

91106

HDDV Exhaust - Composite

Diesel

91113

Nonroad Gasoline Exhaust - Composite

Gasoline

91156

Residential Natural Gas Combustion

CNG and LPG
(California only)

95219

CNG Transit Bus Exhaust

CNG and LPG

3.2.3 NOx speciation

NOx emission factors and therefore NOx inventories are developed on a NO2 weight basis. For air quality
modeling, NOx is speciated into NO, NO2, and/or HONO. For the non-mobile sources, the EPA used a
single profile "NHONO" to split NOx into NO and NO2.

The importance of HONO chemistry, identification of its presence in ambient air and the measurements of
HONO from mobile sources have prompted the inclusion of HONO in NOx speciation for mobile
sources. Based on tunnel studies, a HONO to NOx ratio of 0.008 was chosen (Sarwar, 2008). For the
mobile sources, except for onroad (including nonroad, cmv, rail, othon sectors), and for specific SCCs in
othar and ptnonipm, the profile "HONO" is used. Table 3-16 gives the split factor for these two profiles.
The onroad sector does not use the "HONO" profile to speciate NOx. MOVES2014 produces speciated
NO, NO2, and HONO by source, including emission factors for these species in the emission factor tables
used by SMOKE-MOVES. Within MOVES, the HONO fraction is a constant 0.008 of NOx. The NO
fraction varies by heavy duty versus light duty, fuel type, and model year.

The NO2 fraction = 1 - NO - HONO. For more details on the NOx fractions within MOVES, see EPA
report "Use of data from 'Development of Emission Rates for the MOVES Model,'

Sierra Research, March 3, 2010" available at
https://nepis.epa.gov/Exe/ZyPDF.cgi?Dockev=P100F lA5.pdf.

Table 3-16. NOx speciation profiles

Profile

pollutant

species

split factor

HONO

NOX

N02

0.092

HONO

NOX

NO

0.9

HONO

NOX

HONO

0.008

NHONO

NOX

N02

0.1

NHONO

NOX

NO

0.9

3.2.4 Creation of Sulfuric Acid Vapor (SULF)

Since at least the 2002 Platform, sulfuric acid vapor (SULF) has been estimated through the SMOKE
speciation process for coal combustion and residual and distillate oil fuel combustion sources. Profiles
that compute SULF from SO2 are assigned to coal and oil combustion SCCs in the GSREF ancillary file.

120


-------
The profiles were derived from information from AP-42 (EPA, 1998), which identifies the fractions of
sulfur emitted as sulfate and SO2 and relates the sulfate as a function of S02.

Sulfate is computed from SO2 assuming that gaseous sulfate, which is comprised of many components, is
primarily H2SO4. The equation for calculating FhSO/ds given below.

Emissions of SULF (as H2S04)	Equation 3-1

fraction of S emitted as sulfate MW H2S04

= S07 emissions x —				-	x	

fraction of S emitted as S02 MW S02

In the above, AfWis the molecular weight of the compound. The molecular weights of H2SO4 and SO2
are 98 g/mol and 64 g/mol, respectively.

This method does not reduce SO2 emissions; it solely adds gaseous sulfate emissions as a function of S02
emissions. The derivation of the profiles is provided in Table 3-17; a summary of the profiles is provided
in Table 3-18.

Table 3-17. Sulfate split factor computation

fuel

SCCs

Profile
Code

Fraction
as S02

Fraction as
sulfate

Split factor (mass
fraction)

Bituminous

1-0X-002-YY, where X is 1,
2 or 3 and YY is 01 thru 19
and 21-ZZ-002-000 where
ZZ is 02,03 or 04

95014

0.95

0.014

.014/.95 * 98/64 =
0.0226

Subbituminous

1-0X-002-YY, where X is 1,
2 or 3 and YY is 21 thru 38

87514

.875

0.014

.014/.875 * 98/64 =
0.0245

Lignite

1-0X-003-YY, where X is 1,
2 or 3 and YY is 01 thru 18
and 21-ZZ-002-000 where
ZZ is 02,03 or 04

75014

0.75

0.014

.014/.75 * 98/64 =
0.0286

Residual oil

1-0X-004-YY, where X is 1,
2 or 3 and YY is 01 thru 06
and 21-ZZ-005-000 where
ZZ is 02,03 or 04

99010

0.99

0.01

.01/. 99 * 98/64 =
0.0155

Distillate oil

1-0X-005-YY, where X is 1,
2 or 3 and YY is 01 thru 06
and 21-ZZ-004-000 where
ZZ is 02,03 or 04

99010

0.99

0.01

Same as residual oil

Table 3-18. SO2 speciation profiles

Profile

pollutant

species

split factor

95014

S02

SULF

0.0226

95014

S02

S02

1

87514

S02

SULF

0.0245

87514

S02

S02

1

75014

S02

SULF

0.0286

121


-------
75014

S02

S02

1

99010

S02

SULF

0.0155

99010

S02

S02

1

3.3 Temporal Allocation

Temporal allocation is the process of distributing aggregated emissions to a finer temporal resolution,
thereby converting annual emissions to hourly emissions as is required by CMAQ. While the total
emissions are important, the timing of the occurrence of emissions is also essential for accurately
simulating ozone, PM, and other pollutant concentrations in the atmosphere. Many emissions inventories
are annual or monthly in nature. Temporal allocation takes these aggregated emissions and distributes the
emissions to the hours of each day. This process is typically done by applying temporal profiles to the
inventories in this order: monthly, day of the week, and diurnal, with monthly and day-of-week profiles
applied only if the inventory is not already at that level of detail.

The temporal factors applied to the inventory are selected using some combination of country, state,
county, SCC, and pollutant. Table 3-19 summarizes the temporal aspects of emissions modeling by
comparing the key approaches used for temporal processing across the sectors. In the table, "Daily
temporal approach" refers to the temporal approach for getting daily emissions from the inventory using
the SMOKE Temporal program. The values given are the values of the SMOKE L TYPE setting. The
"Merge processing approach" refers to the days used to represent other days in the month for the merge
step. If this is not "all," then the SMOKE merge step runs only for representative days, which could
include holidays as indicated by the right-most column. The values given are those used for the SMOKE
M TYPE setting (see below for more information).

Table 3-19. Temporal settings used for the platform sectors in SMOKE

Platform sector
short name

Inventory
resolutions

Monthly

profiles

used?

Daily

temporal

approach

Merge

processing

approach

Process holidays
as separate days

afdust adj

Annual

Yes

week

All

Yes

afdust ak adj

Annual

Yes

week

All

Yes

ag

Monthly

No

all

All

No

airports

Annual

Yes

week

week

Yes

beis

Hourly

No

n/a

All

No

cmv clc2

Annual

Yes

aveday

aveday

No

cmv c3

Annual

Yes

aveday

aveday

No

nonpt

Annual

Yes

week

week

Yes

nonroad

Monthly

No

mwdss

mwdss

Yes

np oilgas

Annual

Yes

aveday

aveday

No

onroad

Annual & monthly1

No

all

all

Yes

onroad ca adj

Annual & monthly1

No

all

all

Yes

onroad nonconus

Annual & monthly1

No

all

all

Yes

othafdust adj

Annual

Yes

week

all

No

othar

Annual & monthly

Yes

week

week

No

onroad can

Monthly

No

week

week

No

122


-------
Platform sector
short name

Inventory
resolutions

Monthly

profiles

used?

Daily

temporal

approach

Merge

processing

approach

Process holidays
as separate days

onroad mex

Monthly

No

week

week

No

othpt

Annual & monthly

Yes

mwdss

mwdss

No

othptdust adi

Monthly

No

week

all

No

pt oilgas

Annual

Yes

mwdss

mwdss

Yes

ptegu

Annual & hourly

Yes2

all

all

No

ptnonipm

Annual

Yes

mwdss

mwdss

Yes

ptagfire

Daily

No

all

all

No

ptfire

Daily

No

all

all

No

ptfire othna

Daily

No

all

all

No

rail

Annual

Yes

aveday

aveday

No

rwc

Annual

No3

met-based3

all

No3

'Note the annual and monthly "inventory" actually refers to the activity data (VMT, hoteling, and VPOP) for onroad.
VMT and hoteling is monthly and VPOP is annual. The actual emissions are computed on an hourly basis.

2Only units that do not have matching hourly CEMS data use monthly temporal profiles.

3Except for 2 SCCs that do not use met-based speciation

The following values are used in the table. The value "all" means that hourly emissions are computed for
every day of the year and that emissions potentially have day-of-year variation. The value "week" means
that hourly emissions computed for all days in one "representative" week, representing all weeks for each
month. This means emissions have day-of-week variation, but not week-to-week variation within the
month. The value "mwdss" means hourly emissions for one representative Monday, representative
weekday (Tuesday through Friday), representative Saturday, and representative Sunday for each month.
This means emissions have variation between Mondays, other weekdays, Saturdays and Sundays within
the month, but not week-to-week variation within the month. The value "aveday" means hourly
emissions computed for one representative day of each month, meaning emissions for all days within a
month are the same. Special situations with respect to temporal allocation are described in the following
subsections.

In addition to the resolution, temporal processing includes a ramp-up period for several days prior to
January 1, 2016, which is intended to mitigate the effects of initial condition concentrations. The ramp-up
period was 10 days (December 22-31, 2015). For most sectors, emissions from December 2016
(representative days) were used to fill in emissions for the end of December 2015. For biogenic
emissions, December 2015 emissions were processed using 2015 meteorology.

3.3.1 Use of FF10 format for finer than annual emissions

The FF10 inventory format for SMOKE provides a consolidated format for monthly, daily, and hourly
emissions inventories. With the FF10 format, a single inventory file can contain emissions for all 12
months and the annual emissions in a single record. This helps simplify the management of numerous
inventories. Similarly, daily and hourly FF10 inventories contain individual records with data for all days
in a month and all hours in a day, respectively.

SMOKE prevents the application of temporal profiles on top of the "native" resolution of the inventory.
For example, a monthly inventory should not have annual-to-month temporal allocation applied to it;
rather, it should only have month-to-day and diurnal temporal allocation. This becomes particularly

123


-------
important when specific sectors have a mix of annual, monthly, daily, and/or hourly inventories. The
flags that control temporal allocation for a mixed set of inventories are discussed in the SMOKE
documentation. The modeling platform sectors that make use of monthly values in the FF10 files are ag,
nonroad, onroad, onroad can, onroadmex, othar, and othpt.

3.3.2 Electric Generating Utility temporal allocation (ptegu)

3.3.2.1 Base year temporal allocation of EGUs

The temporal allocation procedure for EGUs in the base year is differentiated by whether or not the unit
could be directly matched to a unit with CEMS data via its ORIS facility code and boiler ID. Note that
for units matched to CEMS data, annual totals of their emissions input to CMAQ may be different than
the annual values in the 2016 annual inventory because the CEMS data replaces the NOx and SO2 annual
inventory data for the seasons in which the CEMS are operating. If a CEMS-matched unit is determined
to be a partial year reporter, as can happen for sources that run CEMS only in the summer, emissions
totaling the difference between the annual emissions and the total CEMS emissions are allocated to the
non-summer months. Prior to use of the CEMS data in SMOKE it is processed through the CEMCorrect
tool. The CEMCorrect tool identifies hours for which the data were not measured as indicated by the data
quality flags in the CEMS data files. Unmeasured data can be filled in with maximum values and thereby
cause erroneously high values in the CEMS data. When data were flagged as unmeasured and the values
were found to be more than three times the annual mean for that unit, the data for those hours are replaced
with annual mean values (Adelman et al., 2012). These adjusted CEMS data were then used for the
remainder of the temporal allocation process described below (see Figure 3-5 for an example).

Figure 3-5. Eliminating unmeasured spikes in CEMS data

2016 January CEMs for 6068 3

Date

124


-------
In modeling platforms prior to 2016 beta, unmatched EGUs were temporally allocated using daily and
diurnal profiles weighted by CEMS values within an IPM region, season, and by fuel type (coal, gas, and
other). All unit types (peaking and non-peaking) were given the same profile within a region, season and
fuel bin. Units identified as municipal waste combustors (MWCs) or cogeneration units (cogens) were
given flat daily and diurnal profiles. Beginning with the 2016 beta platform and continuing for the 2016vl
platform, the small EGU temporalization process was improved to also consider peaking units.

The region, fuel, and type (peaking or non-peaking) were identified for each input EGU with CEMS data
that are used for generating profiles. The identification of peaking units was based on hourly heat input
data from the 2016 base year and the two previous years (2014 and 2015). The heat input was summed for
each year. Equation 3-2 shows how the annual heat input value is converted from heat units (BTU/year) to
power units (MW) using the unit-level heat rate (BTU/kWh) derived from the NEEDS v6 database. In
Equation 3-3 a capacity factor is calculated by dividing the annual unit MW value by the NEEDS v6 unit
capacity value (MW) multiplied by the hours in the year. A peaking unit was defined as any unit that had
a maximum capacity factor of less than 0.2 for every year (2014, 2015, and 2016) and a 3-year average
capacity factor of less than 0.1.

Annual Unit Power Output

Annual Unit Output (MW) =

y8760

Hourly HI	(MW\

(BTU) ' 1000 (W)	Equation 3-2

NEEDS Heat Rate (frrrr)

Unit Capacity Factor

Annual Unit Output (MW)

Capacity Factor

NEEDS Unit Capacity	* 8760 (h)	Equation 3-3

Input regions were determined from one of the eight EGU modeling regions based on MJO and climate
regions. Regions were used to group units with similar climate-based load demands. Region assignment is
made on a state level, where all units within a state were assigned to the appropriate region. Unit fuel
assignments were made using the primary NEEDS v6 fuel. Units fueled by bituminous, subbituminous, or
lignite are assigned to the coal fuel type. Natural gas units were assigned to the gas fuel type. Distillate
and residual fuel oil were assigned to the oil fuel type. Units with any other primary fuel were assigned
the "other" fuel type. The number of units used to calculate the daily and diurnal EGU temporal profiles
are shown in Figure 3-6 by region, fuel, and for peaking/non-peaking. Currently there are 64 unique
profiles available based on 8 regions, 4 fuels, and 2 for peaking unit status (peaking and non-peaking).

125


-------
Figure 3-6. Temporal Profile Input Unit Counts by Fuel and Peaking Unit Classification

Small EGU 2016 Temporal Profile Input Unit Counts

(peaknaftw
a»:0 /1 |

ml: fl / 0 j
CChei: 0 / D

1 Wqt" HofttfCentral ^
(iK^r^roTpciiiing):

'coal C,'41	

.nai:4S,.',24

EM0IB

ulfitr. O'-'O

HMIE-VU	.

;coK'3 '! { f \
1 J
ij I— <¦
-:L-rr:

¦West—1		

(pesfcrijirw^jeaSiM^];
ant : 0/3
fasiW /137
"oil: 01 0
OOfer: 0 V <1

S—.
01: 74/S \ \
other: O / S3 \ /

South

(pBjkirtgi'narpeakifig):

0197 I
9WJ537-JJ7~L,

at-, 18/0 	

Other: fl / 4 k

WOCO

(pciitangi'rKnprsiun^}:

— 155
4L.U3&

EGU Regions

¦	LADCO

¦	MMIE-VU
I I Northwest

~	SESARM
i I South

I I Southwest
I 1 west

~	West North Central

The daily and diurnal profiles were calculated for each region, fuel, and peaking type group from the year
2016 CEMS heat input values. The heat input values were summed for each input group to the annual
level at each level of temporal resolution: monthly, month-of-day, and diurnal. The sum by temporal
resolution value was then divided by the sum of annual heat input in that group to get a set of
temporalization factors. Diurnal factors were created for both the summer and winter seasons to account
for the variation in hourly load demands between the seasons. For example, the sum of all hour 1 heat
input values in the group was divided by the sum of all heat inputs over all hours to get the hour 1 factor.
Each grouping contained 12 monthly factors, up to 31 daily factors per month, and two sets of 24 hourly
factors. The profiles were weighted by unit size where the units with more heat input have a greater
influence on the shape of the profile. Composite profiles were created for each region and type across all
fuels as a way to provide profiles for a fuel type that does not have hourly CEMS data in that region.
Figure 3-7 shows peaking and non-peaking daily temporal profiles for the gas fuel type in the LADCO
region. Figure 3-8 shows the diurnal profiles for the coal fuel type in the Mid-Atlantic Northeast Visibility
Union (MANE-VU) region.

126


-------
Figure 3-7. Example Daily Temporal Profiles for the LADCO Region and the Gas Fuel Type

0.040 -
0.035 -
0.030 -
c 0.025 -

O

-M

u

£ 0.020 -
>.

Q 0.015 -
0.010 -
0.005 -
0.000 -

Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec
2016

na\(

Figure 3-8. Example Diurnal Temporal Profiles for the MANE-VU Region and the Coal Fuel Type

0.10

0.08 -

0.06 -

h

0.04-

0.02 -

0.00

0	5	10	15	20

Hour

SMOKE uses a cross-reference file to select a monthly, daily, and diurnal profile for each source. For the
2016 beta and vl platforms, the temporal profiles were assigned in the cross-reference at the unit level to
EGU sources without hourly CEMS data. An inventory of all EGU sources without CEMS data was used
to identify the region, fuel type, and type (peaking/non-peaking) of each source. As with the input unit the
regions are assigned using the state from the unit FIPS. The fuel was assigned by SCC to one of the four
fuel types: coal, gas, oil, and other. A fuel type unit assignment is made by summing the VOC, NOX,

Daily Small EGU Profile for LADCO gas

Diurnal Small EGU Profile for MANE-VU coal

127


-------
PM2.5, and S02 for all SCCs in the unit. The SCC that contributed the highest total emissions to the unit
for selected pollutants was used to assign the unit fuel type. Peaking units were identified as any unit with
an oil, gas, or oil fuel type with a NAICS of 22111 or 221112. Some units may be assigned to a fuel type
within a region that does not have an available input unit with a matching fuel type in that region. These
units without an available profile for their group were assigned to use the regional composite profile.
MWC and cogen units were identified using the NEEDS primary fuel type and cogeneration flag,
respectively, from the NEEDS v6 database. The number of EGU units assigned each profile group are
shown by region in Figure 3-9.

Figure 3-9. Non-CEMS EGU Temporal Profile Application Counts

Small EGU 2016 Temporal Profile Application Counts

LADCO

KnrfMwKl1

3.3.2.2 Future year temporal allocation of EGUs

For future year temporal allocation of unit-level EGU emissions, estimates of average winter
(representing December through February), average winter shield (October through November and March
through April), and average summer (May through September) values were provided by the Integrated
Planning Model (IPM) for all units. The seasonal emissions for the 2023 and 2028 EGU future year cases
were produced by post processing of the IPM outputs. The unit-level data were converted into hourly
values through the temporal allocation process using a 3-step methodology: annualized summer/winter
value to month, month to day, and day to hour. CEMS data from the air quality analysis year (e.g., 2016)
is used as much as possible to temporally allocate the EGU emissions.

The goal of the temporal allocation process is to reflect the variability in the unit-level emissions that can
impact air quality over seasonal, daily, or hourly time scales, in a manner compatible with incorporating
future-year emission projections into future-year air quality modeling. The temporal allocation process is

128


-------
applied to the seasonal emission projections for two seasons: summer (May through September) and
winter (October through April). The winter shield emissions are summed with the winter emissions for
consistency with previous platforms that did not have separate values for the winter shield season. The
Flat File used as the input to the temporal allocation process contains unit-level emissions and stack
parameters (i.e., stack location and other characteristics consistent with information found in the NEI).
When the flat file is produced from post-processed IPM outputs, a cross reference is used to map the units
in version 6 of the NEEDS database to the stack parameter and facility, unit, release point, and process
identifiers used in the NEI. This cross reference also maps sources to the hourly CEMS data used to
temporally allocate the emissions in the base year air quality modeling.

All units have seasonal information provided in the future year Flat File, the monthly values in the Flat
File input to the temporal allocation process are computed by multiplying the average summer day,
average winter shield day, and average winter day emissions by the number of days in the respective
month. When generating seasonal emissions totals from the Flat File winter shield emissions are summed
with the winter emissions to create a total winter season. In summary, the monthly emission values shown
in the Flat File are not intended to represent an actual month-to-month emission pattern. Instead, they are
interim values that have translated IPM's seasonal projections into month-level data that serve as a
starting point for the temporal allocation process.

The monthly emissions within the Flat File undergo a multi-step temporal allocation process to yield the
hourly emission values at each unit, as is needed for air quality modeling: summer or winter value to
month, month to day, and day to hour. For sources not matched to unit-specific CEMS data, the first two
steps are done outside of SMOKE and the third step to get to hourly values is done by SMOKE using the
daily emissions files created from the first two steps. For each of these three temporal allocation steps,
NOx and SO2 CEMS data are used to allocate NOxand SO2 emissions, while CEMS heat input data are
used to allocate all other pollutants. The approach defined here gives priority to temporalization based on
the base year CEMS data to the maximum extent possible for both base and future year modeling.

Prior to using the 2016 CEMS data to develop monthly, daily, and hourly profiles, the CEMS data were
processed through the CEMCorrect tool to make adjustments for hours for which data quality flags
indicated the data were not measured and that the reported values were much larger than the annual mean
emissions for the unit. These adjusted CEMS data were used to compute the monthly, daily, and hourly
profiles described below.

For units that have CEMS data available and that have CEMS units matched to the NEI sources, the
emissions are temporalized according to the base year (i.e., 2016) CEMS data for that unit and pollutant.
For units that are not matched to the NEI or for which CEMS data are not available, the allocation of the
seasonal emissions to months is done using average fuel-specific season-to-month factors for both
peaking and non-peaking units generated for each of the eight regions shown in Figure 5. These factors
are based on a single year of CEMS data for the modeling base year associated with the air quality
modeling analysis being performed, such as 2016. The fuels used for creating the profiles for a region
were coal, natural gas, oil, and "other". The "other "fuels category is a broad catchall that includes fuels
such as wood and waste. Separate profiles are computed for NOx, SO2, and heat input, where heat input is
used to temporally allocate emissions for pollutants other than NOx and SO2. An overall composite
profile across all fuels is also computed and can be used in the event that a region has too few units of a
fuel type to make a reasonable average profile, or in the case when a unit changes fuels between the base
and future year and there were previously no units with that fuel in the region containing the unit. A
complete description of the generation and application of these regional fuel profiles is available in the
base year temporalization section.

129


-------
The monthly emission values in the Flat File were first reallocated across the months in that season to
align the month-to-month emission pattern at each stack with historic seasonal emission patterns. While
this reallocation affects the monthly pattern of each unit's future-year seasonal emissions, the seasonal
totals are held equal to the IPM projection for that unit and season. Second, the reallocated monthly
emission values at each stack are disaggregated down to the daily level consistent with historic daily
emission patterns in the given month at the given stack using separate profiles for NOx, SO2, and heat
input. This process helps to capture the influence of meteorological episodes that cause electricity
demand to vary from day-to-day, as well as weekday-weekend effects that change demand during the
course of a given week. Third, this data set of emission values for each day of the year at each unit is
input into SMOKE, which uses temporal profiles to disaggregate the daily values into specific values for
each hour of the year.

For units without or not matched to CEMS data, or for which the CEMS data are found to be unsuitable
for use in the future year, emissions were allocated from month to day using IPM-region and fuel-specific
average month-to-day factors based on CEMS data from the base year of the air quality modeling
analysis. These instances include units that did not operate in the base year or for which it may not have
been possible to match the unit to a specific unit in the NEI. Regional average profiles may be used for
some units with CEMS data in the base year when one of the following cases is true: (1) units are
projected to have substantially increased emissions in the future year compared to its emissions in the
base (historic) year; (2) CEMS data were only available for a limited number of hours in that base year;
(3) the unit is new in the future year; (4) when there were no CEMS data for one season in the base year
but IPM runs the unit during both seasons; or (5) units experienced atypical conditions during the base
year, such as lengthy downtimes for maintenance or installation of controls.

The temporal profiles that map emissions from days to hours were computed based on the region and
fuel-specific seasonal (i.e., winter and summer) average day-to-hour factors derived from the CEMS data
for heat input for those fuels and regions and for that season. Heat input was used because it is the
variable that is the most complete in the CEMS data and should be present for all of the hours in which
the unit was operating. SMOKE uses these diurnal temporal profiles to allocate the daily emissions data
to hours of each day. Note that this approach results in each unit having the same hourly temporal
allocation for all the days of a season.

The emissions from units for which unit-specific profiles were not used were temporally allocated to
hours reflecting patterns typical of the region in which the unit is located. Analysis of CEMS data for
units in each of the 8 regions shown in Figure 3-6 revealed that there were differences in the temporal
patterns of historic emission data that correlate with fuel type (e.g., coal, gas, oil, and other), time of year,
pollutant, season (i.e., winter versus summer) and region of the country. The correlation of the temporal
pattern with fuel type is explained by the relationship of units' operating practices with the fuel burned.
For example, coal units take longer to ramp up and ramp down than natural gas units, and some oil units
are used only when electricity demand cannot otherwise be met. Geographically, the patterns were less
dependent on state location than they were on regional location. Figure 3-7 provides an example of daily
profiles for gas fuel in the LADCO region. The EPA developed seasonal average emission profiles, each
derived from base year CEMS data for each season across all units sharing both IPM region and fuel type.
Figure 3-8 provides an example of seasonal profiles that allocate daily emissions to hours in the MANE-
VU region. These average day-to-hour temporal profiles were also used for sources during seasons of the
year for which there were no CEMS data available, but for which IPM predicted emissions in that season.
This situation can occur for multiple reasons, including how the CEMS was run at each source in the base
year.

130


-------
For units that do have CEMS data in the base year and were matched to units in the IPM output, the base
year CEMS data were scaled so that their seasonal emissions match the IPM-projected totals. The scaling
process used the fraction of the unit's seasonal emissions in the base year as computed for each hour of
the season, and then applied those fractions to the seasonal emissions from the future year Flat File. Any
pollutants other than NOx and SO2 were temporally allocated using heat input. Through the temporal
allocation process, the future year emissions will have the same temporal pattern as the base year CEMS
data, where available, while the future-year seasonal total emissions for each unit match the future-year
unit-specific projection for each season (see example in Figure 3-10). Note that the future year IPM
output for 2030 also maps to the year 2028 and was therefore used for the 2028 modeling case.

In cases when the emissions for a particular unit are projected to be substantially higher in the future year
than in the base year, the proportional scaling method to match the emission patterns in the base year
described above can yield emissions for a unit that are much higher than the historic maximum emissions
for that unit. To help address this issue in the future case, the maximum measured emissions of NOx and
SO2 in the period of 2014-2017 were computed. The temporally allocated emissions were then evaluated
at each hour to determine whether they were above this maximum. The amount of "excess emissions"
over the maximum were then computed. For units for which the "excess emissions" could be reallocated
to other hours, those emissions were distributed evenly to hours that were below the maximum. Those
hourly emissions were then reevaluated against the maximum, and the procedure of reallocating the
excess emissions to other hours was repeated until all of the hours had emissions below the maximum,
whenever possible (see example in Figure 3-11).

Figure 3-10. Future Year Emissions Follow the Pattern of Base Year Emissions

2030 and 2016 Summer CEMs for 2277 1

May
2016

¦	2016 CEMs
2030 CEMs

¦	2030 Adjusted CEMs
Annual unit max

131


-------
Figure 3-11. Excess Emissions Apportioned to Hours Less than the Historic Maximum

12000
10000
8000

.c

§ 6000

rN
O
ui

4000
2000

May	Jun	Jul	Aug	Sep

2016

Date

Using the above approach, it was not always possible to reallocate excess emissions to hours below the
historic maximum, such as when the total seasonal emissions of NOx or SO2 for a unit divided by the
number of hours of operation are greater than the 2014-2017 maximum emissions level. For these units,
the regional fuel-specific average profiles were applied to all pollutants, including heat input, for the
respective season (see example in Figure 3-12). It was not possible for SMOKE to use regional profiles
for some pollutants and adjusted CEMS data for other pollutants for the same unit and season, therefore,
all pollutants in the unit and season are assigned to regional profiles when regional profiles are needed.
For some units, hourly emissions values still exceed the 2014-2017 annual maximum for the unit even
after regional profiles were applied (see example in Figure 3-13).

2030 and 2016 Summer CEMs for 3943 2

2016 CEMs
2030 CEMs
2030 Adjusted CEMs
Annual unit max

rm

132


-------
Figure 3-12. Regional Profile Applied due to not being able to Adjust below Historic Maximum

2030 and 2016 Summer CEMs for 6095 2

2016 CEMs
2030 CEMs
2030 Adjusted CEMs
2030 Season Fuel
Annual unit max

May
2016

Figure 3-13. Regional Profile Applied, but Exceeds Historic Maximum in Some Hours

2030 and 2016 Summer CEMs for 2103 1

2016 CEMs
2030 CEMs
2030 Adjusted CEMs
2030 Season Fuel
Annual unit max

133


-------
3.3.3 Airport Temporal allocation (airports)

Airport temporal profiles were updated in 2014v7.0 and were kept the same for the 2016vl platform. All
airport SCCs (i.e., 2275*, 2265008005, 2267008005, 2268008005 and 2270008005) were given the same
hourly, weekly and monthly profile for all airports other than Alaska seaplanes (which are not in the
CMAQ modeling domain). Hourly airport operations data were obtained from the Aviation System
Performance Metrics (ASPM) Airport Analysis website (https://aspm.faa.gov/apm/svs/AnalvsisAP.asp).
A report of 2014 hourly Departures and Arrivals for Metric Computation was generated. An overview of
the ASPM metrics is at

http://aspmhelp.faa.gov/index.php/Aviation Performance Metrics %28APM%29. Figure 3-14 shows
the diurnal airport profile.

Weekly and monthly temporal profiles are based on 2014 data from the FAA Operations Network Air
Traffic Activity System (http://aspm.faa.gov/opsnet/sys/Terminal.asp). A report of all airport operations
(takeoffs and landings) by day for 2014 was generated. These data were then summed to month and day-
of-week to derive the monthly and weekly temporal profiles shown in Figure 3-14, Figure 3-15, and
Figure 3-16. An overview of the Operations Network data system is at
http://aspmhelp.faa.gov/index.php/Operations Network %28QPSNET%29.

Alaska seaplanes, which are outside the CONUS domain use the same monthly profile as in the 2011
platform shown in Figure 3-17. These were assigned based on the facility ID.

Figure 3-14. Diurnal Profile for all Airport SCCs
Diurnal Airport Profile

Hour

134


-------
Figure 3-15. Weekly profile for all Airport SCCs

Weekly Airport Profile

0.18

Figure 3-16. Monthly Profile for all Airport SCCs

Monthly Airport Profile

0.05
0.04
0.03
0.02
0.01
0

Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec

135


-------
Figure 3-17. Alaska Seaplane Profile

0.14
0.12
0.10
0.08
0.06
0.04

0.02
0.00

Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec

3.3.4 Residential Wood Combustion Temporal allocation (rwc)

There are many factors that impact the timing of when emissions occur, and for some sectors this includes
meteorology. The benefits of utilizing meteorology as a method for temporal allocation are: (1) a
meteorological dataset consistent with that used by the AQ model is available (e.g., outputs from WRF);
(2) the meteorological model data are highly resolved in terms of spatial resolution; and (3) the
meteorological variables vary at hourly resolution and can, therefore, be translated into hour-specific
temporal allocation.

The SMOKE program Gentpro provides a method for developing meteorology-based temporal allocation.
Currently, the program can utilize three types of temporal algorithms: annual-to-day temporal allocation
for residential wood combustion (RWC); month-to-hour temporal allocation for agricultural livestock
NH3; and a generic meteorology-based algorithm for other situations. Meteorological-based temporal
allocation was used for portions of the rwc sector and for the entire ag sector.

Gentpro reads in gridded meteorological data (output from MCIP) along with spatial surrogates and uses
the specified algorithm to produce a new temporal profile that can be input into SMOKE. The
meteorological variables and the resolution of the generated temporal profile (hourly, daily, etc.) depend
on the selected algorithm and the run parameters. For more details on the development of these
algorithms and running Gentpro, see the Gentpro documentation and the SMOKE documentation at
http://www.cmascenter.0rg/smoke/documentation/3.l/GenTPRO Technical Summary Aug2012 Final, pd
f and https://www.cmascenter.Org/smoke/documentation/4.5/html/ch05s03s05.html respectively.

For the RWC algorithm, Gentpro uses the daily minimum temperature to determine the temporal
allocation of emissions to days of the year. Gentpro was used to create an annual-to-day temporal profile
for the RWC sources. These generated profiles distribute annual RWC emissions to the coldest days of
the year. On days where the minimum temperature does not drop below a user-defined threshold, RWC
emissions for most sources in the sector are zero. Conversely, the program temporally allocates the
largest percentage of emissions to the coldest days. Similar to other temporal allocation profiles, the total
annual emissions do not change, only the distribution of the emissions within the year is affected. The
temperature threshold for RWC emissions was 50 °F for most of the country, and 60 °F for the following

136


-------
states: Alabama, Arizona, California, Florida, Georgia, Louisiana, Mississippi, South Carolina, and
Texas. The algorithm is as follows:

IfTd >=Tt: no emissions that day
If Td < Tt: daily factor = 0.79*(Tt -Td)

where (Td = minimum daily temperature; Tt = threshold temperature, which is 60 degres F in southern
states and 50 degrees F elsewhere).

Once computed, the factors are normalized to sum to 1 to ensure that the total annual emissions are
unchanged (or minimally changed) during the temporal allocation process.

Figure 3-18 illustrates the impact of changing the temperature threshold for a warm climate county. The
plot shows the temporal fraction by day for Duval County, Florida, for the first four months of 2007. The
default 50 °F threshold creates large spikes on a few days, while the 60 °F threshold dampens these spikes
and distributes a small amount of emissions to the days that have a minimum temperature between 50 and

60 °F.

Figure 3-18. Example of RWC temporal allocation in 2007 using a 50 versus 60 °F threshold

RWC temporal profile, Duval County, FL, Jan - Apr

0.04
0.035
0.03
\ 0.025
| 0.02

i

3 0.015
0.01
0.005



1
1

A.



i
i

J

i i

l v







/»
»

} I

I







h

i\i
:<









'\

\ 	

1

K



1 , /A
lrv

\ 11





1 .1

\

/\ *



Ai A /

\ r\ \











\ a r\ f\ a



Mr v v



V\ \ A / A Ji V
> \/ — /; \\r: \ A

60F, alternate formula

) Oi O C O O O <

The diurnal profile used for most RWC sources (see Figure 3-19) places more of the RWC emissions in
the morning and the evening when people are typically using these sources. This profile is based on a
2004 MANE-VU survey based temporal profiles (https://s3.amazonaws.com/marama.org/wp-
content/uploads/2019/11/04184303/Qpen Burning Residential Areas Emissions Report-2004.pdf). This
profile was created by averaging three indoor and three RWC outdoor temporal profiles from counties in
Delaware and aggregating them into a single RWC diurnal profile. This new profile was compared to a
concentration-based analysis of aethalometer measurements in Rochester, New York (Wang et al. 2011)
for various seasons and days of the week and was found that the new RWC profile generally tracked the
concentration based temporal patterns.

137


-------
Figure 3-19. RWC diurnal temporal profile

Comparison of RWC diurnal profile

The temporal allocation for "Outdoor Hydronic Heaters" (i.e., "OHH," SCC=2104008610) and "Outdoor
wood burning device, NEC (fire-pits, chimneas, etc.)" (i.e., "recreational RWC," SCC=21040087000) is
not based on temperature data, because the meteorologically-based temporal allocation used for the rest of
the rwc sector did not agree with observations for how these appliances are used.

For OHH, the annual-to-month, day-of-week and diurnal profiles were modified based on information in
the New York State Energy Research and Development Authority's (NYSERDA) "Environmental,

Energy Market, and Health Characterization of Wood-Fired Hydronic Heater Technologies, Final Report"
(NYSERDA, 2012), as well as a Northeast States for Coordinated Air Use Management (NESCAUM)
report "Assessment of Outdoor Wood-fired Boilers" (NESCAUM, 2006). A Minnesota 2008 Residential
Fuelwood Assessment Survey of individual household responses (MDNR, 2008) provided additional
annual-to-month, day-of-week, and diurnal activity information for OHH as well as recreational RWC
usage.

Data used to create the diurnal profile for OHH, shown in Figure 3-20, are based on a conventional single-
stage heat load unit burning red oak in Syracuse, New York. As shown in Figure 3-21, the NESCAUM
report describes how for individual units, OHH are highly variable day-to-day but that in the aggregate,
these emissions have no day-of-week variation. In contrast, the day-of-week profile for recreational RWC
follows a typical "recreational" profile with emissions peaked on weekends.

Annual-to-month temporal allocation for OHH as well as recreational RWC were computed from the
MDNR 2008 survey and are illustrated in Figure 3-22. The OHH emissions still exhibit strong seasonal
variability, but do not drop to zero because many units operate year-round for water and pool heating. In
contrast to all other RWC appliances, recreational RWC emissions are used far more frequently during the
warm season.

138


-------
Figure 3-20. Data used to produce a diurnal profile for OHH, based on heat load (BTU/hr)

Figure 3-21. Day-of-week temporal profiles for OHH and Recreational RWC

139


-------
Figure 3-22. Annual-to-month temporal profiles for OHH and recreational RWC

3.3.5 Agricultural Ammonia Temporal Profiles (ag)

For the agricultural livestock NFb algorithm, the GenTPRO algorithm is based on an equation derived by
Jesse Bash of the EPA's ORD based on the Zhu, Henze, et al. (2013) empirical equation. This equation is
based on observations from the TES satellite instrument with the GEOS-Chem model and its adjoint to
estimate diurnal NFb emission variations from livestock as a function of ambient temperature,
aerodynamic resistance, and wind speed. The equations are:

Et.h = [161500/T, /, x eM380 x AR,,/,	Equation 3-4

PE;,/; = Euh / Sum(E,,/,)	Equation 3-5

where

•	PE;,/; = Percentage of emissions in county i on hour h

•	Eij, = Emission rate in county i on hour h

•	Tin = Ambient temperature (Kelvin) in county i on hour h

•	AR;,/; = Aerodynamic resistance in county i

GenTPRO was run using the "BASH NH3" profile method to create month-to-hour temporal profiles for
these sources. Because these profiles distribute to the hour based on monthly emissions, the monthly
emissions are obtained from a monthly inventory, or from an annual inventory that has been temporalized
to the month. Figure 3-23 compares the daily emissions for Minnesota from the "old" approach (uniform
monthly profile) with the "new" approach (GenTPRO generated month-to-hour profiles) for 2014.
Although the GenTPRO profiles show daily (and hourly variability), the monthly total emissions are the
same between the two approaches.

140


-------
Figure 3-23. Example of animal NH3 emissions temporal allocation approach, summed to daily

emissions

2014fd Minnesota ag NH3 livestock daily temporal profiles

1600
1400
~ 1200
t? 1000

on

g 800
* 600
400
200
0

m
X





























































j

rl





























1

,



















Ailtr

,

ft



i iA















j

s



A rtjl









Hi"* r



AW



|

y



w V





1/1/2014 2/1/2014 3/4/2014 4/4/2014 5/5/2014 6/5/2014 7/6/2014 8/6/2014 9/6/2014 10/7/201411/7/201412/8/2014

-months
approach

¦ hourly
approach

For the 2016 platform, the GenTPRO approach is applied to all sources in the ag sector, NFb and non-
NFb, livestock and fertilizer. Monthly profiles are based on the daily-based EPA livestock emissions and
are the same as were used in 2014v7.0. Profiles are by state/SCC_category, where SCC_category is one
of the following: beef, broilers, layers, dairy, swine.

3.3.6 Oil and gas temporal allocation (np_oilgas)

Monthly oil and gas temporal profiles by county and SCC were updated to use 2016 activity information
for the 2016vl platform. Weekly and diurnal profiles are flat and are based on comments received on a
version of the 2011 platform.

3.3.7 Onroad mobile temporal allocation (onroad)

For the onroad sector, the temporal distribution of emissions is a combination of traditional temporal
profiles and the influence of meteorology. This section will discuss both the meteorological influences
and the development of the temporal profiles for this platform.

The "inventories" referred to in Table 3-19 consist of activity data for the onroad sector, not emissions.
For the off-network emissions from the rate-per-profile (RPP) and rate-per-vehicle (RPV) processes, the
VPOP activity data is annual and does not need temporal allocation. For rate-per-hour (RPH) processes
that result from hoteling of combination trucks, the HOTELING inventory is annual and was
temporalized to month, day of the week, and hour of the day through temporal profiles.

For on-roadway rate-per-distance (RPD) processes, the VMT activity data is annual for some sources and
monthly for other sources, depending on the source of the data. Sources without monthly VMT were
temporalized from annual to month through temporal profiles. VMT was also temporalized from month
to day of the week, and then to hourly through temporal profiles. The RPD processes require a speed
profile (SPDPRO) that consists of vehicle speed by hour for a typical weekday and weekend day. For
onroad, the temporal profiles and SPDPRO will impact not only the distribution of emissions through
time but also the total emissions. Because SMOKE-MOVES (for RPD) calculates emissions based on the

141


-------
VMT, speed and meteorology, if one shifted the VMT or speed to different hours, it would align with
different temperatures and hence different emission factors. In other words, two SMOKE-MOVES runs
with identical annual VMT, meteorology, and MOVES emission factors, will have different total
emissions if the temporal allocation of VMT changes. Figure 3-24 illustrates the temporal allocation of
the onroad activity data (i.e., VMT) and the pattern of the emissions that result after running SMOKE-
MOVES. In this figure, it can be seen that the meteorologically varying emission factors add variation on
top of the temporal allocation of the activity data.

Meteorology is not used in the development of the temporal profiles, but rather it impacts the calculation
of the hourly emissions through the program Movesmrg. The result is that the emissions vary at the
hourly level by grid cell. More specifically, the on-network (RPD) and the off-network parked vehicle
(RPV, RPH, and RPP) processes use the gridded meteorology (MCIP) either directly or indirectly. For
RPD, RPV, and RPH, Movesmrg determines the temperature for each hour and grid cell and uses that
information to select the appropriate emission factor for the specified SCC/pollutant/mode combination.
For RPP, instead of reading gridded hourly meteorology, Movesmrg reads gridded daily minimum and
maximum temperatures. The total of the emissions from the combination of these four processes (RPD,
RPV, RPH, and RPP) comprise the onroad sector emissions. The temporal patterns of emissions in the
onroad sector are influenced by meteorology.

Figure 3-24. Example of temporal variability of NOx emissions



4

	

3.5

U



O
_c

3

l/l



oj

2.5

E



c

2

o



=

1.5





1-

1

>



>

0.5

0

7/8/140:00

	VMT

	NOX

2014v2 onroad RPD hourly NOX and VMT: Wake County, NC

7/9/140:00 7/10/140:00 7/11/140:00 7/12/140:00 7/13/140:00 7/14/140:00

Date and time (GMT)

0

7/15/140:00

New VMT day-of-week and hour-of-day temporal profiles were developed for use in the 2014NEIv2 and
later platforms as part of the effort to update the inputs to MOVES and SMOKE-MOVES under CRC A-
100 (Coordinating Research Council, 2017). CRC A-100 data includes profiles by region or county, road
type, and broad vehicle category. There are three vehicle categories: passenger vehicles (11/21/31),
commercial trucks (32/52), and combination trucks (53/61/62). CRC A-100 does not cover buses, refuse
trucks, or motor homes, so those vehicle types were mapped to other vehicle types for which CRC A-100
did provide profiles as follows: 1) Intercity/transit buses were mapped to commercial trucks; 2) Motor
homes were mapped to passenger vehicles for day-of-week and commercial trucks for hour-of-day; 3)
School buses and refuse trucks were mapped to commercial trucks for hour-of-day and use a new custom
day-of-week profile called LOWSATSUN that has a very low weekend allocation, since school buses and
refuse trucks operate primarily on business days. In addition to temporal profiles, CRC A-100 data were
also used to develop the average hourly speed data (SPDPRO) used by SMOKE-MOVES. In areas where
CRC A-100 data does not exist, hourly speed data is based on MOVES county databases.

142


-------
The CRC A-100 dataset includes temporal profiles for individual counties, Metropolitan Statistical Areas
(MSAs), and entire regions (e.g. West, South). For counties without county or MSA temporal profiles
specific to itself, regional temporal profiles are used. Temporal profiles also vary by each of the MOVES
road types, and there are distinct hour-of-day profiles for each day of the week. Plots of hour-of-day
profiles for passenger vehicles in Fulton County, GA, are shown in Figure 3-25. Separate plots are shown
for Monday, Friday, Saturday, and Sunday, and each line corresponds to a particular MOVES road type
(i.e., road type 2 = rural restricted, 3 = rural unrestricted, 4 = urban restricted, and 5 = urban unrestricted)
Figure 3-26 shows which counties have temporal profiles specific to that county, and which counties use
MSA or regional average profiles. Figure 3-27 shows the regions used to coput regional average profiles.

Figure 3-25. Sample onroad diurnal profiles for Fulton County, GA

o.oi	o.oi ^

0			0

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24	1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24

road 2	road 3	road 4	road 5	road 2	road 3	road 4	road 5

Saturday	Fulton Co	passenger	Sunday	Fulton Co	passenger

0.09	0.1

road 2	road 3	road 4	road 5	road 2	road 3	road 4	road 5

143


-------
Figure 3-26. Methods to Populate Onroad Speeds and Temporal Profiles by Road Type

Figure 3-27. Regions for computing Region Average Speeds and Temporal Profiles

144


-------
For hoteling, day-of-week profiles are the same as non-hoteling for combination trucks, while hour-of-day
non-hoteling profiles for combination trucks were inverted to create new hoteling profiles that peak
overnight instead of during the day. The combination truck profiles for Fulton County are shown in
Figure 3-28.

The CRC A-100 temporal profiles were used in the entire contiguous United States, except in California.
All California temporal profiles were carried over from 2014v7.0, although California hoteling uses CRC
A-100-based profiles just like the rest of the country, since CARB didn't have a hoteling-specific profile.
Monthly profiles in all states (national profiles by broad vehicle type) were also carried over from
2014v7.0 and applied directly to the VMT. For California, CARB supplied diurnal profiles that varied by
vehicle type, day of the week,29 and air basin. These CARB-specific profiles were used in developing
EPA estimates for California. Although the EPA adjusted the total emissions to match California-
submitted emissions for 2016, the temporal allocation of these emissions took into account both the state-
specific VMT profiles and the SMOKE-MOYES process of incorporating meteorology.

Monday

Figure 3-28. Example of Temporal Profiles for Combination Trucks

Fulton Co	combo	Friday	Fulton Co	combo

5 6 7 8 9 10 11 12 13 14 15 16 17 18 13 20 21 22 23 24
road 2	road 3	road 4	road 5

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24
road 2	road 3	road 4	road 5

Saturday

Fulton Co

combo

Sunday

Fulton Co

combo

5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24
road 2	road 3	road 4	road 5

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24
road 2	road 3	road 4	road 5

29 California's diurnal profiles varied within the week. Monday, Friday, Saturday, and Sunday had unique profiles and
Tuesday, Wednesday, Thursday had the same profile.

145


-------
3.3.8 Nonroad mobile temporal allocation(nonroad)

For nonroad mobile sources, temporal allocation is performed differently for different SCCs. Beginning
with the final 2011 platform and continued int the 2016 platform, some improvements to temporal
allocation of nonroad mobile sources were made to make the temporal profiles more realistically reflect
real-world practices. Some specific updates were made for agricultural sources (e.g., tractors),
construction, and commercial residential lawn and garden sources.

Figure 3-29 shows two previously existing temporal profiles (9 and 18) and a new temporal profile (19)
which has lower emissions on weekends. In the 2016 platform, construction and commercial lawn and
garden sources were updated from profile 18 to the new profile 19 which has lower emissions on
weekends. Residental lawn and garden sources continue to use use profile 9 and agricultural sources
continue to use profile 19.

Figure 3-29. Example Nonroad Day-of-week Temporal Profiles

Day of Week Profiles

0.24
0.22
0.2
0.18
0.16
0.14
0.12
0.1
0.08
0.06
0.04
0.02
0

mond^ tuesday Wednesday thursday friday Saturday Sunday
	9	18	19

146


-------
Figure 3-30 shows the previously existing temporal profiles 26 and 27 along with new temporal profiles
(25a and 26a) which have lower emissions overnight. In the 2016 platform, construction sources
previously used profile 26 and were upated to use profile 26a. Commercial lawn and garden and
agriculture sources also previously used profile 26 but were updated to use the new profiles 26a and 25a,
respectively. Residental lawn and garden sources were updated from profile 26 to use profile 27.

Figure 3-30. Example Nonroad Diurnal Temporal Profiles

Hour of Day Profiles

26a-New 	27 	 25 a-New	26

3.3.9 Additional sector specific details (afdust, beis, cmv, rail, nonpt,
ptnonipm, ptfire)

For the afdust sector, meteorology is not used in the development of the temporal profiles, but it is used to
reduce the total emissions based on meteorological conditions. These adjustments are applied through
sector-specific scripts, beginning with the application of land use-based gridded transport fractions and
then subsequent zero-outs for hours during which precipitation occurs or there is snow cover on the
ground. The land use data used to reduce the NEI emissions explains the amount of emissions that are
subject to transport. This methodology is discussed in (Pouliot et al., 2010), and in "Fugitive Dust
Modeling for the 2008 Emissions Modeling Platform" (Adelman, 2012). The precipitation adjustment is
applied to remove all emissions for hours where measurable rain occurs, or where there is snow cover.
Therefore, the afdust emissions vary day-to-day based on the precipitation and/or snow cover for each
grid cell and hour. Both the transport fraction and meteorological adjustments are based on the gridded
resolution of the platform; therefore, somewhat different emissions will result from different grid
resolutions. For this reason, to ensure consistency between grid resolutions, afdust emissions for the
36US3 grid are aggregated from the 12US1 emissions. Application of the transport fraction and
meteorological adjustments prevents the overestimation of fugitive dust impacts in the grid modeling as
compared to ambient samples.

Biogenic emissions in the beis sector vary by every day of the year because they are developed using
meteorological data including temperature, surface pressure, and radiation/cloud data. The emissions are
computed using appropriate emission factors according to the vegetation in each model grid cell, while
taking the meteorological data into account.

147


-------
For the cmv sectors, most areas use hourly emission inventories derived from the 5-minute AIS data. In
some areas where AIS data are not available, such as in Canada between the St. Lawrence Seaway and the
Great Lakes and in the southern Carribbean, the flat temporal profiles are used for hourly and day-of-
week values. Most regions without AIS data also use a flat monthly profile, with some offshore areas
using an average monthly profile derived from the 2008 ECA inventory monthly values. These areas
without AIS data also use flat day of week and hour of day profiles.

For the rail sector, new monthly profiles were developed for the 2016 platform. Monthly temporal
allocation for rail freight emissions is based on AAR Rail Traffic Data, Total Carloads and Intermodal, for
2016. For passenger trains, monthly temporal allocation is flat for all months. Rail passenger miles data
is available by month for 2016 but it is not known how closely rail emissions track with passenger activity
since passenger trains run on a fixed schedule regardless of how many passengers are aboard, and so a flat
profile is chosen for passenger trains. Rail emissions are allocated with flat day of week profiles, and
most emissions are allocated with flat hourly profiles.

For the ptagfire sector, the inventories are in the daily point fire format FF10 PTDAY. The diurnal
temporal profile for ag fires reflects the fact that burning occurs during the daylight hours - see Figure
3-31 (McCarty et al., 2009). This puts most of the emissions during the work day and suppresses the
emissions during the middle of the night.

Figure 3-31. Agricultural burning diurnal temporal profile



Comparison of Agricultural Burning Temporal Profiles

0.18



0.16



0.14

I \ 	New McCarty Profile

0.12

j \ ..— OLD EPA

1 0.1

/ ^

I0"08

/

£

/ \

0.06



0.04



0.02



0

123456789 101112131415161718192021222324

Industrial processes that are not likely to shut down on Sundays, such as those at cement plants, use
profiles that include emissions on Sundays, while those that would shut down on Sundays use profiles that
reflect Sunday shutdowns.

For the ptfire sectors, the inventories are in the daily point fire format FF10 PTDAY. Separate hourly
profiles for prescribed and wildfires were used. Figure 3-32 below shows the profiles used for each state
for the 2014v7.0 and 2014v7.1 modeling platforms. They are similar but not the same and vary according

148


-------
to the average meteorological conditions in each state. The 2016 alpha platform uses the ptfire diurnal
profiles form 2014v7.1 platform.

Figure 3-32. Prescribed and Wildfire diurnal temporal profiles

For the nonroad sector, while the NEI only stores the annual totals, the modeling platform uses monthly
inventories from output from MOVES. For California, CARB's annual inventory was temporalized to
monthly using monthly temporal profiles applied in SMOKE by SCC. This is an improvement over the
2011 platform, which applied monthly temporal allocation in California at the broader SCC7 level.

3.4 Spatial Allocation

The methods used to perform spatial allocation are summarized in this section. For the modeling
platform, spatial factors are typically applied by county and SCC. As described in Section 3.1, spatial
allocation was performed for national 36-km and 12-km domains. To accomplish this, SMOKE used
national 36-km and 12-km spatial surrogates and a SMOKE area-to-point data file. For the U.S., the EPA
updated surrogates to use circa 2014 data wherever possible. For Mexico, updated spatial surrogates were
used as described below. For Canada, updated surrogates were provided by Environment Canada for the
2016v7.2 platform. The U.S., Mexican, and Canadian 36-km and 12-km surrogates cover the entire
CONUS domain 12US1 shown in Figure 3-1. The 36US3 domain includes a portion of Alaska, and since
Alaska emissions are typically not included in air quality modeling, special considerations are taken to
include Alaska emissions in 36-km modeling.

Documentation of the origin of the spatial surrogates for the platform is provided in the workbook
US SpatialSurrogate_Workbook_v07172018 which is available with the reports for the 2014v7.1
platform. The remainder of this subsection summarizes the data used for the spatial surrogates and the
area-to-point data which is used for airport refueling.

3.4.1 Spatial Surrogates for U.S. emissions

There are more than 100 spatial surrogates available for spatially allocating U.S. county-level emissions
to the 36-km and 12-km grid cells used by the air quality model. As described in Section 3.4.2, an area-
to-point approach overrides the use of surrogates for an airport refueling sources. Table 3-20 lists the
codes and descriptions of the surrogates. Surrogate names and codes listed in italics are not directly

149


-------
assigned to any sources for the 2016 alpha platform, but they are sometimes used to gapfill other
surrogates, or as an input for merging two surrogates to create a new surrogate that is used.

Many surrogates were updated or newly developed for use in the 2014v7.0 platform (Adelman, 2016).
They include the use of the 2011 National Land Cover Database (the previous platform used 2006) and
development of various development density levels such as open, low, medium high and various
combinations of these. These landuse surrogates largely replaced the FEMA category surrogates that
were used in the 2011 platform. Additionally, onroad surrogates were developed using average annual
daily traffic counts from the highway monitoring performance system (HPMS). Previously, the "activity"
for the onroad surrogates was length of road miles. This and other surrogates are described in a reference
(Adelman, 2016).

Several surrogates were updated or developed as new surrogates for the 2016v7.1 (aka alpha) platform:
Oil and gas surrogates were updated to represent 2016;

Onroad spatial allocation uses surrogates that do not distinguish between urban and rural road
types, correcting the issue arising in some counties due to the inconsistent urban and rural
definitions between MOVES and the surrogate data and were further updated for the 2016vl
platform;

Correction was made to the water surrogate to gap fill missing counties using the 2006 National
Land Cover Database (NLCD).

In addition, spatial surrogates 201 through 244, which concern road miles, annual average daily traffic
(AADT), and truck stops, were further updated for the 2016 beta and regional haze platforms. The
surrogates for the U.S. were mostly generated using the Surrogate Tool to drive the Spatial Allocator, but
a few surrogates were developed directly within ArcGIS or using scripts that manipulate spatial data in
PostgreSQL. The tool and documentation for the Surrogate Tool is available at
https://www.cmascenter.Org/sa-tools/documentation/4.2/SurrogateToolUserGuide 4 2.pdf.

Table 3-20. U.S. Surrogates available for the 2016vl modeling platforms

Code

Surrogate Description

Code

Surrogate Description

N/A

Area-to-point approach (see 3.6.2)

506

Education

100

Population

507

Heavy Light Construction Industrial Land

110

Housing

510

Commercial plus Industrial

131

urban Housing

515

Commercial plus Institutional Land

132

Suburban Housing

520

Commercial plus Industrial plus Institutional

134

Rural Housing

525

Golf Courses plus Institutional plus
Industrial plus Commercial

137

Housing Change

526

Residential - Non-Institutional

140

Housing Change and Population

527

Single Family Residential

150

Residential Heating - Natural Gas

535

Residential + Commercial + Industrial +
Institutional + Government

160

Residential Heating - Wood \

540

Retail Trade (COM1)

170

Residential Heating - Distillate Oil

545

Personal Repair (COM3)

180

Residential Heating - Coal

555

Professional/Technical (COM4) plus General
Government (GOV1)

190

Residential Heating - LP Gas

560

Hospital (COM6)

201

Urban Restricted Road Miles

575

Light and High Tech Industrial (1ND2 +
IND5)

202

Urban Restricted AADT

580

Food Drug Chemical Industrial (1ND3)

150


-------
Code

Surrogate Description

Code

Surrogate Description

205

Extended Idle Locations

585

Metals and Minerals Industrial (IND4)

211

Rural Restricted Road Miles

590

Heavy Industrial (IND1)

212

Rural Restricted AADT

595

Light Industrial (IND2)

221

Urban Unrestricted Road Miles ;

596

Industrial plus Institutional plus Hospitals

222

Urban Unrestricted AADT ]

650

Refineries and Tank Farms

231

Rural Unrestricted Road Miles

670

Spud Count - CBM Wells

232

Rural Unrestricted AADT \

671

Spud Count - Gas Wells

239

Total Road AADT

672

Gas Production at Oil Wells

240

Total Road Miles

673

Oil Production at CBM Wells

241

Total Restricted Road Miles =

674

Unconventional Well Completion Counts

242

All Restricted AADT

676

Well Count - All Producing

243

Total Unrestricted Road Miles

677

Well Count - All Exploratory

244

All Unrestricted AADT

678

Completions at Gas Wells

258

Intercity Bus Terminals

679

Completions at CBM Wells

259

Transit Bus Terminals

681

Spud Count - Oil Wells

260

Total Railroad Miles

683

Produced Water at All Wells

261

NT AD Total Railroad Density

685

Completions at Oil Wells

271

NT AD Class 12 3 Railroad Density

686

Completions at All Wells

272

NTAD Amtrak Railroad Density

687

Feet Drilled at All Wells

273

NTAD Commuter Railroad Density

691

Well Counts - CBM Wells

275

ERTACRail Yards

692

Spud Count-All Wells

280

Class 2 and 3 Railroad Miles i

693

Well Count - All Wells

300

NLCD Low Intensity Development

694

Oil Production at Oil Wells

301

NL CD Med Intensity Development

695

Well Count - Oil Wells

302

NLCD High Intensity Development \

696

Gas Production at Gas Wells

303

NLCD Open Space

697

Oil Production at Gas Wells

304

NLCD Open + Low

698

Well Count - Gas Wells

305

NLCD Low + Med

699

Gas Production at CBM Wells

306

NLCD Med + High

710

Airport Points

307

NLCD All Development

711

Airport Areas

308

NLCD Low + Med + High

801

Port Areas

309

NLCD Open + Low + Med

802

Shipping Lanes

310

NLCD Total Agriculture

805

Offshore Shipping Area

318

NLCD Pasture Land

806

Offshore Shipping NEI2014 Activity

319

NLCD Crop Land

807

Navigable Waterway Miles

320

NLCD Forest Land

808

2013 Shipping Density

321

NLCD Recreational Land

820

Ports NEI2014 Activity

340

NLCD Land

850

Golf Courses

350

NLCD Water

860

Mines

500

Commercial Land

890

Commercial Timber

505

Industrial Land





For the onroad sector, the on-network (RPD) emissions were allocated differently from the off-network
(RPP and RPV). On-network used AADT data and off network used land use surrogates as shown in
Table 3-21. Emissions from the extended (i.e., overnight) idling of trucks were assigned to surrogate 205,
which is based on locations of overnight truck parking spaces. This surrogate's underlying data were
updated for use in the 2016 platforms to include additional data sources and corrections based on
comments received.

151


-------
Table 3-21. Off-Network Mobile Source Surrogates

Source type

Source Type name

Surrogate ID

Description

11

Motorcycle

307

NLCD All Development

21

Passenger Car

307

NLCD All Development

31

Passenger Truck

307

NLCD All Development







NLCD Low + Med +

32

Light Commercial Truck

308

High

41

Intercity Bus

258

Intercity Bus Terminals

42

Transit Bus

259

Transit Bus Terminals

43

School Bus

506

Education

51

Refuse Truck

306

NLCD Med + High

52

Single Unit Short-haul Truck

306

NLCD Med + High

53

Single Unit Long-haul Truck

306

NLCD Med + High

54

Motor Home

304

NLCD Open + Low

61

Combination Short-haul Truck

306

NLCD Med + High

62

Combination Long-haul Truck

306

NLCD Med + High

For the oil and gas sources in the np oilgas sector, the spatial surrogates were updated to those shown in
Table 3-22 using 2016 data consistent with what was used to develop the 2016 beta nonpoint oil and gas
emissions. The primary activity data source used for the development of the oil and gas spatial
surrogates was data from Drilling Info (DI) Desktop's HPDI database (Drilling Info, 2017). This
database contains well-level location, production, and exploration statistics at the monthly level.

Due to a proprietary agreement with DI Desktop, individual well locations and ancillary
production cannot be made publicly available, but aggregated statistics are allowed. These data were
supplemented with data from state Oil and Gas Commission (OGC) websites (Alaska, Arizona, Idaho,
Illinois, Indiana, Kentucky, Louisiana, Michigan, Mississippi, Missouri, Nevada, Oregon and
Pennsylvania, Tennessee). In cases when the desired surrogate parameter was not available (e.g., feet
drilled), data for an alternative surrogate parameter (e.g., number of spudded wells) was downloaded and
used. Under that methodology, both completion date and date of first production from HPDI were used to
identify wells completed during 2016. In total, over 1 million unique wells were compiled from the above
data sources. The wells cover 34 states and over 1,100 counties. (ERG, 2018).

Table 3-22. Spatial Surrogates for Oil and Gas Sources

Surrogate Code

Surrogate Description

670

Spud Count - CBM Wells

671

Spud Count - Gas Wells

672

Gas Production at Oil Wells

673

Oil Production at CBM Wells

674

Unconventional Well Completion Counts

676

Well Count - All Producing

677

Well Count - All Exploratory

678

Completions at Gas Wells

679

Completions at CBM Wells

681

Spud Count - Oil Wells

683

Produced Water at All Wells

152


-------
Surrogate Code

Surrogate Description

685

Completions at Oil Wells

686

Completions at All Wells

687

Feet Drilled at All Wells

691

Well Counts - CBM Wells

692

Spud Count - All Wells

693

Well Count - All Wells

694

Oil Production at Oil Wells

695

Well Count - Oil Wells

696

Gas Production at Gas Wells

697

Oil Production at Gas Wells

698

Well Count - Gas Wells

699

Gas Production at CBM Wells

Not all of the available surrogates are used to spatially allocate sources in the modeling platform; that is,
some surrogates shown in Table 3-20 were not assigned to any SCCs, although many of the "unused"
surrogates are actually used to "gap fill" other surrogates that are used. When the source data for a
surrogate has no values for a particular county, gap filling is used to provide values for the surrogate in
those counties to ensure that no emissions are dropped when the spatial surrogates are applied to the
emission inventories. Table 3-23 shows the CAP emissions (i.e., NH3, NOx, PM2.5, SO2, and VOC) by
sector assigned to each spatial surrogate.

Table 3-23. Selected 2016 CAP emissions by sector for U.S. Surrogates (short tons in 12US1)

Sector

ID

Description

NH3

NOX

PM2 5

S02

VOC

afdust

240

Total Road Miles





294,379





afdust

304

NLCD Open + Low





1,053,145





afdust

306

NLCD Med + High





43,633





afdust

308

NLCD Low + Med + High





123,524





afdust

310

NLCD Total Agriculture





988,012





ag

310

NLCD Total Agriculture

3,409,761







194,779

nonpt

100

Population

0

0

0

0

1,240,692

nonpt

150

Residential Heating - Natural Gas

42,973

219,189

3,632

1,442

13,296

nonpt

170

Residential Heating - Distillate Oil

1,563

31,048

3,356

41,193

1,051

nonpt

180

Residential Heating - Coal

20

101

53

1,086

111

nonpt

190

Residential Heating - LP Gas

111

33,230

175

705

1,292

nonpt

239

Total Road AADT

0

25

551

0

274,266

nonpt

240

Total Road Miles

0

0

0

0

34,027

nonpt

242

All Restricted AADT

0

0

0

0

5,451

nonpt

244

All Unrestricted AADT

0

0

0

0

96,232

nonpt

271

NT AD Class 12 3 Railroad Density

0

0

0

0

2,252

nonpt

300

NLCD Low Intensity Development

5,198

27,727

104,108

3,722

71,770

nonpt

306

NLCD Med + High

27,518

180,692

207,536

62,698

950,022

nonpt

307

NLCD All Development

25

46,331

126,722

14,185

601,828

nonpt

308

NLCD Low + Med + High

1,027

171,603

16,096

13,527

65,123

153


-------
Sector

ID

Description

NH3

NOX

PM2 5

S02

voc

nonpt

310

NLCD Total Agriculture

0

0

37

0

204,819

nonpt

319

NLCD Crop Land

0

0

95

71

293

nonpt

320

NLCD Forest Land

69

378

1,289

9

474

nonpt

505

Industrial Land

0

0

0

0

174

nonpt

535

Residential + Commercial + Industrial +
Institutional + Government

5

2

130

0

39

nonpt

560

Hospital (COM6)

0

0

0

0

0

nonpt

650

Refineries and Tank Farms

0

22

0

0

99,564

nonpt

711

Airport Areas

0

0

0

0

271

nonpt

801

Port Areas

0

0

0

0

8,194

nonroad

261

NT AD Total Railroad Density

3

2,154

227

2

425

nonroad

304

NLCD Open + Low

4

1,824

159

5

2,727

nonroad

305

NLCD Low + Med

94

15,985

3,832

126

114,513

nonroad

306

NLCD Med + High

305

183,591

11,873

421

93,596

nonroad

307

NLCD All Development

99

31,526

15,340

125

169,943

nonroad

308

NLCD Low + Med + High

498

338,083

28,585

487

51,865

nonroad

309

NLCD Open + Low + Med

119

21,334

1,257

162

45,498

nonroad

310

NLCD Total Agriculture

422

378,388

28,387

425

40,707

nonroad

320

NLCD Forest Land

15

5,910

703

15

3,939

nonroad

321

NLCD Recreational Land

83

11,616

6,517

104

246,154

nonroad

350

NLCD Water

188

115,175

5,952

240

353,189

nonroad

850

Golf Courses

13

2,001

117

18

5,613

nonroad

860

Mines

2

2,691

281

3

521

np oilgas

670

Spud Count - CBM Wells

0

0

0

0

112

np oilgas

671

Spud Count - Gas Wells

0

0

0

0

6,284

np oilgas

674

Unconventional Well Completion Counts

12

18,802

720

9

1,264

np oilgas

678

Completions at Gas Wells

0

5,315

136

2,488

16,615

np oilgas

679

Completions at CBM Wells

0

3

0

80

395

np oilgas

681

Spud Count - Oil Wells

0

0

0

0

15,164

np oilgas

683

Produced Water at All Wells

0

11

0

0

47,271

np oilgas

685

Completions at Oil Wells

0

255

0

769

27,935

np oilgas

687

Feet Drilled at All Wells

0

36,162

1,309

22

2,664

np oilgas

691

Well Counts - CBM Wells

0

32,971

490

13

27,566

np oilgas

693

Well Count - All Wells

0

0

0

0

159

np oilgas

694

Oil Production at Oil Wells

0

4,165

0

15,385

1,062,178

np oilgas

695

Well Count - Oil Wells

0

134,921

2,953

32

566,235

np oilgas

696

Gas Production at Gas Wells

0

16,339

1,847

164

428,206

np oilgas

698

Well Count - Gas Wells

0

320,688

6,217

258

582,442

np oilgas

699

Gas Production at CBM Wells

0

2,413

312

25

7,602

onroad

205

Extended Idle Locations

230

78,126

794

36

13,711

onroad

239

Total Road AADT









5,755

onroad

242

All Restricted AADT

34,545

1,175,197

38,140

8,744

194,836

onroad

244

All Unrestricted AADT

65,543

1,773,993

67,525

17,788

477,839

onroad

258

Intercity Bus Terminals



147

2

0

34

154


-------
Sector

ID

Description

NH3

NOX

PM2 5

S02

voc

onroad

259

Transit Bus Terminals



53

3

0

149

onroad

304

NLCD Open + Low



829

29

1

3,874

onroad

306

NLCD Med + High



15,209

333

17

19,917

onroad

307

NLCD All Development



546,312

10,195

910

1,073,380

onroad

308

NLCD Low + Med + High



40,054

722

62

62,127

onroad

506

Education



629

15

1

637

rail

261

NT AD Total Railroad Density

13

33,389

996

15

1,647

rail

271

NT AD Class 12 3 Railroad Density

313

525,992

14,823

442

24,435

rwc

300

NLCD Low Intensity Development

15,439

31,282

316,943

7,703

340,941

For 36US3 modeling in the 2016 platforms, most U.S. emissions sectors were processed using 36-km
spatial surrogates, and if applicable, 36-km meteorology. Exceptions include:

- For the onroad and onroad ca adj sectors, 36US3 emissions were aggregated from 12US1 by
summing emissions from a 3x3 group of 12-km cells into a single 36-km cell. Differences in 12-
km and 36-km meteorology can introduce differences in onroad emissions, and so this approach
ensures that the 36-km and 12-km onroad emissions are consistent. However, this approach means
that 36US3 onroad does not include emissions in Southeast Alaska; therefore, Alaska onroad
emissions are included in a separate sector called onroadnonconus that is processed for only the
36US3 domain. The 36US3 onroad nonconus emissions are spatially allocated using 36-km
surrogates and processed with 36-km meteorology.

Similarly to onroad, because afdust emissions incorporate meteorologically-based adjustments,
afdust adj emissions for 36US3 were aggregated from 12US1 to ensure consistency in emissions
between modeling domains. Again, similarly to onroad, this means 36US3 afdust does not include
emissions in Southeast Alaska; therefore, Alaska afdust emissions are processed in a separate
sector called afdustakadj. The 36US3 afdustakadj emissions are spatially allocated using 36-
km surrogates and adjusted with 36-km meteorology.

The ag and rwc sectors are processed using 36-km spatial surrogates, but using temporal profiles
based on 12-km meteorology.

3.4.2	Allocation method for airport-related sources in the U.S.

There are numerous airport-related emission sources in the NEI, such as aircraft, airport ground support
equipment, and jet refueling. The modeling platform includes the aircraft and airport ground support
equipment emissions as point sources. For the modeling platform, the EPA used the SMOKE "area-to-
point" approach for only jet refueling in the nonpt sector. The following SCCs use this approach:
2501080050 and 2501080100 (petroleum storage at airports), and 2810040000 (aircraft/rocket engine
firing and testing). The ARTOPNT approach is described in detail in the 2002 platform documentation:
http://www3.epa.gov/scram001/reports/Emissions%20TSD%20Voll 02-28-08.pdf. The ARTOPNT file
that lists the nonpoint sources to locate using point data were unchanged from the 2005-based platform.

3.4.3	Surrogates for Canada and Mexico emission inventories

Spatial surrogates for allocating Mexico municipio level emissions have been updated in the 2014v7.1
platform and carried forward into the 2016 alpha platform. For the 2016v7.2 platform, a new set of
Canada shapefiles were provided by Environment Canada along with cross references spatially allocate
the year 2015 Canadian emissions. Gridded surrogates were generated using the Surrogate Tool

155


-------
(previously referenced); Table 3-24 provides a list. Due to computational reasons, total roads (1263) were
used instead of the unpaved rural road surrogate provided. The population surrogate was recently updated
for Mexico; surrogate code 11, which uses 2015 population data at 1 km resolution, replaces the previous
population surrogate code 10. The other surrogates for Mexico are circa 1999 and 2000 and were based
on data obtained from the Sistema Municipal de Bases de Datos (SIMBAD) de INEGI and the Bases de
datos del Censo Economico 1999. Most of the CAPs allocated to the Mexico and Canada surrogates are
shown in Table 3-25.

Table 3-24. Canadian Spatial Surrogates

Code

Canadian Surrogate Description

Code

Description







TOTAL INSTITUTIONAL AND

100

Population

923

GOVERNEMNT

101

total dwelling

924

Primary Industry

104

capped total dwelling

925

Manufacturing and Assembly

106

ALL INDUST

926

Distribution and Retail (no petroleum)

113

Forestry and logging

927

Commercial Services

200

Urban Primary Road Miles

932

CANRAIL

210

Rural Primary Road Miles

940

PAVED ROADS NEW

211

Oil and Gas Extraction

945

Commercial Marine Vessels

212

Mining except oil and gas

946

Construction and mining

220

Urban Secondary Road Miles

948

Forest

221

Total Mining

951

Wood Consumption Percentage

222

Utilities

955

UNPAVED ROADS AND TRAILS

230

Rural Secondary Road Miles

960

TOTBEEF

233

Total Land Development

970

TOTPOUL

240

capped population

980

TOTS WIN

308

Food manufacturing

990

TOTFERT

321

Wood product manufacturing

996

urban area

323

Printing and related support activities

1251

OFFR TOTFERT

324

Petroleum and coal products manufacturing

1252

OFFR MINES

326

Plastics and rubber products manufacturing

1253

OFFR Other Construction not Urban

327

Non-metallic mineral product manufacturing

1254

OFFR Commercial Services

331

Primary Metal Manufacturing

1255

OFFR Oil Sands Mines

350

Water

1256

OFFR Wood industries CANVEC

412

Petroleum product wholesaler-distributors

1257

OFFR UNPAVED ROADS RURAL

448

clothing and clothing accessories stores

1258

OFFR Utilities

482

Rail transportation

1259

OFFR total dwelling

562

Waste management and remediation services

1260

OFFR water

901

AIRPORT

1261

OFFR ALL INDUST

902

Military LTO

1262

OFFR Oil and Gas Extraction

903

Commercial LTO

1263

OFFR ALLROADS

904

General Aviation LTO

1265

OFFR CANRAIL

921

Commercial Fuel Combustion

9450

Commercial Marine Vessel Ports

156


-------
Table 3-25. CAPs Allocated to Mexican and Canadian Spatial Surrogates (short tons in 36US3)

Sector

Code

Mexican or Canadian Surrogate Description

nh3

NOx

pm25

so2

voc

othafdust

106

CAN ALL INDUST

—

—

5,632

"

"

othafdust

212

CAN Mining except oil and gas

—

—

684

—

—

othafdust

221

CAN Total Mining

—

—

142,940

—

—

othafdust

222

CAN Utilities

—

—

23,640

—

—

othafdust

940

CAN Paved Roads New

—

—

210,336

—

—

othafdust

955

CANUNPAVED ROADS AND TRAILS

—

—

389,775

—

—

othafdust

960

CAN TOTBEEF

—

—

1,289

—

—

othafdust

970

CAN TOTPOUL

—

—

184

—

—

othafdust

980

CAN TOTS WIN

—

—

792

—

—

othafdust

990

CAN TOTFERT

—

—

321

—

—

othafdust

996

CAN urban area

—

—

617

—

—

othar

11

MEX 2015 Population

164,464

168,447

13,521

1,164

291,178

othar

14

MEX Residential Heating - Wood

0

23,842

305,597

3,658

2,101,03
3

othar

16

MEX Residential Heating - Distillate Oil

2

58

1

16

2

othar

20

MEX Residential Heating - LP Gas

0

26,526

838

0

505

othar

22

MEX Total Road Miles

1

1,046

2

7

2,308

othar

24

MEX Total Railroads Miles

0

63,136

1,407

551

2,494

othar

26

MEX Total Agriculture

713,253

399,070

80,458

18,650

33,742

othar

32

MEX Commercial Land

0

457

7,719

0

106,077

othar

34

MEX Industrial Land

8

3,383

4,833

1

563,953

othar

36

MEX Commercial plus Industrial Land

0

0

0

0

272,155

othar

38

MEX Commercial plus Institutional Land

3

6,740

235

3

148

othar

40

MEX Residential (RESl-4)+Commercial+
Industrial+Institutional+Government

0

16

39

0

331,216

othar

42

MEX Personal Repair (COM3)

0

0

0

0

26,261

othar

44

MEX Airports Area

0

13,429

306

1,561

3,766

othar

50

MEX Mobile sources - Border Crossing

5

161

1

3

293

othar

100

CAN Population

761

54

669

15

241

othar

101

CAN total dwelling

0

0

0

0

150,892

othar

104

CAN Capped Total Dwelling

421

37,205

2,766

206

1,952

othar

113

CAN Forestry and logging

185

2,210

11,310

45

6,246

othar

211

CAN Oil and Gas Extraction

0

31

60

22

925

othar

212

CAN Mining except oil and gas

0

0

3,079

0

0

othar

221

CAN Total Mining

0

0

43

0

0

othar

222

CAN Utilities

34

1,858

0

386

22

othar

308

CAN Food manufacturing

0

0

20,185

0

10,324

othar

321

CAN Wood product manufacturing

874

4,822

1,646

383

16,606

othar

323

CAN Printing and related support activities

0

0

0

0

11,770

othar

324

CAN Petroleum and coal products manufacturing

0

1,205

1,542

486

9,304

othar

326

CAN Plastics and rubber products manufacturing

0

0

0

0

23,283

othar

327

CAN Non-metallic mineral product manufacturing

0

0

6,695

0

0

othar

331

CAN Primary Metal Manufacturing

0

158

5,595

30

72

othar

350

CAN Water

0

120

2

0

4

othar

412

CAN Petroleum product wholesaler-distributors

0

0

0

0

45,257

157


-------
Sector

Code

Mexican or Canadian Surrogate Description

nh3

NOx

pm25

so2

voc

othar

448

CAN clothing and clothing accessories stores

0

0

0

0

149

othar

482

CAN Rail Transportation

2

4,980

106

12

310

othar

562

CAN Waste management and remediation services

271

1,977

2,710

2,528

13,138

othar

901

CAN Airport

0

109

11

0

11

othar

921

CAN Commercial Fuel Combustion

243

23,628

2,333

2,821

1,091

othar

923

CAN TOTAL INSTITUTIONAL AND
GOVERNEMNT

0

0

0

0

14,859

othar

924

CAN Primary Industry

0

0

0

0

40,376

othar

925

CAN Manufacturing and Assembly

0

0

0

0

71,198

othar

926

CAN Distribution and Retail (no petroleum)

0

0

0

0

7,461

othar

927

CAN Commercial Services

0

0

0

0

32,167

othar

932

CAN CANRAIL

61

132,985

3,107

485

6,567

othar

946

CAN Construction and Mining

0

0

0

0

4,359

othar

951

CAN Wood Consumption Percentage

1,950

21,662

179,087

3,095

253,523

othar

990

CAN TOTFERT

48

4,456

0

9,881

164

othar

1251

CAN OFFR TOTFERT

81

77,166

5,671

58

7,176

othar

1252

CAN OFFR MINES

1

1,004

70

1

138

othar

1253

CAN OFFR Other Construction not Urban

66

53,671

6,096

47

12,159

othar

1254

CAN OFFR Commercial Services

40

17,791

2,552

34

44,338

othar

1255

CAN OFFR Oil Sands Mines

18

9,491

311

10

1,025

othar

1256

CAN OFFR Wood industries CANVEC

9

5,856

476

7

1,318

othar

1257

CAN OFFR Unpaved Roads Rural

32

11,866

1,169

28

49,975

othar

1258

CAN OFFR Utilities

8

5,579

349

7

1,087

othar

1259

CAN OFFR total dwelling

16

5,768

773

14

15,653

othar

1260

CAN OFFR water

15

4,356

451

29

28,411

othar

1261

CAN OFFR ALL INDUST

4

5,770

253

3

1,049

othar

1262

CAN OFFR Oil and Gas Extraction

0

368

29

0

143

othar

1263

CAN OFFR ALLROADS

3

2,418

244

2

582

othar

1265

CAN OFFR CANRAIL

0

85

9

0

15

onroad_
can

200

CAN Urban Primary Road Miles

1,619

85,558

2,851

329

8,396

onroad_
can

210

CAN Rural Primary Road Miles

683

51,307

1,673

139

3,807

onroad_
can

220

CAN Urban Secondary Road Miles

3,021

136,582

5,708

690

22,374

onroad_
can

230

CAN Rural Secondary Road Miles

1,769

96,911

3,238

374

10,370

onroad_
can

240

CAN Total Road Miles

43

57,401

1,355

77

103,658

onroad_
mex

11

MEX 2015 Population

0

281,317

1,873

533

291,992

onroad_
mex

22

MEX Total Road Miles

10,321

1,208,461

54,823

25,855

251,931

onroad_
mex

36

MEX Commercial plus Industrial Land

0

7,975

142

29

9,192

158


-------
3.5

Preparation of Emissions for the CAMx model

3.5.1 Development of CAMx Emissions for Standard CAMx Runs

To perform air quality modeling with the Comprehensive Air Quality Model with Extensions (CAMx
model), the gridded hourly emissions output by the SMOKE model are output in the format needed by the
CMAQ model, but must be converted to the format required by CAMx. For "regular" CAMx modeling
(i.e., without two-way nesting), the CAMx conversion process consists of the following:

1)	Convert all emissions file formats from the I/O API NetCDF format used by CMAQ to the UAM
format used by CAMx, including the merged, gridded low-level emissions files that include
biogenics

2)	Shift hourly emissions files from the 25 hour format used by CMAQ to the averaged 24 hour
format used by CAMx

3)	Rename and aggregate model species for CAMx

4)	Convert 3D wildland and agricultural fire emissions into CAMx point format

5)	Merge all inline point source emissions files together for each day, including layered fire
emissions originally from SMOKE

6)	Add sea salt aerosol emissions to the converted, gridded low-level emissions files

Conversion of file formats from I/O API to UAM is performed using a program called "cmaq2uam". In
the CAMx conversion process, all SMOKE outputs are passed through this step first. Unlike CMAQ, the
CAMx model does not have an inline biogenics option, and so for the purposes of CAMx modeling,
emissions from SMOKE must include biogenic emissions.

One difference between CMAQ-ready emissions files and CAMx-ready emissions files involves hourly
temporalization. A daily emissions file for CMAQ includes data for 25 hours, where the first hour is 0:00
GMT of a given day, and the last hour is 0:00 GMT of the following day. For the CAMx model, a daily
emissions file must only include data for 24 hours, not 25. Furthermore, to match the hourly configuration
expected by CAMx, each set of consecutive hourly timesteps from CMAQ-ready emissions files must be
averaged. For example, the first hour of a CAMx-ready emissions file will equal the average of the first
two hours from the corresponding CMAQ-ready emissions file, and the last (24th) hour of a CAMx-ready
emissions file will equal the average of the last two hours (24th and 25th) from the corresponding CMAQ-
ready emissions file. This time conversion is incorporated into each step of the CAMx-ready emissions
conversion process.

The CAMx model uses a slightly different version of the CB6 speciation mechanism than does the
CMAQ model. SMOKE prepares emissions files for the CB6 mechanism used by the CMAQ model
("CB6-CMAQ"), and therefore, the emissions must be converted to the CB6 mechanism used by the
CAMx model ("CB6-CAMx") during the CAMx conversion process. In addition to the mechanism
differences, CMAQ and CAMx also occasionally use different species naming conventions. For CAMx
modeling, we also create additional tracer species. A summary of the differences between CMAQ input
species and CAMx input species for CB6 (VOC), AE6 (PM2.5), and other model species, is provided in
Table 3-26. Each step of the CAMx-ready emissions conversion process includes conversion of CMAQ
species to CAMx species using a species mapping table which includes the mappings in Table 3-26.

159


-------
Table 3-26. Emission model species mappings for CMAQ and CAMx

Inventory Pollutant

CMAQ Model Species

CAMx Model Species

Cl2

CL2

CL2

HC1

HCL

HCL

CO

CO

CO

NOx

NO

NO



N02

N02



HONO

HONO

S02

S02

S02



SULF

SULF

nh,

NH3

NH3



NH3 FERT

n/a (not used in CAMx)

voc

ACET

ACET



ALD2

ALD2



ALDX

ALDX



BENZ

BENZ and BNZA (duplicate species)



CH4

CH4



ETH

ETH



ETHA

ETHA



ETHY

ETHY



ETOH

ETOH



FORM

FORM



IOLE

IOLE



ISOP

ISOP and ISP (duplicate species)



KET

KET



MEOH

MEOH



NAPH + XYLMN (sum)

XYL



NVOL

n/a (not used in CAMx)



OLE

OLE



PAR

PAR



PRPA

PRPA



SESQ

SQT



SOAALK

n/a (not used in CAMx)



TERP

TERP and TRP (duplicate species)



TOL

TOL and TOLA (duplicate species)



UNR + NR (sum)

NR

PM10

PMC

CPRM

PM2.5

PEC

PEC



PN03

PN03



POC

POC



PS04

PS04



PAL

PAL



PCA

PCA



PCL

PCL



PFE

PFE



PK

PK



PH20

PH20



PMG

PMG



PMN

PMN



PMOTHR

PMOTHR and FPRM (duplicate species)



PNA

NA

160


-------
Inventory Pollutant

CMAQ Model Species

CAMx Model Species



PNCOM

PNCOM



PNH4

PNH4



PSI

PSI



PTI

PTI



POC + PNCOM (sum)

POA1



PAL + PCA + PFE +

FCRS1



PMG + PK + PMN +





PSI + PTI (sum)



1 The POA species, which is the sum of POC and PNCOM, is passed to the CAMx model in addition to individual species POC
and PNCOM. The FCRS species, which is also a sum of multiple PM species, is passed to CAMx in addition to each of the
individual component species.

One feature which is part of CMAQ and is not part of CAMx involves plume rise for fires. For CMAQ
modeling, we process fire emissions through SMOKE as inline point sources, and plume rise for fires is
calculated within CMAQ using parameters from the inline emissions files (heat flux, etc). This is similar
to how non-fire point sources are handled, except that the fire parameters are used to calculate plume rise
instead of traditional stack parameters. The CAMx model supports inline plume rise calculations using
traditional stack parameters, but, does not support inline plume rise for fire sources. Therefore, for the
purposes of CAMx modeling, we must have SMOKE calculate plume rise for fires using the Laypoint
program. In this modeling platform, this must be done for the ptfire, ptfire othna, and ptagfire sectors. To
distinguish these layered fire emissions from inline fire emissions, layered fire emissions are processed
with the sector names "ptfire3D", "ptfire_othna3D", and "ptagfire3D". When converting layered fire
emissions files to CAMx format, stack parameters are added to the CAMx-ready fire emissions files to
force the correct amount of fire emissions into each layer for each fire location.

CMAQ modeling uses one gridded low-level emissions file, plus multiple inline point source emissions
files, per day. CAMx modeling also uses one gridded low-level emissions file per day - but instead of
reading multiple inline point source emissions files at once, CAMx can only read a single point source file
per day. Therefore, as part of the CAMx conversion process, all inline point source files are merged into a
single "mrgpt" file per day. The mrgpt file includes the layered fire emissions described in the previous
paragraph, in addition to all non-fire elevated point sources from the cmv_c3, othpt, ptegu, ptnonipm, and
pt oilgas sectors.

The remaining step in the CAMx emissions process is to generate sea salt aerosol emissions, which are
distinct from ocean chlorine emissions. Sea salt emissions do not need to be included in CMAQ-ready
emissions because they are calculated by the model, but, do need to be included in CAMx-ready
emissions. After the merged low-level emissions are converted to CAMx format, sea salt emissions are
generated using a program called "seasalt" and added to the low-level emissions. Sea salt emissions
depend on meteorology, vary on a daily and hourly basis, and exist for model species PCL, NA, PS04,
and SS (i.e., sea salt).

3.5.2 Development of CAMx Emissions for Source Apportionment
CAMx Runs

The CAMx model supports source apportionment modeling for ozone and PM sources using techniques
called Ozone Source Apportionment Technology (OSAT) and Particulate Matter Source Apportionment
Technology (PSAT). These source apportionment techniques allow emissions from different types of
sources to be tracked through the CAMx model. For the Revised CSAPR Update study, OSAT modeling
was performed in CAMx for 2023 and 2028 using one-way nesting (i.e., the inner 12km grid takes

161


-------
boundary information from the outer 36km grid but the inner grid does not feed any concentration
information back to the outer grid). The emissions developed specifically for OSAT modeling used the
case names "2023fhl_ussa_16j" and "2028fhl_ussa_16j".

Source Apportionment modeling involves assigning tags to different categories of emissions. These tags
can be applied by region (e.g., state), by emissions type (e.g., SCC or sector), or a combination of the two.
For the Revised CSAPR Update study, emissions tagging was applied by state. All emissions from US
states, except for biogenics, fires, and fugitive dust (afdust), were assigned a state-specific tag. Emissions
from tribal lands were also assigned a separate tag, as well as offshore emissions. Other tags include a tag
for biogenics and afdust; a tag for all fires, both inside and outside the US; and a tag for all anthropogenic
emissions from Canada and Mexico. A full list of tags is provided in Table 3-27. State-level tags 2
through 51 exclude emissions from biogenics, fugitive dust, and fires, which are included in other tags.

Table 3-27. State tags for 2023fhl, 2028fhl USSA modeling

Tag

Emissions applied to tag

1

All biogenics (beis sector) and US fugitive dust (afdust sector)

2

Alabama

3

Arizona

4

Arkansas

5

California

6

Colorado

7

Connecticut

8

Delaware

9

District of Columbia

10

Florida

11

Georgia

12

Idaho

13

Illinois

14

Indiana

15

Iowa

16

Kansas

17

Kentucky

18

Louisiana

19

Maine

20

Maryland

21

Massachusetts

22

Michigan

23

Minnesota

24

Mississippi

25

Missouri

26

Montana

27

Nebraska

28

Nevada

29

New Hampshire

30

New Jersey

31

New Mexico

32

New York

162


-------
Tag

Emissions applied to tag

33

North Carolina

34

North Dakota

35

Ohio

36

Oklahoma

37

Oregon

38

Pennsylvania

39

Rhode Island

40

South Carolina

41

South Dakota

42

Tennessee

43

Texas

44

Utah

45

Vermont

46

Virginia

47

Washington

48

West Virginia

49

Wisconsin

50

Wyoming

51

Tribal Data

52

Canada and Mexico (except fires)

53

Offshore

54

All fires from US, Canada, and Mexico, including ag fires

For OSAT and PSAT modeling, all emissions must be input to CAMx in the form of a point source
(mrgpt) file, including low level sources that are found in gridded files for regular CAMx runs. In
addition, for two-way nested modeling, all emissions must be input in a single mrgpt file, rather than
separate mrgpt files for each of the two domains (36US3 and 12US2). Note that fire emissions require
special consideration in two-way nested model runs and for PSAT and OSAT modeling. That same
consideration must be given to any sector in which emissions are being gridded by SMOKE.

There are two main approaches for tagging emissions for CAMx modeling. One approach is to tag
emissions within SMOKE. Here, SMOKE will output tagged point source files (SGINLN files), which
can then be converted to CAMx point source format with the tags applied by SMOKE carried forward
into the CAMx inputs. The second approach is to, if necessary, depending on the nature of the tags, split
sectors into multiple components by tag so that each sector corresponds to a single tag. Then, the gridded
and/or point source format SMOKE outputs from those split sectors are converted to CAMx point source
format, and then merged into the full mrgpt file, with the tags applied at that last step. In some situtations,
a mix of the two approaches is appropriate.

For the Revised CSAPR Update study the first approach was used for most sectors, meaning tags were
applied in SMOKE. The exceptions were sectors where the entire sector receives only one tag: afdust,
beis, onroad ca adj, ptfire, ptagfire, ptfire othna, and all Canada and Mexico sectors. Afdust emissions
are not tagged by state because the current tagging methodology does not support applying transportable
fraction and meteorological adjustments to tagged emissions.

163


-------
Once the individual sector tagging is complete, the point source files for all of the sectors are merged
together to create the mrgpt file which includes all emissions, with the desired tags and appropriate
resolution throughout the domain for OSAT or PSAT modeling.

164


-------
4 Development of Future Year Emissions

The emission inventories for future years of 2023 and 2028 have been developed using projection
methods that are specific to the type of emissions source. Future emissions are projected from the 2016
base case either by running models to estimate future year emissions from specific types of emission
sources (e.g., EGUs, and onroad and nonroad mobile sources), or for other types of sources by adjusting
the base year emissions according to the best estimate of changes expected to occur in the intervening
years (e.g., non-EGU point and nonpoint sources). For some sectors, the same emissions are used in the
base and future years, such as biogenic and fire. For the remaining sectors, rules and specific legal
obligations that go into effect in the intervening years, along with changes in activity for the sector, are
considered when possible. These sectors have been projected to 2023 and 2028 as summarized in Table
4-1. The development of the 2021fi emissions for each sector is also discussed.

Table 4-1. Overview of projection methods for the 2023 and 2028 regional cases

Platform Sector:

abbreviation

Description of Projection Methods for 2023 and 2028

EGU units:

Ptegu

The Integrated Planning Model (IPM) was run to create the 2023 and 2028
emissions. IPM outputs from the January, 2020 version of the IPM platform were
used (httDs://www.eDa.gov/airmarkets/eDas-DOwer-sector-modeling-
platform-v6-using-ipm-ianuarv-2020-reference-case). For 2023. the 2023 IPM

output year was used and for 2028 the 2030 output year was used because the year
2028 maps to the 2030 output year. Emission inventory Flat Files for input to
SMOKE were generated using post-processed IPM output data. Temporal
allocation for future year emissions is discussed in the EGU-IPM specification
sheet for the 2016vl platform. For 2021fi, an engineering analysis-based inventory
was used. The inventory is available in Docket ID No. EPA-HQ-OAR-2020-0272
as "Final Rule State Emission Budgets Calculations and Engineering Analytics".

Point source oil and
gas:

ptoilgas

First, known closures were applied to the 2016 pt_oilgas sources. Production-
related sources were then grown from 2016 to 2017 using historic production data.
The production-related sources were then grown to 2023 and 2028 based on
growth factors derived from the Annual Energy Outlook (AEO) 2019 data for oil,
natural gas, or a combination thereof. The grown emissions were then controlled
to account for the impacts of relevant New Source Performance Standards (NSPS).
For 202 lfi, a set of projection and control factors for 2021 were developed
consistently with those used for 2023fh and applied to 2016fh inventories.

Remaining non-
EGU point:

Ptnonipm

First, known closures were applied to the 2016 ptnonipm sources. Closures were
obtained from the Emission Inventory System (EIS) and also submitted by the
states of Alabama, North Carolina, Ohio, Pennsylvania, and Virginia. Industrial
sources were grown using factors derived from the AEO 2019. Rail yard emissions
were grown using the same factors as line haul locomotives in the rail sector.
Controls were then applied to account for relevant NSPS for reciprocating internal
combustion engines (RICE), gas turbines, and process heaters. Reductions due to
consent decrees that had not been fully implemented by 2016 were also applied,
along with specific comments received by S/L/T agencies. For 20216, most
emissions were interpolated between 2016fi and 2023, additional closures were
implemented and new sources were added based on 2018NEI, and Pennsylvania
emissions were updated based on feedback from MARAMA. Rail yards were
interpolated between 2016 and 2023.

165


-------
Platform Sector:

abbreviation

Description of Projection Methods for 2023 and 2028

Airports

Starts with 2017 NEI. Airport emissions were grown using factors derived from
the Terminal Area Forecast (TAF) (see

https://www.faa.aov/data research/av iation/taf/). For 2021. a set of projection
factors consistent with 2023fhl were developed, and then applied to the corrected
2017 NEI emissions. Corrections to emissions for ATL from the state of Georgia
were also implemented.

Agricultural:

Ag

Livestock were projected based on factors created from USDA National livestock
inventory projections published in February 2018

(https://www.ers.usda.aov/webdocs/outlooks/87459/oce-2018-1.pdf?v=7587).
Fertilizer emissions were held constant at year 2016 levels. For 20216, the
emissions were interpolated between 2016 and 2023.

Area fugitive dust:

afdust, afdust ak

Paved road dust was grown to 2023 and 2028 levels based on the growth in VMT
from 2016 to 2023 and 2028. The remainder of the sector including building
construction, road construction, agricultural dust, and unpaved road dust was held
constant, except in the MARAMA region where some factors were provided for
categories other than paved roads. The projected emissions are reduced during
modeling according to a transport fraction (newly computed for the beta platform)
and a meteorology-based (precipitation and snow/ice cover) zero-out as they are
for the base year. For 20216, the emissions were interpolated between 2016 and
2023.

Category 1, 2 CMV:

cmv_clc2

Category 1 and category 2 (C1C2) CMV emissions sources outside of California
were projected to 2023 and 2028 based on factors from the Regulatory Impact
Analysis (RIA) Control of Emissions of Air Pollution from Locomotive Engines
and Marine Compression Ignition Engines Less than 30 Liters per Cylinder.
California emissions were projected based on factors provided by the state. For
20216, projection factors consistent with 2023fhl were developed and applied to
the 2016fh emissions. Canada emission were interpolated between 2015 and 2023.

Category 3 CMV:

cmv_c3

Category 3 (C3) CMV emissions were projected using a forthcoming EPA report
on projected bunker fiiel demand. The report projects bunker fiiel consumption by
region out to the year 2030. Bunker fiiel usage was used as a surrogate for marine
vessel activity. Factors based on the report were used for all pollutants except
NOx. Growth factors for NOx emissions were handled separately to account for
the phase in of Tier 3 vessel engines. The NOx growth rates from the EPA C3
Regulatory Impact Assessment (RIA) were refactored to use the new bunker fuel
usage growth rates. The assumptions of changes in fleet composition and
emissions rates from the C3 RIA were preserved and applied to the new bunker
fuel demand growth rates for 2023 and 2028 to arrive at the final growth rates. For
20216, projection factors consistent with 2023fhl were developed and applied to
the 2016fh emissions. Canada emission were interpolated between 2015 and 2023.

Locomotives:
rail

Passenger and freight were projected using separate factors. Freight emissions
were computed for future years based on future year fuel use values for 2020,
2023, and 2028. Specifically, they were based on AEO2018 freight rail energy use
growth rate projections and emission factors, which are based on historic
emissions trends that reflect the rate of market penetration of new locomotive
engines. For 20216, the emissions were interpolated between 2016 and 2023.

166


-------
Platform Sector:

abbreviation

Description of Projection Methods for 2023 and 2028

Remaining
nonpoint:

nonpt

Industrial emissions were grown according to factors derived from AEO2019.
Portions of the nonpt sector were grown using factors based on expected growth in
human population. Controls were applied to reflect relevant NSPS rules (i.e.,
reciprocating internal combustion engines (RICE), natural gas turbines, and
process heaters). Emissions were also reduced to account for fuel sulfur rules in
the mid-Atlantic and northeast. For 20216, most emissions were interpolated
between 2016 and 2023 and cellulosic emissions were removed after consultation
with the EPA Office of Transportation and Air Quality.

Nonpoint source oil
and gas:
npoilgas

Production-related sources were grown starting from an average of 2014 and 2016
production data. Emissions were initially projected to 2017 using historical data
and then grown to 2023 and 2028 based on factors generated from AEO2019.
Based on the SCC, factors related to oil, gas, or combined growth were used.
Coalbed methane SCCs were projected independently. Controls were then applied
to account for NSPS for oil and gas and RICE. For 2021fi, a set of projection and
control factors for 2021 were developed consistently with those used for 2023fh
and applied to 2016fh inventories.

Residential Wood
Combustion:

rwc

RWC emissions were projected from 2016 to 2023 and 2028 based on growth and
control assumptions compatible with EPA's 201 lv6.3 platform, which accounts
for growth, retirements, and NSPS, although implemented in the Mid-Atlantic
Regional Air Management Association (MARAMA)'s growth tool. RWC
emissions in California, Oregon, and Washington were held constant. For 202 lfi,
emissions were interpolated between 2016 and 2023.

Nonroad:

nonroad

Outside California, the MOVES2014b model was run to create nonroad emissions
for 2023 and 2028 without any state inputs. The fuels used are specific to the
future year, but the meteorological data represented the year 2016. For California,
datasets provided by the California Air Resources Board (CARB) circa 2017 were
used. For 20216, MOVES2014b was run for 2020 and the 2021 emissions were
interpolated between 2020 and 2023. Texas 2021 emissions were interpolated
between 2020 and 2023. California 2021 emissions were interpolated between
2016 and 2023.

Onroad:

onroad,

onroadnonconus

Activity data were projected from 2016 to 2023 and 2028 based on factors derived
from AEO2019. Where S/Ls provided activity data, those data were used. To
create the emission factors, MOVES2014b was run for the years 2023 and 2028,
with 2016 meteorological data and fuels, but with age distributions projected to
represent future years, and the remaining inputs consistent with those used in
2014NEIv2. The future year activity data and emission factors were then
combined using SMOKE-MOVES to produce the 2023 and 2028 emissions.
Section 4.3.2 describes the applicable rules that were considered when projecting
onroad emissions. For 20216, MOVES2014b was run for 2020 and 2020 activity
data were developed by interpolating between 2016 and 2023. Adjustment factors
from 2020 to 2021 were developed by SCC and pollutant from national runs of
MOVES2014b for those two years.

Onroad California:

onroadcaadj

CARB-provided emissions were used for California, but they were gridded and
temporalized using MOVES2014b-based data output from SMOKE-MOVES.
Volatile organic compound (VOC) HAP emissions derived from California-
provided VOC emissions and MOVES-based speciation. For 20216, emissions
were interpolated between 2016 and 2023.

167


-------
Platform Sector:

abbreviation

Description of Projection Methods for 2023 and 2028

Other Area Fugitive
dust sources not
from the NEI:
othafdust

Othafdust emissions for future years were provided by ECCC. The emissions were
extracted from a broader nonpoint source inventory. Adjustments to construction
dust were made to make those more consistent with the 2016 and ECCC 2010
inventories. Mexico emissions are not included in this sector. For 20216,
emissions were interpolated between 2016 and 2023

Other Point Fugitive
dust sources not
from the NEI:
othptdust

Wind erosion emissions were removed from the point fugitive dust inventory prior
to regional haze modeling. Base year 2015 inventories with the rotated grid pattern
removed were projected to 2023 and 2028 based on factors provided by ECCC. A
transport fraction adjustment is applied to the projected inventories along with a
meteorology-based (precipitation and snow/ice cover) zero-out. For 20216,
emissions were interpolated between 2016 and 2023.

Other point sources
not from the NEI:
othpt

For agricultural sources that were originally developed on the rotated 10-km grid,
the reallocated base year emissions were projected to 2023 and 2028 using
projection factors based on data provided by ECCC and applied by province,
pollutant, and ECCC sub-class code. Airports were also projected from 2016 using
ECCC-based factors. For the remaining sources in this sector, ECCC provided
future year inventories. For Mexico sources, inventories projected from Mexico's
2008 inventory to 2018, 2025, and 2030 were interpolated to the years 2023 and
2028. For 20216, emissions were interpolated between 2016 and 2023 except 2023
emissions were used for three inventories provided by ECCC that had unique
sources for each year.

Other non-NEI
nonpoint and
nonroad:

othar

Future year nonpoint inventories for many parts of this sector were provided by
ECCC and were split into sectors to match those in the base year inventory. For
Canadian nonroad sources, factors were provided from which the future year
inventories could be derived. For Mexico nonpoint and nonroad sources,
inventories projected to 2018, 2025, and 2030 from their 2008 inventory were
interpolated to 2023 and 2028. For 20216, emissions were interpolated between
2016 and 2023 except for one ECCC inventory for which 2023 emissions were
used directly because only 2023 emissons were available.

Other non-NEI
onroad sources:

onroadcan

For Canadian mobile onroad sources, fiiture year inventories were derived from
the base year 2015 inventory and data provided by ECCC. Projection factors were
applied by province, sub-class code, and pollutant. For 20216, emissions were
interpolated between 2016 and 2023.

Other non-NEI
onroad sources:

onroad mex

Monthly year Mexico (municipio resolution) onroad mobile inventories were
developed based runs of MOVES-Mexico for 2023 and 2028. For 20216,
emissions were interpolated between 2016 and 2023.

168


-------
4.1 EGU Point Source Projections (ptegu)

The original 2023fh and 2028fh EGU emissions inventories were developed from the output of the v6
platform using the May 2019 reference case run, while the 2023fhl and 2028fhl emissions are based on
the January 2020 reference case run of the Integrated Planning Model (IPM). IPM is a linear
programming model that accounts for variables and information such as energy demand, planned unit
retirements, and planned rules to forecast unit-level energy production and configurations. The following
specific rules and regulations are included in IPM v6 platform run from May 2019:

•	The Cross-State Air Pollution Rule (CSAPR) Update, a federal regulatory measure to address
transport of ozone and its precursors under the 1997 and 2008 National Ambient Air Quality
Standards (NAAQS) for ozone.

•	The Standards of Performance for Greenhouse Gas Emissions from New, Modified, and
Reconstructed Stationary Sources: Electric Utility Generating Units.

•	The Mercury and Air Toxics Rule (MATS), which was initially finalized in 2011 and later revised
(https://www.epa.gov/mats/regulatorv-actions-final-mercury-and-air-toxics-standards-mats-power-
plants). MATS establishes National Emissions Standards for Hazardous Air Pollutants (NESHAP)
for the "electric utility steam generating unit" source category.

•	Current and existing state regulations.

•	The final actions EPA has taken to implement the Regional Haze Regulations and Guidelines for
Best Available Retrofit Technology (BART) Determinations Final Rule. This regulation requires
states to submit revised State Implementation Plans (SIPs) that include (1) goals for improving
visibility in Class I areas on the 20% worst days and allowing no degradation on the 20% best
days and (2) assessments and plans for achieving BART emission targets for sources placed in
operation between 1962 and 1977. Since 2010, EPA has approved SIPs or, in a very few cases, put
in place regional haze Federal Implementation Plans for several states. The BART limits approved
in these plans (as of summer 2017) that will be in place for EGUs are represented in the EPA
Platform v6.

•	Three non-air federal rules affecting EGUs: National Pollutant Discharge Elimination System-
Final Regulations to Establish Requirements for Cooling Water Intake Structures at Existing
Facilities and Amend Requirements at Phase I Facilities, Hazardous and Solid Waste Management
System; Disposal of Coal Combustion Residuals From Electric Utilities; and the Effluent
Limitation Guidelines and Standards for the Steam Electric Power Generating Point Source
Category.

Some additional updates were made to IPM for the January 2020 case which includes rules that were in
effect by September 2019 along with other updates that are reflected in the 2023fhl and 2028fhl
emissions inventories:

•	Updated NEEDS to the December 2019 version. This included more than 10 GW of
retirements, 4 GW of which were coal plants, along with some unit-level rate changes in Utah,
Nebraska, Kentucky, and New York.

•	Updated (i.e., lowered) storage and renewal energy technology costs based on the National
Renewable Energy Laboratory (NREL) Annual Technology Baseline 2019 mid case.

•	Implemented offshore wind power mandates in Maryland, New Jersey,

Connecticut, Massachusetts, and New York .

169


-------
•	Incorporated clean energy standards in California, New Mexico, Nevada, New York, and
Washington.

•	Implemented renewable portfolio standard updates in California, Washington D.C., Maryland,
Maine, New Mexico, Nevada, New York, Ohio, and Washington.

•	Reflected the Affordable Clean Energy (ACE) rule (June 19, 2019).

•	Incorporated the 26 U.S. Code § 45Q. Credit for carbon oxide sequestration
(https://www.energv.gov/sites/prod/files/2019/10/f67/Internal%20Revenue%20Code%20Tax
%20Fact%20Sheet.pdf).

IPM is run for a set of years, including the 2023 and 202830 future years used in the 2016vl platform.
Further documentation of the IPM model and the v6 platform can be found on the CAMD website
(https://www.epa.gov/airmarkets/documentation-epas-power-sector-modeling-platform-v6-ianuarv-202Q-
reference-case).

The EGU missions are calculated for the inventory using the output of the IPM model for the forecast
year. Units that are identified to have a primary fuel of landfill gas, fossil waste, non-fossil waste, residual
fuel oil, or distillate fuel oil may be missing emissions values for certain pollutants in the generated
inventory flat file. Units with missing emissions values are gapfilled using projected base year values. The
projections are calculated using the ratio of the future year seasonal generation in the IPM parsed file and
the base year seasonal generation at each unit for each fuel type in the unit as derived from the 2016 EIA-
923 tables. New controls identified at a unit in the IPM parsed file are accounted for with appropriate
emissions reductions in the gapfill projection values. When base year unit-level generation data cannot be
obtained no gapfill value is calculated for that unit. Additionally, some units, such as landfill gas, may not
be assigned a valid SCC in the initial flat file. The SCCs for these units are updated based on the base
year SCC for the unit-fuel type.

Combined cycle units produce some of their energy from process steam that turns a steam turbine. The
IPM model assigns a fraction of the total combined cycle production to the steam turbine. When the
emissions are calculated these steam units are assigned emissions values that come from the combustion
portion of the process. In the base year NEI steam turbines are usually implicit to the total combined cycle
unit. To achieve the proper plume rise for the total combined cycle emissions, the stack parameters for the
steam turbine units are updated with the parameters from the combustion release point.

Large EGUs in the IPM-derived flat file inventory are associated with hourly CEMS data for NOX and
S02 emissions values in the base year. To maintain a temporal pattern consistent with the 2016 base year,
the NOX and S02 values in the hourly CEMS inventories are projected to match the total seasonal
emissions values in the future years.

The EGU sector NOx emissions by state are listed in Table 4-2 for 2023 and 2028 regional cases. The
designation "fh" here refers to the May 2019 IPM case and "fhl" refers to the January 2020 IPM case.

30 2028 is not a specific output year for IPM, but 2028 maps to the 2030 output year. The IPM inputs were adjusted to make it
more suitable for modeling of 2028.

170


-------
Table 4-2. EGU sector NOx emissions by State for the 2023 and 2028 regional cases

State

2016fh

2023fh

2023fhl

2028fh

2028fhl

Alabama

28,596

9,545

9,954

11,812

12,376

Arizona

18,786

10,909

11,175

9,259

9,011

Arkansas

26,808

11,579

17,461

15,318

17,074

California

6,908

7,501

5,808

2,707

1,719

Colorado

30,152

17,965

16,561

18,616

15,448

Connecticut

4,088

4,359

4,365

4,249

4,202

Delaware

1,487

367

488

407

544

District of Columbia

NA

1

1

1

1

Florida

65,059

32,327

32,684

33,282

31,488

Georgia

29,384

14,292

13,760

15,950

15,666

Idaho

1,369

469

469

949

419

Illinois

30,250

31,189

21,321

32,474

21,668

Indiana

83,425

44,029

45,169

44,971

45,328

Iowa

22,971

23,069

24,264

22,976

23,379

Kansas

14,959

15,669

15,725

15,684

14,528

Kentucky

57,342

14,411

14,316

11,761

14,495

Louisiana

47,931

17,223

18,145

16,179

16,909

Maine

4,935

3,016

3,005

2,557

2,945

Maryland

10,448

5,387

5,436

5,115

5,599

Massachusetts

8,121

5,851

5,819

5,626

5,683

Michigan

37,149

30,141

28,344

31,948

32,895

Minnesota

21,737

15,565

17,497

15,364

12,665

Mississippi

16,414

5,749

5,604

6,248

6,135

Missouri

57,647

46,714

48,809

46,528

45,433

Montana

15,819

9,186

9,186

9,193

9,018

Nebraska

20,734

21,428

21,451

21,508

21,468

Nevada

3,949

2,215

2,368

1,458

1,531

New Hampshire

2,158

601

590

533

529

New Jersey

5,723

5,771

5,889

6,135

6,582

New Mexico

20,222

8,246

9,332

6,532

6,542

New York

13,770

14,740

14,552

13,699

13,707

North Carolina

27,892

30,088

29,482

21,685

24,320

North Dakota

38,400

25,458

25,772

25,314

24,151

Ohio

55,581

40,029

45,211

38,572

43,345

171


-------
State

2016fh

2023fh

2023fhl

2028fh

2028fhl

Oklahoma

25,084

17,877

17,396

17,342

16,375

Oregon

4,067

1,560

1,827

1,665

1,791

Pennsylvania

84,086

33,301

31,707

31,326

28,769

Rhode Island

261

769

764

739

737

South Carolina

13,734

13,460

13,474

13,053

13,048

South Dakota

1,095

692

756

832

776

Tennessee

18,752

4,285

5,896

4,753

5,958

Texas

111,612

81,051

82,699

80,579

77,506

Tribal Data

35,057

6,897

6,907

6,902

6,854

Utah

27,450

21,063

14,455

20,991

13,986

Vermont

302

21

21

20

20

Virginia

26,387

10,183

10,050

11,217

11,899

Washington

8,860

1,760

1,909

1,809

1,875

West Virginia

50,984

41,891

41,992

39,495

39,601

Wisconsin

16,148

10,238

10,467

10,048

9,293

Wyoming

36,095

15,216

17,463

13,300

13,371

4.2 Non-EGU Point and Nonpoint Sector Projections

To project all U.S. non-EGU stationary sources, facility/unit closures information and growth
(PROJECTION) factors and/or controls were applied to certain categories within the afdust, ag, cmv, rail,
nonpt, np oilgas, ptnonipm, pt oilgas and rwc platform sectors. Some facility or sub-facility-level
closure information was also applied to the point sources. There are also a handful of situations where
new inventories were generated for sources that did not exist in the 2014v2 NEI (e.g., biodiesel and
cellulosic plants, yet-to-be constructed cement kilns). This subsection provides details on the data and
projection methods used for these sectors.

Because much of the projections and controls data are developed independently from how the EPA
defines its emissions modeling sectors, this section is organized primarily by the type of projections data,
with secondary consideration given to the emissions modeling sector (e.g., industrial source growth
factors are applicable to four emissions modeling sectors). The rest of this section is organized in the
order that the EPA uses the Control Strategy Tool (CoST) in combination with other methods to produce
future year inventories: 1) for point sources, apply plant (facility or sub-facility-level) closure information
via CoST; 2) apply all PROJECTION packets via CoST (multiplicative factors that could cause increases
or decreases); 3) apply all percent reduction-based CONTROL packets via CoST; and 4) append all other
future-year inventories not generated via CoST. This organization allows consolidation of the discussion
of the emissions categories that are contained in multiple sectors, because the data and approaches used
across the sectors are consistent and do not need to be repeated. Sector names associated with the CoST
packets are provided in parentheses. The projection and control factors applied by CoST to prepare the
2023fhl and 2028fhl emissions are provided on the 2016vl FTP site and in the docket for the final
Revised Cross-state Air Pollution Rule Update (RCU) (see https://regulations.gov EPA-HQ-OAR-2020-
0272).

172


-------
4.2.1 Background on the Control Strategy Tool (CoST)

CoST is used to apply most non-EGU projection/growth factors, controls and facility/unit/stack-level
closures to the 2016-based emissions modeling inventories to create future year inventories for the
following sectors: afdust, ag, cmv, rail, nonpt, np oilgas, ptnonipm, ptoilgas and rwc. Information
about CoST and related data sets is available from https://www.epa.gov/economic-and-cost-analvsis-air-
pollution-regulations/cost-analvsis-modelstools-air-pollution.

CoST allows the user to apply projection (growth) factors, controls and closures at various geographic
and inventory key field resolutions. Each of these CoST datasets, also called "packets" or "programs,"
provides the user with the ability to perform numerous quality assurance assessments as well as create
SMOKE-ready future year inventories. Future year inventories are created for each emissions modeling
sector via a CoST "strategy" and each strategy includes all base year 2016 inventories and applicable
CoST packets. For reasons discussed later, some emissions modeling sectors require multiple CoST
strategies to account for the compounding of control programs that impact the same type of sources.

There are also available linkages to existing and user-defined control measures databases and it is up to
the user to determine how control strategies are developed and applied. The EPA typically creates
individual CoST packets that represent specific intended purposes (e.g., aircraft projections for airports
are in a separate PROJECTION packet from residential wood combustion sales/appliance turnover-based
projections). CoST uses three packet types as described below:

1.	CLOSURE: Applied first in CoST. This packet can be used to zero-out (close) point source
emissions at resolutions as broad as a facility to as specific as a stack. The EPA uses these types
of packets for known post-2016 controls as well as information on closures provided by states on
specific facilities, units or stacks. This packet type is only used in the ptnonipm and pt oilgas
sectors.

2.	PROJECTION: This packet allows the user to increase or decrease emissions for virtually any
geographic and/or inventory source level. Projection factors are applied as multiplicative factors
to the 2011 emissions inventories prior to the application of any possible subsequent CONTROLS.
A PROJECTION packet is necessary whenever emissions increase from 2011 and is also desirable
when information is based more on activity assumptions rather than known control measures. The
EPA uses PROJECTION packet(s) in every non-EGU modeling sector.

3.	CONTROL: These packets are applied after any/all CLOSURE and PROJECTION packet entries.
The user has similar level of control as PROJECTION packets regarding specificity of geographic
and/or inventory source level application. Control factors are expressed as a percent reduction (0
to 100) and can be applied in addition to any pre-existing inventory control, or as a replacement
control where inventory controls are first backed out prior to the application of a more-stringent
replacement control.

All of these packets are stored as data sets within the Emissions Modeling Framework and use comma-
delimited formats. As mentioned above, CoST first applies any/all CLOSURE information for point
sources, then applies PROJECTION packet information, followed by CONTROL packets. A hierarchy is
used by CoST to separately apply PROJECTION and CONTROL packets. In short, in a separate process
for PROJECTION and CONTROL packets, more specific information is applied in lieu of less-specific
information in ANY other packets. For example, a facility-level PROJECTION factor will be replaced by
a unit-level, or facility and pollutant-level PROJECTION factor. It is important to note that this hierarchy
does not apply between packet types (e.g., CONTROL packet entries are applied irrespective of
PROJECTION packet hierarchies). A more specific example: a state/SCC-level PROJECTION factor
will be applied before a stack/pollutant-level CONTROL factor that impacts the same inventory record.

173


-------
However, an inventory source that is subject to a CLOSURE packet record is removed from consideration
of subsequent PROJECTION and CONTROL packets.

The implication for this hierarchy and intra-packet independence is important to understand and quality
assure when creating future year strategies. For example, with consent decrees, settlements and state
comments, the goal is typically to achieve a targeted reduction (from the 2011NEI) or a targeted future-
year emissions value. Therefore, as encountered with this future year base case, consent decrees and state
comments for specific cement kilns (expressed as CONTROL packet entries) needed to be applied instead
of (not in addition to) the more general approach of the PROJECTION packet entries for cement
manufacturing. By processing CoST control strategies with PROJECTION and CONTROL packets
separated by the type of broad measure/program, it is possible to show actual changes from the base year
inventory to the future year inventory as a result of applying each packet.

Ultimately, CoST concatenates all PROJECTION packets into one PROJECTION dataset and uses a
hierarchal matching approach to assign PROJECTION factors to the inventory. For example, a packet
entry with Ranking=l will supersede all other potential inventory matches from other packets. CoST then
computes the projected emissions from all PROJECTION packet matches and then performs a similar
routine for all CONTROL packets. Therefore, when summarizing "emissions reduced" from CONTROL
packets, it is important to note that these reductions are not relative to the 2011 inventory, but rather to the
intermediate inventory after application of any/all PROJECTION packet matches (and CLOSURES). A
subset of the more than 70 hierarchy options is shown in Table 4-3, although the fields in the table are not
necessarily named the same in CoST, but rather are similar to those in the SMOKE FF10 inventories. For
example, "REGIONCD" is the county-state-county FIPS code (e.g., Harris county Texas is 48201) and
"STATE" would be the 2-digit state FIPS code with three trailing zeroes (e.g., Texas is 48000).

Table 4-3. Subset of CoST Packet Matching Hierarchy

Rank

Matching Hierarchy

Inventory Type

1

REGION CD, FACILITY ID, UNIT ID, REL POINT ID, PROCESS ID, SCC, POLL

point

2

REGION CD, FACILITY ID, UNIT ID, REL POINT ID, PROCESS ID, POLL

point

3

REGION CD, FACILITY ID, UNIT ID, REL POINT ID, POLL

point

4

REGION CD, FACILITY ID, UNIT ID, POLL

point

5

REGION CD, FACILITY ID, SCC, POLL

point

6

REGION CD, FACILITY ID, POLL

point

7

REGION CD, FACILITY ID, UNIT ID, REL POINT ID, PROCESS ID, SCC

point

8

REGION CD, FACILITY ID, UNIT ID, REL POINT ID, PROCESS ID

point

9

REGION CD, FACILITY ID, UNIT ID, REL POINT ID

point

10

REGION CD, FACILITY ID, UNIT ID

point

11

REGION CD, FACILITY ID, SCC

point

12

REGION CD, FACILITY ID

point

13

REGION CD, NAICS, SCC, POLL

point, nonpoint

14

REGION CD, NAICS, POLL

point, nonpoint

15

STATE, NAICS, SCC, POLL

point, nonpoint

16

STATE, NAICS, POLL

point, nonpoint

17

NAICS, SCC, POLL

point, nonpoint

18

NAICS, POLL

point, nonpoint

19

REGION CD, NAICS, SCC

point, nonpoint

20

REGION CD, NAICS

point, nonpoint

21

STATE, NAICS, SCC

point, nonpoint

22

STATE, NAICS

point, nonpoint

23

NAICS, SCC

point, nonpoint

174


-------
Rank

Matching Hierarchy

Inventory Type

24

NAICS

point, nonpoint

25

REGION CD, SCC, POLL

point, nonpoint

26

STATE, SCC, POLL

point, nonpoint

27

SCC, POLL

point, nonpoint

28

REGION CD, SCC

point, nonpoint

29

STATE, SCC

point, nonpoint

30

SCC

point, nonpoint

31

REGION CD, POLL

point, nonpoint

32

REGION CD

point, nonpoint

33

STATE, POLL

point, nonpoint

34

STATE

point, nonpoint

35

POLL

point, nonpoint

The contents of the controls, local adjustments and closures for the future year base case are described in
the following subsections. Year-specific projection factors (PROJECTION packets) for the future year
were used to create the future year base case, unless noted otherwise in the specific subsections. The
contents of a few of these projection packets (and control reductions) are provided in the following
subsections where feasible. However, most sectors used growth or control factors that varied
geographically and their contents could not be provided in the following sections (e.g., facilities and units
subject to the Boiler MACT reconsideration has thousands of records). The remainder of Section 4.2 is
divided into several subsections that are summarized in Table 4-4. Note that future year inventories were
used rather than projection or control packets for some sources.

Table 4-4. Summary of non-EGU stationary projections subsections

Subsection

Title

Sector(s)

Brief Description

4.2.2

CoST Plant CLOSURE
packet

ptnonipm,
ptoilgas

All facility/unit/stack closures information,
primarily from Emissions Inventory System (EIS),
but also includes information from states and other
organizations.

4.2.3

CoST PROJECTION
packets

All

Introduces and summarizes national impacts of all
CoST PROJECTION packets to the future year.

4.2.3.1

Fugitive dust growth

afdust

PROJECTION packet: county-level resolution,
primarily based on VMT growth.

4.2.3.2

Livestock population
growth

ag

PROJECTION packet: national, by-animal type
resolution, based on animal population
projections.

4.2.3.3

Category 1, 2, and 3
commercial marine
vessels

cmv

PROJECTION packet: Category 1 & 2: CMV uses
SCC/poll for all states except Calif.

4.2.3.4

Category 3 commercial
marine vessels

cmv

PROJECTION packet: Category 3: region-level
by-pollutant, based on cumulative growth and
control impacts from rulemaking.

4.2.3.5

Oil and gas and industrial
source growth

nonpt,
npoilgas,
ptnonipm,
ptoilgas

Several PROJECTION packets: varying
geographic resolutions from state, county, to
oil/gas play-level and by-process/fuel-type
applications. Data derived from AEO2019 with
several modifications.

175


-------
Subsection

Title

Sector(s)

Brief Description

4.2.3.6

Non-IPM Point Sources

ptnonipm

Several PROJECTION packets: specific
projections from MARAMA region and states,
EIA-based projection factors for industrial sources
for non-MARAMA states.

4.2.3.7

Nonpoint sources

nonpt

Several PROJECTION packets: MARAMA states
projection for Portable Fuel Containers and for all
other nonpt sources. Non-MARAMA states
projected with EIA-based factors for industrial
sources. Evaporative Emissions from Finished
Fuels projected using EIA-based factors. Human
population used as growth for applicable sources.

4.2.3.8

Airport Sources

ptnonipm

PROJECTION packet: by-airport for all direct
matches to FAA Terminal Area Forecast data,
with state-level factors for non-matching NEI
airports.

4.2.3.9

Residential wood
combustion

rwc

PROJECTION packet: national with exceptions,
based on appliance type sales growth estimates
and retirement assumptions and impacts of recent
NSPS.

4.2.4

CoST CONTROL packets

ptnonipm,
nonpt,
npoilgas,
pt oilgas

Introduces and summarizes national impacts of all
CoST CONTROL packets to the future year.

4.2.4.1

Oil and Gas NSPS

npoil

gas,

pt oilgas



4.2.4.2

RICE NSPS

ptnonipm,
nonpt,
npoilgas,
pt oilgas

CONTROL packet: applies reductions for lean
burn, rich burn, and combined engines for
identified SCCs.

4.2.4.3

Fuel Sulfur Rules

ptnonipm,
nonpt

CONTROL packet: updated by MARAMA,
applies reductions to specific units in ten states.

4.2.4.4

Natural Gas Turbines
NOx NSPS

ptnonipm

CONTROL packet: applies NOx emission
reductions established by the NSPS.

4.2.4.5

Process Heaters NOx
NSPS

ptnonipm

CONTROL packet: applies NOx emission limits
established by the NSPS.

4.2.4.6

CISWI

ptnonipm

CONTROL packet: applies controls to specific
CISWI units in 11 states.

4.2.4.7

Petroleum Refineries
NSPS Subpart JA

ptnonipm

CONTROL packet: control efficiencies are
applied to identified delayed coking and storage
tank units.

4.2.4.8

State-Specific Controls

ptnonipm

CONTROL packets and comments submitted by
individual states for rules that may only impact
their state or corrections noted from previous
review.

176


-------
4.2.2 CoST Plant CLOSURE Packet (ptnonipm, pt_oilgas)

Packets:

CLOSURES2016_beta_platform_04oct2019_v 1 (for2023fhl and2028fhl)
CLOSURES2016_beta_platform_l9aug2020_nf_v2 (for 202lfi)

The CLOSURES packet contains facility, unit and stack-level closure information derived from an
Emissions Inventory System (EIS) unit-level report from March 5, 2019, with closure status equal to "PS"
(permanent shutdown; i.e., post-2016 permanent facility/unit shutdowns known in EIS as of the date of
the report). In addition, comments on past modeling platforms received by states and other agencies
specified additional closures, as well as some previously specified closures which should remain open, in
the following states: Alabama, North Carolina, Ohio, Pennsylvania, and Virginia. The list of closures for
2021fi also includes two Pennsylvania facilities that were only partially closed in prior runs, but in 2021fi
are completely closed: Pittsburgh Corning Corp - Port Allegany (ID 3025211), and Osram Sylvania Inc. -
Wellsboro Plant (ID 5490611). Ultimately, all data were updated to match the SMOKE FF10 inventory
key fields, with all duplicates removed, and a single CoST packet was generated. These changes impact
sources in the ptnonipm and ptoilgas sectors. The cumulative reduction in emissions for ptnonipm are
shown in Table 4-5.

Table 4-5. Reductions from all facility/unit/stack-level closures in 2016vl

Pollutant

ptnonipm

ptoilgas

CO

1,010

187

NH3

59

0

NOX

1,373

284

PM10

447

9

PM2.5

358

9

S02

727

178

VOC

2,211

106

4.2.3 CoST PROJECTION Packets (afdust, ag, cmv, rail, nonpt,
np_°ilgas, ptnonipm, pt_oilgas, rwc)

As previously discussed, for point inventories, after application of any/all CLOSURE packet information,
the next step in running a CoST control strategy is the application of all CoST PROJECTION packets.
Regardless of inventory type (point or nonpoint), the PROJECTION packets applied prior to the CoST
packets. For several emissions modeling sectors (i.e., afdust, ag, cmv, rail and rwc), there is only one
CoST PROJECTION packet. For all other sectors, there are several different sources of PROJECTIONS
data and, therefore, there are multiple PROJECTION packets that are concatenated and quality-assured
for duplicates and applicability to the inventories in the CoST strategy. The PROJECTION (and
CONTROL) packets were separated into a few "key" control program types to allow for quick summaries
of these distinct control programs. The remainder of this section is broken out by CoST packet, with the
exception of discussion of the various packets used for oil and gas and industrial source projections; these
packets are a mix of different sources of data that target similar sources.

MARAMA provided PROJECTION and CONTROL packets for years 2023 and 2028 for states
including: Connecticut, Delaware, Maryland, Massachusetts, New Hampshire, New York, New Jersey,
North Carolina, Pennsylvania, Rhode Island, Vermont, Virginia, West Virginia, Maine, and the District of

177


-------
Columbia. MARAMA only provided pt oilgas and np oilgas packets for Rhode Island, Maryland and
Massachusetts. For the 202lfi case, new projection factors for sources affected by the Pennsylvania
Reasonably Available Control Technology (RACT) II were included in the projections. Also for 202lfi,
MARAMA provided 2023 emissions directly for one Pennsylvania facility (Anchor Hocking LLC,
Monaca Plant) affected by the rule; for that facility, emissions values were swapped in after applying all
other projections and controls. For states not covered by the MARAMA packets, projection factors were
developed using nationally available data and methods.

4.2.3.1 Fugitive dust growth (afdust)

Packets:

Proj ection_2016_2023_afdust_version l_platform_MARAMA_04oct2019_v 1
Proj ection_2016_2023_afdust_version l_platform_NJ_l 3 sep2019_v0
Proj ection_2016_2023_afdust_version l_platform_national_04oct2019_v 1
Proj ection_2016_2023_all_nonpoint_version l_platform_NC_04oct2019_v2
Proj ection_2016_2028_afdust_version l_platform_MARAMA_04oct2019_v 1
Proj ection_2016_2028_afdust_version l_platform_NJ_l 3 sep2019_v0
Proj ection_2016_2028_afdust_version l_platform_national_04oct2019_v 1
Proj ection_2016_2028_all_nonpoint_version l_platform_NC_04oct2019_v2

MARAMA States

MARAMA submitted projection factors for their states to project 2016 afdust emissions to future years
2023 and 2028. These county-specific projection factors impacted paved roads (SCC 2294000000),
residential construction dust (SCC 2311010000), industrial/commercial/institutional construction dust
(SCC 2311020000), road construction dust (SCC 2311030000), dust from mining and quarrying (SCC
2325000000), agricultural crop tilling dust (SCC 2801000003), and agricultural dust kick-up from beef
cattle hooves (SCC 2805001000). Other afdust emissions, including unpaved road dust emissions, were
held constant in future year projections. Note that North Carolina and New Jersey provided their own
packets for this sector.

Non-MARAMA States

For paved roads (SCC 2294000000), the 2016 afdust emissions were projected to future years 2023 and
2028 based on differences in county total VMT:

Future year afdust paved roads = 2016 afdust paved roads * (Future year county total VMT) / (2016
county total VMT)

The VMT projections are described in the onroad section.

All emissions other than paved roads are held constant in future year projections. The impacts of the
projections are shown in Table 4-6.

Table 4-6. Increase in total afdust PM2.5 emissions from projections in 2016vl

2016 Emissions

2023 Emissions

percent Increase
2023

2028 Emissions

percent Increase
2028

2,530,625

2,557,970

1.09%

2,570,714

1.60%

178


-------
4.2.3.2 Livestock population growth (ag)

Packets:

Proj ection_2016_2023_all_nonpoint_version l_platform_NC_04oct2019_v2
Proj ection_2016_2028_all_nonpoint_version l_platform_NC_04oct2019_v2
Proj ection_2017_2023_ag_version l_platform_l 1 sep2019_v0
Proj ection_2017_2023_ag_version l_platform_NJ_l 1 sep2019_v0
Proj ection_2017_2028_ag_version l_platform_l 1 sep2019_v0
Proj ection_2017_2028_ag_version l_platform_NJ_l 1 sep2019_v0

The 2017NEI livestock emissions were projected to year 2023 and 2028 using projection factors created
from USDA National livestock inventory projections published in March 2019
(https://www.ers.usda.gov/publications/pub-details/?pubid=92599) and are shown in Table 4-7. For
emission projections to 2023, a ratio was created between animal inventory counts for 2023 and 2017 to
create a projection factor. This process was completed for the animal categories of beef, dairy, broilers,
layers, turkeys, and swine. The projection factor was then applied to the 2017NEI base emissions for the
specific animal type to estimate 2023 NH3 and VOC emissions. For emission projections to 2028, the
same projection method was used. New Jersey (NJ) provided NJ-specific projection factors that were used
to grow livestock waste emissions from 2017 to 2023 and 2028. North Carolina (NC) provided NC-
specific projection factors that used a 2016-based projection, therefore, NC's livestock waste emissions
are projected from the 2016 back-casted base year emissions to 2023 and 2028.

Table 4-7. National projection factors for livestock: 2016 to 2023 and 2028

Animal

2023

2028

beef

-0.02%

-2.87%

swine

+7.47%

+10.36%

broilers

+8.60%

+12.50%

turkeys

-0.03%

+1.57%

layers

+9.28%

+15.93%

dairy

+0.92%

+1.24%

4.2.3.3 Category 1, Category 2 Commercial Marine Vessels (cmv_c1c2)

Packets:

Proj ection_2016_2023_cmv_c 1 c2_version l_platform_04oct2019_v 1
Proj ection_2016_2023_cmv_Canada_versionl_platform_24sep2019_v0
Proj ection_2016_2028_cmv_c 1 c2_version l_platform_04oct2019_v 1
Proj ection_2016_2028_cmv_Canada_versionl_platform_24sep2019_v0

The cmv_clc2 emissions outside of California were projected from 2016 to 2023 and 2028 using factors
derived from the Regulatory Impact Analysis (RIA) Control of Emissions of Air Pollution from
Locomotive Engines and Marine Compression Ignition Engines Less than 30 Liters per Cylinder
(https://www.epa.gov/regulations-emissions-vehicles-and-engines/final-rule-control-emissions-air-
pollution-locomotive). Table 4-8 lists the pollutant-specific projection factors to 2023, and 2028 that were
used for cmv_clc2 sources outside of California. California sources were projected to 2023 and 2028
using the factors in Table 4-9, which are based on data provided by CARB.

179


-------
Table 4-8. National projection factors for cmv_clc2

Pollutant

20l6-to-2023 (%)

20I6-1O-2028 (%)

20I6-1O-2023

20I6-1O-2028

CO

-1.3%

0.3%

0.987

1.003

NOX

-29.3%

-44.6%

0.707

0.554

PM10

-28.3%

-43.4%

0.717

0.566

PM2.5

-28.3%

-43.4%

0.717

0.566

S02

-65.3%

-65.9%

0.347

0.341

VOC

-31.5%

-47.2%

0.685

0.528

Table 4-9. California projection factors for cmv_clc2

Pollutant

20l6-to-2023 (%)

20I6-1O-2028 (%)

20I6-1O-2023

20I6-1O-2028

CO

20.1%

25.3%

1.201

1.253

NOX

-15.0%

-17.7%

0.850

0.823

PM10

-29.9%

-33.5%

0.701

0.665

PM2.5

-29.9%

-33.5%

0.701

0.665

S02

24.1%

48.7%

1.241

1.487

VOC

1.5%

1.9%

1.015

1.019

4.2.3.4 Category 3 Commercial Marine Vessels (cmv_c3)

Packets:

Proj ection_2016_2023_cmv_c3_version l_platform_04oct2019_v2_Mexico
Proj ection_2016_2023_cmv_c3_version l_platform_24sep2019_v 1
Proj ection_2016_2023_cmv_Canada_versionl_platform_24sep2019_v0
Proj ection_2016_2028_cmv_c3_version l_platform_04oct2019_v2_Mexico
Proj ection_2016_2028_cmv_c3_version l_platform_24sep2019_v 1
Proj ection_2016_2028_cmv_Canada_versionl_platform_24sep2019_v0

Growth rates for cmv_c3 emissions from 2016 to 2023 and 2028 were developed using a forthcoming
EPA report on projected bunker fuel demand. The report projects bunker fuel consumption by region out
to the year 2030. Bunker fuel usage was used as a surrogate for marine vessel activity. To estimate future
year emissions of CO, C02, hydrocarbons, PM10, and PM2.5, the bunker fuel growth rate from 2016 to
2023, and 2028 were directly applied to the estimated 2016 emissions.

Growth factors for NOx emissions were handled separately to account for the phase in of Tier 3 vessel
engines. To estimate these emissions, the NOx growth rates from the EPA C3 Regulatory Impact
Assessment (RIA)31 were refactored to use the new bunker fuel usage growth rates. The assumptions of
changes in fleet composition and emissions rates from the C3 RIA were preserved and applied to the new
bunker fuel demand growth rates for 2023, and 2028 to arrive at the final growth rates. The Category 3
marine diesel engines Clean Air Act and International Maritime Organization standards from April, 2010
(https://www.epa.gov/regulations-emissions-vehicles-and-engines/final-rule-control-emissions-new-
marine-compression-O) were also considered for emission estimates.

31 https://nepis.epa.gov/Exe/ZvPURL.cgi?Dockev=P1005ZGH.TXT

180


-------
The 2023 and 2028 projection factors are shown in Table 4-10. Some regions for which 2016 projection
factors were available did not have 2023 or 2028 projection factors specific to that region, so factors from
another region were used as follows:

•	Alaska was projected using North Pacific factors.

•	Hawaii was projected using South Pacific factors.

•	Puerto Rico and Virgin Islands were projected using Gulf Coast factors.

•	Emissions outside Federal Waters (FIPS 98) were projected using the factors given in
Table 4-10 for the region "Other".

•	California was projected using a separate set of state-wide projection factors based on
CMV emissions data provided by the California Air Resources Board (CARB). These
factors are shown in Table 4-11

Table 4-10. 2016-to-2023 and 2016-2028 CMV C3 projection factors outside of California

Region

2016-io-2023

2016-lo-2023

2016-lo-2028

2016-lo-2028



\()\

other polliiliinls

\()\

oilier polliiliinls

US East Coast

-6.05%

27.71%

-7.54%

49.71%

US South Pacific









(ex. California)

-24.79%

20.89%

-33.97%

45.86%

US North Pacific

-3.37%

22.57%

-4.07%

41.31%

US Gulf

-6.88%

20.82%

-12.40%

36.41%

US Great Lakes

8.71%

14.55%

19.80%

28.29%

Other

23.09%

23.09%

42.58%

42.58%

Non-IVilernl Waters

2016-lo-2023

2016-lo-2028

S02

-77.21%

-73.60%

PM (main engines)

-36.06%

-25.93%

PM (aux. engines)

-39.69%

-30.14%

Other pollutants

+23.09%

+42.58%

Table 4-11. 2016-to-2023 and 2016-2028 CMV C3 projection factors for California

I'olllltillll

20l6-lo-2023

20I6-IO-2028

CO

1.180

1.340

Nox

1.156

1.327

PMio / PM2.5

1.205

1.381

S02

1.183

1.332

VOC

1.242

1.461

181


-------
4.2.3.5 Oil and Gas Sources (pt_oilgas, np_oilgas)

Packets:

Proj ection_2016_202X_pt_oilgas_P A_NGtrans_fromMARAMA_09sep2019_v0

Proj ection_2016_2023_oilgas_version l_platform_09sep2019_v0

Proj ection_2016_2023_pt_oilgas_version l_platform_VA_NGtrans_l 6sep2019_v0

Proj ection_2016_2028_oilgas_version l_platform_09sep2019_v0

Proj ection_2016_2028_pt_oilgas_version l_platform_VA_NGtrans_l 6sep2019_v0

Proj ection_2016_2023_oilgas_version l_platform_09sep2019_v0

Proj ection_2016_2028_oilgas_version l_platform_09sep2019_v0

Future year projections for the 2016vl platform were generated for point oil and gas sources for years
2023 and 2028. These projections consisted of three components: (1) applying facility closures to the
ptoilgas sector using the CoST CLOSURE packet; (2) using historical and/or forecast activity data to
generate future-year emissions before applicable control technologies are applied using the CoST
PROJECTION packet; and (3) estimating impacts of applicable control technologies on future-year
emissions using the CoST CONTROL packet. Applying the CLOSURE packet to the pt oilgas sector
resulted in small emissions changes to the national summary shown inTable 4-5. Note the closures for
years 2023 and 2028 are the same.

For pt oilgas growth to 2023 and 2028, the oil and gas sources were separated into production-related and
exploration-related sources by SCC. These sources were further subdivided by fuel-type by SCC into
either OIL, natural gas (NGAS), BOTH oil-natural gas fuels possible, or coal-bed methane (CBM). The
next two subsections describe the growth component process.

For npoilgas growth to 2023 and 2028, oil and gas sources were separated into production-related,
transmission-related, and all other point sources by NAICS. These sources are further subdivided by fuel-
type by SCC into either OIL, natural gas (NGAS), or BOTH oil-natural gas fuels possible.

Production-related Sources (pt oilgas, np oilgas)

The growth factors for the production-related NAICS-SCC combinations were generated in a two-step
process. The first step used historical production data at the state-level to get state-level short-term trends
or factors from 2016 to year 2017. In some cases, historical data for year 2018 were available for a state,
in these cases a 2016 to 2018 factor was calculated. These historical data were acquired from EIA from
the following links:

•	Historical Natural Gas: http://www.eia.gov/dnav/ng/ng sum lsum a epgO fgw mmcf a.htm

•	Historical Crude Oil: http://www.eia.gov/dnav/pet/pet crd crpdn adc mbbl a.htm

•	Historical CBM: https://www.eia.gov/dnav/ng/ng prod coalbed si a.htm

The second step involved using the Annual Energy Outlook (AEO) 2019 reference case for the Lower 48
forecast production tables to project from year 2017 to the years of 2023 and 2028. Specifically, AEO
2019 Table 60 "Lower 48 Crude Oil Production and Wellhead Prices by Supply Region " and AEO 2019
Table 61 "Lower 48 Natural Gas Production and Supply Prices by Supply Region " were used in this
projection process. The AEO2019 forecast production is supplied for each EIA Oil and Gas Supply
region shown in Figure 4-1.

182


-------
Figure 4-1. EI A Oil and Gas Supply Regions as of AEO2019

Pacific

The result of this second step is a growth factor for each Supply Region from 2017 (or 2018) to 2023 and
from 2017 (or 2018) to 2028. A Supply Region mapping to FIPS cross-walk was developed so the
regional growth factors could be applied for each FIPS (for pt_oilgas) or to the county-level np_oilgas
inventories. Note that portions of Texas are in three different Supply Regions and portions of New
Mexico are in two different supply regions. The state-level historical factor (2016 to 2017 or 2018) was
then multiplied by the Supply Region factor (2017 or 2018 to future years) to produce a state-level or
FlPS-level factor to grow from 2016 to 2023 and from 2016 to 2028. This process was done using crude
production forecast information to generate a factor to apply to oil-production related SCCs or NAICS-
SCC combinations and it was also done using natural gas production forecast information to generate a
factor to apply to natural gas-production related NAICS-SCC combinations. For the N AICS-SCC
combinations that are designated "BOTH" the average of the oil-production and natural-gas production
factors was calculated and applied to these specific combinations.

The state of Texas provided specific technical direction for growth of production-related point sources.
Texas provided updated basin specific production for 2016 and 2017 to allow for a better calculation of
the estimated growth for this one-year period. The AEO2019 was used as described above for the three
AEO Oil and Gas Supply Regions that include Texas counties to grow from 2017 to 2023 and 2028 years.
However, Texas only wanted these growth factors applied to sources in the Permian and Eagle Ford
basins. The oil and gas production point sources in the other basins in Texas were not grown (i.e.,
2016vl=2023=2028 emissions).

Transmission-related Sources (pt oilgas)

Projection factors were generated using the same AEO2019 tables used for production sources. The
growth factors for transmission sources were developed solely using AEO 2019 data by Oil and Gas
Supply Regions shown in Figure 4-1. Additionally, limits were put on these regional factors where the
minimum factor was set to l.Oand the maximum factor was set to 1.5. The states of Virginia and

183


-------
Pennsylvania provided source specific growth factors for natural gas transmission sources to be used in
place of the AEO regional factors.

Exploration-related Sources (npoilgas)

Due to Year 2016 being a low exploration activity year when compared to exploration activity in other
recent years, Years 2014 through 2017 exploration activity data were averaged and the average activity
input into EPA's Oil and Gas Tool to produce "averaged" emissions for exploration sources (Table 4-12).
This four-year average (2014-2017) activity data were used because they were readily available for use
with the 2016vl platform. These averaged emissions were used for both the 2023 and 2028 future years in
the 2016vl emissions modeling platform. Colorado, Pennsylvania, California, and Oklahoma submitted
inventories for use. Note CoST was not used for this step for exploration sources.

Table 4-12. Year 2014-2017 high-level summary of national oil and gas exploration activity

Parameter (all US states)

Year2014

Year2015

Year2016

Year2017

4-year
average

Total Well Completions

40,306

22,754

15,605

21,850

25,129

Unconventional Well
Completions

20,896

11,673

7,610

11,617

12,949

Total Oil Spuds

36,104

17,240

7,014

14,322

18,670

Total Natural Gas Spuds

4,750

3,168

4,244

4,025

4,047

Total Coalbed Methane Spuds

239

130

141

222

183

Total Spuds

41,093

20,538

11,399

18,569

22,900

Total Feet Drilled

327,832,580

178,297,779

106,468,774

181,164,800

198,440,983

4.2.3.6 Non-EGU point sources (ptnonipm)

Packets:

Proj ection_2016_202X_ptnonipm_version l_platform_WI_supplement_25 sep2019_v0
Proj ection_2016_2023_corn_ethanol_E0B0_Volpe_27 sep2019_v0
Proj ection_2016_2023_finished_fuels_volpe_04oct2019_v2

Proj ection_2016_2023_industrial_byNAIC S_SCC_version l_platform_l 3 sep2019_v0

Proj ection_2016_2023_industrial_by SCC_version l_platform_20sep2019_v 1

Projection_2016_2023_ptnonipm_airports_railyards_versionl_platform_NC_nopoll_26sep2019_v0

Proj ection_2016_2023_ptnonipm_version l_platform_MARAMA_l 1 sep2019_nf_v 1

Proj ection_2016_2023_ptnonipm_version l_platform_NJ_l 0sep2019_v0

Proj ection_2016_2023_ptnonipm_version l_platform_VA_04oct2019_v 1

proj ection_2016_2028_corn_ethanol_E0B0_Volpe_l 1 sep2019_v0

Proj ection_2016_2028_finished_fuels_volpe_04oct2019_vl

Proj ection_2016_2028_industrial_byNAIC S_SCC_version l_platform_l 3 sep2019_v0

Proj ection_2016_2028_industrial_by SCC_version l_platform_20sep2019_v 1

Projection_2016_2028_ptnonipm_airports_railyards_versionl_platform_NC_nopoll_26sep2019_v0

Proj ection_2016_2028_ptnonipm_version l_platform_MARAMA_l 1 sep2019_nf_v 1

Proj ection_2016_2028_ptnonipm_version l_platform_NJ_l 0sep2019_v0

Proj ection_2016_2028_ptnonipm_version l_platform_VA_04oct2019_v 1

184


-------
The 2023 and 2028 ptnonipm projections involved several growth and projection methods described here.
The projection of all oil and gas sources is explained in the oil and gas specification sheet and will not be
discussed in these methods.

2023 and 2028 Point Inventory - inside MARAMA region

2016-to-2023 and 2016-to-2028 projection packets for point sources were provided by MARAMA for the
following states: CT, DE, DC, ME, MD, MA, NH, NJ, NY, NC, PA, RI, VT, VA, and WV.

The MARAMA projection packets were used throughout the MARAMA region, except in North
Carolina, New Jersey, and Virginia. Those three states provided their own projection packets for the
ptnonipm sector, and those projection packets were used instead of the MARAMA packets in those states.
The Virginia growth factors for one facility were edited to incorporate emissions limits provided by
MARAMA for that facility.

2023 and 2028 Point Inventory - outside MARAMA region

The Energy Information Administration's (EIA) AEO for year 2019 was used as a starting point for
projecting industrial sources in this sector. SCC's were mapped to AEO categories and projection factors
were created using a ratio between the base year and projection year estimates from each specific AEO
category. Table 4-13 below details the 2019 AEO tables used to map SCCs to AEO categories for the
projections of industrial sources. Depending on the category, a projection factor may be national or
regional. The maximum projection factor was capped at 1.25 and the minimum projection factor was
capped at 0.5. MARAMA states were not projected using this method, nor were aircraft and rail sources.

An SCC-NAICS projection was also developed using AEO2019. SCC/NAICS combinations with
emissions >100tons/year for any CAP were mapped to AEO sector and fuel. Projection factors for this
method were capped at a maximum of 2.5 and a minimum of 0.5.

Table 4-13. EIA's 2019 Annual Energy Outlook (AEO) tables used to project industrial sources

Table #

Table name

2

Energy Consumption by Sector and Source

25

Refining Industry Energy Consumption

26

Food Industry Energy Consumption

27

Paper Industry Energy Consumption

28

Bulk Chemical Industry Energy Consumption

29

Glass Industry Energy Consumption

30

Cement Industry Energy Consumption

31

Iron and Steel Industries Energy Consumption

32

Aluminum Industry Energy Consumption

33

Metal Based Durables Energy Consumption

34

Other Manufacturing Sector Energy Consumption

35

Nonmanufacturing Sector Energy Consumption

185


-------
The state of Wisconsin provided source-specific growth factors for four facilities in the state. For those
facilities, the growth factors provided by Wisconsin were used instead of those derived from the AEO.

4.2.3.7 Nonpoint Sources (nonpt)

Packets:

Proj ection_2016_2023_all_nonpoint_version l_platform_NC_04oct2019_v2

Proj ection_2016_2023_finished_fuels_volpe_04oct2019_v2

Proj ection_2016_2023_industrial_by SCC_version l_platform_20sep2019_v 1

Proj ection_2016_2023_nonpt_other_version l_platform_MARAMA_20sep2019_v 1

Proj ection_2016_2023_nonpt_PFC_version l_platform_MARAMA_20sep2019_v 1

Proj ection_2016_2023_nonpt_population_beta_platform_ext_20sep2019_v 1

Proj ection_2016_2023_nonpt_version l_platform_NJ_04oct2019_v 1

Proj ection_2016_2028_all_nonpoint_version l_platform_NC_04oct2019_v2

Proj ection_2016_2028_finished_fuels_volpe_04oct2019_vl

Proj ection_2016_2028_industrial_by SCC_version l_platform_20sep2019_v 1

Proj ection_2016_2028_nonpt_other_version l_platform_MARAMA_20sep2019_v 1

Proj ection_2016_2028_nonpt_PFC_version l_platform_MARAMA_20sep2019_v 1

Proj ection_2016_2028_nonpt_population_beta_platform_ext_20sep2019_v 1

Proj ection_2016_2028_nonpt_version l_platform_NJ_04oct2019_v 1

Inside MARAMA region

2016-to-2023 and 2016-to-2028 projection packets for all nonpoint sources were provided by MARAMA
for the following states: CT, DE, DC, ME, MD, MA, NH, NJ, NY, NC, PA, RI, VT, VA, and WV.
MARAMA provided one projection packet per year for portable fuel containers (PFCs), and a second
projection packet per year for all other nonpt sources.

The MARAMA projection packets were used throughout the MARAMA region, except in North Carolina
and New Jersey. Both NC and NJ provided separate projection packets for the nonpt sector, and those
projection packets were used instead of the MARAMA packets in those two states. New Jersey did not
provide projection factors for PFCs, and so NJ PFCs were projected using the MARAMA PFC growth
packet.

Industrial Sources outside MARAMA region

The EIA's AEO for year 2019 was used as a starting point for projecting industrial sources in this sector.
SCC's were mapped to AEO categories and projection factors were created using a ratio between the base
year and projection year estimates from each specific AEO category. For the nonpoint sector, only 2018
AEO Table 2 was used to map SCCs to AEO categories for the projections of industrial sources.
Depending on the category, a projection factor may be national or regional. The maximum projection
factor was capped at a factor of 1.25 and the minimum projection factor was capped at 0.5. Aircraft and
rail sources were not projected using this method. Sources within the MARAMA region were not
projected with these factors, but with the MARAMA-provided growth factors.

Evaporative Emissions from Transport of Finished Fuels outside MARAMA region

Estimates on growth of evaporative emissions from transporting finished fuels are partially covered in the
nonpoint and point oil and gas projection packets. However, there are some processes with evaporative

186


-------
emissions from storing and transporting finished fuels which are not included in the nonpoint and point
oil and gas projection packets, e.g., withdrawing fuel from tanks at bulk plants, filling tanks at service
stations, etc., and those processes are included in nonpoint other. The EIA's AEO for year 2018 was used
as a starting point for projecting volumes of finished fuel that would be transported in future years, i.e.,
2023 and 2028. Then these volumes were used to calculate inventories associated with evaporative
emissions in 2016, 2023, and 2028 using the upstream modules. Those emission inventories were
mapped to the appropriate SCCs and projection packets were generated from 2016 to 2023 and 2016 to
2028 using the upstream modules. Sources within the MARAMA region were not projected with these
factors, but with the MARAMA-provided growth factors.

Human Population Growth outside MARAMA region

For SCCs that are projected based on human population growth, population projection data were available
from the Benefits Mapping and Analysis Program (BenMAP) model by county for several years,
including 2017, 2023, and 2028. These human population data were used to create modified county-
specific projection factors. Note that 2017 is being used as the base year since 2016 human population is
not available in this dataset. A newer human population dataset was assessed but it did not have
trustworthy near-term (e.g., 2023/2028) projections, and was not used; for example, rural areas of NC
were projected to have more growth than urban areas, which is the opposite of what one would expect.
Growth factors were limited to a range of 0.9-1.35 for 2023 and 0.85-1.6 for 2028, but none of the factors
fell outside that range. (The 1.35 and 1.6 caps are based on 5% annual growth.) Sources within the
MARAMA region were not projected with these factors, but with the MARAMA-provided growth
factors.

4.2.3.8 Airport sources (airports)

Packets:

airport_proj ections_itn_2017_2023_09sep2019_v0

airport_proj ections_itn_2017_2028_09sep2019_v0

Airport emissions were projected from the 2017 NEI April 2020 release, the original source of the airport
inventory, to 2023 and 2028 mostly using 2018 Terminal Area Forecast (TAF) data available from the
Federal Aviation Administration (https://www.faa.gov/data research/aviation/taf/Y Projection factors
were computed using the ratio of the itinerant (ITN) data from the Airport Operations table between the
base and projection year. For airports not matching a unit in the TAF data, state default growth factors by
itinerant class (commercial, air taxi, and general) were created from the collection of airports unmatched.
Emission growth for facilities is capped at 500% and the state default growth is capped at 200%. Military
state default projection values were kept flat (i.e., equal to 1.0) to reflect uncertainly in the data regarding
these sources. Note: the 2016fh, 2023fhl and 2028fhl cases as modeled for the RCU had commercial
aircraft emissions that were up to twice as high as they should have been due to an error in the 2017 NEI
(April 2020 version) airport emissions.

4.2.4 CoST CONTROL Packets (nonpt, np_oilgas, ptnonipm, pt_oilgas)

The final step in the projection of emissions to a future year is the application of any control technologies
or programs. For future-year New Source Performance Standards (NSPS) controls (e.g., oil and gas,
Reciprocating Internal Combustion Engines (RICE), Natural Gas Turbines, and Process Heaters), we
attempted to control only new sources/equipment using the following equation to account for growth and
retirement of existing sources and the differences between the new and existing source emission rates.

187


-------
Qn = Qo { [ (1 + Pf) t- 1 ] Fn + (1 - Ri) t Fe + [ 1 - (1 - Ri) t] Fn ] }	Equation 4-1

where:

Qn = emissions in projection year
Qo = emissions in base year

Pf = growth rate expressed as ratio (e.g., 1.5=50 percent cumulative growth)
t = number of years between base and future years
Fn = emission factor ratio for new sources

Ri = retirement rate, expressed as whole number (e.g., 3.3 percent=0.033)

Fe = emission factor ratio for existing sources

The first term in Equation 4-1 represents new source growth and controls, the second term accounts for
retirement and controls for existing sources, and the third term accounts for replacement source controls.
For computing the CoST % reductions (Control Efficiency), the simplified Equation 4-2 was used for
2023 and 2028 projections:

r. , r-rr- ¦ /n/A -,nn (a \(Pf202x-i)xFn+(i-Ri)12+(i-(i-Ri)12)xFn]\	Equation 4-2

Control Efficiency202*(%) = 100 x 1 - L			—-1—*—		'—L	L)

V	Pj202X	'

For example, to compute the control efficiency for 2028 from a base year of 2015 the existing source
emissions factor (Fe) is set to 1.0, 2028 (future year) minus 2016 (base year) is 12, and new source
emission factor (Fn) is the ratio of the NSPS emission factor to the existing emission factor. Table 4-14
shows the values for Retirement rate and new source emission factors (Fn) for new sources with respect to
each NSPS regulation and other conditions within. For the nonpt sector, the RICE NSPS control program
was applied when estimating year 2023 and 2028 emissions for the 2016vl modeling platform. Further
information about the application of NSPS controls can be found in Section 4 of the Additional Updates
to Emissions Inventories for the Version 6.3, 2011 Emissions Modeling Platform for the Year 2023
technical support document (https://www.epa.gov/sites/production/files/2017-
ll/documents/2011v6.3 2023en update emismod tsd oct2017.pdf).

Table 4-14. Assumed retirement rates and new source emission factor ratios for NSPS rules

NSPS Rule

Sector(s)

Retirement
Rate years
(%/year)

Pollutant
Impacted

Applied where?

New Source
Emission Factor
(Fn)

Oil and
Gas

np_oilgas,
pt_oilgas

No

assumption

VOC

Storage Tanks: 70.3% reduction in
growth-only (>1.0)

0.297

Gas Well Completions: 95% control
(regardless)

0.05

Pneumatic controllers, not high-bleed
>6scfm or low-bleed: 77% reduction in
growth-only (>1.0)

0.23

Pneumatic controllers, high-bleed
>6scfm or low-bleed: 100% reduction in
growth-only (>1.0)

0.00

Compressor Seals: 79.9% reduction in
growth-only (>1.0)

0.201

188


-------
NSPS Rule

Sector(s)

Retirement
Rate years
(%/year)

Pollutant
Impacted

Applied where?

New Source
Emission Factor
(Fn)









Fugitive Emissions: 60% Valves, flanges,
connections, pumps, open-ended lines,
and other

0.40

Pneumatic Pumps: 71.3%; Oil and Gas

0.287

RICE

np_oilgas,
pt_oilgas,
nonpt,
ptnonipm

40, (2.5%)

NOx

Lean burn: PA, all other states

0.25, 0.606

Rich Burn: PA, all other states

0.1, 0.069

Combined (average) LB/RB: PA, other
states

0.175, 0.338

CO

Lean burn: PA, all other states

1.0 (n/a), 0.889

Rich Burn: PA, all other states

0.15, 0.25

Combined (average) LB/RB: PA, other
states

0.575, 0.569

VOC

Lean burn: PA, all other states

0.125, n/a

Rich Burn: PA, all other states

0.1, n/a

Combined (average) LB/RB: PA, other
states

0.1125,n/a

Gas

Turbines

pt_oilgas,
ptnonipm

45 (2.2%)

NOx

California and NOx SIP Call states

0.595

All other states

0.238

Process
Heaters

pt_oilgas,
ptnonipm

30 (3.3%)

NOx

Nationally to Process Heater SCCs

0.41

4.2.4.1 Residential Wood Combustion (rwc)

Packets:

Proj ection_2016_2023_all_nonpoint_version l_platform_NC_04oct2019_v2
Proj ection_2016_2023_rwc_version l_platform_fromMARAMA_20aug2019_v0
Proj ection_2016_2028_all_nonpoint_version l_platform_NC_04oct2019_v2
Proj ection_2016_2028_rwc_version l_platform_fromMARAMA_20aug2019_v0

For residential wood combustion, the growth and control factors are computed together into merged
factors in the same packets. For states other than California, Oregon, and Washington, RWC emissions
from 2016 were projected to 2023 and 2028 using projection factors derived using the MARAMA tool
that is based on the projection methodology from EPA's 201 lv6.3 platform. The development of
projected growth in RWC emissions to year 2023 starts with the projected growth in RWC appliances
derived from year 2012 appliance shipments reported in the Regulatory Impact Analysis (RIA) for
Proposed Residential Wood Heaters NSPS Revision Final Report available at:

http://www2.epa.gov/sites/production/files/2013-12/documents/ria-20140103.pdf. The 2012 shipments
are based on 2008 shipment data and revenue forecasts from a Frost & Sullivan Market Report (Frost &
Sullivan, 2010). Next, to be consistent with the RIA, growth rates for new appliances for certified wood
stoves, pellet stoves, indoor furnaces and OHH were based on forecasted revenue (real GDP) growth rate
of 2.0% per year from 2013 through 2023 and 2028 as predicted by the U.S. Bureau of Economic
Analysis (BEA, 2012). While this approach is not perfectly correlated, in the absence of specific
shipment projections, the RIA assumes the overall trend in the projection is reasonable. The growth rates
for appliances not listed in the RIA (fireplaces, outdoor wood burning devices (not elsewhere classified)

189


-------
and residential fire logs) are estimated based on the average growth in the number of houses between
2002 and 2012, about 1% (U.S. Census, 2012).

In addition to new appliance sales and forecasts extrapolating beyond 2012, assumptions on the
replacement of older, existing appliances are needed. Based on long lifetimes, no replacement of
fireplaces, outdoor wood burning devices (not elsewhere classified) or residential fire logs is assumed. It
is assumed that 95% of new woodstoves will replace older non-EPA certified freestanding stoves (pre-
1988 NSPS) and 5% will replace existing EPA-certified catalytic and non-catalytic stoves that currently
meet the 1988 NSPS (Houck, 2011).

Equation 4-1 was applied with RWC-specific factors from the rule. The EPA RWC NSPS experts assume
that 10% of new pellet stoves and OHH replace older units and that because of their short lifespan, that
10%) of indoor furnaces are replaced each year; these are the same assumptions used since the 2007
emissions modeling platform (EPA, 2012d). The resulting growth factors for these appliance types varies
by appliance type and also by pollutant because the emission rates, from EPA RWC tool (EPA, 2013rwc),
vary by appliance type and pollutant. For EPA certified units, the projection factors for PM are lower
than those for all other pollutants. The projection factors also vary because the total number of existing
units in 2016 varies greatly between appliance types.

Table 4-15 contains the factors to adjust the emissions from 2016 to 2023 and 2028. California, Oregon,
and Washington RWC were held constant at NEI2014v2 levels for 2016, 2023, and 2028 due to the
unique control programs those states have in place.

Table 4-15. Projection factors for RWC

S( (

SC'C description

I'olliililiH"

2016-1 o-
2023

2016-to-
202S

2104008100

Fireplace: general



7.19%

12.36%

2104008210

Woodstove: fireplace inserts; non-EPA certified



-13.92%

-17.97%

2104008220

Woodstove: fireplace inserts; EPA certified; non-
catalytic

PM10-PRI

4.09%

5.08%

2104008220

Woodstove: fireplace inserts; EPA certified; non-
catalytic

PM25-PRI

4.09%

5.08%

2104008220

Woodstove: fireplace inserts; EPA certified; non-
catalytic



8.34%

10.28%

2104008230

Woodstove

fireplace inserts; EPA certified; catalytic

PM10-PRI

6.06%

7.68%

2104008230

Woodstove

fireplace inserts; EPA certified; catalytic

PM25-PRI

6.06%

7.68%

2104008230

Woodstove

fireplace inserts; EPA certified; catalytic



12.08%

15.27%

2104008310

Woodstove

freestanding, non-EPA certified

CO

-12.09%

-15.72%

2104008310

Woodstove

freestanding, non-EPA certified

PM10-PRI

-12.67%

-16.52%

2104008310

Woodstove

freestanding, non-EPA certified

PM25-PRI

-12.67%

-16.52%

2104008310

Woodstove

freestanding, non-EPA certified

VOC

-11.40%

-14.84%

2104008310

Woodstove

freestanding, non-EPA certified



-12.09%

-15.72%

2104008320

Woodstove

freestanding, EPA certified, non-catalytic

PM10-PRI

4.09%

5.08%

2104008320

Woodstove

freestanding, EPA certified, non-catalytic

PM25-PRI

4.09%

5.08%

2104008320

Woodstove

freestanding, EPA certified, non-catalytic



8.34%

10.28%

2104008330

Woodstove

freestanding, EPA certified, catalytic

PM10-PRI

6.07%

7.69%

2104008330

Woodstove

freestanding, EPA certified, catalytic

PM25-PRI

6.07%

7.69%

2104008330

Woodstove

freestanding, EPA certified, catalytic



12.08%

15.27%

2104008400

Woodstove
insert)

pellet-fired, general (freestanding or FP

PM10-PRI

30.09%

38.02%

190


-------
S( (

SC'C description

Polliiiiinr

2016-1 o-
2023

2016-to-
202S

2104008400

Woodstove: pellet-fired, general (freestanding or FP
insert)

PM25-PRI

30.09%

38.02%

2104008400

Woodstove: pellet-fired, general (freestanding or FP
insert)



26.96%

33.85%

2104008510

Furnace: Indoor, cordwood-fired, non-EPA certified

CO

-64.93%

-84.78%

2104008510

Furnace: Indoor, cordwood-fired, non-EPA certified

PM10-PRI

-62.99%

-82.89%

2104008510

Furnace: Indoor, cordwood-fired, non-EPA certified

PM25-PRI

-62.99%

-82.89%

2104008510

Furnace: Indoor, cordwood-fired, non-EPA certified

VOC

-65.02%

-84.89%

2104008510

Furnace: Indoor, cordwood-fired, non-EPA certified



-64.93%

-84.78%

2104008610

Hydronic heater: outdoor

PM10-PRI

0.06%

-0.40%

2104008610

Hydronic heater: outdoor

PM25-PRI

0.06%

-0.40%

2104008610

Hydronic heater: outdoor



-0.73%

-1.30%

2104008700

Outdoor wood burning device, NEC (fire-pits,
chimineas, etc)



7.19%

9.25%

2104009000

Fire log total



7.19%

9.25%

* If no pollutant is specified, facture is used for any pollutants that do not have a pollutant-specific factor

4.2.4.2 Oil and Gas NSPS (np_oilgas, pt_oilgas)

Packets:

Control_2016_2023_OilGas_N SPS_pt_oilgas_v l_platform_l 7 sep2019_v0

Control_2016_2028_OilGas_N SPS_pt_oilgas_v l_platform_l 7 sep2019_v0

For oil and gas NSPS controls, except for gas well completions (a 95 percent control), the assumption of
no equipment retirements through year 2028 dictates that NSPS controls are applied to the growth
component only of any PROJECTION factors. For example, if a growth factor is 1.5 for storage tanks
(indicating a 50 percent increase activity), then, using Table 4-14, the 70.3 percent VOC NSPS control to
this new growth will result in a 23.4 percent control: 100 *(70.3 * (1.5 -1) / 1.5); this yields an "effective"
growth rate (combined PROJECTION and CONTROL) of 1.1485, or a 70.3 percent reduction from 1.5 to
1.0. The impacts of all non-drilling completion VOC NSPS controls are therefore greater where growth
in oil and gas production is assumed highest. Conversely, for oil and gas basins with assumed negative
growth in activity/production, VOC NSPS controls will be limited to well completions only. These
reductions are year-specific because projection factors for these sources are year-specific. Table 4-16
(npoilgas) and Table 4-18 (ptoilgas) list the SCCs where Oil and Gas NSPS controls were applied; note
controls are applied to production and exploration-related SCCs. Table 4-17 (np oilgas) and Table 4-19
(pt oilgas) shows the reduction in VOC emissions after the application of the Oil and Gas NSPS
CONTROL packet for both future years 2023 and 2028.

Table 4-16. Non-point (np oilgas) SCCs in 2016vl modeling platform where Oil and Gas NSPS

controls applied

see

SRC TYPE

OILGAS NSPS
CATEGORY

TOOL OR
STATE

see

SRC CAT TYPE

SCCDESC

2310010200

OIL

1. Storage Tanks

TOOL

PRODUCTION

Industrial Processes; Oil and Gas Exploration and
Production; Crude Petroleum; Oil Well Tanks -
Flashing & Standing/Working/Breathing

2310010300

OIL

3. Pnuematic
controllers: not high
or low bleed

TOOL

PRODUCTION

Industrial Processes; Oil and Gas Exploration and
Production; Crude Petroleum; Oil Well Pneumatic
Devices

191


-------




OILGAS NSPS

TOOL OR
STATE





see

SRC TYPE

CATEGORY

see

SRC CAT TYPE

SCCDESC

2310011500

OIL

5. Fugitives

STATE

PRODUCTION

Industrial Processes; Oil and Gas Exploration and
Production; On-Shore Oil Production; Fugitives: All
Processes

2310011501

OIL

5. Fugitives

TOOL

PRODUCTION

Industrial Processes; Oil and Gas Exploration and
Production; On-Shore Oil Production; Fugitives:
Connectors

2310011502

OIL

5. Fugitives

TOOL

PRODUCTION

Industrial Processes; Oil and Gas Exploration and
Production; On-Shore Oil Production; Fugitives:
Flanges

2310011503

OIL

5. Fugitives

TOOL

PRODUCTION

Industrial Processes; Oil and Gas Exploration and
Production; On-Shore Oil Production; Fugitives: Open
Ended Lines

2310011505

OIL

5. Fugitives

TOOL

PRODUCTION

Industrial Processes; Oil and Gas Exploration and
Production; On-Shore Oil Production; Fugitives:
Valves

2310021010

NGAS

1. Storage Tanks

TOOL

PRODUCTION

Industrial Processes; Oil and Gas Exploration and
Production; On-Shore Gas Production; Storage Tanks:
Condensate

2310021300

NGAS

3. Pnuematic
controllers: not high
or low bleed

TOOL

PRODUCTION

Industrial Processes; Oil and Gas Exploration and
Production; On-Shore Gas Production; Gas Well
Pneumatic Devices

2310021310

NGAS

6. Pneumatic Pumps

STATE

PRODUCTION

Industrial Processes; Oil and Gas Exploration and
Production; On-Shore Gas Production; Gas Well
Pneumatic Pumps

2310021501

NGAS

5. Fugitives

TOOL

PRODUCTION

Industrial Processes; Oil and Gas Exploration and
Production; On-Shore Gas Production; Fugitives:
Connectors

2310021502

NGAS

5. Fugitives

TOOL

PRODUCTION

Industrial Processes; Oil and Gas Exploration and
Production; On-Shore Gas Production; Fugitives:
Flanges

2310021503

NGAS

5. Fugitives

TOOL

PRODUCTION

Industrial Processes; Oil and Gas Exploration and
Production; On-Shore Gas Production; Fugitives: Open
Ended Lines

2310021505

NGAS

5. Fugitives

TOOL

PRODUCTION

Industrial Processes; Oil and Gas Exploration and
Production; On-Shore Gas Production; Fugitives:
Valves

2310021506

NGAS

5. Fugitives

TOOL

PRODUCTION

Industrial Processes; Oil and Gas Exploration and
Production; On-Shore Gas Production; Fugitives:
Other

2310021509

NGAS

5. Fugitives

STATE

PRODUCTION

Industrial Processes; Oil and Gas Exploration and
Production; On-Shore Gas Production; Fugitives: All
Processes

2310021601

NGAS

2. Well Completions

STATE

EXPLORATION

Industrial Processes; Oil and Gas Exploration and
Production; On-Shore Gas Production; Gas Well
Venting - Initial Completions

2310030300

NGAS

1. Storage Tanks

STATE

PRODUCTION

Industrial Processes; Oil and Gas Exploration and
Production; Natural Gas Liquids; Gas Well Water Tank
Losses

2310111401

OIL

6. Pneumatic Pumps

TOOL

PRODUCTION

Industrial Processes; Oil and Gas Exploration and
Production; On-Shore Oil Exploration; Oil Well
Pneumatic Pumps

2310111700

OIL

2. Well Completions

TOOL

EXPLORATION

Industrial Processes; Oil and Gas Exploration and
Production; On-Shore Oil Exploration; Oil Well
Completion: All Processes

2310121401

NGAS

6. Pneumatic Pumps

TOOL

PRODUCTION

Industrial Processes; Oil and Gas Exploration and
Production; On-Shore Gas Exploration; Gas Well
Pneumatic Pumps

192


-------
see

SRC TYPE

OILGAS NSPS
CATEGORY

TOOL OR
STATE

see

SRC CAT TYPE

SCCDESC

2310121700

NGAS

2. Well Completions

TOOL

EXPLORATION

Industrial Processes; Oil and Gas Exploration and
Production; On-Shore Gas Exploration; Gas Well
Completion: All Processes

2310421010

NGAS

1. Storage Tanks

STATE

PRODUCTION

Industrial Processes; Oil and Gas Exploration and
Production; On-Shore Gas Production -
Unconventional; Storage Tanks: Condensate

2310421700

NGAS

2. Well Completions

STATE

EXPLORATION

Gas Well Completion: All Processes Unconventional

Table 4-17. Emissions reductions for npoilgas sector due to application of Oil and Gas NSPS

year

poll

2016vl

2016

pre-CoST

emissions

emissions
change from
2016

%

change

2023

VOC

2817303

2881217

-863524

-30.0%

2028

VOC

2817303

2881217

-1077514

-37.4%

Table 4-18. Point source SCCs in pt oilgas sector where Oil and Gas NSPS controls were applied.



FUEL





see

PRODUCED

OILGAS NSPS CATEGORY

SCCDESC







Industrial Processes; Oil and Gas Production; Crude Oil

31000101

Oil

2. Well Completions

Production; Well Completion







Industrial Processes; Oil and Gas Production; Crude Oil

31000130

Oil

4. Compressor Seals

Production; Fugitives: Compressor Seals







Industrial Processes; Oil and Gas Production; Crude Oil

31000133

Oil

1. Storage Tanks

Production; Storage Tank





3. Pnuematic controllers:

Industrial Processes; Oil and Gas Production; Crude Oil

31000151

Oil

high or low bleed

Production; Pneumatic Controllers, Low Bleed





3. Pnuematic controllers:

Industrial Processes; Oil and Gas Production; Crude Oil

31000152

Oil

high or low bleed

Production; Pneumatic Controllers High Bleed >6 scfh







Industrial Processes; Oil and Gas Production; Natural Gas

31000207

Gas

5. Fugitives

Production; Valves: Fugitive Emissions







Industrial Processes; Oil and Gas Production; Natural Gas







Production; All Equipt Leak Fugitives (Valves, Flanges,

31000220

Gas

5. Fugitives

Connections, Seals, Drains







Industrial Processes; Oil and Gas Production; Natural Gas

31000222

Gas

2. Well Completions

Production; Well Completions







Industrial Processes; Oil and Gas Production; Natural Gas

31000225

Gas

4. Compressor Seals

Production; Compressor Seals





3. Pnuematic controllers:

Industrial Processes; Oil and Gas Production; Natural Gas

31000233

Gas

high or low bleed

Production; Pneumatic Controllers, Low Bleed







Industrial Processes; Oil and Gas Production; Natural Gas

31000309

Gas

4. Compressor Seals

Processing; Compressor Seals





3. Pnuematic controllers:

Industrial Processes; Oil and Gas Production; Natural Gas

31000324

Gas

high or low bleed

Processing; Pneumatic Controllers Low Bleed





3. Pnuematic controllers:

Industrial Processes; Oil and Gas Production; Natural Gas

31000325

Gas

high or low bleed

Processing; Pneumatic Controllers, High Bleed >6 scfh







Industrial Processes; Oil and Gas Production; Fugitive Emissions;

31088811

Both

5. Fugitives

Fugitive Emissions

193


-------
Table 4-19. VOC reductions (tons/year) for the ptoilgas sector after application of the Oil and Gas
NSPS CONTROL packet for both future years 2023 and 2028.

Year

Pollutant

2016vl

Emissions Reductions

% change

2023

VOC

129,253

-2,523

-2.0%

2028

VOC

129,253

-2,808

-2.2%

4.2.4.3 RICE NSPS (nonpt, ptnonipm, np_oilgas, pt_oilgas)

Packets:

CONTROL_2016_2023_RICE_NSPS_nonpt_ptnonipm_beta_platform_extended_04oct2019_vl
CONTROL2016_2023_RICE_NSPS_ptnonipm_v l_platform_MARAMA_l 0sep2019_v0
CONTROL_2016_2028_RICE_NSPS_nonpt_ptnonipm_beta_platform_extended_04oct2019_vl
CONTROL2016_2028_RICE_NSPS_ptnonipm_v l_platform_MARAMA_l 0sep2019_v0

For RICE NSPS controls, the EPA emission requirements for stationary engines differ according to
whether the engine is new or existing, whether the engine is located at an area source or major source, and
whether the engine is a compression ignition or a spark ignition engine. Spark ignition engines are further
subdivided by power cycle, two-stroke versus four-stroke, and whether the engine is rich burn or lean
burn. We applied NSPS reduction for lean burn, rich burn and "combined" engines using Equation 4-2
and information listed in Table 4-14. Table 4-20, Table 4-21 and Table 4-25 list the SCCs where RICE
NSPS controls were applied for the 2016vl platform. Table 4-22, Table 4-23, Table 4-24 and Table 4-26
show the reductions in emissions in the nonpoint, ptnonipm, and nonpoint oil and gas sectors after the
application of the RICE NSPS CONTROL packet for both future years 2023 and 2028. Note that for
nonpoint oil and gas, VOC reductions were only appropriate in the state of Pennsylvania.

Table 4-20. SCCs and Engine Type in 2016vl modeling platform where RICE NSPS controls

applied for nonpt and ptnonipm sectors.

see

Lean, Rich, or
Combined

SCCDESC

20200202

Combined

Internal Combustion Engines; Industrial; Natural Gas; Reciprocating

20200253

Rich

Internal Combustion Engines; Industrial; Natural Gas; 4-cycle Rich Burn

20200254

Lean

Internal Combustion Engines; Industrial; Natural Gas; 4-cycle Lean Burn

20200256

Lean

Internal Combustion Engines; Industrial; Natural Gas; 4-cycle Clean Burn

20300201

Combined

Internal Combustion Engines; Commercial/Institutional; Natural Gas; Reciprocating

2102006000

Combined

Stationary Source Fuel Combustion; Industrial; Natural Gas; Total: Boilers and IC
Engines

2102006002

Combined

Stationary Source Fuel Combustion; Industrial; Natural Gas; All IC Engine Types

2103006000

Combined

Stationary Source Fuel Combustion; Commercial/Institutional; Natural Gas; Total:
Boilers and IC Engines

194


-------
Table 4-21. Non-point Oil and Gas SCCs in 2016vl modeling platform where RICE NSPS controls

applied

see

Lean, Rich,
or Combined
category

SRC_TYPE

TOOL OR
STATE

see

SRC CAT TYPE

SCCDESC

2310000220

Combined

BOTH

TOOL

EXPLORATION

Industrial Processes; Oil and Gas
Exploration and Production; All
Processes; Drill Rigs

2310000660

Combined

BOTH

TOOL

EXPLORATION

Industrial Processes; Oil and Gas
Exploration and Production; All
Processes; Hydraulic Fracturing
Engines

2310020600

Combined

NGAS

STATE

PRODUCTION

Industrial Processes; Oil and Gas
Exploration and Production;
Natural Gas; Compressor Engines

2310021202

Lean

NGAS

TOOL

PRODUCTION

Industrial Processes; Oil and Gas
Exploration and Production; On-
shore Gas Production; Natural Gas
Fired 4Cycle Lean Burn Compressor
Engines 50 To 499 HP

2310021251

Lean

NGAS

TOOL

PRODUCTION

Industrial Processes; Oil and Gas
Exploration and Production; On-
shore Gas Production; Lateral
Compressors 4 Cycle Lean Burn

2310021302

Rich

NGAS

TOOL

PRODUCTION

Industrial Processes; Oil and Gas
Exploration and Production; On-
shore Gas Production; Natural Gas
Fired 4Cycle Rich Burn Compressor
Engines 50 To 499 HP

2310021351

Rich

NGAS

TOOL

PRODUCTION

Industrial Processes; Oil and Gas
Exploration and Production; On-
shore Gas Production; Lateral
Compressors 4 Cycle Rich Burn

2310023202

Lean

CBM

TOOL

PRODUCTION

Industrial Processes; Oil and Gas
Exploration and Production; Coal
Bed Methane Natural Gas; CBM
Fired 4Cycle Lean Burn Compressor
Engines 50 To 499 HP

2310023251

Lean

CBM

TOOL

PRODUCTION

Industrial Processes; Oil and Gas
Exploration and Production; Coal
Bed Methane Natural Gas; Lateral
Compressors 4 Cycle Lean Burn

2310023302

Rich

CBM

TOOL

PRODUCTION

Industrial Processes; Oil and Gas
Exploration and Production; Coal
Bed Methane Natural Gas; CBM
Fired 4Cycle Rich Burn Compressor
Engines 50 To 499 HP

2310023351

Rich

CBM

TOOL

PRODUCTION

Industrial Processes; Oil and Gas
Exploration and Production; Coal
Bed Methane Natural Gas; Lateral
Compressors 4 Cycle Rich Burn

2310400220

Combined

BOTH

STATE

EXPLORATION

Industrial Processes; Oil and Gas
Exploration and Production; All

195


-------
see

Lean, Rich,
or Combined
category

SRC_TYPE

TOOL OR
STATE

see

SRC CAT TYPE

SCCDESC











Processes - Unconventional; Drill
Rigs

Table 4-22. Nonpoint Emissions reductions after the application of the RICE NSPS

year

poll

2016vl
(tons)

emissions

reductions

(tons)

%

change

2023

CO

2,688,250

-16,982

-0.6%

2023

NOX

718,766

-23,704

-3.3%

2028

CO

2,688,250

-23,145

-0.9%

2028

NOX

718,766

-33,621

-4.7%

Table 4-23. Ptnonipm Emissions reductions after the application of the RICE NSPS

year

poll

2016vl
(tons)

emissions

reductions

(tons)

%

change

2023

CO

1,446,353

-2,756

-0.2%

2023

NOX

952,181

-3,400

-0.4%

2023

VOC

774,289

-2

0.0%

2028

CO

1,446,353

-3,295

-0.2%

2028

NOX

952,181

-4,232

-0.4%

2028

VOC

774,289

-3

0.0%

Table 4-24. Oil and Gas Emissions reductions for np oilgas sector due to application of RICE NSPS

year

poll

2016vl

2016

pre-CoST
emissions

emissions
reduction

%

change

2023

CO

762706

767414

-106005

-13.8%

2023

NOX

574133

598738

-93806

-15.7%

2023

VOC

2817303

2881217

-525

-0.02%

2028

CO

762706

767414

-145622

-19.0%

2028

NOX

574133

598738

-134144

-22.4%

2028

VOC

2817303

2881217

-785

-0.03%

Table 4-25. Point source SCCs in pt oilgas sector where RICE NSPS controls applied.

see

Lean, Rich, or
Combined

SCCDESC

20200202

Combined

Internal Combustion Engines; Industrial; Natural Gas; Reciprocating

20200253

Rich

Internal Combustion Engines; Industrial; Natural Gas;4-cycle Rich Burn

196


-------
see

Lean, Rich, or
Combined

SCCDESC

20200254

Lean

Internal Combustion Engines; Industrial; Natural Gas;4-cycle Lean Burn

20200256

Combined

Internal Combustion Engines; Industrial; Natural Gas;4-cycle Clean Burn

20300201

Combined

Internal Combustion Engines; Commercial/Institutional; Natural Gas; Reciprocating

31000203

Combined

Industrial Processes; Oil and Gas Production; Natural Gas Production; Compressors
(See also 310003-12 and -13)

Table 4-26. Emissions reductions (tons/year) in ptoilgas sector after the application of the RICE
NSPS CONTROL packet for future years 2023 and 2028.

Year

Pollutant

2016vl

Emissions Reductions

% change

2023

CO

177,690

-20,258

-11.4%

2023

NOX

379,866

-53,694

-14.1%

2023

VOC

129,253

-436

-0.3%

2028

CO

177,690

-26,095

-14.7%

2028

NOX

379,866

-70,659

-18.6%

2028

VOC

129,253

-512

-0.4%

4.2.4.4 Fuel Sulfur Rules (nonpt, ptnonipm)

Packets:

Control_2016_202X_MANEVU_Sulfur_fromMARAMA_v l_platform_23 sep2019_v0

Fuel sulfur rules, based on web searching and the 2011 emissions modeling notice of data availability
(NODA) comments, are currently limited to the following states: Connecticut, Delaware, Maine,
Massachusetts, New Hampshire, New Jersey, New York, Pennsylvania, Rhode Island, and Vermont. The
fuel limits for these states are incremental starting after year 2012, but are fully implemented by July 1,
2018, in all of these states. The control packet representing these controls was updated by MARAMA for
version 1 platform.

Summaries of the sulfur rules by state, with emissions reductions are provided in Table 4-27 and Table
4-28. These tables reflect the impacts of the MARAMA packet only, as these reductions are not estimated
in non-MARAMA states. Most of these reductions occur in the nonpt sector; a small amount of reductions
occurs in the ptnonipm sector, and a negligible amount of reductions occur in the pt oilgas sector.

Table 4-27. Summary of fuel sulfur rule impacts on nonpoint S02 emissions for 2023 and 2028

year

poll

2016vl
(tons)

emissions

reductions

(tons)

%

change

2023

S02

140,469

-28,137

-20.0%

2028

S02

140,469

-24,200

-17.2%

197


-------
Table 4-28. Summary of fuel sulfur rule impacts on ptnonipm S02 emissions for 2023 and 2028

year

poll

2016vl
(tons)

emissions

reductions

(tons)

%

change

2023

S02

658,204

-1,183

-0.2%

2028

S02

658,204

-1,241

-0.2%

4.2.4.5 Natural Gas Turbines NOx NSPS (ptnonipm, pt_oilgas)

Packets:

CONTROL_2016_2023_Natural_Gas_Turbines_NSPS_ptnonipm_beta_platform_extended_04oct2019_v 1
CONTROL_2016_2023_NG_Turbines_NSPS_ptnonipm_v l_platform_MARAMA_l 0sep2019_v0
CONTROL_2016_2028_Natural_Gas_Turbines_NSPS_ptnonipm_beta_platform_extended_04oct2019_v 1
CONTROL_2016_2028_NG_Turbines_NSPS_ptnonipm_v l_platform_MARAMA_l 0sep2019_v0

Natural Gas Turbines NSPS controls were generated based on examination of emission limits for
stationary combustion turbines that are not in the power sector. In 2006, the EPA promulgated standards
of performance for new stationary combustion turbines in 40 CFR part 60, subpart KKKK. The standards
reflect changes in NOx emission control technologies and turbine design since standards for these units
were originally promulgated in 40 CFR part 60, subpart GG. The 2006 NSPSs affecting NOx and SO2
were established at levels that bring the emission limits up-to-date with the performance of current
combustion turbines. Stationary combustion turbines were also regulated by the NOx State
Implementation Plan (SIP) Call, which required affected gas turbines to reduce their NOx emissions by
60 percent. Table 4-29 compares the 2006 NSPS emission limits with the NOx Reasonably Available
Control Technology (RACT) regulations in selected states within the NOx SIP Call region. The map
showing the states and partial-states in the NOx SIP Call Program can be found at:

http://www3.epa.gov/airmarkets/progress/reports/program basics.html. The state NOx RACT regulations
summary (Pechan, 2001) is from a year 2001 analysis, so some states may have updated their rules since
that time.

Table 4-29. Stationary gas turbines NSPS analysis and resulting emission rates used to compute

controls

NOx Emission Limits for New Stationary Combustion Turbines

Firing Natural Gas

<50 MMBTU/hr

50-850
MMBTU/hr

>850

MMBTU/hr



Federal NSPS

100

25

15

ppm











State RACT Regulations

5-100

MMBTU/hr

100-250
MMBTU/hr

>250

MMBTU/hr



Connecticut

225

75

75

ppm

Delaware

42

42

42

ppm

Massachusetts

65*

65

65

ppm

New Jersey

50*

50

50

ppm

New York

50

50

50

ppm

New Hampshire

55

55

55

ppm

198


-------
NOx Emission Limits for New Stationary Combustion Turbines

* Only applies to 25-100 MMBTU/hr

Notes: The above state RACT table is from a 2001 analysis. The current NY State regulations have the

same emission limits.







New source emission rate (Fn)

NOx ratio (Fn)

Control (%)

NOx SIP Call states plus CA

= 25 / 42 =

0.595

40.5%

Other states

= 25 / 105 =

0.238

76.2%

For control factor development, the existing source emission ratio was set to 1.0 for combustion turbines.
The new source emission ratio for the NOx SIP Call states and California is the ratio of state NOx
emission limit to the Federal NSPS. A complicating factor in the above is the lack of size information in
the stationary source SCCs. Plus, the size classifications in the NSPS do not match the size differentiation
used in state air emission regulations. We accepted a simplifying assumption that most industrial
applications of combustion turbines are in the 100-250 MMBtu/hr size range and computed the new
source emission rates as the NSPS emission limit for 50-850 MMBtu/hr units divided by the state
emission limits. We used a conservative new source emission ratio by using the lowest state emission
limit of 42 ppmv (Delaware). This yields a new source emission ratio of 25/42, or 0.595 (40.5 percent
reduction) for states with existing combustion turbine emission limits. States without existing turbine
NOx limits would have a lower new source emission ratio -the uncontrolled emission rate (105 ppmv via
AP-42) divided into 25 ppmv = 0.238 (76.2 percent reduction). This control was then plugged into
Equation 4-2 as a function of the year-specific projection factor. Also, Natural Gas Turbines control
factors supplied by MARAMA were used within the MARAMA region.

Table 4-30 and Table 4-32 list the point source SCCs where Natural Gas Turbines NSPS controls were
applied for the 2016vl platform. Table 4-31 and Table 4-33 show the reduction in NOx emissions after
the application of the Natural Gas Turbines NSPS CONTROL packet for both future years 2023 and
2028. The values in Table 4-31 and Table 4-33 include emissions both inside and outside the MARAMA
region.

Table 4-30. Ptnonipm SCCs in 2016vl modeling platform where Natural Gas Turbines NSPS

controls applied

see

SCC description

20200201

Internal Combustion Engines

Industrial; Natural Gas; Turbine

20200203

Internal Combustion Engines

Industrial; Natural Gas; Turbine: Cogeneration

20200209

Internal Combustion Engines

Industrial; Natural Gas; Turbine: Exhaust

20200701

Internal Combustion Engines

Industrial; Process Gas; Turbine

20200714

Internal Combustion Engines

Industrial; Process Gas; Turbine: Exhaust

20300202

Internal Combustion Engines

Commercial/Institutional; Natural Gas; Turbine

20300203

Internal Combustion Engines
Cogeneration

Commercial/Institutional; Natural Gas; Turbine:

Table 4-31. Ptnonipm emissions reductions after the application of the Natural Gas Turbines NSPS

year

poll

2016vl (tons)

emissions
reduction (tons)

0/

/O

change

2023

NOX

952,181

-2,531

-0.3%

199


-------
2028

NOX

952,181

-3,346

-0.4%

Table 4-32. Point source SCCs in ptoilgas sector where Natural Gas Turbines NSPS control

applied.

see

SCC description

20200201

Internal Combustion Engines; Industrial; Natural Gas; Turbine

20200209

Internal Combustion Engines; Industrial; Natural Gas; Turbine: Exhaust

20300202

Internal Combustion Engines; Commercial/Institutional; Natural Gas; Turbine

20300209

Internal Combustion Engines; Commercial/Institutional; Natural Gas; Turbine: Exhaust

20200203

Internal Combustion Engines; Industrial; Natural Gas; Turbine: Cogeneration

20200714

Internal Combustion Engines; Industrial; Process Gas; Turbine: Exhaust

20300203

Internal Combustion Engines; Commercial/Institutional; Natural Gas; Turbine:
Cogeneration

Table 4-33. Emissions reductions (tons/year) for pt oilgas after the application of the Natural Gas
Turbines NSPS CONTROL packet for future years 2023 and 2028.

Year

Pollutant

2016vl

Emissions
Reduction

%

change

2023

NOX

379,866

-8,079

-2.1%

2028

NOX

379,866

-11,282

-3.0%

4.2.4.6 Process Heaters NOx NSPS (ptnonipm, pt_oilgas)

Packets:

Control_2016_2023_Process_Heaters_NSPS_ptnonipm_beta_platform_ext_25sep2019_v0
Control_2016_2028_Process_Heaters_NSPS_ptnonipm_beta_platform_ext_25sep2019_v0

Process heaters are used throughout refineries and chemical plants to raise the temperature of feed
materials to meet reaction or distillation requirements. Fuels are typically residual oil, distillate oil,
refinery gas, or natural gas. In some sense, process heaters can be considered as emission control devices
because they can be used to control process streams by recovering the fuel value while destroying the
VOC. The criteria pollutants of most concern for process heaters are NOx and SO2.

In 2016, it is assumed that process heaters have not been subject to regional control programs like the
NOx SIP Call, so most of the emission controls put in-place at refineries and chemical plants have
resulted from RACT regulations that were implemented as part of SIPs to achieve ozone NAAQS in
specific areas, and refinery consent decrees. The boiler/process heater NSPS established NOx emission
limits for new and modified process heaters. These emission limits are displayed in Table 4-41.

200


-------
Table 4-34. Process Heaters NSPS analysis and 2016vl new emission rates used to estimate controls

NOx emission rate Existing (Fe)

Fraction at this rate

Average

PPMV

Natural
Draft

Forced
Draft

80

0.4

0



100

0.4

0.5



150

0.15

0.35



200

0.05

0.1



240

0

0.05



Cumulative, weighted: Fe

104.5

134.5

119.5

NSPS Standard

40

60



New Source NOx ratio (Fn)

0.383

0.446

0.414

NSPS Control (%)

61.7

55.4

58.6

For computations, the existing source emission ratio (Fe) was set to 1.0. The computed (average) NOx
emission factor ratio for new sources (Fn) is 0.41 (58.6 percent control). The retirement rate is the inverse
of the expected unit lifetime. There is limited information in the literature about process heater lifetimes.
This information was reviewed at the time that the Western Regional Air Partnership (WRAP) developed
its initial regional haze program emission projections, and energy technology models used a 20-year
lifetime for most refinery equipment. However, it was noted that in practice, heaters would probably have
a lifetime that was on the order of 50 percent above that estimate. Therefore, a 30-year lifetime was used
to estimate the effects of process heater growth and retirement. This yields a 3.3 percent retirement rate.
This control was then plugged into Equation 4-2 as a function of the year-specific projection factor. Table
4-35 and Table 4-37 list the point source SCCs where Process Heaters NSPS controls were applied for the
2016vl platform. Table 4-36 and Table 4-38 show the reduction in NOx emissions after the application
of the Process Heaters NSPS CONTROL packet for both future years 2023 and 2028.

Table 4-35. Ptnonipm SCCs in 2016vl modeling platform where Process Heaters NSPS controls

applied.

see

sccdesc

30190003

Industrial Processes; Chemical Manufacturing; Fuel Fired Equipment; Process Heater:
Natural Gas

30190004

Industrial Processes; Chemical Manufacturing; Fuel Fired Equipment; Process Heater:
Process Gas

30590002

Industrial Processes; Mineral Products; Fuel Fired Equipment; Residual Oil: Process
Heaters

30590003

Industrial Processes; Mineral Products; Fuel Fired Equipment; Natural Gas: Process
Heaters

30600101

Industrial Processes; Petroleum Industry; Process Heaters

Oil-fired

30600102

Industrial Processes; Petroleum Industry; Process Heaters

Gas-fired

30600103

Industrial Processes; Petroleum Industry; Process Heaters

Oil

30600104

Industrial Processes; Petroleum Industry; Process Heaters

Gas-fired

30600105

Industrial Processes; Petroleum Industry; Process Heaters

Natural Gas-fired

30600106

Industrial Processes; Petroleum Industry; Process Heaters

Process Gas-fired

201


-------
see

sccdesc

30600107

Industrial Processes; Petroleum Industry; Process Heaters; Liquified Petroleum Gas (LPG)

30600199

Industrial Processes; Petroleum Industry; Process Heaters; Other Not Classified

30990003

Industrial Processes; Fabricated Metal Products; Fuel Fired Equipment; Natural Gas:
Process Heaters

31000401

Industrial Processes; Oil and Gas Production; Process Heaters; Distillate Oil (No. 2)

31000402

Industrial Processes; Oil and Gas Production; Process Heaters; Residual Oil

31000403

Industrial Processes; Oil and Gas Production; Process Heaters; Crude Oil

31000404

Industrial Processes; Oil and Gas Production; Process Heaters; Natural Gas

31000405

Industrial Processes; Oil and Gas Production; Process Heaters; Process Gas

31000406

Industrial Processes; Oil and Gas Production; Process Heaters; Propane/Butane

31000413

Industrial Processes; Oil and Gas Production; Process Heaters; Crude Oil: Steam
Generators

31000414

Industrial Processes; Oil and Gas Production; Process Heaters; Natural Gas: Steam
Generators

31000415

Industrial Processes; Oil and Gas Production; Process Heaters; Process Gas: Steam
Generators

39900501

Industrial Processes; Miscellaneous Manufacturing Industries; Process Heater/Furnace;
Distillate Oil

39900601

Industrial Processes; Miscellaneous Manufacturing Industries; Process Heater/Furnace;
Natural Gas

39990003

Industrial Processes; Miscellaneous Manufacturing Industries; Miscellaneous
Manufacturing Industries; Natural Gas: Process Heaters

Table 4-36. Ptnonipm emissions reductions after the application of the Process Heaters NSPS

year

poll

2016vl
(tons)

emissions

reductions

(tons)

%

change

2023

NOX

952,181

-9,511

-1.0%

2028

NOX

952,181

-12,692

-1.3%

Table 4-37. Point source SCCs in pt oilgas sector where Process Heaters NSPS controls were

applied

see

SCC Description

30190003

Industrial Processes; Chemical Manufacturing; Fuel Fired Equipment; Process Heater:
Natural Gas

30600102

Industrial Processes; Petroleum Industry; Process Heaters; Gas-fired

30600104

Industrial Processes; Petroleum Industry; Process Heaters; Gas-fired

30600105

Industrial Processes; Petroleum Industry; Process Heaters; Natural Gas-fired

30600106

Industrial Processes; Petroleum Industry; Process Heaters; Process Gas-fired

30600199

Industrial Processes; Petroleum Industry; Process Heaters; Other Not Classified

30990003

Industrial Processes; Fabricated Metal Products; Fuel Fired Equipment; Natural Gas:
Process Heaters

31000401

Industrial Processes; Oil and Gas Production; Process Heaters; Distillate Oil (No. 2)

31000402

Industrial Processes; Oil and Gas Production; Process Heaters; Residual Oil

31000403

Industrial Processes; Oil and Gas Production; Process Heaters; Crude Oil

202


-------
see

SCC Description

31000404

Industrial Processes; Oil and Gas Production; Process Heaters; Natural Gas

31000405

Industrial Processes; Oil and Gas Production; Process Heaters; Process Gas

31000413

Industrial Processes; Oil and Gas Production; Process Heaters; Crude Oil: Steam
Generators

31000414

Industrial Processes; Oil and Gas Production; Process Heaters; Natural Gas: Steam
Generators

31000415

Industrial Processes; Oil and Gas Production; Process Heaters; Process Gas: Steam
Generators

39900501

Industrial Processes; Miscellaneous Manufacturing Industries; Process Heater/Furnace;
Distillate Oil

39900601

Industrial Processes; Miscellaneous Manufacturing Industries; Process Heater/Furnace;
Natural Gas

Table 4-38. NOx emissions reductions (tons/year) in ptoilgas sector after the application of the
Process Heaters NSPS CONTROL packet for futures years 2023 and 2028.

Year

Poll

2016vl

Emissions Reductions

%

change

2023

NOX

379,866

-1,698

-0.4%

2028

NOX

379,866

-2,376

-0.6%

4.2.4.7 CISWI (ptnonipm)

Packets:

Control_2016_202X_CISWI_ptnonipm_beta_platform_ext_25 sep2019_v0

On March 21, 2011, the EPA promulgated the revised NSPS and emission guidelines for Commercial and
Industrial Solid Waste Incineration (CISWI) units. This was a response to the voluntary remand that was
granted in 2001 and the vacatur and remand of the CISWI definition rule in 2007. In addition, the
standards redevelopment included the 5-year technology review of the new source performance standards
and emission guidelines required under Section 129 of the Clean Air Act. The history of the CISWI
implementation is documented here: https://www.epa.gov/stationary-sources-air-pollution/commercial-
and-industrial-solid-waste-incineration-units-ciswi-new. Baseline and CISWI rule impacts associated with
the CISWI rule are documented here: https://www.regulations.gov/document?D=EPA-HQ-OAR-2003-
0119-2559. The EPA mapped the units from the CISWI baseline and controlled dataset to the 2014 NEI
inventory and computed percent reductions such that our future year emissions matched the CISWI
controlled dataset values. Table 4-39 summarizes the total impact of CISWI controls for 2023 and 2028.
Note that this rule applies to specific units in 11 states: Alaska, Arkansas, Illinois, Iowa, Louisiana,

Maine, Oklahoma, Oregon, Pennsylvania, Tennessee, and Texas for CO, S02, and NOX.

Table 4-39. Summary of CISWI rule impacts on ptnonipm emissions for 2023 and 2028

year

poll

2016vl
(tons)

emissions

reductions

(tons)

%

change

2023

CO

1,446,353

-2,745

-0.2%

2023

NOX

952,181

-1,711

-0.2%

203


-------
2023

S02

658,204

-1,807

-0.3%

2028

CO

1,446,353

-2,937

-0.2%

2028

NOX

952,181

-1,722

-0.2%

2028

S02

658,204

-1,933

-0.3%

4.2.4.8 Petroleum Refineries NSPS Subpart JA (ptnonipm)

Packets:

Control_2016_202X_NSPS_Subpart_Ja_ptnonipm_beta_platform_ext_25sep2019_v0

On June 24, 2008, EPA issued final amendments to the Standards of Performance for Petroleum
Refineries. This action also promulgated separate standards of performance for new, modified, or
reconstructed process units after May 14, 2007 at petroleum refineries. The final standards for new
process units included emissions limitations and work practice standards for fluid catalytic cracking units,
fluid coking units, delayed coking units, fuel gas combustion devices, and sulfur recovery plants. In 2012,
EPA finalized the rule after some amendments and technical corrections. See
https://www.epa.gov/stationarv-sources-air-pollution/petroleum-refineries-new-source-performance-
standards-nsps-40-cfr for more details on NSPS - 40 CFR 60 Subpart Ja. These NSPS controls were
applied to petroleum refineries in the ptnonipm sector for years 2023 and 2028. Units impacted by this
rule were identified in the 2016vl inventory. For delayed coking units, an 84% control efficiency was
applied and for storage tanks, a 49% control efficiency was applied. The analysis of applicable units was
completed prior to the 2014v2 NEI and the 2016vl platform. Therefore, to ensure that a control was not
applied to a unit that was already in compliance with this rule, we compared emissions from the 2016vl
inventory and the 2011 en inventory (the time period of the original analysis). Any unit that demonstrated
a 55+%> reduction in VOC emissions from 201 len to 2016vl would be considered compliant with the rule
and therefore not subject to this control. Table 4-40 below reflects the impacts of these NSPS controls on
the ptnonipm sector. This control is applied to all pollutants; Table 4-40 summarizes reductions for the
years 2023 and 2028 for NOX, S02, and VOC.

Table 4-40. Summary of NSPS Subpart JA rule impacts on ptnonipm emissions for 2023 and 2028





2016vl

emissions



year

poll

(tons)

reductions (tons)

% change

2023

NOX

952,181

-1

0.0%

2023

S02

658,204

-3

0.0%

2023

VOC

774,289

-5,269

-0.7%

2028

NOX

952,181

-1

0.0%

2028

S02

658,204

-3

0.0%

2028

VOC

774,289

-5,233

-0.7%

4.2.4.9 Ozone Transport Commission Rules (nonpt)

Packets:

Control_2016_202X_nonpt_OTC_v l_platform_MARAMA_04oct2019_v 1
Control_2016_202X_nonpt_PF C_v l_platform_MARAMA_04oct2019_v 1

204


-------
Several MARAMA states have adopted rules reflecting the recommendations of the Ozone Transport
Commission (OTC) for reducing VOC emissions from consumer products, architectural and industrial
maintenance coatings, and various other solvents. The rules affected 27 different SCCs in the surface
coatings (2401xxxxxx), degreasing (2415000000), graphic arts (2425010000), miscellaneous industrial
(2440020000), and miscellaneous non-industrial consumer and commercial (246xxxxxxx) categories.
Not all states adopted all rules.

The OTC also developed a model rule to address VOC emissions from portable fuel containers (PFCs) via
performance standards and phased-in PFC replacement that was implemented in two phases. Some states
adopted one or both phases of the OTC rule, while others relied on the Federal rule. MARAMA
calculated control factors to reflect each state's compliance dates and, where states implemented one or
both phases of the OTC requirements prior to the Federal mandate, accounted for the early reductions in
the control factors. The rules affected permeation, evaporation, spillage, and vapor displacement for
residential (250101 lxxx) and commercial (2501012xxx) portable gas can SCCs.

MARAMA provided control packets to apply the solvent and PFC rule controls.
4.2.4.10 State-Specific Controls (ptnonipm)

Packets:

Control_2016_202X_ptnonipm_NC_BoilerMACT_beta_platform_ext_25sep2019_v0

Control_2016_202X_AZ_Regional_Haze_ptnonipm_beta_platform_ext_25sep2019_v0

CONTROL_2016_202X_Consent_Decrees_other_state_comments_beta_platform_extended_04oct2019_v 1

CONTROL_2016_202X_Consent_Decrees_ptnonipm_v l_platform_MARAMA_l 0sep2019_v0

CONTROL_2016_202X_DC_supplemental_ptnonipm_v l_platform_04oct2019_v 1

ICI Boilers - North Carolina

The Industrial/Commercial/Institutional Boilers and Process Heaters MACT Rule, hereafter simply
referred to as the "Boiler MACT," was promulgated on January 31, 2013, based on reconsideration.
Background information on the Boiler MACT can be found at: https://www.epa.gov/stationary-sources-
air-pollution/industrial-commercial-and-institutional-boilers-and-process-heaters. The Boiler MACT
promulgates national emission standards for the control of HAPs (NESHAP) for new and existing
industrial, commercial, and institutional (ICI) boilers and process heaters at major sources of HAPs. The
expected cobenefit for CAPs at these facilities is significant and greatest for S02 with lesser impacts for
direct PM, CO and VOC. This control addresses only the expected cobenefits to existing ICI boilers in the
State of North Carolina. All other states previously considered for this rule are assumed to be in
compliance with the rule and therefore the emissions need no further estimated controls applied. The
control factors applied here were provided by North Carolina.

Arizona Regional Haze Controls

U.S. EPA Region 9 provided regional haze FIP controls for a few industrial facilities. Information on
these controls are available in the docket https://www.regulations.gov/document?D=EPA-R09-OAR-
2013-0588-0072. These non-EGU controls have implementation dates between September 2016 and
December 2018.

205


-------
Consent Decrees

MARAMA provided a list of controls relating to consent decrees to be applied to specific units within the
MARAMA region. This list includes sources in North Carolina that were subject to controls in the beta
version of this emission modeling platform. Outside of the MARAMA region, controls related to consent
decrees were applied to several sources, including the LaFarge facility in Michigan (8127411), for which
NOX emissions must be reduced by 18.633% to meet the decree; and the Cabot facilities in Louisiana and
Texas, which had been subject to consent decree controls in the 2011 platforms, and 2016 emissions
values suggest controls have not yet taken effect. Other facilities subject to a consent decree were
determined to already be in compliance based on 2016 emissions values.

State Comments

A comment from the State of Illinois that was included in the 2011 platform was carried over for the
2016vl platform. The data accounts for three coal boilers being replaced by two gas boilers not in the
inventory and results in a large S02 reduction.

The State of Ohio reported that the P. H. Glatfelter Company facility (8131111) has switched fuels after
2016, and so controls related to the fuel switch were applied. This is a new control for version 1 platform.

Comments relating to Regional Haze in the 2011 platform were analyzed for potential use in the 2016vl
platform. For those comments that are still applicable, control efficiencies were recalculated so that
2016vl post-control emissions (without any projections) would equal post-control emissions for the 2011
platform (without any projections). This is to ensure that controls which may already be applied are
accounted for. Some facilities' emissions were already less than the 2011 post-control value in 2016vl
and therefore did not need further controls here. For facility 3982311 (Eastman Chemical in Tennessee),
one unit has a control efficiency of 90 in 2016vl and the others have no control; a replacement control of
91.675 was applied for this facility so that the unit with control efficiency=90 is not double controlled.

Wisconsin provided alternate emissions to use as input to 2023vl/2028vl CoST. Wisconsin provided new
emissions totals for three facilities and requested that these new totals be used as the basis for 2023vl and
2028vl projections, instead of 2016vl. The provided emissions were facility-level only, therefore 2016vl
emissions were scaled at these facilities to match the new provided totals.

The District of Columbia provided a control packet to be applied to three ptnonipm facilities in all 2016vl
platform projections.

4.3 Projections Computed Outside of CoST

Projections for some sectors are not calculated using CoST. These are discussed in this section.

4.3.1 Nonroad Mobile Equipment Sources (nonroad)

Outside California and Texas, the MOVES2014b model was run separately for each future year, including
2023 and 2028, resulting in a separate inventory for each year. The fuels used are specific to each future
year, but the meteorological data represented the year 2016. The 2023 and 2028 nonroad emission factors
account for regulations such the Emissions Standards for New Nonroad Spark-Ignition Engines,
Equipment, and Vessels (https://www.epa.gov/regulations-emissions-vehicles-and-engines/final-rule-
control-emissions-nonroad-spark-ignition). Locomotives and Marine Compression-Ignition Engines Less
than 30 Liters per Cylinder (https://www.epa.gov/regulations-emissions-vehicles-and-engines/final-rule-

206


-------
control-emissions-air-pollution-locomotive), and Clean Air Nonroad Diesel Final Rule - Tier 4
(https://www.epa.gov/regulations-emissions-vehicles-and-engines/final-rule-control-emissions-air-
pollution-nonroad-dieseD. The resulting future year inventories were processed into the format needed by
SMOKE in the same way as the base year emissions. Inside California and Texas, CARB and TCEQ
provided separate datasets for each future year. Because the CARB and TCEQ inventories already reflect
future year emissions, no additional work related to projections was required except to include them as
SMOKE input files.

4.3.2 Onroad Mobile Sources (onroad)

The MOVES2014b model was run separately for each future year, including 2023 and 2028, resulting in
separate emission factors for each year. The 2023 and 2028 onroad emission factors account for changes
in activity data and the impact of on-the-books rules that are implemented into MOVES2014b. These
include regulations such as the Light Duty Vehicle GHG Rule for Model-Year 2017-2025, and the Tier 3
Motor Vehicle Emission and Fuel Standards Rule (https://www.epa.gov/regulations-emissions-vehicles-
and-engines/final-rule-control-air-pollution-motor-vehicles-tier-3). Local inspection and maintenance
(I/M) and other onroad mobile programs are included such as California LEVIII, the National Low
Emissions Vehicle (LEV) and Ozone Transport Commission (OTC) LEV regulations
(https://www.epa.gov/regulations-emissions-vehicles-and-engines/final-rule-control-air-pollution-new-
motor-vehicles-and-2). local fuel programs, and Stage II refueling control programs. Regulations finalized
after the year 2014 are not included, such as the Safer Affordable Fuel Efficient (SAFE) Vehicles Final
Rule for Model Years 2021-2026 and the Final Rule for Phase 2 Greenhouse Gas Emissions Standards
and Fuel Efficiency Standards for Medium- and Heavy-Duty Engines and Vehicles (HD GHG P2).

The fuels used are specific to each future year, the age distributions were projected to the future year, and
the meteorological data represented the year 2016. The resulting emission factors were combined with
future year activity data using SMOKE-MOVES run in a similar way as the base year. The development
of the future year activity data is described later in this section. CARB provided separate emissions
datasets for each future year. The CARB-provided emissions were adjusted to match the temporal and
spatial patterns of the SMOKE-MOVES based emissions. Additional information about the development
of future year onroad emission and on how SMOKE was run to develop the emissions can be found in the
2016vl platform onroad sector specification sheet.

Where state and local agencies did not provide future year activity data, future year VMT were computed
based on annual VMT data from the AEO2019 reference case for VMT by fuel and vehicle type.
Specifically, the following two AEO2019 tables were used:

•	Light Duty (LD): Light-Duty VMT by Technology Type (table #51:
https://www.eia. gov/outlooks/aeo/data/browser/#/?id=51-
AEQ2019&cases=ref2019&sourcekev=0)

•	Heavy Duty (HD): Freight Transportation Energy Use (table #58:
https://www.eia. gov/outlooks/aeo/data/browser/#/?id=58-
AEQ2019&cases=ref2019&sourcekev=0)

Total VMT for each MOVES fuel and vehicle grouping was calculated for the years 2016, 2020, 2023,
and 2028 based on the AEO-to-MOVES mappings above. From these totals, 2016-2023 and 2016-2028
VMT trends were calculated for each fuel and vehicle grouping. Those trends became the national VMT
projection factors. The AEO2019 tables include data starting from the year 2017. Since we were

207


-------
projecting from 2016, 2016-to-2017 projection factors were calculated from AEO2018, and then
multiplied by 2017-to-future projection factors from AEO2019. MOVES fuel and vehicle types were
mapped to AEO fuel and vehicle classes. The resulting 2016-to-future year national VMT projection
factors used for the 2016vl platform are provided in Table 4-41. These factors were adjusted to prepare
county-specific projection factors for light duty vehicles based on human population data available from
the BenMAP model by county for the years 2017, 2023, and 2028 (https://www.woodsandpoole.com/
circa 2015). The purpose of this adjustment based on population changes helps account for areas of the
country that are growing more than others. Where agencies provided future year VMT data, those data
were used.

Table 4-41. Factors used to Project 2016 VMT to 2023 and 2028

SCC6

description

2023
factor

2028 factor

220111

LD gas

5.99%

6.99%

220121

LD gas

5.99%

6.99%

220131

LD gas

5.99%

6.99%

220132

LD gas

5.99%

6.99%

220142

Buses gas

8.43%

19.86%

220143

Buses gas

8.43%

19.86%

220151

MHDgas

8.43%

19.86%

220152

MHDgas

8.43%

19.86%

220153

MHDgas

8.43%

19.86%

220154

MHDgas

8.43%

19.86%

220161

HHDgas

-51.15%

-64.99%

220221

LD diesel

86.79%

177.3%

220231

LD diesel

86.79%

177.3%

220232

LD diesel

86.79%

177.3%

220241

Buses diesel

14.30%

21.23%

220242

Buses diesel

14.30%

21.23%

220243

Buses diesel

14.30%

21.23%

220251

MHD diesel

14.30%

21.23%

220252

MHD diesel

14.30%

21.23%

220253

MHD diesel

14.30%

21.23%

220254

MHD diesel

14.30%

21.23%

220261

HHD diesel

12.91%

17.85%

220262

HHD diesel

12.91%

17.85%

220342

Buses CNG

65.57%

88.00%

220521

LD E-85

-0.70%

-10.03%

220531

LD E-85

-0.70%

-10.03%

220532

LD E-85

-0.70%

-10.03%

220921

LD Electric

1258%

2695%

220931

LD Electric

1258%

2695%

220932

LD Electric

1258%

2695%

Future year VPOP data were projected using calculations of VMT/VPOP ratios for each county, fuel, and
vehicle type from the 2016 VMT and VPOP data. Those ratios were then applied to the future year

208


-------
projected VMT to estimate future year VPOP. Future year VPOP data submitted by state and local
agencies were then incorporated into the VPOP projections. Future year VPOP data were provided by
state and local agencies in NH, NJ, NC, WI, Pima County, AZ, and Clark County, NV. All of these
submissions were the same as for the 2016beta platform except for New Jersey, which provided a new
submission for the 2016vl platform. For Pima County, just like with the VMT, future year VPOP was
only provided for 2022 (used directly for 2023) and not for 2028. Where necessary, VPOP was split to
SCC (full FF10) using SCC distributions from the EPA projection. Both VMT and VPOP were
redistributed between the LD car and truck vehicle types (21/31/32) based on splits from the EPA
projection, and used the EPA projection for buses in North Carolina and state-provided VPOP for all
other vehicles in North Carolina.

Hoteling hours were projected to the future years by calculating 2016 inventory HOTELING/VMT ratios
for each county for combination long-haul trucks on restricted roads only. Those ratios were then applied
to the future year projected VMT for combination long-haul trucks on restricted roads to calculate future
year hoteling. Some counties had hoteling activity but did not have combination long-haul truck restricted
road VMT in 2016; in those counties, the national AEO2018-based projection factor for diesel
combination trucks was used to project 2016 hoteling to the future years. This procedure gives county-
total hoteling for the future years. Each future year also has a distinct APU percentage based on MOVES
input data that was used to split county total hoteling to each SCC: 22.6% APU for 2023, and 25.9% APU
for 2028.

4.3.3 Locomotives (rail)

Rail emissions were computed for future years based on future year fuel use values for 2020, 2023, and
2028 were based on the Energy Information Administration's 2018 Annual Energy Outlook (AEO)
freight rail energy use growth rate projections for 2016 thru 2028 (see Table 4-42) and emission factors
based on historic emissions trends that reflect the rate of market penetration of new locomotive engines.

A correction factor was added to adjust the AEO projected fuel use for 2017 to match the actual 2017 R-l
fuel use data. The additive effect of this correction factor was carried forward for each subsequent year
from 2018 thru 2028. The modified AEO growth rates were used to calculate future year Class I line-haul
fuel use totals for 2020, 2023, and 2028. As shown in Table 4-42 the future year fuel use values ranged
between 3.2 and 3.4 billion gallons, which matched up well with the long-term line-haul fuel use trend
between 2005 and 2018. The emission factors for NOx, PM10 and VOC were derived from trend lines
based on historic line-haul emission factors from the period of 2007 through 2017.

Table 4-42. Class I Line-haul Fuel Projections based on 2018 AEO Data

Year

AEO Kreighl
Eador

Projection
Kaclor

Corrected AEO Kuel

Uaw AEO l uel

2016

1

1

3,203,595,133

3,203,595,133

2017

1.0212

1.0346

3,314,384,605

3,271,393,249

2018

1.0177

1.0311

3,303,215,591

3,260,224,235

2019

1.0092

1.0226

3,275,939,538

3,232,948,182

2020

1.0128

1.0262

3,287,479,935

3,244,488,580

2021

1.0100

1.0235

3,278,759,301

3,235,767,945

2022

0.9955

1.0090

3,232,267,591

3,189,276,235

209


-------
Year

AKO Kreight
Kaclor

Projection
Kador

Corrected AKO Kuel

Uaw AEO l uel

2023

0.9969

1.0103

3,236,531,624

3,193,540,268

2024

1.0221

1.0355

3,317,383,183

3,274,391,827

2025

1.0355

1.0489

3,360,367,382

3,317,376,026

2026

1.0410

1.0544

3,377,946,201

3,334,954,845

2027

1.0419

1.0553

3,380,697,189

3,337,705,833

2028

1.0356

1.0490

3,360,491,175

3,317,499,820

The projected fuel use data was combined with the emission factor estimates to create future year link-
level emission inventories based on the MGT traffic density values contained in the FRA's 2016
shapefile. The link-level data created for 2020, 2023, and 2028 was aggregated to create county, state, and
national emissions estimates (see Table 4-43) which were then converted into FF10 format for use in the
2016vl emissions platform.

Table 4-43. Class I Line-haul Historic and Future Year Projected Emissions

Inventory

CO

IK

Ml 13

NOx

I'M 10

IWI2.5

S()2

2007 (2008 NEI)

110,969

37,941

347

754,433

25,477

23,439

7,836

2014 NEI

107,995

29,264

338

609,295

19,675

18,101

381

2016 vl

94,020

21,727

294

489,562

14,538

14,102

332

2017 NEI

97,272

21,560

304

492,385

14,411

13,979

343

2020 Projected

96,482

19,133

302

448,924

12,800

12,415

340

2023 Projected

94,987

16,550

297

404,329

11,059

10,728

335

2028 Projected

98,625

13,847

309

361,914

9,236

8,959

348

2016 vs 2028

4.90%

-36.27%

4.90%

-26.07%

-36.47%

-36.47%

4.90%

Other rail emissions were projected based on AEO growth rates as shown in Table 4-44. See the 2016vl
rail specification sheet for additional information on rail projections.

Table 4-44. AEO growth rates for rail sub-groups

Sector

2016

2020

2023

2028

Rail Yards

1.0

0.97513

0.947802

0.952483

Class II/III Railroads

1.0

0.97513

0.947802

0.952483

C ommuter/Pas senger

1.0

1.033858

1.071348

1.136023

4.3.1 Sources Added in the 2021fi Case

New units were identified in the 2018 NEI point source inventory which were not in the 2016fi inventory.
These four units were included in the ptnonipm sector of the 2021fi case with emissions values from
2018. The sources added in the 2021fi case are listed in Table 4-45.

210


-------
Table 4-45. Sources Added in the 2021fi Case

II PS

( ounlv/Slale

1'acililv id

1'acilitv name

\()\

I'M 2.5

YOC

08081

Moffat Co, CO

1839411

COLOWYO COAL CO -
COLOWYO & COLLOM
MINES

725

20

0

27137

St Louis Co, MN

13598411

US Steel Corp - Keetac

5,005

443

49

28141

Tishomingo Co, MS

17942211

MISSISSIPPI SILICON
LLC

837

79

3

30031

Gallatin Co, MT

7766011

TRIDENT

1,081

29

0

4.3.2 Sources Outside of the United States (onroad_can, onroad_mex,
othpt, ptfire_othna, othar, othafdust, othptdust)

This section discusses the projection of emissions from Canada and Mexico and other areas outside of the
U.S. Information about the base inventory used for these projections or the the naming conventions can be
found in Section 2.7. Emissions for Mexico are based on the Inventario Nacional de Emisiones de
Mexico, 2008 projected to years 2023 and 2028 (ERG, 2014a). Additional details for these sectors can be
found in the 2016vl platform specification sheets.

4.3.2.1 Canadian fugitive dust sources (othafdust, othptdust)

Canadian area source dust (othafdust)

ECCC provided area stationary source inventories for the years 2023 and 2028. Unlike in their 2015
inventories in which area dust emissions were grouped into a separate file, these sources were not
provided as separate inventories for the future years, and so othafdust sector emissions were extracted
from that single area source inventory. As with 2015, the future year dust emissions are pre-adjusted, so
future year othafdust follows the same emissions processing methodology as the base year. To make the
future year emissions consistent with the base year, the same 2015->2010 adjustment factors for
construction dust that were applied to the base year inventory were also applied to the future year
projected inventories.

Canadian point source dust (othptdust)

ECCC had provided their own future year projections of the harvest and tillage point ag dust inventories,
but those inventories exhibited the same waffle pattern as 2015, so we instead decided to project the
improved 2015 inventories. ECCC separately provided data from which future year projections could be
derived in a file called "Projected_CAN2015_2023_2028.xlsx", which includes emissions data for 2015,
2023, and 2028 by pollutant, province, ECCC sub-class code, and other source categories. This data was
used to calculate 2015-to-2023 and 2015-to2028 projection factors, which were then applied to the
improved 2015 Canada point ag dust inventories to create projections for 2023 and 2028. Emissions
values from these in-house projections were found to be close in magnitude to ECCC's own projections.
Projection factors were applied by province, sub-class code, and pollutant. The ECCC projection
workbook included additional source information which provides more detail than do the subclass codes,
but that more detailed information could not be easily mapped to the inventory, and the level of detail
offered by the sub-class codes was considered sufficient for projection purposes. For the othptdust sector,
there are separate sub-class codes for each of the two inventories (harvest and tillage).

211


-------
4.3.2.2 Point Sources in Canada and Mexico (othpt)

Canada point airport and agriculture emissions

Future year airport and agriculture emission inventories from ECCC were not available in time for
inclusion in the platform. Instead, ECCC provided data from which future year projections of these
inventories could be derived. This data, provided by ECCC in a file called

"Projected_CAN2015_2023_2028.xlsx", includes emissions data for 2015, 2023, and 2028 by pollutant,
province, ECCC sub-class code, and other source categories. This data was used to calculate 2015-to-
2023 and 2015-to-2028 projection factors, which were then applied to the improved 2015 point airport
and ag inventories to create projections of Canadian emissions for 2023 and 2028. Projection factors were
applied by province, sub-class code, and pollutant. The ECCC projection workbook included additional
source information which provides more detail than do the subclass codes, but that more detailed
information could not be easily mapped to the inventory, and the level of detail offered by the sub-class
codes was considered sufficient for projection purposes. For the ag inventories, the sub-class codes are
similar in detail to SCCs: fertilizer has a single sub-class code, and animal emissions categories (broilers,
dairy, horses, sheep, etc) each have a separate sub-class code. Sub-class codes for airport emissions are
similar in detail to SCCs, with separate codes for piston and turbine emissions from military aircraft,
commercial aircraft, and general aviation.

Other Canada point sources

Future year projections for stationary point sources (excluding ag) were provided by ECCC for 2023 and
2028. ECCC provided emissions inventories for upstream oil and gas sources (UOG) and for all other
stationary point sources, including electric power generation. These inventories were generally used as-is,
with the following exceptions. The 2015 non-UOG stationary point source inventories included monthly
emissions as well as annual emissions. In the future year projected inventories provided by ECCC,
monthly emissions were included not included for EPG (electric power generation) sources, but were for
the rest of the non-UOG sources. For consistency with the base year, monthly emissions were added to
the EPG sources in the inventory, using facility-specific monthly temporal profiles derived from the 2015
inventory. For new facilities that were not in 2015, monthly emissions were left blank in the inventory,
and monthly temporalization is applied SMOKE using profiles assigned by SCC. For 2015, ECCC
provided a pre-speciated point source inventory including species for the CB6 mechanism. For the future
years, ECCC did not provide a pre-speciated inventory, but advised that speciation for the future years is
unchanged from the base year. Because the baseline VOC emissions are different in the future year
projections, it was necessary to develop a prespeciated CB6 inventory for the future years which is
consistent with the 2015 inventory but is based on future year projections of VOC. For this, speciation
profiles for each facility-SCC in 2015 were calculated using the 2015 CB6 inventory, and these profiles
were applied to future year VOC to create a CB6 future year inventory. Speciation profiles were also
developed by SCC from 2015, for application to future year facility-SCC combinations which could not
be matched to 2015. The future year inventories also include SCCs which were not in the 2015 inventory
all; for those sources, we apply standard speciation profiles in SMOKE. To prevent double counting of
VOC speciated within SMOKE with pre-speciated VOC, the point source inventory has VOC emissions
represented as VOCINV for sources that are in the pre-speciated CB6 inventory, and as VOC for sources
that are not pre-speciated. Only the VOC and not the VOC INV is speciated within SMOKE. Changes to
point source IDs in the stationary source inventory were necessary for the PMC calculation, which is
based on inventory PM10 and PM2.5. This SMOKE calculation requires that PM10 and PM2.5 emissions
are assigned to the same point source IDs, but that was not always the case with respect to the
rel_point_id and process id fields for each unit. This was also an issue with the 2015 inventory, but the
procedure that was used to fix 2015 did not help resolve this issue in the future year inventories, and so a
more robust fix was implemented for 2023 and 2028. All rel_point_id and processed values in the 2023

212


-------
and 2028 Canada stationary point inventories were redefined, such that all records with the same FIPS
code, latitude, longitude, and stack parameters (implying emissions from the same stack) were assigned
the same rel_point_id and process id for all pollutants. This fixed all instances in which PM10 and PM2.5
from the same source were assigned different point source IDs, but there are still sources in the future
year inventories in which PM10 emissions are less than the PM2.5 emissions from the same source.

Mexico

The othpt sector includes a general point source inventory in Mexico. This inventory is based on
projections of a 2008 inventory. The inventory was originally projected to years 2018, 2025, and 2030 by
ERG1 . For the beta and vl platform future year projections, emissions values from 2018 and 2025 were
interpolated to 2023, and values from 2025 and 2030 were interpolated to 2028. These inventories are
unchanged from the 2011 platform.

4.3.2.3 Nonpoint sources in Canada and Mexico (othar)

Canadian stationary sources

ECCC provided area stationary source inventories for the years 2023 and 2028. Unlike in their 2015
inventories in which dust and agricultural emissions were grouped into separate files, these sources were
not provided as separate inventories for the future years. Therefore, dust emissions from the othafdust and
othptdust sectors, and ag emissions from the othpt sector, needed to be removed from the future year area
source inventory to prevent a double count. PM emissions for all SCCs in the othafdust inventory (see
othafdust sector document) were moved to a separate inventory. Then, most emissions from agricultural
SCCs (2801- and 2805-) were removed, since the NH3 and VOC emissions overlap the point format ag
inventories which are part of the othpt sector, and the PM emissions were either already moved to the
othafdust sector, overlap the othptdust sector, or were not present in 2015 (see note about fertilizer
below). One ag SCC was partially retained in the area source inventory according to both the SCC and
ECCC's 5-digit "sub-class codes". SCC 2805000000 for sub-class code 80104, which represents
agricultural fuel combustion, was not removed from the area source inventory, since these emissions were
part of the othar sector in 2016ff and are not included in any of the other inventories. PM emissions from
fertilizer were not present in any 2015 ECCC inventory, but did appear in the future year area source
inventory. According to ECCC, this was an error in 2015, and the 2015 inventories should have included
approximately 7,000 tons per year of PM emissions from fertilizer. Fertilizer PM emissions were also
excluded from in future year modeling to preserve consistency between modeling years. ECCC provided
an additional stationary area source inventory for 2023 and 2028 representing electric power generation
(EPG). According to ECCC, this inventory's emissions were covered by the point source EPG inventory
in 2015 and does not double count the 2023 and 2028 point source inventories, and it is appropriate to
include this new area source EPG inventory in the othar sector.

Canadian mobile sources

For mobile nonroad sources, including rail and CMV, future year inventories from ECCC were not
available in time for inclusion in beta platform. Instead, ECCC provided data from which future year
projections of these inventories could be derived. This data, provided by ECCC in a file called
"Projected_CAN2015_2023_2028.xlsx", includes emissions data for 2015, 2023, and 2028 by pollutant,
province, ECCC sub-class code, and other source categories. This data was used to calculate 2015-to-
2023 and 2015-to-2028 projection factors, which were then applied to the 2015 mobile source inventories
to create projections of Canadian mobile source emissions for 2023 and 2028. Projection factors were
applied by province, sub-class code, and pollutant. The ECCC projection workbook included additional
source information which provides more detail than do the subclass codes, but that more detailed
information could not be easily mapped to the inventory, and the level of detail offered by the sub-class

213


-------
codes was considered sufficient for projection purposes. For the nonroad inventory, the sub-class code is
analogous to the SCC7 level in U.S. inventories. For example, there are separate sub-class codes for fuels
(e.g. 2-stroke gasoline, diesel, LPG) and category (e.g. construction, lawn and garden) but not for
individual vehicle types within each category (e.g. snowmobiles, tractors). For CMV and rail, the sub-
class code is closer to full SCC, because there are separate codes for port and underway emissions, and for
freight and passenger rail emissions.

Mexico

The othar sector includes two Mexico inventories, an area inventory and a nonroad inventory. Similar to
2016, the future year Mexico inventories are based on projections of a 2008 inventory, but are based on
different interpolations. In addition to the 2014 and 2018 projections that were the basis for 2016, these
inventories were also originally projected to years 2025 and 2030. For future year projections, emissions
values from 2018 and 2025 were interpolated to 2023, and emissions values from 2025 and 2030 were
interpolated to 2028. These emissions are unchanged from the 2011 platform, except that CMV emissions
were removed from the nonroad inventory to prevent a double count with the Mexico CMV inventory,
which was not part of the 2011 platform.

4.3.2.1 Onroad sources in Canada and Mexico (onroad_can,
onroad_mex)

For Canadian mobile onroad sources, future year inventories from ECCC were not available in time for
inclusion in the vl platform. Instead, ECCC provided data from which future year projections of these
inventories could be derived. This data, provided by ECCC in a file called

"Projected_CAN2015_2023_2028.xlsx", includes emissions data for 2015, 2023, and 2028 by pollutant,
province, ECCC sub-class code, and other source categories. This data was used to calculate 2015-to-
2023 and 2015-to-2028 projection factors, which were then applied to the 2015 mobile source inventories
to create projections of Canadian mobile source emissions for 2023 and 2028. Projection factors were
applied by province, sub-class code, and pollutant. The ECCC projection workbook included additional
source information which provides more detail than do the subclass codes, but that more detailed
information could not be easily mapped to the inventory, and the level of detail offered by the sub-class
codes was considered sufficient for projection purposes. For the onroad inventory, the sub-class code is
analogous to the SCC6+process level in U.S. inventories, in that it specifies fuel type, vehicle type, and
process (e.g. brake, tire, exhaust, refueling), but not road type.

For Mexican mobile onroad sources, MOVES-Mexico was run to create emissions inventories for years
2023 and 2028. Results from those runs are used in future year emissions processing for the vl platform.
These emissions are unchanged from the 2011 platform.

214


-------
5 Emission Summaries

Tables 5-1 through 5-6 summarize emissions by sector for the 2016fh, 2023fhl, and 2028fhl cases. These
summaries are provided at the national level by sector for the contiguous U.S. and for the portions of
Canada and Mexico inside the larger 12km domain (12US1) discussed in Section 3.1 and for the 36-km
domain (36US3). Note that totals for the 12US2 domain are not available here, but the sum of the U.S.
sectors would be essentially the same and only the Canadian and Mexican emissions would change
according to how far north/south the grids go. Tables 5-7 and 5-8 summarize emissions for the 2016fi and
2021fi cases. Note that the afdust sector emissions here represent the emissions after application of both
the land use (transport fraction) and meteorological adjustments; therefore, this sector is called
"afdust adj" in these summaries. The afdust emissions in the 36km domain are smaller than those in the
12km domain due to how the adjustment factors are computed and the size of the grid cells. The onroad
sector totals are post-SMOKE-MOVES totals, representing air quality model-ready emission totals, and
include CARB emissions for California. The cmv sectors include U.S. emissions within state waters only;
these extend to roughly 3-5 miles offshore and includes CMV emissions at U.S. ports. "Offshore"
represents CMV emissions that are outside of U.S. state waters. Canadian CMV emissions are included in
the other sector. The total of all US sectors is listed as "Con U.S. Total."

Tables 5-9 and 5-10 show national total ozone season NOx and VOC emissions, respectively. A
spreadsheet of these emissions that includes state totals is included in the docket EPA-HQ-OAR-2020-0272
on on https://regulations.gov as "State totals of ozone season NOx emissions across years" (i.e.,
state_totals_2016-2021 -2023-2028_maysep_calc202 l_updated_airports_v3 .xlsx).

State totals and other summaries are available in the reports area on the web and FTP sites for the 2016vl
platform (https://www.epa.gov/air-emissions-modeling/2016vl-platform.

ftp://newftp.epa.gov/air/emismod/2016/vl/). If you cannot access the FTP site through the provided link,
this link points to the same data: https://gaftp.epa.gov/Air/emismod/2016/v 1 /.

215


-------
Table 5-1. National by-sector CAP emissions summaries for the 2016fh case, 12US1 grid (tons/yr)

Sector

CO

NH3

NOX

PM10

PM2 5

S02

voc

afdustadj







7,203,692

1,006,446





ag



3,409,761









194,779

airports

674,176

0

185,454

11,068

9,805

25,412

85,768

cmv_clc2

23,548

83

162,502

4,457

4,320

634

6,436

cmv_c3

13,956

39

110,462

2,201

2,025

4,528

8,600

nonpt

2,629,755

78,509

710,918

570,314

463,807

138,650

3,695,093

nonroad

10,593,274

1,845

1,110,277

109,196

103,230

2,133

1,128,691

npoilgas

759,771

12

572,043

14,050

13,984

19,243

2,792,092

onroad

19,889,617

100,318

3,630,693

239,997

117,758

27,559

1,852,260

ptagfire

262,645

51,276

10,240

38,688

26,951

3,694

17,181

ptegu

658,346

23,976

1,290,190

163,981

133,517

1,540,589

33,739

ptfire

13,717,466

239,605

227,337

1,461,693

1,234,062

111,291

3,109,465

ptnonipm

1,439,081

63,731

940,031

396,884

254,386

654,527

770,204

ptoilgas

167,531

4,338

339,280

11,301

10,784

33,227

127,565

rail

104,551

326

559,381

16,344

15,819

457

26,082

rwc

2,119,402

15,439

31,282

317,469

316,943

7,703

340,941

















Con. U.S. Total

53,053,119

3,989,258

9,880,090

10,561,336

3,713,836

2,569,647

14,188,893

















beis

7,167,921



965,761







42,133,700

CONUS + beis

60,221,040

3,989,258

10,845,852

10,561,336

3,713,836

2,569,647

56,322,592

















Can./Mex./Offshore















Sector

CO

NH3

NOX

PM10

PM2 5

S02

VOC

Canada othafdust







1,060,979

187,228





Canada othar

2,727,917

4,842

397,394

313,494

248,467

19,939

832,491

Canada onroadcan

1,665,792

6,877

404,856

25,204

14,076

1,556

143,213

Canada othpt

1,081,673

503,214

657,348

115,280

46,765

993,944

797,611

Canada othptdust







150,832

55,539





Canada ptfireothna

761,402

13,032

16,359

84,476

71,745

6,731

185,476

Canada CMV

10,741

37

93,456

1,682

1,563

2,984

5,184

Mexico othar

241,571

201,994

220,491

115,460

54,294

7,717

522,236

Mexico onroad mex

1,828,101

2,789

442,410

15,151

10,836

6,247

158,812

Mexico othpt

171,065

5,049

371,671

67,173

51,791

436,802

67,343

Mexico ptfire othna

383,162

7,436

16,604

44,992

38,176

2,785

131,499

Mexico CMV

0

0

0

0

0

0

0

Offshore cmv in Federal
waters

33,224

128

293,102

7,188

6,658

28,060

16,209

Offshore cmv outside
Federal waters

23,338

440

257,615

24,828

22,848

181,941

11,083

Offshore pt oilgas

50,052

15

48,691

668

667

502

48,210

Non-US Total

8,978,039

745,854

3,219,997

2,027,409

810,652

1,689,208

2,919,366

216


-------
Table 5-2. National by-sector CAP emissions summaries for the 2023fhl case, 12US1 grid (tons/yr)

Sector

CO

NH3

NOX

PM10

PM2 5

S02

voc

afdustadj







7,255,011

1,016,777





ag



3,543,157









205,451

airports

738,835

0

219,766

11,358

10,127

30,208

92,473

cmv_clc2

23,570

59

116,344

3,191

3,093

242

4,527

cmv_c3

16,709

48

104,555

2,623

2,413

5,380

10,397

nonpt

2,644,789

79,342

709,268

579,169

472,935

106,355

3,756,888

nonroad

10,581,376

2,032

737,625

71,457

66,940

1,527

856,474

npoilgas

788,072

20

585,230

16,221

16,102

31,269

3,203,738

onroad

13,773,993

89,285

1,751,007

199,979

72,468

12,484

1,098,966

ptagfire

262,645

51,276

10,240

38,688

26,951

3,694

17,181

ptegu

659,538

36,544

996

144,758

124,433

18,820

35,922

ptfire

13,717,466

239,605

227,337

1,461,693

1,234,062

111,291

3,109,465

ptnonipm

1,448,566

63,739

928,896

400,192

257,145

572,494

771,838

ptoilgas

186,242

4,377

361,166

13,602

12,973

38,125

156,725

rail

105,988

330

469,157

12,778

12,376

460

20,436

rwc

2,046,853

14,793

31,902

304,464

303,920

7,010

329,017

















Con. U.S. Total

46,994,644

4,124,607

6,253,489

10,515,185

3,632,716

939,358

13,669,497

















beis

7,167,921



965,761







42,133,700

CONUS + beis

54,162,565

4,124,607

7,219,250

10,515,185

3,632,716

939,358

55,803,196

















Can./Mex./Offshore















Sector

CO

NH3

NOX

PM10

PM2 5

S02

VOC

Canada othafdust







1,178,439

207,111





Canada othar

2,689,047

4,702

310,393

303,854

228,992

19,477

823,199

Canada onroadcan

1,418,143

6,043

234,813

25,849

10,996

752

87,466

Canada othpt

1,094,900

610,668

541,448

87,726

46,205

868,739

684,095

Canada othptdust







150,854

55,547





Canada ptfireothna

760,345

13,015

16,337

84,366

71,652

6,721

185,224

Canada CMV

11,597

40

67,837

1,819

1,690

3,158

5,525

Mexico othar

263,826

198,635

240,372

118,422

56,685

7,993

583,403

Mexico onroad mex

1,772,026

3,266

427,900

17,023

11,764

7,556

161,115

Mexico othpt

200,105

6,273

380,429

75,143

57,034

365,518

84,277

Mexico ptfire othna

384,764

7,466

16,665

45,198

38,354

2,798

131,980

Mexico CMV

0

0

0

0

0

0

0

Offshore cmv in Federal
waters

39,846

150

257,244

8,460

7,815

34,951

19,345

Offshore cmv outside
Federal waters

28,551

277

314,614

15,644

14,397

41,490

13,542

Offshore pt oilgas

50,052

15

48,691

668

667

502

48,210

Non-US Total

8,713,201

850,550

2,856,743

2,113,463

808,909

1,359,655

2,827,380

217


-------
Table 5-3. National by-sector CAP emissions summaries for the 2028fhl case, 12US1 grid (tons/yr)

Sector

CO

NH3

NOX

PM10

PM2 5

S02

voc

afdustadj







7,279,406

1,021,715





ag



3,564,066









207,123

airports

803,407

0

245,192

11,871

10,622

33,866

100,258

cmv_clc2

24,002

47.404946

92,763

2,549

2,471

243.87567

3,574

cmv_c3

19,175

53.299262

104,503

3,010

2,770

6,160

11,990

nonpt

2,665,492

79,603

708,891

593,878

485,092

106,954

3,800,741

nonroad

10,892,398

2,104

611,510

58,356

54,323

1,545

801,819

npoilgas

774,404

20.377326

560,267

16,462

16,343

33,574

3,331,524

onroad

10,308,234

87,913

1,246,069

189,838

58,925

11,703

836,112

ptagfire

262,645

51,276

10,240

38,688

26,951

3,694

17,181

ptegu

648,829

35,883

748,663

140,100

120,420

781,397

33,831

ptfire

13,717,466

239,605

227,337

1,461,693

1,234,062

111,291

3,109,465

ptnonipm

1,460,891

63,990

933,843

402,471

258,983

575,210

772,997

ptoilgas

186,008

4,383

355,109

14,119

13,477

40,437

160,295

rail

110,026

342.97954

423,103

10,953

10,611

472.9168

17,558

rwc

2,023,977

14,612

32,049

300,378

299,829

6,788

325,390

















Con. U.S. Total

43,896,953

4,143,899

6,299,537

10,523,775

3,616,594

1,713,335

13,529,856

















beis

7,167,921



965,761







42,133,700

CONUS + beis

51,064,874

4,143,899

7,265,298

10,523,775

3,616,594

1,713,335

55,663,555

















Can./Mex./Offshore















Sector

CO

NH3

NOX

PM10

PM2 5

S02

VOC

Canada othafdust







1,267,025

222,026





Canada othar

2,687,318

4,670

282,912

301,578

221,810

19,502

849,301

Canada onroadcan

1,303,551

5,492

168,631

26,129

9,498

698

60,932

Canada othpt

1,133,173

695,896

443,884

93,439

49,576

855,167

752,057

Canada othptdust







151,228

55,685





Canada ptfireothna

760,345

13,015

16,337

84,366

71,652

6,721

185,224

Canada CMV

12,247

42

73,084

1,921

1,785

3,361

5,832

Mexico othar

277,263

200,038

252,523

120,590

58,294

8,206

628,715

Mexico onroad mex

1,615,412

3,732

393,339

18,728

12,667

8,530

164,793

Mexico othpt

215,237

7,273

423,250

85,626

64,575

394,409

98,420

Mexico ptfire othna

384,764

7,466

16,665

45,198

38,354

2,798

131,980

Mexico CMV

0

0

0

0

0

0

0

Offshore cmv in Federal
waters

45,623

171

240,686

9,623

8,879

40,870

22,153

Offshore cmv outside
Federal waters

32,972

320

363,173

18,088

16,645

48,061

15,638

Offshore pt oilgas

50,052

15

48,691

668

667

502

48,210

Non-US Total

8,517,957

938,131

2,723,176

2,224,208

832,112

1,388,825

2,963,253

218


-------
Table 5-4. National by-sector CAP emissions summaries for the 2016fh case, 36US3 grid (tons/yr)

Sector

CO

NH3

NOX

PM10

PM2 5

S02

voc

afdustadj







7,205,579

1,006,637





ag



3,409,762









194,779

airports

675,321

0

185,708

11,097

9,832

25,452

85,912

cmv_clc2

23,786

84

164,075

4,498

4,360

636

6,489

cmv_c3

14,296

40

113,795

2,260

2,080

4,666

8,743

nonpt

2,631,492

78,565

711,375

570,526

463,960

138,883

3,695,797

nonroad

10,596,610

1,846

1,110,476

109,228

103,260

2,134

1,129,520

npoilgas

759,771

12

572,043

14,050

13,984

19,243

2,792,092

onroad

19,894,976

100,332

3,631,843

240,071

117,803

27,562

1,853,073

ptagfire

262,645

51,276

10,240

38,688

26,951

3,694

17,181

ptegu

658,346

23,976

1,290,190

163,981

133,517

1,540,589

33,739

ptfire

13,717,466

239,605

227,337

1,461,693

1,234,062

111,291

3,109,465

ptnonipm

1,439,095

63,731

940,048

396,913

254,394

654,527

770,205

ptoilgas

167,531

4,338

339,280

11,301

10,784

33,227

127,565

rail

104,551

326

559,381

16,344

15,819

457

26,082

rwc

2,119,890

15,442

31,291

317,537

317,011

7,704

341,020

















36US3 U.S. Total

53,065,776

3,989,335

9,887,082

10,563,766

3,714,454

2,570,065

14,191,662

















beis

7,232,588



968,624







42,374,150

36US3 U.S. Total + beis

60,298,364

3,989,335

10,855,706

10,563,766

3,714,454

2,570,065

56,565,812

















Can./Mex./Offshore















Sector

CO

NH3

NOX

PM10

PM2 5

S02

VOC

Canada othafdust







1,101,762

194,352





Canada othar

2,933,979

5,152

437,979

327,343

260,341

20,590

885,639

Canada onroadcan

1,730,052

7,125

425,462

26,286

14,757

1,606

148,376

Canada othpt

1,312,748

521,088

826,476

149,520

56,407

1,116,771

979,359

Canada othptdust







150,320

54,747





Canada ptfireothna

6,282,821

104,683

134,301

685,135

580,928

60,914

1,501,988

Canada CMV

13,802

49

121,859

2,292

2,126

5,172

6,760

Mexico othar

2,684,115

878,370

707,975

585,933

415,474

25,671

3,739,965

Mexico onroad mex

6,273,194

10,319

1,497,028

74,169

56,782

26,400

552,952

Mexico othpt

743,265

36,318

698,064

256,840

179,384

2,110,426

340,352

Mexico ptfire othna

7,133,496

120,584

346,990

1,155,522

745,819

45,208

2,259,747

Mexico CMV

64,730

0

204,997

16,286

15,087

109,778

8,817

Offshore cmv in Federal
waters

36,317

163

322,293

9,143

8,466

40,888

17,404

Offshore cmv outside
Federal waters

88,556

1,178

1,008,678

92,681

85,293

685,101

40,344

Offshore pt oilgas

50,052

15

48,691

668

667

502

48,210

Non-US Total

29,347,127

1,685,043

6,780,791

4,633,898

2,670,630

4,249,027

10,529,914

219


-------
Table 5-5. National by-sector CAP emissions summaries for the 2023fhl case, 36US3 grid (tons/yr)

Sector

CO

NH3

NOX

PM10

PM2 5

S02

voc

afdustadj







7,256,900

1,016,968





ag



3,543,158









205,451

airports

740,248

0

220,047

11,394

10,161

30,253

92,649

cmv_clc2

23,806

60

117,456

3,220

3,122

243

4,563

cmv_c3

17,126

49

107,776

2,696

2,480

5,549

10,572

nonpt

2,646,550

79,408

709,732

579,371

473,087

106,585

3,757,585

nonroad

10,584,399

2,033

737,782

71,479

66,960

1,527

857,041

npoilgas

788,072

20

585,230

16,221

16,102

31,269

3,203,738

onroad

13,777,542

89,297

1,751,649

200,035

72,495

12,486

1,099,467

ptagfire

262,645

51,276

10,240

38,688

26,951

3,694

17,181

ptegu

659,538

36,544

996

144,758

124,433

18,820

35,922

ptfire

13,717,466

239,605

227,337

1,461,693

1,234,062

111,291

3,109,465

ptnonipm

1,448,583

63,739

928,917

400,219

257,153

572,494

771,839

ptoilgas

186,242

4,377

361,166

13,602

12,973

38,125

156,725

rail

105,988

330

469,157

12,778

12,376

460

20,436

rwc

2,047,318

14,796

31,911

304,528

303,984

7,011

329,092

















36US3 U.S. Total

47,005,523

4,124,692

6,259,396

10,517,582

3,633,307

939,807

13,671,726

















beis

7,232,588



968,624







42,374,150

36US3 U.S. Total + beis

54,238,111

4,124,692

7,228,020

10,517,582

3,633,307

939,807

56,045,876

















Can./Mex./Offshore















Sector

CO

NH3

NOX

PM10

PM2 5

S02

VOC

Canada othafdust







1,222,521

214,760





Canada othar

2,896,925

5,004

351,959

316,554

239,499

20,395

875,086

Canada onroadcan

1,471,769

6,260

247,154

26,948

11,536

778

90,813

Canada othpt

1,306,333

631,845

682,142

99,818

53,521

977,647

851,263

Canada othptdust







150,273

54,730





Canada ptfireothna

6,282,821

104,683

134,301

685,165

580,958

60,914

1,501,988

Canada CMV

14,789

52

88,545

2,463

2,285

5,507

7,134

Mexico othar

2,873,134

864,397

767,216

610,423

438,710

26,588

4,050,948

Mexico onroad mex

6,053,503

12,083

1,447,199

94,407

72,468

31,838

560,284

Mexico othpt

930,547

44,909

777,407

303,309

210,038

2,111,906

427,407

Mexico ptfire othna

7,136,168

120,627

347,132

1,155,991

746,107

45,222

2,260,695

Mexico CMV

79,677

0

252,331

20,046

18,571

19,304

10,853

Offshore cmv in Federal
waters

43,338

191

280,425

10,740

9,920

50,540

20,650

Offshore cmv outside
Federal waters

108,334

741

1,234,211

58,177

53,538

155,668

49,468

Offshore pt oilgas

50,052

15

48,691

668

667

502

48,210

Non-US Total

29,247,390

1,790,809

6,658,712

4,757,504

2,707,306

3,506,810

10,754,799

220


-------
Table 5-6. National by-sector CAP emissions summaries for the 2028fhl case, 36US3 grid (tons/yr)

Sector

CO

NH3

NOX

PM10

PM2 5

S02

voc

afdustadj







7,281,296

1,021,906





ag



3,564,067









207,123

airports

804,754

0

245,466

11,900

10,649

33,910

100,417

cmv_clc2

24,241

47

93,634

2,572

2,494

245

3,602

cmv_c3

19,655

54

107,701

3,094

2,847

6,354

12,192

nonpt

2,667,254

79,670

709,358

594,080

485,244

107,185

3,801,426

nonroad

10,895,363

2,105

611,654

58,375

54,340

1,545

802,328

npoilgas

774,404

20

560,267

16,462

16,343

33,574

3,331,524

onroad

10,310,777

87,925

1,246,494

189,887

58,944

11,705

836,476

ptagfire

262,645

51,276

10,240

38,688

26,951

3,694

17,181

ptegu

648,829

35,883

748,663

140,100

120,420

781,397

33,831

ptfire

13,717,466

239,605

227,337

1,461,693

1,234,062

111,291

3,109,465

ptnonipm

1,460,908

63,990

933,863

402,498

258,991

575,210

772,998

ptoilgas

186,008

4,383

355,109

14,119

13,477

40,437

160,295

rail

110,026

343

423,103

10,953

10,611

473

17,558

rwc

2,024,434

14,615

32,058

300,440

299,891

6,789

325,463

















36US3 U.S. Total

43,906,764

4,143,984

6,304,947

10,526,157

3,617,170

1,713,809

13,531,879

















beis

7,232,588



968,624







42,374,150

36US3 U.S. Total + beis

51,139,352

4,143,984

7,273,571

10,526,157

3,617,170

1,713,809

55,906,029

















Can./Mex./Offshore















Sector

CO

NH3

NOX

PM10

PM2 5

S02

VOC

Canada othafdust







1,314,491

230,228





Canada othar

2,896,712

4,968

319,942

313,751

231,705

20,393

902,227

Canada onroadcan

1,353,512

5,692

177,653

27,234

9,960

723

63,284

Canada othpt

1,344,360

719,520

564,509

106,041

57,167

965,763

928,552

Canada othptdust







150,646

54,865





Canada ptfireothna

6,282,821

104,683

134,301

685,165

580,958

60,914

1,501,988

Canada CMV

15,570

55

95,172

2,598

2,409

5,866

7,502

Mexico othar

2,995,073

871,163

800,519

627,824

454,427

27,308

4,263,367

Mexico onroad mex

5,496,594

13,807

1,336,088

108,810

83,255

36,064

574,688

Mexico othpt

1,007,430

51,510

870,465

346,653

239,665

2,188,067

495,677

Mexico ptfire othna

7,136,168

120,627

347,132

1,155,991

746,107

45,222

2,260,695

Mexico CMV

92,295

0

292,291

23,221

21,512

22,361

12,572

Offshore cmv in Federal
waters

49,577

218

261,208

12,259

11,309

59,247

23,628

Offshore cmv outside
Federal waters

125,652

858

1,424,152

67,233

61,846

180,627

57,032

Offshore pt oilgas

50,052

15

48,691

668

667

502

48,210

Non-US Total

28,845,814

1,893,116

6,672,122

4,942,583

2,786,081

3,613,056

11,139,423

221


-------
Table 5-7. National by-sector CAP emissions summaries for the 2016fi case, 12US1 grid (tons/yr)

Sector

CO

NH3

NOX

PM10

PM2 5

S02

voc

afdust adj







7,203,692

1,006,446





ag



3,409,761









194,779

airports

486,237

0

126,713

10,011

8,733

15,245

54,191

cmv clc2

23,548

83

162,502

4,457

4,320

634

6,436

cmv c3

13,956

39

110,462

2,201

2,025

4,528

8,600

nonpt

2,629,755

78,509

710,918

570,314

463,807

138,650

3,695,093

nonroad

10,593,274

1,845

1,110,277

109,196

103,230

2,133

1,128,691

np oilgas

759,771

12

572,043

14,050

13,984

19,243

2,792,092

onroad

19,889,617

100,318

3,630,693

239,997

117,758

27,559

1,852,260

pt oilgas

167,531

4,338

339,280

11,301

10,784

33,227

127,565

ptagfire

262,645

51,276

10,240

38,688

26,951

3,694

17,181

ptegu

658,346

23,976

1,319,553

163,981

133,517

1,565,446

33,739

ptfire

13,717,466

239,605

227,337

1,461,693

1,234,062

111,291

3,109,465

ptnonipm

1,436,952

63,731

938,248

396,857

254,364

654,417

770,177

rail

104,551

326

559,381

16,344

15,819

457

26,082

rwc

2,119,402

15,439

31,282

317,469

316,943

7,703

340,941

Grand Total

52,863,051

3,989,258

9,848,929

10,560,252

3,712,741

2,584,228

14,157,289

* Only the emissions for airports, ptegu, and ptnonipm are different from 2016fh

222


-------
Table 5-8. National by-sector CAP emissions summaries for the 2021fi case, 12US1 grid (tons/yr)

Sector

CO

NH3

NOX

PM10

PM2 5

S02

VOC

afdust ad)







7,240,348

1,013,825





ag



3,505,044









202,401

airports

505,328

0

140,174

9,950

8,712

17,004

56,616

cmv clc2

23,438

65

128,204

3,482

3,375

239

5,017

cmv c3

16,709

45

104,555

2,623

2,413

5,380

10,397

nonpt

2,638,873

79,104

707,398

576,267

470,174

115,476

3,739,021

nonroad

10,518,831

1,997

829,445

80,691

75,820

1,527

909,600

np oilgas

801,948

20

597,124

16,115

15,997

31,299

3,203,182

onroad

14,816,054

87,838

2,020,269

205,721

80,499

12,675

1,202,768

pt oilgas

187,415

4,377

364,905

13,523

12,896

37,859

156,053

ptagfire

262,645

51,276

10,240

38,688

26,951

3,694

17,181

ptegu

534,284

28,546

928,956

175,815

135,329

985,418

30,198

ptfire

13,717,466

239,605

227,337

1,461,693

1,234,062

111,291

3,109,465

ptnonipm

1,444,231

63,698

923,229

398,559

255,393

583,384

769,284

rail

105,578

329

494,935

13,797

13,360

459

22,049

rwc

2,067,581

14,978

31,725

308,180

307,641

7,208

332,424

Con. U.S. Total

47,640,381

4,076,923

7,508,497

10,545,452

3,656,446

1,912,913

13,765,657

















beis

7,167,921



965,761







42,133,700

CONUS + beis

54,808,302

4,076,923

8,474,258

10,545,452

3,656,446

1,912,913

55,899,357

















Canada/Mexico/offshore (12US1)











Sector

CO

NH3

NOX

PM10

PM2 5

S02

VOC

Canada othafdust







1,149,074

202,140





Canada othar

2,700,443

4,739

333,690

306,299

233,888

19,792

825,525

Canada onroad can

1,480,052

6,252

277,315

25,688

11,766

953

101,399

Canada othpt

1,108,562

584,643

563,863

87,892

46,372

869,684

714,401

Canada othptdust







150,926

55,585





Canada ptfire othna

760,345

13,015

16,337

84,366

71,652

6,721

185,224

Canada CMV

11,383

39

74,242

1,784

1,658

3,114

5,440

Mexico othar

257,467

199,595

234,691

117,575

56,001

7,914

565,928

Mexico onroad mex

1,787,920

3,130

432,042

16,488

11,499

7,182

160,451

Mexico othpt

191,807

5,924

377,918

72,865

55,535

385,884

79,438

Mexico ptfire othna

384,764

7,466

16,665

45,198

38,354

2,798

131,980

Mexico CMV

0

0

0

0

0

0

0

CMV - Offshore ECA

37,466

142

264,343

7,978

7,375

32,504

18,196

CMV - outside ECA

26,696

259

294,251

14,617

13,452

38,728

12,664

Offshore pt oilgas

50,052

15

48,691

668

667

502

48,210

Non. U.S. Total

8,796,957

825,218

2,934,048

2,081,419

805,946

1,375,777

2,848,856

223


-------
Table 5-9. National by-sector Ozone Season NOx emissions summaries 12US1 grid (tons/o.s.)

Socio r

2(11 (.Hi

20l(.li

202 in

20231111

2!!2Xlli 1

20231111_
l'i\;iir

202X1111_
l'i\;iir

airports

82,400

56,300

62,281

97,645

108,942

64,685

68.-T

cmv clc2 12

90,624

90,624

71,370

64,719

51,424

64,719

51,424

cmv c3 12

264,816

264,816

270,721

277,635

294,186

277,635

294,186

nonpt

204,293

204,293

203,506

204,554

205,760

204,554

205,760

nonroad

566,218

566,218

424,735

377,911

312,399

377,911

312,399

np oilgas

237,354

237,354

248,007

243,092

232,869

243,092

232,869

onroad

1,436,216

1,436,216

790,537

689,145

481,066

689,145

481,066

onroad ca adj

100,197

100,197

62,845

47,973

42,323

47,973

42,323

pt oilgas

162,562

162,562

173,295

171,730

169,199

171,730

169,199

ptagfire

3,193

3,193

3,193

3,193

3,193

3,193

3,193

ptegu

590,601

605,064

409,870

366,285

358,597

366,285

358,597

ptnonipm

393,846

393,102

386,810

389,030

390,948

389,030

390,948

rail

236,771

236,771

209,477

198,559

179,051

198,559

179,051

i'\vc

2,705

2,705

2,796

2,833

2,868

2,833

2,868

loliil I .S. Amlim



4.35'UI5

3.31 y.443

3.134.303

2.X32.X2"7

3.101.343

2."792.(.XI

bds

581,479

581,479

581,479

581,479

581,479

581,479

581,479

pi fne

"5.S5I

"5.S5 1

"5.S5 1

"5.S5 1

"5.S5 1

"5.S5I

"5.S5 1

(ii'iind loliil

5.02'U 25

5.111 (.."'45

3.«J"7f»."7"'3

3/"JI. (.33

3.4')O.I5"7

3."75X.(>"73

3.450.012

* The 2023fhl_fixair and 2028fhl_fixair cases include airport emissions consistent with the corrected 2017NEI for those
years.

224


-------
Table 5-10. National by-sector Ozone Season VOC emissions summaries 12US1 grid (tons/o.s.)

Seelor

201 (.Hi

20l(.li

202 in

20231111

2028lli 1

20231111_
l'i\air

2028lli l_
I'ixair

ag

137,555

137,555

142,962

145,124

146,354

145,124

146,354

airports

38,108

24,078

25,155

41,087

44,546

25,614

26,751

cmv clc2 12

3,538

3,538

2,749

2,476

1,946

2,476

1,946

cmv c3 12

14,553

14,553

16,776

17,966

20,834

17,966

20,834

nonpt

1,550,432

1,550,432

1,568,595

1,575,983

1,594,820

1,575,983

1,594,820

nonroad

570,765

570,765

448,509

418,518

386,522

418,518

386,522

np oilgas

1,127,829

1,127,829

1,287,481

1,288,459

1,336,473

1,288,459

1,336,473

onroad

753,557

753,557

486,080

446,342

331,068

446,342

331,068

onroad ca ad)

45,633

45,633

32,977

27,926

23,048

27,926

23,048

pt oilgas

73,625

73,625

85,556

85,837

87,331

85,837

87,331

ptagfire

6,314

6,314

6,314

6,314

6,314

6,314

6,314

ptegu

16,215

16,212

14,133

16,746

16,070

16,746

16,070

ptfire

1,277,287

1,277,287

1,277,287

1,277,287

1,277,287

1,277,287

1,277,287

ptnonipm

322,200

322,189

321,771

322,833

323,270

322,833

323,270

rail

11,039

11,039

9,331

8,648

7,429

8,648

7,429

rwc

25,674

25,674

26,040

26,186

26,315

26,186

26,315

Total U.S. Anthro

5,974,324

5,960,279

5,751,716

5,707,731

5,629,630

5,692,258

5,611,834

beis

32,291,364

32,291,364

32,291,364

32,291,364

32,291,364

32,291,364

32,291,364

ptfire

1,277,287

1,277,287

1,277,287

1,277,287

1,277,287

1,277,287

1,277,287

Grand Total

39,542,975

39,528,930

39,320,367

39,276,382

39,198,280

39,260,908

39,180,485

* The 2023fhl_fixair and 2028fhl_fixair cases include airport emissions consistent with the corrected 2017NEI for those
years.

225


-------
6 References

Adelman, Z. 2012. Memorandum: Fugitive Dust Modeling for the 2008 Emissions Modeling Platform.
UNC Institute for the Environment, Chapel Hill, NC. September 28, 2012.

Adelman, Z. 2016. 2014 Emissions Modeling Platform Spatial Surrogate Documentation. UNC Institute
for the Environment, Chapel Hill, NC. October 1, 2016. Available at
ftp://newftp.epa.gov/Air/emismod/2014/v 1/spatial surrogates/

Adelman, Z., M. Omary, Q. He, J. Zhao and D. Yang, J. Boylan, 2012. "A Detailed Approach for
Improving Continuous Emissions Monitoring Data for Regulatory Air Quality Modeling."
Presented at the 2012 International Emission Inventory Conference, Tampa, Florida. Available
from http://www.epa.gOv/ttn/chief/conference/ei20/index.html#ses-5.

Appel, K.W., Napelenok, S., Hogrefe, C., Pouliot, G., Foley, K.M., Roselle, S.J., Pleim, J.E., Bash, J.,
Pye, H.O.T., Heath, N., Murphy, B., Mathur, R., 2018. Overview and evaluation of the
Community Multiscale Air Quality Model (CMAQ) modeling system version 5.2. In Mensink C.,
Kallos G. (eds), Air Pollution Modeling and its Application XXV. ITM 2016. Springer
Proceedings in Complexity. Springer, Cham. Available at https://doi.org/10.1007/978-3-319-
57645-9 11.

Bash, J.O., Baker, K.R., Beaver, M.R., Park, J.-H., Goldstein, A.H., 2016. Evaluation of improved land
use and canopy representation in BEIS with biogenic VOC measurements in California. Available
from http J/www, geosci-model-dev. net/9/2191/2016/.

BEA, 2012. "2013 Global Outlook projections prepared by the Conference Board in November 2012".
U.S. Bureau of Economic Analysis. Available from: http://www.conference-
b oard. org/data/ gl ob al outl ook. cfm.

Bullock Jr., R, and K. A. Brehme (2002) "Atmospheric mercury simulation using the CMAQ model:

formulation description and analysis of wet deposition results." Atmospheric Environment 36, pp
2135-2146. Available at https://doi.org/10.1016/S1352-2310(02)00220-0.

Coordinating Research Council (CRC). Report A-100. Improvement of Default Inputs for MOVES and
SMOKE-MOVES. Final Report. February 2017. Available at http://crcsite.wpengine.com/wp-
content/uploads/2019/05/ERG FinalReport CRCA100 28Feb2017.pdf.

Drillinginfo, Inc. 2015. "DI Desktop Database powered by HPDI." Currently available from
https://www.enverus.com/.

England, G., Watson, J., Chow, J., Zielenska, B., Chang, M., Loos, K., Hidy, G. 2007. "Dilution-Based
Emissions Sampling from Stationary Sources: Part 2— Gas-Fired Combustors Compared with
Other Fuel-Fired Systems," Journal of the Air & Waste Management Association, 57:1, 65-78,
DOI: 10.1080/10473289.2007.10465291. Available
athttps://www.tandfonline.com/doi/abs/10.1080/10473289.2007.10465291.

EPA, Light-Duty Vehicle, Light-Duty Truck, and Medium-Duty Passenger Vehicle Tier 2 Exhaust

Emission Standards. Office of Transportation and Air Quality, Ann Arbor, MI 48105. Available
at: https://www.epa.gov/emission-standards-reference-guide/epa-emission-standards-light-dutv-
vehicles-and-trucks-and .

226


-------
EPA, 2012d. Preparation of Emission Inventories for the Version 5.0, 2007 Emissions Modeling Platform
Technical Support Document. Available from:

http://epa.gov/ttn/chief/emch/2007v5/20Q7v5 2020base EmisMod TSD 13dec2012.pdf.

EPA, 2013rwc. "2011 Residential Wood Combustion Tool version 1.1, September 2013", available from
US EPA, OAQPS, EIAG.

EPA, 2015b. Draft Report Speciation Profiles and Toxic Emission Factors for Nonroad Engines. EPA-
420-R-14-028. Available at

https://cfpub.epa.gov/si/si public record Report.cfm?dirEntryId=309339&CFID=83476290&CF
TOKEN=35281617.

EPA, 2015c. Speciation of Total Organic Gas and Particulate Matter Emissions from On-road Vehicles in
MOVES2014. EPA-420-R-15-022. Available at
https://nepis. epa.gov/Exe/ZyPDF. cgi?Dockev=P 100NQJG.pdf

EPA, 2016. SPECIATE Version 4.5 Database Development Documentation, U.S. Environmental

Protection Agency, Office of Research and Development, National Risk Management Research
Laboratory, Research Triangle Park, NC 27711, EPA/600/R-16/294, September 2016. Available
at https://www.epa.gov/sites/production/files/2016-Q9/documents/speciate 4.5.pdf.

EPA, 2018. AERMOD Model Formulation and Evaluation Document. EPA-454/R-18-003. U.S.

Environmental Protection Agency, Research Triangle Park, North Carolina 27711. Available at
https://www3.epa.gov/ttn/scram/models/aermod/aermod mfed.pdf.

ERG, 2014a. Develop Mexico Future Year Emissions Final Report. Available at

ftp://ftp.epa.gov/EmisInventory/201 Iv6/v2platform/2011 emissions/Mexico Emissions WA%204
-09 final report 121814.pdf.

ERG, 2016b. "Technical Memorandum: Modeling Allocation Factors for the 2014 Oil and Gas Nonpoint
Tool." Available at ftp://newftp.epa.gov/air/emismod/2014/vl/spatial surrogates/oil and gas/.

ERG, 2017. "Technical Report: Development of Mexico Emission Inventories for the 2014 Modeling

Platform." Available at ftp://newftp.epa.gov/Air/emismod/2014/v2/2014fd/emissions/EPA%205-
18%20Report Clean%20Final 01042017.pdf.

ERG, 2018. Technical Report: "2016 Nonpoint Oil and Gas Emission Estimation Tool Version 1.0".
Available at

ftp://newftp.epa.gov/air/emismod/2016/vl/reports/2016%20Nonpoint%20Qil%20and%20Gas%20
Emission%20Estimation%20Tool%20Vl 0%20December 2018.pdf.

Frost & Sullivan, 2010. "Project: Market Research and Report on North American Residential Wood
Heaters, Fireplaces, and Hearth Heating Products Market (P.O. # PO1-IMP403-F&S). Final
Report April 26, 2010", pp. 31-32. Prepared by Frost & Sullivan, Mountain View, CA 94041.

Houck, 2011; "Dirty- vs. Clean-Burning? What percent of freestanding wood heaters in use in the U.S.
today are still old, uncertified units?" Hearth and Home, December 2011.

McCarty, J.L., Korontzi, S., Jutice, C.O., and T. Loboda. 2009. The spatial and temporal distribution of
crop residue burning in the contiguous United States. Science of the Total Environment, 407 (21):
5701-5712. Available at https://doi.Org/10.1016/i.scitotenv.2009.07.009.

MDNR, 2008; "A Minnesota 2008 Residential Fuelwood Assessment Survey of individual household
responses". Minnesota Department of Natural Resources. Available from
http://files.dnr.state.mn.us/forestry/um/residentialfuelwoodassessment07 08.pdf.

227


-------
NCAR, 2016. FIRE EMISSION FACTORS AND EMISSION INVENTORIES, FINN Data, downloaded
2014 SAPRC99 version from http://bai.acom.ucar.edu/Data/fire/.

NESCAUM, 2006; "Assessment of Outdoor Wood-fired Boilers". Northeast States for Coordinated Air
Use Management (NESCAUM) report. Available from

http://www.nescaum.org/documents/assessment-of-outdoor-wood-fired-boilers/20Q6-1031-owb-
report revised-iune2006-appendix.pdf.

NYSERDA, 2012; "Environmental, Energy Market, and Health Characterization of Wood-Fired
Hydronic Heater Technologies, Final Report". New York State Energy Research and
Development Authority (NYSERDA). Available from:
http://www.nvserda.ny.gov/Publications/Case-Studies/-

/media/Files/Publications/Research/Environmental/Wood-Fired-Hvdronic-Heater-Tech.ashx.

Pechan, 2001. E.H. Pechan & Associates, Inc., Control Measure Development Support—Analysis of
Ozone Transport Commission Model Rules, Springfield, VA, prepared for the Ozone Transport
Commission, Washington, DC, March 31, 2001. Available at

https://otcair.Org/upload/Documents/Reports/Control%20Measure%20Development%20Support.p
df.

Pouliot, G., H. Simon, P. Bhave, D. Tong, D. Mobley, T. Pace, and T. Pierce. (2010) "Assessing the
Anthropogenic Fugitive Dust Emission Inventory and Temporal Allocation Using an Updated
Speciation of Particulate Matter." International Emission Inventory Conference, San Antonio, TX.
Available at http://www3.epa.gov/ttn/chief/conference/eil9/session9/pouliot pres.pdf.

Pouliot, G. and J. Bash, 2015. Updates to Version 3.61 of the Biogenic Emission Inventory System
(BEIS). Presented at Air and Waste Management Association conference, Raleigh, NC, 2015.

Pouliot G, Rao V, McCarty JL, Soja A. Development of the crop residue and rangeland burning in the

2014 National Emissions Inventory using information from multiple sources. Journal of the Air &
Waste Management Association. 2017 Apr 27;67(5):613-22.

Raffuse, S., D. Sullivan, L. Chinkin, S. Larkin, R. Solomon, A. Soja, 2007. Integration of Satellite-
Detected and Incident Command Reported Wildfire Information into BlueSky, June 27, 2007.
Available at: http://getblueskv.org/smartfire/docs.cfm.

Reichle, L.,R. Cook, C. Yanca, D. Sonntag, 2015. "Development of organic gas exhaust speciation

profiles for nonroad spark-ignition and compression-ignition engines and equipment", Journal of
the Air & Waste Management Association, 65:10, 1185-1193, DOI:

10.1080/10962247.2015.1020118. Available at https://doi.org/10.1080/10962247.2015.102Q118.

Reff, A., Bhave, P., Simon, H., Pace, T., Pouliot, G., Mobley, J., Houyoux. M. "Emissions Inventory of
PM2.5 Trace Elements across the United States", Environmental Science & Technology 2009 43
(15), 5790-5796, DOI: 10.1021/es802930x. Available at https://doi.org/10.1021/es802930x.

Sarwar, G., S. Roselle, R. Mathur, W. Apel, R. Dennis, "A Comparison of CMAQ HONO predictions
with observations from the Northeast Oxidant and Particle Study", Atmospheric Environment 42
(2008) 5760-5770). Available at https://doi.Org/10.1016/i.atmosenv.2007.12.065.

Schauer, J., G. Lough, M. Shafer, W. Christensen, M. Arndt, J. DeMinter, J. Park, "Characterization of

Metals Emitted from Motor Vehicles," Health Effects Institute, Research Report 133, March 2006.

228


-------
Available at https://www.healtheffects.org/publication/characterization-metals-emitted-motor-
vehicles.

Skamarock, W., J. Klemp, J. Dudhia, D. Gill, D. Barker, M. Duda, X. Huang, W. Wang, J. Powers, 2008.
A Description of the Advanced Research WRF Version 3. NCAR Technical Note. National
Center for Atmospheric Research, Mesoscale and Microscale Meteorology Division, Boulder, CO.
June 2008. Available at: http://www2.mmm.ucar.edu/wrf/users/docs/arw v3 bw.pdf.

Sullivan D.C., Raffuse S.M., Pryden D.A., Craig K.J., Reid S.B., Wheeler N.J.M., Chinkin L.R., Larkin
N.K., Solomon R., and Strand T. (2008) Development and applications of systems for modeling
emissions and smoke from fires: the BlueSky smoke modeling framework and SMARTFIRE: 17th
International Emissions Inventory Conference, Portland, OR, June 2-5. Available at:
http ://www. epa. gov/ttn/chief/conferences, html.

Wang, Y., P. Hopke, O. V. Rattigan, X. Xia, D. C. Chalupa, M. J. Utell. (2011) "Characterization of

Residential Wood Combustion Particles Using the Two-Wavelength Aethalometer", Environ. Sci.
Technol., 45 (17), pp 7387-7393. Available at https://doi.org/10.1021/es2013984.

Wiedinmyer, C., S.K. Akagi, R.J. Yokelson, L.K. Emmons, J.A. Al-Saadi3, J. J. Orlando1, and A. J. Soja.
(2011) "The Fire INventory from NCAR (FINN): a high resolution global model to estimate the
emissions from open burning", Geosci. Model Dev., 4, 625-641. http://www.geosci-model-
dev.net/4/625/2011/ doi: 10.5194/gmd-4-625-2011.

Yarwood, G., J. Jung,, G. Whitten, G. Heo, J. Mellberg, and M. Estes,2010: Updates to the Carbon Bond
Chemical Mechanism for Version 6 (CB6). Presented at the 9th Annual CMAS Conference,

Chapel Hill, NC. Available at

https://www.cmascenter.org/conference/2010/abstracts/emery updates carbon 2010.pdf.

Zhu, Henze, et al, 2013. "Constraining U.S. Ammonia Emissions using TES Remote Sensing
Observations and the GEOS-Chem adjoint model", Journal of Geophysical Research:
Atmospheres, 118: 1-14. Available at https://doi.org/10.1002/igrd.50166.

229


-------
Appendix A: CB6 Assignment for New Species

September 27, 2016

MEMORANDUM

To: Alison Eyth and Madeleine Strum, QftQPS, EPA
From: Ross Beardsiey and Greg Yarwood, Rambofl Environ

Species r^sp^ngs fo» Csb ana CaGa fa,- use v».th SFfOATE 4.5

Summarf

Ramholl Environ fREf reviewed version 4,5 of the SPEClfeTE database,, and created C6Q5 an; C5i
mechanism species mappings fat newly added GDir.pc_r.ds. In addition, the map: mi guids Sines for
Carbon Bond {CB} mechanisms were expanded to promote consistency in euroanc fu: - 'e • v ori.

Background

The :n.- r: rarer.: a Pr:tec:.o • Age—y 5 s?ECU~£	gas r: ; 3 tk.'cte matter

•: = :i;:;,on woffles of air pollution sources, which are used in the generation of emissions data for air
quality nodels{AQMjisucr == cvaq, ;*tt?://www.amascenter.arg/cinaq/]| and GAMx
! f«Kp://'A"Aw.cam*.co!n:|. However, the condensed dterrical ¦nachenisms aiitf wthi" these
phctDchs "ileal models utilize fewer species than SPfClATE tc rep'esen* gas ;h=== chert =:ry, a-d^ ami CBS

.'t:::.- ¦ =q'c.:= =r ^axai edv: -o.=:t ~f:-~riZ_1Z- 'I I Oil. il-ai2%20FinafK2DR£port.pdf).

Methods
CB Mode Species

Organic g£s== are "lapisr :: ?-= C5 rrechan -r eiT" =r as explicit v •ep.-E = E"ted individual
compounds ;=.g. ,m_D2 for scera-dehyie;, or as a comb reticr cf r ode' ipecias that represent
:ommo" structural graucs (E.g. flLDX *"cr other aldehydes, par for a kyl groupsi "able- I lists all of
the exploit ard structura mode spec as in CEC5 ard CB6 nachanisms, each cf '.v^ ch -ep^se-ts 3
defiiec r.^ce- ~r csr:;n s-.z-m si :-.vi *% r:r :=-bc - :: be :onse~ved n 3 csie:. Cif cctahs f:-u"
rooneeKp' :i: r?ds ipse =it~-b-> CEC5 =«-d a~ =r Di- D*a str. grru; tc -ep-a-s'-t ¦-:e:c"-n. T- =
€B05 rec-esentafon of tr,e five add tional CB6 spsries "s provided in t-a *."nc.'vifstf in C6C5 column of
Table 1.

%r*£«ti Ewi,ir«\ 773 San f/li-rin Oriix, 5afte IMS, fiowtlro, CA34996
VTl«l3,SSS>B7lia F+l 415.2890707

230


-------
ENVIRON

In addition to the explicit and structural species, there are two model species that are used to

represent organic gases that are not treated by the CB mechanism:

NVOL- Very low volatility SPEC!ATE compounds that reside predominantly in the particle phase and
should be excluded from the gas phase mechanism. These compounds are mapped by setting
NVOL equal to the melecula' weight (e.g. decsbromodiphenyl oxide is mapped as 9E-9.2
hJVOL|, v/ftich allov.sforthe total mass of all NVOL to be determined.

UNK-Compounds that are unable tc be mapped to CB using the ava labia model species. This

approach should be avoided unless absolutely necessary, and will lead to a warning message
in the speciation tool.

Table l. Model species in the CBQ5 and CBS chemical mechanisms.	







Included in



Vncel



Plumber

CB05



S pee es



~1

(structural

Included

llsire

Desertion

Cartons

mapping)

in CBS

Esplict model species

ACET

f> cetane (proparone |

1

Mo (a PAP|

res

ald:

Acet: Idehyde |ettrsnalj

2_

rej

res

BENZ

Serpens

6

Ho(l PAR.3
UNR|

res

CH4

Methane

1

res

res

ETH

Etl-.ene (ethylene]

2

res

res

ET HA

Ethane

2

res

res

ETHV

Efl-,»ne |acetylene)

2

Mo (1 PAR... 1
JHR|

res

E70H

Ettisnei

2

res

res

=ORM

=ormalctehvde imetrianalj'

1

res

res

SOP

.saprene [l-rnettiyl-i^a-tiutBdene)

3

res

res

MEOH

Methanol

1

res

res

PRFA

'rapane

3

Wo (1.3 PA 3,
1.5 UNRi

res

Commofi Structural groups

ALDX

- igtier: dehyde ^raup [-C-tHO'

2

res

res

OLE

ntenal cletin grcup |3;R..>C=C<3 RHJ

4

res

res

C=C|

2

res

res

aAR

^araffinic group (R <
-------
ENVIRON

Mapping guidelines for nan-explicit organicgases using CB model species

5PECIATE compounds that are not treated explicitly ore mapped to CB mcdel species that represent
common structural groups. Table 2 lists the cart>on number and general mapping guidelines for each
of the structure model species.

Table2. General Guidelines for mapping using CB6 structural model species.

CB4

Species

Name

dumber ~!
Cirtofis

Represents

ALOX

2

Aldehyde group ALOX represents 2 carbcns and additicnal carbons are represented as
alkyl groups fnncsfiy PAR), e.g. prapionaldehv:ie is AL3X + PAR

OLE

4

-itemal ciefin grcup. IOLE represents 4 cartons and additicnal mrtons are represented as
alkyl groups [mastsy PAR|, e.g. 2-pentene isomers are I0LE+ PAR.
trcepCwni:

• OLE with 2 carton branches cr. both sides of the double bend are downgraded to

OLE

SET

1


-------
Tac-lr 2. Mapp nf gu :'e: n==fo'sc-iE r!ffi:l it :o ma: cc"" pound :i-3=£s= = _d =ti.c:.o grcups

:c "i •'
I!s: £
: :

¦:2 ¦ : Jic ?; rt: -t:s -i?i : -

lilDTobenienes and
other halogensteif
Herat-f:

Stiifcline:

« 3 or less hslogsns - J =>A^ 5 'IS

*	- :-r -1:-; nalog:-.: -6 5
Star-:!?:

*	. i S-lf i-rnfeen::te -

*	~:t~::l-l:-Dber\:er:s - 6 J'.®



Suiitefine:

•	1 '.-OLE with adriitiomri cartons represented as aflcyl |rn«ps (generally

Shf::;:

¦ Mettiytcytfopentsdisie- . : .= : 3AS

*	1 0-E 3 °«

fumafPfireHB

Suirt:

¦	2 OLE with additional carteru represertfc k s »,l 5r31.p1 (generally

:=4J)

Btamplei:

*	1-oj*

•	i-ftsfityaiirin —: :-.e : :ar

¦	: :

¦ l'-fi1eth-|fp|TlBlE - 2 OLE, 1 PAR

-tte'3c,cl: areTiitie
:ciF:jr3:

:c -its lie 2 r:n-
:»rt:n atari:

Siiiaains:

¦	i OLE with mnir> i; zartmris, repress ned ei iltfl pgup (generally

~ ^yrezine —1 OLE - :--

¦	l-meHiylpyrazoie-1 ;.E : r-R

« C5-3*iiethfIoiniQls - i C.£ S *«



SiiitJeiine:

"• : ?::-3:s ;7*vi; ); "... ™!:m

fu«tlion»lgrayp. f 1 campe-uni contiira mors tlsir cie tr all fccnrt
¦nd no alterreactive fund-ions! groups, ihencne C fie tp Pie tenrts
ia treated as OLE with additional carbons trials a a: s'lrvl 5*: up s.
£*artples:

¦	"i - i -I .1 I

¦	. :1 par

¦	i-S—ec:BB if! - „ O.E. 5

These guidelines weee used to mao tne new species from SPEICATC4.5, and also to revise same
crs-.ic.s ,¦ '•aiD'sd ::nrpc.rd; ever5 , 3ts:s c- :~E rev.* ssross*x*i 5=-=«::iTEv^ 5 •. •„-«-= rsoped

ami 7 previously mapped species were revised based on the new guidelines.

Ilii-fceli Eii'kircci US Capaattm, 7/S a»n -Marin Ilfwe, S«te 2il;3. Hnvoto, IA1MB8B
V -4 415,1155,0700 »+l4SESSU?W

233


-------
Recommiendation

Complete a systematic review of the mapping of all species to ensure conformity with current
-¦¦srrra ^ " = = ¦ ~~= =:= | ts i* e- =TPj -:rrpc_*"ds f=t ar= s-rri ; •:: new =redes '-ere
••ev ev>ed 3*: -=viisc tc r-rc'ite :?nsi=t=rcv i* '"•szprg azsrca-:*-*-. ru* the	o*

existing species mappings- were net: reviewed as ft was outside the scope of this work.

De -:c :p e. 'lef.octicgv T3' ziassryhi e :¦=:)• '•= l3f.ge.' ? \zii : cc-z-o.- -esed :n trei ¦
vc =ti t-. i-sf' intsfned sti. :r o-..= -:c a:;! tv. tc nrpT-.-E jwpzrrrf** •siots-*. organ : eer-sol
;5CA. irccs ;Fig f«= t"= .-c =:i tr'b=sis =et i.\ 55; SOA nc:=:. v.-h'rK i avs^a? = r bcr-- C'.'AQ

ami crntx.. A preliminary* investigation of the possibility of doing-so has been performed, and is
discussed in a separate memorandum.

ills,

234


-------
Appendix B: Profiles (other than onroad) that are new or revised in SPECIATE4.5 that were used

in the 2016 alpha platform





Profile



SPECIATE

comment

Sector

Pollutant

code

Profile description

version











5.0 (not

Replacement for v4.5









yet

profile 95223; Used 70%









released)

methane, 20% ethane,
and the 10% remaining







Poultry Production - Average of Production



VOC is from profile

nonpt

voc

G95223TOG

Cycle with gapfilled methane and ethane



95223









5.0 (not

Replacement for v4.5









yet

profile 95240. Used 70%









released)

methane, 20% ethane;

Nonpt,





Beef Cattle Farm and Animal Waste with



the 10% remaining VOC

ptnonipm

voc

G95240TOG

gapfilled methane and ethane



is from profile 95240.









5.0 (not

Replacement for v4.5









yet

profile 95241. Used 70%









released)

methane, 20% ethane;
the 10% remaining VOC

nonpt

voc

G95241TOG

Swine Farm and Animal Waste



is from profile 95241

nonpt,







5.0 (not

Composite of AE6-ready

ptnonipm,







yet

versions of SPECIATE4.5

pt_oilgas,





Composite -Refinery Fuel Gas and Natural

released)

profies 95125, 95126,

ptegu

PM2.5

95475

Gas Combustion



and 95127







Spark-Ignition Exhaust Emissions from 2-

4.5









stroke off-road engines - E10 ethanol





nonroad

VOC

95328

gasoline











Spark-Ignition Exhaust Emissions from 4-

4.5









stroke off-road engines - E10 ethanol





nonroad

VOC

95330

gasoline











Diesel Exhaust Emissions from Pre-Tier 1

4.5



nonroad

VOC

95331

Off-road Engines











Diesel Exhaust Emissions from Tier 1 Off-

4.5



nonroad

VOC

95332

road Engines











Diesel Exhaust Emissions from Tier 2 Off-

4.5



nonroad

VOC

95333

road Engines











Oil and Gas - Composite - Oil Field - Oil

4.5



nP_oilgas

VOC

95087a

Tank Battery Vent Gas











Oil and Gas - Composite - Oil Field -

4.5



nP_oilgas

voc

95109a

Condensate Tank Battery Vent Gas











Composite Profile - Oil and Natural Gas

4.5



nP_oilgas

voc

95398

Production - Condensate Tanks





nP_oilgas

voc

95403

Composite Profile - Gas Wells

4.5









Oil and Gas Production - Composite Profile

4.5



np_oilgas

voc

95417

- Untreated Natural Gas, Uinta Basin











Oil and Gas Production - Composite Profile

4.5



np_oilgas

voc

95418

- Condensate Tank Vent Gas, Uinta Basin











Oil and Gas Production - Composite Profile

4.5



np_oilgas

voc

95419

- Oil Tank Vent Gas, Uinta Basin











Oil and Gas Production - Composite Profile

4.5



np_oilgas

voc

95420

- Glycol Dehydrator, Uinta Basin





235


-------






Oil and Gas -Denver-Julesburg Basin

4.5









Produced Gas Composition from Non-CBM





nP_oilgas

VOC

DJVNT R

Gas Wells





nP_oilgas

VOC

FLR99

Natural Gas Flare Profile with DRE >98%

4.5









Oil and Gas -Piceance Basin Produced Gas

4.5



np_oilgas

VOC

PNC01 R

Composition from Non-CBM Gas Wells











Oil and Gas -Piceance Basin Produced Gas

4.5



np_oilgas

VOC

PNC02 R

Composition from Oil Wells











Oil and Gas -Piceance Basin Flash Gas

4.5



nP_oilgas

VOC

PNC03 R

Composition for Condensate Tank











Oil and Gas Production - Composite Profile

4.5



np_oilgas

VOC

PNCDH

- Glycol Dehydrator, Piceance Basin











Oil and Gas -Powder River Basin Produced

4.5



np_oilgas

VOC

PRBCB R

Gas Composition from CBM Wells











Oil and Gas -Powder River Basin Produced

4.5



np_oilgas

VOC

PRBCO R

Gas Composition from Non-CBM Wells











Oil and Gas -Permian Basin Produced Gas

4.5



np_oilgas

VOC

PRM01 R

Composition for Non-CBM Wells











Oil and Gas -South San Juan Basin

4.5









Produced Gas Composition from CBM





nP_oilgas

VOC

SSJCB R

Wells











Oil and Gas -South San Juan Basin

4.5









Produced Gas Composition from Non-CBM





np_oilgas

VOC

SSJCO R

Gas Wells











Oil and Gas -SW Wyoming Basin Flash Gas

4.5



np_oilgas

VOC

SWFLA R

Composition for Condensate Tanks











Oil and Gas -SW Wyoming Basin Produced

4.5



np_oilgas

VOC

SWVNT R

Gas Composition from Non-CBM Wells











Oil and Gas -Uinta Basin Produced Gas

4.5



np_oilgas

VOC

UNT01 R

Composition from CBM Wells











Oil and Gas -Wind River Basin Produced

4.5



np_oilgas

VOC

WRBCO R

Gas Composition from Non-CBM Gas Wells











Chemical Manufacturing Industrywide

4.5



pt_oilgas

VOC

95325

Composite





pt_oilgas

VOC

95326

Pulp and Paper Industry Wide Composite

4.5



pt_oilgas,







4.5



ptnonipm

VOC

95399

Composite Profile - Oil Field - Wells





pt_oilgas

VOC

95403

Composite Profile - Gas Wells

4.5









Oil and Gas Production - Composite Profile

4.5



pt_oilgas

VOC

95417

- Untreated Natural Gas, Uinta Basin











Oil and Gas -Denver-Julesburg Basin

4.5









Produced Gas Composition from Non-CBM





pt_oilgas

VOC

DJVNT R

Gas Wells





pt_oilgas,







4.5



ptnonipm

VOC

FLR99

Natural Gas Flare Profile with DRE >98%











Oil and Gas -Piceance Basin Produced Gas

4.5



pt_oilgas

VOC

PNC01 R

Composition from Non-CBM Gas Wells











Oil and Gas -Piceance Basin Produced Gas

4.5



pt_oilgas

VOC

PNC02 R

Composition from Oil Wells











Oil and Gas Production - Composite Profile

4.5



pt_oilgas

VOC

PNCDH

- Glycol Dehydrator, Piceance Basin





pt_oilgas,





Oil and Gas -Powder River Basin Produced

4.5



ptnonipm

VOC

PRBCO_R

Gas Composition from Non-CBM Wells





236


-------
pt_oilgas,





Oil and Gas -Permian Basin Produced Gas

4.5



ptnoniom

VOC

PRM01 R

Composition for Non-CBM Wells











Oil and Gas -South San Juan Basin

4.5



pt_oilgas,





Produced Gas Composition from Non-CBM





ptnonipm

VOC

SSJCO R

Gas Wells





pt_oilgas,





Oil and Gas -SW Wyoming Basin Produced

4.5



ptnonipm

VOC

SWVNT R

Gas Composition from Non-CBM Wells











Composite Profile - Prescribed fire

4.5



ptfire

VOC

95421

southeast conifer forest











Composite Profile - Prescribed fire

4.5



ptfire

VOC

95422

southwest conifer forest











Composite Profile - Prescribed fire

4.5



ptfire

VOC

95423

northwest conifer forest











Composite Profile - Wildfire northwest

4.5



ptfire

VOC

95424

conifer forest





ptfire

VOC

95425

Composite Profile - Wildfire boreal forest

4.5









Chemical Manufacturing Industrywide

4.5



ptnonipm

VOC

95325

Composite





ptnonipm

VOC

95326

Pulp and Paper Industry Wide Composite

4.5



onroad

PM2.5

95462

Composite - Brake Wear

4.5

Used in SMOKE-MOVES

onroad

PM2.5

95460

Composite - Tire Dust

4.5

Used in SMOKE-MOVES

237


-------
Appendix C: Mapping of Fuel Distribution SCCs to BTP, BPS and RBT

The table below provides a crosswalk between fuel distribution SCCs and classification type for portable
fuel containers (PFC), fuel distribution operations associated with the bulk-plant-to-pump (BTP), refinery
to bulk terminal (RBT) and bulk plant storage (BPS).

see

Typ
e

Description

40301001

RBT

Petroleum and Solvent Evaporation; Petroleum Product Storage at Refineries; Fixed Roof Tanks
(Varying Sizes); Gasoline RVP 13: Breathing Loss (67000 Bbl. Tank Size)

40301002

RBT

Petroleum and Solvent Evaporation; Petroleum Product Storage at Refineries; Fixed Roof Tanks
(Varying Sizes); Gasoline RVP 10: Breathing Loss (67000 Bbl. Tank Size)

40301003

RBT

Petroleum and Solvent Evaporation; Petroleum Product Storage at Refineries; Fixed Roof Tanks
(Varying Sizes); Gasoline RVP 7: Breathing Loss (67000 Bbl. Tank Size)

40301004

RBT

Petroleum and Solvent Evaporation; Petroleum Product Storage at Refineries; Fixed Roof Tanks
(Varying Sizes); Gasoline RVP 13: Breathing Loss (250000 Bbl. Tank Size)

40301006

RBT

Petroleum and Solvent Evaporation; Petroleum Product Storage at Refineries; Fixed Roof Tanks
(Varying Sizes); Gasoline RVP 7: Breathing Loss (250000 Bbl. Tank Size)

40301007

RBT

Petroleum and Solvent Evaporation; Petroleum Product Storage at Refineries; Fixed Roof Tanks
(Varying Sizes); Gasoline RVP 13: Working Loss (Tank Diameter Independent)

40301101

RBT

Petroleum and Solvent Evaporation; Petroleum Product Storage at Refineries; Floating Roof Tanks
(Varying Sizes); Gasoline RVP 13: Standing Loss (67000 Bbl. Tank Size)

40301102

RBT

Petroleum and Solvent Evaporation; Petroleum Product Storage at Refineries; Floating Roof Tanks
(Varying Sizes); Gasoline RVP 10: Standing Loss (67000 Bbl. Tank Size)

40301103

RBT

Petroleum and Solvent Evaporation; Petroleum Product Storage at Refineries; Floating Roof Tanks
(Varying Sizes); Gasoline RVP 7: Standing Loss (67000 Bbl. Tank Size)

40301105

RBT

Petroleum and Solvent Evaporation; Petroleum Product Storage at Refineries; Floating Roof Tanks
(Varying Sizes); Gasoline RVP 10: Standing Loss (250000 Bbl. Tank Size)

40301151

RBT

Petroleum and Solvent Evaporation; Petroleum Product Storage at Refineries; Floating Roof Tanks
(Varying Sizes); Gasoline: Standing Loss - Internal

40301202

RBT

Petroleum and Solvent Evaporation; Petroleum Product Storage at Refineries; Variable Vapor
Space; Gasoline RVP 10: Filling Loss

40301203

RBT

Petroleum and Solvent Evaporation; Petroleum Product Storage at Refineries; Variable Vapor
Space; Gasoline RVP 7: Filling Loss

40400101

RBT

Petroleum and Solvent Evaporation; Petroleum Liquids Storage (non-Refinery); Bulk Terminals;
Gasoline RVP 13: Breathing Loss (67000 Bbl Capacity) - Fixed Roof Tank

40400102

RBT

Petroleum and Solvent Evaporation; Petroleum Liquids Storage (non-Refinery); Bulk Terminals;
Gasoline RVP 10: Breathing Loss (67000 Bbl Capacity) - Fixed Roof Tank

40400103

RBT

Petroleum and Solvent Evaporation; Petroleum Liquids Storage (non-Refinery); Bulk Terminals;
Gasoline RVP 7: Breathing Loss (67000 Bbl. Capacity) - Fixed Roof Tank

40400104

RBT

Petroleum and Solvent Evaporation; Petroleum Liquids Storage (non-Refinery); Bulk Terminals;
Gasoline RVP 13: Breathing Loss (250000 Bbl Capacity)-Fixed Roof Tank

40400105

RBT

Petroleum and Solvent Evaporation; Petroleum Liquids Storage (non-Refinery); Bulk Terminals;
Gasoline RVP 10: Breathing Loss (250000 Bbl Capacity)-Fixed Roof Tank

40400106

RBT

Petroleum and Solvent Evaporation; Petroleum Liquids Storage (non-Refinery); Bulk Terminals;
Gasoline RVP 7: Breathing Loss (250000 Bbl Capacity) - Fixed Roof Tank

40400107

RBT

Petroleum and Solvent Evaporation; Petroleum Liquids Storage (non-Refinery); Bulk Terminals;
Gasoline RVP 13: Working Loss (Diam. Independent) - Fixed Roof Tank

40400108

RBT

Petroleum and Solvent Evaporation; Petroleum Liquids Storage (non-Refinery); Bulk Terminals;
Gasoline RVP 10: Working Loss (Diameter Independent) - Fixed Roof Tank

40400109

RBT

Petroleum and Solvent Evaporation; Petroleum Liquids Storage (non-Refinery); Bulk Terminals;
Gasoline RVP 7: Working Loss (Diameter Independent) - Fixed Roof Tank

40400110

RBT

Petroleum and Solvent Evaporation; Petroleum Liquids Storage (non-Refinery); Bulk Terminals;
Gasoline RVP 13: Standing Loss (67000 Bbl Capacity)-Floating Roof Tank

40400111

RBT

Petroleum and Solvent Evaporation; Petroleum Liquids Storage (non-Refinery); Bulk Terminals;
Gasoline RVP 10: Standing Loss (67000 Bbl Capacity)-Floating Roof Tank

238


-------
see

Typ
e

Description

40400112

RBT

Petroleum and Solvent Evaporation; Petroleum Liquids Storage (non-Refinery); Bulk Terminals;
Gasoline RVP 7: Standing Loss (67000 Bbl Capacity)- Floating Roof Tank

40400113

RBT

Petroleum and Solvent Evaporation; Petroleum Liquids Storage (non-Refinery); Bulk Terminals;
Gasoline RVP 13: Standing Loss (250000 Bbl Cap.) - Floating Roof Tank

40400114

RBT

Petroleum and Solvent Evaporation; Petroleum Liquids Storage (non-Refinery); Bulk Terminals;
Gasoline RVP 10: Standing Loss (250000 Bbl Cap.) - Floating Roof Tank

40400115

RBT

Petroleum and Solvent Evaporation; Petroleum Liquids Storage (non-Refinery); Bulk Terminals;
Gasoline RVP 7: Standing Loss (250000 Bbl Cap.) - Floating Roof Tank

40400116

RBT

Petroleum and Solvent Evaporation; Petroleum Liquids Storage (non-Refinery); Bulk Terminals;
Gasoline RVP 13/10/7: Withdrawal Loss (67000 Bbl Cap.) - Float RfTnk

40400117

RBT

Petroleum and Solvent Evaporation; Petroleum Liquids Storage (non-Refinery); Bulk Terminals;
Gasoline RVP 13/10/7: Withdrawal Loss (250000 Bbl Cap.) - Float RfTnk

40400118

RBT

Petroleum and Solvent Evaporation; Petroleum Liquids Storage (non-Refinery); Bulk Terminals;
Gasoline RVP 13: Filling Loss (10500 Bbl Cap.) - Variable Vapor Space

40400119

RBT

Petroleum and Solvent Evaporation; Petroleum Liquids Storage (non-Refinery); Bulk Terminals;
Gasoline RVP 10: Filling Loss (10500 Bbl Cap.) - Variable Vapor Space

40400120

RBT

Petroleum and Solvent Evaporation; Petroleum Liquids Storage (non-Refinery); Bulk Terminals;
Gasoline RVP 7: Filling Loss (10500 Bbl Cap.) - Variable Vapor Space

40400130

RBT

Petroleum and Solvent Evaporation; Petroleum Liquids Storage (non-Refinery); Bulk Terminals;
Specify Liquid: Standing Loss - External Floating Roof w/ Primary Seal

40400131

RBT

Petroleum and Solvent Evaporation; Petroleum Liquids Storage (non-Refinery); Bulk Terminals;
Gasoline RVP 13: Standing Loss - Ext. Floating Roof w/ Primary Seal

40400132

RBT

Petroleum and Solvent Evaporation; Petroleum Liquids Storage (non-Refinery); Bulk Terminals;
Gasoline RVP 10: Standing Loss - Ext. Floating Roof w/ Primary Seal

40400133

RBT

Petroleum and Solvent Evaporation; Petroleum Liquids Storage (non-Refinery); Bulk Terminals;
Gasoline RVP 7: Standing Loss - External Floating Roof w/ Primary Seal

40400140

RBT

Petroleum and Solvent Evaporation; Petroleum Liquids Storage (non-Refinery); Bulk Terminals;
Specify Liquid: Standing Loss - Ext. Float Roof Tank w/ Secondy Seal

40400141

RBT

Petroleum and Solvent Evaporation; Petroleum Liquids Storage (non-Refinery); Bulk Terminals;
Gasoline RVP 13: Standing Loss - Ext. Floating Roof w/ Secondary Seal

40400142

RBT

Petroleum and Solvent Evaporation; Petroleum Liquids Storage (non-Refinery); Bulk Terminals;
Gasoline RVP 10: Standing Loss - Ext. Floating Roof w/ Secondary Seal

40400143

RBT

Petroleum and Solvent Evaporation; Petroleum Liquids Storage (non-Refinery); Bulk Terminals;
Gasoline RVP 7: Standing Loss - Ext. Floating Roof w/ Secondary Seal

40400148

RBT

Petroleum and Solvent Evaporation; Petroleum Liquids Storage (non-Refinery); Bulk Terminals;
Gasoline RVP 13/10/7: Withdrawal Loss - Ext. Float Roof (Pri/Sec Seal)

40400149

RBT

Petroleum and Solvent Evaporation; Petroleum Liquids Storage (non-Refinery); Bulk Terminals;
Specify Liquid: External Floating Roof (Primary/Secondary Seal)

40400150

RBT

Petroleum and Solvent Evaporation; Petroleum Liquids Storage (non-Refinery); Bulk Terminals;
Miscellaneous Losses/Leaks: Loading Racks

40400151

RBT

Petroleum and Solvent Evaporation; Petroleum Liquids Storage (non-Refinery); Bulk Terminals;
Valves, Flanges, and Pumps

40400152

RBT

Petroleum and Solvent Evaporation; Petroleum Liquids Storage (non-Refinery); Bulk Terminals;
Vapor Collection Losses

40400153

RBT

Petroleum and Solvent Evaporation; Petroleum Liquids Storage (non-Refinery); Bulk Terminals;
Vapor Control Unit Losses

40400160

RBT

Petroleum and Solvent Evaporation; Petroleum Liquids Storage (non-Refinery); Bulk Terminals;
Specify Liquid: Standing Loss - Internal Floating Roof w/ Primary Seal

40400161

RBT

Petroleum and Solvent Evaporation; Petroleum Liquids Storage (non-Refinery); Bulk Terminals;
Gasoline RVP 13: Standing Loss - Int. Floating Roof w/ Primary Seal

40400162

RBT

Petroleum and Solvent Evaporation; Petroleum Liquids Storage (non-Refinery); Bulk Terminals;
Gasoline RVP 10: Standing Loss - Int. Floating Roof w/ Primary Seal

40400163

RBT

Petroleum and Solvent Evaporation; Petroleum Liquids Storage (non-Refinery); Bulk Terminals;
Gasoline RVP 7: Standing Loss - Internal Floating Roof w/ Primary Seal

239


-------
see

Typ
e

Description

40400170

RBT

Petroleum and Solvent Evaporation; Petroleum Liquids Storage (non-Refinery); Bulk Terminals;
Specify Liquid: Standing Loss - Int. Floating Roof w/ Secondary Seal

40400171

RBT

Petroleum and Solvent Evaporation; Petroleum Liquids Storage (non-Refinery); Bulk Terminals;
Gasoline RVP 13: Standing Loss - Int. Floating Roof w/ Secondary Seal

40400172

RBT

Petroleum and Solvent Evaporation; Petroleum Liquids Storage (non-Refinery); Bulk Terminals;
Gasoline RVP 10: Standing Loss - Int. Floating Roof w/ Secondary Seal

40400173

RBT

Petroleum and Solvent Evaporation; Petroleum Liquids Storage (non-Refinery); Bulk Terminals;
Gasoline RVP 7: Standing Loss - Int. Floating Roof w/ Secondary Seal

40400178

RBT

Petroleum and Solvent Evaporation; Petroleum Liquids Storage (non-Refinery); Bulk Terminals;
Gasoline RVP 13/10/7: Withdrawal Loss - Int. Float Roof (Pri/Sec Seal)

40400179

RBT

Petroleum and Solvent Evaporation; Petroleum Liquids Storage (non-Refinery); Bulk Terminals;
Specify Liquid: Internal Floating Roof (Primary/Secondary Seal)

40400199

RBT

Petroleum and Solvent Evaporation; Petroleum Liquids Storage (non-Refinery); Bulk Terminals;

40400201

BTP
/BPS

Petroleum and Solvent Evaporation; Petroleum Liquids Storage (non-Refinery); Bulk Plants;
Gasoline RVP 13: Breathing Loss (67000 Bbl Capacity) - Fixed Roof Tank

40400202

BTP
/BPS

Petroleum and Solvent Evaporation; Petroleum Liquids Storage (non-Refinery); Bulk Plants;
Gasoline RVP 10: Breathing Loss (67000 Bbl Capacity) - Fixed Roof Tank

40400203

BTP
/BPS

Petroleum and Solvent Evaporation; Petroleum Liquids Storage (non-Refinery); Bulk Plants;
Gasoline RVP 7: Breathing Loss (67000 Bbl. Capacity) - Fixed Roof Tank

40400204

BTP
/BPS

Petroleum and Solvent Evaporation; Petroleum Liquids Storage (non-Refinery); Bulk Plants;
Gasoline RVP 13: Working Loss (67000 Bbl. Capacity) - Fixed Roof Tank

40400205

BTP
/BPS

Petroleum and Solvent Evaporation; Petroleum Liquids Storage (non-Refinery); Bulk Plants;
Gasoline RVP 10: Working Loss (67000 Bbl. Capacity) - Fixed Roof Tank

40400206

BTP
/BPS

Petroleum and Solvent Evaporation; Petroleum Liquids Storage (non-Refinery); Bulk Plants;
Gasoline RVP 7: Working Loss (67000 Bbl. Capacity) - Fixed Roof Tank

40400207

BTP
/BPS

Petroleum and Solvent Evaporation; Petroleum Liquids Storage (non-Refinery); Bulk Plants;
Gasoline RVP 13: Standing Loss (67000 Bbl Cap.) - Floating Roof Tank

40400208

BTP
/BPS

Petroleum and Solvent Evaporation; Petroleum Liquids Storage (non-Refinery); Bulk Plants;
Gasoline RVP 10: Standing Loss (67000 Bbl Cap.) - Floating Roof Tank

40400210

BTP
/BPS

Petroleum and Solvent Evaporation; Petroleum Liquids Storage (non-Refinery); Bulk Plants;
Gasoline RVP 13/10/7: Withdrawal Loss (67000 Bbl Cap.) - Float RfTnk

40400211

BTP
/BPS

Petroleum and Solvent Evaporation; Petroleum Liquids Storage (non-Refinery); Bulk Plants;
Gasoline RVP 13: Filling Loss (10500 Bbl Cap.) - Variable Vapor Space

40400212

BTP
/BPS

Petroleum and Solvent Evaporation; Petroleum Liquids Storage (non-Refinery); Bulk Plants;
Gasoline RVP 10: Filling Loss (10500 Bbl Cap.) - Variable Vapor Space

40400213

BTP
/BPS

Petroleum and Solvent Evaporation; Petroleum Liquids Storage (non-Refinery); Bulk Plants;
Gasoline RVP 7: Filling Loss (10500 Bbl Cap.) - Variable Vapor Space

40400230

BTP
/BPS

Petroleum and Solvent Evaporation; Petroleum Liquids Storage (non-Refinery); Bulk Plants;
Specify Liquid: Standing Loss - External Floating Roof w/ Primary Seal

40400231

BTP
/BPS

Petroleum and Solvent Evaporation; Petroleum Liquids Storage (non-Refinery); Bulk Plants;
Gasoline RVP 13: Standing Loss - Ext. Floating Roof w/ Primary Seal

240


-------
see

Typ
e

Description

40400232

BTP
/BPS

Petroleum and Solvent Evaporation; Petroleum Liquids Storage (non-Refinery); Bulk Plants;
Gasoline RVP 10: Standing Loss - Ext. Floating Roof w/ Primary Seal

40400233

BTP
/BPS

Petroleum and Solvent Evaporation; Petroleum Liquids Storage (non-Refinery); Bulk Plants;
Gasoline RVP 7: Standing Loss - External Floating Roof w/ Primary Seal

40400240

RBT

Petroleum and Solvent Evaporation; Petroleum Liquids Storage (non-Refinery); Bulk Plants;
Specify Liquid: Standing Loss - Ext. Floating Roof w/ Secondary Seal

40400241

BTP
/BPS

Petroleum and Solvent Evaporation; Petroleum Liquids Storage (non-Refinery); Bulk Plants;
Gasoline RVP 13: Standing Loss - Ext. Floating Roof w/ Secondary Seal

40400248

BTP
/BPS

Petroleum and Solvent Evaporation; Petroleum Liquids Storage (non-Refinery); Bulk Plants;
Gasoline RVP 10/13/7: Withdrawal Loss - Ext. Float Roof (Pri/Sec Seal)

40400249

RBT

Petroleum and Solvent Evaporation; Petroleum Liquids Storage (non-Refinery); Bulk Plants;
Specify Liquid: External Floating Roof (Primary/Secondary Seal)

40400250

BTP
/BPS

Petroleum and Solvent Evaporation; Petroleum Liquids Storage (non-Refinery); Bulk Plants;
Loading Racks

40400251

BTP
/BPS

Petroleum and Solvent Evaporation; Petroleum Liquids Storage (non-Refinery); Bulk Plants;
Valves, Flanges, and Pumps

40400252

BTP
/BPS

Petroleum and Solvent Evaporation; Petroleum Liquids Storage (non-Refinery); Bulk Plants;
Miscellaneous Losses/Leaks: Vapor Collection Losses

40400253

BTP
/BPS

Petroleum and Solvent Evaporation; Petroleum Liquids Storage (non-Refinery); Bulk Plants;
Miscellaneous Losses/Leaks: Vapor Control Unit Losses

40400260

RBT

Petroleum and Solvent Evaporation; Petroleum Liquids Storage (non-Refinery); Bulk Plants;
Specify Liquid: Standing Loss - Internal Floating Roof w/ Primary Seal

40400261

BTP
/BPS

Petroleum and Solvent Evaporation; Petroleum Liquids Storage (non-Refinery); Bulk Plants;
Gasoline RVP 13: Standing Loss - Int. Floating Roof w/ Primary Seal

40400262

BTP
/BPS

Petroleum and Solvent Evaporation; Petroleum Liquids Storage (non-Refinery); Bulk Plants;
Gasoline RVP 10: Standing Loss - Int. Floating Roof w/ Primary Seal

40400263

BTP
/BPS

Petroleum and Solvent Evaporation; Petroleum Liquids Storage (non-Refinery); Bulk Plants;
Gasoline RVP 7: Standing Loss - Internal Floating Roof w/ Primary Seal

40400270

BTP
/BPS

Petroleum and Solvent Evaporation; Petroleum Liquids Storage (non-Refinery); Bulk Plants;
Specify Liquid: Standing Loss - Int. Floating Roof w/ Secondary Seal

40400271

BTP
/BPS

Petroleum and Solvent Evaporation; Petroleum Liquids Storage (non-Refinery); Bulk Plants;
Gasoline RVP 13: Standing Loss - Int. Floating Roof w/ Secondary Seal

40400272

BTP
/BPS

Petroleum and Solvent Evaporation; Petroleum Liquids Storage (non-Refinery); Bulk Plants;
Gasoline RVP 10: Standing Loss - Int. Floating Roof w/ Secondary Seal

40400273

BTP
/BPS

Petroleum and Solvent Evaporation; Petroleum Liquids Storage (non-Refinery); Bulk Plants;
Gasoline RVP 7: Standing Loss - Int. Floating Roof w/ Secondary Seal

40400278

BTP
/BPS

Petroleum and Solvent Evaporation; Petroleum Liquids Storage (non-Refinery); Bulk Plants;
Gasoline RVP 10/13/7: Withdrawal Loss - Int. Float Roof (Pri/Sec Seal)

241


-------
see

Typ
e

Description

40400279

BTP
/BPS

Petroleum and Solvent Evaporation; Petroleum Liquids Storage (non-Refinery); Bulk Plants;
Specify Liquid: Internal Floating Roof (Primary/Secondary Seal)

40400401

BTP
/BPS

Petroleum and Solvent Evaporation; Petroleum Liquids Storage (non-Refinery); Petroleum Products
- Underground Tanks; Gasoline RVP 13: Breathing Loss

40400402

BTP
/BPS

Petroleum and Solvent Evaporation; Petroleum Liquids Storage (non-Refinery); Petroleum Products
- Underground Tanks; Gasoline RVP 13: Working Loss

40400403

BTP
/BPS

Petroleum and Solvent Evaporation; Petroleum Liquids Storage (non-Refinery); Petroleum Products
- Underground Tanks; Gasoline RVP 10: Breathing Loss

40400404

BTP
/BPS

Petroleum and Solvent Evaporation; Petroleum Liquids Storage (non-Refinery); Petroleum Products
- Underground Tanks; Gasoline RVP 10: Working Loss

40400405

BTP
/BPS

Petroleum and Solvent Evaporation; Petroleum Liquids Storage (non-Refinery); Petroleum Products
- Underground Tanks; Gasoline RVP 7: Breathing Loss

40400406

BTP
/BPS

Petroleum and Solvent Evaporation; Petroleum Liquids Storage (non-Refinery); Petroleum Products
- Underground Tanks; Gasoline RVP 7: Working Loss

40600101

BTP
/BPS

Petroleum and Solvent Evaporation; Transportation and Marketing of Petroleum Products; Tank
Cars and Trucks; Gasoline: Splash Loading

40600126

BTP
/BPS

Petroleum and Solvent Evaporation; Transportation and Marketing of Petroleum Products; Tank
Cars and Trucks; Gasoline: Submerged Loading

40600131

BTP
/BPS

Petroleum and Solvent Evaporation; Transportation and Marketing of Petroleum Products; Tank
Cars and Trucks; Gasoline: Submerged Loading (Normal Service)

40600136

BTP
/BPS

Petroleum and Solvent Evaporation; Transportation and Marketing of Petroleum Products; Tank
Cars and Trucks; Gasoline: Splash Loading (Normal Service)

40600141

BTP
/BPS

Petroleum and Solvent Evaporation; Transportation and Marketing of Petroleum Products; Tank
Cars and Trucks; Gasoline: Submerged Loading (Balanced Service)

40600144

BTP
/BPS

Petroleum and Solvent Evaporation; Transportation and Marketing of Petroleum Products; Tank
Cars and Trucks; Gasoline: Splash Loading (Balanced Service)

40600147

BTP
/BPS

Petroleum and Solvent Evaporation; Transportation and Marketing of Petroleum Products; Tank
Cars and Trucks; Gasoline: Submerged Loading (Clean Tanks)

40600162

BTP
/BPS

Petroleum and Solvent Evaporation; Transportation and Marketing of Petroleum Products; Tank
Cars and Trucks; Gasoline: Loaded with Fuel (Transit Losses)

40600163

BTP
/BPS

Petroleum and Solvent Evaporation; Transportation and Marketing of Petroleum Products; Tank
Cars and Trucks; Gasoline: Return with Vapor (Transit Losses)

40600199

BTP
/BPS

Petroleum and Solvent Evaporation; Transportation and Marketing of Petroleum Products; Tank
Cars and Trucks; Not Classified

40600231

RBT

Petroleum and Solvent Evaporation; Transportation and Marketing of Petroleum Products; Marine
Vessels; Gasoline: Loading Tankers: Cleaned and Vapor Free Tanks

40600232

RBT

Petroleum and Solvent Evaporation; Transportation and Marketing of Petroleum Products; Marine
Vessels; Gasoline: Loading Tankers

242


-------
see

Typ
e

Description

40600233

BTP
/BPS

Petroleum and Solvent Evaporation; Transportation and Marketing of Petroleum Products; Marine
Vessels; Gasoline: Loading Barges: Cleaned and Vapor Free Tanks

40600234

RBT

Petroleum and Solvent Evaporation; Transportation and Marketing of Petroleum Products; Marine
Vessels; Gasoline: Loading Tankers: Ballasted Tank

40600235

BTP
/BPS

Petroleum and Solvent Evaporation; Transportation and Marketing of Petroleum Products; Marine
Vessels; Gasoline: Ocean Barges Loading - Ballasted Tank

40600236

RBT

Petroleum and Solvent Evaporation; Transportation and Marketing of Petroleum Products; Marine
Vessels; Gasoline: Loading Tankers: Uncleaned Tanks

40600237

RBT

Petroleum and Solvent Evaporation; Transportation and Marketing of Petroleum Products; Marine
Vessels; Gasoline: Ocean Barges Loading - Uncleaned Tanks

40600238

RBT

Petroleum and Solvent Evaporation; Transportation and Marketing of Petroleum Products; Marine
Vessels; Gasoline: Loading Barges: Uncleaned Tanks

40600239

RBT

Petroleum and Solvent Evaporation; Transportation and Marketing of Petroleum Products; Marine
Vessels; Gasoline: Tankers: Ballasted Tank

40600240

RBT

Petroleum and Solvent Evaporation; Transportation and Marketing of Petroleum Products; Marine
Vessels; Gasoline: Loading Barges: Average Tank Condition

40600241

BTP
/BPS

Petroleum and Solvent Evaporation; Transportation and Marketing of Petroleum Products; Marine
Vessels; Gasoline: Tanker Ballasting

40600299

RBT

Petroleum and Solvent Evaporation; Transportation and Marketing of Petroleum Products; Marine
Vessels; Not Classified

40600301

BTP
/BPS

Petroleum and Solvent Evaporation; Transportation and Marketing of Petroleum Products; Gasoline
Retail Operations - Stage I; Splash Filling

40600302

BTP
/BPS

Petroleum and Solvent Evaporation; Transportation and Marketing of Petroleum Products; Gasoline
Retail Operations - Stage I; Submerged Filling w/o Controls

40600305

BTP
/BPS

Petroleum and Solvent Evaporation; Transportation and Marketing of Petroleum Products; Gasoline
Retail Operations - Stage I; Unloading

40600306

BTP
/BPS

Petroleum and Solvent Evaporation; Transportation and Marketing of Petroleum Products; Gasoline
Retail Operations - Stage I; Balanced Submerged Filling

40600307

BTP
/BPS

Petroleum and Solvent Evaporation; Transportation and Marketing of Petroleum Products; Gasoline
Retail Operations - Stage I; Underground Tank Breathing and Emptying

40600399

BTP
/BPS

Petroleum and Solvent Evaporation; Transportation and Marketing of Petroleum Products; Gasoline
Retail Operations - Stage I; Not Classified **

40600401

BTP
/BPS

Petroleum and Solvent Evaporation; Transportation and Marketing of Petroleum Products; Filling
Vehicle Gas Tanks - Stage II; Vapor Loss w/o Controls

40600501

RBT

Petroleum and Solvent Evaporation; Transportation and Marketing of Petroleum Products; Pipeline
Petroleum Transport - General - All Products; Pipeline Leaks

40600502

RBT

Petroleum and Solvent Evaporation; Transportation and Marketing of Petroleum Products; Pipeline
Petroleum Transport - General - All Products; Pipeline Venting

40600503

RBT

Petroleum and Solvent Evaporation; Transportation and Marketing of Petroleum Products; Pipeline
Petroleum Transport - General - All Products; Pump Station

40600504

RBT

Petroleum and Solvent Evaporation; Transportation and Marketing of Petroleum Products; Pipeline
Petroleum Transport - General - All Products; Pump Station Leaks

40600602

BTP
/BPS

Petroleum and Solvent Evaporation; Transportation and Marketing of Petroleum Products;
Consumer (Corporate) Fleet Refueling - Stage II; Liquid Spill Loss w/o Controls

243


-------
see

Typ
e

Description

40600701

BTP
/BPS

Petroleum and Solvent Evaporation; Transportation and Marketing of Petroleum Products;
Consumer (Corporate) Fleet Refueling - Stage I; Splash Filling

40600702

BTP
/BPS

Petroleum and Solvent Evaporation; Transportation and Marketing of Petroleum Products;
Consumer (Corporate) Fleet Refueling - Stage I; Submerged Filling w/o Controls

40600706

BTP
/BPS

Petroleum and Solvent Evaporation; Transportation and Marketing of Petroleum Products;
Consumer (Corporate) Fleet Refueling - Stage I; Balanced Submerged Filling

40600707

BTP
/BPS

Petroleum and Solvent Evaporation; Transportation and Marketing of Petroleum Products;
Consumer (Corporate) Fleet Refueling - Stage I; Underground Tank Breathing and Emptying

40688801

BTP
/BPS

Petroleum and Solvent Evaporation; Transportation and Marketing of Petroleum Products; Fugitive
Emissions; Specify in Comments Field

2501050120

RBT

Storage and Transport; Petroleum and Petroleum Product Storage; Bulk Terminals: All Evaporative
Losses; Gasoline

2501055120

BTP
/BPS

Storage and Transport; Petroleum and Petroleum Product Storage; Bulk Plants: All Evaporative
Losses; Gasoline

2501060050

BTP
/BPS

Storage and Transport; Petroleum and Petroleum Product Storage; Gasoline Service Stations; Stage
1: Total

2501060051

BTP
/BPS

Storage and Transport; Petroleum and Petroleum Product Storage; Gasoline Service Stations; Stage
1: Submerged Filling

2501060052

BTP
/BPS

Storage and Transport; Petroleum and Petroleum Product Storage; Gasoline Service Stations; Stage
1: Splash Filling

2501060053

BTP
/BPS

Storage and Transport; Petroleum and Petroleum Product Storage; Gasoline Service Stations; Stage
1: Balanced Submerged Filling

2501060200

BTP
/BPS

Storage and Transport; Petroleum and Petroleum Product Storage; Gasoline Service Stations;
Underground Tank: Total

2501060201

BTP
/BPS

Storage and Transport; Petroleum and Petroleum Product Storage; Gasoline Service Stations;
Underground Tank: Breathing and Emptying

2501995000

BTP
/BPS

Storage and Transport; Petroleum and Petroleum Product Storage; All Storage Types: Working
Loss; Total: All Products

2505000120

RBT

Storage and Transport; Petroleum and Petroleum Product Transport; All Transport Types; Gasoline

2505020120

RBT

Storage and Transport; Petroleum and Petroleum Product Transport; Marine Vessel; Gasoline

2505020121

RBT

Storage and Transport; Petroleum and Petroleum Product Transport; Marine Vessel; Gasoline -
Barge

2505030120

BTP
/BPS

Storage and Transport; Petroleum and Petroleum Product Transport; Truck; Gasoline

2505040120

RBT

Storage and Transport; Petroleum and Petroleum Product Transport; Pipeline; Gasoline

2660000000

BTP
/BPS

Waste Disposal, Treatment, and Recovery; Leaking Underground Storage Tanks; Leaking
Underground Storage Tanks; Total: All Storage Types

244


-------
245


-------