%
\
National Scale Modeling of Air Toxics
for the Mobile Source Air Toxics Rule;
Technical Support Document
-------
United States Office of Air Quality Planning and Standards Publication No. EPA 454/R-06-002
Environmental Protection Emissions, Monitoring and Analysis Division January 2006
Agency Research Triangle Park, NC
-------
TECHNICAL REPORT DATA
(Please read Instructions on reverse before completing)
1. REPORT NO. 2.
EPA-454/R-06-002
4. TITLE AND SUBTITLE
National Scale Modeling of Air Toxics for the Mobile
Source Air Toxics Rule; Technical Support Document
7 . AUTHOR ( S )
9. PERFORMING ORGANIZATION NAME AND ADDRESS
12. SPONSORING AGENCY NAME AND ADDRESS
U.S. Environmental Protection Agency
Office of Air Quality Planning and Standards
Emissions, Monitoring & Analysis Division
Research Triangle Park, NC 27711
3. RECIPIENT'S ACCESSION NO.
5 . REPORT DATE
January 2006
6. PERFORMING ORGANIZATION CODE
8. PERFORMING ORGANIZATION REPORT NO.
10. PROGRAM ELEMENT NO.
11. CONTRACT/GRANT NO. EPA COntr3Ct
NO.IAG47939482-01 (CSC)
13. TYPE OF REPORT AND PERIOD COVERED
Final Report
14. SPONSORING AGENCY CODE
SUPPLEMENTARY NOTES
EPA Work Assignment Manager: Madeleine Strum
16. ABSTRACT
The purpose of the work described in this technical document was to project emissions
for mobile source hazardous air pollutants (HAPs) to 2007, 2010, 2015, 2020, and 2030
from the 1999 National Emissions Inventory Version 3 (NEI), conduct air quality and
exposure modeling, and estimate cancer and non-cancer risk for select future years.
Air quality modeling utilized the Assessment System for Population Exposure Nationwide
(ASPEN) model. Exposure modeling utilized the Hazardous Air Pollutant Exposure Model,
Version 5 (HAPEM5). Modeling was done for reference cases, which included programs
currently planned and in place, as well as control scenarios that evaluated potential
impacts of additional control programs. This work was done to support regulatory needs
related to the 2006 proposed mobile source air toxics rule. Intermediate year
inventories for 2002 through 2010, inclusive, were also developed to support other
program needs in the Office of Air and Radiation.
KEY WORDS AND DOCUMENT ANALYSIS
a. DESCRIPTORS
Air Pollution
Air Quality Dispersion Models
Meteorology
Air Toxics
Urban Area Modeling
18. DISTRIBUTION STATEMENT
Release Unlimited
b. IDENTIFIERS/OPEN ENDED TERMS
19. SECURITY CLASS (Report)
Unclassified
20. SECURITY CLASS (Page)
Unclassified
c. COSATI Field/Group
21. NO. OF PAGES
222
22. PRICE
EPA Form 2220-1 (Rev. 4-77)
PREVIOUS EDITION IS OBSOLETE
-------
EPA-454/ R-06-002
January 2006
National Scale Modeling of Air Toxics for the Mobile Source Air Toxics Rule;
Technical Support Document
U.S. Environmental Protection Agency
Office of Air Quality Planning and Standards
Emissions, Monitoring and Analysis Division
Research Triangle Park, North Carolina
-------
Disclaimer
The information in this document has been reviewed in accordance with the U.S. EPA
administrative review policies and approved for publication. Mention of trade names or
commercial products does not constitute endorsement or recommendation for their use.
The following trademarks appear in this document:
UNIX is a registered trademark of AT&T Bell Laboratories.
Linux is a registered trademark of Red Hat
SAS® is a registered trademark of SAS Institute
SUN is a registered trademark of Sun Microsystems, Inc.
-------
Table of Contents
1. Purpose of Work 1
2. 1999 base inventories 5
2.1 1999 HAP inventories 5
2.2 1999 Precursor inventories 9
3. Development of Future Year Mobile and Mobile-Related Emissions 11
3.1 Locomotive and commercial marine vessels 11
3.2 Aircraft and Aviation gasoline 19
3.3 Projection of onroad and nonroad categories using NMIM 25
3.3.1 Description of NMIM 25
3.3.2 Onroad projections using NMIM 25
3.3.3 Nonroad projections using NMIM (excluding aircraft, locomotives, and commercial
marine vessels) 32
3.3.4 Projection of onroad refueling emissions 41
3.4 Projection of HAP Precursor Emissions from Mobile Sources 44
3.4.1 Locomotive and Commercial Marine Vessel Precursor Emissions 44
3.4.2 Aircraft Precursor Emissions 44
3.4.3 Onroad Precursor Emissions 45
3.4.4 Nonroad Precursor Emissions (excluding aircraft, locomotives, and commercial
marine vessels) 45
4. Development of Future Year Stationary Source Emissions 47
4.1 Growth factors 47
4.1.1 MACT based growth factors 47
4.1.2 SIC based growth factors 51
4.1.3 SCC based growth factors 52
4.2 Reduction factors 53
4.3 Application of growth and reductions to project stationary source emissions 55
5. EMS-HAP Processing for HAPs 57
5.1 Point sources 57
5.2 Non-point sources 58
5.3 Onroad sources 58
5.4 Nonroad sources 59
5.4.1 Aircraft sources 59
5.4.2 Airport Support Equipment 59
5.4.3 Remaining nonroad sources 59
5.5 EMS-HAP for precursors 60
6. ASPEN Processing 63
6.1 MSAT HAPs 63
6.2 Precursors 63
6.3 Post-processing of ASPEN concentrations 67
7. HAPEM5 Model and Post-Processing 77
7.1 HAPEM model 77
7.2 Summaries of annual HAPEM5 output 79
8. Cancer and non-cancer risk calculations 83
8.1 Cancer risk calculations 84
-------
Table of Contents
8.2 Non-cancer risk calculations 85
8.3 Cancer and non-cancer risk population statistics using 2000 and projected population .... 88
8.3.1 Allocation of future county level populations to tract level 88
8.3.2 Population statistic calculations for cancer risk 89
8.3.3 Population statistic calculations for non-cancer respiratory hazard index 90
8.3.4 Population statistic calculations using 2000 population for all years 92
9. Background concentration sensitivity analysis 93
10. Benzene Control Scenario 97
10.1 Stationary gasoline distribution and vehicle gasoline refueling inventory 97
10.2 Highway gasoline vehicle inventory 107
10.3 Nonroad gasoline inventory 108
10.4 EMS-HAP Processing 109
10.5 ASPEN Processing and Post-Processing 110
10.6 HAPEM Processing and Post-Processing Ill
10.7 Cancer and Non-cancer Calculations 112
10.7.1 Cancer 113
10.7.2 Non-cancer 114
10.7.3 Population statistics 115
References 119
Appendix A: Documentation of NMEVI Runs Used to Develop Inventories for MS AT Rule Air
Quality Modeling A-l
Appendix B: Steps and Example calculations of onroad projections B-l
Appendix C: Example calculations of nonroad projections C-l
Appendix D: Risk Calculations D-l
Appendix E: Control of stationary refueling and gasoline marketing emissions E-l
Appendix F: Control of onroad gasoline emissions F-l
Appendix G: Development of controlled nonroad inventory G-l
11
-------
List of Tables
Table 1. Pollutants of interest in MS AT study. CAS numbers in italics are in the stationary
inventories only; otherwise they are in mobile and stationary inventories 2
Table 2. Changes made to the 1999 NEI HAP inventories prior to processing for 1999 NATA or
projections 6
Table 3. Emissions (tons) for MSAT HAPs in the 1999 NEI inventories. Totals include Puerto
Rico and the Virgin Islands 8
Table 4. Non-HAP precursors for the MSAT secondary HAPs with source sector emissions for
1999. Totals include Puerto Rico and the Virgin Islands 10
Table 5. Locomotive SCC codes in the 1999 NEI nonroad inventory 11
Table 6. Locomotive 50-State annual emissions trends (tons per year) and future year ratios... 12
Table 7. Locomotive HAPs. HAPs not in bold are not emphasized in the MSAT study but are
projected 13
Table 8. Commercial marine vessel 50-State annual emissions trends (tons per year) and future
year ratios used as projection factors 14
Table 9. Commercial marine vessel SCC codes, HAPs, and basis of projection factors. HAPs in
bold are emphasized in the MSAT study 15
Table 10. National locomotive emissions (rounded) by SCC for selected HAPs and across all HAPs. 17
Table 11. National commercial marine vessel emissions (rounded) by SCC for selected HAPs
and across all HAPs 18
Table 12. TAP landing and take-off data for 2002 through 2020, 2015, and 2020 19
Table 13. Aircraft growth factors for MSAT study years 20
Table 14. Airport related SCC codes and assigned growth factor basis 21
Table 15. Airport related emissions (excluding airport support equipment) for selected HAPs
and all HAPs by SCC. Non-point SCC emissions for 2030 are set equal to 2020 24
Table 16. HDDV SCC codes used to calculate HDDV emissions for NEI projections 28
Table 17. Motorcycle (MC) SCC codes not in NMIM output for Alpine, Modoc, and Sierra
Counties California 29
Table 18. National summary of projected onroad emissions by vehicle type for 1999, 2007,
2010, 2015, 2020, and 2030 across all HAPs and for 1,3-butadiene, acetaldehyde, acrolein,
benzene, formaldehyde, and naphthalene 30
Table 19. SCC codes in the 1999 NEI inventory and not in the NMIM inventory 33
Table 20. National engine emissions for selected HAPs and total MSAT HAPs for 1999, 2007,
2010, 2015, 2020, and 2030 35
Table 21. National equipment emissions for selected HAPs and all MSAT HAPs for 1999, 2007,
2010, 2015, 2020, and 2030 37
Table 22. National engine/equipment emissions for MSAT HAPs 40
Table 23. Onroad refueling SCC codes 42
Table 24. Onroad refueling emissions by SCC for 1999, 2007, 2010, 2015 and 2020 43
The precursors from nonroad emission categories covered by the NONROAD model were
processed using a similar methodology as the emissions for HAPs. However, instead of
HAP specific projection ratios, we used VOC ratios from NMEVI.Table 25. HDDV SCC
codes used to calculate HDDV emissions in the precursor inventory 45
Table 25. HDDV SCC codes used to calculate HDDV emissions in the precursor inventory.... 46
Table 26. National level MACT growth factors for 2015 and 2020 49
iii
-------
List of Tables
Table 27. Utility Boilers: Coal (MACT= 1808-1) state level growth factors for 2015 and 2020.50
Table 28. SIC codes changed due to unrealistic growth factors 51
Table 29. Summary of Categories for which reductions were applied in EMS-HAP 54
Table 30. 1999 and projected stationary emissions for selected HAPs and total MSAT HAPs... 56
Table 31. ASPEN emission groups for MSAT for future years 60
Table 32. Reactivity classes for MSAT HAPs and precursors 64
Table 33. Description of emissions files for the stationary and mobile divisions used for ASPEN
simulations 65
Table 34. National average background, stationary, and mobile ASPEN concentrations (|j,g m"3)
for each MS AT HAP for 2015, 2020, and 2030 70
Table 35. National average stationary and mobile HAPEM concentrations (ng m"3) for 2015,
2020, and 2030 by HAP 81
Table 36. MSAT HAPs carcinogenic class, URE, non-cancer target organ systems, and Rfc.
N/A denotes HAP is neither a cancer or non-cancer HAP 83
Table 37. National average inhalation cancer risks for stationary and mobile sources for MSAT
HAPs, each carcinogenic class and total risk (all MSAT HAPs) 85
Table 38. National average non-cancer hazard quotient (HQ) for MSAT HAPs and hazard index
(HI) for organ systems for stationary and mobile sources 88
Table 39. Population risk classes for mobile total risk for 2015, 2020, and 2030 using projected
populations for each year 90
Table 40. Population respiratory HI classes for mobile sources for 2015, 2020, and 2030 using
projected populations for each year 92
Table 41. Total benzene emissions of counties within 300 km of Wake County, NC for 1999,
2015, 2020 and 2030, 1999 background benzene concentration for Wake County, and
scaled background concentrations for Wake County for 2015, 2020, and 2030 94
Table 42. National average 1999 background and scaled backgrounds for 1,3-butadiene,
acetaldehyde, benzene, formaldehyde, and xylenes 95
Table 43. National average total concentrations (all sources and background) for 2015, 2020,
and 2030 using both the 1999 background and the scaled backgrounds 95
Table 44. Benzene gasoline marketing and distribution SCC codes to be controlled 98
Table 45. Change in Average Fuel Benzene Level (Volume Percent) by PADD with
Implementation of Proposed Fuel Benzene Standard (CG - Conventional Gasoline; RFG -
Reformulated Gasoline) 104
Table 46. Benzene stationary emissions (tons) before and after applying controls to reference
case gasoline marketing and distribution emissions (non refueling gasoline) and vehicle
refueling emissions. Also shown are the percent differences (control-reference). 1999 NEI
emissions are shown for comparison 105
Table 47. National MSAT reference and controlled emissions (nearest ton) for gasoline powered
vehicles by HAP for 2015, 2020, and 2030 108
Table 48. 2015, 2020, and 2030 reference and controlled emissions for the five HAPS for
nonroad gasoline categories 109
Table 49. National average 1999 and future year reference and controlled benzene stationary
concentrations 110
IV
-------
List of Tables
Table 50. National average reference and controlled onroad gasoline concentrations for the five
HAPs for 2015, 2020, and 2030 Ill
Table 51. National average reference and controlled nonroad gasoline concentrations for the five
HAPs for 2015, 2020, and 2030 Ill
Table 52. National average 1999 and future reference and controlled benzene HAPEM
stationary concentrations 112
Table 53. National average reference and controlled HAPEM onroad gasoline concentrations for
the five HAPs for 2015, 2020, and 2030 112
Table 54. National average reference and controlled HAPEM nonroad gasoline concentrations
for the five HAPs for 2015, 2020, and 2030 112
Table 55. National average risks from stationary sources for 1999 and future year reference and
controlled benzene, carcinogen class A, and total (all MS AT HAPs) 113
Table 56. Reference and controlled HAPEM onroad gasoline risks for 2015, 2020, and 2030 for
individual HAPs and carcinogen classes A, Bl, and B2 and total risk (all MSAT HAPs,
including HAPs not controlled) 113
Table 57. Reference and controlled HAPEM nonroad gasoline risks for 2015, 2020, and 2030
for individual HAPs and carcinogen classes A, Bl, and B2 and total risk (all MSAT HAPs,
including HAPs not controlled) 114
Table 58. 1999 and future year reference and controlled stationary benzene hazard quotients and
immune system hazard indices for MSAT HAPs for 2015 and 2020 114
Table 59. Reference and controlled HAPEM onroad gasoline HQ for controlled HAPs and HI
for immune, reproductive, and respiratory systems (including MSAT HAPs not controlled)
for 2015, 2020, and 2030 115
Table 60. Reference and controlled HAPEM nonroad gasoline HQ for controlled MSAT HAPs
and HI for immune, reproductive, and respiratory systems (from MSAT HAPs including
those HAPs not controlled) for 2015, 2020, and 2030 115
Table 61. Population risk classes for stationary and mobile total risk for 2015, 2020, and 2030
for reference and controlled risks from MSAT HAPs using projected populations for each
year. The total category includes background contributions 116
Table 62. Population HI classes for mobile and total respiratory HI for 2015, 2020, and 2030 for
reference and controlled risks using projected populations for each year. The total category
includes background contributions 117
-------
List of Figures
Figure 1. Format of the aircraft growth factor file. 2010 growth factors shown as example 22
Figure 2. Flowchart of aircraft and aviation gasoline emissions projections 23
Figure 3. 2015 county level refueling projection factors 41
Figure 4. Annual emissions by source sector at the national level 56
Figure 5. Box and whisker plots of ratios of stationary secondary contributions to total
concentrations (white boxes) and ratios of mobile secondary contributions to total
concentrations (gray boxes) for 1999 acetaldehyde and formaldehyde concentrations. Dots
represent the national mean ratios 61
Figure 6. Stationary source emission input files and ASPEN output files for each reactivity class
forMSATHAPs 65
Figure 7. Mobile source emission input files and ASPEN output files for each reactivity class for
MSATHAPs 66
Figure 8. Mobile source emission input files and ASPEN output files for each reactivity class for
MS AT precursors 67
Figure 9. 1999 County level median total (all sources and background) concentrations (ng m"3)
for benzene 71
Figure 10. 2015 County level median total (all sources and background) concentrations (|j,g m"3)
for benzene 72
Figure 11. 2020 County level median total (all sources and background) concentrations (ng m"3)
for benzene 73
Figure 12. 2030 County level median total (all sources and background) concentrations (|j,g m"3)
for benzene 74
Figure 13. Sample records of the Runl 2015 HAPEM input air quality file con45201_runl.txt
for benzene. Note that each set of concentrations for a tract is one record. More records
appear due to of "wrapping" of text in word processor 75
Figure 14. Sample records of the Run2 2015 HAPEM input air quality file con45201_run2.txt
for benzene 75
Figure 15. Sample records showing HAPEM5 output for Benzene Runl. Filename is
2015_45201_runl.HAPEM5-TRACT.txt. Variables are FIPS, tract id, major, area & other,
onroad gasoline, nonroad gasoline, total and background concentrations 79
Figure 16. Sample records showing HAPEM5 output for Benzene Run2. Filename is
2015_45201_run2.HAPEM5-TRACT.txt. Variables are FIPS, tract id, major, area and
other, onroad diesel, nonroad other, total and background concentrations 79
Figure 17. 2015 HAPEM county median total concentrations (all sources) for benzene 82
Figure 18. County median total inhalation cancer risks for all MSAT HAPs for 2015. Risk is
characterized as N in a million 85
Figure 19. County median total (all sources) non-cancer hazard index for MSAT HAPs affecting
the respiratory system 87
Figure 20. Counties within 300 km of the centroid of Wake County, North Carolina (county in
gray). Dots represent county centroids 94
Figure 21. Benzene background concentrations (ug m"3) for a) 1999 background, b) 2015 scaled
background c) 2020 scaled background and d)2030 scaled background 96
Figure 22. Xylenes background concentrations (ug m"3) for a) 1999 background, b) 2015 scaled
background c) 2020 scaled background and d) 2030 scaled background 96
vi
-------
List of Figures
Figure 23. PADD regions for the U.S 103
Figure 24. RFG counties (dark gray) for the U.S 104
vn
-------
List of Acronyms
AEO Annual Energy Outlook
ASPEN Assessment System for Population Exposure Nationwide
BEA Bureau of Economic Analysis
CAS Chemical Abstract Service
EGAS Economic Growth Analysis System
EMS-HAP The Emissions Modeling System for Hazardous Air Pollutants
EPA United States Environmental Protection Agency
HAP Hazardous Air Pollutant
HAPEM5 Hazardous Air Pollutant Exposure Model, Version 5
HI non-cancer Hazard Index for a target organ system
HQ non-cancer Hazard Quotient for an individual HAP
MACT Maximum Available Control Technology standards for HAP, established under
Section 112 of the Clean Air Act
MSAT Mobile Source Air Toxics
NATA National Air Toxics Assessment
NEI EPA's National Emission Inventory
NMIM National Mobile Inventory Model
OAQPS EPA's Office of Air Quality Planning and Standards
OTAQ EPA's Office of Transportation and Air Quality
REMI Regional Economic Model, Inc.
SAROAD Air pollution chemical species classification system used in EPA's initial data
base for "Storage and Retrieval of Aerometric Data"
SCC Source Classification Code
SIC Standard Industrial Classification code used for Federal economic statistics
TAP Terminal Area Forecast
URE Unit risk estimate for cancer risk
VI
-------
List of files referenced in document
File
onroad_0923.xls
onroad_pivot.xls
nonroad_0923.xls
nonroad_pivot. xls
mwi.sas
concentrations.xls
ASPEN_medians.ppt
hapem_concentrations.xls
HAPEM_medians.ppt
hapem_risks.xls
risk_030305.ppt
hapem_hq.xls
hq_030305.ppt
pop_stats_risk.xls
pop_stats_hi_respiratory.xls
cty cntr99.sas7bdat
background_test.xls
background_acetaldehyde_0 111 .ppt
background_butadiene_0 111 .ppt
background_test_0 111 .ppt
background_formaldehyde_0 111 .ppt
background_xylenes_0 111 .ppt
benzene_gas_scc.xls
onroad_controls.xls
onroad_controls_pivot. xls
Description
Excel workbook of onroad emissions by vehicle type for state and
national level
Excel workbook containing pivot table of onroad emissions by
vehicle type for state and national
Excel workbook of nonroad emissions by engine, equipment, and
engine/equipement type for state and national level
Excel workbook containing pivot table of nonroad emissions by
engine, equipment, and engine/equipment type for state and
national level
SAS® program to substitute 2002 MWI point emissions in the
1999 point inventory
Excel workbook of national and state mean concentrations and
concentration distribution for ASPEN results
PowerPoint file containing national maps of county median total
ASPEN concentrations
Excel workbook of national and state mean concentrations and
concentration distribution for HAPEM results
PowerPoint file containing national maps of county median total
HAPEM concentrations
Excel workbook of national and state mean risks and risk
distribution for HAPEM based results
PowerPoint file containing national maps of county median total
HAPEM based risks
Excel workbook of national and state mean HQ and HI and HQ
and HI distribution for HAPEM based results
PowerPoint file containing national maps of county median total
HAPEM based HQ and respiratory HI
Excel workbook of population statistics using both 2000 and
projected populations for cancer risk
Excel workbook of population statistics using both 2000 and
projected populations for respiratory HI
S AS® dataset of county centroids
Excel workbook containing national and county mean
concentrations using 1999 background and scaled backgrounds
PowerPoint file containing county maps of concentrations using
1999 and scaled backgrounds for acetaldehyde.
PowerPoint file containing county maps of concentrations using
1999 and scaled backgrounds for 1,3 -butadiene.
PowerPoint file containing county maps of concentrations using
1999 and scaled backgrounds for benzene.
Powerpoint file containing county maps of concentrations using
1999 and scaled backgrounds for formaldehyde.
PowerPoint file containing county maps of concentrations using
1999 and scaled backgrounds for xylenes.
Excel workbook containing benzene gasoline distribution
reference and controlled emissions by SCC
Excel workbook of controlled onroad emissions by vehicle type
for state and national level
Excel workbook containing pivot table of controlled onroad
emissions by vehicle type for state and national
Section
o o ^>
3.3.2
3.3.2
3.3.3
3.3.3
4.3
6.3
6.3
7.2
7.2
8.1
8.1
8.2
8.2
8.3
8.3
9
9
9
9
9
9
9
10.1
10.2
10.2
Vll
-------
List of files referenced in document
File
nonroad_ controls.xls
nonroad_pivot_ controls.xls
aspen_conc_controls.xls
ASPEN_median_cntrl.ppt
hapem_concentrations_cntrl.xls
HAPEM_median_cntrl.ppt
hapem_risks_control. xls
risk_cntrll.ppt
hapem_hq_control. xls
hq_cntrll.ppt
pop_stats_risk_cntrl. xls
pop_stats_hi_resp_cntrl.xls
onroad.sas
loco_marine.sas
marine_locomotive_growth. sas
nonroad.sas
cancer risk, sas
noncancer.sas
project_stationary_benz.sas
control onroad.sas
calc_factors.sas
control nonroad.sas
Description
Excel workbook of controlled nonroad emissions by engine,
equipment, and engine/equipment type for state and national level
Excel workbook containing pivot table of controlled onroad
emissions by engine, equipment, and engine/equipment type for
state and national level
Excel workbook of national and state mean controlled
concentrations and concentration distribution for ASPEN results
PowerPoint file containing national maps of county median total
ASPEN controlled concentrations
Excel workbook of national and state mean concentrations and
concentration distribution for HAPEM controlled results
PowerPoint file containing national maps of county median total
HAPEM controlled concentrations
Excel workbook of national and state mean risks and risk
distribution for controlled HAPEM based results
PowerPoint file containing national maps of county median total
controlled HAPEM based risks
Excel workbook of national and state mean HQ and HI and HQ
and HI distribution for controlled HAPEM based results
PowerPoint file containing national maps of county median total
controlled HAPEM based HQ and respiratory HI
Excel workbook of population statistics using both 2000 and
projected populations for controlled cancer risk
Excel workbook of population statistics using both 2000 and
projected populations for controlled respiratory HI
SAS® program to project 1999 onroad emissions to future years
SAS® program to create locomotive and CMV projection factor
files
SAS® program to project 1999 locomotive and CMV emissions to
future years.
SAS® program to project nonroad emissions (excluding aircraft,
locomotives, and CMV) to future years
SAS® program to calculate risk estimates from HAPEM results
SAS® program to calculate non-cancer estimates from HAPEM
results
SAS® program to apply controls to projected benzene gasoline
distribution emissions
SAS® program to apply controls to projected onroad emissions
SAS® program to calculate exhaust and evaporative factors for
use in controlling projected nonroad emissions
SAS® program to control projected nonroad emissions
Section
10.3
10.3
10.5
10.5
10.6
10.6
10.7.1
10.7.1
10.7.2
10.7.2
10.7.3
10.7.3
App. B
App.C
App.C
App.C
App. D
App. D
App. E
App. F
App. G
App. G
Vlll
-------
1. Purpose of Work
The purpose of the work described in this technical document was to project emissions for
mobile source hazardous air pollutants (HAPs) to 2007, 2010, 2015, 2020, and 2030 from the
1999 National Emissions Inventory Version 3 (NEI) (U. S. EPA, 2004a), conduct air quality and
exposure modeling, and estimate cancer and non-cancer risk for select future years. Air quality
modeling utilized the Assessment System for Population Exposure Nationwide (ASPEN) model
(U. S. EPA, 2000). Exposure modeling utilized the Hazardous Air Pollutant Exposure Model,
Version 5 (HAPEM5) (U.S. EPA, 2005d). Cancer risk and non-cancer risk were estimated for
2015, 2020, and 2030. Modeling was done for reference cases, which included programs
currently planned and in place, as well as control scenarios that evaluated potential impacts of
additional control programs. This work was done to support regulatory needs related to the 2006
proposed mobile source air toxics rule. Intermediate year inventories for 2002 through 2010,
inclusive, were also developed to support other program needs in the Office of Air and
Radiation.
The pollutants modeled in this study, in support of the mobile source air toxics rule, are shown in
Table 1. They are referenced in the document as MSAT HAPs. These pollutants are all included
in the NEI and are on EPA's list of hazardous pollutants in Section 112 of the Clean Air Act.
They are also emitted by mobile sources. In this assessment, projected inventories were
developed for both the mobile and stationary emission sources in the 1999 NEI. There are
additional hazardous air pollutants in the 1999 NEI with a mobile source emissions estimate that
are not included in Table 1. Some of these were pollutants found only in data submitted by
individual States. Others were generated by EPA through the use of speciation factors obtained
from a non-mobile source process (e.g., commercial marine vessels, residual oil). More
information on the 1999 NEI development can be found at www.epa.gov/ttn/chief/.
Emission projection methods for other HAPs (non-MSAT HAPs) are also discussed in several
places in this document. These projections were done to support other OAR program needs.
After inventory projection, these pollutants were modeled in ASPEN and HAPEM5, following
the same general methods used in the 1999 National Air Toxics Assessment
(www. epa. gov/ttn/nata99).
The remainder of this document describes the methodology used for the inventory projections
and subsequent air quality modeling. Section 2 describes the 1999 base HAP and precursor
inventories, Section 3 describes the development of the projected mobile inventories and
refueling projection factors. Section 4 describes the development of the projected stationary
inventories. Sections 5, 6, and 7 describe the emissions processing, air quality modeling and
exposure modeling. Section 8 describes the calculation of cancer risk and non-cancer risk
(hazard quotients and hazard indices). Section 9 describes the methodology to adjust future year
background concentrations based on projected emissions, and Section 10 describes the
methodology to develop future year inventories that incorporate the benzene control scenario.
Appendices also follow describing the calculations in more detail, and providing sample
calculations and additional supporting data.
1
-------
Table 1. Pollutants of interest in MS AT study. CAS numbers in italics are in the stationary
inventories only; otherwise they are in mobile and stationary inventories.
HAP
CAS or pollutant code in 1999 NEI
SAROAD(s)
Organic gaseous HAPs (excluding those assessed as POM group)
1,3 -Butadiene
2,2,4-Trimethylpentane
Acetaldehyde
Acrolein
Benzene
Ethyl Benzene
Formaldehyde
Hexane
Methyl tert-butyl ether (MTBE)
Naphthalene
Propionaldehyde
Styrene
Toluene
Xylenes
106990
540841
75070
107028
71432
100414
50000
110543
1634044
91203
123386
100425
108883
106423, 108383, 1330207, 95476
43218
43250
43503
43505
45201
45203
43502
43231
43376
46701, 46702
43504
45220
45202
45102
Metal HAPs
Chromium III
Chromium VI
Manganese
Nickel
10060125, 12018018, 1308389, 136, 16065831, 21679312,
7440473
10294403, 10588019, 11103869, 11115745, 1308130,
1333820, 13530659, 136, 13765190, 14307358, 18454121,
18540299, 7440473, 7738945, 7758976, 7775113,
7778509, 7789006, 7789062
10101505, 1313139, 1317346, 1317357, 198,7439965,
7722647, 7783166, 7785877
10101970, 12054487, 13138459, 1313991, 1314063,
13462889, 13463393, 13770893, 226, 373024, 7440020,
7718549, 7786814, NY059280
59992, 59993
69992, 69993
80196, 80396
80216,80316
HAPs grouped as POM
Acenaphthene
Acenaphthylene
Anthracene
Benzo(g,h,i)perylene
Fluoranthene
Fluorene
Phenanthrene
Pyrene
Benzo(a)pyrene
Dibenzo(a,h)anthracene
Benz(a)anthracene
Benzo(b)fluoranthene
Benzo(k)fluoranthene
Indeno(l,2,3,c,d)-pyrene
Chrysene
83329
208968
120127
191242
206440
86737
85018
129000
50328
53703
56553
205992
207089
193395
218019
72002
72002
72002
72002
72002
72002
72002
72002
75002
75002
76002
76002
76002
76002
77002
Description of POM groups by SAROAD
72002: POM, Group 2: no URE data
75002: POM, Group 5: 5.0E-4 < URE < 5.0E-3
76002: POM, Group 6: 5.0E-5 < URE < 5.0E-4
77002: POM, Group 7: 5.0E-6 < URE < 5.0E-5
-------
The following notes apply to Table 1:
• Although designated as SAROAD code in the EMS-HAP User's Guide (U.S. EPA,
2004b) and ASPEN User's Guide (U.S. EPA, 2000), the SAROAD code value in Table 1
is not the actual SAROAD code for the HAP. Rather, it is a 5-digit code used by ASPEN
and EMS-HAP to represent the specific pollutant or pollutant group that is modeled in
ASPEN.
• For HAPs with two SAROAD codes, the lower numbered code represents the fine
particle mode and the other number represents the coarse particle mode. For
naphthalene, the lower numbered code represents the gas mode while the higher number
represents the fine particle mode. For chromium III and chromium VI, CAS numbers
136 and 7440473 are used for both HAPs. These two CAS numbers represent non-
speciated chromium. During emissions processing, the non-speciated chromium is
speciated to chromium III and chromium VI. For mobile sources, eighteen percent of the
chromium was assumed to be hexavalent, based on combustion data from stationary
combustion turbines that burn diesel fuel (Taylor, 2003).
• The NEI contains additional POM pollutants including unspeciated POM groups such as
7-PAH or other specific POM that are not on the MSAT list but are emitted from
stationary sources and were thus modeled as a POM group. These other POM pollutants,
with CAS in parentheses, are listed below along with the POM group that they fall in.
• POM group 71002: total PAH (234), POM (246), 16-PAH - 7-PAH (75040)1,
16-PAH (40), Benz(a)Anthracene/Chrysene (103)
• POM group 72002: Benzo(e)pyrene (192972), Perylene (198550), 2-
Methylnaphthalene (91576), Benzofluoranthenes (56832736), 2-
Chloronaphthalene (91587), Methylanthracene (26914181), Methylchrysene
(248), 12-Methylbenz(a)Anthracene (2422799), 1-Methylpyrene (2381217), 1-
Methylphenanthrene (832699), Methylbenzopyrenes (247), 9-
Methylbenz(a)Anthracene (779022), Benzo(a)fluoranthene (203338),
Benzo(g,h,i)Fluoranthene (203123), Benzo(c)phenanthrene (195197)
• POM group 73002: 7,12-Dimethylbenz(a)anthracene (57976)
• POM group 74002: Dibenzo(a,i)pyrene (189559), D(a,h)pyrene (189640), 3-
Methylcholanthrene (56495)
• POM group 75002: D(a,e)pyrene (192654), 5-Methylchrysene (3697243)
• POM group 76002: B(j)fluoranthene (205823), D(a,j)acridine (224420),
Benzo(b+k)fluoranthene (102)
• POM group 78002: 7-PAH (75)
1 See Table 2 for explanation of CAS 75040.
-------
Preceeding Page Blank
2. 1999 base inventories
2.1 1999 HAP inventories
The inventories used in development of the future mobile and stationary inventories are from the
1999 National Emissions Inventory (NET) Version 3
(http://www.epa.gov/ttn/chief/net/1999inventory.html): this is the inventory used for the 1999
NATA. The HAP emissions are provided for the following four inventory sectors: point, non-
point, onroad mobile, and nonroad mobile. Point and non-point inventories contain the
stationary source emissions, and onroad and nonroad contain the mobile emissions. For details
about each inventory see
http ://www. epa. gov/ttn/chief/net/1999inventory.html.
For the 1999 NATA, emissions, concentrations and risks are also summarized by emission
source sector: major, area & other, onroad and nonroad. The inventory sectors onroad and
nonroad map directly to the corresponding NATA source sectors. The stationary sources (point
and non-point) map as follows: point sources contain both major and area sources; non-point
sources contain area and area & other sources.
Before processing the inventories for the 1999 NATA and/or the future year projections, some
changes (or fixes) were made to the inventories before processing in EMS-HAP. Table 2 lists
these changes.
-------
Table 2. Changes made to the 1999 NEI HAP inventories prior to processing for 1999 NAT A or projections.
Inventory
Point
Non-point
Onroad
Change
Changed stack diameter for siteid 4200300899 to 0.67 ft
Corrected emissions from pounds to tons for siteid 31109-0217 for methylene chloride
Convert stack parameters from English units to metric units
Removed dashes from SCC code and convert all lowercase characters in SCC code to uppercase. Also if SCC is 0, 00000000,
"NONE", or "N/A" make SCC blank
If SIC code is "NONE" or "XXXX" make SIC blank
If Maximum Achievable Control Technology (MACT) code is "NONE" make MACT blank
For sources defaulted to county centroids (DEFAULT_LOC_FLAG='CNTYCENT) make the location coordinates equal to
missing so that EMS-HAP will default location to tracts.
If MACT code = 723 then MACT = 0723
If MACT code = 724 then MACT = 0724
Remove mercury emissions
Remove all emissions for Puerto Rico and Virgin Islands since we conducted the MS AT analysis for the 50 states
Replace 1999 MACT 1801 emissions with 2002 draft 1801 emissions
Remove all emissions for Puerto Rico and Virgin Islands since we conducted the MSAT analysis for the 50 states.
Change MACT 1801 emissions to 0 for projections.
Remove mercury emissions (note onroad and nonroad inventories contain no mercury)
For FIPS/SCC/ combinations where there were both 16-PAH emissions (CAS=40) and 7-PAH emissions (CAS=75) and the 16-
PAH emissions are larger than the 7-PAH emissions, subtract the 7-PAH emissions from the 16-PAH emissions and assign the
CAS 75040 to the emissions. For the FIPS/SCC combinations being changed delete the 16-PAH emissions but retain the 7-PAH
emissions. For FIPS/SCC combinations that have both 16-PAH and 7-PAH, but 7-PAH emissions are larger than 16-PAH
emissions, make no changes. Also make no changes where there are 7-PAH emissions but no 16-PAH and vice versa.
Remove all emissions for Puerto Rico and Virgin Islands since we conducted the MSAT analysis for the 50 states
Change Chromium III and Chromium VI CAS numbers to the unspeciated Chromium CAS (7440473). Once making the change,
sum up the chromium emissions by FIPS/SCC. The chromium was summed so that EMS-HAP would use an 82/18 chromium
Ill/chromium VI split. Before the summation, the chromium Ill/chromium VI split was not 82/18% as desired.
Reason
(NATA or
projections)
Both
Both
Both
Both
Both
Both
Both
Projections
Projections
Projections
Projections
Projections
Projections
Projections
Projections
Both
Projections
Both
-------
Table 2. Continued
Inventory
Nonroad
Change
Remove all emissions for Puerto Rico and Virgin Islands since we conducted the MS AT analysis for the 50 states
For FIPS/SCC/ combinations where there were both 16-PAH emissions (CAS=40) and 7-PAH emissions (CAS=75) and the 16-
PAH emissions are larger than the 7-PAH emissions, subtract the 7-PAH emissions from the 16-PAH emissions and assign the
CAS 75040 to the emissions. For the FIPS/SCC combinations being changed delete the 16-PAH emissions but retain the 7-PAH
emissions. For FIPS/SCC combinations that have both 16-PAH and 7-PAH, but 7-PAH emissions are larger than 16-PAH
emissions, make no changes. Also make no changes where there are 7-PAH emissions but no 16-PAH and vice versa.
Change Chromium III and Chromium VI CAS numbers to the unspeciated Chromium CAS (7440473). Once making the change,
sum up the chromium emissions by FIPS/SCC. The chromium was summed so that EMS-HAP would use an 82/18 chromium
Ill/chromium VI split. Before the summation, the chromium Ill/chromium VI split was not 82/18% as desired.
Corrected FIPS for aircraft emissions in a few counties in which we found the underlying NEI geographic data to be erroneous.
Section C.2.1 of the EMS-HAP V3 User's Guide provides details, in particular, see Table C-4.
Reason
(NATA or
projections)
Projections
Both
Both
Both
-------
Table 3 lists the inventory emissions for each of the MSAT HAPs prior to EMS-HAP processing.
Table 3. Emissions (tons) for MSAT HAPs in the 1999 NEI inventories. Totals include Puerto
Rico and the Virgin Islands.
HAP
1,3 -butadiene
2,2,4-Trimethylpentane
Acetaldehyde
Acrolein
Benzene
Ethyl Benzene
Formaldehyde
Hexane
MTBE
Naphthalene
Propionaldehyde
Styrene
Toluene
Xylenes
Chromium (total)
Manganese
Nickel
Acenaphthene
Acenaphthylene
Anthracene
Benzo(g,h,i)perylene
Fluoranthene
Fluorene
Phenanthrene
Pyrene
Benzo(a)pyrene
Dibenzo(a,h)anthracene
Benz(a)anthracene
Benzo(b)fluoranthene
Benzo(k)fluoranthene
Indeno(l,2,3,c,d)-pyrene
Chrysene
7-PAH*
16-PAH*
16-PAH - 7-PAH*
Total PAH*
Total POM*
Inventory
Point
2,068
7,211
12,141
959
12,529
12,026
35,571
46,698
4,398
2,652
1,964
40,713
93,912
62,588
846
2,844
1,304
40
1
38
2
232
60
175
399
16
1
109
5
2
0.4
30
66
13
0
55
3,315
Non-point
22,133
7,076
26,474
21,083
98,961
27,805
121,738
94,755
13,915
11,433
3,480
9,673
231,196
181,248
47
356
176
310
1,745
333
284
524
254
1,085
617
1,074
7
434
86
144
178
388
40
320
132
977
644
Onroad
23,785
167,576
30,068
4,013
171,644
70,075
81,081
65,898
82,777
4,008
4,231
13,266
460,240
269,500
21
16
16
26
139
32
9
33
55
91
46
5
0.001
8
5
5
3
4
0
0
0
0
0
Nonroad
9,923
94,546
24,099
3,183
67,642
44,137
57,520
30,063
24,338
1,257
4,833
4,319
211,095
198,748
22
6
34
26
66
15
10
30
51
101
35
3
0.07
5
3
2
3
3
1
1
0
0
0
Total
57,909
276,409
92,782
29,238
350,776
154,043
295,910
237,414
125,428
19,350
14,508
67,971
996,443
712,084
936
3,222
1,530
402
1,951
418
305
819
420
1,452
1097
1,098
8.071
556
99
153
184.4
425
107
334
132
1032
3,959
*Some portion of these could be MSAT HAPs but are not sufficiently speciated in the inventory to determine what
portion is MSAT POM HAP
One change made to modeled concentrations for NATA (and this effort) after EMS-HAP and
ASPEN was to the POM group 75002 concentrations for Oregon for area & other sources. The
area & other emissions for benzo(a)pyrene were incorrect in the 1999 NEI for Oregon. In order
to alleviate the problem, the national median area & other concentration (excluding Oregon) was
substituted for Oregon's area & other tract level concentrations.
-------
2.2 1999 Precursor inventories
In order to calculate secondary concentrations for acetaldehyde, acrolein, formaldehyde, and
propionaldehyde after ASPEN simulations for the primary concentrations, the emissions for the
precursors also had to be processed through EMS-HAP and subsequently ASPEN for later
secondary contribution calculations. For those precursors that are not HAPs themselves (non-
HAP precursors) a separate precursor inventory was used. The precursor inventory used was the
same as that used for the 1999 NATA and is Version 2 of the NEI for VOC. Precursor emissions
were obtained by speciating VOC emissions from Version 2 of the NEI. The speciation profiles
are the same as those used for the 1996 NATA (see Section D.I.2 in EMS-HAP Version 2 User's
Guide, [U.S. EPA, 2002]). All point sources in the precursor inventory were treated as major
sources. Table 4 lists the non-HAP precursors for the four secondary HAPs being modeled in
this study.
-------
Table 4. Non-HAP precursors for the MSAT secondary HAPs with source sector emissions for
1999. Totals include Puerto Rico and the Virgin Islands.
Precursor
1-Butene
1-2,3-Dimethyl butene
1-2-Ethylbutene
1-2-Methyl butene
1-3-Methyl butene
2-Butene
2-2-Methyl butene
1-Decene
Ethanol
Ethene
1-Heptene
2-Heptene
1-Hexene
2-Hexene
3-Hexene
Isoprene
1-Nonene
2-Nonene
1-Octene
2-Octene
1-Pentene
1-2,4,4-Trimethyl
1-2-Methyl pentene
1-3-Methyl pentene
1-4-Methyl pentene
2-Pentene
2-3-Methyl pentene
2-4-Methyl pentene
Propene
2-Methyl propene
Precursor for
Formaldehyde,
Formaldehyde
Formaldehyde
Formaldehyde
Formaldehyde
Acetaldehyde
Acetaldehyde
Formaldehyde
Acetaldehyde
Formaldehyde
Formaldehyde
Acetaldehyde
Formaldehyde
Acetaldehyde
Propionaldehy de
Formaldehyde
Formaldehyde
Acetaldehyde
Formaldehyde
Acetaldehyde
Formaldehyde
Formaldehyde
Formaldehyde
Formaldehyde
Formaldehyde
Acetaldehyde, Propionaldehyde
Acetaldehyde
Acetaldehyde
Acetaldehyde, Formaldehyde
Formaldehyde
Inventory
Point
5,337
110
0.0005
104
81
2,170
78
762
43,907
24,481
602
579
1,248
94
516
316
56
2
37
18
2,693
27
124
87
96
2,699
66
132
15,705
602
Non-point
24,245
3,004
0
4,750
7,527
7,503
4,750
0
201,811
349,992
177
177
1,332
1,332
1,332
401
301
0
0.3
0.3
18,199
0
1,784
1,782
1,782
4,550
1,782
1,782
64,853
8,896
On road
41,481
1,902
0
30,647
4,625
42,270
89,651
0
22,820
411,045
1,524
2,822
15,498
13,399
5,370
6,280
35,806
0
1,961
1,961
32,313
21,064
37,937
55,197
6,072
70,835
14,328
37,937
171,468
93,943
Nonroad
12,503
0
255
11,898
2,947
7,521
19,722
166
330
165,071
1,805
2,196
6,507
10,077
4,370
5,289
7,082
82
1,828
837
8,959
0
4,780
4,127
2,275
22,234
10,916
8,950
66,043
19,171
Total
83,566
5,016
255.0005
47,399
15,180
59,464
114,201
928
268,868
950,589
4,108
5,774
24,585
24,902
11,588
12,286
43,245
84
3,826.3
2,816.3
62,164
21,091
44,625
61,193
10,225
100,318
27,092
48,801
318,069
122,612
10
-------
3. Development of Future Year Mobile and Mobile-Related Emissions
3.1 Locomotive and commercial marine vessels
Emissions from locomotive and commercial marine vessels were projected similarly; using ratios
computed from previously projected, multi-year, national-level, criteria pollutant emission data.
Because these previously projected emissions account for both activity growth and reductions
due to control programs, the term "projection factor" is used rather than "growth factor" to
describe the factor used to multiply base year emissions to obtain future year emissions. Table 5
shows the eight locomotive SCC codes in the 1999 NEI.
Table 5. Locomotive SCC codes in the 1999 NEI nonroad inventory.
SCC
2285000000
2285002000
2285002005
2285002006
2285002007
2285002008
2285002009
2285002010
Description
Mobile Sources, Railroad Equipment, All Fuels, Total
Mobile Sources, Railroad Equipment, Diesel, Total
Mobile Sources, Railroad Equipment, Diesel, Line Haul Locomotives
Mobile Sources, Railroad Equipment, Diesel, Line Haul Locomotives:
Mobile Sources, Railroad Equipment, Diesel, Line Haul Locomotives:
Mobile Sources, Railroad Equipment, Diesel, Line Haul Locomotives:
(Amtrak)
Mobile Sources, Railroad Equipment, Diesel, Line Haul Locomotives:
Class I operations
Class II/III operations
Passenger Trains
Commuter lines
Mobile Sources, Railroad Equipment, Diesel, Yard Locomotives
Projection factors for the locomotive emissions, which account for both growth and reductions
due to control programs, were developed from the VOC and PM10 projected emissions shown in
Table 6. These were derived as part of the EPA's 2004 Clean Air Nonroad Diesel Rule (U.S.
EPA, 2004d).
11
-------
Table 6. Locomotive 50-State annual emissions trends (tons per year) and future year ratios
Year
1999
2007
2010
2015
2020
2030
voc
emissions
34,579
32,646
31,559
31,072
30,170
28,622
VOC ratio
(future year/1999)
1.0000
0.9441
0.9127
0.8986
0.8725
0.8277
PM10
emissions
20,869
17,657
15,109
14,461
13,652
12,061
PM10 ratio
(future year/1999)
1.0000
0.8461
0.7240
0.6929
0.6542
0.5779
In that they were computed using 50-state total emission sums, the projection factors were
national level. In addition, they were applied to each pollutant across all SCC codes. That is, all
locomotive SCC codes with pollutants deemed VOC received the same projection factor. The
pollutants associated with locomotive emissions are shown in Table 7 with their assigned
projection factor for locomotives.
12
-------
Table 7. Locomotive HAPs. HAPs not in bold are not emphasized in the MSAT study but are
projected.
HAP
1,3-Butadiene
2,2,4-
Trimethylpentane
Acet aldehyde
Acrolein
Antimony
Benzene
Beryllium
Cadmium
Chlorine
Chromium
Cobalt
Cumene
Ethyl benzene
Formaldehyde
Growth factor basis
VOC ratios
VOC ratios
VOC ratios
VOC ratios
Metals (projection factors =1.0)
VOC ratios
Metals (projection factors = 1.0)
Metals (projection factors =1.0)
VOC ratios
Metals (projection factors = 1.0)
Metals (projection factors = 1.0)
VOC ratios
VOC ratios
VOC ratios
HAP
Hexane
Lead
Manganese
Methanol
Methyl ethyl
ketone
Naphthalene
Nickel
POM (excluding
Naphthalene)
Phosphorus
Propionaldehyde
Selenium
Styrene
Toluene
Xylene
Growth factor basis
VOC ratios
Metals (projection factors = 1.0)
Metals (growth factors =1.0)
VOC ratios
VOC ratios
PM ratios
Metals (projection factors = 1.0)
PM ratios
PM ratios
VOC ratios
Metals (projection factors =1.0)
VOC ratios
VOC ratios
VOC ratios
Metals were set to no growth (projection factor = 1.0, metals remain at 1999 levels) because little
activity change is expected in locomotives in the future. This is because metal emissions were
most likely the result of impurities in fuel and engine oil, and from engine wear, and it is not
known how these emissions would be impacted by controls, if it all. Several of the metals were
estimated using emission factor (EF) x Activity, and several were estimated as fractions of PM
emissions.
Projection factors for commercial marine vessels (CMV) were computed similarly to locomotive
projection factors, using 50-state emission summaries for various future years that were
developed as part of the EPA's 2004 Clean Air Nonroad Diesel Rule (U.S. EPA, 2004d). These
emissions summaries are shown in Table 8.
One difference, however, is that the projection factors for CMV were specific to both SCC
(diesel, residual, or no fuel information) and pollutant specific (VOC or PM). The SCC
dependence on the projection factor was based on whether the SCC was related to diesel
emissions or residual emissions. There were three SCC codes used to assign the basis of the
13
-------
projection factor for the SCC. Within the SCC, the projection factor used was dependent on
whether the HAP was VOC or PM. Table 8 lists the projection factors computed from the 50-
state total emission summaries commercial marine vessels.
Projection factors computed for the SCC codes in Table 8 were assigned to the five SCC codes
corresponding to the commercial marine vessels in the 1999 NEI. Each HAP within the SCC
category was then assigned the projection factor for VOC or PM. Table 9 lists the SCC codes
and HAPs associated with the commercial marine vessel emissions in the 1999 NEI.
Table 8. Commercial marine vessel 50-State annual emissions trends (tons per year) and future
year ratios used as projection factors
SCC
2280000000
2280002000
iicnmnnn
zzoUUJUUU
Description
Mobile Sources,
Marine Vessels,
Commercial, All
Fuels, Total, All
Vessel Types
Mobile Sources,
Marine Vessels,
Commercial, Diesel,
Total, All Vessel
Types
Mobile Sources,
RpsiHiial Tntal All
Vessel Types
Year
1999
2007
2010
2015
2020
2030
1999
2007
2010
2015
2020
2030
1999
2007
2010
2015
2020
2030
VOC
emissions
32,133
35,951
36,990
39,543
43,395
55,083
23,403
24,530
24,568
24,695
25,268
27,546
8,730
11,421
12,421
14,848
18,127
27,537
VOC ratio
(year/ 1999)
.0000
.1188
.1511
.2306
.3505
.7142
.0000
.0482
.0498
.0552
.0797
.7700
.0000
.3083
.4229
.7009
2.0765
3.1544
PM10
emissions
39,012
43,247
43,717
47,456
53,496
72,489
19,927
19,133
17,721
16,900
16,795
18,258
19,085
24,115
25,996
30,556
36,701
54,231
PM10 ratio
(year/ 1999)
.0000
.1086
.1206
.2165
.3713
.8581
.0000
0.9601
0.8893
0.8481
0.8428
.9162
.0000
.2635
.3621
.6011
.9230
2.8416
14
-------
Table 9. Commercial marine vessel SCC codes, HAPs, and basis of projection factors. HAPs in
bold are emphasized in the MSAT study.
SCC
2280000000
2280002100
2280002200
2280003100
2280003200
Description
Mobile Sources, Marine
Vessels, Commercial, All
Fuels, Total, All Vessel Types
Mobile Sources, Marine
Vessels, Commercial, Diesel,
Diesel- port emissions
Mobile Sources, Marine
Vessels, Commercial, Diesel,
Diesel- underway emissions
Mobile Sources, Marine
Vessels, Commercial,
Residual, Residual - port
emissions
Mobile Sources, Marine
Vessels, Commercial,
Residual, Residual -underway
emissions
VOC and PM HAPs
1,3-Butadiene, 2,2,4-
Trimethylpentane, Acetaldehyde,
Benzene, Chlorine, Cumene, Ethyl
Benzene, Formaldehyde, Hexane,
Methanol, Methyl Ethyl Ketone,
Propionaldehyde, Styrene,
Toluene, Xylenes
Antimony, Cadmium, Chromium,
Cobalt, Lead, Manganese,
Naphthalene, Nickel, Phosphorus,
POM, Selenium
1,3-Butadiene, 2,2,4-
Trimethylpentane, Acetaldehyde,
Acrolein, Benzene, Chlorine,
Cumene, Ethyl Benzene,
Formaldehyde, Hexane, Methanol,
Methyl Ethyl Ketone,
Propionaldehyde, Styrene,
Toluene, Xylenes
Antimony, Cadmium, Chromium,
Cobalt, Lead, Manganese,
Naphthalene, Nickel, Phosphorus,
POM, Selenium
2,2,4-Trimethylpentane,
Acetaldehyde, Acrolein, Benzene,
Ethyl Benzene, Formaldehyde,
Hexane, Propionaldehyde,
Styrene, Toluene, Xylenes
Chromium, Lead, Manganese,
Naphthalene, Nickel, POM
2,2,4-Trimethylpentane,
Acetaldehyde, Acrolein, Benzene,
Chlorobenzene, Ethyl Benzene,
Formaldehyde, Hexane,
Propionaldehyde, Styrene,
Toluene, Xylenes
Beryllium, Cadmium, Chromium,
Lead, Manganese, Naphthalene,
Nickel, POM, Selenium
2,2,4-Trimethylpentane,
Acetaldehyde, Acrolein, Benzene,
Ethyl Benzene, Formaldehyde,
Hexane, Propionaldehyde,
Styrene, Toluene, Xylenes
Beryllium, Cadmium, Chromium,
Lead, Manganese, Naphthalene,
Nickel, POM, Selenium
Projection factor basis
VOC ratios for
2280000000 (all fuels)
PM ratios for
2280000000 (all fuels)
VOC ratios for
2280002000 (diesel)
PM ratios for
2280002000 (diesel)
VOC ratios for
2280002000 (diesel)
PM ratios for
2280002000 (diesel)
VOC ratios for
2280003000 (residual)
PM ratios for
2280003000 (residual)
VOC ratios for
2280003000 (residual)
PM ratios for
2280003000 (residual)
15
-------
As can be seen from Tables 7 and 9, there were other HAPs being projected other than those of
interest for the MSAT study. As mentioned previously, these were HAPs found only in data
submitted by States or in surrogate profiles for vessels running on residual fuel. These HAPs
were not removed from the inventories for MSAT because these HAPs would need to be
projected for other projections work; by not removing pollutants there were fewer datasets to
manage.
Appendix C (C. 1.1) describes the steps involved in the development of the projection factor files
used for the locomotives and commercial marine vessels emission projections and (C.I.2)
the steps taken to apply the factors and produce projected emissions.
Tables 10 and 11 present the nationwide 1999 and projected emissions for locomotives and
commercial marine vessels, respectively. "All HAPs" refer to the sum of MSAT and non-MSAT
HAPs.
16
-------
Table 10. National locomotive emissions (rounded) by SCC for selected HAPs and across all HAPs.
SCC
2285000000
2285002000
2285002005
2285002006
2285002007
2285002008
2285002009
2285002010
Total
Locomotive
HAP
Acrolein
All HAPs
Acetaldehyde
Acrolein
Formaldehyde
All HAPs
1,3-Butadiene
Acetaldehyde
Benzene
Formaldehyde
Naphthalene
All HAPs
1,3-Butadiene
Acetaldehyde
Acrolein
Benzene
Formaldehyde
Naphthalene
All HAPs
1,3-Butadiene
Acetaldehyde
Acrolein
Benzene
Formaldehyde
Naphthalene
All HAPs
1,3-Butadiene
Acetaldehyde
Acrolein
Benzene
Formaldehyde
Naphthalene
All HAPs
1,3-Butadiene
Acetaldehyde
Acrolein
Benzene
Formaldehyde
Naphthalene
All HAPs
1,3-Butadiene
Acetaldehyde
Acrolein
Benzene
Formaldehyde
Naphthalene
All HAPs
1,3-Butadiene
Acetaldehyde
Acrolein
Benzene
Formaldehyde
Naphthalene
All HAPs
Emissions (tons/yr)
1999
24
151
1
0.3
2
5
5
174
47
349
2
717
90
519
87
71
1,196
48
2,709
6
35
6
5
80
3
184
2
11
2
2
26
1
59
2
10
1
1
22
1
49
8
66
7
12
143
5
371
111
815
128
139
1,817
60
4,246
2007
23
143
1
0.3
2
5
4
165
45
330
2
677
85
490
83
67
1,129
40
2,550
6
33
6
5
75
3
173
2
11
2
1
24
1
55
2
9
1
1
21
1
46
7
62
6
12
135
5
350
105
770
121
131
1,716
51
4,000
2010
22
138
1
0.3
2
5
4
159
43
319
2
654
82
474
80
65
1,091
35
2,458
5
32
6
4
73
2
167
2
10
2
1
24
1
53
2
9
1
1
20
1
45
7
60
6
11
131
4
337
102
744
117
127
1,659
44
3,858
2015
22
136
1
0.3
2
5
4
157
43
314
1
644
81
466
79
64
1,074
33
2,419
5
31
6
4
72
2
164
2
10
2
1
23
1
53
2
9
1
1
20
1
44
7
59
6
11
129
4
332
100
732
115
125
1,634
42
3,796
2020
21
132
1
0.3
2
5
4
152
41
305
1
625
78
453
76
62
1,043
31
2,347
5
30
5
4
70
2
160
2
10
2
1
23
1
51
1
8
1
1
20
1
43
7
57
6
11
125
3
322
97
711
112
121
1,586
39
3,684
2030
20
125
1
0.3
1
4
4
144
39
289
1
593
74
429
72
59
990
28
2,223
5
29
5
4
66
2
151
2
9
2
1
21
1
48
1
8
1
1
19
1
40
6
54
5
10
118
3
305
92
675
106
115
1,505
35
3,491
17
-------
Table 11. National commercial marine vessel emissions (rounded) by SCC for selected HAPs
and across all HAPs.
SCC
2280000000
2280002100
2280002200
2280003100
2280003200
Total CMV
HAP
1,3 -Butadiene
Acetaldehyde
Benzene
Formaldehyde
Naphthalene
All HAPs
1,3 -Butadiene
Acetaldehyde
Acrolein
Benzene
Formaldehyde
Naphthalene
All HAPs
Acetaldehyde
Acrolein
Benzene
Formaldehyde
Naphthalene
All HAPs
Acetaldehyde
Acrolein
Benzene
Formaldehyde
Naphthalene
All HAPs
Acetaldehyde
Acrolein
Benzene
Formaldehyde
Naphthalene
All HAPs
1,3 -Butadiene
Acetaldehyde
Acrolein
Benzene
Formaldehyde
Naphthalene
All HAPs
Emissions (tons/yr)
1999
0.1
5
1
10
0.1
21
6
1,478
54
404
2,973
33
5,478
459
22
126
925
11
1,694
300
17
80
561
16
1,126
122
6
33
247
6
466
6
2364
98
644
4715
65
8,786
2007
0.2
6
2
12
0.1
24
6
1,549
56
424
3,116
31
5,736
481
23
132
969
10
1,774
392
23
104
733
20
1,467
160
8
44
323
7
607
6
2588
109
705
5153
69
9,609
2010
0.2
6
2
12
0.1
25
6
1,551
56
424
3,121
29
5,740
482
23
132
971
10
1,776
426
24
113
798
22
1,594
174
8
48
351
8
660
6
2639
112
719
5252
68
9,795
2015
0.2
6
2
13
0.1
26
6
1,559
57
427
3,136
28
5,766
485
23
133
976
9
1,785
510
29
135
954
26
1,901
208
10
57
420
9
787
6
2768
118
753
5499
72
10,266
2020
0.2
7
2
14
0.1
29
6
1,595
58
437
3,210
27
5,898
496
23
136
998
9
1,826
622
36
165
1,164
31
2,316
254
12
69
513
11
959
6
2974
129
809
5899
79
11,028
2030
0.2
9
2
18
0.1
37
7
1,739
63
476
3,499
30
6,430
540
25
148
1,088
10
1,990
945
54
251
1,768
46
3,506
386
18
105
779
16
1,451
7
3619
161
982
7152
102
13,414
18
-------
3.2 Aircraft and Aviation gasoline
Aircraft emissions were projected by using growth factors based on activity growth. These
growth factors were also used to project aviation gasoline source categories that are inventoried
in the NEI as stationary sources. Note that the projection of airport support equipment source
categories did not use this approach; they were projected using the National Mobile Inventory
Model (NMIM) as described in Section 3.3.3.
Aircraft growth factors were developed using data on itinerant (landing and take-off) operations
from the Terminal Area Forecast System (TAP) (FAA, 2004), http://www.apo.data.faa.gov/.
These data were accessed from the website in February 2004.
The TAP model provides itinerant activity for commercial aircraft, general aviation, air taxis,
and military aircraft. The four categories map directly to inventory categories for aircraft
emissions. We used the growth factors for general aviation for aviation gasoline emissions since
most aircraft gasoline is used with general aviation aircraft. Although the TAF model provides
activity at individual airports, the TAF data were summed to create growth factors at the national
level. This was done to smooth out the large-scale year-to-year changes in individual airport
itinerant data that were questionable. The same approach was used in the modeling for the Clean
Air Interstate (CAIR) rule (EPA, 2005b). Table 12 provides the nationally aggregated TAF
intinerant data for 2002-2010, inclusive, 2015 and 2020. Note that the "all operations" data is
simply the sum of commercial aircraft, air taxi, general aviation, and military operations.
Table 12. TAF landing and take-off data for 2002 through 2020, 2015, and 2020.
Year
1999
2000
2001
2002
2003
2004
2005
2006
2007
2008
2009
2010
2011
2012
2013
2014
2015
2020
Commercial
14,769,055
15,239,632
14,806,610
13,823,811
12,877,845
13,144,840
13,587,336
13,931,296
14,244,149
14,559,169
14,875,408
15,199,253
15,516,407
15,837,083
16,167,397
16,503,853
16,844,216
18,584,876
Air Taxi
14,177,496
14,474,434
14,559,287
13,976,494
15,192,330
15,812,801
16,177,234
16,478,077
16,754,880
17,022,228
17,292,401
17,566,653
17,839,971
18,113,540
18,388,892
18,666,510
18,943,304
20,347,985
General
Aviation
44,413,777
45,075,619
44,199,554
44,096,192
42,946,739
43,302,577
43,672,459
44,041,238
44,409,867
44,778,567
45,147,517
45,516,733
45,886,238
46,255,950
46,626,125
46,996,330
47366,817
49,223,017
Military
3,977,646
4,023,672
4,145,459
4,168,042
4,141,986
4,148,717
4,155,567
4,155,999
4,156,432
4,156,864
4,157,297
4,157,730
4,158,162
4,158,595
4,159,028
4,159,460
4,159,893
4,162,058
All
Operations
77,337,974
78,813,357
77,710,910
76,064,539
75,158,900
76,408,935
77,592,596
78,606,610
79,565,328
80,516,828
81,472,623
82,440,369
83,400,778
84,365,168
85,341,442
86,326,153
87,314,230
92,317,936
19
-------
Growth factors were computed for 2002-2010, inclusive, 2015 and 2020 by dividing each year's
TAP data by the TAP data for 1999. The TAP data did not cover 2030; growth factor for 2030
was calculated by using the same rate of growth between 2015 and 2020 and extrapolating to
2030 using Equation 1:
GF2030 = GF2020 + ((2030- 2020) x (GF2020 - GF2015) + (2020- 2015))
where GF is the growth factor for the respective years.
(1)
The growth factors for the MS AT study years are shown in Table 13. The growth factor
assignments for each of the airport related SCC codes are shown in Table 14.
Table 13. Aircraft growth factors for MSAT study years.
Aviation type
Commercial
Air Taxi
General
Military
All operations
Growth Factors
2007
0.9645
1.1818
0.9999
1.0449
1.0288
2010
1.0291
1.2391
1.0248
1.0453
1.0660
2015
1.1405
1.3362
1.0665
1.0458
1.1290
2020
1.2584
1.4352
1.1083
1.0464
1.1937
2030
1.4941
1.6334
1.1919
1.0475
1.3231
20
-------
Table 14. Airport related SCC codes and assigned growth factor basis.
SCC
2265008000
2265008005
2267008000
2267008005
2268008000
2270008000
2270008005
2275000000
2275060000
2275020000
2275070000
2275050000
2275900000
2501080000*
2501080050*
2501080100*
2275001000
Description
Airport Support Equipment, Total, Off-highway 4-
stroke
Airport Support Equipment, Off-highway 4-stroke
Airport Ground Support Equipment, All, LPG
Airport Ground Support Equipment, LPG
Airport Ground Support Equipment, CNG, All
Airport Service Equipment, Total, Off-highway
Diesel
Airport Service Equipment, Airport Support
Equipment, Off-highway Diesel
All Aircraft Types and Operations
Air Taxi, Total
Commercial Aircraft, Total
Aircraft Auxiliary Power Units, Total
General Aircraft, Total
Aircraft Refueling: All Fuels, All Processes
Aviation Gasoline Distribution: Stage 1 & II
Aviation Gasoline Storage -Stage I
Aviation Gasoline Storage -Stage II
Military Aircraft, Total
Growth factor basis
No factor. Projected emissions in NMIM
(see 3.3. 3)
No factor. Projected emissions in NMIM
(see 3.3. 3)
No factor. Projected emissions in NMIM
(see 3.3. 3)
No factor. Projected emissions in NMIM
(see 3.3. 3)
No factor. Projected emissions in NMIM
(see 3.3. 3)
No factor. Projected emissions in NMIM
(see 3.3. 3)
No factor. Projected emissions in NMIM
(see 3.3. 3)
Growth factor for All operations
Growth factor for Air Taxi
Growth factor for Commercial Aviation
Growth factor for Commercial Aviation
Growth factor for General Aviation
Growth factor for General Aviation
Growth factor for General Aviation
Growth factor for General Aviation
Growth factor for General Aviation
Growth factor for Military Aviation
# Stationary sources in the non-point inventory. All others are nonroad sources.
Growth factor files were created for each year, 2002-2010 inclusive, 2015, 2020, and 2030, using
the SCC growth factor file format for EMS-HAP Version 3.0 described in Appendix B of the
EMS-HAP Version 3.0 User's Guide (U.S. EPA, 2004b). For this format, each SCC was
assigned a code describing its growth method, basically the "growth factor basis" column in
Table 14. The format of the file is shown in Figure 1. The naming convention of the aircraft and
aviation gasoline growth factor files is gf99scca_XX.txt where XX is the two-digit year for 2002
through 2010 inclusive, 2015, 2020, and 2030.
21
-------
1999 Base Year EGAS SCC Growth Factors for 2010, Created 12APRIL04 BEGIN SCC-REMIXREF on line 3.
GROWTH FACTORS BEGIN ON LINE 18.
2265008000 N/A projected emissions will be supplied with NMIM
2265008005 N/A projected emissions will be supplied with NMIM
2267008000 N/A projected emissions will be supplied with NMIM
2267008005 N/A projected emissions will be supplied with NMIM
2268008000 N/A projected emissions will be supplied with NMIM
2270008000 N/A projected emissions will be supplied with NMIM
2270008005 N/A projected emissions will be supplied with NMIM
2275000000 TAP for ALL OPERATIONS (p_tot)
2275060000 TAP for Air Taxi
2275020000 TAP for Commercial Aviation
2275070000 TAP for Commercial Aviation
2275050000 TAP for General Aviation
2275900000 TAP for General Aviation
2501080000 TAP for General Aviation
2501080050 TAP for General Aviation
2501080100 TAP for General Aviation
2275001000 TAP for Military Aviation
00 000 1.0000 N/A projected emissions will be supplied with NMIM
00 000 1.0660 TAP for ALL OPERATIONS (p_tot)
00 000 1.2391 TAP for Air Taxi
00 000 1.0291 TAP for Commercial Aviation
00 000 1.0248 TAP for General Aviation
00 OOP 1.0453 TAP for Military Aviation
Figure 1. Format of the aircraft growth factor file. 2010 growth factors shown as example.
EMS-HAP V3 was used to apply the growth factors to the aircraft and aviation gasoline sources.
Aircraft emissions were projected by first subserting the nonroad airport-related emissions to
exclude the airport support equipment emissions, which were projected using NMIM future
emissions data as described in Section 3.3.3. The subserting of the data was done on the
temporally allocated 1999 NEI emissions for NATA (National Air Toxics Assessment). These
emissions had previously been processed through the appropriate EMS-HAP programs, COP AX,
PtDataProc, PtModelProc, and PtTemporal for the 1999 NATA (see EMS-HAP User's Guide for
details, (U.S. EPA, 2004b)). After the subsetting was completed, the emissions were processed
through the EMS-HAP program PtGrowCntl for 2002 through 2010, 2015, 2020, and 2030,
using the TAF-derived growth factors described above.
Aviation gasoline emissions (SCCs shown in Table 14 with # footnotes) that had been processed
through the appropriate EMS-HAP programs for the 1999 NATA, were projected using the
EMS-HAP program PtGrowCntl for 2002 through 2010, 2015 and 2020, using the TAF-derived
growth factors described above. Aviation gasoline emissions were projected to 2030 but it was
decided to use 2020 projected emissions for 2030 for all stationary sources because of
uncertainty in the 2030 projection and growth factors.
A flowchart of the projection processing is shown in Figure 2.
22
-------
Airports (nonroad)
PtTemporal output
(temporal)
| Subset inventory not include airport support I
| equipment (SCO: 2265008000, I
I 2265008005,2267008000,2267008005, j
2268008000, 2270008000, 2270008005)
Temporal_no_supp
I 1
1 PtGrowCntl |
Airports (nonpoint)
PtTemporal output
(temporal)
Figure 2. Flowchart of aircraft and aviation gasoline emissions projections.
Projected airport related emissions (excluding airport support equipment) by SCC are shown in
Table 15.
23
-------
Table 15. Airport related emissions (excluding airport support equipment) for selected HAPs
and all HAPs by SCC. Non-point SCC emissions for 2030 are set equal to 2020.
see
2275000000
2275001000
2275020000
2275050000
2275060000
2501080000
2501080050
2501080100
Total
HAP
1,3-Butadiene
Acetaldehyde
Acrolein
Benzene
Formaldehyde
Naphthalene
All HAPs
1,3-Butadiene
Acetaldehyde
Acrolein
Benzene
Formaldehyde
Naphthalene
All HAPs
1,3-Butadiene
Acetaldehyde
Acrolein
Benzene
Formaldehyde
Naphthalene
All HAPs
1,3-Butadiene
Acetaldehyde
Acrolein
Benzene
Formaldehyde
Naphthalene
All HAPs
1,3-Butadiene
Acetaldehyde
Acrolein
Benzene
Formaldehyde
Benzene
All HAPs
Benzene
Naphthalene
All HAPs
Benzene
Naphthalene
All HAPs
Benzene
Naphthalene
All HAPs
1,3-Butadiene
Acetaldehyde
Acrolein
Benzene
Formaldehyde
Naphthalene
All HAPs
Emissions (tons/yr)
1999
6
12
6
10
40
1
116
193
506
247
213
1,635
62
3,171
525
1,357
662
574
4,383
164
8,455
60
85
33
186
285
109
2,053
40
58
22
119
206
121
1,077
0.04
0.01
1
287
16
1,676
20
1
117
824
2,019
968
1,409
6,549
473
16,666
2007
6
12
6
10
41
1
119
202
529
258
223
1,708
64
3,313
506
1,309
638
554
4,227
158
8,155
60
85
33
186
285
109
2,053
47
68
26
141
243
143
1,272
0.04
0.01
1
287
16
1,676
20
1
117
821
2,004
960
1,420
6,505
492
16,706
2010
6
13
6
10
43
1
123
202
529
258
223
1,709
64
3,315
540
1,397
681
591
4,510
169
8,701
61
87
34
191
293
111
2,104
49
72
27
147
255
150
1,334
0.04
0.01
1
294
16
1,718
20
1
119
859
2,098
1,005
1,477
6,809
513
17,415
2015
6
14
6
11
45
1
131
202
529
258
223
1,710
64
3,316
599
1,548
754
655
4,998
187
9,643
64
91
35
199
304
116
2,190
53
77
29
159
275
162
1,439
0.04
0.01
1
306
17
1,788
21
1
124
924
2,259
1,083
1,574
7,333
548
18,632
2020
7
14
7
12
48
2
138
202
530
258
223
1,711
64
3,318
661
1,708
832
722
5,515
206
10,640
66
95
36
206
316
120
2,275
57
83
31
171
296
174
1,545
0.05
0.01
1
318
18
1,858
22
1
129
993
2,430
1,165
1,674
7,885
585
19,904
2030
7
16
7
13
53
2
153
202
530
258
223
1,712
64
3,322
784
2028
988
858
6548
245
12,633
71
102
39
222
340
129
2,447
65
94
35
194
336
198
1,758
0.05
0.01
1
318
18
1,858
22
1
129
1,131
2,770
1,329
1,850
8,990
657
22,301
24
-------
3.3 Projection of onroad and nonroad categories using NMIM
3.3.1 Description of NMIM
For all mobile source categories except commercial marine vessels, locomotives, and aircraft
(Sections 3.1 and 3.2), EPA's Office of Transportation and Air Quality's (OTAQ) new emissions
inventory modeling system for highway and nonroad sources, the National Mobile Inventory
Model (NMIM) (Michaels et al. 2005; Cook et al. 2004) was used to generate emission data for
projections. NMIM develops county level inventories using MOBILE6.2, NONROAD, and
model inputs stored in data files. The version of NMIM used in this assessment includes
NONROAD2004, which was also used in the recent Clean Air Nonroad Diesel Rule (U.S. EPA,
2004d). More details on the inputs and data files used in the modeling can be found in Appendix
A. In addition to criteria pollutants, NMEVI can currently produce 13 gaseous hydrocarbons, 16
poly cyclic aromatic hydrocarbons, 4 metal compounds and 17 dioxin and furan congeners, for
any calendar year 1999 through 2050.
Future year MOBILE6.2 and NONROAD inputs include future year vehicle miles traveled
(VMT) and fuel parameters, and future year equipment populations. Future year VMT for years
2010, 2020 and 2030 were developed at the county-level using data from the Energy Information
Administration's National Energy Modeling System (NEMS) Transportation Model and
Regional Economic Models Inc. population growth (Mullen and Neumann, 2004). VMT for
intermediate years were interpolated, using 1999 as the base year. This same approach and
projected VMT were used for the CAIR rule. Projection year fuel parameters were developed
using results of several refinery modeling analyses conducted to assess impacts of fuel control
programs on fuel properties (MathPro, 1998; 1999a, 1999b). The projection year fuel parameters
were calculated by applying adjustment factors to the base year parameters (Eastern Research
Group, 2003). In addition, NMIM uses monthly rather than seasonal fuel parameters, and
parameters for spring and fall months were estimated by interpolating from summer and winter
data. Documentation of the fuel parameters used in NMIM was compiled in 2003 (Eastern
Research Group, 2003) and then subsequently, a number of changes were made, based on
comments from States. These changes are documented in the change log for NMIM, dated May
14, 2004. This change log is included in the docket for this rule (EPA-HQ-OAR-2005-0036),
along with the original documentation. In general, multiplicative adjustment factors were used
to calculate future year gasoline parameters (i.e., future year parameter = base year parameter x
adjustment factor). However, additive adjustment factors were used to calculate future year
parameters for E200, E300, and oxygenate market shares (i.e., future year parameter = base year
parameter + adjustment factor). The database used for this assessment assumes no Federal ban
on MTBE, but does include State bans. Also, it did not include the renewable fuels mandate in
the recent Energy Policy Act.
3.3.2 Onroad projections using NMIM
The 1999 NEI, which contains some State reported data for California and Texas, served as the
base year inventory for the emission projections. In order to preserve the State reported data, it
25
-------
was decided to compute projection factors from NMIM output for 1999 and each future year of
interest.
Equation 2 shows the computation of the projection factor
where PF20xx is the projection factor for 2007, 2010, 2015, 2020, or 2030, E20xx is the emissions
for the corresponding year and £1999 is the 1999 emissions.
Under this approach, the projection factor is computed at the detailed inventory level, for each
FIPS/SCC/CAS combination where 1999 emissions are nonzero. The FIPS represents the
specific state and county of the emission; the SCC is the source category code and the CAS is the
particular HAP emitted.
Before calculating the projection factors, the NMIM emissions for each year had to be summed
by FIPS/SCC/CAS to remove the emissions type from the NMEVI SCC. This was because the
NMEVI onroad emissions SCCs were broken out by exhaust and evaporative (non refueling)
emissions for several of the HAPs. The NEI emissions were not broken out into exhaust and
evaporative emissions. Once 1999 and future years' NMIM results were summed by
FIPS/SCC/CAS for each year, then the projection factors were calculated using Equation 2.
The projection factors were then applied to the same FIPS/SCC/CAS combinations in the 1999
NEI. Only combinations in the 1999 NEI were kept. However, before the projection factors
could be applied, the NMIM output needed to be processed because for some situations, the
NMEVI FIPS/SCC/CAS combinations did not match the NEI FIPS/SCC/CAS combinations. The
bullets below describe the necessary processing. More details on the programming steps and
example calculations are provided in Appendix B (B.I).
• Since the NEI contained the CAS for chromium, 7440473, the NMEVI chromium III and
chromium VI emissions were summed for each FIPS/SCC to give a total chromium
number. New projection factors were calculated for the summed chromium and the CAS
7440473 was assigned to each record. This was done for all FIPS/SCC combinations
with chromium III or chromium VI in NMIM.
• For xylenes, manganese, and nickel, NMEVI results and projection factors were assigned
to the CAS associated with total xylenes (1330207) and elemental manganese (7439965)
and nickel (7440020). The 1999 NEI used CAS numbers 106423, 108383, and 95476
for p-xylene, m-xylene, and o-xylene respectively. In addition to 7439965 and 7440020
for manganese and nickel, the NEI also reported emissions using 198 as a manganese
CAS number and 226 as a nickel CAS number. NMEVI xylenes, manganese, and nickel
observations were copied to observations with the same FIPS/SCC and emissions but
26
-------
replacing the CAS numbers with one of the other xylenes, manganese, or nickel CAS
numbers while still retaining the original NMIM emissions.
• The 1999 NEI contained emissions for SCC codes 2230070YYY where YYY is the 3-
digit road type descriptor (110, 130, 150, 170, 190, 210, 230, 250, 270, 290, 310, and
330). These SCC codes were for heavy duty diesel vehicles (HDDV). In NMIM, there
were no SCC codes with 2230070 as the first seven numbers. NMIM contained
emissions for SCC codes beginning with 2230071, 2230072, 2230073, 2230074, and
2230075. In order to calculate a projection factor for SCC codes beginning with
2330070 in the NEI, the emissions for 2230071 YYY, 2230072YYY, 2230073 YYY,
2230074YYY, and 2230075 YYY were summed together for each road type YYY (as
described above) for each FIPS/CAS across the HDDV SCC emissions. The summed
emissions were assigned to an SCC code 223007XYYY where YYY is the road type.
Table 16 shows the NMIM SCC codes used to create each of the 223007XYYY SCC
emissions. Emissions were assigned to an SCC code 223007X instead of 2230070 for
ease of visual QA of the emissions, given the quantity of data being processed for the
onroad emissions. Once the 223007XYYY emissions were created from the NMIM
results (for 1999 and future years), a projection factor was calculated using Equation 2.
In this case, ENMiM,2oxx represents the sum of the 5 HDDV types by road type for each
FIPS/CAS for a future year and ENMiM,i999 represents the sum of the 5 HDDV types by
road type for each FIPS/CAS for 1999. For example, for SCC 2230070130 for benzene
in Autauga County, AL for 2007 the projection factor used would be:
F + F + F + F + F
_ -^2230071130,2007 T -^2230072130,2007 T -^2230073130,2007 T -^2230074130,2007 T -^2230075130,2007 ,^
^2230070130,2007 = £^ ^ ^ ^ V)
-^2230071130,1999 T -^2230072130,1999 T -^2230073130,1999 T -^2230074130,1999 T -^2230075130,1999
where E represents the benzene emissions. For examples of the calculations see Table B-
2 in Appendix B.
• After preliminary processing, it was found that three counties in California, had data for
motorcycle SCC codes, 2201080YYY, (where YYY is the road type as described in the
above HDDV discussion) in the 1999 NEI which were not in NMIM and thus had no
FIPS/SCC/CAS projection factor. These counties were Alpine County (06003), Modoc
County (06049), and Sierra County (06091). The SCC codes are shown in Table 17. To
project the 1999 NEI emissions in these counties, the future year onroad emissions were
summed across all SCC codes for each FIPS/CAS, resulting in a county-HAP specific
emissions number. The same was done for 1999. To calculate a county level HAP-
specific projection factor the summed future year emissions were divided by the summed
1999 emissions. The county level HAP specific projection factors were then applied to
the 1999 NEI motorcycle emissions for the appropriate HAPs.
27
-------
Table 16. HDDV SCC codes used to calculate HDDV emissions for NEI ]
HDDVtype
(First 7 characters of SCC code)
2230071, 2230072, 2230073, 2230074, 2230075
2230071, 2230072, 2230073, 2230074, 2230075
2230071, 2230072, 2230073, 2230074, 2230075
2230071, 2230072, 2230073, 2230074, 2230075
2230071, 2230072, 2230073, 2230074, 2230075
2230071, 2230072, 2230073, 2230074, 2230075
2230071, 2230072, 2230073, 2230074, 2230075
2230071, 2230072, 2230073, 2230074, 2230075
2230071, 2230072, 2230073, 2230074, 2230075
2230071, 2230072, 2230073, 2230074, 2230075
2230071, 2230072, 2230073, 2230074, 2230075
2230071, 2230072, 2230073, 2230074, 2230075
Road types
(last 3 digits of SCC code)
110 (Rural Interstate)
130 (Other Principal Arterial)
150 (Rural Minor Arterial)
170 (Rural Major Collector)
190 (Rural Minor Collector)
2 10 (Rural Local)
230 (Urban Interstate)
250 (Urban Other Freeways and
Expressways)
270 (Urban Other Principal
Arterial)
290 (Urban Minor Arterial)
3 10 (Urban Collector)
330 (Urban Local)
Drojections.
SCC in NEI which
projections are applied
2230070110
2230070130
2230070150
2230070170
2230070190
2230070210
2230070230
2230070250
2230070270
2230070290
2230070310
2230070330
HDDV descriptions:
2230070: HDDV
2230071: HDDV Class 2B
2230072: HDDV Class 3, 4, and 5
2230073: HDDV Class 6 and 7
2230074: HDDV Class 8A and 8B
2230075: HDDV Buses (School and Transit)
28
-------
Table 17. Motorcycle (MC) SCC codes not in NMIM output for Alpine, Modoc, and Sierra
Counties California.
SCC
2201080110
2201080130
2201080150
2201080170
2201080190
2201080210
2201080330
Description
Mobile
Mobile
Arterial
Mobile
Mobile
Mobile
Mobile
Mobile
Sources,
Sources,
: Total
Sources,
Sources,
Sources,
Sources,
Sources,
Highway
Highway
Highway
Highway
Highway
Highway
Highway
Vehicles
Vehicles
Vehicles
Vehicles
Vehicles
Vehicles
Vehicles
- Gasoline,
- Gasoline,
- Gasoline,
- Gasoline,
- Gasoline,
- Gasoline,
- Gasoline,
Motorcycles
Motorcycles
Motorcycles
Motorcycles
Motorcycles
Motorcycles
Motorcycles
(MC),
(MC),
(MC),
(MC),
(MC),
(MC),
(MC),
Rural
Rural
Rural
Rural
Interstate: Total
Other Principal
Minor Arterial: Total
Major
Rural Minor
Rural
Local:
Urban Local
Collector:
Collector:
Total
Total
Total
Total
A summary of national-level onroad projected emissions is provided in Table 18. Summary
emissions were calculated nationwide by vehicle type. The vehicle types summarized are: 1)
heavy duty gasoline vehicles (HDGV), 2) heavy duty diesel vehicles (HDDV), 3) light duty
diesel trucks (LDDT), 4) light duty diesel vehicles (LDDV), 5) light duty gasoline trucks 1
(LDGT1), 6) light duty gasoline trucks 2 (LDGT2), 7) light duty gasoline vehicles (LDGV) and
8) motorcycles (MC).
More detailed summaries of onroad projected emissions can be found in the MSAT rule docket:
EPA-HQ-OAR-2005-0036. The State and HAP specific summaries can be found in
onroad_0923.xls and as a pivot table in onroad_pivot.xls
29
-------
Table 18. National summary of projected onroad emissions by vehicle type for 1999, 2007,
2010, 2015, 2020, and 2030 across all HAPs and for 1,3-butadiene, acetaldehyde, acrolein,
benzene, formaldehyde, and naphthalene.
Vehicle
Type
HDDV
HDGV
LDDT
LDDV
LDGT1
LDGT2
HAP
1,3 -Butadiene
Acetaldehyde
Acrolein
Benzene
Formaldehyde
Naphthalene
All HAPs
1,3 -Butadiene
Acetaldehyde
Acrolein
Benzene
Formaldehyde
Naphthalene
All HAPs
1,3 -Butadiene
Acetaldehyde
Acrolein
Benzene
Formaldehyde
Naphthalene
All HAPs
1,3 -Butadiene
Acetaldehyde
Acrolein
Benzene
Formaldehyde
Naphthalene
All HAPs
1,3 -Butadiene
Acetaldehyde
Acrolein
Benzene
Formaldehyde
Naphthalene
All HAPs
1,3 -Butadiene
Acetaldehyde
Acrolein
Benzene
Formaldehyde
Naphthalene
All HAPs
Emissions (tons/yr)
1999
1,431
7,568
807
2,674
19,887
172
38,534
1,507
1,569
714
7,967
6,648
752
80,227
64
200
24
167
495
6
1,279
44
164
16
120
391
7
977
5,132
5,766
661
42,433
14,907
766
342,839
3,483
3,433
357
20,638
9,809
491
186,078
2007
995
5,310
561
1,872
13,921
98
26,923
483
722
177
4,041
2,242
540
35,096
38
120
14
100
297
3
766
6
24
2
17
56
1
139
3,218
3,947
434
30,773
8,540
612
239,534
1,919
2,411
255
17,701
5,264
260
139,447
2010
877
4,682
494
1,650
12,272
67
23,707
260
465
79
2,970
1,309
388
24,838
31
96
12
82
283
2
617
3
10
1
7
24
0.3
60
2,801
3,265
368
27,498
6,787
4,164
208,636
1,735
2,023
222
16,805
4,164
268
126,396
2015
760
4,071
429
1,434
10,663
33
20,570
130
297
25
2,152
741
241
17,342
29
84
11
74
211
1
552
1
6
1
4
14
0.2
34
2,307
2,714
306
23,835
5,572
702
177,486
1,524
1,789
198
15,694
3,628
274
114,204
2020
755
4,049
425
1,426
10,601
20
20,435
103
245
18
1,760
599
189
13,666
26
73
10
67
186
1
491
1
4
0.4
3
9
0. 1
23
2,291
2,682
302
23,346
5,516
774
170,855
1,503
1,726
191
14,897
3,513
281
105,843
2030
859
4,633
483
1,629
12,109
16
23,336
84
209
12
1,539
498
170
12,023
23
57
9
57
148
1
402
1
4
0.4
3
9
0.1
22
2,447
2,899
326
24,856
5,975
906
179,122
1,486
1,710
188
14,505
3,509
316
102,085
30
-------
Table 18. Continued.
Vehicle
Type
LDGV
MC
Total
Onroad
HAP
1,3 -Butadiene
Acetaldehyde
Acrolein
Benzene
Formaldehyde
Naphthalene
All HAPs
1,3 -Butadiene
Acetaldehyde
Acrolein
Benzene
Formaldehyde
Naphthalene
All HAPs
1,3 -Butadiene
Acetaldehyde
Acrolein
Benzene
Formaldehyde
Naphthalene
All HAPs
Emissions (tons/yr)
1999
11,743
11,057
1,396
955,951
27,957
1,757
778,772
220
171
18
764
582
26
8,826
23,623
29,928
3,993
170,355
80,677
3,978
1,437,532
2007
3,983
4,311
511
40,478
9,239
950
317,021
234
204
19
784
609
27
8,691
10,876
17,049
1,974
95,766
40,168
2,490
767,617
2010
2,855
3,155
374
29,722
6,811
831
232,547
244
214
20
817
635
28
9,035
8,807
13,909
1,570
79,550
32,240
2,229
625,836
2015
1,895
2,123
251
19,835
4,628
726
153,050
266
233
22
892
693
30
9,854
6,913
11,317
1,242
63,920
26,150
2,007
493,092
2020
1,500
1,690
199
15,643
3,705
678
118,762
288
253
24
967
751
33
10,673
6,468
10,721
1,170
58,109
24,879
1,976
440,748
2030
1,614
1,831
215
16,895
4,028
807
128,305
350
309
29
1177
912
40
12,957
6,864
11,651
1,263
60,660
27,188
2,255
458,252
HDDV: Heavy Duty Diesel Vehicles; HDGV: Heavy Duty Gasoline Vehicles
LDDT: Light Duty Diesel Tracks; LDDV: Light Duty Diesel Vehicles
LDGT1: Light Duty Gasoline Tracks 1; LDGT2: Light Duty Gasoline Tracks 2
LDGV: Light Duty Gasoline Vehicles; MC: Motorcycles
Once the onroad inventory had been projected, emissions for intermediate years not included in
assessments done for the MSAT rule, 2002 through 2006, inclusive, 2008, and 2009 were
interpolated from the MSAT projections and 1999 base emissions for each FIPS/SCC/CAS. For
years between 1999 and 2007, the following formula was used to interpolate the projection
factors for non-MSAT years:
PF2QXX = 1+ ((20XX- 1999) x (PF2007 - l)/(2007- 1999))
(4)
where PF2oxx is the interpolated projection factor for 2002 through 2006, 20XX is the year 2002
through 2006, PF2oo? is the projection factor for 2007 calculated from Equation 2 and 1 is the
projection factor for 1999 (base year, no growth PF=1).
For 2008 and 2009, 2010 replaces 2007, and 2007 replaces 1999.
PF20XX = 1+ ((20XX- 2007) x (PF20W - PF2007)/(2010- 2007))
(5)
-------
3.3.3 Nonroadprojections using NMIM (excluding aircraft, locomotives, and commercial
marine vessels)
The projection of the portion of the nonroad inventory that is developed using the NONROAD
model followed a similar methodology as for the onroad. Projection factors (FIPS/SCC/CAS
specific) were developed using the 1999 and future year NMIM runs using equation (2) above,
and were applied to nonroad categories in the 1999 NEI. Similar to onroad, some processing
took place, as described by the bullets below, to create FIPS/SCC/CAS projection factors to map
to the 1999 NEI.
• The same processing was done for the nonroad as for onroad to create summed
chromium, xylenes, manganese, and nickel projection factors. See first two bullets of
Section 3.3.2.
• NMEVI SCC emissions were summed to a "Total" or aggregated category (first 7 digits
of SCC followed by 3 zeros) for each FIPS/HAP/SCC since numerous emission records
in the 1999 NEI contained these aggregated categories and thus needed NMIM-based
projection ratios. See Table 19 for a list.
• NMIM pleasure craft emissions, SCC codes beginning with 2282, were summed to
provide a projection factor for the SCC code 2282000000 for each FIPS/CAS.
• For remaining FIPS/SCC/CAS combinations that did not match the NMIM emissions,
county-level HAP specific projection factors were created based on engine/fuel type by
summing emissions for 1999 NMIM and future year NMIM for each
FIPS/CAS/engine/fuel type. These were applied to all SCC codes with the relevant
engine/fuel type by HAP and by county. The engine fuel types were 2-stroke gasoline,
4-stroke gasoline, diesel, LPG, CNG, and miscellaneous.
• For CNG and LPG emissions in California and Texas (SCC codes beginning with
226800, 226801, and 226700) from the 1999 NEI without an NMIM based
FIPS/SCC/CAS specific projection factor, used the VOC or PM county level ratios for
CNG and LPG as fuel types for the HAPs in the inventory. Particulate HAPs received
the PM ratios, and gaseous HAPs received the VOC ratios.
Appendix C provides the steps used to develop the projected emissions and contains sample
calculations.
32
-------
Table 19. SCC codes in the 1999 NEI inventory and not in the NMIM inventory.
SCC
2260001000
2260002000
2260003000
2260004000
2260005000
2260006000
2260007000
2265001000
2265001020
2265002000
2265003000
2265004000
2265005000
2265006000
2265007000
2270001000
2270002000
2270003000
2270004000
2270005000
2270006000
2270007000
Description
Mobile Sources, Off-highway Vehicle
Gasoline, 2-Stroke, Recreational Equipment,
Total
Mobile Sources, Off-highway Vehicle
Gasoline, 2-Stroke, Construction and Mining
Equipment, Total
Mobile Sources, Off-highway Vehicle
Gasoline, 2-Stroke, Industrial Equipment, Total
Mobile Sources, Off-highway Vehicle
Gasoline, 2-Stroke, Lawn and Garden
Equipment, All
Mobile Sources, Off-highway Vehicle
Gasoline, 2-Stroke, Agricultural Equipment,
Total
Mobile Sources, Off-highway Vehicle
Gasoline, 2-Stroke, Commercial Equipment,
Total
Mobile Sources, Off-highway Vehicle
Gasoline, 2-Stroke, Logging Equipment, Total
Mobile Sources, Off-highway Vehicle
Gasoline, 4-Stroke, Recreational Equipment,
Total
Mobile Sources, Off-highway Vehicle
Gasoline, 4-Stroke, Recreational Equipment,
Snowmobiles
Mobile Sources, Off-highway Vehicle
Gasoline, 4-Stroke, Construction and Mining
Equipment, Total
Mobile Sources, Off-highway Vehicle
Gasoline, 4-Stroke, Industrial Equipment, Total
Mobile Sources, Off-highway Vehicle
Gasoline, 4-Stroke, Lawn and Garden
Equipment, All
Mobile Sources, Off-highway Vehicle
Gasoline, 4-Stroke, Agricultural Equipment,
Total
Mobile Sources, Off-highway Vehicle
Gasoline, 4-Stroke, Commercial Equipment,
Total
Mobile Sources, Off-highway Vehicle
Gasoline, 4-Stroke, Logging Equipment, Total
Mobile Sources, Off-highway Vehicle Diesel,
Recreational Equipment, Total
Mobile Sources, Off-highway Vehicle Diesel,
Construction and Mining Equipment, Total
Mobile Sources, Off-highway Vehicle Diesel,
Industrial Equipment, Total
Mobile Sources, Off-highway Vehicle Diesel,
Lawn and Garden Equipment, All
Mobile Sources, Off-highway Vehicle Diesel,
Agricultural Equipment, Total
Mobile Sources, Off-highway Vehicle Diesel,
Commercial Equipment, Total
Mobile Sources, Off-highway Vehicle Diesel,
Logging Equipment, Total
SCC
2265008000
2265010000
2267001000
2267002000
2267003000
2267004000
2267005000
2267006000
2267008000
2268002000
2268003000
2268005000
2268006000
2268008000
2268010000
2270009000
2270010000
2282000000
2282005000
2282010000
2282020000
2270008000
Description
Airport Support Equipment, Total, Off-
highway 4-stroke
Mobile Sources, Off-highway Vehicle
Gasoline, 4-Stroke, Industrial Equipment, All
Mobile Sources, LPG, Recreational
Equipment, All
Mobile Sources, LPG, Construction and
Mining Equipment, All
Mobile Sources, LPG, Industrial Equipment,
All
Mobile Sources, LPG, Lawn and Garden
Equipment, All
Mobile Sources, LPG, Agricultural Equipment,
All
Mobile Sources, LPG, Commercial Equipment,
All
Airport Ground Support Equipment, All, LPG
Mobile Sources, CNG, Construction and
Mining Equipment, All
Mobile Sources, CNG, Industrial Equipment,
All
Mobile Sources, CNG, Agricultural
Equipment, All
Mobile Sources, CNG, Commercial
Equipment, All
Airport Ground Support Equipment, CNG, All
Mobile Sources, CNG, Industrial Equipment,
All
Mobile Sources, Off-highway Vehicle Diesel,
Underground Mining Equipment, All
Mobile Sources, Off-highway Vehicle Diesel,
Industrial Equipment, All
Mobile Sources, Pleasure Craft, All Fuels,
Total, All Vessel Types
Mobile Sources, Pleasure Craft, Gasoline 2-
Stroke, Total
Mobile Sources, Pleasure Craft, Gasoline 4-
Stroke, Total
Mobile Sources, Pleasure Craft, Diesel, Total
Airport Service Equipment, Total, Off-
highway Diesel
33
-------
In addition to the MS AT HAPs, there were several other HAPs in the 1999 NEI nonroad
inventory. These HAPs were also projected for other projections work and the processing is
described in Appendix C.
Summaries of national-level nonroad projected emissions for the MS AT HAPS by engine,
equipment, and engine/equipment type are provided in Tables 20, 21 and 22, respectively.
Engine types include 4-stroke gasoline, 2-stroke gasoline, diesel, aircraft, CNG (natural gas),
LPG (liquid propane), miscellaneous, and residual fuel. Equipment types projected include
agriculture, aircraft, airport support, commercial, commercial marine vessel, construction,
industrial, lawn & garden, logging, pleasure craft, railroad, recreation, and underground mining.
Engine and equipment type definitions were based on the NMIM definitions found in the NMIM
tables. These tables include the nonroad categories (locomotives, commercial marine and
aircraft) that did not utilize NMIM for projections; these were discussed in Sections 3.1 and 3.2.
More detailed summaries of nonroad projected emissions can be found in the MS AT rule docket:
EPA-HQ-OAR-2005-0036. The State and HAP specific summaries, including non-MSAT
HAPs, can be found in nonroad_0923.xls and nonroad_pivot.xls.
34
-------
Table 20. National engine emissions for selected HAPs and total MS AT HAPs for 1999, 2007,
2010, 2015, 2020, and 2030.
Engine type
2-stroke gas
4-stroke gas
Aircraft
CNG
Diesel
LPG
Miscellaneous
HAP
1,3 -Butadiene
Acetaldehyde
Acrolein
Benzene
Formaldehyde
Naphthalene
MSAT HAPs
1,3 -Butadiene
Acetaldehyde
Acrolein
Benzene
Formaldehyde
Naphthalene
MSAT HAPs
1,3 -Butadiene
Acetaldehyde
Acrolein
Benzene
Formaldehyde
Naphthalene
MSAT HAPs
1,3 -Butadiene
Acetaldehyde
Acrolein
Benzene
Formaldehyde
MSAT HAPs
1,3 -Butadiene
Acetaldehyde
Acrolein
Benzene
Formaldehyde
Naphthalene
MSAT HAPs
1,3 -Butadiene
Acetaldehyde
Acrolein
Benzene
Formaldehyde
MSAT HAPs
1,3 -Butadiene
Acetaldehyde
Acrolein
Benzene
Formaldehyde
Naphthalene
MSAT HAPs
Emissions (tons/yr)
1999
3,182
2,560
559
29,998
6,701
82
501,265
5,157
2,968
458
28,713
8,759
487
171,323
824
2,019
968
1,102
6,549
456
14,276
8
5
1
64
38
373
526
15,472
1,048
5,190
33,311
207
67,113
22
12
2
174
45
957
1
22
24
6
44
0.3
238
2007
2,513
1,974
472
25,165
5,235
72
437,667
4,148
2,848
367
23,717
7,043
493
139,306
821
2,004
960
1,114
6,505
475
14,315
2
2
0.2
20
17
115
404
11,926
801
3,930
25,598
168
51,410
17
9
1
134
34
697
0.4
19
23
5
37
0.3
215
2010
2,240
1,755
426
22,545
4,728
69
395,319
3,330
2,309
300
19,531
5,736
470
115,264
859
2,098
1,005
1,163
6,809
496
14,965
1
1
0.1
11
10
61
359
10,604
709
3,467
22,729
149
45,569
9
5
1
72
19
376
0.4
19
22
4
34
0.3
205
2015
1,847
1,467
344
18,582
3,928
71
318,999
3,224
2,196
289
19,165
5,495
497
113,151
924
2,259
1,083
1,247
7,333
530
16,081
1
0.5
0.1
6
5
33
299
8,766
583
2,812
18,739
125
37,457
o
5
2
0.3
23
6
121
0.4
18
22
4
32
0.3
200
2020
1,604
1,293
292
16,287
3,415
74
270,889
3,379
2,265
300
20,153
5,688
531
118,789
993
2,430
1,165
1,335
7,885
566
17,256
1
0.4
0.1
5
4
28
259
7,633
504
2,410
16,274
105
32,414
2
1
0.2
16
4
83
0.4
18
21
4
32
0.3
196
2030
1,595
1,296
291
16,457
3,414
80
270,706
3,805
2,511
334
22,705
6,326
597
133,591
1131
2,770
1,329
1,511
8,990
638
19,603
1
0.3
0.05
4
3
24
231
7,059
459
2,199
15,000
87
29,718
2
1
0.2
15
4
77
0.4
20
20
4
36
0.3
196
35
-------
Table 20. Continued.
Engine type
Residual Oil
HAP
Acetaldehyde
Acrolein
Benzene
Formaldehyde
Naphthalene
MSAT HAPs
Emissions (tons/yr)
1999
422
23
113
807
22
1,591
2007
552
30
148
1,056
28
2,074
2010
600
33
161
1,149
30
2,253
2015
717
39
192
1,373
35
2,687
2020
876
47
235
1,677
42
3,274
2030
1330
72
356
2,547
62
4,956
36
-------
Table 21. National equipment emissions for selected HAPs and all MSAT HAPs for 1999,
2007, 2010, 2015, 2020, and 2030.
Equipment type
Agriculture
Aircraft
Airport Support
Commercial
Commercial
Marine Vessel
Construction
HAP
1,3 -Butadiene
Acetaldehyde
Acrolein
Benzene
Formaldehyde
Naphthalene
MSAT HAPs
1,3 -Butadiene
Acetaldehyde
Acrolein
Benzene
Formaldehyde
Naphthalene
MSAT HAPs
1,3 -Butadiene
Acetaldehyde
Acrolein
Benzene
Formaldehyde
Naphthalene
MSAT HAPs
1,3 -Butadiene
Acetaldehyde
Acrolein
Benzene
Formaldehyde
Naphthalene
MSAT HAPs
1,3 -Butadiene
Acetaldehyde
Acrolein
Benzene
Formaldehyde
Naphthalene
MSAT HAPs
1,3 -Butadiene
Acetaldehyde
Acrolein
Benzene
Formaldehyde
Naphthalene
MSAT HAPs
Emissions (tons/yr)
1999
243
4,493
285
2,203
9,816
49
23,098
824
2,019
968
1,102
6,549
456
14,276
7
63
6
44
139
1
421
1,140
1,400
156
6,809
3,418
98
46,990
6
2,364
98
644
4,715
65
8,736
407
5,723
392
3,601
12,417
61
39,675
2007
176
3,058
194
1,569
6,671
36
15,954
821
2,004
960
1,114
6,505
475
14,315
5
49
4
33
105
1
311
892
1,270
127
5,323
2,907
106
33,732
6
2,588
109
705
5,153
69
9,557
259
4,138
280
2,310
8,958
46
25,138
2010
148
2,581
164
1,323
5,630
32
13,476
859
2,098
1,005
1,163
6,809
496
14,965
o
J
42
4
26
90
1
251
683
1,071
105
4,206
2,435
98
27,281
6
2,639
112
719
5,252
68
9,742
214
3,578
241
1,957
7,742
42
21,702
2015
120
1,966
125
1,058
4,288
26
10,546
924
2,259
1,083
1,247
7,333
530
16,081
3
33
3
21
71
1
206
738
975
99
4,529
2,236
108
29,004
6
2,768
118
753
5,499
72
10,213
182
2,745
186
1,639
5,937
32
17,937
2020
101
1,542
98
877
3,363
21
8,530
993
2,430
1,165
1,335
7,885
566
17,256
3
29
3
20
63
1
191
813
920
98
4964
2131
119
31,451
6
2,974
129
809
5,899
79
10,973
165
2,210
151
1,450
4,779
23
15,609
2030
85
1,260
81
744
2,749
15
7,129
1131
2,770
1,329
1,511
8,990
638
19,603
o
J
30
o
J
22
65
1
205
972
906
102
5,906
2,128
142
36,981
7
3,619
161
982
7,152
102
13,354
156
1,883
130
1,348
4,074
16
14,303
37
-------
Table 21. Continued.
Equipment type
Industrial
Lawn/Garden
Logging
Pleasure Craft
Railroad
Recreational
HAP
1,3 -Butadiene
Acetaldehyde
Acrolein
Benzene
Formaldehyde
Naphthalene
MSAT HAPs
1,3 -Butadiene
Acetaldehyde
Acrolein
Benzene
Formaldehyde
Naphthalene
MSAT HAPs
1,3 -Butadiene
Acetaldehyde
Acrolein
Benzene
Formaldehyde
Naphthalene
MSAT HAPs
1,3 -Butadiene
Acetaldehyde
Acrolein
Benzene
Formaldehyde
Naphthalene
MSAT HAPs
1,3 -Butadiene
Acetaldehyde
Acrolein
Benzene
Formaldehyde
Naphthalene
MSAT HAPs
1,3 -Butadiene
Acetaldehyde
Acrolein
Benzene
Formaldehyde
Naphthalene
MSAT HAPs
Emissions (tons/yr)
1999
302
1,350
119
1,976
3,046
30
14,559
3,423
2,478
388
20,451
6,867
261
196,257
44
176
16
267
432
4
3,816
2071
1703
316
20304
4136
112
258,190
114
853
131
162
1,901
61
4,416
1,136
820
206
7,781
2,731
56
146,526
2007
143
857
71
986
1,790
18
7,456
2,445
1,920
252
14,729
4,727
245
115,652
29
102
9
185
248
4
2,325
1423
1179
212
14177
2848
103
172,930
107
805
124
150
1,793
51
4,143
1,600
1,330
312
12,938
3,743
81
244,129
2010
88
676
55
633
1,404
15
5,114
1,933
1,548
207
12,112
3,830
224
99,485
29
85
8
180
214
4
2,339
1201
1002
179
13113
2447
100
144,245
104
776
120
144
1,730
44
3,984
1,530
1,264
295
12,365
3,562
90
231,291
2015
50
459
38
368
963
9
3,157
1,887
1,480
201
12,039
3,678
232
101,535
29
62
7
177
167
4
2,394
1018
854
152
10507
2105
101
122,057
102
758
117
140
1,690
42
3,896
1,238
1,041
228
9,544
2,890
101
171,593
2020
39
389
33
291
832
6
2,573
2,030
1,546
212
12,960
3,856
251
109,328
31
55
7
187
155
4
2,562
928
782
139
9787
1932
104
111,936
99
731
113
134
1,629
40
3,758
1,029
886
180
7,622
2,404
105
128,661
2030
33
381
34
258
837
4
2,382
2,342
1,748
241
14,941
4,371
289
125,823
36
55
8
221
163
5
3,054
895
757
134
9598
1879
110
108,260
94
686
107
125
1,529
35
3,531
1,009
870
176
7,587
2,333
109
124,142
38
-------
Table 21. Continued.
Equipment type
Underground
Mining
HAP
1,3 -Butadiene
Acetaldehyde
Acrolein
Benzene
Formaldehyde
Naphthalene
MSAT HAPs
Emissions (tons/yr)
1999
1
39
2
15
87
0.2
176
2007
1
34
2
13
77
0.2
155
2010
1
31
2
12
68
0.2
138
2015
1
25
1
10
55
0.1
112
2020
1
22
1
9
50
0.1
100
2030
1
23
1
9
51
0.1
104
39
-------
Table 22. National engine/equipment emissions for MSAT HAPs
Engine Type
2-stroke gas
2-stroke gas
2-stroke gas
2-stroke gas
2-stroke gas
2-stroke gas
2-stroke gas
2-stroke gas
4-stroke gas
4-stroke gas
4-stroke gas
4-stroke gas
4-stroke gas
4-stroke gas
4-stroke gas
4-stroke gas
4-stroke gas
4-stroke gas
Aircraft
CNG
CNG
CNG
CNG
CNG
Diesel
Diesel
Diesel
Diesel
Diesel
Diesel
Diesel
Diesel
Diesel
Diesel
Diesel
Diesel
LPG
LPG
LPG
LPG
LPG
LPG
LPG
LPG
Miscellaneous
Miscellaneous
Miscellaneous
Residual
Equipment Type
Agriculture
Commercial
Construction
Industrial
Lawn/Garden
Logging
Pleasure Craft
Recreational
Agriculture
Airport Support
Commercial
Construction
Industrial
Lawn/Garden
Logging
Pleasure Craft
Railroad
Recreational
Aircraft
Agriculture
Airport Support
Commercial
Construction
Industrial
Agriculture
Airport Support
Commercial
Commercial Marine Vessel
Construction
Industrial
Lawn/Garden
Logging
Pleasure Craft
Railroad
Recreational
Underground Mining
Agriculture
Airport Support
Commercial
Construction
Industrial
Lawn/Garden
Railroad
Recreational
Commercial Marine Vessel
Pleasure Craft
Railroad
Commercial Marine Vessel
Total nonroad
Emissions (ton/yr)
1999
358
8,758
9,852
127
107,470
2,753
241,175
130,772
3,186
162
34,652
5,327
8,571
86,647
418
16,724
51
15,595
14,276
10
0.21
55
.04
307
19,553
257
3,485
7,125
24,479
4,671
2,127
645
225
4,213
159
176
0.24
2
39
18
883
14
0.04
1
20
66
151
1,591
757,136
2007
81
2,330
4,001
17
45,085
1,629
158,105
226,419
2,620
116
28,153
3,366
3,695
68,917
340
14,515
33
17,551
14,315
o
J
0.11
20
0.02
93
13,249
193
3,201
7,460
17,758
3,009
1,640
356
262
3,967
159
155
0.19
2
29
13
642
10
0.03
0.40
23
49
143
2,074
645,797
2010
79
2,487
3,774
14
43,643
1,736
129,809
213,777
2,210
81
21,883
2,537
2,251
54,462
319
14,131
28
17,361
14,965
1
0.07
11
0.01
48
11,184
169
2,884
7,466
15,384
2,455
1,374
284
262
3,818
153
138
0.12
1
16
7
346
5
0.02
0.22
23
44
138
2,253
574,012
2015
85
2,769
3,825
8
47,298
1,882
108,270
154,861
1,967
74
23,931
2,374
1,269
53,090
331
13,491
29
16,595
16,081
1
0.03
6
0.005
27
8,493
131
2,292
7,500
11,735
1,742
1,145
181
257
3,732
136
112
0.05
0.31
5
2
111
2
0.02
0.08
25
39
136
2,687
488,730
2020
91
3,072
3,877
4
51,110
2,051
98,520
112,163
1,803
77
26,534
2,350
920
57,199
366
13,129
30
16,383
17,256
1
0.03
5
0.004
23
6,636
114
1,836
7,672
9,381
1,550
1,018
145
250
3,596
115
100
0.04
0.21
4
2
76
1
0.02
0.05
27
36
132
3,274
442,928
2030
102
3,698
3,980
0.002
58,731
2,483
94,557
107,154
1,621
88
31,956
2,384
669
66,074
444
13,407
33
16,915
19,603
1
0.02
4
0.003
20
5,405
117
1,320
8,363
7,939
1,622
1,017
127
260
3,373
72
104
0.02
0.19
3
1
71
1
0.02
0.05
35
36
125
4,956
458,871
40
-------
3.3.4 Projection of onroad refueling emissions
Onroad refueling emissions are inventoried as stationary sources, although the emissions are
related to mobile sources, and can be estimated using NMIM. As such, the onroad refueling
emissions were projected using ratios developed from 1999, 2007, 2010, 2015, 2020, and 2030
refueling emissions developed from NMIM (Michaels et. al, 2005). More details on the NMEVI
refueling runs can be found in Appendix A. The ratios, used as projection factors, were
calculated by dividing the future year NMIM onroad refueling emissions (2007 and beyond) by
1999 NMIM onroad refueling emissions in each county. These factors were then assigned to the
onroad refueling SCC codes in the 1999 point and non-point inventory shown in Table 23. A
map of the county-level 2015 growth factors is shown as an example in Figure 3. As with the
aviation refueling emissions, the 2020 projected emissions were used for 2030 as well because of
uncertainty in other stationary source projection information for 2030. The onroad refueling
projection factors were included in the SCC growth factor files described in Section 4.1.3, and
the onroad refueling emissions were projected at the same time as the other stationary sources as
described in Section 4.3. Results are presented here however, since the projection factors were
derived from NMEVI.
Projection factors
0.1000-0.2012
0.2013-0.2961
0.2962-0.4370
0.4371-0.6639
0.6640-1.0647
1.0648-2.1957
Figure 3. 2015 county level refueling projection factors.
41
-------
Table 23. Onroad refueling SCC codes.
SCC
2501060000
2501060100
2501060101
2501060102
2501060103
40600401
40600402
40600403
40600499
40600601
40600602
40600603
Description
Storage and Transport, Petroleum and Petroleum Product Storage, Gasoline Service Stations,
Total: All Gasoline/ All Processes
Storage and Transport, Petroleum and Petroleum Product Storage, Gasoline Service Stations,
Stage 2: Total
Storage and Transport, Petroleum and Petroleum Product Storage, Gasoline Service Stations,
Stage 2: Displacement Loss/Uncontrolled
Storage and Transport, Petroleum and Petroleum Product Storage, Gasoline Service Stations,
Stage 2: Displacement Loss/Controlled
Storage and Transport, Petroleum and Petroleum Product Storage, Gasoline Service Stations,
Stage 2: Spillage
Petroleum and Solvent Evaporation, Transportation and Marketing of Petroleum Products, Filling
Vehicle Gas Tanks - Stage II, Vapor Loss w/o Controls
Petroleum and Solvent Evaporation, Transportation and Marketing of Petroleum Products, Filling
Vehicle Gas Tanks - Stage II, Liquid Spill Loss w/o Controls
Petroleum and Solvent Evaporation, Transportation and Marketing of Petroleum Products, Filling
Vehicle Gas Tanks - Stage II, Vapor Loss w/o Controls
Petroleum and Solvent Evaporation, Transportation and Marketing of Petroleum Products, Filling
Vehicle Gas Tanks - Stage II, Not Classified
Petroleum and Solvent Evaporation, Transportation and Marketing of Petroleum Products,
Consumer (Corporate) Fleet Refueling - Stage II, Vapor Loss w/o Controls
Petroleum and Solvent Evaporation, Transportation and Marketing of Petroleum Products,
Consumer (Corporate) Fleet Refueling - Stage II, Liquid Spill Loss w/o Controls
Petroleum and Solvent Evaporation, Transportation and Marketing of Petroleum Products,
Consumer (Corporate) Fleet Refueling - Stage II, Vapor Loss w/controls
A national summary of onroad refueling emissions by SCC is shown in Table 24.
42
-------
Table 24. Onroad refueling emissions by SCC for 1999, 2007, 2010, 2015 and 2020.
sec
2501060000
2501060100
2501060101
2501060102
2501060103
40600401
40600402
40600403
40600499
40600601
40600602
40600603
HAP
1,3 -Butadiene
Benzene
Naphthalene
All HAPs
Ren/ene
Naphthalene
All HAPs
Ren/ene
Naphthalene
All HAPs
Benzene
All HAPs
Ren/ene
Naphthalene
All HAPs
Ren/ene
Naphthalene
All HAPs
Benzene
All HAPs
Benzene
Naphthalene
All HAPs
Benzene
Naphthalene
All HAPs
Benzene
All HAPs
Benzene
All HAPs
Benzene
All HAPs
Emissions (tons)
1999
3
93
3
1,329
993
151
10,882
223
7
1,336
21
1,171
138
24
4.672
84
0.001
111
6
9
8
0.2
215
0.01
0.002
0.23
0.004
0.021
0.001
0.006
0.1
0.36
2007
2
65
2
905
08
94
6,775
140
4
839
17
998
112
19
3.838
109
0.002
144
8
11
10
0.2
281
0.01
0.002
0.30
0.004
0.021
0.001
0.006
0.1
0.46
2010
2
54
2
747
452
73
5,146
105
3
625
16
940
103
18
3.585
119
0.003
156
8
12
11
0.3
295
0.01
0.002
0.31
0.004
0.022
0.001
0.007
0.2
0.52
2015
1
46
2
633
322
57
3,821
75
2
445
16
942
102
18
3.565
142
0.003
185
10
14
13
0.3
342
0.01
0.002
0.33
0.005
0.025
0.002
0.007
0.2
0.61
2020
1
45
2
629
291
54
3,544
67
2
401
17
1,017
109
19
3.825
164
0.003
214
12
16
14
0.3
386
0.01
0.002
0.36
0.005
0.028
0.002
0.008
0.2
0.70
43
-------
3.4 Projection of HAP Precursor Emissions from Mobile Sources
In order to calculate secondary concentrations for acetaldehyde, acrolein, formaldehyde, and
propionaldehyde after ASPEN simulations for the primary concentrations for those HAPs, the
emissions for the precursors also had to be projected to 2015, 2020, and 2030 (see Table 4 for
non-HAP precursors). The total number of precursors is thirty-four. The precursor inventory
used was the same as that used for the 1999 NAT A and was derived from Version 2 of the NEI
for VOC. In addition to the non-HAP precursors listed in Table 4 there are four HAPs that are
precursors as well: 1,3-butadiene, acetaldehyde, MTBE, and methanol. The first three were
HAPs already in the projected nonroad, aircraft, and locomotive/commercial marine inventories.
Methanol (a HAP, but not an MSAT HAP) was projected as described in Appendix C. The
precursors which themselves are HAPs were projected separately from the other non-HAP
precursors and were appended to the remaining non-HAP precursors' inventories prior to
processing through EMS-HAP. Following is the methodology used to project the mobile non-
HAP precursors for locomotive and commercial marine vessels, aircraft, onroad, and remaining
nonroad sources.
3.4.1 Locomotive and Commercial Marine Vessel Precursor Emissions
Locomotive and commercial marine vessel precursor emissions were projected in the same way
as the HAP locomotive and commercial marine vessel emissions. For locomotives, the VOC
ratios shown in Table 6 were applied to each precursor. For commercial marine vessels, the
VOC ratios (Table 8) were applied to the same SCC codes shown in Table 9. In addition to the
SCC codes in Table 9, there were two other SCC codes in the precursor inventory, 2280001000
(commercial marine vessels, coal) and 2283002000 (military marine vessels, diesel). The coal
fueled marine vessel emissions were projected using the VOC ratio for all vessel types (VOC
factors for 2280000000). The military marine vessel emissions were projected using the VOC
ratios for diesel (VOC factors for 2280002000).
After the precursor emissions were assigned ratios and projected to 2015, 2020, and 2030, the
locomotive and commercial marine 1,3-butadiene, acetaldehyde, MTBE, and methanol air toxics
emissions from the projections were appended to the projected precursor locomotive and
commercial marine inventory.
3.4.2 Aircraft Precursor Emissions
Aircraft precursor emissions were projected using the same methodology and growth factors as
discussed in Section 3.2. The temporally allocated nonroad airport precursor emissions output
from PtTemporal for 1999 NAT A were subset to: 1) include only the pollutants shown in Table
4, other than the precursors that were MSAT HAPs (1,3-Butadiene, acetaldehyde, MTBE, and
methanol); and 2) exclude airport support equipment SCC codes. They were then projected for
2015, 2020, and 2030 using PtGrowCntl.
44
-------
Non-point airport-related precursor emissions (i.e., aviation gasoline) were not projected. This is
because these are stationary source emissions and it was decided to use the 1999 NATA
precursor secondary concentrations for all future year stationary precursor concentrations, except
for acrolein, which utilizes 1,3-butadiene as the sole precursor. Since acrolein's precursor is an
MSAT HAP, its secondary formation could be reasonably calculated with some confidence.
There were several reasons for not projecting the other stationary precursors: projection data
were not readily available for the stationary precursors as they were for the mobile precursors
and the approach for estimating secondary concentrations is approximate and generally shows
secondary concentrations from stationary sources to be a small portion of the total concentration
as discussed in Section 5.5.
3.4.3 Onroad Precursor Emissions
Onroad emissions for the precursors were projected using the ratio of VOC emissions for each
FIPS/SCC of 2015, 2020, or 2030 to 1999 NMIM results, in a similar fashion to that done for the
MSAT HAPs. The precursor inventory's SCC codes were classified as either exhaust or
evaporative emissions, i.e., HDDV emissions for rural interstates were divided into exhaust and
evaporative emissions. The NMIM results also were divided by exhaust or evaporative
emissions. It was decided to calculate VOC projection ratios for exhaust and evaporation
separately for each FIPS/SCC.
As with the onroad processing for MSAT HAPs, the heavy-duty diesel vehicle emissions in
NMIM were summed to create a total HDDV emission number for each FlPS/road type/exhaust
or evaporative emission type. New ratios were calculated and were applied to SCC codes
beginning with 2230070 for each FIPS/CAS in the 1999 NEI precursor inventory for the same
road type and exhaust/evaporation emission type. Table 25 lists the HDDV SCC codes in the
precursor inventory.
3.4.4 Nonroad Precursor Emissions (excluding aircraft, locomotives, and commercial marine
vessels)
The precursors from nonroad emission categories covered by the NONROAD model were
processed using a similar methodology as the emissions for HAPs. However, instead of HAP
specific projection ratios, we used VOC ratios from NMIM.
45
-------
Table 25. HDDV SCC codes used to calculate HDDV emissions in the precursor inventory.
HDDV type
(First 7 characters of
SCC code)
Road types
(last 3 digits of SCC code)
SCC in precursor
inventory which
projections are
applied
2230071,2230072,
2230073,2230074,
2230075
1IX (Rural Interstate, Exhaust)
1IV (Rural Interstate, Evaporation)
22300701IX
223007011V
2230071,2230072,
2230073,2230074,
2230075
13X (Other Principal Arterial, Exhaust)
13V (Other Principal Arterial, Evaporation)
223007013X
223007013V
2230071, 2230072,
2230073,2230074,
2230075
15X (Rural Minor Arterial, Exhaust)
15V (Rural Minor Arterial, Evaporation)
223007015X
223007015V
2230071,2230072,
2230073,2230074,
2230075
17X (Rural Major Collector, Exhaust)
17V (Rural Major Collector, Evaporation)
223007017X
223007017V
2230071, 2230072,
2230073, 2230074,
2230075
19X (Rural Minor Collector, Exhaust)
19V (Rural Minor Collector, Evaporation)
223007019X
223007019V
2230071,2230072,
2230073,2230074,
2230075
2IX (Rural Local, Exhaust)
21V (Rural Local, Evaporation)
22300702IX
223007021V
2230071,2230072,
2230073,2230074,
2230075
23X (Urban Interstate, Exhaust)
23V (Urban Interstate, Evaporation)
223007023X
223007023V
2230071,2230072,
2230073,2230074,
2230075
25X (Urban Other Freeways and Expressways, Exhaust)
25V (Urban Other Freeways and Expressways, Evaporation)
223007025X
223007025V
2230071,2230072,
2230073,2230074,
2230075
27X (Urban Other Principal Arterial, Exhaust)
27V (Urban Other Principal Arterial, Evaporation)
223007027X
223007027V
2230071, 2230072,
2230073, 2230074,
2230075
29X (Urban Minor Arterial, Exhaust)
29V (Urban Minor Arterial, Evaporation)
223007029X
223007029V
2230071,2230072,
2230073,2230074,
2230075
3IX (Urban Collector, Exhaust)
31V (Urban Collector, Evaporation)
22300703IX
223007031V
2230071,2230072,
2230073,2230074,
2230075
33X (Urban Local, Exhaust)
33V (Urban Local, Evaporation)
223007033X
223007033V
46
-------
4. Development of Future Year Stationary Source Emissions
This section describes the methodology used to develop growth factors, reduction factors, and
other inventory changes used to project the stationary (point and non-point inventories) to
various future years, including 2015 and 2020, which are the MS AT years of interest. As
previously noted, 1999 stationary source emissions were not projected to 2030 because of
uncertainty in 2030 projection information; 2020 stationary emissions were used for both 2020
and 2030.
The general approach was to develop growth and reduction factors, and apply them using EMS-
HAP Version 3.0. For one category (medical waste incineration), however, a draft 2002
emission inventory was used to represent emissions for all future years (Section 4.3).
4.1 Growth factors
Three sets of growth factors (GFs) were developed for input into EMS-HAP for use in growing
stationary source emissions: Maximum Achievable Control Technology (MACT)-based GFs,
Standard Industrial Classification (SlC)-based GFs and SCC-based GFs. Depending upon the
particular code (i.e., MACT, SCC, SIC), the GFs were national, state-level or county level.
EMS-HAP uses the most specific level of data (county) available within a particular GF file.
Thus, if a SIC-based GF file contained state and county GFs for the same SIC, and if the county
in the GF file matched the county in the inventory, EMS-HAP would apply the county SIC-based
GF. Also, in EMS-HAP, if an inventory record matches to GFs in multiple files, the MACT-
based GF overrides the SIC-based GF, which overrides the SCC-based one.
For stationary sources, growth factors were developed using three sources of information:
• Regional Economic Model, Inc. (REMI) Policy Insight® model, version 5.5 (REMI,
2004; Fan et al., 2000),
• Regional and National fuel-use forecast data from the U.S. Department of Energy,
Annual Energy Outlook for the years 2004, 2001 and 2002 (Energy Information
Administration, 2005), and
• Rule development leads or economists who had obtained economic information in the
process of rule development.
The first two sources of information were also used in projecting criteria pollutant emissions for
the Clean Air Interstate Rule (U.S. EPA, 2005b). Earlier versions of REMI and AEO were used
to develop the EGAS 4.0, which provides growth factors from 1996 up to 2020 (E.H. Pechan and
Associates, 2001).
4.1.1 MACT based growth factors
The MACT-based growth factors used in the projections are shown in Tables 26 (national level
growth factors) and Table 27 (state level growth factors for utility boilers, coal, which is
47
-------
MACT= 1808-1). Most growth factors were based on data from rule development project leads.
Some leads estimated that particular categories were not expected to experience any growth, and
were assigned growth factors of 1.0. Some leads provided a per year rate, which resulted in a
formula of raising a percent growth to a power, where the power was the number of years
between the future year and 1999. In one case, for primary aluminum production
(MACT=0201), year-specific growth factors based on a 1996 base year were provided; we
determined the 1999 base year growth factors as the ratio of the future year's growth factor and
1999 growth factor from the 1996 base year information (Table 26). All MACT-based growth
factors in the files were national level growth factors with the exception of 1808-1 (coal burning
utility boilers). These growth factors were developed at the state level, using Integrated Planning
Model (IPM) run results from the IAQR proposal (http://www.epa.gov/airmarkets/epa-
ipm/iaqr.html) (U. S. EPA, 2004c). The IPM data were available for 2010 and 2015; thus
growth factors for 1808-1 for other years were computed using interpolation, with 2020 being set
equal to 2015. For years prior to 2010 the interpolation formula was:
GFX = 1 + ((X - 1999) x (GF2010 - 1) / (2010 - 1999)) (6)
where X is 2015 or 2020.
48
-------
Table 26. National level MACT growth factors for 2015 and 2020.
MACT
0101-2
0105
0108
0201
0302
0303
0409
0412
0415
0705
0707
0802
1001
1101
1609
1614
1621
1631
1643
1801
1802
1808-2
1808-3
Description
Rocket Engine Test
Firing
Stationary RICE
Stationary
Combustion Turbines
Primary Aluminum
Production
Coke Ovens:
Charging, Top Side,
and Door Leaks
Coke Ovens:
Pushing, Quenching,
& Battery Stacks
Mineral Wool
Production
Wool Fiberglass
Manufacturing
Clay Ceramics
Manufacturing
Magnetic Tapes
(Surface Coating)
Metal Can (Surface
Coating)
Municipal Landfills
Aery lic/Modacry lie
Fibers Production
Manufacture of
Nutritional Yeast
Commercial
Sterilization Facilities
Halogenated Solvent
Cleaners
Paint Stripping
Operations
Rubber Tire
Production
Dry Cleaning:
Perchloroethylene
Medical Waste
Incinerators
Municipal Waste
Combustors
Utility Boilers:
Natural Gas
Utility Boilers: Oil
Methodology*
no growth
5% growth per year
0.8% growth per year
Future year's 1996 based
growth factor divided by 1999
growth factor based on 1996
4% decline per year
4% decline per year
no growth
no growth
no growth
no growth
no growth
no growth
no growth before 2007, 1%
growth after 2007
growth factors based on 2020
GF=1.14
0.5% growth per year
no growth
decline by 40% from 1999 to
2010 Keep same growth factor
as 2010 for all future years
thereafter.
increase by 2% per year from
1999 to 2020.
no growth
no growth; future set to 2002
emissions. See Section 4. 3
no growth
no growth
no growth
Equation
GF= 1
GF=1.05(year-1999)
GF=1.008(year-1999)
GF=GF1996/(1999 GF1996)
1999 GF1996 = 0.832
2015 GF1996= 1.025
2020 GF1996= 1.11
GF=0.96(year-1999)
GF=0.96(year-1999)
GF= 1
GF= 1
GF= 1
GF= 1
GF= 1
GF= 1
GF=1 before 2007
QF=1 oi(year~2007)
1.006258947(year-1999)
L005(year-1999)
GF= 1
GF=0. 954623 b—1999)
GF=0.6for2010and
beyond
I Q2(year-1999)
GF= 1
GF= 1
GF= 1
GF= 1
GF=1
Growth Factors
2015
1.0000
2.1829
1.1360
1.2320
0.5204
0.5204
1.0000
1.0000
1.0000
1.0000
1.0000
1.0000
1.0829
1.1045
1.0831
1.0000
0.6
1.3728
1.0000
1.0000
1.0000
1.0000
1.0000
2020
1.0000
2.7860
1.1821
1.3341
0.4243
0.4243
1.0000
1.0000
1.0000
1.0000
1.0000
1.0000
1.1381
1.1400
1.1104
1.0000
0.6
1.5157
1.0000
1.0000
1.0000
1.0000
1.0000
* growth factor methodologies provided by project leads
49
-------
Table 27. Utility Boilers: Coal (MACT= 1808-1) state level growth factors for 2015 and 2020
State
FIPS
01
02
04
05
06
08
09
10
11
12
13
15
16
17
18
19
20
State
Alabama
Alaska
Arizona
Arkansas
California
Colorado
Connecticut
Delaware
District of
Columbia
Florida
Georgia
Hawaii
Idaho
Illinois
Indiana
Iowa
Kansas
Growth
Factor
1.0124
1.0291
0.8722
1.0505
1.1607
0.9969
2.9294
1.1898
1.0000
0.9407
1.1779
1.0291
1.0000
1.1783
1.0211
0.9547
1.1285
State
FIPS
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
State
Kentucky
Louisiana
Maine
Maryland
Massachusetts
Michigan
Minnesota
Mississippi
Missouri
Montana
Nebraska
Nevada
New
Hampshire
New Jersey
New Mexico
New York
North Carolina
Growth
Factor
1.0061
0.7512
0.8222
0.8925
0.6548
1.0635
1.0894
1.1299
1.1095
0.9568
1.1353
1.1310
0.9262
1.3554
0.9538
1.1976
1.1753
State
FIPS
38
39
40
41
42
44
45
46
47
48
49
50
51
53
54
55
56
State
North Dakota
Ohio
Oklahoma
Oregon
Pennsylvania
Rhode Island
South Carolina
South Dakota
Tennessee
Texas
Utah
Vermont
Virginia
Washington
West Virginia
Wisconsin
Wyoming
Growth
Factor
0.8446
1.1332
0.9677
0.9657
1.1294
1.0000
1.1315
0.9049
1.0324
0.8056
0.8566
1.0000
0.9378
1.0034
1.0764
1.2966
0.8366
In Table 27, Alaska and Hawaii were set equal to the average 48 state growth factor.
Note, MACT codes in the NEI that are not listed in Tables 26 and 27 were not assigned a
MACT-based growth factor. Instead growth for sources with those MACT codes were grown
using the SIC or SCC based growth factors, described in the next sections.
The actual MACT-based growth factors files containing the data described above are provided
with the EMS-HAP version 3.0 projection-related ancillary files, at
http://www.epa.gov/ttn/chief/emch/projection/emshapS0.html and also in the MSAT rule docket
(EPA-HQ-OAR-2005-0036).
50
-------
4.1.2 SIC based growth factors
State-specific SIC-based growth factors, for specific standard industrial codes (SIC) were
developed using the Regional Economic Model, Inc. (REMI) Policy Insight® model, version 5.5
(being used in the development of the Economic Growth Analysis System (EGAS), version 5.0,
(U.S. EPA, 2005c)). The REMI model forecasts economic activity by region and for individual
sectors of the economy. By making assumptions about which economic indicators can represent
emissions growth, growth factors can be developed for projecting emission inventories. A
review of these growth factors for the development of the Clean Air Interstate Rule (U.S. EPA,
2005b) projected inventories, led to changes to about thirty SIC-based growth factors where they
were unrealistic or highly uncertain (U.S. EPA 2005b). They were replaced with data (national-
level) from industry forecasts, bureau of labor statistics (BLS) projections and Bureau of
Economic Analysis (BEA) historical growth from 1986 - 2002 (U. S. EPA, 2005b). These SIC
codes are shown in Table 28. Also SIC 1041 (Mining of gold ores) was set to no growth
(GF=1.0). Growth factors for 3322 (Malleable iron foundries) and 3324 (Steel investment
foundries) were set equal to the growth factors for SIC 3321.
Table 28. SIC codes changed due to unrealistic growth factors.
SIC
Description
1311
Oil And Gas Extraction, Crude Petroleum And Natural Gas, Crude petroleum and natural gas
1321
Oil And Gas Extraction, Natural Gas Liquids, Natural gas liquids
2821
Chemicals And Allied Products, Plastics Materials and Synthetics, Plastics materials and resins
2822
Chemicals And Allied Products, Plastics Materials and Synthetics, Synthetic rubber
2823
Chemicals And Allied Products, Plastics Materials and Synthetics, Cellulosic manmade fibers
2851
Chemicals And Allied Products, Paints and Allied Products, Paints and allied products
2873
Chemicals And Allied Products, Agricultural Chemicals, Nitrogenous fertilizers
2874
Chemicals And Allied Products, Agricultural Chemicals, Phosphatic fertilizers
2895
Chemicals And Allied Products, Miscellaneous Chemical Products, Carbon black
3011
Rubber And Misc. Plastics Products, Tires and Inner Tubes, Tires and inner tubes
3211
Stone, Clay, And Glass Products, Flat Glass, Flat glass
3221
Stone, Clay, And Glass Products, Glass and Glassware, Pressed Or Blown, Glass containers
3229
Stone, Clay, And Glass Products, Glass and Glassware, Pressed Or Blown, Pressed and blown glass, nee
3241
Stone, Clay, And Glass Products, Cement, Hydraulic, Cement, hydraulic
3321
Primary Metal Industries, Iron and Steel Foundries, Gray and ductile iron foundries
5325
Primary Metal Industries, Iron and Steel Foundries, Steel foundries, nee
5331
Primary Metal Industries, Primary Nonferrous Metals, Primary copper
5334
Primary Metal Industries, Primary Nonferrous Metals, Primary aluminum
5339
Primary Metal Industries, Primary Nonferrous Metals, Primary nonferrous metals, nee
5411
Fabricated Metal Products, Metal Cans and Shipping Containers, Metal cans
5441
Fabricated Metal Products, Fabricated Structural Metal Products, Fabricated structural metal
5471
Fabricated Metal Products, Metal Services, Nee, Plating and polishing
5479
Fabricated Metal Products, Metal Services, Nee, Metal coating and allied services
3497
Fabricated Metal Products, Misc. Fabricated Metal Products, Metal foil & leaf
3499
Fabricated Metal Products, Misc. Fabricated Metal Products, Fabricated metal products, nee
5711
Transportation Equipment, Motor Vehicles and Equipment, Motor vehicles and car bodies
5713
Transportation Equipment, Motor Vehicles and Equipment, Truck and bus bodies
5714
Transportation Equipment, Motor Vehicles and Equipment, Motor vehicle parts and accessories
5715
Transportation Equipment, Motor Vehicles and Equipment, Truck trailers
51
-------
The actual SIC-based growth factors files containing the data described above are provided with
the EMS-HAP version 3.0 projection-related ancillary files, at
http://www.epa.gov/ttn/chief/emch/proj ection/emshapS0.html and in the MSAT rule docket
(EPA-HQ-OAR-2005-0036).
4.1.3 SCC based growth factors
SCC based growth factors for stationary sources were derived from four sources: 1) REMI
model, 2) Energy Information Administration's National Energy Modeling System (Energy
Information Administration, 2005), and 3) NMEVI derived onroad refueling future-to-1999
emission ratios. The REMI model is discussed in Section 4.1.2 and the onroad refueling factors
are discussed in Section 2.3; and 4) aviation gasoline emissions (discussed in Section 3.2). The
National Energy Modeling system was used to calculate growth factors for emission sources
related to energy use such as residential heating. The data are provided at a division level, with
the country divided into nine divisions, for some sectors (e.g., residential fuel use), and at the
national level for more detailed industrial sectors (e.g., paper). Growth factors were developed
at the most detailed geographic scale (e.g., developed State-level growth factors from the
division information) and sectors available. The AEO data were then mapped to SCC codes
(Bollman, 2004).
In addition to the three sources of data above, emissions for fires (wild and prescribed) were
assumed to remain flat, i.e. no.
For all SCC codes, with the exception of the onroad refueling SCC codes, growth factors were at
national or state level. The refueling factors were at county level.
In the growth factor files that are input into EMS-HAP, instead of listing growth factors by SCC,
each SCC is assigned a growth indicator group. These groups consist of related SCC codes that
shared common growth factors. For example, for the onroad refueling SCC codes, instead of
listing the growth factor for each of the 12 SCC codes by FIPS, the onroad refueling SCC codes
are assigned the growth indicator group "NMIM Refueling" and the growth factors cross-
referenced by growth indicator group instead of SCC. This cuts down on the number of records
in the SCC-based growth factor files. Example records showing the SCC based growth factor
file format are shown in Figure 1 in Section 3.2.
The actual SCC-based growth factors files containing the data described above are provided with
the EMS-HAP version 3.0 projection-related ancillary files, at
http://www.epa.gov/ttn/chief/emch/proj ection/emshap30.html and in the MSAT rule docket
(EPA-HQ-OAR-2005-0036).
52
-------
4.2 Reduction factors
Not only does EMS-HAP allow the user to specify the growth factors for emissions sources,
EMS-HAP also allows for reduction of emissions. Reduction factors were applied to the grown
stationary source emissions to account for regulatory impacts and plant closures.
The percent reductions were primarily based on estimates of national average reductions for
specific HAPs or for groups of HAPs from a source category or subcategory as a result of
regulatory efforts. These efforts are primarily the MACT and Section 129 standards, mandated
in Title III of the 1990 Clean Air Act Amendments. Percent reductions were determined by, as
well as information on applicability and compliance dates, whether they apply to "major" only or
both "major" and "area" sources. With regards to applicability it was necessary to gather
information for the various rules from rule preambles, fact sheets and through the project leads
(questionnaire and phone calls). A major source is defined as any stationary source or group of
stationary sources located within a contiguous area and under common control that has the
potential to emit, considering controls, in the aggregate, 10 tons per year or more of any
hazardous air pollutant or 25 tons per year or more of any combination of hazardous air
pollutants; the status of a point source as "major" is indicated in the NEI by the field called
"FACILITY CATEGORY". For some rules, percent reductions were provided for specific
HAPs or groups of HAPs (e.g., all metals, or all volatiles) rather than a single number for all
HAPs in the categories. Information was also received on plant closures for several categories
such as coke ovens and municipal waste combustors. For the "utility boilers coal" category, it
was assumed that the acid gases (hydrochloric acid, hydrogen fluoride and chlorine) would be
reduced by the same amount as SC>2 due to co-benefits of potential controls. State-level SC>2
reductions were calculated using 862 projected emissions from the Integrated Planning Model
(IPM) runs done for proposed CAIR (U. S. EPA, 2004c) and applied these reductions to the acid
gas emissions. At the time of the projections, the IPM runs for the final CAIR rule were not
available.
Emission reductions were applied in EMS-HAP by MACT code; some were HAP and MACT
specific, some were SCC and MACT specific. Site specific reductions such as plant closures or
estimations of reductions expected from particular facilities in the source category, were applied
by the EMS-HAP site id; process specific, site specific reductions used the SCC as well.
A list of the source categories to which reductions were applied in EMS-HAP, either to facilities
in the category or the entire category, is presented in Table 29. Note that this does not include the
impacts of all of the rules, only those for which HAP emission reductions were able to be
estimated and for which the compliance date was later than 1999, or for which information on
closures was obtained. In addition, if the inventory did not have emissions for which the rule
was expected to impact, then that was also left out of the table. It also does not include
reductions from MWI, as discussed in the next section.
The actual reduction information for these source categories is provided with the EMS-HAP
version 3.0 projection-related ancillary files, at
http://www.epa.gov/ttn/chief/emch/proj ection/emshapS0.html along with more detailed
53
-------
descriptions and summaries of the data. The reduction information and detailed summaries and
descriptions can also be found in the MSAT rule docket (EPA-HQ-OAR-2005-0036).
Table 29. Summary of Categories for which reductions were applied in EMS-HAP.
Category Category
Amino/Phenolic Resins Production: POLYMERS &
RESINS III
Ammonium Sulfate - Caprolactam By-Product Plants:
THE MON
Asphalt roofing and Processing
Boat Manufacturing
Brick and Structural Clay Products Manufacturing
Carbon Black Production
Carbonyl Sulfide (COS) Production
Cellulose products manufacturing
Commercial/Industrial Solid Waste Incineration
(ClSWI)Coke Ovens: Charging, Topside and Door
Leaks Coke Ovens: Pushing, Quenching, & Battery
Stacks Cyanide Chemicals Manufacturing Ethylene
Processes Flexible Polyurethane Foam Production
Friction Products Manufacturing
Hazardous Waste Incineration and its subcategories:
Commercial Haz. Waste Incinerators, On-Site Haz.
Waste Incinerators, Cement Kilns, Lightweight
Aggregate Kilns
Industrial/Commercial/ Institutional Boilers & Process
Heaters
Industrial/Commercial/ Institutional Boilers & Process
Heaters (Coal)
Integrated Iron & Steel Manufacturing
Iron Foundries
Leather Tanning & Finishing Operations
Lime Manufacturing
Manufacturing of Nutritional Yeast
Mineral Wool Production
Municipal Solid Waste Landfills
Miscellaneous Organic Chemical Products & Processes
Miscellaneous Coatings Manufacturing
Municipal Waste Combustors
Primary Aluminum Production
Primary Copper Smelting
Primary Magnesium Refining
Secondary Aluminum Production
Stationary Reciprocating Internal Combustion Engines
Natural Gas Transmission & Storage
Off-Site Waste and Recovery Operations
Oil & Natural Gas Production
Organic Liquids Distribution (Non-Gasoline)
Pesticide Active Ingredient Production
Petroleum Refineries - Catalytic Cracking, Catalytic
Reforming, & Sulfur Plant Units (10 yr)
Petroleum Refineries - Other Sources Not Distinctly
Listed (4yr)
Pharmaceuticals Production
Reinforced Plastic Composites Production
Phosphate Fertilizers Production& Phosphoric Acid
Manufacturing
Plywood and Composite Wood Products
Polyether Polyols Production
Portland Cement Manufacturing
Pulp & Paper Production - Combustion &
Noncombustion.
Refractories Products Manufacturing
Rubber Tire Production
Secondary Aluminum Production
Secondary Lead Smelting
Site Remediation
Solvent Extraction for Vegetable Oil Production
Stationary Reciprocating Internal Combustion Engines
Surface coating related categories:
• Auto & Light Duty Truck
• Wood Building Products
• Large Appliances
• Metal Can
• Metal Coil
• Metal Furniture
• Miscellaneous Metal Parts
• Paper & Other Webs
• Plastic Parts & Products
• Fabric Coating Dying and Printing
• Printing/Publishing
Steel Pickling - HCL Process
Taconite Iron Ore Processing
Viscose Process Manufacturing
Wet-Formed Fiberglass Mat Production
Wool Fiberglass Manufacturing
Utility Boilers: Coal
54
-------
4.3 Application of growth and reductions to project stationary source emissions
For stationary sources, EMS-HAP was used to project the emissions, including onroad refueling,
with the lone exception of Medical Waste Incinerator (MWI, MACT=1801) emissions which
utilized draft 2002 MWI emissions as advised by the MWI project lead. For this category, it was
expected that emissions would remain at 2002 levels into the future.
For point sources, the PtTemporal output from the 1999 NATA EMS-HAP run was adjusted (via
a program called mwi.sas, which is available in the docket for this rule [EPA-HQ-OAR-2005-
0036] to change the 1999 medical waste incineration (MWI) emissions to 2002 emissions (U.S.
EPA, 2005a). The adjusted emissions then processed through PtGrowCntl, using the growth and
reduction factors described in Sections 3.3.5, 4.1 and 4.2, to project the inventory to 2002
through 2010 inclusive, 2015, and 2020.
The substitution of the 2002 MWI emissions for the 1999 emissions resulted in a change from
727 tons to 31.5 tons.
Note that the aviation gasoline point sources were run separately through EMS-HAP, using the
aircraft growth factors as described in Section 3.2.
For the non-point inventory, the EMS-HAP program CountyProc was run using the growth and
reduction factors described in Sections 3.3.4, 4.1 and 4.2, to project the inventory to 2002
through 2010 inclusive, 2015, and 2020. For all non-point projection years except for 2015 and
2020, ASPEN ready files for the non-point inventory were not needed so the GCFLAG variable
in CountyProc was set to 0, creating projected emissions without the other ASPEN-specific
steps. This was done to decrease run time.
There were no 2002 MWI emissions in the non-point inventory, so 1999 MWI non-point
emissions were removed. The amount of emissions removed was 220 tons.
Summaries of major and area & other emissions for 1999, 2007, 2010, 2015, and 2020 for
selected MSAT HAPs and the sum across all MSAT HAPs are shown in Table 30. For all
MS AT HAPs, major source emissions initially decrease from 1999 to 2007 but then increase
with time to 2020. Area & other source emissions increase with all years.
55
-------
Table 30. 1999 and projected stationary emissions for selected HAPs and total MS AT HAPs.
HAP
1,3-butadiene
Acetaldehyde
Acrolein
Benzene
Formaldehyde
Naphthalene
ALL MSAT
HAPs*
Year
1999
Major
1,982
11,578
899
9,820
30,611
2,245
290,498
Area &
other
22,164
26,990
21,097
101,362
126,365
11,831
925,042
2007
Major
1,731
9,299
763
7,671
30,857
1,850
230,800
Area &
other
22,819
28,277
21,808
108,123
131,649
13,162
1,020,953
2010
Major
1,805
9,225
731
7,877
30,970
1,919
242,641
Area &
other
22,961
28,715
21,896
109,628
133,283
13,570
1,067,558
2015
Major
2,011
10,695
819
8,696
35,367
2,146
277,173
Area &
other
23,068
29,419
21,990
111,634
136,008
14,314
1,143,706
2020
Major
2,247
12,111
904
9,634
40,657
2,398
313,831
Area &
other
23,212
30,142
22,088
114,161
139,095
15,137
1,225,530
* POM groups 2, 5, 6, and 7 may include emissions of HAPs that are not MSAT HAPs but part of those POM
groups in the stationary inventories. Non MSAT HAPs may be included due to processing in EMS-HAP when
POM HAPs are grouped into POM groups.
Figure 4 shows the comparison of stationary and mobile emissions, nationwide, after all
projections for 1999, 2007, 2010, 2015, 2020, and 2030. With all source groups considered, it
can be seen that total MSAT HAP emissions were projected to decrease with time from 1999 to
2030 with a slight increase between 2020 and 2030, due to mobile sources. It can also be seen
that non-gasoline mobile emissions are a very small part of the total emissions for all years.
D Major
D Area & Other
D Onroad Diesel
D Onroad Gasoline
• Other Nonroad
D Nonroad Gasoline
0
1999 2007 2010 2015 2020 2030
Year
Figure 4. Annual emissions by source sector at the national level.
56
-------
5. EMS-HAP Processing for HAPs
Prior to conducting air quality modeling using the ASPEN model, the emissions were processed
in the Emissions Modeling System for Hazardous Air Pollutants (EMS-HAP) Version 3 (U.S.
EPA, 2004b). EMS-HAP creates the emissions input files that are used by ASPEN to calculate
the air quality concentrations. Following are brief descriptions of the EMS-HAP processing.
The reader is referred to the EMS-HAP User's Guide (U.S. EPA, 2004b) for more details.
5.1 Point sources
Point sources (including major and area sources) are processed through four EMS-HAP
programs to create ASPEN ready files: PtDataProc, PtModelProc, PtTemporal, and
PtFinal_ASPEN. A fifth point source program, PtGrowCntl is used to apply growth factors and
reduction information to a base year inventory to develop future year emissions inventories.
This program is run between PtTemporal and PtFinal_ASPEN.
For the MSAT study, the point inventory had already been processed through PtDataProc,
PtModelProc and PtTemporal for the 1999 National Air Toxics Assessment (NATA).
Geographic locations and stack parameters' quality assurance was done in PtDataProc. See Ch.
3, EMS-HAP User's Guide for details.
In PtModelProc, the individual POM HAPs were grouped into eight POM groups, based on
cancer risk (See Section C.4.2 in Appendix C of the EMS-HAP User's Guide for POM
groupings). Also in PtModelProc, the metals (chromium, nickel, and manganese) were split into
fine and coarse particle emissions. Also, unspeciated chromium was speciated into chromium III
and chromium VI based on MACT codes. For naphthalene, emissions were split into gaseous
and particle mode. For descriptions of these two processes see Ch. 4, EMS-HAP User's Guide.
Urban/rural dispersion parameters, vent type, and building parameters are also assigned in
PtModelProc.
PtTemporal allocated the annual emissions to eight 3-hour time blocks based on the category of
the emissions. PtTemporal output was adjusted to change the 1999 medical waste incineration
(MWI) emissions to 2002 emissions (see Section 4.3) which were used as the projected MWI
emissions for all future years.
As discussed in 4.3, PtGrowCntl was run to project the inventory to 2002 through 2010
inclusive, 2015, and 2020. For 2015 and 2020, the PtGrowCntl output was subset to MSAT
HAPs and then processed through PtFinal_ASPEN to create ASPEN ready emissions files
(including reactivity/particle size information) for the point inventory. EMS-HAP also allows
for grouping of the emissions so that the contribution of different source groups can be
quantified when calculating concentrations in ASPEN. As for the 1999 NAT A, the point sources
were binned into two groups, major (group=0) and area & other sources (group=l). Source
groupings for stationary and mobile sources can be seen in Table 31.
57
-------
5.2 Non-point sources
For the 1999 NAT A, the non-point emissions inventory was first processed through the EMS-
HAP COP AX program to separate the airport related emissions from other non-point emissions
(see Table 14 for airport related SCC codes). COP AX allocated the airport related emissions to
point source locations at the airports (See Ch. 2 in the EMS-HAP User's Guide). The airport
related emissions were then processed through the same programs as the point source inventory.
The growth factors used for PtGrowCntl are documented in Section 2.2.
For the remaining non-point inventory, after removing the MWI (MACT=1801) emissions, the
emissions were projected to 2002 through 2010 inclusive, 2015, and 2020 using the EMS-HAP
program CountyProc. This program also spatially allocated county level emissions to census
tracts, temporally allocated emissions to 3-hour time blocks, assigned urban/rural dispersion
parameters, assigned reactivity classes/particle size information for ASPEN, and grouped certain
pollutants together such as the POM groups, and metals (See Ch. 9 of EMS-HAP User's Guide).
For 2015 and 2020, not all HAPs were needed for ASPEN files. Therefore, the 1999 inventory
was subset to MS AT HAPs and all POM HAPs only and CountyProc run again to project
emissions, this time with the GCFLAG set to 1, resulting in projected ASPEN-ready emissions.
With GCFLAG=1, ASPEN ready files are created with the projected emissions.
For both the non-point airport related emissions and remaining non-point sources, the emissions
were grouped into area & other sources (group=l).
5.3 Onroad sources
The emission inventories for 2015, 2020, and 2030 were projected outside of EMS-HAP using
the methodology in Section 2.4.2. Therefore, EMS-HAP was only used to create the ASPEN
ready files. For the onroad inventory, the CountyProc program was used to create the ASPEN
ready files. As with the non-point inventory, CountyProc spatially allocated county level
emissions to census tracts, temporally allocated emissions to 3-hour time blocks, assigned
urban/rural dispersion parameters, assigned reactivity classes/particle size information for
ASPEN, and grouped certain pollutants together such as the POM groups, and metals (See Ch. 9
of EMS-HAP User's Guide for details). Onroad emissions were grouped into two onroad
groups: onroad gasoline emissions (group=2) and onroad diesel emissions (group=4). SCC
codes beginning with 2201 were assigned to group 2 and SCC codes beginning with 2230 were
assigned to group 4.
58
-------
5.4 Nonroad sources
5.4.1 Aircraft sources
Aircraft emissions had been previously extracted from the 1999 inventory for NATA using
COP AX in order to be modeled in ASPEN as point sources. The projected aircraft emissions
were processed in PtFinal_ASPEN to create ASPEN ready files. Aircraft emissions, SCC codes
beginning with 2275, were grouped into non-gasoline nonroad emissions (group=3).
5.4.2 Airport Support Equipment
The projected nonroad inventories discussed in Section 3.3.3 contained emissions related to
airport support equipment. Therefore, the projected nonroad inventories were processed through
COP AX to separate the airport related emissions from the remaining nonroad emissions. See
Table 14 for airport support equipment SCC codes (those denoted as being projected in NMEVI).
After the COP AX program, the airport support equipment emissions were processed through the
point source programs PtDataProc, PtmodelProc, PtTemporal, and PtFinal_ASPEN. Note that
unlike the non-point airport emissions and aircraft emissions, the airport support equipment
emissions were not processed through the PtGrowCntl program since emissions had already been
projected outside of EMS-HAP.
5.4.3 Remaining nonroad sources
The remaining nonroad emissions were processed through CountyProc in a similar fashion to the
onroad emissions.
Both airport support equipment emissions and remaining nonroad emissions were binned into
two groups, non-gasoline nonroad emissions (group=3) and nonroad gasoline (group=5) (Table
31). SCC codes beginning with 2267, 2268, 2270, 2280, and 2285 were assigned to group 3 and
SCC codes beginning with 2260, 2265, and 2282 were assigned to group 5. The exceptions to
this were SCC codes 22882020000, 2282020005, and 2282020010, which are diesel pleasure
craft emissions. These codes were assigned to group 5 by mistake and should have been
assigned to group 3. This mistake was found after EMS-HAP and ASPEN modeling. It was
determined however, that these emissions were small when compared to the nonroad emissions
and changes were not made, and EMS-HAP and ASPEN were not rerun.
59
-------
Table 31. ASPEN emission groups for MSAT for future years2
Group
0
1
2
3
4
5
Source Sector
Major sources
Area & other sources
Onroad gasoline sources
Non-gasoline nonroad
sources
Onroad diesel sources
Nonroad gasoline
sources
Description
Any stationary source or group of stationary
sources located within a contiguous area and
under common control that emits or has the
potential to emit considering controls, in the
aggregate, 10 tons per year or more of any
hazardous air pollutant or 25 tons per year or
more of any combination of hazardous air
pollutants
Any stationary source of hazardous air
pollutants that is not a major source. Does not
include motor vehicles or nonroad vehicles.
Onroad vehicles burning gasoline
Nonroad vehicles burning fuels other than
gasoline such as diesel, natural gas, aviation
fuel, LP gas, residual oils, and miscellaneous
fuel sources.
Onroad vehicles burning diesel
Nonroad vehicles burning gasoline
Inventories*
Point
Point, and
non-point
Onroad
Nonroad
Onroad
Nonroad
Non-point and nonroad include airport related emissions.
5.5 EMS-HAP for precursors
EMS-HAP was run for 2015, 2020, and 2030 for the precursor emissions from the mobile
inventory only (i.e. not stationary sources) with the exception of 1,3 butadiene, which is both a
HAP and a precursor to acrolein, and was thus projected and run for both stationary and mobile
sources. Mobile EMS-HAP processing followed the same steps as described in Sections 5.3 and
5.4
Secondary concentrations for stationary sources for all HAPs other than acrolein were taken as
secondary concentrations from the 1999 NAT A for all years. Stationary precursors were not
projected due to the small contribution of stationary secondary contributions to the total
concentrations for acetaldehyde and formaldehyde. An analysis of the secondary contributions
of the 1999 precursor concentrations for acetaldehyde and formaldehyde revealed that stationary
secondary contributions were small when compared to the total concentrations (secondary and
background included). Figure 5 shows box and whisker plots for acetaldehyde and
formaldehyde for ratios of tract level stationary secondary concentrations to total concentrations
(white boxes) and ratios of tract level mobile secondary concentrations to total concentrations
(gray boxes) for 1999. The ratios for the stationary secondary contributions are much less than
the mobile ratios, since acetaldehyde and formaldehyde are mobile dominant. Note that even
though propionaldehyde is an MSAT HAP, it has no cancer or non-cancer risks associated with it
and was not included in the analysis of the secondary concentrations.
1999 NATA source groups were: 0=major, l=area & other, 2=all onroad mobile, and 3=all nonroad mobile.
60
-------
10
Ratios of stationary and mobile secondary contributions to total concentration
10
o
'•§
cc
10
10
95th percentile
75th percentile
Median
25th percentile
5th percentile
Acetaldehyde
Formaldehyde
Figure 5. Box and whisker plots of ratios of stationary secondary contributions to total
concentrations (white boxes) and ratios of mobile secondary contributions to total concentrations
(gray boxes) for 1999 acetaldehyde and formaldehyde concentrations. Dots represent the
national mean ratios.
61
-------
Preceeding Page Blank
6. ASPEN Processing
6.1 MSAT HAPs
Once the emissions were processed, they were input into ASPEN (U.S. EPA, 2000) to calculate
ambient air quality concentrations. In addition to the emissions, ASPEN needs meteorological
parameters, and census tract centroid locations for concentration calculations. For the MSAT
years, 2015, 2020, and 2030, the 1999 meteorology and year 2000 census tract locations were
used as for the 1999 NAT A.
In EMS-HAP, emissions are divided into nine files, one for each HAP reactivity class, 1-9, as
defined for ASPEN (Reactivity classes 6 and 8 are not used for HAPs) based on decay rates or
paniculate sizes (See ASPEN User's Guide [U.S. EPA, 2000] for details). For example, the
emissions file for reactivity class 1 would contain the emissions information (location, emissions,
stack parameters, etc.) for all of the HAPs processed through EMS-HAP with reactivity class 1.
The reactivity classes for each MSAT HAP are listed in Table 32.
ASPEN runs were set up such that the stationary and mobile concentrations were calculated in
two separate runs, one for stationary and one for mobile. Table 33 shows how the separate sets
of emission files (one file for each reactivity class) are provided to ASPEN.
ASPEN is composed of two modules, ASPENA and ASPENB. ASPENA calculates
concentrations at receptors arranged in rings around an emission source up to 50 km away.
ASPENB then reads the ASPENA output and interpolates the concentrations to census tract
centroids. ASPEN is run for each reactivity class for mobile sources and for each reactivity class
for stationary sources. The output from ASPEN is a binary file for each SAROAD code in the
emissions input file (see Table 1 for MSAT HAPs' SAROAD codes). Figures 6 and 7
graphically show the input/output for each reactivity class for stationary and mobile sources
respectively, including which HAPS are in each reactivity class.
Once ASPEN has been run, the programs AVGDAT and EXTRAVG were used to convert the
binary output from ASPEN to ASCII text. Concentrations are annual average concentrations for
each source sector and are at the census tract level. For details of the two programs see the
ASPEN User's Guide (U.S. EPA, 2000).
6.2 Precursors
Precursor emissions were processed through ASPEN in the same manner as for the HAPs but in
a separate model run. Reactivity classes for precursors are also shown in Table 32 with
input/output files shown in Figure 8 for mobile sources. As discussed in Section 5.5, the 1999
stationary secondary concentrations were used for 2015 and 2020 secondary concentrations,
excluding acrolein whose precursor emissions were projected to 2030 for both stationary and
mobile sources.
63
-------
Table 32. Reactivity classes for MSAT HAPs and precursors.
Pollutant
1,3 -Butadiene
2,2,4-Trimethylpentane
Acetaldehyde, primary
Acrolein, primary
Benzene
Chromium III, fine
Chromium III, coarse
Chromium VI, fine
Chromium VI, coarse
Ethyl Benzene
Formaldehyde, primary
Hexane
Manganese, fine
Manganese, coarse
MTBE
Naphthalene, gas
Naphthalene, fine PM
Nickel, fine
Nickel, coarse
SAROAD
43218
43250
43503
43505
45201
59992
59993
69992
69993
45203
43502
43231
80196
80396
43376
46701
46702
80216
80316
Reactivity
7
1
5
5
1
2
o
J
2
o
J
4
5
9
2
3
1
5
2
2
3
Pollutant
Propionaldehyde, primary
Styrene
Toluene
Xylenes
POM1
POM 2
POM 3
POM 4
POM 5
POM 6
POM 7
POMS
Acetaldehyde precursors,
reactive
Formaldehyde precursors,
reactive
Propionaldehyde precursors,
reactive
Acetaldehyde precursors,
inert
Acrolein precursor, inert
Formaldehyde precursors,
inert
Propionaldehyde precursors,
inert
SAROAD
43505
45220
45202
45102
71002
72002
73002
74002
75002
76002
77002
78002
80100
80180
80234
80301
80302
80303
80305
Reactivity
5
7
4
5
2
2
2
2
2
2
2
2
7
6
7
1
1
1
1
POM 1: POM, Group 1: Unspeciated
POM 2: POM, Group 2: no URE data
POM 3: POM, Group 3: 5.0E-2 < URE <= 5.0E-1
POM 4: POM, Group 4: 5.0E-3 < URE <= 5.0E-2
POM 5: POM, Group 5: 5.0E-4 < URE <= 5.0E-3
POM 6: POM, Group 6: 5.0E-5 < URE <= 5.0E-4
POM 7: POM, Group 7: 5.0E-6 < URE <= 5.0E-5
POM 8: POM, Group 8: Unspeciated (7-PAH only)
REACTIVITY CLASSES:
1 non reactive
2 fine paniculate (2.5 microns and less)
3 coarse paniculate (2.5 to 10 microns)
4 medium low reactivity
5 medium reactivity
6 medium high reactivity
7 very high reactivity
8 high reactivity
9 low reactivity
64
-------
Reactivity 1
Point
reactivity 1
input file
Non-point
reactivity 1
input file
Airport
reactivity 1
input file
r.J
->l ASI
43376.exp
43250.exp
>EN !
j
45201.exp
80302.exp
Reactivitv 3
Point
reactivity 3
input file
Non-point
reactivity 3
input file
59993.exp 4
-. i
Airport
reactivity 3
input file
r r
> 80396. exp
Reactivitv 7
Reactivitv 2
46702. exp
t
\
59992. exp
L i
' 1
72002. exp
69992.exp
t j
r i
73002.exp
71002.exp
L t
r 1
74002.exp
k
r
75002.exp
Point
reactivity2
input file
Non-point
reactivity 2
input file
H ASPEN.
Airport
reactivity 2
input file
76002.exp
770Q2.exp 78002.exp 8Q196.exp
80216.exp
Reactivitv 4
Reactivity 5
Non-point
reactivity 5
input file
Non-point
reactivity 4
input file
Figure 6. Stationary source emission input files and ASPEN output files for each reactivity class
for MSAT HAPs.
Table 33. Description of emissions files for the stationary and mobile divisions used for ASPEN
simulations.
Division
Stationary
Mobile
Emissions type
point sources (major and area & other)
non-point airport emissions (i.e., aviation gasoline categories) assigned to
point sources by COP AX
non-point emissions (excluding non-point airport emissions)
Aircraft emissions assigned to point sources by COP AX
Airport support equipment emissions assigned to point sources by COP AX.
Onroad mobile sources
Nonroad mobile sources (excluding aircraft and airport support equipment)
65
-------
Reactivity 1
Reactivity 2
Onroad reacti\
input file
ityl
Remaining nonroad
reactivity 1
input file
4337 6. exD U — i
1
r
Aircraft reactivity 1
input file
Airport support
reactivity 1
input file
>
80302. exD
4325Q.exp
Onroad reactivity 2
input file
Remaining nonroad
reactivity 2
input file
59992.exp
"1 rJ
r
Aircraft reactivity 2
input file
Airport support
reactivity 2
input file
1
-W 45201.exp
69992.C
Reactivity 3
| 8Q196.exp
Reactivity 4
O216.exp 72002.exp 75002.exp 76002.exp
ReactMtv 5
Onroad reactivity 4
input file
Remaining nonroad
reactivity 4
input file
i
Aircraft reactivity 4
input file
Airport support
reactivity 4
input file
Onroad reactivity 5
input file
Remaining nonroad
reactivity 5
input file
r
i
r
Aircraft reactivity 5
input file
Airport support
input file
I ASPEN I
i n I
45102.exp
exp
45202. exp
45203.exp
43502.exp 43503.exp 43504.exp 43505..exp
ReactMtv 9
Onroad reactivity 9
input file
Remaining nonroad
reactivity 9
input file
43231.exp
Figure 7. Mobile source emission input files and ASPEN output files for each reactivity class
for MSAT HAPs.
66
-------
Reactivity 1
Reactivity 6
Onroad reactivity 1
input file
Remaining nonroad
reactivity 1
input file
i
Aircraft reactivity 1
input file
Airport support
reactivity 1
input file
! ASPEN !
Onroad reactivity 6
input file
Remaining nonroad
reactivity 6
input file
i
r
Aircraft reactivity
input file
Airport support
reactivity 6
input file
6
! ASPEN |
Reactivity 7
Onroad reactivity 7
input file
Remaining nonroad
reactivity 7
innut file
i
Aircraft reactivity 7
input file
Airport support
reactivity 7
innut file
Figure 8. Mobile source emission input files and ASPEN output files for each reactivity class
for MS AT precursors.
6.3 Post-processing of ASPEN concentrations
ASPEN output concentrations were calculated for each SAROAD associated with the MSAT
HAPs (see Table 1 for SAROADs). Post-processing of the ASPEN concentrations for each year
included the following:
• Adjusting the SAROAD 75002 (POM Group 5) area & other concentrations in Oregon as
described in Section 2.1.
• Merging the stationary and mobile concentrations together at tract level. For 2030, 2020
stationary concentrations were used.
• Summing the fine and coarse metal concentrations (i.e., fine and coarse nickel) at census
tract level for each source sector.
67
-------
• Summing the particle and gas modes of naphthalene at census tract level for each source
category.
• Adding secondary concentrations for each source category using the appropriate
precursor concentrations for acetaldehyde, acrolein, formaldehyde, and propionaldehyde
at the census tract level. Total concentrations were calculated by adding the primary
concentrations (SAROADS in Table 1) and secondary concentrations. Secondary
concentrations are computing by subtracting the reactive component of the precursor
from the inert component of the precursor, and multiplying by a factor (if needed). The
following equations show how the ASPEN-modeled3 concentrations for each of the
HAPs with secondary components were calculated:
V ')
acrolem = X 43503 + 1-04(X 803Q1 - X 8moo) (8)
formaldehyde ~ ^ 43502 + ^ 80303 ~ ^ 80180 I")
- propionaldehyde ~ 43504 ~~ 80305 ~ 80234
Where:
Xacetaidehyde = Acetaldehyde concentrations with secondary contributions included.
X43503 = Primary acetaldehyde concentrations due to directly emitted acetaldehyde.
= Inert precursor concentrations for acetaldehyde (reactivity class 1).
= Reactive precursor concentrations for acetaldehyde (reactivity class 7).
acetaldehyde ~ 43503 ~~ 80301 ~ 80100
Xacroiein = Acrolein concentrations with secondary contributions included.
X43505 = Primary acrolein concentrations due to directly emitted acrolein.
Xgo302 = Inert precursor concentrations for acrolein (reactivity class 1)- note that 1,3
butadiene is the sole precursor for acrolein and that 80302 represents 1,3 butadiene, inert.
X43218 = Reactive precursor concentrations for acrolein (reactivity class 7))- note that 1,3
butadiene is the sole precursor for acrolein and that SAROAD=43218 represents 1,3
butadiene.
= Formaldehyde concentrations with secondary contributions included.
X43502 = Primary formaldehyde concentrations due to directly emitted formaldehyde.
X80303 = Inert precursor concentrations for formaldehyde (reactivity class 1).
= Reactive precursor concentrations for formaldehyde (reactivity class 6).
= Propionaldehyde concentrations with secondary contributions included.
3 These equations provide the ASPEN-modeled concentrations prior to the addition of the background concentration,
discussed in the next bullet.
68
-------
X43504 = Primary propionaldehyde concentrations due to directly emitted
propi onal dehy de.
X80305 = Inert precursor concentrations for propionaldehyde (reactivity class 1).
X80234 = Reactive precursor concentrations for propionaldehyde (reactivity class 6).
• Adding county level background concentrations to total concentrations (all sources) for
HAPs with background. The MSAT HAPs with nonzero background are: 1,3-butadiene,
acetaldehyde, benzene, formaldehyde, and xylenes. Each of the three model years used
1999 background. For details about the 1999 background see
http://www.epa.gov/ttn/atw/natal999/background.html or Batelle (2003). Because it
would be expected that background levels would likely change in the future due to
emissions changes, a sensitivity analysis was done to evaluate the potential impact of
changing the constant background assumption. This analysis is detailed in Section 9.
After post-processing of the concentrations, summary statistics for the concentrations for each
year, including 1999 were calculated. They included:
• Average concentrations for major, area & other, onroad gasoline, onroad diesel, nonroad
gasoline, remaining nonroad, background, and total at the county, state, state urban/rural,
state RFG/non-RFG, national, national urban/rural, and national RFG/non-RFG levels.
• Distributions (5th, 10th, 25th, 50th, 75th, 90th, and 95th percentiles) for total concentrations
at the county, state, state urban/rural, state RFG/non-RFG levels.
• Maps of county median concentrations for 1,3-butadiene, acetaldehyde, acrolein,
benzene, formaldehyde, and naphthalene were generated for 1999, 2015, 2020, and 2030.
National level average concentrations for 1999, 2015, 2020, and 2030 are shown in Table 34 and
county maps for the same years for benzene are in Figures 9 through 12.
The complete summaries and maps described above are in concentrations.xls and
ASPEN_medians.ppt respectively in the MSAT rule docket EPA-HQ-OAR-2005-0036. Note
that in Table 34 and in concentrations.xls, the national average background concentration differs
from the national average background concentration for NATA even though the same county
level background concentrations were used for all years. This is because Puerto Rico and the
Virgin Islands were not included in the 2015, 2020, and 2030 analyses but were included in the
NATA analysis.
69
-------
Table 34. National average background, stationary, and mobile ASPEN concentrations
for each MSAT HAP for 2015, 2020, and 2030.
rn-3)
HAP
1,3 -Butadiene
2,2,4-
Trimethylpentane
Acetaldehyde
Acrolein
Benzene
Chromium III
Chromium VI
Ethyl Benzene
Formaldehyde
Hexane
MTBE
Manganese
Naphthalene
Nickel
POM*
Propionaldehyde
Styrene
Toluene
Xylenes
Back-
ground
S.llxlO"2
0
S.lVxlO"1
0
3.94X10"1
0
0
0
7.62X10'1
0
0
0
0
0
0
0
0
0
l.VOxlO"1
2015
Stationary
2.27xlO"2
4.19xlO'2
8.67xlO"2
2.97xlO'2
2.04X10"1
1.66xlO'3
4.08xlO"4
1.24X10'1
1.48X10'1
5.91XKT1
7.04xlO'2
6.14xlO"3
6.15xlO'2
2.50xlO"3
2.25xlO'2
3.32xlO"2
4.90xlQ-2
1.19
8.62X10"1
Mobile
2.44xlO"2
2.86X10'1
3.62X10"1
3.41xlO'2
3.16X10"1
2.11X10'4
4.64xlO"5
1. 33X10'1
3.13X10'1
1.35XKT1
1.26X10'1
1.85xlO"4
1.24xlO'2
2.50xlO"4
1.47xlO'3
9.20xlO"2
1.23xlQ-2
7.24X10'1
5.22X10"1
2020
Stationary
2.28xlO"2
4.44xlO'2
8.93xlO"2
2.93xlO'2
2.13X10"1
1.86xlO'3
4.61xlO"4
1.37X10"1
1.60X10'1
6.38X10'1
7.41xlO'2
6.80xlO"3
6.57xlO'2
2.74xlO"3
2.34xlO'2
3.39xlO"2
5.53xlQ-2
1.31
9.52X10"1
Mobile
2.38xlO'2
2.59X10"1
3.29X10'1
3.37xlO"2
2.95X10'1
2.30xlO"4
5.05xlO'5
1.23X10"1
3.03XKT1
LlSxlO'1
l.OSxlO"1
2.08xlO'4
1.26xlO"2
2.71xlO'4
1.50xlO"3
8.21xlO'2
1.19X10'2
6.63X10"1
4.87X10'1
2030
Stationary
2.28xlO'2
4.44xlO"2
8.93xlO'2
2.93xlO"2
2.13X10'1
1.87xlO"3
4.61xlO'4
1.37X10"1
1.60X10'1
6.38X10'1
7.41xlO"2
6.80xlO'3
6.57xlO"2
2.74xlO'3
2.34xlO"2
3.39xlO'2
5.53xlO'2
1.31
9.52X10'1
Mobile
2.63xlO'2
2.75X10"1
3.50X10'1
3.74xlO"2
S.lSxlO'1
2.73xlO"4
6.00X10'5
1.32X10"1
3.31X10'1
1.23X10'1
l.OSxlO"1
2.61xlO'4
1.46xlO"2
3.19xlO'4
1.71xlO"3
8.62xlO'2
1.29xlO'2
7.07X10"1
5.24X10'1
*POM is the sum of all POM groups.
70
-------
Figure 9. 1999 County level median total (all sources and background) concentrations (ng m" )
for benzene.
71
-------
Figure 10. 2015 County level median total (all sources and background) concentrations
for benzene.
rn-3)
72
-------
Figure 11. 2020 County level median total (all sources and background) concentrations
for benzene.
rn-3)
73
-------
Figure 12. 2030 County level median total (all sources and background) concentrations
for benzene.
rn-3)
ASPEN results were also processed for input to HAPEM5. ASPEN outputs annual average
tract-level concentrations for eight 3 -hour time blocks. The concentrations were extracted from
the binary ASPEN output, .exp files, using the AVGDAT program and written to an ASCII text
file. The concentrations were then processed in a similar fashion as for the annual average
concentrations: stationary and mobile concentrations merged together, fine and coarse
components of the metals added together, gas and parti culate phases of naphthalene added
together and secondary concentrations added to the secondary HAPs. Once these steps were
done there were eight 3 -hour concentrations for major, area & other, onroad gasoline, onroad
diesel, nonroad gasoline and, nonroad other. Background was also added for each tract.
Normally HAPEM input files, also called air quality files, contain major, area & other, onroad,
nonroad, and background concentrations, as done for the 1999 NATA. For MS AT, the onroad
and nonroad concentrations were broken into two categories each, resulting in a total of seven
concentration categories. In order to avoid receding the FORTRAN programs of HAPEM, the
HAPEM input files were split into two different files, called "runl" and "run2." Runl files
contained major, area & other, onroad gasoline, nonroad gasoline and background. Run2 files
contained placeholders (values of zero concentration) for major, area & other and background.
The onroad diesel and nonroad other concentrations were added to the run2 files. Figures 13 and
14 show sample records for the runl and run2 HAPEM input files.
74
-------
Annual average concentrations in ug/m3 by time blocks
Pollutant species: Benzene
Source category: ALL
FIPS Tract Conc_blockl Conc_block2 Conc_block3 Conc_block4 Conc_block5
Conc_block6 Conc_block7 Conc_block8 Conc_blockl Conc_block2 Conc_block3
Conc_block4 Conc_block5 Conc_block6 Conc_block7 Conc_block8 Conc_blockl
Conc_block2 Conc_block3 Conc_block4 Conc_block5 Conc_block6 Conc_block7
Conc_block8 Conc_blockl Conc_block2 Conc_block3 Conc_block4 Conc_block5
Conc_block6 Conc_block7 Conc_block8 Bconc_block8
OIOOI 020100 I.014290E-02 7.904850E-03 27722090E-03 1.851860E-03 2.127090E-03
4.262630E-03
7.934730E-02
7.547470E-02
1.726230E-01
8.620350E-02
01001 020200
1.014880E-02
8.579260E-02
8.572060E-02
7.336690E-03
4.950200E-02
1.926930E-02
1.014940E-02
1.807510E-01
6.858460E-02
6.009230E-03
7.228590E-03
1.436350E-02
6.959740E-02
2.245080E-01
8.586170E-02
3.060390E-02
3.212943E-01
4.269420E-03
1.451020E-01
6.572720E-02
2.021050E-01
3.728000E-02
1.125670E-01
9.880010E-02
2.630900E-01
4.081490E-02
3.102840E-03 3.331600E-03
697180E-03
081070E-01
409060E-01
009450E-01
2.436150E-01
1.928040E-02
3.301660E-01
2.053260E-01
7.639370E-03
9.773740E-02
1.928210E-02
5.106280E-01
1.727360E-01
6.252210E-03
7.526810E-03
1.090870E-01
4.794080E-01
2.178950E-01
1.075850E-01
2.961870E-01
037620E-01
099920E-01
475200E-01
4.099370E-01
1.762320E-01
4.509380E-01
1.605220E-01
3.212943E-01
Figure 13. Sample records of the Runl 2015 HAPEM input air quality file con45201_runl.txt
for benzene. Note that each set of concentrations for a tract is one record. More records appear
due to of "wrapping" of text in word processor.
Annual average concentrations in ug/m3 by time blocks
Pollutant species: Benzene
Source category: ALL
Conc_block4
Conc_block2
Conc_block8
Cone block6
FIPS Tract Conc_blockl Conc_block2 Conc_block3 Conc_block4 Conc_block5
Conc_block6 Conc_block7 Conc_block8 Conc_blockl Conc_block2 Conc_block3
Conc_block7 Conc_block8 Conc_blockl
Conc_block5 Conc_block6 Conc_block7
Conc_block3 Conc_block4 Conc_block5
_ _ _ Bconc_block8
OIOOI 020100 CI.OOOOOOE+00 C>.OOOOOOE+00 o7oOOOOOE+00 O.OOOOOOE+00 O.OOOOOOE+00
O.OOOOOOE+00 O.OOOOOOE+00 O.OOOOOOE+00 O.OOOOOOE+00 O.OOOOOOE+00 O.OOOOOOE+00
O.OOOOOOE+00 O.OOOOOOE+00 O.OOOOOOE+00
2.433580E-03 2.331080E-03 2.200300E-03 5.139870E-03
3.162220E-04 2.568330E-04 1.723930E-03 1.608990E-03
O.OOOOOOE+00
O.OOOOOOE+00
7.163460E-04
9.321060E-04
3.808400E-03
Conc_block5
Conc_block3
Conc_blockl
Cone block7
3.525590E-03
Conc_block6
Conc_block4
Conc_block2
Cone blocks
O.OOOOOOE+00 5.437490E-04
2.213920E-03
1.761980E-03
01001 020200 O.OOOOOOE+00 O.OOOOOOE+00 O.OOOOOOE+00 O.OOOOOOE+00 O.OOOOOOE+00
O.OOOOOOE+00 O.OOOOOOE+00 O.OOOOOOE+00 O.OOOOOOE+00 O.OOOOOOE+00 O.OOOOOOE+00
O.OOOOOOE+00 O.OOOOOOE+00 O.OOOOOOE+00 9.786940E-04
6.259710E-03 5.902090E-03 1.110560E-02 3.984490E-03
3.463900E-04 4.707650E-03 4.879280E-03 5.317610E-03
1.999970E-03 O.OOOOOOE+00
O.OOOOOOE+00
1.364310E-03
1.677710E-03
8.468020E-03
O.OOOOOOE+00
6.234640E-03
4.114270E-04
5.789880E-03
Figure 14. Sample records of the Run2 2015 HAPEM input air quality file con45201_run2.txt
for benzene.
75
-------
This page intentionally blank
76
-------
7. HAPEM5 Model and Post-Processing
7.1 HAPEM model
The HAPEM5 (U.S. EPA, 2005d) model was originally compiled to run on a DOS PC, one
pollutant (HAP) at a time. In early 2004, HAPEM5 was compiled on a Linux cluster in order to
run the model for multiple pollutants simultaneously (using multiple computer processors),
thereby reducing the overall time required to run HAPEM5 for a long list of pollutants. The
post-processing FORTRAN code that converts HAPEM5 output into the input for the risk
calculations was also compiled for Linux.
The HAPEM model consists of 5 modules, to be run in order: 1) DURAV5, 2) INDEXPOP5, 3)
COMMUTES, 4) AIRQUAL5 and 5) HAPEM5. The first three modules only need to be run
once for the whole domain, while the last two modules must be run for each HAP (generally
distinguished by SAROAD code). Refer to the HAPEM5 User's Guide for more detailed
information about the model and formats of input and output files (U.S. EPA, 2005e).
For NATA (and thus 1999 results), as noted in Section 6.3, the emissions source categories were
major, area & other, onroad, nonroad and background. However, as discussed previously in 5.3,
for the future year MSAT analyses described here, the onroad and nonroad mobile source
categories were split into two categories each (highway gasoline, highway diesel, nonroad
gasoline and remaining nonroad), making the total number of non-background source categories
equal to six. Because HAPEM is limited to a maximum of five source categories per run, the
runs for each projection year were split into two runs per HAP. In other words, there were two
runs (2) for each year (3) and for each HAP (26), a grand total of 156 individual HAPEM runs.
The first run for each HAP for each year ("runl") included ASPEN-derived air quality input data
for the major, area & other, highway gasoline, and nonroad gasoline groups and background,
while "run2" only included highway diesel and remaining nonroad (no background). The output
for the two onroad groups were added together after post-processing and prior to the risk
calculations, as were the two nonroad groups.
The HAPEM5 runs for MSAT used 2000 census data, 1990 commuting data adjusted to reflect
the 2000 census tract designations, and activity pattern data from the Consolidated Human
Activity Database (CHAD) (Glen et al. 1997). HAPEM5 output for the first three modules from
a previous set of HAPEM5 runs done for the 1999 NATA were included, so only the HAP-
specific modules (AIRQUAL5 and HAPEM5) were run for MSAT.
The input files were the same for all pollutants, except for:
• a) the HAP-specific parameter file used for the AIRQUAL5 and HAPEM5 modules
(p2_MSAT_XXXXX_YYYY_runZ.txt, where XXXXX is the 5-digit pollutant
SAROAD code or text identifier, YYYY is the 4-digit projection year and Z is the run
number"!" or"2"),
• b) the ASPEN-derived air quality (AQ) concentration file (conXXXXX_runZ.txt, for
each pollutant, located in the appropriate year- and run-specific run directories), and
77
-------
• c) the microenvironmental factors file used (gas, particulate or mixed).
There are three microenvironments factor files provided with the HAPEM5 model: gas,
particulate and mixed. The factor file used for each HAP is determined from a HAP factors
lookup file also provided with the model. The factor file is listed in each HAP-specific
parameter file (p2 files) for the AIRQUAL5 and HAPEM5 modules.
The final output files generated by HAPEM5 are in the form ofXXXXX.YY.dat, where
XXXXX is the 5-digit SAROAD (for gaseous HAPs excluding secondary HAPs) or text HAP
identifier (for metals or secondary HAPs) and YY is the 2-digit state FIPS code. There are 53
output files for each SAROAD (HAP), one for each state FIPS code. This includes Puerto Rico
and Virgin Islands even though those areas were not projected. Puerto Rico and the Virgin
Islands are included because HAPEM5 was run as set up for the 1999 NAT A. Given that there
would be many files to edit (because of the number of HAPs and years) to exclude Puerto Rico
and the Virgin Islands, and that runtimes would not be significantly decreased if they were
excluded, they remained in the setup files for AIRQUAL and ignored after post-processing.
The raw HAPEM5 output had to be post-processed prior to making the risk calculations. The
post-processing code we used basically takes the exposure concentrations for each demographic
group (10 demographic groups, 5 age groups x 2 genders) and builds a lifetime exposure (about
70 years) for an individual at that tract for each HAP. At each tract and demographic group,
there are 30 replicates. For the total exposure (all sources), for each tract and demographic
group, the median exposure is used to adjust the mean exposure at each tract for the difference
source categories (major, area & other, etc.) by dividing the mean source category exposure
concentration for the tract by the median total concentration. This operation causes stationary
exposure concentrations for 2020 and 2030 to differ even though the same input concentrations
were used. The mobile concentrations between the two years changed, causing the total
concentrations to change, changing the median total concentrations at the tract level. This in
turn, changed the adjusted stationary source exposure concentrations. The final output file is in
the form XXXX_SAROAD_runY.HAPEM5-TRACT.txt where XXXX is 2015, 2020, or 2030,
SAROAD is the SAROAD or HAP name, and Y is 1 or 2 for runl or run2. Sample records of
post-processed HAPEM5 output for 2015 Benzene runl and run2 are shown in Figures 15 and
16 respectively.
78
-------
01001
01001
01001
01001
01001
01001
01001
01001
01001
01001
01001
01003
01003
01003
01003
01003
01003
01003
01003
020100
020200
020300
020400
020500
020600
020700
020800
020900
021000
021100
010100
010200
010300
010400
010500
010600
010701
010703
5
1
1
1
1
4
2
3
1
1
3
3
1
9
4
5
6
1
9
592E-03
002E-02
158E-02
597E-02
446E-02
544E-02
747E-02
978E-03
757E-03
300E-03
073E-03
347E-03
019E-02
414E-03
961E-03
667E-03
178E-03
852E-02
308E-03
1
2
1
1
1
8
7
5
4
4
4
3
5
4
7
1
1
4
4
027E-01
417E-01
763E-01
413E-01
044E-01
300E-02
944E-02
112E-02
429E-02
273E-02
065E-02
120E-02
030E-02
720E-02
028E-02
279E-01
192E-01
337E-02
769E-02
1.
2.
2.
2.
2.
2.
1.
9.
6.
5.
6.
4.
6.
6.
6.
7.
7.
8.
9.
715E-01
838E-01
403E-01
582E-01
155E-01
014E-01
601E-01
521E-02
416E-02
122E-02
345E-02
427E-02
172E-02
464E-02
688E-02
744E-02
016E-02
418E-02
990E-02
3.346E-02
7.260E-02
5.643E-02
4.394E-02
3.762E-02
3.387E-02
3.409E-02
2.191E-02
1.158E-02
1.045E-02
1.440E-02
1.870E-02
2.341E-02
3.340E-02
2.827E-02
4.564E-02
4.180E-02
5.977E-02
5.949E-02
5
8
7
7
6
6
5
4
3
3
3
3
3
4
4
5
4
4
4
653E-01
586E-01
346E-01
132E-01
250E-01
139E-01
496E-01
208E-01
648E-01
525E-01
736E-01
471E-01
949E-01
035E-01
210E-01
071E-01
884E-01
595E-01
656E-01
2.521E-01
2.505E-01
2.500E-01
2.538E-01
2.531E-01
2.502E-01
2.485E-01
2.486E-01
2.430E-01
2.468E-01
2.520E-01
2.496E-01
2.493E-01
2.488E-01
2.506E-01
2.504E-01
2.511E-01
2.537E-01
2.492E-01
Figure 15. Sample records showing HAPEM5 output for Benzene Runl. Filename is
2015_45201_runl.HAPEM5-TRACT.txt. Variables are FIPS, tract id, major, area & other,
onroad gasoline, nonroad gasoline, total and background concentrations.
01001
01001
01001
01001
01001
01001
01001
01001
01001
01001
01001
01003
01003
01003
01003
01003
01003
01003
01003
020100
020200
020300
020400
020500
020600
020700
020800
020900
021000
021100
010100
010200
010300
010400
010500
010600
010701
010703
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
OOOE+00
OOOE+00
OOOE+00
OOOE+00
OOOE+00
OOOE+00
OOOE+00
OOOE+00
OOOE+00
OOOE+00
OOOE+00
OOOE+00
OOOE+00
OOOE+00
OOOE+00
OOOE+00
OOOE+00
OOOE+00
OOOE+00
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
OOOE+00
OOOE+00
OOOE+00
OOOE+00
OOOE+00
OOOE+00
OOOE+00
OOOE+00
OOOE+00
OOOE+00
OOOE+00
OOOE+00
OOOE+00
OOOE+00
OOOE+00
OOOE+00
OOOE+00
OOOE+00
OOOE+00
2.
5.
4.
4.
3.
3.
2.
1.
9.
7.
1.
3.
6.
8.
6.
1.
8.
1.
1.
836E-03
176E-03
209E-03
552E-03
569E-03
553E-03
575E-03
513E-03
994E-04
444E-04
016E-03
194E-04
738E-04
175E-04
943E-04
069E-03
959E-04
063E-03
256E-03
1.629E-03
3.097E-03
2.422E-03
1.975E-03
1.984E-03
1.895E-03
1.941E-03
1.089E-03
6.306E-04
4.991E-04
7.317E-04
7.389E-04
1.490E-03
2.639E-03
2.991E-03
2.462E-03
2.692E-03
7.130E-03
9.177E-03
4
8
6
6
5
5
4
2
1
1
1
1
2
3
3
3
3
8
1
465E-03
273E-03
631E-03
527E-03
554E-03
448E-03
516E-03
602E-03
630E-03
243E-03
748E-03
058E-03
164E-03
456E-03
685E-03
531E-03
588E-03
193E-03
043E-02
0. OOOE+00
0. OOOE+00
0. OOOE+00
0. OOOE+00
0. OOOE+00
0. OOOE+00
0. OOOE+00
0. OOOE+00
0. OOOE+00
0. OOOE+00
0. OOOE+00
0. OOOE+00
0. OOOE+00
0. OOOE+00
0. OOOE+00
0. OOOE+00
0. OOOE+00
0. OOOE+00
0. OOOE+00
Figure 16. Sample records showing HAPEM5 output for Benzene Run2. Filename is
2015_45201_run2.HAPEM5-TRACT.txt. Variables are FIPS, tract id, major, area and other,
onroad diesel, nonroad other, total and background concentrations.
7.2 Summaries of annual HAPEM5 output
After post-processing of the HAPEM5 concentrations, summary statistics for the concentrations
for each year, including 1999 were calculated as done for ASPEN results. The statistics
included:
79
-------
• Average concentrations for major, area & other, onroad gasoline, onroad diesel, nonroad
gasoline, remaining nonroad, background, and total at the county, state, state urban/rural,
state RFG/non-RFG, national, national urban/rural, and national RFG/non-RFG levels.
• Distributions (5th, 10th, 25th, 50th, 75th, 90th, and 95th percentiles) for total concentrations
at the county, state, state urban/rural, state RFG/non-RFG levels.
• Maps of county median concentrations for 1,3-butadiene, acetaldehyde, acrolein,
benzene, formaldehyde, and naphthalene were generated for 1999, 2015, 2020, and 2030.
Table 35 lists national average HAPEM concentrations for each MS AT HAP for stationary and
mobile sources for each year. Figure 17 shows the county median total concentration for
Benzene for 2015.
The summaries and maps described above can be found in the MSAT rule docket: EPA-HQ-
OAR-2005-0036 in the excel file hapem_concentrations.xls. County median maps are also in the
docket; the file name is: HAPEM_medians.ppt.
80
-------
Table 35. National average stationary and mobile HAPEM concentrations (|jg m'3) for 2015, 2020, and 2030 by HAP.
HAP
1,3 -Butadiene
2,2,4-
Trimethylpentane
Acetaldehyde
Acrolein
Benzene
Chromium III
Chromium VI
Ethyl Benzene
Formaldehyde
Hexane
MTBE
Manganese
Naphthalene
Nickel
POM
Propionaldehyde
Styrene
Toluene
Xylenes
Stationary
1.91X10'2
3.61xlO"2
7.40xlO"2
2.53xlO'2
l.VSxlO"1
6.68xlO'4
1.687xlO"4
l.OSxKT1
1.27X10'1
5.12X10"1
5.95xlO'2
2.48xlO"3
5.25xlO'2
1.03xlO"3
1.40xl(r2
2.80xlO'2
4.09xl(T2
1.03
7.52X10"1
2015
Mobile
2.77xlO'2
3.16X10"1
3.84X10"1
3.40xlO'2
3.54X10"1
1.21xl(r4
2.70xlO"5
1.44X10-1
3.25X10'1
l.SOxlO"1
1.40X10'1
1.23xlO"4
1.34xlO'2
1.35xlO"4
1.12xl(r3
9.68xlO'2
1.36xl(r2
8.06X10'1
5.61X10"1
Background
3.90xlO'2
0
4.00X10"1
0
S.OlxlO"1
0
0
0
5.91X10'1
0
0
0
0
0
0
0
0
0
1.28X10"1
Stationary
1.92xlO'2
3.83xlO"2
7.62xlO"2
2.50xlO'2
1.86X10"1
7.51xlO'4
1.90xlO"4
1.19X10'1
l.SSxlO'1
5.53X10"1
6.27xlO'2
2.74xlO"3
5.61xlO'2
1.12xlO"3
1.46xlO'2
2.86xlO'2
4.61xl(r2
1.14
S.SlxlO"1
2020
Mobile
2.68xlO'2
2.86X10"1
3.47X10"1
3.33xlO'2
3.29X10"1
1.33xlO'4
2.96xlO"5
1.32X10'1
3.12X10'1
l.SOxlO"1
1.14X10'1
1.39xlO"4
1.36xlO'2
1.48xlO"4
1.14X10'3
8.58xlO'2
1.30xlQ-2
7.35X10'1
5.20X10"1
Background
3.90xlO'2
0
4.00X10"1
0
S.OlxlO"1
0
0
0
5.91X10'1
0
0
0
0
0
0
0
0
0
1.27X10"1
Stationary
1.92xlO'2
3.83xlO"2
7.62xlO"2
2.50xlO'2
1.86X10"1
7.51xlO'4
1.90xlO"4
1.19X10'1
l.SSxlO'1
5.53X10"1
6.27xlO'2
2.74xlO"3
5.61xlO'2
1.12xlO"3
1.46xlO'2
2.86xlO'2
4.61xlQ-2
1.14
S.SlxlO"1
2030
Mobile
2.94xlO'2
S.OSxlO"1
3.67X10"1
3.68xlO'2
3.53X10"1
1.60X10'4
3.57xlO"5
1.41X10'1
3.40X10'1
.36X10"1
.14X10'1
.74xlO"4
.57xlO'2
.77xlO"4
.31xlO'3
8.98xlO'2
1.42xlQ-2
7.83X10'1
5.59X10"1
Background
3.90xlO'2
0
4.00X10"1
0
S.OlxlO"1
0
0
0
5.91X10'1
0
0
0
0
0
0
0
0
0
1.27X10"1
81
-------
Note that even though the 2020 and 2030 stationary concentrations input into HAPEM were
identical, the output stationary concentrations for the two years were not exactly the same. This
is due to the post-processing of the raw HAPEM output. The average tract-level source category
concentrations are adjusted by dividing by the tract-level median total concentration. Between
2020 and 2030, the mobile concentrations change, therefore changing the total concentrations.
Therefore, the stationary HAPEM output concentrations change, even though the ASPEN
concentrations are no different between the two years.
0.000-0.293
0.294-0.510
0.511 -0.825
0.826- 1.310
1.311 -2.206
2.207-4.784
Figure 17. 2015 HAPEM county median total concentrations (all sources) for benzene.
82
-------
8. Cancer and non-cancer risk calculations
Once HAPEM runs were completed, cancer and non-cancer risks were calculated for each of the
MSAT HAPs. Table 36 lists the MSAT HAPS with their respective unit risk estimates (URE)
for cancer calculations and non-cancer reference concentrations (Rfc) for non-cancer
calculations, resulting from long term (chronic) inhalation exposure to these HAPS. Also listed
are the HAPs appropriate carcinogenic class and target organ system(s) for non-cancer effects.
Table 36. MSAT HAPs carcinogenic class,
N/A denotes HAP is neither a cancer or non-
URE, non-cancer target organ systems, and Rfc.
•cancer HAP
HAP
1,3 -Butadiene
2,2,4-Trimethylpentane
Acetaldehyde
Acrolein
Benzene
Chromium III
Chromium VI
Ethyl Benzene
Formaldehyde
Hexane
Manganese
MTBE
Naphthalene
Nickel
Propionaldehyde
POM1
POM2
POMS
POM4
POM5
POM6
POM7
POMS
Styrene
Toluene
Xylenes
Carcinogen
Class
A
N/A
B2
A
N/A
A
B
C
A
N/A
B2
B2
B2
B2
B2
B2
B2
B2
URE
S.OxlO'5
N/A
2.2xlO'6
N/A
7.8xlO'6
N/A
1.2xl(r2
N/A
5.5xlO'9
N/A
N/A
N/A
3.4xlO'5
1.6xl(r4
N/A
5.5xlO'5
5.5xlO'5
.OxlO'1
.OxlO'2
.OxlO'3
.OxlO'4
.OxlO'5
2.0xlO'4
N/A
N/A
N/A
Organ systems
Reproductive
N/A
Respiratory
Respiratory
Immune
N/A
Respiratory
Developmental
Respiratory
Respiratory, Neurological
Neurological
Liver, Kidney, Ocular
Respiratory
Respiratory, Immune
N/A
Neurological
Respiratory, Neurological
Neurological
Rfc
2.0xlO'3
N/A
9.0xlO'3
2.0X10'5
S.OxlO'2
N/A
l.OxlO'4
1.0
9.8xlO'3
2.0X10'1
S.OxlO'5
3.0
S.OxlO'3
6.5X10'5
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
1.0
4.0X10'1
l.OxlO'1
URE and Rfc estimates were obtained from hazard and dose-response information that EPA's
Office of Air Quality Planning and Standards posts on the internet ("OAQPS Toxicity Values")
at the following link: www.epa.gov/ttn/fera. This information is updated as new data become
available; the version of the table used for this paper is the same as used for the 1999 NATA (U.
S. EPA, 2005d).
Prior to computing risks, the HAPEM results from runl and run2 were combined into a single
data set containing the major, area & other, onroad gasoline, onroad diesel, nonroad gasoline,
nonroad other, background, and total (all sources) exposure concentrations for each census tract.
For each modeling year, each HAP was contained in its own dataset.
83
-------
In that the 1999 NATA approach involved the use of ACCESS to store exposure results and
®
compute risks, and the MSAT approach involved the use of a series of SAS programs, quality
assurance was performed on the SAS® programs to ensure the same results were obtained as
would have been under the ACCESS approach.
8.1 Cancer risk calculations
To calculate cancer risks and summary statistics for 1999, 2015, 2020, and 2030, the HAPEM
concentrations for each HAP were multiplied by the corresponding URE for the HAP. This was
done for each source sector (and background) in each census tract to produce tract-level, source
sector level cancer risks. Appendix D (D.I) provides the programming steps.
After calculating the risks, the tract level risks for each HAP, risk for each carcinogen class, and
risk across all HAPs were summarized at the same levels as done for the ASPEN and HAPEM5
outputs. Once statistics were calculated for each year and level (county, state, etc.) they were
merged together. Maps of county median risk were generated for several HAPs and total risk.
The total risk map for 2015 is shown in Figure 18. Table 37 lists national and stationary and
mobile risks for the cancer risk HAPs, carcinogen class, and total risk across all HAPs.
More detailed summaries can be found in the MSAT rule docket: EPA-HQ-OAR-2005-0036 in
the excel file named hapem_risks.xls. County median risk maps are also in the docket; the file
name is: risk_030305.ppt.
84
-------
Figure 18. County median total inhalation cancer risks for all MS AT HAPs for 2015. Risk is
characterized as N in a million.
Table 37. National avera
HAPs, each carcinogenic
ge inhalation cancer risks for stationary and mobile sources for MSAT
class and total risk (all MSAT HAPs).
HAP
1,3 -Butadiene
Benzene
Chromium VI
Nickel
Class A
Formaldehyde
Class Bl
Acetaldehyde
POM
Class B2
Naphthalene
Class C
Total Risk
Carcinogen
Class
A
A
A
A
A
Bl
Bl
B2
B2
B2
C
C
All
2015
stationary
5.73xlO"7
1.38xlO"6
2.01xlO"6
1.64xlO"7
4.14X106
V.OlxlO'10
7.01 xlO'10
1.63xlO"7
1.31xlO"6
1.47X106
1.79xlO"6
1.79X106
7.39X10"6
mobile
8.30xl(T7
2.76xl(T6
3.24xl(T7
2.17X10'8
3.94X106
1.79X10'9
1.79X109
8.45 xlO'7
7.29xl(T8
9.18X107
4.57xl(T7
4.57X107
5.31X106
2020
stationary
5.76xl(T7
1.45X10'6
2.28xl(T6
l.SOxlO'7
4.48X106
7.57xlO"10
7.57X1Q-10
1.68xlO'7
1.36xlO'6
1.53X106
1.91X1Q-6
1.91X106
7.92X10'6
mobile
8.03 xlO'7
2.57xl(T6
3.56xlO'7
2.37xlO'8
3.75X10'6
1.73 xlO'9
1.73X10'9
7.64xlO'7
7.44xlO'8
8.38X107
4.63 xl(T7
4.63X107
5.05X10'6
2030
stationary
5.76xlO'7
1.45X10'6
2.28xlO'6
l.SOxlO'7
4.48X106
7.57X10'10
7.57X1Q-10
1.68xlO'7
1.36xlO'6
1.53X106
1.91X1Q-6
1.91X106
7.92X10'6
mobile
8.82xl(T7
2.75 xlO'6
4.29xlO'7
2.83X10'8
4.09X106
1.87xlO'9
1.87X10'9
S.OSxlO'7
8.51X10'8
8.931 xlO7
6.35xl(T7
6.35X107
5.52X10"6
8.2 Non-cancer risk calculations
Tract level non-cancer hazard quotients (HQ) for each HAP were calculated by dividing, for
each HAP and each source sector, the exposure concentration by the Rfc. A hazard index was
85
-------
computed for each organ system by summing the HQs over all HAPs that share the same target
organ system. Appendix D (D.2) describes the programming steps.
The tract level HQ and HI estimates for each HAP and organ system respectively, were
summarized at the same levels as done for the ASPEN and HAPEM5 outputs. Once statistics
were calculated for each year and level (county, state, etc.) they were merged together. Maps of
county median HQ for several HAPs and respiratory HI were generated. The respiratory system
HI map for 2015 is shown in Figure 19. Table 38 lists national and stationary and mobile HI for
the non-cancer HAPs, and the organ systems affected by the MSAT HAPs.
More detailed summaries can be found in the MSAT rule docket: EPA-HQ-OAR-2005-0036 in
the excel file named hapem_hq.xls. County median non-cancer risk maps are also in the docket;
the file name is hq_030305.ppt.
86
-------
Hazard Index
0.000-0.883
J 0.884-2.175
J 2.176-5.099
J 5.100- 10.182
| 10.183-16.707
I 16.708-31.635
Figure 19. County median total (all sources) non-cancer hazard index for MSAT HAPs
affecting the respiratory system.
87
-------
Table 38. National average non-cancer hazard quotient (HQ) for MSAT HAPs and hazard index
(HI) for organ systems for stationary and mobile sources.
HAP
1,3 -Butadiene
Acetaldehyde
Acrolein
Benzene
Chromium VI
Ethyl Benzene
Formaldehyde
Hexane
MTBE
Manganese
Naphthalene
Nickel
Styrene
Toluene
Xylenes
Organ
Systems
Developmental
Immunological
Kidney
Liver
Neurological
Ocular
Reproductive
Respiratory
2015
stationary
9.55xlO-3
8.22xlO-3
1.27
5.92xlO-3
1.68xlO-3
l.OSxlO'4
l.SOxlO'2
2.56xl(T3
1.98X10'5
4.95 xlO-2
1.75xlO-2
1.58xlO-2
4.09xlO-5
2.58xlO-3
7.52xlO-3
mobile
1.38xlO'2
4.27xlO'2
1.70
LlSxlO'2
2.70xlO-4
1.44X10'4
3.32xlO'2
7.52xlO-4
4.67xlO'5
2.46xlO'3
4.48xl(T3
2.08xlO-3
1.36X10'5
2.01 xlO-3
5.61X10'3
2015
stationary
l.OSxlO'4
2.17X10'2
1.98X10'5
1.98X10'5
6.22xl(T2
1.98X10'5
9.55xlO-3
1.33
mobile
1.44X10'4
1.39xlO'2
4.67xlO'5
4.67xlO'5
1.09X10'2
4.67xlO'5
1.38xlO'2
1.79
2020
stationary
9.60xl(T3
8.47xlO'3
1.25
6.19X10'3
1.90X10'3
1.19X10'4
1.40X10'2
2.77xlO'3
2.09X10'5
5.48xl(T2
1.87xlO'2
1.73 xlO'2
4.61 xlO'5
2.84xlO'3
8.31 xl(T3
mobile
1.39X10'2
3.86xlO'2
1.67
LlOxlO'2
2.97xlO'4
1.32xlO'4
3.19X10'2
6.51X10'4
3.82xl(T5
2.77xlO'3
4.54xlO'3
2.28xlO'3
l.SOxlO'5
1.84xlO'3
5.20xlO'3
2020
stationary
1.19X10'4
2.35xlO'2
2.09xl(T5
2.09X10'5
6.87xlO'2
2.09X10'5
9.60xlO'3
1.31
mobile
1.32xlO'4
1.32xlO'2
3.82xl(T5
3.82xlO'5
l.OSxlO'2
3.82xlO'5
1.39xlO'2
1.75
2030
stationary
9.60xl(T3
8.47xlO'3
1.25
6.19X10'3
1.90X10'3
1.19X10'4
1.40X10'2
2.77xl(T3
2.09X10'5
5.48xlO'2
1.87xlO'2
1.73 xl(T2
4.61 xlO'5
2.84xlQ-3
8.31xlQ-3
mobile
1.47X1Q-2
4.08xlQ-2
1.84
LlSxlQ-2
3.57xlQ-4
1.41X1Q-4
3.47xlQ-2
6.80xlQ-4
3.80xlQ-5
3.47xlQ-3
5.24xlQ-3
2.72xlQ-3
1.42X1Q-5
1.96X1Q-3
5.59xlQ-3
2030
stationary
1.19X1Q-4
2.35xlQ-2
2.09xlQ-5
2.09xlO-5
6.87xlQ-2
2.09xlQ-5
9.60xlQ-3
1.31
mobile
1.41X1Q-4
1.45X1Q-2
3.80xlQ-5
3.80xlO-5
1.17X1Q-2
3.80xlQ-5
1.47X1Q-2
1.93
8.3 Cancer and non-cancer risk population statistics using 2000 and projected population
Tract level population and risks were used to develop population statistics for base and future
years. We used the same county-level projected population data as is used in BenMAP (Abt,
2005) for 2015, 2020, and 2030, which originated from Woods and Poole
(www.woodsandpoole.com). The projected population data was for the contiguous 48 states, not
Alaska and Hawaii. Therefore, the statistics were for the contiguous 48 state region.
Also, populations statistics were calculated using 2000 census population for all years' future
year risks. Following is the methodology used for the population statistic calculations.
8.3.1 Allocation of future county level populations to tract level
As previously noted, the projected populations for 2015, 2020, and 2030 were at the county
level. Since the calculated risks and non-cancer HQ or HI estimates were at tract level, the
county level future year populations were allocated to the tracts using the 2000 census based
tract-level to county-level population ratio (this is also spatial surrogate code 100 [population]
used by EMS-HAP Version 3.0). Each tract's ratio was applied to the county's projected
88
-------
population to calculate a future year's tract population. For example, for Wake County, NC
(FIPS=37183), the 2000 county population was 627,846. The 2000 population of tract 050100
was 1,847, resulting in a ratio of 0.003. The projected 2015 population of Wake County was
867,680.75. Applying the 2000 population ratio to the 2015 county population resulted in a
projected tract population of 2,552.55. This was done for each tract in the 48 contiguous U.S.
states.
8.3.2 Population statistic calculations for cancer risk
Once the 2015, 2020, and 2030 populations had been allocated to tract level, the populations
were then merged by FIPS and tract with the tract level total risk (across all MSAT HAPs)
estimates. Next, for each year and source sector, population totals were calculated based on
three risk benchmark values: IxlO"4, IxlO"5, and IxlO"6.
For example, for major source total risk for 2015, each tract level major risk was checked to see
how it compared to the three values listed above. The following formulas show the checks and
the population calculations:
If tract major source total risk > 10"4 then
major _4 = major _4 + poptract (11)
If tract major source total risk > 10"5 then
major _5= major _5+pop tmct (12)
If tract major source total risk > 10"6 then
major _6 = major _6 + poptract (13)
If tract major source total risk < 10"6 then
major _7 = major _7 + poptract (14)
where major_4, major_5, major_6, and major_7 are the running totals and poptract is the tract
population for 2015. Note, initially, before any checks of the tracts, major_4, major_5, major_6,
and major_7 are initialized to zero.
From the logic, a tract's 2015 population was added to each major_4, major_5, and major_6. For
example if the major source risk was 1.1 then the tract's population was added to major_4,
major_5, and major_6. After all tracts in the contiguous 48 states had been checked, each
running total represented a national population affected by each of the three risk classes listed.
These populations were then binned so that they became mutually exclusive, i.e., no overlap, and
were plotted on charts or summarized. The following formulas were used to bin the populations:
For population affected by risk > 10"5 but < 10"4
major _5a = major _5 - major _4
r
(15)
89
-------
For population affected by risk > 10" but < 10"
major _6a = major _6 - major _5
(16)
where major_5a is the population affected by risk > 10"5 but < 10"4 and major_6a is the
population affected by risk > 10"6 but < 10"5. Note that major_4 and major_7 do not need to be
modified.
The steps described in this section were performed for each source sector, major, area & other,
onroad gasoline, onroad diesel, total onroad, nonroad gasoline, non-gasoline nonroad, total
nonroad, background, and total risk (all source sectors). Table 39 shows the results for onroad
nonroad for 1999, 2015, 2020, and 2030. Note that Alaska and Hawaii populations are not
included.
Table 39. Population risk classes for mobile total risk for 2015, 2020, and 2030 using projected
populations for each year.
Source
Category
Onroad
Nonroad
Population Class
Risk > 10'4
10"5 < Risk <10"4
lO'6 < Risk <10'5
Risk <10"6
Total Population
Risk > lO'4
10"5 < Risk <10"4
lO'6 < Risk <10'5
Risk <10'6
Total Population
Populations
Year
1999
208,150
112,848,474
145,060,999
21,465,809
279,583,432
22,272
2,630,188
180,439,149
96,491,823
279,583,432
2015
0
19,596,469
241,185,986
56,122,217
316,904,672
23,710
1,365,537
150,013,784
165,501,640
316,904,672
2020
0
16,703,891
249,373,492
63,615,359
329,692,742
25,123
1,584,116
159,142,708
168,940,795
329,692,742
2030
0
21,839,016
269,464,226
64,592,322
355,895,564
27,986
2,215,401
18,553,8098
168,114,078
355,895,564
8.3.3 Population statistic calculations for non-cancer respiratory hazard index
A similar procedure was used for the respiratory system hazard index. The threshold HI values
used for binning purposes were 10, 1, and 0.1. As described above, each respiratory HI for each
source sector and year were compared against these values and a running total kept for three
populations. For example for 2015 for major source HI values, the following conditions and
equations are used:
If tract major source respiratory HI > 10 then
major _10 = major _10+ poptract
If tract major source respiratory HI > 1 then
major _1 = major _1 + poptract
(17)
(18)
90
-------
If tract major source respiratory HI > 0.1 then
major _01 = major _01+ poptract
If tract major source respiratory HI< 0.1 then
major _0 = major _0 + poptract
(19)
(20)
where major_10, major_l, major_01, and major_0 are the running totals and poptract is the tract
population for 2015. Note, initially the four running totals are set to zero before any checks.
As with the risk calculations, from the logic, a tract's 2015 population can be added to each
major_10, major_l, and major_01. For example if the major source respiratory HI was 11 then
the tract's population would be added to major_10, major_l, and major_01. After all tracts in
the contiguous 48 states have been checked, each running total represented a national population
affected by each of the three HI classes listed. These populations were then binned so that they
become mutually exclusive, i.e., no overlap. The following formulas were used to bin the
populations:
For population affected by HI > 1 but < 10
major_la = major_1- major_10 (21)
For population affected by HI > 0.1 but < 1
major_01a = major_01- major_1
(22)
where major_la is the population affected by HI > 1 but < 10 and major_01a is the population
affected by HI > 0.1 but < 1. Major_10 and major_0 were not modified.
The steps described in this section were performed for each source sector, major, area & other,
onroad gasoline, onroad diesel, total onroad, nonroad gasoline, non-gasoline nonroad, total
nonroad, background, and total HI (all source sectors). Table 40 shows the results for onroad
nonroad for 1999, 2015, 2020, and 2030. Note that Alaska and Hawaii populations are not
included.
91
-------
Table 40. Population respiratory HI classes for mobile sources for 2015, 2020, and 2030 using
projected populations for each year.
Source
Category
Onroad
Nonroad
Population Class
HI> 10
1 < HI < 10
0.110
1 < HI < 10
0.1
-------
9. Background concentration sensitivity analysis
For the air quality modeling, background concentrations that were added to the ASPEN-modeled
concentrations for subsequent HAPEM modeling (Section 7) and risk calculations (Section 8)
were assumed to remain the same across the modeled future years, 2015, 2020, and 2030. The
values used were the same background values used for the 1999 NATA background
concentrations. The background concentrations were at the county level so for a given HAP with
a background value, all tracts in a county received the same background. Details about the
development of the 1999 background concentrations can be found in Batelle (2003).
Because background concentrations added are assumed to account for medium-to-long range
transport of emissions, it is expected that they would decrease or increase as emissions increased
or decreased. A sensitivity analysis to determine the potential magnitude of such background
concentration changes was made for 1,3-butadiene, acetaldehyde, benzene, formaldehyde, and
xylenes for the following years: 2015, 2020, and 2030. The sensitivity analysis included
adjusting the 1999 background concentrations using the change in emissions between 1999 and
these future years. Following is the methodology and results of the analysis.
For each county in the U.S. (excluding Puerto Rico and Virgin Islands), the emissions from all
sources in the county were summed together for each year, 1999, 2015, 2020 and 2030. Next,
the emissions for that county and the surrounding counties whose county centroids (supplied
from the EMS-HAP ancillary file cty_cntr99.sas7bdat, found in the MS AT rule docket EPA-HQ-
OAR-2005-0036) were within 300 km of the county were summed together for each year,
resulting in emission totals for 1999, 2015, 2020, and 2030. These summed emissions were
assigned to the county being analyzed. After summation, for each year, 2015, 2020, and 2030,
the emissions were divided by the 1999 emissions to create a scaling factor to multiply with the
1999 background concentration.
As an example, consider Wake County, NC (FIPS=37183). Figure 20 shows the 209 counties
(including Wake County) whose centroids are within 300 km of Wake County's centroid. These
counties cover most of North Carolina with some in Virginia and South Carolina. The total
benzene emissions for 1999, 2015, 2020, and 2030 can be seen in Table 41. The ratio to apply to
the 1999 background was calculated by dividing the future year's emissions (2015, 2020, or
2030) by the 1999 emissions. This ratio was then applied to the 1999 background, and the new
scaled background was added to the total model concentrations at each tract. Table 41 also
shows the scaled backgrounds used for 2015, 2020, and 2030 for benzene.
93
-------
Figure 20. Counties within 300 km of the centroid of Wake County, North Carolina (county in
gray). Dots represent county centroids.
Table 41. Total benzene emissions of counties within 300 km of Wake County, NC for 1999,
2015, 2020 and 2030, 1999 background benzene concentration for Wake County, and scaled
background concentrations for Wake County for 2015, 2020, and 2030.
Year
1999
2015
2020
2030
Emissions
(tons)
15,692.404
9,364.748
9,225.224
9,552.035
Emissions Ratios
(Future year/1999)
1
0.59677
0.58788
0.608704
Background
Concentration (jig m 3)
(1999 background x Ratio)
0.403794
0.242117
0.238277
0.246566
Table 42 shows national average background concentrations for 1999 and future years. Table 43
shows total concentrations (all sources plus background) using both the 1999 background for
each year and using the scaled background for each year for the five HAPs studied in the
analysis. The analysis showed how much the scaling affected background concentrations but
also showed how little background changed between 2015, 2020, and 2030. This can be seen
both for one county in Table 41 and for the whole country in Table 42. In Figure 21, the changes
for the benzene background between the four years can be seen. The 1999 background
concentrations are generally higher than 2015, 2020, and 2030. One outcome of the analysis was
a change in spatial variability of background for xylenes. For 1999, the entire country received
the same background for xylenes (0.17 |j,g m"3). Scaling background by emissions created a
spatial variability of xylenes background (Figure 22).
94
-------
Detailed summaries of scaled background concentrations for the five HAPs summaries can be
found in the MSAT rule docket: EPA-HQ-OAR-2005-0036 in the excel file named
background_test.xls with maps for the HAPs in background_acetaldehyde_011 l.ppt,
background_butadiene_0111 .ppt, background_test_0111 .ppt (for benzene),
background_formaldehyde_0111 .ppt, and background_xylenes_0111 .ppt.
Table 42. National average 1999 background and scaled backgrounds for 1,3-butadiene,
acetaldehyde, benzene, formaldehyde, and xylenes.
HAP
1,3 -Butadiene
Acetaldehyde
Benzene
Formaldehyde
Xylenes
Background concentrations (jig m 3)
1999
5.12xlO"2
5.17X10"1
3.94X10"1
7.62X10"1
l.VOxlO"1
2015
2.86xlO"2
3.29X10"1
2.38X10"1
4.96X10"1
LlSxlO"1
2020
2.83 xlO'2
3.28X10"1
2.32X10"1
5.05X10"1
l.lVxlO"1
2030
2.95 xlO"2
3.36X10"1
2.40X10"1
5.21X10'1
1.20X10"1
Table 43. National average total concentrations (all sources and background) for 2015, 2020,
and 2030 using both the 1999 background and the scaled backgrounds.
HAP
1,3 -Butadiene
Acetaldehyde
Benzene
Formaldehyde
Xylenes
Total concentrations (jig m 3) using
1999 background
2015
9.81xlO"2
9.66X10"1
9.13X10"1
1.22
1.55
2020
9.77xlO"2
9.36X10"1
9.02X10"1
1.22
1.61
2030
l.OOxlO"1
9.56X10"1
9.24X10"1
1.25
1.65
Total concentrations (jig m 3) using
scaled background concentrations
2015
7.58xlO"2
7.77X10"1
7.57X10"1
9.57X10"1
1.50
2020
7.50xlO"2
7.47X10"1
7.40X10"1
9.68X10"1
1.56
2030
7.86xlO"2
7.78X10"1
7.71 xlO'1
1.01
1.60
95
-------
Figure 21. Benzene background concentrations (ug m" ) for a) 1999 background, b) 2015 scaled
background c) 2020 scaled background and d) 2030 scaled background.
a
Figure 22. Xylenes background concentrations (ug m" ) for a) 1999 background, b) 2015 scaled
background c) 2020 scaled background and d) 2030 scaled background.
While background scaling did not change background concentrations much between 2015, 2020,
and 2030, concentrations did differ between the three future years and 1999. Scaling the
background using the projected emissions can add a spatial variation to background
concentration.
96
-------
10. Benzene Control Scenario
This section details the methodology used to develop the controlled inventories for 2015, 2020
and 2030 for benzene, formaldehyde, acetaldehyde, acrolein, and 1,3-butadiene as part of the
benzene control scenario. Controls were applied to gasoline marketing and distribution
emissions, onroad gasoline refueling emissions, onroad gasoline emissions, and nonroad gasoline
emissions.
10.1 Stationary gasoline distribution and vehicle gasoline refueling inventory
For the stationary inventories, controls were applied to benzene emissions only for gasoline
marketing and distribution and onroad gasoline refueling for 2015 and 2020. For 2030, these
emissions (along with all other stationary source emissions for the reference case) were set equal
to 2020 emissions. Table 44 lists the gasoline distribution SCC codes contained in the reference
case inventories, for which controlled emissions were estimated. Onroad gasoline refueling SCC
codes can be found in Table 23.
Gasoline marketing and distribution emissions were estimated for the control scenario by
applying a county specific control ratio based on the change in average fuel benzene level for the
control and reference case. Average fuel benzene level for the control case was determined from
refinery modeling done for the rule. As part of the refinery modeling, average fuel properties for
each Petroleum Administration for Defense District (PADD) under the new standards were
estimated. Average fuel benzene levels for conventional gasoline (CG) and reformulated
gasoline (RFG) in each PADD before and after implementation of the proposed standards were
used to develop multiplicative factors. These multiplicative factors were used as control ratios
for estimating the controlled gasoline marketing and distribution emissions. They were also
applied to the reference case fuel benzene levels for each county in the NMIM database to use
for generating the NMIM controlled case emissions, which were used to develop control
inventories for the other categories discussed in this section.
The multiplicative factors (control ratios for gasoline marketing and distribution emissions) are
shown in Table 45. Although California is part of PADD5, it was treated separately since
California has its own reformulated gasoline program. PADD regions are shown in Figure 23.
To apply the control ratios to the gasoline marketing and distribution SCCs, it was necessary to
distinguish between the counties in each PADD using RFG versus CG. Figure 24 shows which
counties are RFG counties.
Onroad gasoline refueling emissions were estimated for the control scenario by calculating a
county specific ratio of control to reference case NMEVI refueling emissions for benzene for
2015, 2020 and 2030. NMIM was rerun for refueling emissions for the control case with revised
gasoline input parameters as described in 10.2.
Appendix E describes the steps used to develop the controlled benzene emissions for gasoline
marketing and distribution and onroad gasoline refueling. Sample calculations are also provided.
97
-------
Table 44. Benzene gasoline marketing and distribution SCC codes to be controlled.
SCC
2501000000
2501060050
2501060052
2501060200
2501080000
2501080100
2505000000
2505010000
2505020120
2505030120
40400102
40400104
40400106
40400108
Description
Storage and Transport; Petroleum and Petroleum Product
Storage; All Storage Types: Breathing Loss; Total: All
Products
Storage and Transport; Petroleum and Petroleum Product
Storage; Gasoline Service Stations; Stage 1: Total
Storage and Transport; Petroleum and Petroleum Product
Storage; Gasoline Service Stations; Stage 1: Splash Filling
Storage and Transport; Petroleum and Petroleum Product
Storage; Gasoline Service Stations; Underground Tank: Total
Aviation Gasoline Distribution: Stage 1 & II
Aviation Gasoline Storage -Stage II
Storage and Transport; Petroleum and Petroleum Product
Transport; All Transport Types; Total: All Products
Storage and Transport; Petroleum and Petroleum Product
Transport; Rail Tank Car; Total: All Products
Storage and Transport; Petroleum and Petroleum Product
Transport; Marine Vessel; Gasoline
Storage and Transport; Petroleum and Petroleum Product
Transport; Truck; Gasoline
Petroleum and Solvent Evaporation; Petroleum Liquids
Storage (non-Refinery); Bulk Terminals; Gasoline RVP 10:
Breathing Loss (67000 Bbl Capacity) - Fixed Roof Tank
Petroleum and Solvent Evaporation; Petroleum Liquids
Storage (non-Refinery); Bulk Terminals; Gasoline RVP 13:
Breathing Loss (250000 Bbl Capacity)-Fixed Roof Tank
Petroleum and Solvent Evaporation; Petroleum Liquids
Storage (non-Refinery); Bulk Terminals; Gasoline RVP 7:
Breathing Loss (250000 Bbl Capacity) - Fixed Roof Tank
Petroleum and Solvent Evaporation; Petroleum Liquids
Storage (non-Refinery); Bulk Terminals; Gasoline RVP 10:
Working Loss (Diameter Independent) - Fixed Roof Tank
SCC
2501050120
2501060051
2501060053
2501060201
2501080050
2501995120
2505000120
2505020000
2505020121
40400101
40400103
40400105
40400107
40400109
Description
Storage and Transport; Petroleum and Petroleum Product
Storage; Bulk Stations/Terminals: Breathing Loss; Gasoline
Storage and Transport; Petroleum and Petroleum Product
Storage; Gasoline Service Stations; Stage 1: Submerged
Filling
Storage and Transport; Petroleum and Petroleum Product
Storage; Gasoline Service Stations; Stage 1: Balanced
Submerged Filling
Storage and Transport; Petroleum and Petroleum Product
Storage; Gasoline Service Stations; Underground Tank:
Breathing and Emptying
Aviation Gasoline Storage -Stage I
Storage and Transport; Petroleum and Petroleum Product
Storage; All Storage Types: Working Loss; Gasoline
Storage and Transport; Petroleum and Petroleum Product
Transport; All Transport Types; Gasoline
Storage and Transport; Petroleum and Petroleum Product
Transport; Marine Vessel; Total: All Products
Marine Vessel Operations - Barge Handling of Gasoline
Petroleum and Solvent Evaporation; Petroleum Liquids
Storage (non-Refinery); Bulk Terminals; Gasoline RVP 13:
Breathing Loss (67000 Bbl Capacity) - Fixed Roof Tank
Petroleum and Solvent Evaporation; Petroleum Liquids
Storage (non-Refinery); Bulk Terminals; Gasoline RVP 7:
Breathing Loss (67000 Bbl. Capacity) - Fixed Roof Tank
Petroleum and Solvent Evaporation; Petroleum Liquids
Storage (non-Refinery); Bulk Terminals; Gasoline RVP 10:
Breathing Loss (250000 Bbl Capacity)-Fixed Roof Tank
Petroleum and Solvent Evaporation; Petroleum Liquids
Storage (non-Refinery); Bulk Terminals; Gasoline RVP 13:
Working Loss (Diam. Independent) - Fixed Roof Tank
Petroleum and Solvent Evaporation; Petroleum Liquids
Storage (non-Refinery); Bulk Terminals; Gasoline RVP 7:
Working Loss (Diameter Independent) - Fixed Roof Tank
98
-------
Table 44. Continued.
sec
40400110
40400112
40400114
40400116
40400118
40400120
40400132
40400142
40400148
40400151
40400153
40400161
Description
Petroleum and Solvent Evaporation; Petroleum Liquids
Storage (non-Refinery); Bulk Terminals; Gasoline RVP 13:
Standing Loss (67000 Bbl Capacity)-Floating Roof Tank
Petroleum and Solvent Evaporation; Petroleum Liquids
Storage (non-Refinery); Bulk Terminals; Gasoline RVP 7:
Standing Loss (67000 Bbl Capacity)- Floating Roof Tank
Petroleum and Solvent Evaporation; Petroleum Liquids
Storage (non-Refinery); Bulk Terminals; Gasoline RVP 10:
Standing Loss (250000 Bbl Cap.) - Floating Roof Tank
Petroleum and Solvent Evaporation; Petroleum Liquids
Storage (non-Refinery); Bulk Terminals; Gasoline RVP
13/10/7: Withdrawal Loss (67000 Bbl Cap.) - Float RfTnk
Petroleum and Solvent Evaporation; Petroleum Liquids
Storage (non-Refinery); Bulk Terminals; Gasoline RVP 13:
Filling Loss (10500 Bbl Cap.) - Variable Vapor Space
Petroleum and Solvent Evaporation; Petroleum Liquids
Storage (non-Refinery); Bulk Terminals; Gasoline RVP 7:
Filling Loss (10500 Bbl Cap.) - Variable Vapor Space
Petroleum and Solvent Evaporation; Petroleum Liquids
Storage (non-Refinery); Bulk Terminals; Gasoline RVP 10:
Standing Loss - Ext. Floating Roof w/ Primary Seal
Petroleum and Solvent Evaporation; Petroleum Liquids
Storage (non-Refinery); Bulk Terminals; Gasoline RVP 10:
Standing Loss - Ext. Floating Roof w/ Secondary Seal
Petroleum and Solvent Evaporation; Petroleum Liquids
Storage (non-Refinery); Bulk Terminals; Gasoline RVP
13/10/7: Withdrawal Loss - Ext. Float Roof (Pri/Sec Seal)
Petroleum and Solvent Evaporation; Petroleum Liquids
Storage (non-Refinery); Bulk Terminals; Valves, Flanges, and
Pumps
Petroleum and Solvent Evaporation; Petroleum Liquids
Storage (non-Refinery); Bulk Terminals; Vapor Control Unit
Losses
Petroleum and Solvent Evaporation; Petroleum Liquids
Storage (non-Refinery); Bulk Terminals; Gasoline RVP 13:
Standing Loss - Int. Floating Roof w/ Primary Seal
sec
40400111
40400113
40400115
40400117
40400119
40400131
40400141
40400143
40400150
40400152
40400154
40400162
Description
Petroleum and Solvent Evaporation; Petroleum Liquids
Storage (non-Refinery); Bulk Terminals; Gasoline RVP 10:
Standing Loss (67000 Bbl Capacity)-Floating Roof Tank
Petroleum and Solvent Evaporation; Petroleum Liquids
Storage (non-Refinery); Bulk Terminals; Gasoline RVP 13:
Standing Loss (250000 Bbl Cap.) - Floating Roof Tank
Petroleum and Solvent Evaporation; Petroleum Liquids
Storage (non-Refinery); Bulk Terminals; Gasoline RVP 7:
Standing Loss (250000 Bbl Cap.) - Floating Roof Tank
Petroleum and Solvent Evaporation; Petroleum Liquids
Storage (non-Refinery); Bulk Terminals; Gasoline RVP
13/10/7: Withdrawal Loss (250000 Bbl Cap.) - Float Rf Tnk
Petroleum and Solvent Evaporation; Petroleum Liquids
Storage (non-Refinery); Bulk Terminals; Gasoline RVP 10:
Filling Loss (10500 Bbl Cap.) - Variable Vapor Space
Petroleum and Solvent Evaporation; Petroleum Liquids
Storage (non-Refinery); Bulk Terminals; Gasoline RVP 13:
Standing Loss - Ext. Floating Roof w/ Primary Seal
Petroleum and Solvent Evaporation; Petroleum Liquids
Storage (non-Refinery); Bulk Terminals; Gasoline RVP 13:
Standing Loss - Ext. Floating Roof w/ Secondary Seal
Petroleum and Solvent Evaporation; Petroleum Liquids
Storage (non-Refinery); Bulk Terminals; Gasoline RVP 7:
Standing Loss - Ext. Floating Roof w/ Secondary Seal
Petroleum and Solvent Evaporation; Petroleum Liquids
Storage (non-Refinery); Bulk Terminals; Miscellaneous
Losses/Leaks: Loading Racks
Petroleum and Solvent Evaporation; Petroleum Liquids
Storage (non-Refinery); Bulk Terminals; Vapor Collection
Losses
Petroleum and Solvent Evaporation; Petroleum Liquids
Storage (non-Refinery); Bulk Terminals; Tank Truck Vapor
Leaks
Petroleum and Solvent Evaporation; Petroleum Liquids
Storage (non-Refinery); Bulk Terminals; Gasoline RVP 10:
Standing Loss - Int. Floating Roof w/ Primary Seal
99
-------
Table 44. Continued.
sec
40400163
40400172
40400178
40400202
40400204
40400206
40400208
40400210
40400213
40400241
40400250
40400252
Description
Petroleum and Solvent Evaporation; Petroleum Liquids Storage
(non-Refinery); Bulk Terminals; Gasoline RVP 7: Standing
Loss - Internal Floating Roof w/ Primary Seal
Petroleum and Solvent Evaporation; Petroleum Liquids Storage
(non-Refinery); Bulk Terminals; Gasoline RVP 10: Standing
Loss - Int. Floating Roof w/ Secondary Seal
Petroleum and Solvent Evaporation; Petroleum Liquids Storage
(non-Refinery); Bulk Terminals; Gasoline RVP 13/10/7:
Withdrawal Loss - Int. Float Roof (Pri/Sec Seal)
Petroleum and Solvent Evaporation; Petroleum Liquids Storage
(non-Refinery); Bulk Plants; Gasoline RVP 10: Breathing Loss
(67000 Bbl Capacity) - Fixed Roof Tank
Petroleum and Solvent Evaporation; Petroleum Liquids Storage
(non-Refinery); Bulk Plants; Gasoline RVP 13: Working Loss
(67000 Bbl. Capacity) - Fixed Roof Tank
Petroleum and Solvent Evaporation; Petroleum Liquids Storage
(non-Refinery); Bulk Plants; Gasoline RVP 7: Working Loss
(67000 Bbl. Capacity) - Fixed Roof Tank
Petroleum and Solvent Evaporation; Petroleum Liquids Storage
(non-Refinery); Bulk Plants; Gasoline RVP 10: Standing Loss
(67000 Bbl Cap.) - Floating Roof Tank
Petroleum and Solvent Evaporation; Petroleum Liquids Storage
(non-Refinery); Bulk Plants; Gasoline RVP 13/10/7:
Withdrawal Loss (67000 Bbl Cap.) - Float RfTnk
Petroleum and Solvent Evaporation; Petroleum Liquids Storage
(non-Refinery); Bulk Plants; Gasoline RVP 7: Filling Loss
(10500 Bbl Cap.) - Variable Vapor Space
Petroleum and Solvent Evaporation; Petroleum Liquids Storage
(non-Refinery); Bulk Plants; Gasoline RVP 13: Standing Loss -
Ext. Floating Roof w/ Secondary Seal
Petroleum and Solvent Evaporation; Petroleum Liquids Storage
(non-Refinery); Bulk Plants; Loading Racks
Petroleum and Solvent Evaporation; Petroleum Liquids Storage
(non-Refinery); Bulk Plants; Miscellaneous Losses/Leaks:
Vapor Collection Losses
sec
40400171
40400173
40400201
40400203
40400205
40400207
40400209
40400212
40400231
40400242
40400251
40400253
Description
Petroleum and Solvent Evaporation; Petroleum Liquids
Storage (non-Refinery); Bulk Terminals; Gasoline RVP 13:
Standing Loss - Int. Floating Roof w/ Secondary Seal
Petroleum and Solvent Evaporation; Petroleum Liquids
Storage (non-Refinery); Bulk Terminals; Gasoline RVP 7:
Standing Loss - Int. Floating Roof w/ Secondary Seal
Petroleum and Solvent Evaporation; Petroleum Liquids
Storage (non-Refinery); Bulk Plants; Gasoline RVP 13:
Breathing Loss (67000 Bbl Capacity) - Fixed Roof Tank
Petroleum and Solvent Evaporation; Petroleum Liquids
Storage (non-Refinery); Bulk Plants; Gasoline RVP 7:
Breathing Loss (67000 Bbl. Capacity) - Fixed Roof Tank
Petroleum and Solvent Evaporation; Petroleum Liquids
Storage (non-Refinery); Bulk Plants; Gasoline RVP 10:
Working Loss (67000 Bbl. Capacity) - Fixed Roof Tank
Petroleum and Solvent Evaporation; Petroleum Liquids
Storage (non-Refinery); Bulk Plants; Gasoline RVP 13:
Standing Loss (67000 Bbl Cap.) - Floating Roof Tank
Petroleum and Solvent Evaporation; Petroleum Liquids
Storage (non-Refinery); Bulk Plants; Gasoline RVP 7:
Standing Loss (67000 Bbl Cap.) - Floating Roof Tank
Petroleum and Solvent Evaporation; Petroleum Liquids
Storage (non-Refinery); Bulk Plants; Gasoline RVP 10:
Filling Loss (10500 Bbl Cap.) - Variable Vapor Space
Petroleum and Solvent Evaporation; Petroleum Liquids
Storage (non-Refinery); Bulk Plants; Gasoline RVP 13:
Standing Loss - Ext. Floating Roof w/ Primary Seal
Petroleum and Solvent Evaporation; Petroleum Liquids
Storage (non-Refinery); Bulk Plants; Gasoline RVP 10:
Standing Loss - Ext. Floating Roof w/ Secondary Seal
Petroleum and Solvent Evaporation; Petroleum Liquids
Storage (non-Refinery); Bulk Plants; Valves, Flanges, and
Pumps
Petroleum and Solvent Evaporation; Petroleum Liquids
Storage (non-Refinery); Bulk Plants; Miscellaneous
Losses/Leaks: Vapor Control Unit Losses
100
-------
Table 44. Continued.
sec
40400254
40400262
40400278
40400402
40400404
40400406
40400498
40600101
40600131
40600141
40600147
40600163
Description
Petroleum and Solvent Evaporation; Petroleum Liquids Storage
(non-Refinery); Bulk Plants; Tank Truck Vapor Losses
Petroleum and Solvent Evaporation; Petroleum Liquids Storage
(non-Refinery); Bulk Plants; Gasoline RVP 10: Standing Loss -
Int. Floating Roof w/ Primary Seal
Petroleum and Solvent Evaporation; Petroleum Liquids Storage
(non-Refinery); Bulk Plants; Gasoline RVP 10/13/7:
Withdrawal Loss - Int. Float Roof (Pri/Sec Seal)
Petroleum and Solvent Evaporation; Petroleum Liquids Storage
(non-Refinery); Petroleum Products - Underground Tanks;
Gasoline RVP 13: Working Loss
Petroleum and Solvent Evaporation; Petroleum Liquids Storage
(non-Refinery); Petroleum Products - Underground Tanks;
Gasoline RVP 10: Working Loss
Petroleum and Solvent Evaporation; Petroleum Liquids Storage
(non-Refinery); Petroleum Products - Underground Tanks;
Gasoline RVP 7: Working Loss
Petroleum and Solvent Evaporation; Petroleum Liquids Storage
(non-Refinery); Petroleum Products - Underground Tanks;
Specify Liquid: Working Loss
Petroleum and Solvent Evaporation; Transportation and
Marketing of Petroleum Products; Tank Cars and Trucks;
Gasoline: Splash Loading **
Petroleum and Solvent Evaporation; Transportation and
Marketing of Petroleum Products; Tank Cars and Trucks;
Gasoline: Submerged Loading (Normal Service)
Petroleum and Solvent Evaporation; Transportation and
Marketing of Petroleum Products; Tank Cars and Trucks;
Gasoline: Submerged Loading (Balanced Service)
Petroleum and Solvent Evaporation; Transportation and
Marketing of Petroleum Products; Tank Cars and Trucks;
Gasoline: Submerged Loading (Clean Tanks)
Petroleum and Solvent Evaporation; Transportation and
Marketing of Petroleum Products; Tank Cars and Trucks;
Gasoline: Return with Vapor (Transit Losses)
sec
40400261
40400263
40400401
40400403
40400405
40400497
406001
40600126
40600136
40600144
40600162
406002
Description
Petroleum and Solvent Evaporation; Petroleum Liquids
Storage (non-Refinery); Bulk Plants; Gasoline RVP 13:
Standing Loss - Int. Floating Roof w/ Primary Seal
Petroleum and Solvent Evaporation; Petroleum Liquids
Storage (non-Refinery); Bulk Plants; Gasoline RVP 7:
Standing Loss - Internal Floating Roof w/ Primary Seal
Petroleum and Solvent Evaporation; Petroleum Liquids
Storage (non-Refinery); Petroleum Products - Underground
Tanks; Gasoline RVP 13: Breathing Loss
Petroleum and Solvent Evaporation; Petroleum Liquids
Storage (non-Refinery); Petroleum Products - Underground
Tanks; Gasoline RVP 10: Breathing Loss
Petroleum and Solvent Evaporation; Petroleum Liquids
Storage (non-Refinery); Petroleum Products - Underground
Tanks; Gasoline RVP 7: Breathing Loss
Petroleum and Solvent Evaporation; Petroleum Liquids
Storage (non-Refinery); Petroleum Products - Underground
Tanks; Specify Liquid: Breathing Loss
Petroleum and Solvent Evaporation; Transportation and
Marketing of Petroleum Products; Tank Cars and Trucks
Petroleum and Solvent Evaporation; Transportation and
Marketing of Petroleum Products; Tank Cars and Trucks;
Gasoline: Submerged Loading **
Petroleum and Solvent Evaporation; Transportation and
Marketing of Petroleum Products; Tank Cars and Trucks;
Gasoline: Splash Loading (Normal Service)
Petroleum and Solvent Evaporation; Transportation and
Marketing of Petroleum Products; Tank Cars and Trucks;
Gasoline: Splash Loading (Balanced Service)
Petroleum and Solvent Evaporation; Transportation and
Marketing of Petroleum Products; Tank Cars and Trucks;
Gasoline: Loaded with Fuel (Transit Losses)
Petroleum and Solvent Evaporation; Transportation and
Marketing of Petroleum Products; Marine Vessels
101
-------
Table 44. Continued.
sec
40600231
40600233
40600236
40600238
40600240
40600242
40600302
40600306
40600399
40600707
40688802
Description
Petroleum and Solvent Evaporation; Transportation and
Marketing of Petroleum Products; Marine Vessels; Gasoline:
Ship Loading - Cleaned and Vapor Free Tanks
Petroleum and Solvent Evaporation; Transportation and
Marketing of Petroleum Products; Marine Vessels; Gasoline:
Barge Loading - Cleaned and Vapor Free Tanks
Petroleum and Solvent Evaporation; Transportation and
Marketing of Petroleum Products; Marine Vessels; Gasoline:
Ship Loading - Uncleaned Tanks
Petroleum and Solvent Evaporation; Transportation and
Marketing of Petroleum Products; Marine Vessels; Gasoline:
Barges Loading - Uncleaned Tanks
Petroleum and Solvent Evaporation; Transportation and
Marketing of Petroleum Products; Marine Vessels; Gasoline:
Barge Loading - Average Tank Condition
Petroleum and Solvent Evaporation; Transportation and
Marketing of Petroleum Products; Marine Vessels; Gasoline:
Transit Loss
Petroleum and Solvent Evaporation; Transportation and
Marketing of Petroleum Products; Gasoline Retail Operations -
Stage I; Submerged Filling w/o Controls
Petroleum and Solvent Evaporation; Transportation and
Marketing of Petroleum Products; Gasoline Retail Operations -
Stage I; Balanced Submerged Filling
Petroleum and Solvent Evaporation; Transportation and
Marketing of Petroleum Products; Gasoline Retail Operations -
Stage I; Not Classified **
Petroleum and Solvent Evaporation; Transportation and
Marketing of Petroleum Products; Consumer (Corporate) Fleet
Refueling - Stage I; Underground Tank Breathing and
Emptying
Petroleum and Solvent Evaporation; Transportation and
Marketing of Petroleum Products; Fugitive Emissions; Specify
in Comments Field
sec
40600232
40600234
40600237
40600239
40600241
40600301
40600305
40600307
40600706
40688801
Description
Petroleum and Solvent Evaporation; Transportation and
Marketing of Petroleum Products; Marine Vessels; Gasoline:
Ocean Barges Loading
Petroleum and Solvent Evaporation; Transportation and
Marketing of Petroleum Products; Marine Vessels; Gasoline:
Ship Loading - Ballasted Tank
Petroleum and Solvent Evaporation; Transportation and
Marketing of Petroleum Products; Marine Vessels; Gasoline:
Ocean Barges Loading - Uncleaned Tanks
Petroleum and Solvent Evaporation; Transportation and
Marketing of Petroleum Products; Marine Vessels; Gasoline:
Tanker Ship - Ballasted Tank Condition
Petroleum and Solvent Evaporation; Transportation and
Marketing of Petroleum Products; Marine Vessels; Gasoline:
Tanker Ship - Ballasting
Petroleum and Solvent Evaporation; Transportation and
Marketing of Petroleum Products; Gasoline Retail Operations
- Stage I; Splash Filling
Petroleum and Solvent Evaporation; Transportation and
Marketing of Petroleum Products; Gasoline Retail Operations
- Stage I; Unloading **
Petroleum and Solvent Evaporation; Transportation and
Marketing of Petroleum Products; Gasoline Retail Operations
- Stage I; Underground Tank Breathing and Emptying
Petroleum and Solvent Evaporation; Transportation and
Marketing of Petroleum Products; Consumer (Corporate)
Fleet Refueling - Stage I; Balanced Submerged Filling
Petroleum and Solvent Evaporation; Transportation and
Marketing of Petroleum Products; Fugitive Emissions;
Specify in Comments Field
102
-------
PADD5
Figure 23. PADD regions for the U.S.
103
-------
Figure 24. RFG counties (dark gray) for the U.S.
Table 45. Change in Average Fuel Benzene Level (Volume Percent) by PADD with
Implementation of Proposed Fuel Benzene Standard (CG - Conventional Gasoline; RFG -
Reformulated Gasoline).
Reference
Case
Control Case
Multiplicative
Factor
Gasoline
Type
CG
RFG
CG
RFG
CG
RFG
Region
PADD1
0.91 %
0.59%
0.55%
0.54%
0.60
0.92
PADD 2
1.26%
0.80%
0.68%
0.71%
0.54
0.89
PADD 3
0.95%
0.57%
0.54%
0.55%
0.57
0.96
PADD 4
1.47%
1.05%
0.93%
0.62%
0.63
0.59
PADD 5
1.42%
0.65%
0.85%
0.60%
0.60
0.92
Calif.
0.62%
0.62%
0.61%
0.61%
0.98
0.98
Table 46 shows the reference (no controls) and controlled stationary benzene emissions after
applying the controls to 2015 and 2020 emissions by SCC. Detailed summaries can be found in
benzene_gas_scc.xls in the MSAT rule docket EPA-HQ-OAR-2005-0036
104
-------
Table 46. Benzene stationary emissions (tons) before and after applying controls to reference case gasoline
distribution emissions (non refueling gasoline) and vehicle refueling emissions. Also shown are the percent
reference). 1999 NEI emissions are shown for comparison.
marketing and
differences (control-
Emissions type
All storage types and
all products: total
breathing loss
Bulk
stations/terminals :
gasoline breathing
loss: gasoline
Stage 1 Filling
Underground Tanks:
Gasoline Service
Aviation gasoline
distribution
Transport
Bulk terminals
Bulk plants
Underground tanks
SCC codes
2501000000
2501050120
2501060050, 2501060051, 2501060052, 2501060053
2501060200, 2501060201
2501080050, 2501080000, 2501080100
2501995120, 2505000000, 2505000120, 2505010000,
2505020000, 2505020120, 2505020121, 2505030120
40400101, 40400102, 40400103, 40400104, 40400105,
40400106, 40400107, 40400108, 40400109, 40400110,
40400111, 40400112, 40400113, 40400114, 40400115,
40400116, 40400117, 40400118, 40400119, 40400120,
40400131, 40400132, 40400141, 40400142, 40400143,
40400148, 40400150, 40400151,
40400152, 40400153, 40400154, 40400161,
40400162, 40400163, 40400171, 40400172,
40400173, 40400178
40400201, 40400202, 40400203, 40400204, 40400205,
40400206, 40400207, 40400208, 40400209, 40400210,
40400212, 40400213,
40400231, 40400241, 40400242, 40400250, 40400251,
40400252, 40400253, 40400254, 40400261, 40400262,
40400263, 40400278,
40400401, 40400402, 40400403, 40400404, 40400405,
40400406, 40400497, 40400498
1999
8
1,535
1,785
86
307
915
69
40
3
2015
Reference
12
1,579
1,826
88
327
1,110
107
61
4
Control
7
952
1,176
62
226
908
71
41
3
%
Diff.
-40
-40
-36
-25
-31
-18
-33
-32
-23
2020
Reference
14
1,593
1,835
88
340
1,198
120
68
5
Cont
rol
8
960
1,182
62
236
983
80
46
4
%
Diff.
-10
-40
-36
-30
-31
-18
-33
-32
-23
105
-------
Table 46. Continued.
Emissions type
Tank cars and trucks
Marine vessels
Stage 1 Evaporation
Retail
Stage 1 Evaporation
Fleet
SCC codes
406001, 40600101, 40600126, 40600131, 40600136,
40600141, 40600144, 40600147, 40600162, 40600163
406002, 40600231, 40600232, 40600233, 40600234,
40600236, 40600237, 40600238, 40600239, 40600240,
40600241, 40600242
40600301, 40600302, 40600305, 40600306, 40600307,
40600399
40600706, 40600707, 40688801, 40688802
Total non refueling gasoline
Vehicle refueling
Other stationary sources
Total
1999
149
17
22
35
4,970
1,566
104,645
111,181
2015
Reference
192
24
35
54
5,419
724
114,186
120,329
Control
127
17
22
49
3,663
459
114,186
118,308
%
Diff.
-34
-30
-37
-10
-32
-37
0
-2
2020
Reference
217
27
40
61
5,606
720
117,470
123,796
Control
143
19
25
55
3,804
459
117,470
121,732
%
Diff.
-34
-30
-37
-9
-32
-36
0
-2
106
-------
10.2 Highway gasoline vehicle inventory
To develop the highway vehicle inventories, NMIM was rerun for the controlled case, using
revised gasoline fuel parameter inputs for fuel benzene and aromatics levels. The revised fuel
benzene inputs were described in Section 10.1 (see Table 45). The refinery modeling also
indicated that the reduction in fuel benzene levels would result in small decreases in aromatics
levels as well. Thus, aromatics levels were adjusted using the additive factors calculated as
follows:
Additive Factor = 0.7*(Benzenecontroi - Benzenereference) (23)
A pollutant, county and SCC specific projection factor was computed from the controlled and
reference case NMIM- based emissions as follows:
77
^ NMIMcontrol,20XX ,_,>
T - (24)
NMIMreference , 20 XX
This factor was then applied to the reference case inventories for 2015, 2020 and 2030, at the
county and SCC and pollutant level, for the 1,3-butadiene, benzene, acetaldehyde, acrolein, and
formaldehyde.
Details on the computation of the projection factor and the application to the reference case
inventories are provided in Appendix F. Sample calculations are also provided. Summaries are
shown in Table 47. The complete summaries can be found in onroad_controls.xls or
onroad_controls_pivot.xls in the MSAT rule docket, EPA-HQ-OAR-2005-0036.
107
-------
Table 47. National MSAT reference and controlled emissions (nearest ton) for gasoline
powered vehicles by HAP for 2015, 2020, and 2030.
Vehicle
HDGV
LDGT1
LDGT2
LDGV
MC
HAP
1,3 -Butadiene
Acetaldehyde
Acrolein
Benzene
Formaldehyde
1,3 -Butadiene
Acetaldehyde
Acrolein
Benzene
Formaldehyde
1,3 -Butadiene
Acetaldehyde
Acrolein
Benzene
Formaldehyde
1,3 -Butadiene
Acetaldehyde
Acrolein
Benzene
Formaldehyde
1,3 -Butadiene
Acetaldehyde
Acrolein
Benzene
Formaldehyde
2015 emissions
Base
130
297
25
2,152
741
2,307
2,714
306
23,835
5,572
1,524
1,789
198
15,694
3,628
1,895
2,123
251
19,835
4,628
266
233
22
892
693
Controlled
130
297
25
1,890
741
2,312
2,721
306
21,060
5,591
1,528
1,793
198
13,940
3,639
1,899
2,130
251
17,424
4,643
266
233
22
770
693
2020 emissions
Base
103
245
18
1,760
599
2,291
2,682
302
23,346
5,516
1,503
1,726
191
14,897
3,513
1,500
1,690
199
15,643
3,705
288
253
24
967
751
Controlled
103
245
18
1,557
599
2,297
2,690
302
20,700
5,534
1,506
1,730
191
13,329
3,524
1,503
1,695
199
13,794
3,717
288
253
24
835
751
2030 emissions
Base
84
209
12
1,539
498
2,447
2,899
326
24,856
5,975
1,486
1,710
188
14,505
3,509
1,614
1,831
215
16,895
4,028
350
309
29
1,177
912
Controlled
84
209
12
1,359
498
2,453
2,907
326
22,116
5,994
1,489
1,714
188
13,060
3,519
1,618
1,837
215
14,914
4,041
350
309
29
1,017
912
HDGV: Heavy Duty Gasoline Vehicles
LDGT1: Light Duty Gasoline Trucks 1
LDGT2: Light Duty Gasoline Trucks 2
LDGV: Light Duty Gasoline Vehicles
MC: Motorcycles
10.3 Nonroad gasoline inventory
The approach used to compute controlled inventories for 2015, 2020 and 2030 for all nonroad
gasoline source categories (excluding planes, trains and ships) was to use projection factors
based on NMEVI results for the controlled case and reference case, and apply them to the
reference inventories.
Exhaust and evaporative projection factors for each year, 2015, 2020, and 2030 were obtained
from the NMEVI light duty gasoline vehicle reference and control case inventories. We assumed
that changes in county level exhaust emissions of nonroad gasoline equipment were proportional
to changes in highway light duty gasoline vehicle exhaust emissions, and changes in county level
evaporative emissions of nonroad gasoline equipment were proportional to changes in highway
light duty gasoline vehicle evaporative (refueling and non-refueling) emissions:
108
-------
PF nonroad exhaust
ElDGVexhausi
NMIMControl 2'OXX
20XX
ELDOVExhaustNMIMReference20XX
(25)
PF nonroad evap2
ElDGVeva,
'P NMMControl20XX
20XX
J>/hvaPNMIMReference20XX
(26)
The steps taken to compute the projection factors, along with example calculations, are shown i
Appendix G.
in
Once the projection factors were developed, the reference case nonroad emissions were projected
using the factors. For benzene, the reference MSAT emissions were broken out by exhaust and
evaporative emissions in NMIM, with each type being multiplied by the appropriate projection
factor, exhaust or evaporative. For the other HAPs, the exhaust projection factor was applied to
the reference MSAT emissions, with no exhaust or evaporative breakout of the emissions
because those HAPs did not have an evaporative component.
Appendix G describes the steps taken to develop of the controlled nonroad gasoline emissions.
Example calculations are also provided.
Table 48 summarizes the 2-stroke and 4-stroke emissions for the reference and controlled case
inventories for 2015, 2020, and 2030. Complete nonroad summaries, including emissions not
affected by the controls for the five HAPs are in nonroad_controls.xls and
nonroad_pivot_controls.xls in the MSAT rule docket EPA-HQ-OAR-2005-0036.
Table 48. 2015, 2020, and 2030 reference and controlled emissions for the five HAPS for
nonroad gasoline categories.
Engine
2-stroke
4-stroke
HAP
1,3 -Butadiene
Acetaldehyde
Acrolein
Benzene
Formaldehyde
5 HAP total
1,3 -Butadiene
Acetaldehyde
Acrolein
Benzene
Formaldehyde
5 HAP total
2015 emissions
Reference
1,847
1,467
344
18,582
3,928
26,168
3,224
2,196
289
19,165
5,495
30,368
Controlled
1,852
1,471
344
16,295
3,942
23,904
3,231
2,201
289
16,852
5,510
28,183
2020 emissions
Reference
1,604
1,293
292
16,287
3,415
22,890
3,379
2,265
300
20,153
5,688
31,784
Controlled
1,608
1,297
292
14,198
3,427
20,821
3,386
2,271
300
17,820
5,704
29,480
2030 emissions
Reference
1,595
1,296
291
16,457
3,414
23,054
3,805
2,511
334
22,705
6,326
35,682
Controlled
1,599
1,300
291
14,311
3,426
20,927
3,814
2,517
334
20,089
6,344
33,098
10.4 EMS-HAP Processing
EMS-HAP processing for stationary sources followed that as described in Section 5.1 for point
sources and 5.2 for non-point sources. For onroad emissions, processing followed that as
109
-------
described in Section 5.3, while the nonroad processing followed that as described in Section
5.4.2 for airport support equipment and Section 5.4.3 for remaining nonroad emissions. Aircraft
emissions were unaffected, and EMS-HAP was not rerun. Note that the point, non-point,
onroad, airport support equipment, and remaining nonroad emissions (no aircraft) contained only
the five HAPS being emphasized. The aircraft emissions files input into ASPEN contained other
HAPs.
10.5 ASPEN Processing and Post-Processing
ASPEN processing followed that as described in Section 6.1. For the stationary sources, only
benzene was modeled for the controlled case, as that was the only HAP affected by the benzene
control scenario as described in Section 10.1. For the mobile sources, 1,3-butadiene, benzene,
acetaldehyde, acrolein, and formaldehyde were modeled.
ASPEN post-processing followed that as described in Section 6.3, using the same "runl" and
"run2" file organization as described in that section. For the creation of the HAPEM input files,
only the runl file was affected since it contains the stationary, onroad gasoline, and nonroad
gasoline concentrations. The run2 files contain the onroad diesel and non-gasoline nonroad
concentrations and zeros for major, area & other, and background. The non-gasoline mobile
concentrations were unaffected by the controls on the emissions; thus, the run2 files did not need
to be rerun through HAPEM.
Table 49 presents the national average 1999 and projected reference and controlled stationary
source concentrations for benzene.
Table 49. National average 1999 and future year reference and controlled benzene stationary
concentrations.
Year
1999
2015
2020
Concentration
Type
Base
Reference
Controlled
Reference
Controlled
Benzene Concentrations (ug m 3)
Major
2.24xlO"2
1.60xlO"2
1.58xlO"2
1.75xlO"2
1.73xlO"2
Area & other
1.63X10"1
l.SSxlO"1
l.SlxlO"1
1.95X10"1
l.SSxlO"1
Tables 50 and 51 present the national average reference and controlled concentrations for onroad
gasoline and nonroad gasoline, respectively.
110
-------
Table 50. National average reference and controlled onroad gasoline concentrations for the five
HAPs for 2015, 2020, and 2030.
HAP
1,3 -Butadiene
Acetaldehyde
Acrolein
Benzene
Formaldehyde
Onroad Gasoline Concentrations (jig m 3)
2015
Reference
1.33X10'2
2.38X10'1
1.44X10'2
2.21X10'1
1.12X10'1
Controlled
1.33X10'2
2.38X10'1
1.44X10'2
2.00X10'1
1.12X10'1
2020
Reference
1.22X10'2
2.05X10'1
l.SlxlO'2
1.98X10'1
l.OlxlO'1
Controlled
1.22X10'2
2.05X10'1
l.SlxlO'2
l.SOxlO'1
l.OlxlO'1
2030
Reference
l.SlxlO'2
2.13X10'1
1.41X10'2
2.09X10'1
l.OSxlO'1
Controlled
l.SlxlO'2
2.13X10'1
1.41X10'2
1.91X10'1
l.OSxlO'1
Table 51. National average reference and controlled nonroad gasoline concentrations for the
five HAPs for 2015, 2020, and 2030.
HAP
1,3 -Butadiene
Acetaldehyde
Acrolein
Benzene
Formaldehyde
Nonroad Gasoline Concentrations (jig m 3)
2015
Reference
7.44xlO'3
4.89xlO'2
5.37xlO'3
7.40X10'2
5.31xlO'2
Controlled
7.45xlO'3
4.89xlO'2
5.37xlO'3
6.76xlO'2
5.31X10'2
2020
Reference
7.87xlO'3
4.97xlO'2
5.53xlO'3
7.68xlO'2
5.55xlO'2
Controlled
7.88xlO'3
4.97xlO'2
5.53xlO'3
7.00X10'2
5.56xlO"2
2030
Reference
9.01X10'3
5.53xlO'2
6.18X10'3
8.67xlO'2
6.29xlO'2
Controlled
9.02xlO'3
5.54xlO'2
6.19xlO'3
7.91xlO'2
6.30xlO'2
More detailed summaries can be found in the MSAT rule docket EPA-HQ-OAR-2005-0036 in
the excel file named aspen_conc_control.xls. County median concentration maps are also in the
docket; the file name is: ASPEN_median_cntrl.ppt.
10.6 HAPEM Processing and Post-Processing
HAPEM runs were made for 2015, 2020, and 2030 for the five HAPs modeled in the control
case. Only the runl file was needed, since it contained the stationary and the mobile gasoline
ASPEN concentrations and background concentrations. Run2 files contained zeros for stationary
and background and the non-gasoline nonroad concentrations and onroad diesel concentrations.
After the HAPEM runs, summaries were calculated for the five HAPs. Due to the adjustment to
exposure concentrations in HAPEM (documented in Section 7) by the median total concentration
at each tract, the stationary source concentrations of 1,3-butadiene, acetaldehyde, acrolein, and
formaldehyde stationary concentrations were different from the reference case, even though the
stationary input concentrations into HAPEM were unchanged. This is because they were
contained in the runl file along with the onroad and nonroad gasoline concentrations, which did
change. Since the control case does not impact the stationary source contribution for these
HAPs, we replaced the HAPEM control case concentrations for these four HAPs with the
reference case concentrations. This was not done for benzene, since stationary concentrations
were expected to change resulting from the changes to the stationary benzene gasoline inventory
described in Section 10.1.
Ill
-------
Table 52 presents the national average 1999 and projected reference and controlled stationary
source concentrations for benzene.
Table 52. National average 1999 and future reference and controlled benzene HAPEM
stationary concentrations.
Year
1999
2015
2020
Concentration
Type
Base
Reference
Controlled
Reference
Controlled
Benzene Concentrations (ug m 3)
Major
1.88xlO"2
1.35xlO"2
1.34xlO"2
1.48xlO"2
1.47xlO"2
Area & other
1.42X10"1
1.64X10"1
l.SSxlO"1
l.VlxlO"1
1.65X10"1
Tables 53 and 54 present the national average reference and controlled concentrations for the
five modeled HAPs for onroad gasoline and nonroad gasoline, respectively.
More detailed summaries can be found in the MSAT rule docket EPA-HQ-OAR-2005-0036. in
the excel file named hapem_concentrations_cntrl.xls. County median concentration maps are
also in the docket; the file name is: HAPEM_median_cntrl.ppt.
Table 53. National average reference and controlled HAPEM onroad gasoline concentrations
for the five HAPs for 2015, 2020, and 2030.
HAP
1,3 -Butadiene
Acetaldehyde
Acrolein
Benzene
Formaldehyde
Onroad Gasoline Concentrations (jig m 3)
2015
Reference
1.68xlO"2
2.73X10"1
1.65xlO"2
2.66X10"1
1.40X10"1
Controlled
1.68xlO"2
2.73X10"1
1.66xlO"2
2.41X10"1
1.40X10"1
2020
Reference
1.54xlO"2
2.35X10"1
1.51xlO"2
2.38X10"1
1.26X10"1
Controlled
1.54xlO"2
2.35X10"1
1.51xlO"2
2.17X10"1
1.27X10"1
2030
Reference
1.65xlO"2
2.44X10"1
1.61xlO"2
2.51X10"1
1.35X10"1
Controlled
1.65xlO"2
2.44X10"1
1.62xlO"2
2.30X10"1
1.35X10"1
Table 54. National average reference and controlled HAPEM nonroad gasoline concentrations
for the five HAPs for 2015, 2020, and 2030.
HAP
1,3 -Butadiene
Acetaldehyde
Acrolein
Benzene
Formaldehyde
Nonroad Gasoline Concentrations (jig m 3)
2015
Reference
7.28xlO'3
4.26xlO'2
4.67xlO'3
6.86xlO'2
4.73xlO'2
Controlled
7.29xlO'3
4.26xlO'2
4.68xlO'3
6.28xlO'2
4.74xlO'2
2020
Reference
7.73xlO'3
4.35xlO'2
4.82xlO'3
7.15xlO'2
4.96xlO'2
Controlled
7.75xlO'3
4.36xlO'2
4.82xlO'3
6.54xlO'2
4.96xlO'2
2030
Reference
8.88xlO'3
4.86xlO'2
5.40xlO'3
S.llxlO'2
5.63xlO'2
Controlled
8.89xlO'3
4.86xlO'2
5.40X10'3
7.42xlO'2
5.63xlO'2
10.7 Cancer and Non-cancer Calculations
The cancer and non-cancer risk calculations followed the same general methodology as
discussed in Section 8 with some minor differences in that HAP specific calculations were made
112
-------
for only the five HAPs being modeled. When calculating total risk or organ specific non-cancer
estimates, MS AT results for the other MS AT HAPs were used. Following are the results of the
calculations.
10.7.1 Cancer
Cancer risk estimates for 1,3-butadiene, acetaldehyde, benzene, and formaldehyde were
calculated based on the controlled HAPEM results. Total risks (across all MSAT HAPs) and
risks by carcinogen classes were also recalculated using the newly calculated risks for the four
above HAPs and the other carcinogenic MSAT HAPs reference case MSAT risks.
Table 55 lists the stationary risks for benzene, MSAT HAPs in carcinogen class A (benzene's
carcinogen class), and total risk from MSAT HAPs. Table 56 lists the onroad gasoline risks for
1,3-butadiene, acetaldehyde, benzene, formaldehyde and MSAT HAPs in carcinogenic classes
A, B, B2 and total MSAT HAP risk. Table 57 lists the nonroad gasoline risks for the same
HAPs, carcinogen classes, and total risk.
More detailed summaries can be found in the MSAT rule docket EPA-HQ-OAR-2005-0036 in
the excel file named hapem_risks_control.xls. County median risk maps are also in the docket;
the file name is: risk_cntrll.ppt.
Table 55. National average risks from stationary sources for 1999 and future year reference and
controlled benzene, carcinogen class A, and total (all MSAT HAPs).
Year
1999
2015
2020
Risk Type
Base
Reference
Controlled
Reference
Controlled
Stationary Risks
Benzene
Major
1.47E-07
1.05E-07
1.04E-07
1.15E-07
1.14E-07
Area&
other
1.11E-06
1.28E-06
1.23E-06
1.33E-06
1.28E-06
Carcinogen Class A
Major
7.71E-07
8.85E-07
8.85E-07
9.98E-07
9.97E-07
Area&
other
2.70E-06
3.25E-06
3.20E-06
3.49E-06
3.44E-06
Total Risk (All HAPs)
Major
1.14E-06
1.20E-06
1.20E-06
1.34E-06
1.34E-06
Area&
other
5.19E-06
6.19E-06
6.15E-06
6.57E-06
6.53E-06
Table 56. Reference and controlled HAPEM onroad gasoline risks for 2015, 2020, and 2030 for
individual HAPs and carcinogen classes A, Bl, and B2 and total risk (all MSAT HAPs,
including HAPs not controlled).
HAP
1,3 -Butadiene
Acetaldehyde
Benzene
Formaldehyde
Class A MSAT
Class B IMS AT
Class B2 MSAT
Total MSAT Risk
Onroad Gasoline Risks
2015
Reference
5.04xlO'7
6.00xlO'7
2.07xlO'6
7.71X10'10
2.84xlO'6
7.71X10'10
6.38xlO'7
3.79xlO'6
Controlled
5.05xlO'7
6.00xlO'7
l.SSxlO'6
7.72X10'10
2.64xlO'6
7.72X10'10
6.38xlO'7
3.60xlO'6
2020
Reference
4.62xlO'7
S.lSxlO'7
1.86X10'6
6.96X10'10
2.61X10'6
6.96X10'10
5.57xlO'7
3.48xlO'6
Controlled
4.63xlO'7
5.18xlO'7
1.69X10'6
6.96X10'10
2.44xlO'6
6.96X10'10
5.58xlO'7
3.32X10'6
2030
Reference
4.94xlO'7
5.38xlO'7
1.96X10'6
7.42X10'10
2.81X10'6
7.42X10'10
5.84xlO'7
3.76X10'6
Controlled
4.94xlO'7
5.38xlO'7
1.79X10'6
7.43X10'10
2.64xlO'6
7.43X10'10
5.85xlO'7
3.60xlO'6
113
-------
Table 57. Reference and controlled HAPEM nonroad gasoline risks for
for individual HAPs and carcinogen classes A, Bl, and B2 and total risk
including HAPs not controlled)
2015, 2020, and 2030
(all MSAT HAPs,
HAP
1,3 -Butadiene
Acetaldehyde
Benzene
Formaldehyde
Class A MSAT
Class B IMS AT
Class B2 MSAT
Total MSAT Risk
Nonroad Gasoline Risks
2015
Reference
2.18xlO"7
9.38xlO"8
5.35xlO"7
2.60xlO"10
7.63xlO"7
2.60xlO"10
l.llxlO"7
9.23xlO"7
Controlled
2.19xlO"7
9.38xlO"8
4.90xlO"7
2.60xlO"10
7.18xlO"7
2.60xlO"10
l.llxlO"7
8.78xlO"7
2020
Reference
2.32xlO"7
9.58xlO"8
5.58xlO"7
2.73xlO"10
S.OOxlO"7
2.73xlO"10
1.14xlO"7
9.67xlO"7
Controlled
2.32xlO"7
9.58xlO"8
5.10xlO"7
2.73xlO"10
7.52xlO"7
2.73xlO"10
1.15xlO"7
9.19xlO"7
2030
Reference
2.66xlO"7
1.07xlO"7
6.32xlO"7
3.10xlO"10
9.10xlO"7
S.lOxlO"10
1.28xlO"7
LlOxlO"6
Controlled
2.67xlO"7
1.07xlO"7
5.78xlO"7
S.lOxlO"10
8.56xlO"7
S.lOxlO"10
1.28xlO"7
1.04xlO"6
10.7.2 Non-cancer
Non-cancer hazard quotient estimates for 1,3-butadiene, acetaldehyde, acrolein, benzene, and
formaldehyde were calculated based on the controlled HAPEM results. Hazard indices by organ
system (across all MSAT HAPs) were also recalculated using the newly calculated risks for the
five above HAPs and the other non-cancer MSAT HAPs reference case risks.
Table 58 lists the stationary HQ for benzene and stationary HI for the immune system. Table 59
lists the onroad gasoline HQ for 1,3-butadiene, acetaldehyde, benzene, formaldehyde and HI for
immune, reproductive and respiratory systems. Table 60 lists the nonroad gasoline HQ for 1,3-
butadiene, acetaldehyde, benzene, formaldehyde and HI for immune, reproductive and
respiratory systems.
More detailed summaries can be found in the MSAT rule docket EPA-HQ-OAR-2005-0036 in
the excel file named: hapem_hq_control.xls. County median HQ or HI maps are also in the
docket; the file name is: hq_cntrll.ppt.
Table 58. 1999 and future year reference and controlled stationary benzene hazard quotients and
immune system hazard indices for MSAT HAPs for 2015 and 2020.
Year
1999
2015
2020
Non-cancer
estimate type
Base
Reference
Controlled
Reference
Controlled
Stationary
Benzene
Major
6.27xlO'4
4.49xlO'4
4.46xlO'4
4.93xlO'4
4.89xlO'4
Area&
other
4.72xlO'3
5.47xlO'3
5.27xlO'3
5.70xlO'3
5.49xlO'3
Immune System
Major
5.42xlO'3
5.95xlO'3
5.95xlO'3
6.48xlO'3
6.47xlO'3
Area & other
1.39xlO'2
1.57xlO'2
1.55xlO'2
1.70xlO'2
1.68xlO'2
114
-------
Table 59. Reference and controlled HAPEM onroad gasoline HQ for controlled HAPs and HI
for immune, reproductive, and respiratory systems (including MSAT HAPs not controlled) for
2015, 2020, and 2030.
HAP
1,3 -Butadiene
Acetaldehyde
Acrolein
Benzene
Formaldehyde
Immune System
Reproductive
System
Respiratory System
Onroad Gasoline
2015
Reference
8.40X10'3
S.OSxlO'2
8.27X10'1
8.87X10'3
1.43X10'2
l.OlxlO'2
8.40X10'3
8.78X10'1
Controlled
8.42xlO'3
S.OSxlO'2
8.27X10'1
S.OSxlO'3
1.43X10'2
9.29xlO'3
8.42xlO'3
8.79X10'1
2020
Reference
7.70xlO'3
2.61xlO'2
7.54X10'1
7.94xlO'3
1.29X10'2
9.34xlO'3
7.70xlO'3
S.OOxlO'1
Controlled
7.71xlO'3
2.62xlO'2
7.54X10'1
7.24xlO'3
1.29X10'2
8.63xlO'3
7.71xlO'3
S.OOxlO'1
2030
Reference
8.23xlO'3
2.72xlO'2
8.07X10'1
8.37xlO'3
1.38xl(r2
l.OlxlO'2
8.23xlO'3
8.56X10'1
Controlled
8.24xlO'3
2.72xlO'2
S.OSxlO'1
7.65xlO'3
l.SSxlO'2
9.40xlO'3
8.24xlO'3
8.57X10'1
Table 60. Reference and controlled HAPEM nonroad gasoline HQ for controlled MSAT HAPs
and HI for immune, reproductive, and respiratory systems (from MSAT HAPs including those
HAPs not controlled) for 2015, 2020, and 2030.
HAP
1,3 -Butadiene
Acetaldehyde
Acrolein
Benzene
Formaldehyde
Immune System
Reproductive
System
Respiratory
System
Nonroad Gasoline
2015
Reference
3.64xlO'3
4.74xlO'3
2.34X10'1
2.29xlO'3
4.83xlO'3
2.34xlO'3
3.64xlO'3
2.44X10'1
Controlled
3.64xlO'3
4.74xlO'3
2.34X10'1
2.09X10'3
4.83xlO'3
2.15xlO'3
3.64xlO'3
2.44X10'1
2020
Reference
3.87xlO'3
4.84xlO'3
2.41X10'1
2.38xlO'3
5.06xlO'3
2.44xlO'3
3.87xlO'3
2.52X10'1
Controlled
3.87xlO'3
4.84xlO'3
2.41X10'1
2.18X10'3
5.06X10'3
2.24xlO'3
3.87xlO'3
2.52X10'1
2030
Reference
4.44xlO'3
5.40X10'3
2.70X10'1
2.70xlO'3
5.74xlO'3
2.77xlO'3
4.44xlO'3
2.82X10'1
Controlled
4.44xlO'3
5.40xlO'3
2.70X10'1
2.47xlO'3
5.74xlO'3
2.54xlO'3
4.44xlO'3
2.82X10'1
10.7.3 Population statistics
Population statistics were calculated for total risk across MSAT HAPs and respiratory HI across
MSAT HAPs as documented in Section 8.3. Table 61 lists the total risk populations for 1999,
and for the future years 2015, 2020, and 2030 for reference and control cases. Differences
between reference and control case are also shown. Table 62 lists the respiratory HI populations
for 1999, and for the future years 2015, 2020, and 2030 for reference and control cases as well as
the differences. Major and area & other statistics are not shown for the HI statistics because
benzene is the only stationary source HAP impacted by the controls and is not a respiratory
HAP. Full summaries can be found in pop_stats_risk_cntrl.xls and pop_stats_hi_resp_cntrl.xls
for cancer and non-cancer respectively in the MSAT rule docket EPA-HQ-OAR-2005-0036.
115
-------
Table 61. Population risk classes for stationary and mobile total risk for 2015, 2020, and 2030 for reference and controlled risks from
MSAT HAPs using projected populations for each year. Tne total ™tepnrv indnHes W.WnnnH contributions
Source
Category
Major
Area&
Other
Onroad
Nonroad
Total
Population Class
Risk > 10'4
10"5 < Risk <10"4
10"6 < Risk <10"5
Risk <10'6
Total Population
Risk > 10'4
10"5 < Risk <10"4
10"6 < Risk <10"5
Risk <10'6
Total Population
Risk > 10"4
10'5 < Risk <10'4
10'6 < Risk <10'5
Risk <10"6
Total Population
Risk > 10'4
10'5 < Risk <10'4
10"6 < Risk <10"5
Risk <10'6
Total Population
Risk > 10'4
10"5 < Risk <10"4
10'6 < Risk <10'5
Risk <10"6
Total Population
^ Populations ^
Year
1999
Base
168,437
3,537,717
52,027,786
223,849,492
279,583,432
433,665
28,874,198
210,220,920
40,054,649
279,583,432
208,150
112,848,474
145,060,999
21,465,809
279,583,432
22,272
2,630,188
180,439,149
96,491,823
279,583,432
2,035,482
211,743,744
64,760,978
10,243,228
279,583,432
2015
Reference
192,100
4,244,119
52,174,340
260,294,114
316,904,672
636,991
39,345,554
245,736,898
31,185,230
316,904,672
0
19,596,469
241,185,986
56,122,217
316,904,672
23,710
1,365,537
150,013,784
165,501,640
316,904,672
1,303,148
210,880,893
104,720,624
7
316,904,672
Controlled
192,100
4,244,119
52,019,015
260,449,439
316,904,672
636,991
39,086,924
245,293,550
31,887,208
316,904,672
0
17,860,243
238,194,479
60,849,950
316,904,672
23,710
1,335,534
143,285,777
172,259,651
316,904,672
1,253,210
207,704,745
107,946,710
7
316,904,672
2020
Reference
254,904
5,215,260
57,544,083
266,678,495
329,692,742
739,981
4,417,331
254,558,606
30,276,824
329,692,742
0
16,703,891
249,373,492
63,615,359
329,692,742
25,123
1,584,116
159,142,708
168,940,795
329,692,742
1,547,121
219,257,053
108,888,561
7
329,692,742
Controlled
254,904
5,215,260
57,423,740
266,798,839
329,692,742
739,981
43,749,847
254,305,389
30,987,525
329,692,742
0
15,240,789
245,938,812
68,513,141
329,692,742
25,123
1,548,720
151,982,761
176,136,139
329,692,742
1,489,062
216,237,537
111,966,135
7
329,692,742
2030
Reference
276,017
5,560,051
61,320,162
288,739,334
355,895,564
779012
46500156
276158456
32457940
355,895,564
0
21,839,016
269,464,226
64,592,322
355,895,564
27,986
2,215,401
18,553,8098
168,114,078
355,895,564
1,654,725
239,434,529
114,806,302
9
355,895,564
Controlled
276,017
556,873
61,217,659
288,845,016
355,895,564
779,012
46,107,408
275,864,333
33,144,812
355,895,564
0
20,411,989
265,725,873
69,757,702
355,895,564
27,986
2,129,598
177,885,571
175,852,409
355,895,564
1,620,506
235,991,736
118,283,314
9
355,895,564
116
-------
Table 62. Population HI classes for mobile and total respiratory HI for 2015, 2020, and 2030 for reference and controlled risks using
projected populations for each year. The total category includes background contributions.
Source
Category
Onroad
Nonroad
Total
Population Class
HI> 10
1 < HI < 10
0.1 10
1 < HI < 10
0.110
1 < HI < 10
0.1
-------
This page intentionally blank
118
-------
References
Abt Associates, Inc., 2005: BenMAP (Environmental Benefits Mapping and Analysis Program)
User's Manual, June 2005. Bethesda, MD.
http://www.epa.gov/ttn/ecas/models/BenMAPUserManualTextJune2005.pdf
Batelle, 2003: Estimate background concentrations for the National-Scale Air Toxics
Assessment. Technical Report. Contract No. 68-D-02-061. Work Assignment 1-03.
Bollman A. Technical Memorandum: Development of Growth Factors for Future Year
Modeling Inventories, Prepared by E. H. Pechan and Associates for Marc Houyoux, U.S. EPA,
Contract Number 68-D-00-265, Work Assignment 3-31, April 30, 2004.
http://www.epa.gov/air/interstateairquality/pdfs/Non-EGU_nonpoint_Growth_Development.pdf
Cook, R.; Glover, E.; Michaels, H.; Brzezinski, D. "Modeling of Mobile Source Air Toxics
Using EPA's National Mobile Inventory Model. Proceedings, 2004 Emission Inventory
Conference, Clearwater Beach, Fl.
http://www.epa.gov/ttn/chief/conference/eil3/poster/cook.pdf
Eastern Research Group. National Mobile Inventory Model (NMIM) Base and Future Year
County Database Documentation and Quality Assurance Procedures. Prepared by Eastern
Research Group, Inc., for U.S. Environmental Protection Agency, Office of Transportation and
Air Quality, Ann Arbor, MI, 2003.
E.H. Pechan and Associates. Economic Growth and Analysis System Reference Manual, EGAS
4.0. Prepared for U. S. EPA, Emission Factor and Inventory Group, Emissions Monitoring and
Analysis Division, Office of Air Quality Planning and Standards, EPA, January 6, 2001.
http://www.epa.gov/ttn/chi ef/emch/projection/egas40/ref_man_4.pdf
Energy Information Administration. Annual Energy Outlook 2005 with Projections to 2025.
2005, Report No. DOE/EIA-0383. http://www.eia.doe.gov/oiaf/aeo/index.html.
Fan W, Treyz F, Treyz G. An evolutionary new economic geography model. J Regional Sci
2000 40: 671- 696.
Federal Aviation Administration, 2004. Terminal Area Forecast System.
Glen, G., Y. Lakkadi, J.A. Tippett, and M. del Valle-Torres (prepared by Manteca
Environmental Technology, Inc.), 1997: Development of NERL/CHAD: The National Exposure
Research Laboratory Consolidated Human Activity Database. EPA Contract No. 68-D5-0049.
MathPro, 1998. Costs of Alternative Sulfur Content Standards for Gasoline in PADD IV. Final
Report. Prepared for the National Petrochemical and Refiners Association. December 30 1998.
119
-------
Mathpro, 1999a. Costs of meeting 40 ppm Sulfur Content Standard for Gasoline in PADDs 1-3,
Via MOBIL and CD Tech Desulfurization Processes. Final Report. Prepared for the American
Petroleum Institute. February 26 1999.
MathPro, 1999b. Analysis of California Phase 3 RFG Standards. Prepared for the California
Energy Commission. December 7 1999.
Michaels H, Brzezinski D, Cook R. EPA's National Mobile Inventory Model (NMIM), A
Consolidated Emissions Modeling System for MOBILE6 and NONROAD. U. S. Environmental
Protection Agency, Office of Transportation and Air Quality, Assessment and Standards
Division, Ann Arbor, MI, March 2005; Report No. EPA-420-R-05-003.
Mullen M, Neumann J. Technical Memorandum: Documentation of 2003 VMT Projection
Methodology, Prepared by E. H. Pechan and Associates and Industrial Economics, Inc. for James
DeMocker, Office of Air and Radiation, Office of Policy Analysis and Review, U. S.
EPA, Contract No. 68-W-02-048, WA B-41, March 2004.
Taylor, M. Memorandum: Revised HAP Emission Factors for Stationary Combustion Turbines,
Prepared by Alpha-Gamma Technologies, Inc for Sims Roy, EPA OAQPS ESD Combustion
Group. August, 2003.Docket ID: OAR-2002-0060-0649.
Access via http://www.regulations.gov
U.S. Environmental Protection Agency, 2000. User's Guide for the Assessment System for
Population Exposure Nationwide (ASPEN, Version 1.1) Model. Office of Air Quality Planning
and Standards, Research Triangle Park, NC, Report No. EPA-454/R-00-017.
http://www.epa.gov/scramOO l/dispersion_alt.htm#aspen
U. S. Environmental Protection Agency, 2002. User's Guide for the Emissions Modeling System
for Hazardous Air Pollutants (EMS-HAP, Version 2.0), Office of Air Quality Planning and
Standards, Research Triangle Park, NC, Report No. EPA-454/B-02-001.
http://www.epa.gov/scram001/userg/other/emshapv2ug.pdf
U. S. Environmental Protection Agency, 2004. 1999 Final National Emissions Inventory Data
and Documentation, http://www.epa.gov/ttn/chief/net/1999inventory.html
U. S. Environmental Protection Agency, 2004b. User's Guide for the Emissions Modeling
System for Hazardous Air Pollutants (EMS-HAP, Version 3.0), Office of Air Quality Planning
and Standards, Research Triangle Park, NC, Report No. EPA-454/B-03-006.
http://www.epa.gov/scramOO l/dispersion_related.htm#ems-hap
U. S. Environmental Protection Agency, 2004c. IPM Analysis for the Clean Air Interstate Rule
(CAIR) - Formerly known as the Interstate Air Quality Rule (IAQR).
http://www.epa.gov/airmarkets/epa-ipm/iaqr.html
120
-------
U. S. Environmental Protection Agency, 2004d. Regulatory Impact Analysis: Clean Air
Nonroad Diesel Rule. Office of Transportation and Air Quality, Ann Arbor, MI, Report No.
EPA420-R-04-007. http://www.epa.gov/nonroad-diesel/2004fr.htmfeia
U. S. Environmental Protection Agency, 2005a. 2002 National Emissions Inventory Data and
Documentation, http://www.epa.gov/ttn/chief/net/2002inventory.html
U. S. Environmental Protection Agency, 2005b. Clean Air Interstate Rule Emissions Inventory
Technical Support Document, March 4 2005. http://www.epa.gov/cair/pdfs/finaltech01.pdf
U. S. Environmental Protection Agency, 2005c. EGAS Version 5.0 Beta, 2005.
http://www.epa.gov/ttn/ecas/egas5.htm.
U. S. Environmental Protection Agency, 2005d. Risk - Air Toxics Risk Assessment. 2005.
http://www.epa.gov/ttn/fera/risk_atoxic.html.
U. S. Environmental Protection Agency, 2005e: The HAPEM5 User's Guide - Hazardous Air
Pollutant Exposure Model, Version 5; March 2005, Available at:
http://www.epa.gov/ttn/fera/hapem5/hapem5_guide.pdf; RTF, NC
121
-------
This page intentionally blank
122
-------
Appendix A: Documentation of NMIM Runs Used to Develop Inventories for MSAT Rule
Air Quality Modeling
Harvey Michaels, Rich Cook and David Brzezinski,
Office of Transportation and Air Quality
A-l
-------
Mobile source hazardous air pollutants (HAP) inventories at a county-level resolution were
produced for the Mobile Sources Air Toxics (MSAT) rule analysis by running the National
Mobile Inventory Model (NMIM) for all 50 States and the District of Columbia. Simulations
were made for calendar years 1999, 2007, 2010, 2015, 2020 and 2030. The same temperature
and humidity inputs were used for all calendar years; the determination of these values is
explained in more detail below. Although monthly inputs were used in the model, results were
provided (summed up) to an annual temporal resolution for all calendar years. Resulting
inventories were used to develop future year to 1999 inventory ratios, which were then applied to
the final 1999 National Emissions Inventory (NEI). Thus, the emission trends for projection
years were consistent with the 1999 NEI, the inventory used in the 1999 NATA assessment. The
onroad sources were run separately, with and without refueling emissions for all inventory years.
This allowed the refueling runs to be used to develop future year to 1999 inventory ratios that
could be applied for refueling emissions which are contained in the 1999 NEI as non-point
sources.
Modeling air toxics requires specific fuel parameter inputs, as discussed in NMEVI
documentation. The sources of the fuel parameter inputs used in this modeling, and the methods
used to develop the fuel parameter database are described in the technical document, "Draft
National Mobile Inventory Model (NMIM) Base and Future Year County Database
Documentation and Quality Assurance Procedures," prepared by Eastern Research Group for U.
S. EPA, Office of Transportation and Air Quality, 15 April, 2003. However, the gasoline
parameters as delivered from the contractor do not address the phase out of gasoline containing
MTBE in California, Arizona, New York and Connecticut properly. The difference is important
when calculating HAPs. Thus, a new set of gasoline parameters was derived for areas in these
States using reformulated gasoline with MTBE, for summer and winter in 2004, 2005 and 2006.
An interpolated set of parameters was derived from the summer/winter values for use in
spring/fall months. The 2006 gasoline parameters were used for all 2007 and later calendar years
as well. More information on these revisions can be found in the change log for the NMIM
database. Both the technical document and change log described above can be found in the
docket for the rulemaking.
Three versions of the NMIM code and two versions of the NMIM County database were used to
generate onroad and nonroad inventories. The three code versions represent fixes to minor
problems in the computer code and added features that do not affect the output results. The three
NMIM code versions are equivalent in terms of the emission results they produce.
Two versions of the NMIM County database were used. The May 14, 2004, database is identical
to the April 12, 2004, version of the database, but includes the 1999 National Emission Inventory
(NEI) list of counties with Stage 2 refueling programs for estimating the effects of local control
programs on refueling emissions. Stage 2 refueling programs only affect onroad inventories for
refueling emissions. The Stage 2 used counties are listed in Appendix E-2 County Level
Allocation Values Used for Allocation Schemes 18, 22 and 27 (Stage 2 Control) in the report,
"Documentation for the Final 1999 Nonpoint Area Source National Emission Inventory for
A-2
-------
Hazardous Air Pollutants (Version 3)," August 26, 2003. This document is available on the EPA
web site at: http://www.epa.gov/ttn/chief/net/1999inventory.html
Multiple runs were necessary for the refueling runs due to computer problems during the runs
that resulted in incomplete results. The additional runs (labeled "a" and "b" in the table below)
were needed to fill in for the missing data in the original runs. The entire nationwide run was not
redone to save running time.
All onroad results are based on runs of the MOBILE6.2.03 version of MOBILE6. All nonroad
results are based on runs of the NR2003a version of the NONROAD model.
Output databases have been named to be the same as the run identification, except for
recreational marine, which goes in the same database as the other nonroad output. All run results
aggregate by emission type and power class.
Every run has associated with it a RunSpec and a batch file with the run name and extensions
"nrs" (for NMIM RunSpec) and "bat," respectively. RunSpecs and batch files have been
archived along with copies of the output results from the simulations.
Table A-l describes the specific version of the NMIM code and the version of the input
parameter database used for the analysis. The codes following the table are useful in
understanding the names used in the table.
A-3
-------
Table A-l. Run summary for MSAT mobile source inventories.
Run ID
MSATOH1999c24dl3
MSATOH2007c24dl3
OH2010c22dll
MSATOH2015c24dl3
MSATOH2020c24dl3
MSATOH2030c24dl3
NH1999c22dll
NH2007c22dll
NH2010c22dll
NH2015c22dll
NH2020c22dll
NH2030c22dll
RH1999c22dll
RH2007c22dll
RH2010c22dll
RH2015c22dll
RH2020c22dll
RH2030c22dll
MSATOR1999c24rdl3
MSATOR1999c24rdl3a
MSATOR2001c24rdl3
MSATOR2001c24rdl3a
MSATOR2007c24rdl3
MSATOR2010c24rdl3
MSATOR2015c24rdl3
MSATOR2015c24rdl3a
MS ATOR20 1 5c24rd 1 3b
MSATOR2020c24rdl3
MSATOR2030c24rdl3
Output Database
MSATOH1999c24dl3
MSATOH2007c24dl3
OH2010c22dll
MSATOH2015c24dl3
MSATOH2020c24dl3
MSATOH2030c24dl3
NH1999c22dll
NH2007c22dll
NH2010c22dll
NH2015c22dll
NH2020c22dll
NH2030c22dll
NH1999c22dll
NH2007c22dll
NH2010c22dll
NH2015c22dll
NH2020c22dll
NH2030c22dll
MSATOR1999c24rdl3
MSATOR1999c24rdl3
MSATOR2001c24rdl3
MSATOR2001c24rdl3
MSATOR2007c24rdl3
MSATOR2010c24rdl3
MSATOR2015c24rdl3
MSATOR2015c24rdl3
MSATOR2015c24rdl3
MSATOR2020c24rdl3
MSATOR2030c24rdl3
Description
OnroadHAPS 1999
Onroad HAPS 2007
OnroadHAPS 20 10
Onroad HAPS 20 15
Onroad HAPS 2020
Onroad HAPS 2030
NonroadHAPS 1999
Nonroad HAPS 2007
NonroadHAPS 20 10
Nonroad HAPS 20 15
Nonroad HAPS 2020
Nonroad HAPS 2030
Recreational Marine HAPS 1999
Recreational Marine HAPS 2007
Recreational Marine HAPS 2010
Recreational Marine HAPS 2015
Recreational Marine HAPS 2020
Recreational Marine HAPS 2030
Onroad Refueling 1999
Onroad Refueling 1999
Onroad Refueling 2001
Onroad Refueling 2001
Onroad Refueling 2007
Onroad Refueling 20 10
Onroad Refueling 20 15
Onroad Refueling 2015
Onroad Refueling 2015
Onroad Refueling 2020
Onroad Refueling 2030
NMIM Code Version:
c22 =NMIMSource20040415
c24= NMIMSource20040512
c24r = NMIMSource20040512 altered for refueling emissions only output.
NMIM County Database Version:
dll =County20040412
d!3=County20040514
Run codes:
NH = Nonroad except diesel recreational marine HAPS
RH = Diesel recreational marine HAPS
OH = Onroad HAPS
OR = Onroad Refueling
A-4
-------
Methodology Used to Compute By-County, By-Month, By-Hour Temperature and Relative
Humidity Tables
Both onroad and nonroad emission inventories are affected by changes in temperature. In
addition, onroad estimates for NOx from gasoline fueled vehicles are affected by humidity. A
detailed analysis of climate data was done to produce an estimate for the average hourly
temperatures and humidity (over an approximately 20 year period) to use for each county in the
nation. The results of this analysis are found in the NMEVI County database. Below is a brief
discussion of how the county specific average hourly temperatures were determined for each
month.
1) Hourly temperature and dew point data, as well as location (latitude and longitude), for all 1st
Order weather stations across the United States were obtained from the National Climatic
Data Center (NCDC). (Note: Automated weather stations began being installed in 1996.
Data from these 2nd order stations were used for the more recent, shorter analysis periods.)
2) For each station, an inventory was made as to the number of hours with joint temperature and
dew point data. In order to be included in the 1981-2000 analysis, each station had to have at
least 50% data recovery for each hour of each month, and at least 75% data recovery over the
entire 20 years. (This cutoff was raised to 75% for the 5-year analysis, and to 90% for one-
year analyses). (Note: Climatological averages are usually based on a 30-year period. The
20-year period of 1981-2000 was selected due to hourly data availability constraints. Prior to
1981, limited computer technology forced the NCDC to only store observations for every 3rd
hour. Attempts to interpolate the 3-hour data showed biases and errors (for example dew
point exceeding temperature.)
3) For each station passing the data availability filter, the average temperature and dew point for
each hour of each month over the 20-year period was computed. (Note: Relative humidity
data should never be averaged. Since it depends on the associated temperature and dew point,
relative humidity is not a conservative property of the atmosphere.)
4) Population centroids (latitude and longitude) for each county were obtained from the 2000
United States Census. Population, rather than geographic, centroids were used to provide the
best estimate where the county's VMT would occur. (Note: This selection proved
particularly important for those counties near mountainous or desert areas.)
5) From each county's centroid, the distance and direction to each weather station was
calculated. The shortest distance was computed using the standard great circle navigation
method and the constant course direction was computed using the standard rhumb line
method.
6) Based on the computed directions, the stations were assigned to an octant, as follows: Octant
1: 0°
-------
7) For each octant, the stations were sorted by distance. The station closest to the centroid for
each octant was chosen for further processing. If the closest station was more than 200 miles
away, that octant is ignored. (Such situations occurred near the oceans and the along the
Canadian and Mexican borders.)
8) To remove the effects of differing time zones between counties and stations, the temperature
and dew point data from each octant station were synchronized to the same local hour. Thus,
noon EST is matched up with noon CST, with noon MST with noon PST, etc.
9) The octant (8 or less) temperature and dew point values were merged together using inverse
distance-squared weighting.
10) The corresponding relative humidity was then computed from the weighted temperature and
dew point values. (Note: In keeping with standard meteorological practices, the relative
humidity was always computed with respect to water, even if the temperature was below
freezing.)
11) The above process was repeated for each hour, for each month, and for each county centroid.
As a final check, the results from different times and months were plotted on maps and
contoured.
A-6
-------
Appendix B: Steps and Example calculations of onroad projections
B.I Onroad HAPs (Section 3.3.2)
®
The following steps summarizes the SAS program, onroad. sas (found in the MS AT rule docket
EPA-HQ-OAR-2005-0036) used to project the onroad inventory with sample calculations for
2015 for Autauga County, AL and Modoc County, CA. A detailed flow chart is shown in Figure
B-l.
Read the 1999 NEI onroad inventory and retained only the MSAT HAPs listed in Table
1, Section 1. It was found that all HAPs in the onroad NEI were MSAT HAPs. Also,
NMEVI emissions were initially provided by FIPS/SCC/CAS where SCC categories were
broken out by evaporative and exhaust components. For each FIPS/SCC/CAS, the
exhaust and evaporative components were summed together to give one emission
number.
Table B-l. Partial listing of emissions after merger of 1999 NEI emissions after merger
with MSAT CAS numbers. CAS 71432 is benzene, 1330207 is xylenes, 7440473 is
chromium, and CAS 226 and 7440020 are nickel.
FIPS
01001
01001
01001
01001
01001
01001
06049
06049
06049
06049
SCC
2201001130
2201001130
2201001130
2201080130
2230070130
2230070130
2201001130
2201001130
2201001130
2201080130
CAS
226
71432
7440473
71432
226
71432
71432
7440020
7440473
71432
emis
0.00007
1.15235
0.0001
0.009925
5E-6
0.02079
1.075615
0.000225
0.00032
0.015485
2. Read in NMIM emissions and summed heavy-duty diesel vehicle emissions to create a
total HDDV emission number for each FIPS/CAS/ road type where road type is
represented by the last 3 characters of SCC code and calculate new projection factors
based on the summed emissions for 1999 NMIM and future year NMIM results. These
factors were applied to SCC codes beginning with 2230070 for each FIPS/CAS in the
1999 NEI in step 13. These SCC codes are shown in Table 16 (Section 3.3.2). Output
dataset was named hddv_nmim. The temporary SCC code in the NMIM output was
223007XYYY where YYY is road type.
B-l
-------
Table B-2. Partial listing of Autauga County emissions after creating total HDDV SCC
from NMIM results. Emis99 are the 1999 NMIM emissions, emislS are 2015 emissions,
and ratio 15 is the ratio of 2015 emissions to the 1999 emissions.
FIPS
01001
01001
01001
01001
01001
01001
01001
01001
01001
01001
SCC
223007X130
223007X130
223007X130
223007X130
223007X130
223007X130
223007X130
223007X130
223007X130
223007X130
CAS
71432
7440020
71432
7440020
71432
7440020
71432
7440020
71432
7440020
emis99
0.0009132375
7.8660057E-7
0.0006728924
4.798758E-7
0.0029419617
1.1361038E-6
0.0155974124
4.0112847E-6
0.0006618121
1.7606117E-7
emislS
0.000441322
9.3792589E-7
0.0004738843
7.5895458E-7
0.0017220468
1.7165236E-6
0.0072218833
5.4976188E-6
0.0004315654
3.0090454E-7
ratio 15
0.4832499782
1.1923788539
0.7042497321
1.5815645908
0.5853396615
1.510886278
0.4630180423
1.3705381898
0.6520966767
1.7090908974
Table B-3. Partial listing of Autuaga County HDDV emissions after summing by
FIPS/SCC/CAS with recalculated ratio.
FIPS
01001
01001
SCC
223007X130
223007X130
CAS
71432
7440020
emis99
0.020787316
6.589926E-6
emislS
0.0102907019
9.2119274E-6
ratio 15
0.4950471688
1.3978802578
3. Recombined step 2 output with the original NMIM dataset.
Table B-4. Partial listing of emissions after concatenating HDDV emissions with
original NMIM emissions.
FIPS
01001
01001
01001
01001
01001
01001
01001
01001
01001
01001
06049
06049
06049
06049
SCC
223007X130
223007X130
2201001130
2201001130
2201001130
2201001130
2201080130
2201080130
2201080130
2201080130
2201001130
2201001130
2201001130
2201001130
CAS
71432
7440020
16065831
18540299
71432
7440020
16065831
18540299
71432
7440020
16065831
18540299
71432
7440020
emis99
0.020787316
6.589926E-6
0.0000593097
0.0000395398
1.1247759168
0.0000718905
5.9322425E-7
3.9548283E-7
0.00979517
7.1905967E-7
0.0000394923
0.0000263282
0.506942138
0.0000478695
emislS
0.0102907019
9.2119274E-6
0.0000494437
0.0000329625
0.2370964659
0.0000599318
6.6164804E-7
4.410987E-7
0.0104629781
8.0199761E-7
0.0000369164
0.0000246109
0.136686063
0.0000447472
ratio
0.4950471688
1.3978802578
0.8336535061
0.8336535004
0.2107944012
0.8336535267
1.1153422083
1.1153422209
1.0681772804
1.1153422226
0.9347752652
0.9347752577
0.2696285292
0.9347752469
B-2
-------
4. The NMIM Chromium III and Chromium VI emissions were summed for each FIPS/SCC
to give a total Chromium number. New projection factors were calculated for the
summed chromium and the CAS 7440473 was assigned to each record. This was done
for all FIPS/SCC combinations with Chromium III or Chromium VI.
Table B-5. Partial listing of NMIM chromium emissions (Section 3.4.2, step 4).
FIPS
01001
01001
01001
01001
06049
06049
sec
2201001130
2201001130
2201080130
2201080130
2201001130
2201001130
CAS
16065831
18540299
16065831
18540299
16065831
18540299
emis99
0.0000593097
0.0000395398
5.9322425E-7
3.9548283E-7
0.0000394923
0.0000263282
emislS
0.0000494437
0.0000329625
6.6164804E-7
4.410987E-7
0.0000369164
0.0000246109
ratio
0.8336535061
0.8336535004
1.1153422083
1.1153422209
0.9347752652
0.9347752577
Table B-6. Partial listing of NMIM chromium after summing by FIPS/SCC, assigning a
CAS, and calculating a ratio.
FIPS
01001
01001
06049
sec
2201001130
2201080130
2201001130
CAS
7440473
7440473
7440473
emis99
0.0000988494
9.8870708E-7
0.0000658205
emislS
0.0000824062
1.1027467E-6
0.0000615274
ratio
0.8336535038
1.1153422134
0.9347752622
5. Combined the chromium data with the original NMIM data.
Table B-7. Partial listing of NMIM emissions after concatenating with chromium
emissions.
FIPS
01001
01001
01001
01001
01001
01001
01001
01001
01001
01001
06049
06049
06049
06049
01001
01001
06049
sec
223007X130
223007X130
2201001130
2201001130
2201001130
2201001130
2201080130
2201080130
2201080130
2201080130
2201001130
2201001130
2201001130
2201001130
2201001130
2201080130
2201001130
CAS
71432
7440020
16065831
18540299
71432
7440020
16065831
18540299
71432
7440020
16065831
18540299
71432
7440020
7440473
7440473
7440473
emis99
0.020787316
6.589926E-6
0.0000593097
0.0000395398
1.1247759168
0.0000718905
5.9322425E-7
3.9548283E-7
0.00979517
7.1905967E-7
0.0000394923
0.0000263282
0.506942138
0.0000478695
0.0000988494
9.8870708E-7
0.0000658205
emislS
0.0102907019
9.2119274E-6
0.0000494437
0.0000329625
0.2370964659
0.0000599318
6.6164804E-7
4.410987E-7
0.0104629781
8.0199761E-7
0.0000369164
0.0000246109
0.136686063
0.0000447472
0.0000824062
1.1027467E-6
0.0000615274
ratio
0.4950471688
1.3978802578
0.8336535061
0.8336535004
0.2107944012
0.8336535267
1.1153422083
1.1153422209
1.0681772804
1.1153422226
0.9347752652
0.9347752577
0.2696285292
0.9347752469
0.8336535038
1.1153422134
0.9347752622
-------
7.
Extracted the NMIM xylenes, manganese, and nickel observations from the NMIM
results in preparation for work described in step 7.
Table B-8. Partial list of emissions for nickel.
FIPS
01001
01001
01001
06049
sec
223007X130
2201001130
2201080130
2201001130
CAS
7440020
7440020
7440020
7440020
emis99
6.589926E-6
0.0000718905
7.1905967E-7
0.0000478695
emislS
9.2119274E-6
0.0000599318
8.0199761E-7
0.0000447472
ratio
1.3978802578
0.8336535267
1.1153422226
0.9347752469
Copied the xylenes, manganese, and nickel observations to new observations with new
CAS numbers.
Table B-9. Partial list of nickel emissions after copying observations to duplicate records
and assigning CAS number to 226.
FIPS
01001
01001
01001
06049
sec
223007X130
2201001130
2201080130
2201001130
CAS
226
226
226
226
emis99
6.589926E-6
0.0000718905
7.1905967E-7
0.0000478695
emislS
9.2119274E-6
0.0000599318
8.0199761E-7
0.0000447472
ratio
1.3978802578
0.8336535267
1.1153422226
0.9347752469
8. Appended output from step 7 to output from step 5.
Table B-10. Partial list of emissions after concatening duplicate nickel records with
original data and sorted by FIPS/SCC/CAS.
FIPS
01001
01001
01001
01001
01001
01001
01001
01001
01001
01001
01001
01001
01001
01001
01001
06049
06049
06049
06049
06049
06049
sec
2201001130
2201001130
2201001130
2201001130
2201001130
2201001130
2201080130
2201080130
2201080130
2201080130
2201080130
2201080130
223007X130
223007X130
223007X130
2201001130
2201001130
2201001130
2201001130
2201001130
2201001130
CAS
16065831
18540299
226
71432
7440020
7440473
16065831
18540299
226
71432
7440020
7440473
226
71432
7440020
16065831
18540299
226
71432
7440020
7440473
emis99
0.0000593097
0.0000395398
0.0000718905
1.1247759168
0.0000718905
0.0000988494
5.9322425E-7
3.9548283E-7
7.1905967E-7
0.00979517
7.1905967E-7
9.8870708E-7
6.589926E-6
0.020787316
6.589926E-6
0.0000394923
0.0000263282
0.0000478695
0.506942138
0.0000478695
0.0000658205
emislS
0.0000494437
0.0000329625
0.0000599318
0.2370964659
0.0000599318
0.0000824062
6.6164804E-7
4.410987E-7
8.0199761E-7
0.0104629781
8.0199761E-7
1.1027467E-6
9.2119274E-6
0.0102907019
9.2119274E-6
0.0000369164
0.0000246109
0.0000447472
0.136686063
0.0000447472
0.0000615274
ratio
0.8336535061
0.8336535004
0.8336535267
0.2107944012
0.8336535267
0.8336535038
.1153422083
.1153422209
.1153422226
.0681772804
.1153422226
.1153422134
.3978802578
0.4950471688
1.3978802578
0.9347752652
0.9347752577
0.9347752469
0.2696285292
0.9347752469
0.9347752622
B-4
-------
After making changes to NMIM for total chromium, xylenes, manganese, and nickel, the
NEI and NMIM data were merged by FIPS/SCC/CAS keeping all records from the NEI
inventory. Keep all NEI observations and output data to mergedl.
Table B-ll. Merged NEI and NMEVI emissions by FIPS/SCC/CAS. Emis_nei is the
1999 NEI emissions variable.
FIPS
01001
01001
01001
01001
01001
01001
06049
06049
06049
06049
sec
2201001130
2201001130
2201001130
2201080130
2230070130
2230070130
2201001130
2201001130
2201001130
2201080130
CAS
226
71432
7440473
71432
226
71432
71432
7440020
7440473
71432
emis
0.00007
1.15235
0.0001
0.009925
5E-6
0.02079
1.075615
0.000225
0.00032
0.015485
emis99
0.0000718905
1.1247759168
0.0000988494
0.00979517
0.506942138
0.0000478695
0.0000658205
emislS
0.0000599318
0.2370964659
0.0000824062
0.0104629781
0.136686063
0.0000447472
0.0000615274
ratio
0.8336535267
0.2107944012
0.8336535038
1.0681772804
0.2696285292
0.9347752469
0.9347752622
10. Calculated the projection factors for motorcycles by first summing across all SCC codes
for each FIPS/CAS for the future year and 1999 and dividing the future year summed
emissions by the 1999 summed emissions.
Table B-12. County emissions for benzene for Modoc County. Emissions total includes
SCC emissions not shown in tables.
FIPS
06049
CAS
71432
emis99
4.0984894367
emislS
2.560901195
ratio 1
0.6248402575
11. Merged output from step 10 to output from step 9 for the observations without a
projection factor. Did not do this for the HDDV emissions (1999 NEI SCC beginning
with 223 0070).
Table B-13. Merged NEI and NMEVI emissions with Modoc County benzene county
ratio reset to county ratio (ratio 1).
FIPS
01001
01001
01001
01001
01001
01001
06049
06049
06049
06049
SCC
2201001130
2201001130
2201001130
2201080130
2230070130
2230070130
2201001130
2201001130
2201001130
2201080130
CAS
226
71432
7440473
71432
226
71432
71432
7440020
7440473
71432
emis
0.00007
1.15235
0.0001
0.009925
5E-6
0.02079
1.075615
0.000225
0.00032
0.015485
emis99
0.0000718905
1.1247759168
0.0000988494
0.00979517
0.506942138
0.0000478695
0.0000658205
emislS
0.0000599318
0.2370964659
0.0000824062
0.0104629781
0.136686063
0.0000447472
0.0000615274
ratio
0.8336535267
0.2107944012
0.8336535038
1.0681772804
0.2696285292
0.9347752469
0.9347752622
0.6248402575
ratio 1
0.6248402575
B-5
-------
12. Subset HDDV emissions, 223007XYYY from the NMIM data.
Table B-14. Subsetted HDDV emissions. Ratio has been renamed ratio_07 and SCC to
SCC1.
FIPS
01001
01001
01001
SCC1
223007X130
223007X130
223007X130
CAS
226
71432
7440020
emis99
6.589926E-6
0.020787316
6.589926E-6
emislS
9.2119274E-6
0.0102907019
9.2119274E-6
ratio 07
1.3978802578
0.4950471688
1.3978802578
13. Merged the subsetted HDDV emissions by FIPS, CAS, and where the NEI SCC began
with 2230070 and the NMIM derived SCC began with 223007X.
Table B-15. Merged NEI and NMIM emissions merged with HDDV data by FIPS/CAS
where SCC=SCC1. SCC1, ratiol, emis99 and emislS not shown. For HDDV SCC
emissions, ratio has been set equal to ratio 07
FIPS
01001
01001
01001
01001
01001
01001
06049
06049
06049
06049
SCC
2201001130
2201001130
2201001130
2201080130
2230070130
2230070130
2201001130
2201001130
2201001130
2201080130
CAS
226
71432
7440473
71432
226
71432
71432
7440020
7440473
71432
emis nei
0.00007
1.15235
0.0001
0.009925
5E-6
0.02079
1.075615
0.000225
0.00032
0.015485
ratio
0.8336535267
0.2107944012
0.8336535038
1.0681772804
1.3978802578
0.4950471688
0.2696285292
0.9347752469
0.9347752622
0.6248402575
ratio 07
1.3978802578
0.4950471688
0.6248402575
14. Output to a permanent dataset.
Table B-16. Output to permanent dataset and create 2015 projected emissions by
multiplying the ratio by emis nei, creating a new variable called emis.
FIPS
01001
01001
01001
01001
01001
01001
06049
06049
06049
06049
SCC
2201001130
2201001130
2201001130
2201080130
2230070130
2230070130
2201001130
2201001130
2201001130
2201080130
CAS
226
71432
7440473
71432
226
71432
71432
7440020
7440473
71432
emis
0.0000583557
0.2429089282
0.0000833654
0.0106016595
6.9894013E-6
0.0102920306
0.2900164904
0.0002103244
0.0002991281
0.0096926381
ratio
0.8336535267
0.2107944012
0.8336535038
1.0681772804
1.3978802578
0.4950471688
0.2696285292
0.9347752469
0.9347752622
0.6248402575
emis nei
0.00007
1.15235
0.0001
0.009925
5E-6
0.02079
1.075615
0.000225
0.00032
0.015485
B-6
-------
0 ©
1 999 NEIonroad inventory P
on99_sept04.sas7bdat Create ^ewter
NMIM output
| (xx=05,07,10,15, 20, 30)
mporary SCC \~*\ hddv nmim
are road type !
-* \ \
I mnnn dal ^ | concaLenaLe data <^
By FIPS/SCC/CAS. i
Output matching observations |^ '
and observations from NM1M ; 1
not in NEI to separate datasets. ! /— x
- -t - -J4 ti9
^ -
mergedl emissions by
©1 FIPS/CAS
Y and calculate
Merge by new projection
FIPS/CAS ratios
where SCC ^— i 1
does not begin T
with 2230070 I— cnty sum
merged2 (~T\
t ©
xynimn sum
t
Copy xylenes, nickel, and
manganese observations to
new
observations for the other
xylene, nickel, and
manganese CAS
In the NEI.
04
|
Cll '
. Subset out HDDV summed emissions
' ^y i
Merge with HDDV ratios and
Ratios to FIPS/CAS with SCC
apply ^ ^ hddvl
99^0070
I 0
© |
Recombine NMIM output with
"* summed HDDV emssions
1
vi)
~ Sum up Chromium III and
ChromiumVI for each
FIIPS/SCC
and calculate chromium
projection ratios.
nmim chrom — i ^5^
x
__TL
im_udi ^ ; concatenate data l^-1
1 C^
mn
T.
extract xylenes,
manganese, and
nickel
observations
Figure B-l. Projection of the 1999 NEI onroad inventory to 2007, 2010, 2015, 2020, and 2030.
B-7
-------
B.2 Onroad Precursor Emissions (see 3.4.3)
Following are the steps used to project the onroad precursor emissions:
1. Read in the precursor HAP table used for EMS-HAP and subset the data to the
acetaldehyde, acrolein, formaldehyde, and propionaldehyde precursors, with the
exception of 1,3-butadiene, acetaldehyde, and MTBE.
2. The precursor onroad inventory was subset to the acetaldehyde, acrolein, formaldehyde,
and propionaldehyde precursors by merging with the output of step 1 by CAS. Also
created a variable called CAS1 that is set to a value of "VOC". This was to be used for
merging with the NMEVI inventory.
3. The NMIM data was split into non-HDDV emissions and HDDV emissions. Summed up
the heavy-duty diesel vehicle emissions in NMIM to create a total HDDV emission
number for each FIPS/ HDDV road type (last 3 characters of SCC code) and exhaust or
evaporative type. Created a new SCC, by replacing the seventh digit of the SCC with a
zero. Output dataset for HDDV emissions was hddv and for non-HDDV emissions, voc.
For both datasets created a new variable, CAS1 which was set to "VOC".
4. Summed up the HDDV emissions with the new SCC by FIPS. Calculated new projection
factors for each FIPS/SCC using Equation 2 (Section 3.3.2). These factors would be
applied to SCC codes beginning with 2230070YY# for each FIPS/CAS in the 1999
precursor inventory. These SCC codes are shown in Table 25.
5. Concatenated the HDDV and non-HDDV data with projection factors.
6. Merged step 5 output with the 1999 precursor inventory by FIPS/SCC/CAS1. Output all
observations for 1999 precursor inventory. Some 1999 observations did not have a
matching observation by FIPS/SCC/CAS1 in the NMIM data.
7. Extracted the emissions where the SCC code did not contain X or V, i.e. the total SCC
emissions (exhaust + evaporative).
8. To provide projection factors for the non-matching data, summed all the emissions in
each county across all SCC codes and calculated a new projection factor using Equation
2.
9. Merged the output from step 7 with the output from step 6 by FIPS.
10. Applied the projection factors to each FIPS/SCC/CAS.
11. Extracted the onroad emissions for 1,3-butadiene, acetaldehyde, and MTBE from the
MS AT onroad inventory.
B-8
-------
12. Appended output from step 10 to output from step 9.
13. If emissions were 1,3-butadiene, acetaldehyde, or MTBE, set the emis variable
(emissions variable for EMS-HAP) equal to the appropriate year emissions (emis_xx
where xx is 15, 20, or 30). Otherwise, projected the emissions from 1999 by multiplying
the 1999 emissions by the projection factor.
The flowchart of the projection processing is shown in Figure B-2.
Precursor onroad HAP table
0
Retain only CAS numbers of
MSAT HAP rprecursors
(except for precursors which
are MSAT HAPs)
1999 precursor onroad inventory
on99pre_out.sas7bdat
j Merge by CAS and exclude
I | Puerto Rico and the Virgin
—^ Islands and create new
j variable CAS 1 with value of
i "VOC."
NMIM VOC emissions
by FIPS/SCC containing
projection facotrs. For
each FIPS/SCC, the
dataset contain s the total
SCC emissions as well as
the evaporative and
exhaust components
(Nmim_203_30.sas7bdat)
Split data into non HDDV
emissions and HDDV
emisions.For HDDV
emisisons create a new SCC
by changing the seventh digit
of the SCC to a 0. Create a
new variable CAS 1 with
value "VOC."
onroad_20XX
where XX is
15, 20, or 30.
j Sum up HDDV
| emissions by
! FIPS/SCC/CAS
| 1. Calculate
I new projection
j factors
If an MSAT HAP, set the
emis variable equal to the
appropriate year's
emissions. Otherwise, apply
projection factors to 1999
precursor emissions. Output
to permanent dataset.
Retain FIPS/SCC where
SCC is the total
emissions for the SCC.
Concatenate
datasets
I
Merge 1999
precursor
emissions and
NMIM emissions
by
FIPS/SCC/CAS 1
I
SumbyFIPS/CASl
and calculate new
projection factors
Subset to 1,3-Butadiene,
H Acetaldehyde, and
MTBE
If no projection
factor from first
merge, then use
new factor from
voc sum
Projected emissions
from MSAT
(onroad_proj.sas7bdat)
Figure B-2. Precursor onroad inventory projection processing.
B-9
-------
This page intentionally blank
B-10
-------
Appendix C: Example calculations of nonroad projections
C.I Locomotive and commercial marine vessel emissions
C.1.1 Development of Projection Factor Files for locomotives and commercial marine vessels
emission projections
The development of the projection factor files included the following steps performed in a SAS®
program called loco_marine.sas which can be found in the docket for the MSAT rule (EPA-HQ-
OAR-2005-0036):
1. Read the 1999 NEI nonroad inventory, nonroad_fixed_airports.sas7bdat (found in the
MSAT rule docket EPA-HQ-OAR-2005-0036), and extracted the HAP emissions by
CAS for the locomotive and commercial marine vessel SCC codes. Emissions were
summed by CAS and SCC. HAP names were then assigned by CAS numbers.
Emissions were then summed by HAP name and SCC. This was done because some
HAPs such as Chromium III or Chromium VI were composed of several CAS numbers.
This step was done as a matter of convenience for the user. The summations were done
for the entire U.S. excluding Puerto Rico and the Virgin Islands.
2. The resulting emissions from step 1 were transferred to a PC where they were imported
into an Excel spreadsheet where the projection factors for the various years were added to
the emissions based on the criteria in Tables 4 and 6. The spreadsheet name is
loco_marine.xls
3. The Excel spreadsheet was then imported into SAS® where SAROAD codes were added
based on the HAP name in the SAS® program loco_marine.sas. For HAPs with two
SAROAD codes, i.e. the metals and naphthalene, both SAROAD codes were assigned
and the projection factors were associated with each SAROAD.
4. Output the SAROAD/SCC/projection factors to text files, locomotive_gf.txt for
locomotives and marine_cv.txt for commercial marine vessels. The files contained the
growth factors for all MSAT years.
Figures C-l and C-2 show sample records of the projection factor files for locomotive and
commercial marine vessels.
C-l
-------
228500000043231 0.9962 0.9441 0.9127 0.8986 0.8725 0.8277
228500000043504 0.9962 0.9441 0.9127 0.8986 0.8725 0.8277
228500000059992 1.0000 1.0000 1.0000 1.0000 1.0000 1.0000
228500000059993 1.0000 1.0000 1.0000 1.0000 1.0000 1.0000
Figure C-l. Sample records of the locomotive_gf.txt file. Variables are in the following order:
SCC, SAROAD, 2001 projection factor, 2007 projection factor, 2010 projection factor, 2015
projection factor, 2020 projection factor, and 2030 projection factor. Note that the 2001
projection factor is not used.
2280000000
2280000000
2280000000
2280000000
43218
43231
59992
59993
1
1
1
1
0352
0352
0181
0181
1.1188
1.1188
1.0743
1.0743
1.1511
1.1511
1.1036
1.1036
1.2306
1.2306
1.1541
1.1541
1.3505
1.3505
1.2070
1.2070
1.7142
1.7142
1.3202
1.3202
Figure C-2. As for Figure 1, except for the commercial marine vessel projection factor file,
marine cv.txt.
C. 1.2. Projection of the 1999 locomotive and commercial marine vessel emissions
Projection of the 1999 locomotive and commercial marine vessel emissions was performed in
marine_locomotive_growth.sas (found in the MSAT rule docket EPA-HQ-OAR-2005-0036) and
included the following steps. The methodology of the program is shown in Figure C-3 as a
flowchart with example calculations shown with each step:
1. Combined the commercial marine vessels and locomotive SAROAD/SCC/projection
factor data into one file.
Table C-l. Partial listing of locomotive and commercial marine vessel growth factors by
SAROAD after reading in growth factor files, concatenating, and sorting. Benzene and
unspeciated chromium growth factors are shown.
SCC
2280000000
2280000000
2280000000
2280002200
2280002200
2280002200
2285002005
2285002005
2285002005
2285002010
2285002010
2285002010
SAROAD
45201
80141
80341
45201
80141
80341
45201
80141
80341
45201
80141
80341
gf99_01
1.0352
1.0181
1.0181
1.0196
1.0181
1.0181
0.9962
1.0000
1.0000
0.9962
1.0000
1.0000
gf99_07
.1188
.0743
.0743
.0482
.0743
.0743
0.9441
1.0000
1.0000
0.9441
1.0000
1.0000
gf99_10
.1511
.1036
.1036
.0498
.1036
.1036
0.9127
.0000
.0000
0.9127
.0000
.0000
gf99_15
1.2306
1.1541
1.1541
1.0552
1.1541
1.1541
0.8986
1.0000
1.0000
0.8986
1.0000
1.0000
gf99_20
1.3505
1.207
1.207
1.0797
1.207
1.207
0.8725
1.0000
1.0000
0.8725
1.0000
1.0000
gf99_30
1.7142
1.3202
1.3202
1.177
1.3202
1.3202
0.8277
1.0000
1.0000
0.8277
1.0000
1.0000
C-2
-------
2.
Read in the text file, haptabl_nonroadGEN_toxwt.txt to get the CAS/SAROAD cross
reference for nonroad HAPs. The data was sorted to eliminate duplicate CAS numbers.
For metals, this usually meant the CAS became associated with the fine particle
SAROAD or the lowest numbered SAROAD for the metal.
Table C-2. CAS/SAROAD cross reference for benzene and unspeciated chromium after
sorting by CAS/SAROAD.
CAS
136
136
71432
7440473
7440473
SAROAD
80141
80341
45201
80141
80341
3.
Merged the output from steps 1 and 2 together so that now the projection factors were
associated with CAS and SCC codes instead of SAROAD and SCC codes. Metals were
associated with the fine SAROAD codes usually but duplicate records were put into the
projection factor text files for the coarse SAROAD codes as a precaution.
Table C-3. Partial listing
after sorting by SCC/CAS
merged CAS/SAROAD cross-reference with growth factors
, eliminating duplicate CAS observations.
SCC
2280000000
2280000000
2280000000
2280002200
2280002200
2280002200
2285002005
2285002005
2285002005
2285002010
2285002010
2285002010
SAROAD
80141
45201
80141
80141
45201
80141
80141
45201
80141
80141
45201
80141
gf99_01
1.0181
1.0352
1.0181
1.0181
1.0196
1.0181
1.0000
0.9962
1.0000
1.0000
0.9962
1.0000
gf99_07
1.0743
1.1188
1.0743
1.0743
1.0482
1.0743
1.0000
0.9441
1.0000
1.0000
0.9441
1.0000
gf99_10
1.1036
1.1511
1.1036
1.1036
1.0498
1.1036
1.0000
0.9127
1.0000
1.0000
0.9127
1.0000
gf99_15
.1541
.2306
.1541
.1541
.0552
.1541
.0000
0.8986
.0000
.0000
0.8986
1.0000
gf99_20
1.207
1.3505
1.207
1.207
1.0797
1.207
1.0000
0.8725
1.0000
1.0000
0.8725
1.0000
gf99_30
.3202
.7142
.3202
.3202
1.177
.3202
.0000
0.8277
.0000
.0000
0.8277
1.0000
CAS
136
71432
7440473
136
71432
7440473
136
71432
7440473
136
71432
7440473
4. Summed the locomotive and commercial marine vessel emissions from the 1999 NEI
nonroad inventory for QA purposes.
Total locomotive and commercial marine vessels before processing among all HAPs is
13,085.96 tons
C-2
-------
5. Merged the nonroad emissions with the projection factors by CAS/SCC across all FIPS, retaining only the locomotive and
commercial marine vessel HAP and SCC emissions. Summed the 1999 emissions again to check against the emissions total
from step 4. These emissions should be the same.
Table C-4. Partial listing of merged emissions for Los Angeles County, CA (FIPS=06037) after merging nonroad inventory
with growth factors.
SCC
2280000000
2280002200
2280002200
2285002005
2285002005
2285002010
2285002010
FIPS
06037
06037
06037
06037
06037
06037
06037
CAS
71432
71432
7440473
71432
7440473
71432
7440473
emis
0.764
0.0100924896
2.1818725E-6
5.956
0.001143
1.336
0.0002790113
gf99_01
1.0352
1.0196
1.0181
0.9962
1.0000
0.9962
1.0000
gf99_07
1.1188
1.0482
1.0743
0.9441
1.0000
0.9441
1.0000
gf99_10
1.1511
1.0498
1.1036
0.9127
1.0000
0.9127
1.0000
gf99_15
1.2306
1.0552
1.1541
0.8986
1.0000
0.8986
1.0000
gf99_20
1.3505
1.0797
1.207
0.8725
1.0000
0.8725
1.0000
gf99_30
1.7142
1.177
1.3202
0.8277
1.0000
0.8277
1.0000
Summed emissions of 1999 emissions after merger is 13,085.96.
6. Projected the 1999 emissions to future years at the FIPS/SCC/CAS level by multiplying each future year's growth factor by the
1999 emissions.
®
7. Output to a SAS dataset for later use with the other nonroad emissions for projection to future years.
Table C-5. Partial listing of projected emissions (steps 6 and 7). 1999 base emissions and growth factors not shown.
SCC
2280000000
2280002200
2280002200
2285002005
2285002005
2285002010
2285002010
FIPS
06037
06037
06037
06037
06037
06037
06037
CAS
71432
71432
7440473
71432
7440473
71432
7440473
emis 01
0.7908928
0.0102903024
2.2213644E-6
5.9333672
0.001143
1.3309232
0.0002790113
emis_07
0.8547632
0.0105789476
2.3439856E-6
5.6230596
0.001143
1.2613176
0.0002790113
emis 10
0.8794404
0.0105950955
2.4079145E-6
5.4360412
0.001143
1.2193672
0.0002790113
emis 15
0.9401784
0.010649595
2.518099E-6
5.3520616
0.001143
1.2005296
0.0002790113
emis_20
1.031782
0.010896861
2.6335201E-6
5.19661
0.001143
1.16566
0.0002790113
emis 30
1.3096488
0.0118788602
2.8805081E-6
4.9297812
0.001143
1.1058072
0.0002790113
8. Output first two records of projected dataset to manually QA calculations.
C-4
-------
haptable nonroadGENtoxwt.txt
(cross reference of saroad
and CAS numbers)
Commercial marine
vessel projection factors
by saroad and SCC
(m ari ne_cv_gf. txt)
locomotive projection factors
by saroad and SCC
(1 ocom on' ve_gf. txt)
I Cross reference saroad
I and CAS numbers | " ^_^
, _' Q
Projection factors by I 1999 NEI "onroad inventory
CAS/SCC (non99_oct03_fixed_airports)
(loco_marine_cas)
5
Combine emissions with projection factors j^ | j Calculate total locomotive and j
by SCC and CAS for each county !^ | marine vessel emissions for j
1 ' | QA purposes. Output to 1st file j
1" * I i Calculate total locomotive and | f i
loco_marlne_emls | ^ marlne vessel emlsslons for | . , Visually compare in 1st file j
I QA purposes. Output to 1st file
Apply projection factors for each [
Year to emissions (gf*emis) i
) r .
^j Output to permament
*] dataset jH loco_marine_growth
^
i Output first 2 observations to I I I /"\
I manually QA results 1_^|_jrowth_qaJ {sj
Figure C-3. Projection of 1999 locomotive and commercial marine vessel emissions to 2002,
2007, 2010, 2015, 2020, and 2030. Data is represented by solid boxes with processes or steps
denoted by dashed boxes. Input and final output data denoted by heavy solid boxes. Steps
outlined in text are shown as circled numbers. Unless otherwise denoted, data represents SAS*1
datasets.
C-5
-------
C.2 Remaining nonroad emissions (excluding aircraft, locomotives, and commercial marine
vessels)
The following summarizes the steps used in the SAS® program nonroad. sas (found in the MS AT
rule docket EPA-HQ-OAR-2005-0036) with example calculations for Alameda County, CA
(FIPS=06001). A detailed flow chart is shown in Figure C-4.
1. The 1999 NEI nonroad inventory was subsetted to the MS AT HAPs, excluding aircraft,
marine commercial vessels, and locomotive emissions. These were projected separately
from the other nonroad emissions as documented in Section 3.1.
Table C-6. Partial listing of 1999 NEI nonroad emissions for Alameda County after
subsetting inventory to MSAT HAPs and excluding aircraft, locomotive, and commercial
marine vessel SCC emissions. SCC descriptions are listed for informational purposes.
Emis nei is the emissions variable and polldesc is the pollutant description.
FIPS
06001
06001
06001
06001
06001
06001
06001
06001
06001
06001
06001
06001
06001
06001
06001
06001
06001
06001
06001
06001
SCC
2260001010
2260001010
2260001010
2260002000
2260002000
2260002000
2260002000
2265008000
2265008000
2265008000
2265008000
2268003000
2268003000
2268003000
2268003000
2282000000
2282000000
2282000000
2282000000
2282000000
CAS
1330207
71432
7440473
108383
71432
7440473
95476
108383
71432
7440473
95476
108383
71432
7440473
95476
106423
108383
71432
7440473
95476
emis nei
14.317113886
2.3263908964
0.0000122711
0.30705
0.23736
0.00015
0.10695
0.6141
0.47472
0.00015
0.2139
0.01102
0.12122
0.0013
0.01102
0.0015
0.00915
0.03
2E-6
0.0051
polldesc
Xylenes (mixture of o,
m, and p isomers)
Benzene
Chromium
m-Xylene
Benzene
Chromium
o-Xylene
m-Xylene
Benzene
Chromium
o-Xylene
m-Xylene
Benzene
Chromium
o-Xylene
p-Xylene
m-Xylene
Benzene
Chromium
o-Xylene
SCC description
Mobile Sources,Off-highway
Vehicle Gasoline, 2-
Stroke,Recreational
Equipment,Motorcycles: Off-road
Mobile Sources, Off-highway
Vehicle Gasoline, 2-
Stroke,Construction and Mining
Equipment, Total
Airport Support Equipment, Total,
Off-highway 4-stroke
Mobile Sources,CNG,Industrial
Equipment, All
Mobile Sources, Pleasure Craft, All
Fuels, Total, All Vessel Types
2. NMIM SCC emissions were summed to a "Total" category for each SCC category (first 7
digits of SCC followed by 3 zeros) for each FIPS/HAP/SCC. These SCC codes were
found in a separate SAS® program missing_scc.sas (found in the MSAT rule docket
EPA-HQ-OAR-2005-0036) that was run before any nonroad processing. See Table 19
for list.
C-6
-------
Table C-7. Partial listing of NMIM 1999 and 2015 emissions for Alameda County after
summing SCC emissions to total SCC category for each HAP.
FIPS
06001
06001
06001
06001
SCC
2260002000
2260002000
2260002000
2260002000
CAS
1330207
16065831
18540299
71432
emis99
14.632461572
9.9206348E-6
5.1106299E-6
3.3503477212
emislS
5.9177500529
7.218985E-6
3.718871E-6
1.3761635963
3. NMIM pleasure craft emissions, first four SCC digits 2282, were summed and assigned
SCC 2282000000 for each FIPS/SCC.
Table C-8. Partial listing of NMIM 1999 and 2015 emissions for Alameda County after
summing pleasure craft emissions into one SCC for each HAP (Section 3.4.3, step 3).
FIPS
06001
06001
06001
06001
SCC
2282000000
2282000000
2282000000
2282000000
CAS
1330207
16065831
18540299
71432
emis99
45.980996386
0.0000759904
0.0000391466
9.8014022292
emislS
18.529284389
0.0000757439
0.0000390196
3.6067584954
4. Concatenated the total SCC emissions from step 2 and pleasure craft emissions from step
3.
Table C-9. Partial listing of NMIM emissions and 2015 to 1999 ratios for Alameda
County after concatenating the total SCC emissions and the total pleasure craft emissions.
Ratio 15 has been renamed ratio.
FIPS
06001
06001
06001
06001
06001
06001
06001
06001
SCC
2260002000
2260002000
2260002000
2260002000
2282000000
2282000000
2282000000
2282000000
CAS
1330207
16065831
18540299
71432
1330207
16065831
18540299
71432
emis99
14.632461572
9.9206348E-6
5.1106299E-6
3.3503477212
45.980996386
0.0000759904
0.0000391466
9.8014022292
emislS
5.9177500529
7.218985E-6
3.718871E-6
1.3761635963
18.529284389
0.0000757439
0.0000390196
3.6067584954
ratio
0.4044261469
0.7276737009
0.727673714
0.4107524683
0.4029770089
0.9967563592
0.9967563665
0.3679839283
5. Concatenated the data from step 4 with the original NMIM inventory.
C-7
-------
Table C-10. Partial listing of NMIM emissions after concatenating the total SCC and pleasure
craft emissions with the original NMIM emissions.
FIPS
06001
06001
06001
06001
06001
06001
06001
06001
06001
06001
06001
06001
SCC
2260002000
2260002000
2260002000
2260002000
2282000000
2282000000
2282000000
2282000000
2260001010
2260001010
2260001010
2260001010
CAS
1330207
16065831
18540299
71432
1330207
16065831
18540299
71432
1330207
16065831
18540299
71432
emis99
14.632461572
9.9206348E-6
5.1106299E-6
3.3503477212
45.980996386
0.0000759904
0.0000391466
9.8014022292
9.711512804
5.1118629E-6
2.6333838E-6
1.5561215132
emislS
5.9177500529
7.218985E-6
3.718871E-6
1.3761635963
18.529284389
0.0000757439
0.0000390196
3.6067584954
14.214579582
0.0000109192
5.6250196E-6
2.2015727758
ratio
0.4044261469
0.7276737009
0.727673714
0.4107524683
0.4029770089
0.9967563592
0.9967563665
0.3679839283
1.4636833487
2.1360424041
2.1360424142
1.4147820444
6. Extracted chromium III and chromium VI emissions from the output of step 5.
Table C-ll. Partial listing of extracted chromium emissions.
FIPS
06001
06001
06001
06001
06001
06001
SCC
2260001010
2260001010
2260002000
2260002000
2282000000
2282000000
CAS
16065831
18540299
16065831
18540299
16065831
18540299
emis99
5.1118629E-6
2.6333838E-6
9.9206348E-6
5.1106299E-6
0.0000759904
0.0000391466
emislS
0.0000109192
5.6250196E-6
7.218985E-6
3.718871E-6
0.0000757439
0.0000390196
ratio
2.1360424041
2.1360424142
0.7276737009
0.727673714
0.9967563592
0.9967563665
As with the onroad summed up NMIM chromium III and chromium VI emissions to
create total chromium. For the NEI nonroad inventory, chromium was reported with
either CAS 136 or CAS 74404734. To make sure all FIPS/SCC/CAS combinations are
covered, the summed chromium III and chromium VI emissions were assigned to both
Chromium CAS numbers. Therefore temporarily, chromium emissions were double
counted while processing the NMEVI output.
In the 1999 NEI nonroad inventory (just like in the onroad inventory), chromium was speciated as chromium III
and chromium VI. The emissions were summed and re-speciated by EMS-HAP to use a speciation factor of 18% of
total chromium is chromium VI.
-------
Table C-12. Partial listing of chromium emissions after summing by FIPS/SCC and
assigning unspeciated chromium CAS numbers and calculating a new ratio.
FIPS
06001
06001
06001
06001
06001
06001
sec
2260001010
2260001010
2260002000
2260002000
2282000000
2282000000
CAS
136
7440473
136
7440473
136
7440473
emis99
7.7452467E-6
7.7452467E-6
0.0000150313
0.0000150313
0.000115137
0.000115137
emislS
0.0000165442
0.0000165442
0.0000109379
0.0000109379
0.0001147635
0.0001147635
ratio
2.1360424076
2.1360424076
0.7276737054
0.7276737054
0.9967563617
0.9967563617
8. Concatenated step 7 output with step 5 output.
Table C-13. Partial listing of NMIM emissions after concatenating chromium emissions
with NMIM data.
FIPS
06001
06001
06001
06001
06001
06001
06001
06001
06001
06001
06001
06001
06001
06001
06001
06001
06001
06001
sec
2260002000
2260002000
2260002000
2260002000
2282000000
2282000000
2282000000
2282000000
2260001010
2260001010
2260001010
2260001010
2260001010
2260001010
2260002000
2260002000
2282000000
2282000000
CAS
1330207
16065831
18540299
71432
1330207
16065831
18540299
71432
1330207
16065831
18540299
71432
136
7440473
136
7440473
136
7440473
emis99
14.632461572
9.9206348E-6
5.1106299E-6
3.3503477212
45.980996386
0.0000759904
0.0000391466
9.8014022292
9.711512804
5.1118629E-6
2.6333838E-6
1.5561215132
7.7452467E-6
7.7452467E-6
0.0000150313
0.0000150313
0.000115137
0.000115137
emislS
5.9177500529
7.218985E-6
3.718871E-6
1.3761635963
18.529284389
0.0000757439
0.0000390196
3.6067584954
14.214579582
0.0000109192
5.6250196E-6
2.2015727758
0.0000165442
0.0000165442
0.0000109379
0.0000109379
0.0001147635
0.0001147635
ratio
0.4044261469
0.7276737009
0.727673714
0.4107524683
0.4029770089
0.9967563592
0.9967563665
0.3679839283
1.4636833487
2.1360424041
2.1360424142
1.4147820444
2.1360424076
2.1360424076
0.7276737054
0.7276737054
0.9967563617
0.9967563617
9. Extracted xylenes, nickel, and manganese observations from step 8 output.
Table C-14. Extracted xylenes emissions.
FIPS
06001
06001
06001
sec
2260002000
2282000000
2600010210
CAS
1330207
1330207
1330207
emis99
14.632461572
45.980996386
9.711512804
emislS
5.9177500529
18.529284389
14.214579582
ratio
0.4044261469
0.4029770089
1.4636833487
10. As with the onroad processing, copied the NMIM xylenes, nickel, and manganese NMIM
observations to duplicate observations with the other xylenes, nickel, and manganese
CAS numbers.
C-9
-------
Table C-15. Xylenes emissions after copying records to duplicate records and changing
CAS numbers to 106423, 108383, and 95476.
FIPS
06001
06001
06001
06001
06001
06001
06001
06001
06001
sec
2260002000
2260002000
2260002000
2282000000
2282000000
2282000000
2600010210
2600010210
2600010210
CAS
106423
108383
95476
106423
108383
95476
106423
108383
95476
emis99
14.632461572
14.632461572
14.632461572
45.980996386
45.980996386
45.980996386
9.711512804
9.711512804
9.711512804
emislS
5.9177500529
5.9177500529
5.9177500529
18.529284389
18.529284389
18.529284389
14.214579582
14.214579582
14.214579582
ratio
0.4044261469
0.4044261469
0.4044261469
0.4029770089
0.4029770089
0.4029770089
1.4636833487
1.4636833487
1.4636833487
11. Concatenated step 10 output with step 8 output.
Table C-16. Concatenated NMIM emissions and duplicate xylenes emissions.
FIPS
06001
06001
06001
06001
06001
06001
06001
06001
06001
06001
06001
06001
06001
06001
06001
06001
06001
06001
06001
06001
06001
06001
06001
06001
06001
06001
06001
sec
2260002000
2260002000
2260002000
2260002000
2282000000
2282000000
2282000000
2282000000
2260001010
2260001010
2260001010
2260001010
2260001010
2260001010
2260002000
2260002000
2282000000
2282000000
2260002000
2260002000
2260002000
2282000000
2282000000
2282000000
2600010210
2600010210
2600010210
CAS
1330207
16065831
18540299
71432
1330207
16065831
18540299
71432
1330207
16065831
18540299
71432
136
7440473
136
7440473
136
7440473
106423
108383
95476
106423
108383
95476
106423
108383
95476
emis99
14.632461572
9.9206348E-6
5.1106299E-6
3.3503477212
45.980996386
0.0000759904
0.0000391466
9.8014022292
9.711512804
5.1118629E-6
2.6333838E-6
1.5561215132
7.7452467E-6
7.7452467E-6
0.0000150313
0.0000150313
0.000115137
0.000115137
14.632461572
14.632461572
14.632461572
45.980996386
45.980996386
45.980996386
9.711512804
9.711512804
9.711512804
emislS
5.9177500529
7.218985E-6
3.718871E-6
1.3761635963
18.529284389
0.0000757439
0.0000390196
3.6067584954
14.214579582
0.0000109192
5.6250196E-6
2.2015727758
0.0000165442
0.0000165442
0.0000109379
0.0000109379
0.0001147635
0.0001147635
5.9177500529
5.9177500529
5.9177500529
18.529284389
18.529284389
18.529284389
14.214579582
14.214579582
14.214579582
ratio
0.4044261469
0.7276737009
0.727673714
0.4107524683
0.4029770089
0.9967563592
0.9967563665
0.3679839283
1.4636833487
2.1360424041
2.1360424142
1.4147820444
2.1360424076
2.1360424076
0.7276737054
0.7276737054
0.9967563617
0.9967563617
0.4044261469
0.4044261469
0.4044261469
0.4029770089
0.4029770089
0.4029770089
1.4636833487
1.4636833487
1.4636833487
C-10
-------
12. Merged the NEI and NMIM output, nei_dat and nmim_dat, by FIPS/SCC/CAS, retaining
all NEI observations. Split the data into a dataset that matched (called okay), i.e. has a
projection factor, and into dataset that did not matched (called need_ratio), i.e. no
project!on factors.
Table C-17. Partial listing of merged NEI and NMIM emissions, with all NEI emissions
retained.
FIPS
06001
06001
06001
06001
06001
06001
06001
06001
06001
06001
06001
06001
06001
06001
06001
06001
06001
06001
06001
06001
sec
2260001010
2260001010
2260001010
2260002000
2260002000
2260002000
2260002000
2265008000
2265008000
2265008000
2265008000
2268003000
2268003000
2268003000
2268003000
2282000000
2282000000
2282000000
2282000000
2282000000
CAS
1330207
71432
7440473
108383
71432
7440473
95476
108383
71432
7440473
95476
108383
71432
7440473
95476
106423
108383
71432
7440473
95476
emis_nei
14.317113886
2.3263908964
0.0000122711
0.30705
0.23736
0.00015
0.10695
0.6141
0.47472
0.00015
0.2139
0.01102
0.12122
0.0013
0.01102
0.0015
0.00915
0.03
2E-6
0.0051
emis99
9.711512804
1.5561215132
7.7452467E-6
14.632461572
3.3503477212
0.0000150313
14.632461572
45.980996386
45.980996386
9.8014022292
0.000115137
45.980996386
emislS
14.214579582
2.2015727758
0.0000165442
5.9177500529
1.3761635963
0.0000109379
5.9177500529
18.529284389
18.529284389
3.6067584954
0.0001147635
18.529284389
ratio
1.4636833487
1.4147820444
2.1360424076
0.4044261469
0.4107524683
0.7276737054
0.4044261469
0.4029770089
0.4029770089
0.3679839283
0.9967563617
0.4029770089
Table C-18. Listing of emissions still needing a ratio with six digit SCC, scc6.
FIPS
06001
06001
06001
06001
06001
06001
06001
06001
scc6
226500
226500
226500
226500
226800
226800
226800
226800
SCC
2265008000
2265008000
2265008000
2265008000
2268003000
2268003000
2268003000
2268003000
CAS
108383
71432
7440473
95476
108383
71432
7440473
95476
emis nei
0.6141
0.47472
0.00015
0.2139
0.01102
0.12122
0.0013
0.01102
emis99
emislS
ratio
C-ll
-------
Table C-19. Listing of emissions assigned a ratio with projected emissions, etnis.
FIPS
06001
06001
06001
06001
06001
06001
06001
06001
06001
06001
06001
06001
sec
2260001010
2260001010
2260001010
2260002000
2260002000
2260002000
2260002000
2282000000
2282000000
2282000000
2282000000
2282000000
CAS
1330207
71432
7440473
108383
71432
7440473
95476
106423
108383
71432
7440473
95476
emis nei
14.317113886
2.3263908964
0.0000122711
0.30705
0.23736
0.00015
0.10695
0.0015
0.00915
0.03
2E-6
0.0051
emis99
9.711512804
1.5561215132
7.7452467E-6
14.632461572
3.3503477212
0.0000150313
14.632461572
45.980996386
45.980996386
9.8014022292
0.000115137
45.980996386
emislS
14.214579582
2.2015727758
0.0000165442
5.9177500529
1.3761635963
0.0000109379
5.9177500529
18.529284389
18.529284389
3.6067584954
0.0001147635
18.529284389
ratio
1.4636833487
1.4147820444
2.1360424076
0.4044261469
0.4107524683
0.7276737054
0.4044261469
0.4029770089
0.4029770089
0.3679839283
0.9967563617
0.4029770089
emis
20.955721197
3.2913360685
0.0000262116
0.1241790484
0.0974962059
0.0001091511
0.0432533764
0.0006044655
0.0036872396
0.0110395178
1.9935127E-6
0.0020551827
13. For remaining FIPS/SCC/CAS combinations in the 1999 NEI that did not match the
NMIM results, created county-level HAP specific projection factors based on engine/fuel
type by summing emissions for 1999 NMIM and future year NMIM for each
FIPS/CAS/engine/fuel type. These were applied to all SCC codes with the relevant
engine/fuel type by HAP and by county. The engine fuel types were 2-stroke gasoline, 4-
stroke gasoline, diesel, LPG, CNG, and miscellaneous.
Table C-20. Partial listing of Alameda County emissions for SCC6 of 226500.
FIPS
06001
06001
06001
06001
scc6
226500
226500
226500
226500
SCC
2265008000
2265008000
2265008000
2265008000
CAS
108383
71432
7440473
95476
emis99
428.80604172
324.73697506
0.0056123247
428.80604172
emislS
253.71038624
197.05535819
0.0066134549
253.71038624
ratio
0.5916670046
0.6068152792
1.1783806775
0.5916670046
14. Merged the output from step 13, cnty_sum, with the need_ratio dataset from step 12.
Separated data into two datasets, observations with a projection factor (fill_data2) and
those without a projection factor (need_data2).
Table C-21. Alameda County emissions where a ratio was applied from cnty_sums and
applied to NEI emissions to calculate emis variable.
FIPS
06001
06001
06001
06001
scc6
226500
226500
226500
226500
SCC
2265008000
2265008000
2265008000
2265008000
CAS
108383
71432
7440473
95476
emis_nei
0.6141
0.47472
0.00015
0.2139
emis99
emislS
ratio
0.5916670046
0.6068152792
1.1783806775
0.5916670046
emis
0.3633427075
0.2880673493
0.0001767571
0.1265575723
C-12
-------
Table C-22. Alameda County emissions still needing a ratio with a CAS1 variable
assigned.
FIPS
06001
06001
06001
06001
scc6
226800
226800
226800
226800
sec
2268003000
2268003000
2268003000
2268003000
CAS
108383
71432
7440473
95476
CAS1
voc
voc
PM10-PRI
VOC
emis nei
0.01102
0.12122
0.0013
0.01102
emis99
emis 15
ratio
15. Even after the above step, there remained CNG and LPG emissions for California and
Texas (SCC codes beginning with 226800, 226801, and 226700) from the 1999 NEI
without an NMIM based projection factor. Per discussion with Madeleine Strum and
Rich Cook, the VOC or PM county level ratios for CNG and LPG as fuel types were
calculated and used for the HAPs in the inventory. Particulate HAPs received the PM
ratios and gaseous HAPs received the VOC ratios. Calculated county level projection
factors by summing VOC or PM emissions across all SCC codes that used CNG and LPG
as fuel types for 1999 NMIM and future year NMIM output and dividing the future year
summed emissions by the 1999 summed emissions for each county.
Table C-23. Summed VOC and PM10-PRI NMIM emissions for Alameda County by
six digit SCC code emissions with ratio.
FIPS
06001
06001
scc6
226800
226800
CAS1
PM10-PRI
VOC
emis99
0.6706211131
2.2227693134
emislS
1.0020067495
0.3633906096
ratio_15
1.4941473358
0.1634855257
16. Merged the projection factors from step 15 (ca_tx_sum) with the need_data2 output from
step 14 and apply factors. Output dataset was fill_data3.
Table C-24. Projected emissions for Alameda County using the county sums for LPG
and CNG.
FIPS
06001
06001
06001
06001
scc6
226800
226800
226800
226800
SCC
2268003000
2268003000
2268003000
2268003000
CAS
108383
71432
7440473
95476
CAS1
VOC
voc
PM10-PRI
VOC
emis_nei
0.01102
0.12122
0.0013
0.01102
emis
0.0018016105
0.0198177154
0.0019423915
0.0018016105
ratio
0.1634855257
0.1634855257
1.4941473358.
0.1634855257
17. Concatenated datasets okay, fill_data2, and fill_data3. Output was merged2.
18. Appended the locomotive and commercial marine vessel projected emissions to step 17
output and output data to permanent dataset, nonroad_20xx where xx is 07, 10, 15, 20, or
30.
C-13
-------
Table C-25. Projected nonroad emissions with appended locomotive and commercial
marine vessels after sorting by FIPS/SCC/CAS (steps 17 and 18).
FIPS
06001
06001
06001
06001
06001
06001
06001
06001
06001
06001
06001
06001
06001
06001
06001
06001
06001
06001
06001
06001
06001
06001
06001
06001
06001
06001
06001
06001
06001
sec
2260001010
2260001010
2260001010
2260002000
2260002000
2260002000
2260002000
2265008000
2265008000
2265008000
2265008000
2268003000
2268003000
2268003000
2268003000
2280000000
2280000000
2280000000
2280000000
2282000000
2282000000
2282000000
2282000000
2282000000
2285000000
2285000000
2285000000
2285000000
2285000000
CAS
1330207
71432
7440473
108383
71432
7440473
95476
108383
71432
7440473
95476
108383
71432
7440473
95476
106423
108383
71432
95476
106423
108383
71432
7440473
95476
106423
108383
71432
7440473
95476
emis nei
14.317113886
2.3263908964
0.0000122711
0.30705
0.23736
0.00015
0.10695
0.6141
0.47472
0.00015
0.2139
0.01102
0.12122
0.0013
0.01102
0.0045
0.02745
0.09
0.0153
0.0015
0.00915
0.03
2E-6
0.0051
0.038
0.2318
0.76
0.000192
0.1292
ratio
1.4636833487
1.4147820444
2.1360424076
0.4044261469
0.4107524683
0.7276737054
0.4044261469
0.5916670046
0.6068152792
1.1783806775
0.5916670046
0.1634855257
0.1634855257
1.4941473358.
0.1634855257
1.2306
1.2306
1.2306
1.2306
0.4029770089
0.4029770089
0.3679839283
0.9967563617
0.4029770089
0.8986
0.8986
0.8986
1.0000
0.8986
emis
20.955721197
3.2913360685
0.0000262116
0.1241790484
0.0974962059
0.0001091511
0.0432533764
0.3633427075
0.2880673493
0.0001767571
0.1265575723
0.0018016105
0.0198177154
0.0019423915
0.0018016105
0.0055377
0.03377997
0.110754
0.01882818
0.0006044655
0.0036872396
0.0110395178
1.9935127E-6
0.0020551827
0.0341468
0.20829548
0.682936
0.000192
0.11609912
C-14
-------
©
1999 NEI onroad inventory
non99_oct03_fixed_airports.sas7bdat
Extract MSAT HAP emissions
Excluding aircraft, marine vessel,
and locomotive emissions.
o
[ Sum up Chromium III and
Chromium VI for each FIPS/SCC.
rr©-
NMIM VOC and PM emissions by
FIPS/SCC( voc_pm.sas7bdat)
1
| extract chromium | f 6
T
Extract and summarize CA and
TX CNG and LPG emissions by
FIPS and VOC or PM. Calculate
projection factors.
Summarize emissions by
FlPS/engine type/CAS and
calculate new ratios.
Copyxylenes, nickel, and
manganese emissions to duplicate
observations using other CAS for
xylenes, manganese, and nickel
xylenes_ni_mn
jj
Proj ected locomotive & marine
commercial vessel emissions.
loco marine growth. sas7bdat
1
r
Merge by FlPS/engine
type/VOC/PM flag and project
emissions. Engine type refers to
CNG or LPG
Merge with NEI
ByFIPS/SCC/CAS.
Output all observations from
NEI
Set future year MTBE emissions
and projection factors to zero.
Separate observations with a
projection factor from those that
do not. Project emissions to
future year in okay dataset
Separate observations with a
projection factor from those that
do not. Assign a VOC or PM flag
to the need_data2 observations
and project emissions in fill_data
Figure C-4. Flowchart of nonroad projections. Box tYPes as for Fi§ure C'3
C-15
-------
C.3 Nonroad Precursor Emissions
The following contains the steps used for projecting precursors from nonroad emission
categories covered by the NONROAD model.
1. Subset 1999 precursor nonroad inventory to the precursors for acetaldehyde, acrolein,
formaldehyde, and propionaldehyde using the precursor HAP table from EMS-HAP to
get the CAS numbers associated with the appropriate precursors.
2. Excluded locomotive, commercial marine vessel, and aircraft emissions. Also subset
data to exclude Puerto Rico and the Virgin Islands. Create a variable called CAS1 with
value "VOC."
3. Summed NMIM VOC SCC emissions to a "Total" category for each SCC category (first
7 digits of SCC) for each FIPS/ SCC.
4. Summed up NMEVI VOC pleasure craft emissions, first four SCC digits 2282, and
assigned SCC 2282000000 for each FIPS/SCC.
5. Combined output from steps 3 and 4 with original NMIM output and calculated new
projection factors for the total SCC codes and pleasure craft emissions.
6. Merged the 1999 precursor inventory and NMIM output, by FIPS/SCC/CAS1, retaining
all NEI observations.
7. For remaining non-matching FIPS/SCC/CAS combinations, created a county level HAP
specific projection factor based on engine/fuel type by summing emissions for 1999 and
future year NMIM VOC for each FIPS and calculate a county level projection factor.
These were then assigned to all FIPS/SCC codes for each pollutants based on engine
type. The engine fuel types were: 2-stroke gasoline, 4-stroke gasoline, diesel, LPG,
CNG, residual, and miscellaneous.
8. Merged the output from step 6,with output from step 7.
9. Appended output from step 8 to the matched data from step 6 and applied projection
factors to create 2015, 2020, and 2030 emissions.
10. Extracted precursor locomotive and commercial marine vessel emissions from the
precursor locomotive and commercial marine vessel projected inventory.
11. Appended output from step 10 to step 9 output.
12. Extracted 1,3-butadiene, acetaldehyde, MTBE, and methanol nonroad emissions
(excluding aircraft) from the interpolated nonroad inventory (see Appendix B) that
contains MSAT and non-MSAT HAPs.
C-16
-------
13. Appended output from step 12 to output from step 11.
14. Split data into separate datasets for 2015, 2020, and 2030.
The flowchart of the processing is shown in Figure C-5.
C-17
-------
Precursor nonroad HAP table
haptable_precursor.txt
Retain only CAS numbers of MS AT HAP
rprecursors (except for precursors which are
MSAT HAPs)
1999 precursor nonroad inventory
Non99pr e_out. sas7bdat
Subset inventory to CAS in cas_sarroad and
exclude Puerto Rico and the Virgin Islands
and create new variable CAS1 with value of
"VOC."
0
NMIM VOC emisions (voc pm.sasTbdat)
j Extract 1,3-Butadiene,
| Acetaldehyde, MTBE,
I and Methanol
! Split data (^ all_precursors
•*
merge by FlPS/engine
© type/CAS 1. Use first 6
for engine type
K H ^
-W fi
©
and apply
projection factors
Figure C-5. Nonroad projection processing for precursors.
C-18
-------
C.4 Non-MSAT HAPs non-road processing
The remaining nonroad inventory contained HAPs not covered by the NMIM results. These
included Antimony, Beryllium, Cadmium, Chlorine, Cobalt, Cumene, Lead, Methanol, Methyl
Ethyl Ketone, Phenol, Phosphorus, and Selenium. In order to project these HAPs for the GPRA
project, NMEVIVOC and PM emissions would be used to calculate the projection factors instead
of the actual HAP emissions as done for the MS AT HAPs. The metals, Antimony, Beryllium,
Cadmium, Cobalt, Lead, and Selenium would use the PM emissions for projection and all the
other HAPs would use the VOC emissions.
The processing of the nonroad inventory for non MS AT HAPs followed a very similar procedure
as the nonroad processing for MS AT HAPs:
1. Subset 1999 NEI nonroad inventory to the non MSAT HAPs, excluding aircraft, marine
commercial vessels, and locomotive emissions. These were projected separately from the
other nonroad emissions as documented in Section 3.1. Assign a variable, CAS1
denoting whether the HAP is to use VOC or PM projection factors. CAS1 = VOC for
HAPs using VOC and CAS1 = PM for HAPs using PM factors.
2. Assigned a CAS 1 flag to the NMIM output to be used for merger with the NEI data.
CAS1=CAS.
3. Summed NMIM SCC emissions to a "Total" category for each SCC category (first 7
digits of SCC) for each FIPS/CAS1/SCC.
4. Summed up NMIM pleasure craft emissions by FIPS/CAS1, first four SCC digits 2282,
and assign SCC 2282000000 to these emissions. Combine these emissions and the
emissions from step 3 to the original NMIM output.
5. Merged the NEI and NMIM output, by FIPS/SCC/CAS1, retaining all NEI observations.
Split the data into a dataset that matched i.e. has a projection factor, and into dataset that
did not matched, i.e. no projection factors.
6. For remaining non-matching FIPS/CAS/HAP combinations, created a county level VOC
or PM specific projection factor based on engine/fuel type by summing emissions for
1999 and future year NMIM for each FIPS/HAP where HAP is VOC and PM and
calculate a county level projection factor. These were then assigned to all SCC codes for
each CAS based on the CAS1 value. The engine fuel types were: 2-stroke gasoline, 4-
stroke gasoline, diesel, LPG, CNG, and miscellaneous.
C-19
-------
7. Merged the output from step 6, cnty_sum, with the nonmatched dataset from step 5.
Separate data into two datasets, observations with a projection factor (fill_data2) and
those without a projection factor (need_data2).
8. Concatenated output datasets from step 7 and the matched dataset from step 5.
9. Appended the non-MSAT locomotive and commercial marine vessel emissions and
aircraft projected emissions to the output from step 8.
10. Projected the emissions to non-MSAT years using Equations 3 and 4.
After projecting the non-MSAT emissions, the MSAT HAPs were appended to the non-MSAT
projections. This included the locomotive and commercial marine vessel emissions. Also
MSAT projected aircraft emissions were appended to the data. MSAT HAPs were then
projected to non-MSAT years using Equations 3 and 4.
C-20
-------
Appendix D: Risk Calculations
D.I Cancer risk calculation methodology
The following steps detail the cancer risk calculations in cancer_risk.sas (found in the MS AT
rule docket EPA-HQ-OAR-2005-0036).
1. Read in a sorted a SAS® dataset of census tracts and retained FIPS, state name, county
name, tract ID, and tract population.
2. Read in the URE and carcinogenic class for each MSAT HAP from the SAS® dataset
msat_haps_tox_factors.sas7bdat (found in the MSAT rule docket EPA-HQ-OAR-2005-
0036), keeping only HAPs where the URE is nonzero and nonmissing. The dataset,
msat_haps_tox_factors.sas7bdat was created from the ACCESS table, OToxicity in
Master.mdb from Roy Smith.
3. From the output of step 2, created a list of HAPs by SARD AD to be read in by a SAS®
macro for further processing, beginning with step 4.
4. Read in the tract level HAPEM5 output.
5. Merged the output from step 4 with the URE data from step 2 using PROC SQL.
6. Merged the tract population data from step 1 with the output from step 5.
7. For each source sector in each tract, multiplied the tract level source sector concentration
by the URE.
8. Output the tract level risk estimates to a permanent dataset for the HAP.
9. After computing risk estimates for each HAP, appended the HAP risk estimates to a
master dataset containing risk estimates for all HAPs.
Steps 5 through 9 were executed in the SAS® macro calc_risk.
10. Repeated steps 4 through 9 for each HAP in the MACRO calc_risk.
11. Sorted the master dataset by carcinogen class and FlPS/tract and summed the source
category risks within carcinogen classes for each FIPS and tract. For example, for each
tract, summed the risks for 1,3-butadiene, benzene, and nickel for Class A carcinogens.
12. From the output of step 11, output permanent datasets for each carcinogen class.
13. Sorted the master dataset by FIPS/TRACT and calculate a total risk (across all HAPs) for
each source sector at the tract level.
D-l
-------
14. Output the total risk estimates to a permanent dataset.
15. Calculated national risk distributions (percentiles and median) for each source category
for each carcinogen class across all census tracts.
16. Calculated a national risk distribution for each source category across all tracts and
carcinogen class, i.e. total risk.
17. Concatenated outputs from step 15 and step 16.
18. Output to a permanent dataset.
All 18 steps are done in the SAS® MACRO main for each modeling year and 1999 where
the argument for the macro is the year and the flowchart of the program is shown in
Figure D-l.
D-2
-------
^
URE and Rfc factors by ^ Retain ^ HAp§ ^ nonz£rc
SAROAD ™ - - TTOT, ^ CAO^
i nonmissing URE. Keep SARO
msat haps tox factors. sas7bdat 1 TTr)1-. j • i
- F - - i URE, and carcinogen class
L
census tract data
(tractdat merged. sasTbdat) s~*^
1 f^— *(1)
i
Sort and retain FIPS, state name, county
counties
^ ^ ^ r
^->L, Merge by FIPS j ^ scratch_pop
and tract ID !
j
Irisk SAROAD L-
1 total risk U
Calculate
total dist ^ distribution ot -^ —
total risk (all
sources)
1 distributions 1 V_/
Output to »
' permanent ^ combined dist
dataset f7s\
©
i i
AD, i codes to a list of macro
j variables
,©
I Merge with URE data by ! , .^
i SAROAD |^~ scratch ^~
i •
©Calculate tract level risk for
each source category by ^
multiplying the source ' 1
category concentration by the
^ URE
®i r
Output to permanent ^ , . ,
, ^ hap risk
( l4 | permanent dataset j Cj:/
•T Calculate total
^ risk across all
riskjotal HAPs for each <-
source category
r.a1p.ii1ate national ^
^ . risk (all sources) for r
v — '
1 > null
i 1
Read in tract level V_x
HAPEM5 output
^for individual
HAP
L ±
r~
Repeat steps 4
^>j through 9 in
^ MACRO calc_risk
^1 Append to \^}
i master dataset
' T
1
| &
Calculate
carcinogen class
risk for each
source category
and tract
1
risk class ^— '
1
-T
Output to permanent s^\
datasets v_7
^ + + +
1 risk a II risk b 1 II risk b2 II risk c
Figure D-l. Flowchart of the cancer risk calculations in cancer, sas.
-------
D.2 Non-cancer Risk Calculation methodology
The following steps were used to calculate hazard quotients and hazard indices and summary
statistics for 1999, 2015, 2020, and 2030 in the program noncancer.sas (found in the MSAT rule
docket EPA-HQ-OAR-2005-0036):
1. Read in and sorted a SAS® dataset of census tracts retain FIPS, state name, county name,
tract ID, and tract population.
2. Read in the Rfc and target organ system(s) for each MSAT HAP from the SAS® dataset
msat_haps_tox_factors.sas7bdat, keeping only HAPs where the Rfc was non-zero and
non-missing.
3. From the output of step 1, created a list of HAPs by SAROAD to be read in by a SAS®
macro for further processing, beginning with step 4.
4. The tract level HAPEM5 output was read into a dataset.
5. Merged the output from step 4 with the Rfc data from step 2 using PROC SQL.
6. Merged the tract population data from step 1 with the output from step 5.
7. For each source category, multiplied the tract level source category concentrations by
0.001 and then divided by the Rfc to calculate the HAP's hazard quotient (HQ).
8. Output the tract level HQ estimates to a permanent dataset for the HAP.
9. Appended the HAP HQ estimates to a master dataset.
Steps 4 through 9 were executed in the SAS® MACRO calc_hq.
10. Repeated steps 4 through 9 for each HAP in the MACRO calc_hq.
11. After performing steps 4 through 9 for each HAP, the master dataset was separated into
multiple datasets, one for each target organ system. If a HAP affected more than one
organ system, such as hexane, its HQ estimates would go to both the datasets for
respiratory and neurological organ systems.
12. Sorted each organ system dataset by FlPS/tract and calculated a hazard index (HI) for the
organ system by summing the individual HQ estimates at the FlPS/tract level. This was
done for each source category (major, area, onroad gasoline, etc.).
13. Output each organ system's HI tract level estimates to a permanent dataset.
D-4
-------
14. Calculated a national distribution of HI estimates for each source category for the organ
systems across all census tracts.
15. Output distribution to dataset named for organ system.
16. Repeat steps 12 through 15 in the MACRO stats.
17. Once the HI and distributions had been calculated for each organ system, concatenated all
the datasets into one dataset.
18. Sorted step 17 output by organ system and output to a permanent dataset.
All 18 steps were done for each modeling year and 1999 in the SAS® macro main with
the macro's argument as the year and the processing is shown in Figure D-2.
D-5
-------
(T) (7)
Mi
Retain only HAPs with nonzero or nonmissing Rfc. 1 ^1 1 ^. convert the SAROAD codes to a j
Keep SAROAD, Rfc, and target organ systems | aPs | j ]jst of macro variables I 1
census tract data - — ^
(tractdat merged.sasTbdat) ( 5 )
i
Sort and retain FIPS,
tract ID, and populatic
W
scratch 1 ^ Men>e with Rfc data bvSAROAD !*
tate name, county name, ^
i ^ counties ,^~N
! 1 — ^.- — (.0 k. M
; * +
^ i Merge by FIPS and :
1 i source category by multiplying the
source category concentration by
1 ^j 0.001 and dividing by Rfc
j Concatenate j ( IJJ /" \
^ 1 hq_SAROAn sas7bHat L
hq_dist_development, 1 hqjill systems Sas7bdat 1
hq Hist liver ^^^^^^^^ ^^^^^J
^ hq dist immune.
hq_dist respiratory,
hq_dist kidney,
hq_dist immune ^-i III! 1
1
hu disL neui olouiual ^- i :
1 : R pppat steps 1 2 throueh 1 5 j
hq_dist respiratory < i for each °^an system i
. f "\ i Calculate national
hq_di,L_i^iuduai^ 4 (^) fntal ^ |^_j distribution of total HI
11-11 ^ ^T^ i ' i (all sources).
hq_dist_development ^- I 15} X L_. '.
hq_dist ocular ^- Output to dataset named /""^N
1 ho dist liver *J T T T T
null
©T
Read in tract level
i 1^ HAPEM5 output for
1 | scratch |^ ' individual HAP
1 i
3
( lo) Repeat steps 4 through 9 in j
^-S MACRO calc risk !
.^^::::£:
hap hq ^1 Append to master
1 [ 1
i
Output to permanent dataset j
• | ? —
Break up data into all haps hq
for each organ "T^""^
system \^J
I
y * *
respiratory immune neurological
i i i
t
1 1 ! Calculate hazard index for
^_| sums |^ — j each source category at tract
T ! level
to permanent datasets | ^1 hq_immune
1
t t t
1 hq_liver 1 hq_kidney 1 hq_ocular 1 hq_development 1 hq_reproductive 1 hq_respiratory 1 hq_neurologica
,
1
Figure D-2. Flowchart of the HQ and HI calculations in noncancer.sas.
D-6
-------
Appendix E: Control of stationary refueling and gasoline marketing emissions
Steps used in project_stationary_benz.sas (found in the MS AT rule docket EPA-HQ-OAR-2005-
0036) to develop the controlled gasoline inventories for benzene are listed below with example
calculations for 2015 for Imperial County, CA (FIPS=06025) and Denver County, CO
(FIPS=08031).
1. Read the comma delimited county level refueling emissions for benzene and VOC and
convert the integer state and county FIPS codes to character and combining into one code
for the state/county. Retain records for benzene only.
Perform step 1 for 2015 control emissions, 2015 base emissions, 2020 control emissions,
and 2020 base emissions, resulting in step 1 being executed four times.
2. Once the 2015 and 2020 control and base cases have been read into SAS®, merged the
emissions by FIPS/CAS so that the control and base for both years are in one dataset.
Table E-l. Partial listing of 2015 and 2020 base and controlled refueling emissions after
merger (Steps 1 and 2).
3.
FIPS
06025
08031
CAS
71432
71432
refuel 15 base
0.123459112
1.0662192751
refuel 15 control
0.0740812439
0.6716807497
refuel 20 base
0.1337206137
1.1037468356
refuel 20 control
0.0802398396
0.695301335
Calculated 2015 projection factor by dividing the 2015 control refueling emissions by the
2015 base refueling emissions and calculated 2020 projection factor by dividing the 2020
control refueling emissions by the 2020 base refueling emissions.
Table E-2. Partial listing of refueling projection factors after dividing control emissions
by base emissions.
FIPS
06025
08031
CAS
71432
71432
pf!5 refuel
0.6000467907
0.6299649288
pf20 refuel
0.6000558732
0.6299463904
Read in a text file containing the FIPS codes for the 3,141 counties in the U.S. with their
RFG status. Create a flag denoting the county as CG or RFG, rfg_status. If a county is
an RFG county, rfg_status='RFG, otherwise rfg_status='CG.'
Table E-3. Rfg status of Imperial and Denver counties. RFG=reformulated gasoline,
CG=conventional gasoline.
FIPS
06025
08031
rfg^status
RFG
CG
E-l
-------
5. Sorted a SAS dataset of all 66,300 tracts by FIPS, eliminating double values of FIPS
and Puerto Rico and the Virgin Islands. Retain the FIPS and 2 letter state abbreviation.
6. Assign the PADD region to the counties based on 2-letter state abbreviation.
Table E-4. Partial listing of counties after assign PADD region (Steps 5 and 6).
Region
CA
PADD4
FIPS
06025
08031
state
CA
CO
7. Sorted the output of step 4, the rfg status data, by FIPS.
Table E-5. Rfg status of Imperial and Denver counties after sorting.
FIPS
06025
08031
rfg^status
RFG
CG
Merged the rfg status data (step 4 output) with the PADD/county data (output of step 5)
and the refueling projection factors (step 3 output) by FIPS, retaining matching
observations.
Table E-6. Partial listing of counties after merging counties with rfg status and refueling
projection factors.
Region
CA
PADD4
FIPS
06025
08031
state
CA
CO
rfg^status
RFG
CG
CAS
71432
71432
pf!5 refuel
0.6000467907
0.6299649288
pf20 refuel
0.6000558732
0.6299463904
9. Create a dataset containing PADD region identifiers and emissions to develop projection
factors for the gasoline marketing and distribution emissions (excluding refueling).
10. Calculate the projection factors for the PADD regions output from step 9.
Table E-7. PADD regions and percentages (Steps 9 and 10).
Region
PADD1
PADD2
PADD3
PADD4
PADD5
CA
rfg^status
CG
RFG
CG
RFG
CG
RFG
CG
RFG
CG
RFG
CG
RFG
start
0.91
0.59
1.26
0.80
0.95
0.57
1.47
1.05
1.42
0.65
0.62
0.62
end
0.55
0.54
0.68
0.71
0.54
0.55
0.93
0.62
0.85
0.60
0.61
0.61
Pf
0.6043956044
0.9152542373
0.5396825397
0.8875
0.5684210526
0.9649122807
0.6326530612
0.5904761905
0.5985915493
0.9230769231
0.9838709677
0.9838709677
E-2
-------
11. Merged the output of step 8 with the step 10 output by FIPS using PROC SQL so that
each county was assigned a PADD region and projection factor for gasoline marketing
and distribution.
Table E-8. Refueling projection factors and gasoline marketing projection factors after
merging the county dataset with the PADD regions dataset.
Region
CA
PADD4
FIPS
06025
08031
state
CA
CO
rfg^status
RFC
CG
CAS
71432
71432
pf!5 refuel
0.6000467907
0.6299649288
pf20 refuel
0.6000558732
0.6299463904
Pf
0.9838709677
0.6326530612
12. Read in a text file of SCC codes pertaining to gasoline marketing and distribution.
Created a variable called gas_flag and gave it a value of 1 to help identify these codes in
the inventory later.
Table E-9. Partial listing of gasoline marketing and distribution SCC codes with flag
indicating them as marketing/distribution SCC codes. SCC descriptions added for
reference only.
SCC
2501000000
2501050120
2501060050
2501060051
2501060052
gas flag
1
1
1
1
1
Description
Storage and Transport; Petroleum and Petroleum Product Storage; All
Storage Types: Breathing Loss; Total: All Products
Storage and Transport; Petroleum and Petroleum Product Storage; All
Storage Types: Breathing Loss; Total: All Products
Storage and Transport; Petroleum and Petroleum Product Storage;
Gasoline Service Stations; Stage 1 : Total
Storage and Transport; Petroleum and Petroleum Product Storage;
Gasoline Service Stations; Stage 1 : Submerged Filling
Storage and Transport; Petroleum and Petroleum Product Storage;
Gasoline Service Stations; Stage 1 : Splash Filling
E-2
-------
Steps 13 through 20 are performed in the MACRO point for the following cases: 2015 point inventory, 2015 non-point
airport inventory, 2020 point inventory, and 2020 non-point airport inventory.
13. From the projected point or airport inventory for 2015 or 2020, pull the benzene emissions, based on SAROAD code = 45201,
from the inventory. Calculate a variable, gcemis which is the average of the eight temporally allocated emissions, for QA
purposes.
Table E-10. Partial listing of benzene point source emissions with key variables.
FIPS
06025
06025
06025
08031
08031
08031
site id
06025-13151144
06025-13151177
06025-13151115
08031-1194
08031-1713
08031-13388
emrelpid
111-11-1
3081-308-1
2M-3-2
001-001-01
001-001-03
69583-55912-
69185
src type
AREA
AREA
AREA
AREA
AREA
AREA
sec
40688801
40600403
10300601
40400401
40600401
10100601
temisl
0.00033
0.01006
0.02940
0.01398
0.07731
0.00253
temis2
0.00033
0.01006
0.02940
0.01398
0.07731
0.00252
temis3
0.00033
0.01006
0.02940
0.01398
0.07731
0.00254
temis4
0.00033
0.01006
0.02940
0.01398
0.07731
0.00255
temisS
0.00033
0.01006
0.02940
0.01398
0.07731
0.00267
temis6
0.00033
0.01006
0.02940
0.01398
0.07731
0.00254
temis7
0.00033
0.01006
0.02940
0.01398
0.07731
0.00255
temisS
0.00033
0.01006
0.02940
0.01398
0.07731
0.00253
14. Using PROC SQL, merge the benzene emissions from step 13 with the SCC list created in step 12 by SCC, retaining all
observations and data from the benzene inventory and the gas_flag variable. Emissions that are not gasoline
marketing/distribution will have a missing value for the gas_flag and emissions that are gasoline marketing/distribution will
have a value of 1 for the gas_flag.
Table E-ll. Partial listing of benzene
)oint sources after merging with gasoline marketing/distribution SCC list.
FIPS
06025
06025
06025
08031
08031
08031
site id
06025-
13151144
06025-
13151177
06025-
13151115
08031-1194
08031-1713
08031-13388
emrelpid
111-11-1
3081-308-1
2M-3-2
001-001-01
001-001-03
69583-
55912-
69185
src type
AREA
AREA
AREA
AREA
AREA
AREA
SCC
40688801
40600403
10300601
40400401
40600401
10100601
temisl
0.00033
0.01006
0.02940
0.01398
0.07731
0.00253
temis2
0.00033
0.01006
0.02940
0.01398
0.07731
0.00252
temisS
0.00033
0.01006
0.02940
0.01398
0.07731
0.00254
temis4
0.00033
0.01006
0.02940
0.01398
0.07731
0.00255
temisS
0.00033
0.01006
0.02940
0.01398
0.07731
0.00267
temis6
0.00033
0.01006
0.02940
0.01398
0.07731
0.00254
temis7
0.00033
0.01006
0.02940
0.01398
0.07731
0.00255
temisS
0.00033
0.01006
0.02940
0.01398
0.07731
0.00253
gas flag
1
1
E-4
-------
15. Split the benzene inventory into two datasets: gasoline and others. Output observations to the gasoline dataset if they have a
value of 1 for the gas_flag OR they are a vehicle refueling SCC (shown in Table 23). Otherwise output to the others dataset.
The others dataset contains non-gasoline marketing/distribution or vehicle refueling emissions.
Table E-12. Partial listing of gasoline related emissions after splitting gasoline related emissions and non-gasoline related
emissions.
FIPS
06025
06025
08031
08031
site id
06025-
13151144
06025-
13151177
08031-1194
08031-1713
emrelpid
111-11-1
3081-308-1
001-001-01
001-001-03
src type
AREA
AREA
AREA
AREA
SCC
40688801
40600403
40400401
40600401
temisl
0.00033
0.01006
0.01398
0.07731
temis2
0.00033
0.01006
0.01398
0.07731
temis3
0.00033
0.01006
0.01398
0.07731
temis4
0.00033
0.01006
0.01398
0.07731
temisS
0.00033
0.01006
0.01398
0.07731
temis6
0.00033
0.01006
0.01398
0.07731
temis7
0.00033
0.01006
0.01398
0.07731
temisS
0.00033
0.01006
0.01398
0.07731
gas flag
1
1
Table E-13. Partial listing of non-gasoline related emissions after splitting gasoline related emissions and non-gasoline related
emissions.
FIPS
06025
08031
site id
06025-
13151115
08031-
13388
emrelpid
2M-3-2
69583-
55912-
69185
src type
AREA
AREA
SCC
10300601
10100601
temisl
0.02940
0.00253
temis2
0.02940
0.00252
temisS
0.02940
0.00254
temis4
0.02940
0.00255
temisS
0.02940
0.00267
temis6
0.02940
0.00254
temis7
0.02940
0.00255
temisS
0.02940
0.00253
gas flag
16. Using PROC SQL, merge the gasoline dataset from step 15 with the projection factor data from step 11 by FIPS.
Table E-14. Partial listing of gasoline related emissions with projection factors. Note not all variables shown because of space.
FIPS
06025
06025
08031
08031
site id
06025-
13151144
06025-
13151177
08031-1194
08031-1713
SCC
40688801
40600403
40400401
40600401
temisl
0.00033
0.01006
0.01398
0.07731
temis2
0.00033
0.01006
0.01398
0.07731
temisS
0.00033
0.01006
0.01398
0.07731
temis4
0.00033
0.01006
0.01398
0.07731
temisS
0.00033
0.01006
0.01398
0.07731
temis6
0.00033
0.01006
0.01398
0.07731
temis7
0.00033
0.01006
0.01398
0.07731
temisS
0.00033
0.01006
0.01398
0.07731
gas flag
1
1
Pf
0.98387
0.98387
0.63265
0.63265
pflS refuel
0.60005
0.60005
0.62996
0.62996
E-5
-------
17. Apply the projection factors to each of the eight temporally allocated projected emissions. If a gasoline marketing/distribution
SCC emission (gas_flag=l), apply the projection factor based on the PADD data from step 10. Otherwise, multiply each of the
temporally allocated projected emissions by the appropriate year's projection factor for vehicle refueling.
Table E-15. Partial listing of gasoline related emissions after applying appropriate projection factors to emissions. Emissions
with gas flag=l use the pf number while the others use the pf 15 refuel number.
FIPS
06025
06025
08031
08031
site id
06025-
13151144
06025-
13151177
08031-1194
08031-1713
SCC
40688801
40600403
40400401
40600401
temisl
0.00033
0.00604
0.00885
0.04870
temis2
0.00033
0.00604
0.00885
0.04870
temis3
0.00033
0.00604
0.00885
0.04870
temis4
0.00033
0.00604
0.00885
0.04870
temisS
0.00033
0.00604
0.00885
0.04870
temis6
0.00033
0.00604
0.00885
0.04870
temis7
0.00033
0.00604
0.00885
0.04870
temisS
0.00033
0.00604
0.00885
0.04870
gas flag
1
1
Pf
0.98387
0.98387
0.63265
0.63265
pf!5 refuel
0.60005
0.60005
0.62996
0.62996
18. Concatenate the output of step 17 with the others dataset created in step 15. Note the projected and controlled emissions from
step 17 have the same variable name as the projected only emissions from the dataset others. This is to keep consistency for
PtFinal_ASPEN.
19. Sort the concatenated data from step 18 by FIPS site id emrelpid SAROAD MACT SIC and SCC and output to a permanent
dataset.
Table E-16. Partial listing of point emissions after concatenating controlled emissions with the non-gasoline emissions,
sorting, and output to a permanent dataset (Steps 18 and 19).
FIPS
06025
06025
06025
08031
08031
08031
site id
06025-
13151115
06025-
13151144
06025-
13151177
08031-1194
08031-13388
08031-1713
emrelpid
2M-3-2
111-11-1
3081-308-1
001-001-01
69583-
55912-
69185
001-001-03
src type
AREA
AREA
AREA
AREA
AREA
AREA
SCC
10300601
40688801
40600403
40400401
10100601
40600401
temisl
0.02940
0.00033
0.00604
0.00885
0.00253
0.04870
temis2
0.02940
0.00033
0.00604
0.00885
0.00252
0.04870
temisS
0.02940
0.00033
0.00604
0.00885
0.00254
0.04870
temis4
0.02940
0.00033
0.00604
0.00885
0.00255
0.04870
temisS
0.02940
0.00033
0.00604
0.00885
0.00267
0.04870
temis6
0.02940
0.00033
0.00604
0.00885
0.00254
0.04870
temis7
0.02940
0.00033
0.00604
0.00885
0.00255
0.04870
temisS
0.02940
0.00033
0.00604
0.00885
0.00253
0.04870
E-f
-------
20. Sort the 1999 non-point COP AX output for non-airport emissions by SCC to get the
surrogate codes used in EMS-HAP. Sort only where the CAS = 71432 (benzene). This
data is needed to merge with the controlled non-point emissions for EMS-HAP input.
Table E-17. Partial listing of non-point SCC codes and surrogate codes. SCC and
surrogate descriptions added for informational purposes only.
SCC
2102004000
2501050120
2501060102
spatsurr
505
650
600
SCC description
Stationary Source Fuel Combustion; Industrial; Distillate Oil;
Total: Boilers and 1C Engines
Storage and Transport; Petroleum and Petroleum Product
Storage; Bulk Stations/Terminals: Breathing Loss; Gasoline
Storage and Transport; Petroleum and Petroleum Product
Storage; Gasoline Service Stations; Stage 2: Displacement
Loss/Controlled
Surrogate
description
Industrial land
Refineries and
tank farms
Gas stations
Steps 21-28 are performed in the MACRO nonpoint for the following cases: 2015 non-
point (excluding airport emissions) and 2020 non-point (excluding airport emissions).
21. Extract the benzene emissions from the projected non-point inventory and assign the
benzene CAS number to the observations.
Table E-18. Partial listing of benzene non-point emissions with key variables.
FIPS
06025
06025
06025
CAS
71432
71432
71432
SCC
2102004000
2501050120
2501060102
emisgc
1.3460706
0.08181288
0.067603788
22. Using PROC SQL, merge the benzene emissions from step 22 with the gasoline
marketing/distribution SCC list from step 12, retaining all observations and data from the
benzene inventory and the gas_flag variable. Emissions that are not gasoline
marketing/distribution will have a missing value for the gas_flag and emissions that are
gasoline marketing/distribution will have a value of 1 for the gas_flag.
Table E-19. Partial listing of benzene emissions after merging with the gasoline
marketing/distribution SCC list.
FIPS
06025
06025
06025
CAS
71432
71432
71432
SCC
2102004000
2501050120
2501060102
emisgc
1.3460706
0.08181288
0.067603788
gas flag
1
23. Split the benzene inventory into two datasets: gasoline and others. Output observations
to the gasoline dataset if they have a value of 1 for the gas_flag OR they are a vehicle
refueling SCC (shown in Table X). Otherwise output to the others dataset. The others
dataset contains non-gasoline marketing/distribution or vehicle refueling emissions.
E-7
-------
Table E-20. Partial listing of benzene gasoline related emissions after splitting gasoline
and non-gasoline emissions.
FIPS
06025
06025
CAS
71432
71432
sec
2501050120
2501060102
emisgc
0.08181288
0.067603788
gas flag
1
Table E-21. Partial listing of benzene non-gasoline related emissions after splitting
gasoline and non-gasoline emissions.
FIPS
06025
CAS
71432
sec
2102004000
emisgc
1.3460706
gas flag
24. Using PROC SQL, merge the gasoline dataset from step 24 with the projection factor
data from step 11 by FIPS.
Table E-22. Partial listing of benzene gasoline emissions after merging with projection
factors.
FIPS
06025
06025
CAS
71432
71432
sec
2501050120
2501060102
emisgc
0.08181288
0.067603788
gas flag
1
pf
0.98387096777
0.98387096777
pf!5 refuel
0.6000467907
0.6000467907
25. Apply the projection factors to the annual projected emissions. If a gasoline
marketing/distribution SCC emission (gas_flag=l), apply the projection factor based on
the PADD data from step 10. Otherwise, multiply the annual projected emissions by the
appropriate year's projection factor for vehicle refueling.
Table E-23. Partial listing of benzene gasoline related emissions after applying
appropriate projection factors. If gas_flag =1 apply pf, otherwise apply pf!5_refuel to
emissions.
FIPS
06025
06025
CAS
71432
71432
SCC
2501050120
2501060102
emisgc
0.0804933174
0.040565436
gas flag
1
Pf
0.98387096777
0.98387096777
pf!5 refuel
0.6000467907
0.6000467907
26. Concatenate the output of step 26 with the others dataset created in step 24.
Table E-24. Partial listing of benzene emissions after concatenating controlled emissions
with non-gasoline emissions.
FIPS
06025
06025
06025
CAS
71432
71432
71432
SCC
2501050120
2501060102
2102004000
emisgc
0.0804933174
0.040565436
1.3460706
27. Using PROC SQL, merge the emissions with the SCC/surrogate cross reference created
in step 21 by SCC. This is to assign surrogate codes to the emissions, which is needed for
CountyProc for non-point sources.
E-S
-------
Table E-25. Partial listing of benzene emissions after merging the emissions with the
spatial surrogate codes.
FIPS
06025
06025
06025
CAS
71432
71432
71432
sec
2501050120
2501060102
2102004000
emisgc
0.0804933174
0.040565436
1.3460706
spatsurr
650
600
505
28. Sort step 27 output by FIPS MACT SIC SCC and CAS and output to permanent dataset.
Table E-26. Partial listing of benzene emissions after sorting by FIPS/SCC/CAS and
renaming the emissions variable emisgc to emis and output to permanent dataset.
FIPS
06025
06025
06025
CAS
71432
71432
71432
SCC
2102004000
2501050120
2501060102
emis
1.3460706
0.0804933174
0.040565436
spatsurr
505
650
600
Figure E-l shows the first 19 steps of the program and Figure E-2 shows the non-point steps.
MSATBenzR2015 control, csv
MSATBenzR2015 base.csv
Read refueling comma
delimited files and retain
j only benenze. Repeat
Othis step for each of the
files
T
MSATBenzR2020 control, csv
MSATBenzR2020 base.csv
[*
_counties.txt I I tractdat_merged.sas7bdat
1
4j| read in RFG status of |
! counties j
refuel 15 base
refuel 15 control
refuel 20 base
refuel 20 control
1©
sort by FIPS,
eliminating duplicate
observations
Merge by FIPS/CAS j M merged_refuel ^ calculate projection j—M merged_refuel
^ ' ^^ ! factors !
91 I •> \ '-
sort by FIPS
rfg counties [^i Merge by FIPS
j Sort and output to permanent I i-M ap_pt 2015.sas7bdat I
I dataset 1 I ^^^«^^«i^^«^^J
I ap_np_gair20.sas7bdat
ap_pt_2020.sas7bdat I
Figure E-l. Steps in point gasoline inventory control program.
E-9
-------
| arnonptjroj np20.sas7bdat |— , ^ 1 n°npt99_aspen_ap.sas7bdat |
; Extract benzene, i /T^s
observations
_
i , , . ! " —
1 1 | Ivleige by | f^
1
nonpointl ^.^
I ®
i :
1 1
VlX & J j~ & 0 ^.—^ i n j 1
(21 1
1 r T i 1
gasoline factors ! Concatenate , > all , ,
ate
cc
utto
taset
\ r
1 1 nonpoint 2015.sas7bdat 1 1 nonpoint 2020.sas7bdat 1
r 1
_ | Apply control Ea=olinc fac
(,25) i factors t" gaoOlmc_Iac
emissions
Figure E-2. Non-point steps of the stationary gasoline controls program.
E-10
-------
Appendix F: Control of onroad gasoline emissions
Following are the steps and examples for Alpine County, CA (FIPS=06003) for benzene for the
year 2015 in applying controls to the onroad gasoline emissions. Example calculations are also
shown below. Controls are done in the SAS® program control_onroad.sas (found in the MS AT
rule docket EPA-HQ-OAR-2005-0036). All steps take place within the SAS® MACRO control
where the argument for control is the four digit year. Steps 1 and 2 are performed in the SAS®
MACRO control where the argument for control is the four digit year
1. Read in the projected MSAT emissions (output from onroad.sas in Section 3.3.2) and
subset the emissions to the five HAPs.
Table F-l. Partial listing of 2015 projected benzene emissions for Alpine County. Note
that SCC descriptions are listed here for information purposes only. The variables emis,
ratio, and emis_nei are the 2015 projected emissions, the projection factor for 2015, and
the 1999 NEI emissions respectively.
FIPS
06003
06003
06003
06003
06003
06003
06003
SCC
2201001130
2201001150
2201001170
2201080130
2201080150
2230001130
2230001150
SCC description
Mobile Sources, Highway
Vehicles - Gasoline, Light
Duty Gasoline Vehicles
(LDGV), Rural Other
Principal Arterial: Total
Mobile Sources, Highway
Vehicles - Gasoline, Light
Duty Gasoline Vehicles
(LDGV), Rural Minor
Arterial: Total
Mobile Sources, Highway
Vehicles - Gasoline, Light
Duty Gasoline Vehicles
(LDGV), Rural Major
Collector: Total
Mobile Sources, Highway
Vehicles - Gasoline,
Motorcycles (MC),Rural
Other Principal Arterial: Total
Mobile Sources, Highway
Vehicles - Gasoline,
Motorcycles (MC),Rural
Minor Arterial: Total
Mobile Sources, Highway
Vehicles - Diesel, Light Duty
Diesel Vehicles
(LDDV),Rural Other Principal
Arterial: Total
Mobile Sources, Highway
Vehicles - Diesel, Light Duty
Diesel Vehicles (LDDV),
Rural Minor Arterial: Total
CAS
71432
71432
71432
71432
71432
71432
71432
emis
0.0578743127
0.0330683928
0.0323135532
0.0041931539
0.0024060641
0.0011643267
0.0006597704
ratio
0.269621769
0.2681076114
0.2666134753
0.6348454049
0.6348454049
0.377415447
0.3706575353
emis nei
0.21465
0.12334
0.1212
0.006605
0.00379
0.003085
0.00178
F-l
-------
2. Split the output from step 1 into two datasets, gasoline and diesel. Gasoline contained
gasoline emissions and diesel contained diesel emissions. The first four characters of the
SCC code were used to determine if the observation being read was gasoline or diesel. If
the first four characters were 2201 then the observation was gasoline, otherwise it was
diesel.
Table F-2. Partial listing of Alpine County gasoline emissions after splitting the gasoline
and diesel emissions. The variables ratio and emis_nei have been dropped. SCC
aescn
FIPS
06003
06003
06003
06003
06003
)tion nas aiso oeen dropped.
SCC
2201001130
2201001150
2201001170
2201080130
2201080150
CAS
71432
71432
71432
71432
71432
emis
0.0578743127
0.0330683928
0.0323135532
0.0041931539
0.0024060641
Table F-3. Partial listing of Alpine County diesel emissions after splitting the gasoline
and diesel emissions. The variables ratio and emis_nei have been dropped. SCC
description has also been dropped.
FIPS
06003
06003
SCC
2230001130
2230001150
CAS
71432
71432
emis
0.0011643267
0.0006597704
3. Create a dataset of motorcycle SCC codes to be used later.
Table F-4. Dataset of motorcycle SCC codes. SCC description added for information
purposes.
SCC
2201080110
2201080130
2201080150
2201080170
2201080190
2201080210
2201080230
2201080330
Description
Mobile
Total
Sources,
Hi|
jhway
Mobile Sources, Highway
Principal Arterial: Total
Mobile
Arterial
Sources,
: Total
Mobile Sources,
Collector: Total
Mobile Sources,
Collector: Total
Mobile
Mobile
Total
Mobile
Sources,
Sources,
Sources,
Hij
Hij
Hij
Hi|
Hi!
Hi!
jhway
jhway
jhway
jhway
jhway
jhway
Vehicles
Vehicles
Vehicles
Vehicles
Vehicles
Vehicles
Vehicles
Vehicles
- Gasoline,
- Gasoline,
- Gasoline,
- Gasoline,
- Gasoline,
- Gasoline,
- Gasoline,
- Gasoline,
Motorcycles
Motorcycles
Motorcycles
Motorcycles
Motorcycles
Motorcycles
Motorcycles
Motorcycles
(MC),
(MC),
(MC),
(MC),
(MC),
(MC),
(MC),
(MC),
Rural
Rural
Rural
Rural
Rural
Rural
Interstate:
Other
Minor
Major
Minor
Local:
Total
Urban Interstate:
Urban Local
Total
F-2
-------
Steps 4 through 18 are performed in the SAS MACRO read_nmim.
4. Read in the NMIM emissions for MSAT and subset to the five HAPs and gasoline
emissions only. The first four digits of the SCC code were used to determine if the
emissions were gasoline as in step 2 above.
5. Sorted output from step 4 by FIPS, SCC, and CAS.
Table F-5. Partial listing of Alpine County base MSAT NMIM onroad gasoline
emissions for benzene after sorting. (Steps 4 and 5). The variable nmim_msat is the
NMEVI emissions.
FIPS
06003
06003
06003
SCC
2201001130
2201001150
2201001170
CAS
71432
71432
71432
nmim msat
0.0223670533
0.0129730606
0.0128725145
6. Read the NMIM control emissions comma delimited file and subset to the five HAPs and
gasoline emissions. Create FIPS variable from the integer state and county FIPS
variables.
7. Sorted the output from step 6 by FIPS, SCC, and CAS.
Table F-6. Partial listing of the control NMIM output after sorting (Steps 6 and 7). Note
that each SCC is listed twice, one entry for exhaust emissions and the other for
evaporative emissions. Emissions type, exhaust or evaporative, was not retained.
FIPS
06003
06003
06003
06003
06003
06003
06003
06003
06003
06003
SCC
2201001130
2201001130
2201001150
2201001150
2201001170
2201001170
2201080130
2201080130
2201080150
2201080150
CAS
71432
71432
71432
71432
71432
71432
71432
71432
71432
71432
emis
0.019649687
0.0009109781
0.0113615684
0.0005509424
0.0112324624
0.0005728415
0
0
0
0
F-3
-------
Summed the output from step 7 by FIPS, SCC, and CAS. This was done because the
emissions were broken down by evaporative and exhaust types.
Table F-7. Partial listing of the control NMEVI emissions after summing exhaust and
evaporative components.
FIPS
06003
06003
06003
06003
06003
SCC
2201001130
2201001150
2201001170
2201080130
2201080150
CAS
71432
71432
71432
71432
71432
nmim cntrl
0.0205606651
0.0119125108
0.0118053039
0
0
Merged the output of step 8 with the output of step 4 by FIPS, SCC, and CAS. Output
datasets were merged, no_msat, and no_cntrl. The dataset merged contained matching
observations, no_msat contained observations from step 8 output ? are you sure ? I
thought no_msat had no step 8 (controlled NMIM) emissions, but did have the reference
NMEVI emissions? that were not in the original MSAT NMIM emissions, and no_cntrl
contained observations where there were original MSAT emissions but no matching
observations in the step 8 output (this dataset was always empty). The dataset no_msat
contained observations where the control emissions were 0.
Table F-8. Partial listing of merged base MSAT NMEVI data and control NMIM data
with matching observations (merged).
FIPS
06003
06003
06003
SCC
2201001130
2201001150
2201001170
CAS
71432
71432
71432
nmim cntrl
0.0205606651
0.0119125108
0.0118053039
nmim msat
0.0223670533
0.0129730606
0.0128725145
Table F-9. Partial listing of emissions where there were no base MSAT NMIM
observations but there were control NMIM emissions (no_msat). The control emissions
are zero in this dataset.
FIPS
06003
06003
SCC
2201080130
2201080150
CAS
71432
71432
nmim cntrl
0
0
nmim msat
10. Created a dataset from no_msat where the control emissions were nonzero as a second
check. This dataset was always empty.
11. Subset the output from step 8 to the three California counties (Modoc, Sierra, and Alpine)
that did not have NMIM motorcycle emissions. Note, there are observations for these in
the NMIM data for both the MSAT and control runs, but they are zero for all years.
F-4
-------
12. Sorted the output of step 11 by FIPS and CAS.
Table F-9. Partial listing of emissions after subsetting to California and sorting by CAS
(Steps 11 and 12).
FIPS
06003
06003
06003
sec
2201001130
2201001150
2201001170
CAS
71432
71432
71432
nmim cntrl
0.0205606651
0.0119125108
0.0118053039
nmim msat
0.0223670533
0.0129730606
0.0128725145
13. Summed the output of step 12 by FIPS and CAS to get county level HAP emissions.
Table F-10. Summed emissions for Alpine County.
FIPS CAS nmim cntrl nmim msat
06003 71432
0.3708404994 0.403293093
14. Merged the output of step 13 with the motorcycle SCC data from step 13. This would
basically expand the output of step 138 times.
Table F-ll. Partial listing of emissions after merging county emissions with motorcycle
SCC codes.
FIPS
06003
06003
06003
06003
06003
06003
06003
06003
CAS
71432
71432
71432
71432
71432
71432
71432
71432
nmim cntrl
0.3708404994
0.3708404994
0.3708404994
0.3708404994
0.3708404994
0.3708404994
0.3708404994
0.3708404994
nmim msat
0.403293093
0.403293093
0.403293093
0.403293093
0.403293093
0.403293093
0.403293093
0.403293093
SCC
2201080110
2201080130
2201080150
2201080170
2201080190
2201080210
2201080230
2201080330
15. Concatenated the output of step 9 and step 14.
16. Sorted the output of step 15 by FIPS, SCC, and CAS.
Table F-12. Partial listing of concatenated data shown in Table F-8 and F-l 1 after
sorting (Steps 15 and 16).
FIPS
06003
06003
06003
06003
06003
06003
06003
06003
06003
06003
06003
SCC
2201001130
2201001150
2201001170
2201080110
2201080130
2201080150
2201080170
2201080190
2201080210
2201080230
2201080330
CAS
71432
71432
71432
71432
71432
71432
71432
71432
71432
71432
71432
nmim cntrl
0.0205606651
0.0119125108
0.0118053039
0.3708404994
0.3708404994
0.3708404994
0.3708404994
0.3708404994
0.3708404994
0.3708404994
0.3708404994
nmim msat
0.0223670533
0.0129730606
0.0128725145
0.403293093
0.403293093
0.403293093
0.403293093
0.403293093
0.403293093
0.403293093
0.403293093
F-5
-------
17. Created a projection or control factor by dividing the control NMIM emissions by the
MSAT NMIM emissions for each FIPS, SCC, and CAS.
Table F-13. Partial listing of emissions and projection factors.
FIPS
06003
06003
06003
06003
06003
06003
06003
06003
06003
06003
06003
SCC
2201001130
2201001150
2201001170
2201080110
2201080130
2201080150
2201080170
2201080190
2201080210
2201080230
2201080330
CAS
71432
71432
71432
71432
71432
71432
71432
71432
71432
71432
71432
nmim cntrl
0.0205606651
0.0119125108
0.0118053039
0.3708404994
0.3708404994
0.3708404994
0.3708404994
0.3708404994
0.3708404994
0.3708404994
0.3708404994
nmim msat
0.0223670533
0.0129730606
0.0128725145
0.403293093
0.403293093
0.403293093
0.403293093
0.403293093
0.403293093
0.403293093
0.403293093
Pf
0.9192388834
0.9182498384
0.9170938514
0.9195309959
0.9195309959
0.9195309959
0.9195309959
0.9195309959
0.9195309959
0.9195309959
0.9195309959
18. Created a dataset of the first five observations of step 17 output for QA purposes to
visually check calculations.
19. Merged the gasoline dataset from step 2 with the output of step 17 by FIPS, SCC, and
CAS. Two datasets were created, merged_gas, containing matching observations and
no_nmim, which contained observations from gasoline but not in step 17 output. This
dataset was empty.
Table F-14. Partial listing of emissions after merging the projected emissions shown in
Table F-l with the factors shown in Table F-13.
FIPS
06003
06003
06003
06003
06003
SCC
2201001130
2201001150
2201001170
2201080130
2201080150
CAS
71432
71432
71432
71432
71432
emis
0.0578743127
0.0330683928
0.0323135532
0.0041931539
0.0024060641
nmim cntrl
0.0205606651
0.0119125108
0.0118053039
0.3708404994
0.3708404994
nmim msat
0.0223670533
0.0129730606
0.0128725145
0.403293093
0.403293093
Pf
0.9192388834
0.9182498384
0.9170938514
0.9195309959
0.9195309959
20. Applied the factors calculated in step 17 to the projected MSAT emissions. Created a
variable emis_msat, which was the original MSAT projected emissions.
Table F-15. Partial listing of emissions after applying factor to projected 2015
emissions. Note nmim cntrl and nmim msat not shown but are in dataset.
FIPS
06003
06003
06003
06003
06003
SCC
2201001130
2201001150
2201001170
2201080130
2201080150
CAS
71432
71432
71432
71432
71432
emis
0.0532003187
0.0303650464
0.0296345609
0.003855735
0.0022124505
Pf
0.9192388834
0.9182498384
0.9170938514
0.9195309959
0.9195309959
emis msat
0.0578743127
0.0330683928
0.0323135532
0.0041931539
0.0024060641
F-6
-------
21. Concatenated step 20 output with the diesel data from step 2.
22. For the diesel emissions, set the variable emis_msat equal to the diesel emissions for QA
purposes.
23. Sorted the step 22 output by FIPS, SCC, and CAS to a permanent dataset with the year in
the dataset name.
Table F-16. Partial listing of emissions after concatenating controlled emissions with
diesel emissions and set the variable emis_msat for diesel equal to the emis variable for
diesel, sorting and creating a permanent dataset (Steps 21 through 23).
FIPS
06003
06003
06003
06003
06003
06003
06003
SCC
2201001130
2201001150
2201001170
2201080130
2201080150
2230001130
2230001150
CAS
71432
71432
71432
71432
71432
71432
71432
emis
0.0532003187
0.0303650464
0.0296345609
0.003855735
0.0022124505
0.0011643267
0.0006597704
pf
0.9192388834
0.9182498384
0.9170938514
0.9195309959
0.9195309959
emis msat
0.0578743127
0.0330683928
0.0323135532
0.0041931539
0.0024060641
0.0011643267
0.0006597704
24. Created a dataset of the first five observations of step 23 output to visually check
calculations.
Figure F-l shows the steps of the program.
F-7
-------
onroad_20XX.sas7bdat where
XX is 15, 20, or 30.
I Create a dataset of
j motorcycle SCC codes
ratios_on.sas7bdat where XX
is 15, 20, or 30.
i
I Split into gasoline &
diesel emissions
Output observations
where control
emissions are
nonzero
Subset base NMIM
emissions to the 5 HAPs
and gasoline emissions
MSATBenzO20XX.csv where
XX is 15, 20, or 30.
Read in the control
NMIM emissions and
subset to 5 HAPs and
gasoline emissions.
Figure F-l. Steps in the onroad gasoline controls program.
F-8
-------
Appendix G: Development of controlled nonroad inventory
G.I. Calculation of exhaust and evaporative fractions
Following are the steps taken in the SAS® program calc_factors.sas (found in the MSAT rule
docket EPA-HQ-OAR-2005-0036) to develop the exhaust and evaporative fractions for use in
developing the controlled nonroad inventory. Example calculations for Autauga County, AL are
shown.
1. Read in the base onroad NMIM emissions comma delimited file and subset emissions to
the five HAPs (by CAS) and LDGV emissions (first seven characters of SCC code =
2201001). Emissions were by emtype (exhaust or evaporative).
2. Sorted by FIPS/CAS/emtype.
Table G-l. Partial listing of NMEVI base LDGV emissions after sorting by FIPS, CAS,
and emtype. (Steps 1 and 2). Emtype is exh for exhaust and eva is for evaporative.
3.
FIPS
01001
01001
01001
01001
01001
01001
01001
01001
01001
01001
01001
01001
01001
01001
01001
01001
01001
01001
SCC
2201001110
2201001130
2201001150
2201001110
2201001130
2201001150
2201001110
2201001130
2201001150
2201001110
2201001130
2201001150
2201001110
2201001130
2201001150
2201001110
2201001130
2201001150
CAS
106990
106990
106990
107028
107028
107028
50000
50000
50000
71432
71432
71432
71432
71432
71432
75070
75070
75070
emtype
Exh
Exh
Exh
Exh
Exh
Exh
Exh
Exh
Exh
Eva
Eva
Eva
Exh
Exh
Exh
Exh
Exh
Exh
em is
0.1329114838
0.0224262833
0.0274003661
0.0149620306
0.0025289852
0.0030914513
0.2919313181
0.0491281133
0.0599782676
0.1188655905
0.0226015763
0.0290840303
1.2735248953
0.2144948896
0.2619329412
0.1059866981
0.0178493559
0.0217964613
Summarized emissions by FIPS/CAS/type to give county level LDGV emissions for each
CAS.
G-l
-------
Table G-2. Total base LDGV emissions by CAS and emtype for Autauga County. Note
that total emissions include emissions not shown in Table G-l.
FIPS
01001
01001
01001
01001
01001
01001
CAS
106990
107028
50000
71432
71432
75070
emtype
Exh
Exh
Exh
Eva
Exh
Exh
emis
0.3896782826
0.0442493623
0.8536461103
0.4343823954
3.7298372318
0.3102395715
4. Transposed the output of step 3 by FIPS/CAS so that the exhaust and evaporative
emissions were on the same row.
5. Created a dataset called ldgv_base and renamed the exhaust and evaporative emissions to
exhaust_base and ldgv_evap_base respectively.
Table G-3. Autauga County reference LDGV emissions after transposing the data by
FIPS and CAS and renaming the exhaust and evaporative emissions (Steps 4 and 5).
FIPS
01001
01001
01001
01001
01001
CAS
106990
107028
50000
71432
75070
exhaust base
0.3896782826
0.0442493623
0.8536461103
3.7298372318
0.3102395715
Idgv evap base
0.4343823954
Repeated steps 1 through 5 for the control case with output file called ldgv_control and
emissions called exhaust_control and ldgv_evap_control.
Table G-4. Partial listing of NMIM control LDGV emissions after sorting by FIPS,
CAS, and emtype. (Steps 1 and 2). Emtype is exh for exhaust and eva is for evaporative.
FIPS
01001
01001
01001
01001
01001
01001
01001
01001
01001
01001
01001
01001
01001
01001
01001
01001
01001
01001
sec
2201001110
2201001130
2201001150
2201001110
2201001130
2201001150
2201001110
2201001130
2201001150
2201001110
2201001130
2201001150
2201001110
2201001130
2201001150
2201001110
2201001130
2201001150
CAS
106990
106990
106990
107028
107028
107028
50000
50000
50000
71432
71432
71432
71432
71432
71432
75070
75070
75070
emtype
Exh
Exh
Exh
Exh
Exh
Exh
Exh
Exh
Exh
Eva
Eva
Eva
Exh
Exh
Exh
Exh
Exh
Exh
emis
0.1332807094
0.0224884795
0.0274765282
0.0149620306
0.0025289852
0.0030914513
0.2930203266
0.0493116695
0.0602024053
0.071318306
0.0135608203
0.0174501906
1.1610455513
0.1955535179
0.2388030291
0.1063300385
0.0179073763
0.0218667063
G-2
-------
Table G-5. Total control LDGV emissions by CAS and emtype for Autauga County.
Note that total emissions include emissions not shown in Table G-4.
FIPS
01001
01001
01001
01001
01001
01001
CAS
106990
107028
50000
71432
71432
75070
emtype
Exh
Exh
Exh
Eva
Exh
Exh
em is
0.390761545
0.0442493623
0.8568307054
0.2606270021
3.4006572287
0.3112453358
Table G-6. Autauga County control LDGV emissions after transposing the data by FIPS
and CAS and renaming the exhaust and evaporative emissions (Steps 4 and 5).
FIPS
01001
01001
01001
01001
01001
CAS
106990
107028
50000
71432
75070
exhaust control
0.390761545
0.0442493623
0.8568307054
3.4006572287
0.3112453358
Idgv evap control
0.2606270021
® -
Steps 1 through 5 are performed in the SAS MACRO read_ldgv.
7.
Read the county level refueling emissions for benzene and VOC from comma delimited
file and output to dataset called refuel_base with emissions called refuel_base. Create 5
character FIPS variable and retain only benzene emissions.
Table G-7. Autauga County base refueling emissions for 2015.
FIPS
01001
CAS
71432
refuel base
0.1618722891
Repeated step 7 for the control case, with output file called refuel_control and emissions
called refuel_control.
Table G-8. Autauga County control refueling emissions for 2015.
FIPS
01001
CAS
71432
refuel control
0.0971250232
Step 7 is performed in the SAS MACRO refuel.
9. Merged the ldgv_base and ldgv_control datasets by FIPS/CAS.
G-3
-------
Table G-9. Autauga County LDGV emissions after merging base and control emissions
by FIPS and CAS.
FIPS
01001
01001
01001
01001
01001
CAS
106990
107028
50000
71432
75070
exhaust base
0.3896782826
0.0442493623
0.8536461103
3.7298372318
0.3102395715
Idgv evap base
0.4343823954
exhaust control
0.390761545
0.0442493623
0.8568307054
3.4006572287
0.3112453358
Idgv evap control
0.2606270021
10. Merged refuel_base and refuel_control by FIPS/CAS.
Table G-10. Autauga County refueling emissions after merging base and control
emissions by FIPS and CAS.
FIPS
01001
CAS
71432
refuel control
0.0971250232
refuel base
0.1618722891
G-4
-------
11. Merged output of step 10 and step 11 by FIPS/CAS using PROC SQL retaining all observations from step 10 output.
Table G-ll. Autauga County LDGV and refueling emissions after merging the two datasets together by FIPS and CAS,
retaining all LDGV emissions.
FIPS
01001
01001
01001
01001
01001
CAS
106990
107028
50000
71432
75070
exhaust base
0.3896782826
0.0442493623
0.8536461103
3.7298372318
0.3102395715
Idgv evap base
0.4343823954
exhaust control
0.390761545
0.0442493623
0.8568307054
3.4006572287
0.3112453358
Idgv evap control
0.2606270021
refuel control
0.0971250232
refuel base
0.1618722891
12. Calculated projection factors for exhaust type for all HAPs by dividing exhaust_control by exhaust base. For benzene
calculated projection factors for evaporation type by dividing the sum of ldgv_evap_control and refuel_control by the sum of
ldgv_evap_base and refuel_base.
Table G-12. Autauga County emissions with projection factors. Projection factors rounded for visual purposes
FIPS
01001
01001
01001
01001
01001
CAS
106990
107028
50000
71432
75070
exhaust base
0.3896782826
0.0442493623
0.8536461103
3.7298372318
0.3102395715
Idgv evap base
0.4343823954
exhaust control
0.390761545
0.0442493623
0.8568307054
3.4006572287
0.3112453358
Idgv evap control
0.2606270021
refuel control
0.0971250232
refuel base
0.1618722891
pf exh
1.00278
1
1.00373
0.91174
1.00324
pf evap
0.59999
13. Sorted the output of step 12 by FIPS and CAS and output to a permanent dataset, retaining the FIPS, CAS, exhaust projection
factor and evaporative projection factor.
Table G-13. Autauga County projection factors after sorting and outputting to a permanent dataset.
FIPS
01001
01001
01001
01001
01001
CAS
106990
107028
50000
71432
75070
pf exh
1.00278
1
1.00373
0.91174
1.00324
pf evap
0.59999
G-5
-------
Figure G-l shows the steps of the projection factor development program.
MSATBenz020XX.csv,
control onroad NMIM
emissions
MSATBenzR2015control.csv.
control refueling emssions
Read in files and subset to LDGV
emissions and the 5 HAPs
MSATBenzR2015base.csv.
base refueling emssions
nonroad factors 20XX.sas7bdat
Repeat steps 1 through 4
for control case
Sort by FIPS/CAS and
output to permanent dataset
Calculate exhaust factors for all HAPs.
Calculate evaporative factors for
benzene by dividing the sum of control
LDGV evap and control refuel
emissions by the sum of base LDGV
evap and base refuel emissions.
Figure G-l. Steps of the nonroad projection factor development program.
G-6
-------
G.2 Development of controlled nonroad inventories
The following describes the steps taken to apply the projection factors to the nonroad controlled
inventory in the program control_nonroad.sas (found in the MS AT rules docket EPA-HQ-OAR-
2005-0036) with examples from California and Texas.
1. Read in the SCC/ASPEN source group cross reference text file and retain only the
nonroad group SCC codes. Corrected the airport support equipment group to nonroad
gasoline and corrected pleasure craft SCC codes to other nonroad.
Table G-14. Partial listing of SCC codes with bins after correcting airport support
equipment and pleasure craft bins.
SCC
2260000000
2265008000
2265008005
2282000000
2282020000
2282020005
2282020010
grp
5
5
5
3
3
3
3
2. Read in the NMIM base nonroad exhaust and evaporative emissions for benzene and
VOC. Retained only benzene.
3. Sorted step 2 output by FIPS/SCC/CAS/emtype where emtype is exh for exhaust and eva
for evaporative emissions.
G-7
-------
Table G-15. Partial listing of NMIM base nonroad exhaust and evaporative emissions
for benzene after sorting by FIPS, SCC, and CAS (Steps 2 and 3).
FIPS
06001
06001
06001
06001
06001
06001
06001
06001
06001
06001
06021
06021
06021
06021
06021
06021
06021
06021
06021
06021
48001
48001
48001
48001
48001
48001
48001
48001
SCC
2260001010
2260001010
2260001030
2260001030
2260002006
2260002006
2265001010
2265001010
2265007015
2265007015
2260001010
2260001010
2260001030
2260001030
2260002006
2260002006
2265001050
2265001050
2265007015
2265007015
2260001010
2260001010
2260001030
2260001030
2260002006
2260002006
2265007015
2265007015
CAS
71432
71432
71432
71432
71432
71432
71432
71432
71432
71432
71432
71432
71432
71432
71432
71432
71432
71432
71432
71432
71432
71432
71432
71432
71432
71432
71432
71432
emtype
Eva
Exh
Eva
Exh
Eva
Exh
Eva
Exh
Eva
Exh
Eva
Exh
Eva
Exh
Eva
Exh
Eva
Exh
Eva
Exh
Eva
Exh
Eva
Exh
Eva
Exh
Eva
Exh
emis
0.027924
2.173649
0.0332
1.412159
0.003535
0.36573
0.0106649406
0.1943388665
0
0
0
0
0
0
2.81E-05
0.002958
0.0012738279
0.0632532043
0
0
0
0
0
0
0.00016
0.006782
2.63E-05
0.000281
Transposed step 3 output by FIPS/SCC/CAS so that the exhaust and evaporative
emissions are now on the same observation or row instead of multiple rows.
G-8
-------
Table G-16. Partial listing of NMIM base emissions after transposing by FIPS, SCC,
and CAS.
FIPS
06001
06001
06001
06001
06001
06021
06021
06021
06021
06021
48001
48001
48001
48001
SCC
2260001010
2260001030
2260002006
2265001010
2265007015
2260001010
2260001030
2260002006
2265001050
2265007015
2260001010
2260001030
2260002006
2265007015
CAS
71432
71432
71432
71432
71432
71432
71432
71432
71432
71432
71432
71432
71432
71432
Eva
0.027924
0.0332
0.003535
0.0106649406
0
0
0
2.81E-05
0.0012738279
0
0
0
0.00016
2.63E-05
Exh
2.173649
1.412159
0.36573
0.1943388665
0
0
0
0.002958
0.0632532043
0
0
0
0.006782
0.000281
5. From step 4 output, created a new value for the SCC codes but changing the last three
characters to 000 to create "Total" level SCC codes. Output to new dataset.
6. Sorted step 5 output by FIPS/SCC/CAS.
Table G-17. Partial listing of NMIM base emissions after changing last three characters
of SCC codes to OOP and sorting by FIPS, SCC, and CAS (Steps 5 and 6).
FIPS
06001
06001
06001
06001
06001
06021
06021
06021
06021
06021
48001
48001
48001
48001
SCC
2260001000
2260001000
2260002000
2265001000
2265007000
2260001000
2260001000
2260002000
2265001000
2265007000
2260001000
2260001000
2260002000
2265007000
CAS
71432
71432
71432
71432
71432
71432
71432
71432
71432
71432
71432
71432
71432
71432
Eva
0.027924
0.0332
0.003535
0.0106649406
0
0
0
2.81E-05
0.0012738279
0
0
0
0.00016
2.63E-05
Exh
2.173649
1.412159
0.36573
0.1943388665
0
0
0
0.002958
0.0632532043
0
0
0
0.006782
0.000281
7. Summed the exhaust and evaporative emissions by FIPS/SCC/CAS of the step 6 output.
This created "Total" level SCC emissions for exhaust and evaporative components.
G-9
-------
Table G-18. Partial listing of NMIM base emissions after summing "Total" level SCC
codes by FIPS and CAS. Note emissions include SCC codes not listed in Table G-17.
FIPS
06001
06001
06001
06001
06021
06021
06021
06021
48001
48001
48001
SCC
2260001000
2260002000
2265001000
2265007000
2260001000
2260002000
2265001000
2265007000
2260001000
2260002000
2265007000
CAS
71432
71432
71432
71432
71432
71432
71432
71432
71432
71432
71432
Exh
3.6161273849
1.3666668567
2.1453478962
0
0
0.0110537204
0.0632532043
0
0
0.0253426696
0.015208714
Eva
0.065301121
0.0094967323
0.1289871176
0
0
0.000076092
0.0012738279
0
0
0.000416299
0.008868335
Created a dataset of snow mobile emissions (SCC 265001020) for California from step 7
output by changing the SCC code 2265001000 to 2265001020. This was done because
counties in California had snow mobile emissions but they were not in the NMIM output.
Table G-19. Partial listing of NMIM base emissions after changing SCC code
2265001000 to 2265001020 for California.
FIPS
06001
06001
SCC
2265001020
2265001020
CAS
71432
71432
Exh
2.1453478962
0.0632532043
Eva
0.1289871176
0.0012738279
9. Created a dataset by concatenating step 4, step 7, and step 8 output.
10. Created macro variables for the exhaust and evaporative emissions for SCC 2265001000
for FIPS 06021.
11. For FIPS 06021 and SCC = 2265001030 set the exhaust and evaporative emissions equal
to the exhaust and evaporative emission macro variables created in step 10.
G-10
-------
Table G-20. Partial listing of NMIM base emissions after concatenating base NMIM
output, total SCC emissions, and snow mobile emissions and after substituting
2265001000 emissions in FIPS 06021 (Steps 9, 10, and 11).
FIPS
06001
06001
06001
06001
06001
06021
06021
06021
06021
48001
48001
48001
48001
06001
06001
06001
06001
06021
06021
06021
06021
06021
48001
48001
48001
06001
06001
SCC
2260001010
2260001030
2260002006
2265001010
2265007015
2260001010
2260002006
2265001050
2265007015
2260001010
2260001030
2260002006
2265007015
2260001000
2260002000
2265001000
2265007000
2260001000
2260002000
2265001000
2265001030
2265007000
2260001000
2260002000
2265007000
2265001020
2265001020
CAS
71432
71432
71432
71432
71432
71432
71432
71432
71432
71432
71432
71432
71432
71432
71432
71432
71432
71432
71432
71432
71432
71432
71432
71432
71432
71432
71432
Eva
0.027924
0.0332
0.003535
0.0106649406
0
0
2.81E-05
0.0012738279
0
0
0
0.00016
2.63E-05
0.065301121
0.0094967323
0.1289871176
0
0
0.000076092
0.0012738279
0.0012738279
0
0
0.000416299
0.008868335
0.1289871176
0.0012738279
Exh
2.173649
1.412159
0.36573
0.1943388665
0
0
0.002958
0.0632532043
0
0
0
0.006782
0.000281
3.6161273849
1.3666668567
2.1453478962
0
0
0.0110537204
0.0632532043
0.0632532043
0
0
0.0253426696
0.015208714
2.1453478962
0.0632532043
12. Created a dataset for California and Texas with the engine designation for 2 and 4 stroke
emissions based on the first six characters of the SCC code. This was only done with
SCC codes where the last three characters were not 000 (total level) to avoid double
counts of emissions.
13. Sorted step 12 output by FIPS/CAS/eng where eng = 2 for 2-stroke gasoline and eng:
for 4-stroke gasoline emissions.
5=4
G-ll
-------
Table G-22. Partial listing of NMIM base emissions after adding engine type variable
and sorting by engine type (Steps 12 and 13).
FIPS
06001
06001
06001
06001
06001
06001
06021
06021
06001
06021
06021
06021
48001
48001
48001
48001
sec
2260001010
2260001030
2260002006
2265001010
2265001020
2265007015
2260001010
2260002006
2265001020
2260001030
2265001050
2265007015
2260001010
2260001030
2260002006
2265007015
CAS
71432
71432
71432
71432
71432
71432
71432
71432
71432
71432
71432
71432
71432
71432
71432
71432
Eva
0.027924
0.0332
0.003535
0.0106649406
0.1289871176
0
0
2.81E-05
0.0012738279
0.0012738279
0.0012738279
0
0
0
0.00016
2.63E-05
Exh
2.173649
1.412159
0.36573
0.1943388665
2.1453478962
0
0
0.002958
0.0632532043
0.0632532043
0.0632532043
0
0
0
0.006782
0.000281
eng
2
2
2
4
4
4
2
2
4
4
4
4
2
2
2
4
14. Summed the exhaust and summed the evaporative emissions for by FIPS/CAS/eng.
Table G-23. Partial listing of emissions after summing by FIPS, CAS, and engine type.
(Section.
FIPS
06001
06001
06021
06021
48001
48001
CAS
71432
71432
71432
71432
71432
71432
eng
2
4
2
4
2
4
exhl
20.824142764
93.924771216
0.0570237128
0.9153963121
0.1681255378
1.4996708652
eval
3.1043091582
6.8976984165
0.0113984838
0.0600273041
0.0664583016
0.2811232142
15. Merged the step 14 output and step 9 output by FIPS and CAS and where the first six
digits of the SCC were 226501 or 226500 and eng=4 or where the first six digits of the
SCC code were 226000 and eng=2. Retain all observations of the step 9 dataset and the
FIPS/CAS/eng emissions from step 14.
G-12
-------
Table G-24. Partial listing of emissions after merging with the engine emissions by
FIPS, CAS, and engine type.
FIPS
06001
06001
06001
06001
06001
06001
06001
06001
06001
06001
06021
06021
06021
06021
06021
06001
06021
06021
06021
06021
48001
48001
48001
48001
48001
48001
48001
sec
2260001000
2260001010
2260001030
2260002000
2260002006
2265001000
2265001010
2265001020
2265007000
2265007015
2260001000
2260001010
2260002000
2260002006
2265001000
2265001020
2265001030
2265001050
2265007000
2265007015
2260001000
2260001010
2260001030
2260002000
2260002006
2265007000
2265007015
CAS
71432
71432
71432
71432
71432
71432
71432
71432
71432
71432
71432
71432
71432
71432
71432
71432
71432
71432
71432
71432
71432
71432
71432
71432
71432
71432
71432
Eva
0.065301121
0.027924
0.0332
0.0094967323
0.003535
0.1289871176
0.0106649406
0.1289871176
0
0
0
0
0.000076092
2.81E-05
0.0012738279
0.0012738279
0.0012738279
0.0012738279
0
0
0
0
0
0.000416299
0.00016
0.008868335
2.63E-05
Exh
3.6161273849
2.173649
1.412159
1.3666668567
0.36573
2.1453478962
0.1943388665
2.1453478962
0
0
0
0
0.0110537204
0.002958
0.0632532043
0.0632532043
0.0632532043
0.0632532043
0
0
0
0
0
0.0253426696
0.006782
0.015208714
0.000281
eng
2
2
2
2
2
4
4
4
4
4
2
2
2
2
4
4
4
4
4
4
2
2
2
2
2
4
4
exhl
20.824142764
20.824142764
20.824142764
20.824142764
20.824142764
93.924771216
93.924771216
93.924771216
93.924771216
93.924771216
0.0570237128
0.0570237128
0.0570237128
0.0570237128
0.9153963121
0.9153963121
0.9153963121
0.9153963121
0.9153963121
0.9153963121
0.1681255378
0.1681255378
0.1681255378
0.1681255378
0.1681255378
1.4996708652
1.4996708652
eval
3.1043091582
3.1043091582
3.1043091582
3.1043091582
3.1043091582
6.8976984165
6.8976984165
6.8976984165
6.8976984165
6.8976984165
0.0113984838
0.0113984838
0.0113984838
0.0113984838
0.0600273041
0.0600273041
0.0600273041
0.0600273041
0.0600273041
0.0600273041
0.0664583016
0.0664583016
0.0664583016
0.0664583016
0.0664583016
0.2811232142
0.2811232142
16. Corrected the exhaust and evaporative emissions for SCC codes where the eng variable
was 2 or 4 and the exhaust and evaporative emissions were 0 and the emissions were
California or Texas. They were replaced with exhaust and evaporative emissions for the
eng type as calculated in step 14. This was done because the MSAT emissions for
several SCC codes were not zero in 1999 but were not available in 2015 NMIM output
for base MSAT. The same was true with the exhaust and evaporative emissions.
Therefore county level engine emissions for exhaust and evaporative emissions were
calculated to be used for the fraction calculations in step 17.
G-13
-------
Table G-25. Emissions after correcting zero exhaust and evaporative emissions
engine type exhaust and evaporative emissions.
with
fag
C
C
C
C
C
C
C
C
C
FIPS
06001
06001
06001
06001
06001
06001
06001
06001
06001
06001
06021
06021
06021
06021
06021
06001
06021
06021
06021
06021
48001
48001
48001
48001
48001
48001
48001
sec
2260001000
2260001010
2260001030
2260002000
2260002006
2265001000
2265001010
2265001020
2265007000
2265007015
2260001000
2260001010
2260002000
2260002006
2265001000
2265001020
2265001030
2265001050
2265007000
2265007015
2260001000
2260001010
2260001030
2260002000
2260002006
2265007000
2265007015
CAS
71432
71432
71432
71432
71432
71432
71432
71432
71432
71432
71432
71432
71432
71432
71432
71432
71432
71432
71432
71432
71432
71432
71432
71432
71432
71432
71432
Eva
0.065301121
0.027924
0.0332
0.0094967323
0.003535
0.1289871176
0.0106649406
0.1289871176
6.8976984165
6.8976984165
0.0113984838
0.0113984838
0.000076092
2.81E-05
0.0012738279
0.0012738279
0.0012738279
0.0012738279
0.0600273041
0.0600273041
0.1681255378
0.1681255378
0.1681255378
0.000416299
0.00016
0.008868335
2.63E-05
Exh
3.6161273849
2.173649
1.412159
1.3666668567
0.36573
2.1453478962
0.1943388665
2.1453478962
93.924771216
93.924771216
0.0570237128
0.0570237128
0.0110537204
0.002958
0.0632532043
0.0632532043
0.0632532043
0.0632532043
0.9153963121
0.9153963121
0.0664583016
0.0664583016
0.0664583016
0.0253426696
0.006782
0.015208714
0.000281
eng
2
2
2
2
2
4
4
4
4
4
2
2
2
2
4
4
4
4
4
4
2
2
2
2
2
4
4
exhl
20.824142764
20.824142764
20.824142764
20.824142764
20.824142764
93.924771216
93.924771216
93.924771216
93.924771216
93.924771216
0.0570237128
0.0570237128
0.0570237128
0.0570237128
0.9153963121
0.9153963121
0.9153963121
0.9153963121
0.9153963121
0.9153963121
0.1681255378
0.1681255378
0.1681255378
0.1681255378
0.1681255378
1.4996708652
1.4996708652
eval
3.1043091582
3.1043091582
3.1043091582
3.1043091582
3.1043091582
6.8976984165
6.8976984165
6.8976984165
6.8976984165
6.8976984165
0.0113984838
0.0113984838
0.0113984838
0.0113984838
0.0600273041
0.0600273041
0.0600273041
0.0600273041
0.0600273041
0.0600273041
0.0664583016
0.0664583016
0.0664583016
0.0664583016
0.0664583016
0.2811232142
0.2811232142
17. Calculated exhaust emission fractions by dividing the exhaust emissions by the sum of
the exhaust and evaporative emissions for each FIPS/SCC/CAS from step 16 output.
Calculate the evaporative emissions fractions by dividing the evaporative emissions by
the sum of the exhaust and evaporative emissions for each FIPS/SCC/CAS. Do this only
where both are not zero. One of them could be zero but not both. If both were zero, then
set the fractions equal to 0.
18. Sort step 17 output by FIPS/SCC/CAS.
G-14
-------
Table G-26. Partial listing of emissions after calculating projection factors and sorting by FIPS,
SCC, and CAS (Steps 17 and 18).
flag
C
C
C
C
C
C
C
C
C
FIPS
06001
06001
06001
06001
06001
06001
06001
06001
06001
06001
06021
06021
06021
06021
06021
06001
06021
06021
06021
06021
48001
48001
48001
48001
48001
48001
48001
SCC
2260001000
2260001010
2260001030
2260002000
2260002006
2265001000
2265001010
2265001020
2265007000
2265007015
2260001000
2260001010
2260002000
2260002006
2265001000
2265001020
2265001030
2265001050
2265007000
2265007015
2260001000
2260001010
2260001030
2260002000
2260002006
2265007000
2265007015
CAS
71432
71432
71432
71432
71432
71432
71432
71432
71432
71432
71432
71432
71432
71432
71432
71432
71432
71432
71432
71432
71432
71432
71432
71432
71432
71432
71432
Eva
0.065301121
0.027924
0.0332
0.0094967323
0.003535
0.1289871176
0.0106649406
0.1289871176
6.8976984165
6.8976984165
0.0113984838
0.0113984838
0.000076092
2.81E-05
0.0012738279
0.0012738279
0.0012738279
0.0012738279
0.0600273041
0.0600273041
0.0664583016
0.0664583016
0.0664583016
0.000416299
0.00016
0.008868335
2.63E-05
Exh
3.6161273849
2.173649
1.412159
1.3666668567
0.36573
2.1453478962
0.1943388665
2.1453478962
93.924771216
93.924771216
0.0570237128
0.0570237128
0.0110537204
0.002958
0.0632532043
0.0632532043
0.0632532043
0.0632532043
0.9153963121
0.9153963121
0.1681255378
0.1681255378
0.1681255378
0.0253426696
0.006782
0.015208714
0.000281
eng
2
2
2
2
2
4
4
4
4
4
2
2
2
2
4
4
4
4
4
4
2
2
2
2
2
4
4
exh frac
0.9822620157107
0.987316341543
0.977029928205
0.9930991254413
0.99042692917
0.9432857882338
0.9479768656452
0.9432857882338
0.931585702655
0.931585702655
0.833409560546
0.833409560546
0.9931632270819
0.990589732427
0.9802590037606
0.9802590037606
0.9802590037606
0.9802590037606
0.9384602719239
0.9384602719239
0.7166970164271
0.7166970164271
0.7166970164271
0.9838386774539
0.976951887064
0.6316685238295
0.914415880247
eva frac
0.01773798428934
0.012683658457
0.0229700718
0.006900874558744
0.00957307083
0.05671421176623
0.05202313435476
0.05671421176623
0.06841429734505
0.06841429734505
0.166590439454
0.166590439454
0.006836772918113
0.00941027
0.0197409962394
0.0197409962394
0.0197409962394
0.0197409962394
0.06153972807615
0.06153972807615
0.2833029835729
0.2833029835729
0.2833029835729
0.01616132254612
0.02304811294
0.3683314761705
0.08558412
Steps 2 through 18 were performed in the SAS MACRO exhaust_evap.
19. Read the projected MSAT emissions (output from Section 3.3.3) and subset to 1,3-
butadiene, acetaldehyde, acrolein, benzene, and formaldehyde based on CAS.
Table G-27. Partial listing of reference MSAT emissions after subsetting to the five
HAPs.
FIPS
06001
06001
06001
06001
06001
06001
06001
06001
06001
06001
06001
06001
SCC
2260001010
2260001010
2260001010
2260001010
2260001010
2260001030
2260001030
2260001030
2260001030
2260001030
2260002000
2260002000
CAS
106990
107028
50000
71432
75070
106990
107028
50000
71432
75070
106990
107028
em is
0.219016
0.065565
0.438031
3.291336
0.131409
0.095558
0.028607
0.191117
1.454391
0.057335
0.022212
0.004665
G-15
-------
Table 27. Continued.
FIPS
06001
06001
06001
06001
06001
06001
06001
06001
06001
06021
06021
06021
06021
06021
06021
06021
06021
06021
06021
06021
48001
48001
48001
48001
48001
48001
48001
48001
48001
48001
48001
48001
48001
48001
48001
48001
48001
48001
48001
48001
48001
48001
48001
48001
48001
sec
2260002000
2260002000
2260002000
2265001010
2265001010
2265001010
2265001010
2265001010
2270005045
2260002006
2260002006
2260002006
2260002006
2260002006
2265001020
2265001020
2265001020
2265001020
2265001020
2270005045
2260001000
2260001000
2260001000
2260001000
2260001000
2260002000
2260002000
2260002000
2260002000
2260002000
2265001000
2265001000
2265001000
2265001000
2265001000
2265007015
2265007015
2265007015
2265007015
2265007015
2270001000
2270001000
2270001000
2270001000
2270001000
CAS
50000
71432
75070
106990
107028
50000
71432
75070
107028
106990
107028
50000
71432
75070
106990
107028
50000
71432
75070
107028
106990
107028
50000
71432
75070
106990
107028
50000
71432
75070
106990
107028
50000
71432
75070
106990
107028
50000
71432
75070
106990
107028
50000
71432
75070
emis
0.090633
0.097496
0.041693
0.039095
0.002868
0.067717
0.187604
0.034401
0.000112
0.00028
3.84E-05
0.000398
0.003259
0.000457
0.319786
0.066658
1.220082
1.522297
0.608339
0.001556
0.032812
0.006512
0.058605
0.70733
0.035782
0.001097
0.000219
0.001857
0.018899
0.001217
0.034624
0.003068
0.075164
0.397382
0.017972
5.26E-05
3.86E-06
6.47E-05
0.000304
2.26E-05
0.000114
0.000843
0.010962
0.00152
0.005444
G-16
-------
20. Merged the output of step 19 with step 1 output using PROC SQL by SCC, retaining all
observations from the step 19 output and the ASPEN source group from step 1 output.
21. Based on the source group, output source group=5 emissions to a gasoline dataset and all
other emissions to a non-gasoline dataset for later use.
Table G-28. MSAT gasoline emissions after merging with SCC/bin cross reference and
separate gasoline emissions and non-gasoline emissions (Section 10.3.2, steps 20 and 21).
FIPS
06001
06001
06001
06001
06001
06001
06001
06001
06001
06001
06001
06001
06001
06001
06001
06001
06001
06001
06001
06001
06021
06021
06021
06021
06021
06021
06021
06021
06021
06021
48001
48001
48001
48001
48001
48001
48001
48001
48001
48001
48001
SCC
2260001010
2260001010
2260001010
2260001010
2260001010
2260001030
2260001030
2260001030
2260001030
2260001030
2260002000
2260002000
2260002000
2260002000
2260002000
2265001010
2265001010
2265001010
2265001010
2265001010
2260002006
2260002006
2260002006
2260002006
2260002006
2265001020
2265001020
2265001020
2265001020
2265001020
2260001000
2260001000
2260001000
2260001000
2260001000
2260002000
2260002000
2260002000
2260002000
2260002000
2265001000
CAS
106990
107028
50000
71432
75070
106990
107028
50000
71432
75070
106990
107028
50000
71432
75070
106990
107028
50000
71432
75070
106990
107028
50000
71432
75070
106990
107028
50000
71432
75070
106990
107028
50000
71432
75070
106990
107028
50000
71432
75070
106990
em is
0.219016
0.065565
0.438031
3.291336
0.131409
0.095558
0.028607
0.191117
1.454391
0.057335
0.022212
0.004665
0.090633
0.097496
0.041693
0.039095
0.002868
0.067717
0.187604
0.034401
0.00028
3.84E-05
0.000398
0.003259
0.000457
0.319786
0.066658
1.220082
1.522297
0.608339
0.032812
0.006512
0.058605
0.70733
0.035782
0.001097
0.000219
0.001857
0.018899
0.001217
0.034624
grp
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
G-17
-------
Table G-28. Continued.
FIPS
48001
48001
48001
48001
48001
48001
48001
48001
48001
48001
48001
48001
48001
sec
2265001000
2265001000
2265001000
2265001000
2265007015
2265007015
2265001000
2265001000
2265001000
2265001000
2265001000
2265007015
2265007015
CAS
107028
50000
71432
75070
106990
107028
106990
107028
50000
71432
75070
106990
107028
emis
0.003068
0.075164
0.397382
0.017972
5.26E-05
3.86E-06
0.034624
0.003068
0.075164
0.397382
0.017972
5.26E-05
3.86E-06
grp
5
5
5
5
5
5
5
5
5
5
5
5
5
Table G-29. Partial listing of non-gasoline emissions.
FIPS
06001
06021
48001
48001
48001
48001
48001
sec
2270005045
2270005045
2270001000
2270001000
2270001000
2270001000
2270001000
CAS
107028
107028
106990
107028
50000
71432
75070
emis
0.000112
0.001556
0.000114
0.000843
0.010962
0.00152
0.005444
grp
3
3
3
3
3
3
3
G-18
-------
22. From the gasoline dataset created in step 21, merged the emissions with the fractions created in step 18 by FIPS/SCC/CAS.
Output all observations from the dataset created in step 21 to the dataset.
23. Multiplied the emissions by fractions. If the CAS = 71432 (benzene) multiplied the emissions by the exhaust fraction to get
exhaust emissions and multiply the emissions by the evaporative fraction to get evaporative emissions. If not benzene, these
calculations were not done.
Table G-30. Partial listing of gasoline emissions with projection factors and exhaust and evaporative emissions for benzene
(Steps 22 and 23).
FIPS
06001
06001
06001
06001
06001
06021
06021
06021
06021
06021
48001
48001
48001
48001
48001
sec
2260001010
2260001010
2260001010
2260001010
2260001010
2260002006
2260002006
2260002006
2260002006
2260002006
2260001000
2260001000
2260001000
2260001000
2260001000
CAS
106990
107028
50000
71432
75070
106990
107028
50000
71432
75070
106990
107028
50000
71432
75070
emis
0.219016
0.065565
0.438031
3.291336
0.131409
0.00028
3.84E-05
0.000398
0.003259
0.000457
0.032812
0.006512
0.058605
0.70733
0.035782
exh frac
0.987316341543
0.990589732427
0.7166970164271
eva frac
0.012683658457
0.00941027
0.2833029835729
emis exh
3.249589818309
'0.003228331938
0.5069413006294
emis eva
0.041746181691
0.00003067
0.2003886993706
Steps 19 through 23 were performed in the SAS MACRO read_msat.
G-19
-------
24. Merged the output of step 23 with the projection factors calculated in calc_fabtors.sas by FIPS/CAS using PROC SQL. Retain
all observations from step 23 output.
Table G-31. Partial listing of gasoline emissions with projection factors.
FIPS
06001
06001
06001
06001
06001
06021
06021
06021
06021
06021
48001
48001
48001
48001
48001
sec
2260001010
2260001010
2260001010
2260001010
2260001010
2260002006
2260002006
2260002006
2260002006
2260002006
2260001000
2260001000
2260001000
2260001000
2260001000
CAS
106990
107028
50000
71432
75070
106990
107028
50000
71432
75070
106990
107028
50000
71432
75070
emis
0.219016
0.065565
0.438031
3.291336
0.131409
0.00028
3.84E-05
0.000398
0.003259
0.000457
0.032812
0.006512
0.058605
0.70733
0.035782
emis exh
3.249589818309
0.003228331938
0.5069413006294
emis eva
0.041746181691
0.00003067
0.2003886993706
pf exh
1.0012384215
1.0018241077
0.944296577
1.0015537589
1.0012845671
1
1.0018887224
0.9424016216
1.0016036635
1.0028720014
1
1.0038495361
0.9104735847
1.0033413338
pf evap
0.5999749383
0.5999925419
0.570001502
25. Calculated projected or controlled emissions for step 24 output. For benzene, the projected emissions were the sum of the
exhaust emissions multiplied by the exhaust projection factor (calculated in calc_factors.sas) and the evaporative emissions
multiplied by the evaporative projection factor (calculated in calc_factors.sas) (Equation 23). Otherwise the emissions were
multiplied by the exhaust projection factor (Equation 24).
G-20
-------
Table G-32. Partial listing
emissions and emis orig is
of gasoline emissions after calculating controlled emissions. The variable emis is the controlled
the reference projected emissions for MS AT.
FIPS
06001
06001
06001
06001
06001
06021
06021
06021
06021
06021
48001
48001
48001
48001
48001
sec
2260001010
2260001010
2260001010
2260001010
2260001010
2260002006
2260002006
2260002006
2260002006
2260002006
2260001000
2260001000
2260001000
2260001000
2260001000
CAS
106990
107028
50000
71432
75070
106990
107028
50000
71432
75070
106990
107028
50000
71432
75070
emis
0.2192872341
0.065565
0.438031
0.4388300157
3.093623204868
0.1316131779
0.0002803597
0.0003987517
0.003060787025
0.0004577329
0.0329062361
0.006512
0.0588306021
0.5757785228416
0.0359015596
emis exh
3.249589818309
0.003228331938
0.5069413006294
emis eva
0.041746181691
0.00003067
0.2003886993706
pf exh
1.0012384215
1
1.0018241077
0.944296577
1.0015537589
1.0012845671
1
1.0018887224
0.9424016216
1.0016036635
1.0028720014
1
1.0038495361
0.9104735847
1.0033413338
pf evap
0.5999749383
0.5999925419
0.570001502
emis orig
0.219016
0.065565
0.438031
3.291336
0.131409
0.00028
3.84E-05
0.000398
0.003259
0.000457
0.032812
0.006512
0.058605
0.70733
0.035782
26. Concatenated the output of step 25 with the non-gasoline emissions created in step 21.
27. Sorted step 27 output by FIPS/SCC/CAS and output to a permanent dataset.
G-21
-------
Table G-33. Partial listing of nonroad emissions after concatenating controlled nonroad gasoline emissions with non-gasoline
nonroad emissions, sorting by PIPS, SCC, and CAS and output to a permanent dataset with only needed variables (Steps 26
and 27).
FIPS
06001
06001
06001
06001
06001
06001
06021
06021
06021
06021
06021
06001
48001
48001
48001
48001
48001
48001
48001
48001
48001
CAS
106990
107028
50000
71432
75070
107028
106990
107028
50000
71432
75070
107028
106990
107028
50000
71432
75070
106990
107028
50000
71432
SCC
2260001010
2260001010
2260001010
2260001010
2260001010
2270005045
2260002006
2260002006
2260002006
2260002006
2260002006
2270005045
2260001000
2260001000
2260001000
2260001000
2260001000
2270001000
2270001000
2270001000
2270001000
em is
0.2192872341
0.065565
0.438031
0.4388300157
3.093623204868
0.000112
0.1316131779
0.0002803597
0.0003987517
0.003060787025
0.0004577329
0.001556
0.0329062361
0.006512
0.0588306021
0.5757785228416
0.0359015596
0.000114
0.000843
0.010962
0.00152
emis exh
3.249589818309
0.003228331938
0.5069413006294
emis eva
0.041746181691
0.00003067
0.2003886993706
flag
C
emis orig
0.219016
0.065565
0.438031
3.291336
0.131409
0.00028
3.84E-05
0.000398
0.003259
0.000457
0.032812
0.006512
0.058605
0.70733
0.035782
G-22
-------
®
Steps 24 through 27 were performed in the SAS MACRO apply pf.
All steps were performed in the SAS MACRO control with the four digit year, 2015,
2020, or 2030 as the argument.
Figure G-2 shows the steps to calculate the exhaust and evaporative fractions and Figure G-3
shows the steps of the application of the exhaust and evaporative fractions and projection factors.
am_grp_MSAT.txt
Read SCC/grouping cross
reference file, retaining
nonroad. Correct SCC
codes with wrong group
MSATBenzN20XX. csv
where XX is 2 digit
year
©
Read nonroad NMIM j
emissions broken out into | ^ ] ^J Sort by
exhaust and evaporative I ^ nmim \ FTPS/SCC/CAS/emissions
emissions for VOC and j I type
benzene. Retain benzene, j
Create macro variables of the exhaust
and evaporative emissions for SCC
2265001000 in FTPS 06021
! Extract emissions for 2 and 4 stroke
I emissions, assigning engine type. I
| Keep only SCC codes where 3 last 3 ] '
! characters not 000. !
15) | Merge by FIPS/CAS and
| where emissions are 2 or 4
! stroke
I FIPS/CAS/engine ~I FIPS/CAS/engine
Figure G-2. Steps in the calculation of the exhaust and evaporative emissions fractions (steps 2-
18 in control_nonroad.sas (found in the MSAT rule docket EPA-HQ-OAR-2005-0036).
G-23
-------
nonroad_20XX.sas7bdat
MSAT nonroad inventory
| Merge by
| SCC with
H sec bin
j Merge by
| FIPS/SCC/CAS
with fractions
from step 18
Split into gasoline and
1 non-gasoline
emissions based on
group
nonroad 20XX.sas7bdat
| Sort by FIPS/SCC/CAS and |
| output to permanent dataset I
Calculate exhaust and
evaporative components of
benzene emissions
Apply projection factors to
emissions
nonroad factors 20XX.sas7bdat
(24) \ Merge with projection
H factors by FIPS/CAS
Figure G-3. Steps in calculating controlled nonroad emissions (steps 19-27 of
control_nonroad. sas).
G-24
------- |