Fuel Supply Defaults: Regional Fuels
and the Fuel Wizard in MOVES2014
£% United States
Environmental Protect
Agency
-------
Fuel Supply Defaults: Regional Fuels
and the Fuel Wizard in MOVES2014
Assessment and Standards Division
Office of Transportation and Air Quality
U.S. Environmental Protection Agency
NOTICE
This technical report does not necessarily represent final EPA decisions or
positions. It is intended to present technical analysis of issues using data
that are currently available. The purpose in the release of such reports is to
facilitate the exchange of technical information and to inform the public of
technical developments.
£% United States
Environmental Protection
^'^1 Agency
EPA-420-R-16-002
November 2016
-------
Table of Contents
1 Executive Summary 2
2 Introduction 2
3 Background 3
4 Regional Fuels 5
5 Fuel Properties 10
6 Renewable Fuels Market Share 15
7 NonroadFuel Supply 17
8 Fuel Wizard 19
9 References 21
1
-------
1 Executive Summary
Historically, EPA has used a combination of disparate sources of fuel property data to populate
the local fuel quality database tables included in the MOVES model. These data sources,
including surveys from the Alliance of Automobile Manufacturers (AAM), state and local point
source sampling, and reformulated gasoline (RFG) compliance reports, have always provided a
detailed snapshot of fuel qualities for a given local area at a given time. However, in reviewing
the results of this approach in the context of our knowledge of national gasoline production and
distribution, EPA believes that this approach over-specified the local fuel properties beyond the
capabilities of the available data. We have overhauled and simplified the aggregation of fuel
survey data for use in the MOVES model, as well as now including the extensive refinery gate
batch dataset collected as a part of EPA compliance programs. This has had the effect of
reducing the total number of unique fuel property areas from approximately 425 in MOVES2010
and its minor releases51 (hereafter referred to as MOVES2010) to approximately 45 in
MOVES2014, generally representing the geographic distribution of refined products pipelines
and taking into account state and local fuel programs. These fuel properties are also now
projected to 2050 in MOVES2014 using estimates from the Annual Energy Outlook13'1'2) report,
and includes El 5 and E85 use (a large update from MOVES2010, which held fuels constant
beyond 2012 and did not contain El5 fuel supplies). In doing so, we hope to have 1) clarified the
source of the fuel properties 2) better represented the actual in-use fuel properties and 3) better
represented the actual variation in fuel properties across different regions in the U.S.
Additionally, we have included the fuel wizard tool in MOVES2014 that allows users to adjust
local fuel properties using EPA refinery modeling based on the Tier 3 Motor Vehicle Emissions
and Fuel Standards rulemaking.3 With this tool, states and local areas can streamline the process
to analyze potential fuel control programs, such as RVP, by taking into account the changes to
other fuel properties that would occur with these programs.
2 Introduction
This document describes the background and methodology behind the changes and updates to
the fuel supply components of MOVES2014 and its minor release, MOVES2014a.
Section 3 provides an overview of the fuel supply in MOVES2010 and describes our rationale
for significantly changing the data and structure of the fuel supply in MOVES2014. For more
information on the fuel supply included in the MOVES2010 model, please see the MOVES2009
Fuel Effects report4.
Section 4 defines the large geographical areas used in the new MOVES2014 fuel supplies.
Section 5 discusses the aggregation of the disparate sources of data used in assigning properties
to these areas, the adjustment to these properties for the inclusion of ethanol and other
renewables, and the effects from local fuel control programs. Section 6 discusses the
methodology of applying the AEO projections for renewable fuel (including El 5 and E85), into
future years. Section 7 describes the development of the fuel supply used in MOVES2014
a Including MOVES2010a and MOVES2010b
b MOVES2014 used the AEO2014 Early Release1. MOVES2014a used the AEO2014 Final Release2.
2
-------
specifically for the nonroad component of the model. Finally, Section 8 describes the fuel wizard
tool based on EPA refinery modeling data that can be used to adjust fuel properties in modeling
local fuel control.
We updated both the fuel supply and fuel wizard in MOVES2014a from MOVES2014. The
specifics and the emission impacts of these changes are discussed in Sections 6, 7, and 8. The
report refers to MOVES2014 for the general structure and changes made to MOVES2014. We
refer to MOVES2014a to discuss changes that were made between MOVES2014a and
MOVES2014.
3 Background
The fuel supply in MOVES2010 included extensive fuel property data taken from many sources,
including AAM fuel surveys, state and local point source sampling, and EPA collected
reformulated gasoline compliance data. However, while these data may provide accurate fuel
property information in a specific location and at a specific point in time, they may not have
provided a true reflection for an area over time, given the limitations in the data and how it was
collected. It resulted in significant differences in fuel properties between even neighboring
counties, even though they would have received fuel from the same fuel distributors. These
differences were left largely unresolved in the default fuel database included in MOVES2010,
leaving the model with a patchwork of many disparate regions with large variations in fuel
properties that generally did not follow geographic regions, pipeline and terminal locations, or in
some cases local fuel control areas. See Figure 1 for an example of the large number of differing
fuel property areas included in the fuel database for MOVES2010.
Figure 1 - Disparate fuel property areas in the MOVES2010 model
Note: Each color represents a difference in any single fuel property for a given county. This example is taken from the
\K)\ l-SJVltlh fuel supply for calendar year 2007.
We believe that an improved method should be used to generate these fuel supply areas, as the
MOVES2010 method may exaggerate emission inventory differences among neighboring
counties in different fuel supply areas, and therefore, we sought a new technique to refine the
generation of the default fuel database for MOVES2014.
The new technique relies more heavily on our nationwide refinery gate compliance data set (EPA
collects electronic records of refinery fuel property data as it leaves the facility, for compliance
3
-------
purposes), and less on local retail surveys. This data includes volumes and fuel properties used in
MOVES calculations (e.g., RVP, sulfur, aromatics content, distillation values), the types of fuel
(conventional gasoline (CG), reformulated gasoline (RFG), blendstock for oxygenate blending
(BOB)), as well as additional fuel property data (e.g., API gravity, batch type) that were not
being used in the previous versions of MOVES. This data was used by EPA for the Tier 3
rulemaking analysis.? This dataset includes approximately thirty thousand entries per year (each
entry containing batch-specific properties), with EPA using the years 2007, 2009, and 2011 for
the MOVES2014 analysis.
After the initial aggregation of this compliance batch data, it became apparent that our previous
method of basing large geographic areas on a small number of point-source samples was
inadequate. Batch to batch variation was much larger than we had assumed in the previous
modeling attempts. Figure 2 provides an example of the batch to batch variation that occurs for
E200 (the percentage of fuel that evaporates at 200 °F). E200 was specifically selected for this
example due to its current status as an unregulated fuel property that nevertheless has a
significant impact on exhaust emissions, based on the EPAct models (for more information, see
the MOVES2014 Fuel Effects report6).
Figure 2 - Example of fuel quality variation (E200) from EPA compliance data
c
(Ll
UJ
CfL
E
o
s
r-J
LU
dj
bO
!Z
lT3
10O
M-i- * vtSSic fBJfe'TSiK'-wviyrfWg*cy.iiiBg:¦1
1O0
150
?oo
250 30>0
35Q
400
450
C
dj
CL
Batch No. (in 1000's)
Coo's representa'ffs-entfaSssoffL-ss [CG, RFG., Prem um, and Regular
Of course, selecting any one of these points to represent an area as a whole would be
inappropriate. In addition, due to the prevalence of downstream blending with multiple fuel
deliveries, it would be unlikely that variations of this magnitude would be seen at a retail outlet
(as multiple batches would tend to move the overall fuel qualities to the average). Therefore, a
new aggregation methodology was developed to take advantage of the large amount of data
available from the EPA-collected compliance data, while also preserving the importance of
downstream sampling from AAM and other sources, as well as applying the latest refinery
modeling developed as part of the Tier 3 rulemaking process. With this new methodology, we
believe that the fuels contained in the default MOVES2014 database 1) more closely represent
4
-------
actual in-use fuel qualities for a given area, 2) more closely represent the fuel variations from
one local region to another, and 3) provide simplified interpretation of the fuel supply database
for external users of the model.
4 Regional Fuels
Aggregating fuels into larger, more representative areas was the main goal in the development of
the regional fuels approach. Using this methodology, we have created eleven general fuel regions
for the United States and major territories. We initially based these fuel regions on existing
PADD boundaries (a historic division of fuel supply areas originally developed in the 1950s),
and then adjusted to account for broad fuel distributi on corridors and the presence of bulk fuel
pipelines and terminals. Terminal locations and their associated pipelines were identified using
the Petroleum Terminal Encyclopedia provided by the Oil Price Information Service (OPIS)7
For illustrative purposes, Figure 3 below shows the pipeline locations. Using these terminal
locations, we could group areas sharing connections to similar pipeline networks as a part of
defining the new regional fuel areas. Within these broad fuel regions, we further identified areas
requiring additional unique fuel qualities, due to federal, state, or local fuel quality standards.
Figure 3 - An example of petroleum product pipelines in the continental United States8
f V —. / / „,
f f _y f \ M
* < i V / \
» \ \ I
i \ \ . V 7 ^ \ i
4n
v V,.'* UN| ] \\ j A
taJvy;~r
1 L I \>'/ jJT ft*
• / '• i
/j v ,
jl r ^ I v /fj / / i vi
_ ' / ? <1, it %
V— ' i
HiiMi '
Although an area has not been specifically defined for the historical Geographic Phase-in Area
(GPA) as was done for the MOVES2010a model, we have created a fuel region including the
Rocky Mountain area closely related to this area. Table 1 identifies and briefly describes each
region as used in the MOVES2014 default fuel supply database. Additionally, please see Figure
4 for a graphical representation of these new fuel regions, including areas adjusted from the large
fuel region aggregates to account for state or local fuel control programs. Comparing Figure 4 to
Figure 1 helps illustrate the major changes from the versions of the MOVES2010a model to
MOVES2014.
5
-------
Table 1 - Base fuel region ID numbers and general descriptions
Base
Region
ID#
Base Region
Name
General Description
1
East Coast
East coast states, west to Appalachians; Florida; and Gulf Coast region
2
Midwest
Midwest states, east to Appalachians; Tennessee; Kentucky
3
South
Iowa to Texas (North to South), Alabama to New Mexico (East to West); not including
counties along the Gulf Coast;
4
North
North and South Dakota, Minnesota, Wisconsin
5
Rocky Mtns.
Pacific Northwest, Rocky Mountain states
6
CA/NV/AR/AU
Others
California0, Nevada, Arizona not using RFG, small market areas including Alaska,
Hawaii, Puerto Rico and Virgin Islands
11
East Coast RFG
East Coast states and regions using RFG fuel or under a controlled fuel program
12
MD/VA RFG
Maryland and Virginia regions using RFG fuel or under a controlled fuel program
13
Texas RFG
Texas regions using RFG fuel or under a controlled fuel program
14
Midwest RFG
Midwest regions using RFG fuel or under a controlled fuel program
15
California RFG
California using California fuel, Nevada and Arizona regions using California fuel or
California fuel derivatives
0 Currently, Base Region 6 does not include any market share in California in MOVES2014.
6
-------
Figure 4 - Fuel Regions in the MOVES2014 model
The MOVES2014 regional fuel areas are defined by the regionCounty table in the MOVES
default database. This table contains the counties [coimtylD] included for each defined region
[regionID] for a given year [fuelYearlD], including regions with state or local fuel control
programs.d The regionCounty also has an identifier [regionCodelD] which identifies onroad or
nonroad use, which can be used to model separate fuel regions for onroad and nonroad fuel use.
The fuel properties associated with each of these regions are represented in another table,
fuelFormulation, which is discussed in detail in Section 5 of this document. Furthermore, the
fuelSupply table defines the market share fractions of various fuels sharing the same fuel region,
such as E10, E15, E85, biodiesel, and CNG. For more information regarding the specific design
and properties of these tables in MOVES2014, please refer to the MOVES Module Reference9
The regionJJ) field contains some information regarding the nature of the fuel region as part of
the ID number, in an effort to make this table more easily readable by the user. The ID can be
read as described below:
regionID: AABBCCDDXX
where: A A = base region ID
BB = maximum summer region EVP value, or 00 for ASTM
CC = presence ofRVP waiver
DD = presence of minimum ethanol level
XX = reservedfor future use
d For the October release of MOVES2014, we updated the county to fuel region assignments based on updates to
local fuel programs. We changed the fuel region assignment for the following North Carolina counties (Davidson,
Davie, Durham. Forsyth Granville, and Guilford) from the East RVP control area (RegionID 178000000) to the
East Coast region with no local fuel control programs (RegionID 100000000).
7
-------
The contents of the regionID are shown in Table 2 below. The maximum summer RVP, refers to
the maximum Reid Vapor Pressure, which is a measure of the volatility of the fuel. 10'e Local fuel
programs set a limit to the vapor pressure of the gasoline fuel to reduce evaporative emissions of
volatile organic compounds. If this value is 0, then we assume that the region is using the federal
RVP requirements.10
The RVP waiver refers to the "1.0 psi RVP allowance for gasoline containing ethanol at 9 to 10
volume percent."10 Thus, for regions with an RVP waiver and a summer RVP limit of 7.8 psi, the
RVP of an E10 fuel is assumed to be 8.8 psi. Not all fuel regions allow for the 1.0 psi fuel
waiver; for example, a State Implementation Plan (SIP) may enforce an RVP standard that does
not allow the 1.0 psi waiver. Thus, each fuel region is defined on the presence of an RVP waiver.
The minimum ethanol content field establishes a minimum ethanol level required either by RFG
or local fuel programs.
e State-by-State RVP table with maximum RVP is located at the US EPA Reid Vapor Pressure webpage10
8
-------
Table 2. RegionID Table in MOVES2014
regionID
AA, Base
Region ID#
Base Region Name
BB,
Maximum
summer RVP
(psi) or 00 for
ASTM
CC, E10 RVP
Waiver (00=1
psi waiver,
01=no waiver)
DD,
Minimum
ethanol
volume,
%
XX
(Reserved
for future
use)
0
0.0
0
0
0
100000000
0.0
0
0
0
100010000
1
East Coast
0.0
1
0
0
170000000
7.0
0
0
0
178000000
7.8
0
0
0
178010000
7.8
1
0
0
200000000
0.0
0
0
0
270000000
2
Midwest
7.0
0
0
0
278000000
7.8
0
0
0
278010000
7.8
1
0
0
300000000
0.0
0
0
0
370000000
3
South
7.0
0
0
0
370010000
7.0
1
0
0
400000000
4
North
0.0
0
0
0
500000000
5
Rocky Mtns
0.0
0
0
0
578000000
7.8
0
0
0
600000000
6
CA/NV/AR/A11 Others
0.0
0
0
0
678000000
7.8
0
0
0
1170011000
11
East Coast RFG
7.0
1
10
0
1270011000
12
MD/VA RFG
7.0
1
10
0
1370011000
13
Texas RFG
7.0
1
10
0
1470011000
14
Midwest RFG
7.0
1
10
0
1570011000
15
California
7.0
1
10
0
In total, MOVES2014 has 24 fuel regions, when combining the base fuel areas with the state and
local fuel control programs. We may consider additional fuel regions in the future based on fuels
data and recommendations from users.13 We are aware that fuel regions, and especially local and
state fuel control programs, change over time, both historically and into the future. However, for
this the default fuel database in MOVES2014, we have chosen to keep the fuel regions consistent
(a snapshot in the year 2013) through time. An update to this methodology allowing for historical
and future variation in fuel region definitions is being considered for future versions of the
MOVES model.
9
-------
5 Fuel Properties
The EPA fuel compliance data consists of a set of databases, by year, which contain reports
provided by refiners, downstream blenders, and terminal operators as part of their compliance
process. This data is provided to us periodically, with a complete dataset for the previous year
usually compiled in the subsequent year. These reports (and underlying databases) are
considered Confidential Business Information (CBI) and cannot be provided as part of this
document. For the regional fuel properties in MOVES2014, we chose to use 2007, 2009, and
2011 as the analysis years, considering the completeness of these databases. Future updates to
the MOVES fuel properties will take advantage of new compliance data as it becomes available.
The fuel compliance data includes information on fuel as it leaves its point of production (e.g.,
refinery) and contains fields for a multitude of properties as well as information regarding the
specific type of fuel created (e.g., CG, BOB, RFG). These reported properties form the basis for
the fuel properties used in the regional fuel methodology, and are eventually aggregated into the
large fuel regions described in the previous section. The specific fuel properties contained in
these batch data can be found in Table 3 below:
Table 3 - Properties included in EPA fuel compliance data
Batch Volume
Production Date
Batch Grade
VOC Control
Oxygen
Sulfur
Aromatics
Olefins
Benzene
Methanol
MTBE
Ethanol
ETBE
TAME
t-Butanol
RVP
T50
T90
E200
E300
API Gravity
10
-------
Before aggregating to fuel regions, we processed the dataset to exclude duplicate reporting (i.e.,
a refinery and independent lab results may both report the same fuel). We also repaired or
excluded batches with missing or inappropriate data (e.g., T90 can be correlated to E300, such
that if a T90 value is missing for a batch of fuel but an E300 value is present, T90 can be
estimated and included in the full dataset. See Equation 1 and Equation 2 below for correlation
equations used for T50 and T90 gap filling). Finally, we separated differing types of fuel batches
for further processing (i.e., CG and pre-blended fuels can be included in the dataset without
adjustment, BOB fuels must be adjusted to account for oxygenates added downstream from the
refinery gate). After these processing steps, we were left with a set of data containing between
twenty and thirty-five thousand usable points of batch properties (depending on year), with no
fuel region being represented by less than one thousand batches.
T50 = 2.04 081 63 X ( 147.91 — E200 ) Equation 1. T50 Correlations to E200
T9o = 4.5454 X ( 155.47 — E300 ) Equation 2. T90 Correlations to E300
These data were then aggregated by fuel region, using fuel batch size as a weighting factor.
Initially, fuels were aggregated into four seasonal categories, including summer, winter, and two
transitionary 'shoulder' seasons. After reviewing the results of these categories, EPA determined
that there was not adequate data (<100 batches for some regions) for the shoulder seasons to
remain as separate categories and therefore, these data were re-aggregated into the summer and
winter seasons. Additionally, because Tier 3 refinery modeling results (described later in this
section) only provide adjustments for winter and summer seasons, we felt it was not appropriate
to apply these adjustments to additional seasons at this time. The two aggregation seasons used
for this dataset are summer (May, June, July, and August) and winter (January, February, March,
April, September, October, November, and December).
In addition to the EPA compliance data, we also included point source downstream fuel
sampling data measured by the AAM11. While this data cannot represent fuel properties for a
region on its own (as explained in Section 3), it serves as an important validation to adjustments
made to compliance data to account for BOB fuels, as well as aggregate results. Because this
data is sampled at retail locations, it includes steps in the downstream fuel process that we cannot
fully account for using only compliance data sampled at refineries. By including this downstream
sample data as part of the validation process, we have increased our confidence that our final
result is appropriately representing retail location samples for the given fuel region.
As part of the Tier 3 rulemaking (which includes reductions in fuel sulfur content), adjustments
to secondary fuel properties affected by changes in sulfur, RVP, and ethanol level were
generated using refinery modeling tools.5 EPA understands that fuel property changes cannot
happen independently, and has made an effort to capture these interactive effects in the
MOVES2014 fuel supply. The adjustments for sulfur, RVP, and ethanol level are shown in Table
4, Table 5 and Table 6. For example, the RVP adjustment factor in Table 4, increases the RVP
by 1 psi, when the ethanol is changed from E0 to E10, to account for the RVP 1 psi waiver
discussed in Section 4.
11
-------
Table 4 - Adjustment factors for various ethanol blends
ETHANOL ADJ1
IJSTME]
VT FACTORS (from E0 to E10 or
E15)
FUEL
DESCRIPTION
RVP
SULF
AROM
OLEF
BENZ
E200
E300
T50
T90
E10S
E10 Summer Fuel
1.00
-2.02
-0.46
3.11
0.39
-6.34
-1.77
E10W
E10 Winter Fuel
1.00
-3.65
-2.07
4.88
0.54
-9.96
-2.45
E15 S
E15 Summer Fuel
-3.36
-1.64
9.24
0.91
-18.86
-4.14
E15 W
E15 Winter Fuel
-5.69
-3.27
11.11
1.01
-22.67
-4.59
Table 5 - Adjustment factors for lower RVP blends
RVP ADIUSTM
[ENT FACTORS (
per PSI)
FUEL
DESCRIPTION
RVP
SULF
AROM
OLEF
BENZ
E200
E300
T50
T90
per PSI
Boutique fuel adj.
-1.00
-1.26
-0.50
2.57
2.27
Table 6 - Adjustment factors for lower sulfur blends
SU
LFUR ADIUSTMENT FACTORS (per ppm)
FUEL
DESCRIPTION
RVP
SULF
AROM
OLEF
BENZ
E200
E300
T50
T90
per ppm
Sulfur fuel adj
-1.00
-0.032
Gasoline fuel compliance data suggests that the vast majority of fuel batches (>80% in the 2011
database) produced by refiners are produced as a match blendstock (CBOB/RBOB) for
downstream ethanol blending. Therefore, for calendar years beyond 2011, when making
adjustments to include the addition of ethanol (as E10 or E15 fuel), we did not include the
additional dilution effects on sulfur or other fuel properties occurring from this blending beyond
those required to meet federal, state, or local limits. Table 7 shows the gasoline sulfur levels used
in MOVES2014 for 2011 and later years. We assume gasoline fuels in all regions meet the
requirements of 10 ppm sulfur fuel of the Tier 3 program starting in 2017.3 For calendar years
between 2006 and 2011, we used the batch fuel data as the source of the fuel sulfur content. For
these years, the sulfur level of conventional gasoline and E10 exceeds 30 ppm, and varies
between base fuel regions.
12
-------
Table 7. Gasoline fuel sulfur content (ppm) used in the fuel supply for conventional gasoline and gasoline-
ethanol blends
E0 through E15 gasoline-ethanol blends
E85 Blends
Calendar
Years
All Base Regions (except
California)
California (Base Region
ID #15)f
All Base
Regions
2011-2016
30
9
8
2017-2050
10
Unfortunately, even though MOVES2014 includes the years 1990 and 1999-2050, batch fuel
compliance data was not available for use as part of this analysis for years prior to 2006 or after
2011 (records before this period exist in non-electronic form, for more information see the 2005
Fuels Trends report12). For the years 1990, 1999, and 2000-2005, we used county specific fuel
data contained in MOVES2010 as a surrogate for aggregation. These county level fuel properties
were aggregated into the new fuel regions in a similar way to the fuel compliance data, as
described above, with weighting occurring by VMT in those counties. For fuel property data
after 2011, we assumed no change to current fuel properties until affected by statutory
regulations, such as Tier 3. As such, the database will contain major fuel property changes in
2017, and then remain constant afterwards. This excludes the volumes of renewables included in
the fuel supply, which vary by year, and are described in more detail in Section 0. Of course, in
reality, there will be changes to fuel property data due to market shifts in future years that we are
not able to predict. Therefore, as new fuel property information becomes available through EPA
compliance reports as well as AAM fuel survey data, we will be including this data in future
versions of the MOVES model.
As part of the update to MOVES2014, we have included a new fuel effects model based on the
EPAct/V2/E-89 (EPAct) gasoline fuel effects study.6 The fuel effects model contains
adjustments to gasoline vehicle emission rates based on specific fuel properties such as ethanol,
aromatics, RVP, and distillation properties as well as interactions between these properties and
replaces the Complex Model4 for vehicles model years 2001 and newer. For more information on
the specifics of this new fuel effects model, please refer to the MOVES2014 Fuel Effects report.6
We are aware that for some regions, there may still be MTBE in-use post-2001, a fuel property
that cannot be properly modeled using the EPAct fuel effects model. In MOVES2014, regions
using MTBE fuel have been replaced with ethanol (E10) fuel in order to provide an
approximation of fuel effects in years post-2001. A solution for this limitation in historical
MTBE modeling may be explored in future versions of the MOVES model.13
In MOVES2014, the fuel property information described above is contained in the
fuelFormulation table. This table contains all the fuel properties relevant to the fuel effects
models (e.g., the Complex Model, and now the EPAct Model), including RVP, sulfurLevel,
f California is the only fuel region that has E5 fuel, which is assumed to have 9 ppm sulfur content, starting in 2007.
In 2010, E5 is replaced with E10 fuel which also is assumed to have 9 ppm sulfur content. In MOVES2014a, all
El5 from the California fuel supply was removed to correctly reflect the fuels in California.
13
-------
ETOHvolume, MTBEvolume, TAMEvolume, aromaticContent, olefinContent, benzeneContent,
e200, e300, volToWtPercentOxyg-14, BioDieselEsterVolume, Cetanelndexh, PAHContenth, T50,
and T90 as post-aggregated values for a given fuel region (or regional control program), for a
given time period. Each set of these fuel formulations is given an ID \fuelFormulationID\ and
also assigned to a subtype of fuels to ease model calculations for similar fuel properties
[fuelSubtypelD]. The fuelSupply table is then used to assign these fuel formulations to a month
\monthGroupID\ and year [fuelYearID\ for each fuel region [fuelRegionID], as discussed in
Section 4. In the cases where a given fuel region contains more than one fuel of a given fuel
subtype (e.g., a fuel region containing both an E10 and an E15 fuel), a value for the fraction of
the fuel sold with those properties can be assigned to each of fuels [marketShare]. For more
information regarding the specific design and properties of these tables in MOVES2014, please
refer to the MOVES2014 Module Reference.9
For diesel and compressed natural gas (CNG), we do not use compliance data as the source of
the fuel properties. The sulfur contents for diesel (including biodiesel) across all fuel regions are
shown in Table 8. Starting in 2006, we assumed that refineries are producing diesel fuel at the
maximum 15 ppm sulfur level required by the regulations.15
Table 8. Default diesel sulfur content in MOVES by calendar year across all fuel regions
Year
Sulfur level,
ppm
1990
1000
1999-2005
130
2006-2050
15
For compressed natural gas, we assumed that CNG used in onroad fuels has a sulfur content of
7.6 ppm based on a CNG transit bus study documented in the MOVES fuel effects report.6
The density, energy and carbon content of the fuels in MOVES2014 are not based on fuel
compliance data; these are based on aggregate values that are constant across fuel types and fuel
subtypes as documented in the MOVES GHG and energy report.16,1
g The volToWtPercentOxy values are documented in Table 3-2 of the MOVES Report Speciation of Total Organic
Gas and Particulate Matter Emissions from On-road Vehicles in MOVES201412. MOVES2014a currently does not
use the volToWtPercentOxy values in the fuelFormulation table, but uses identical hardcoded values. Thus,
changing the VolToWtPercentOxy values in the fuelFormulation table will not change the results.
h The Cetanelndex and PAHContent fields in the fuelFormulation table are currently unused and are populated with
NULL values.
1 InMOVES2014a, we updated the energy content of biodiesel to 43.061 (KJ/g) instead of 41.738 (KJ/g) to reflect
B5 energy content, rather than B20.
14
-------
6 Renewable Fuels Market Share
The Annual Energy Outlook report (AEO) generated by the U.S. Energy Information
Administration (EIA) provides the basis for the volumes of biofuels (ethanol and biodiesel) used
in the MOVES2014 fuel supplies.1'2 The AEO report provides year-by-year projections for
biofuel energy consumption by fuel type for ethanol-gasoline blends (EO, E10 and E15), flexible-
fuel vehicle (FFV) blends (E70-E85), and biodiesel blends; currently through the year 2040. We
use these projections in conjunction with the overall fuel consumption numbers projected by the
MOVES model to calculate the marketshares of each of these biofuel types for use in the
MOVES2014 fuel supplies.
Generally, the AEO reports have only described biofuel energy consumption on a national basis.
However, as part of the Tier 3 rulemaking, EIA provided the agency a breakdown, by region, of
ethanol marketshares used in their modeling in conjunction with the AEO2013 final report.17
EPA has continued using this regional breakdown to more accurately identify the differences in
ethanol adoption rates between fuel regions. Currently, ethanol is the only biofuel with regional
variation reflected in the model. We are aware that state and local programs may also cause
significant variation in biodiesel distribution as well, and further updates to the regional
penetration of biofuels may be included in future versions of MOVES.
The fuel regions used in the AEO analysis are defined as part of their economic modeling, and
are not identical to the fuel regions created for MOVES2014, as explained in the previous
section. This necessitated the creation of a mapping scheme for our analysis. Fortunately, the
EIA regions are bounded on a state scale and can be easily allocated to county for our purposes.
Although the AEO reports form the basis of vehicle activity forecasts in the MOVES model,
vehicle energy use in MOVES2014 is not directly taken from the AEO reports; it is calculated as
part of the operation of the model. In order to apply the ethanol and biodiesel data from AEO, it
was necessary to reconcile the differences between the energy use found in those reports with the
energy use reported by the MOVES model. Using county VMT as a scaling factor, MOVES
vehicle energy consumption by fuel type was calculated for each county based on a national
MOVES run producing overall energy consumption. Then, by scaling AEO energy use to match
that generated by MOVES, we can find an appropriate adjustment factor that correctly applies
AEO biofuel volumes to the MOVES fuel supplies. This allows the MOVES model to retain the
correct marketshare of alternative fuels found in the AEO, despite having differing total energy
use (such that if MOVES were to generate exactly the same energy as AEO, the biofuel volumes
would also be identical). The adjustment factor used for estimating MOVES2014a energy
consumption is 0.956 of the AEO energy consumption.
After resolving the energy use between MOVES and AEO, we then calculated an additional
adjustment factor to the regional maximum El5 penetrations (matching the regional variation
reported by EIA in the Tier 3 report discussed above) in order to apply an appropriate E10 to E15
marketshare ratio in the default fuel supply. By using an adjustment factor for El5 penetration,
we preserved the relationship between regional penetrations while still applying the correct total
volume of ethanol to each county. The adjustment factor used for estimating MOVES2014a E15
penetration is 0.114 of the AEO 2014 projection.
15
-------
An example illustrating the phase-in of ethanol (both as E10 and E15, as well as FFV fuel) for a
specific fuel region can be found in Figure 5 below.
Figure 5 - E15 Phase-in as applied in MOVES2014a for a specific fuel region
r\ /ic
5 0.2
*—1
LU
o 0.15
(u
ao
(O
c 0.1
(U
u
&_
CD
a- 0.05
0
20
50
10 2015 2020 2025 2030 2035 2040 2045 20
Year
In MOVES2014a, we updated the fuel supply to reflect the AEO2014 final release2, and made it
consistent with other changes including: removing El5 from the California fuel regionf and new
fuel region assignments based on updates to local fuel programs'1. These changes caused the
marketshare of E15 and E10 to change for all regions. However, for a typical county, the
emissions impact of these changes are minimal, impacting VOC, CO, NOx and PM onroad
emissions by less than 1% between MOVES2014 and MOVES2014a.18
For diesel and biodiesel, we assume that conventional diesel constitutes 100% of the diesel
market share in all fuel regions for 1990 through 2013 calendar years. In 2014 and all later years,
we assume that biodiesel (fuelSubtypelD 21) has 100% of the diesel marketshare. The
MOVES2014 fuelFormulation table contain biodiesel blends with both 3.4, 5, and 20 percent
biodiesel, which a MOVES user can specify in their fuel supply for a county. However, for the
default fuel supply, we assume all biodiesel is B5 (5 percent biodiesel) in all fuel regions.1
16
-------
7 Nonroad Fuel Supply
In MOVES2014, the nonroad fuel supply was based on the fuels used in NMIM 2008, with fuel
supply spatially allocated by individual counties.19
However, in MOVES2014a, we updated the nonroad gasoline fuel supply to be consistent with
the onroad fuel supply, with exceptions made for nonroad fuel regulations. Specifically, ethanol
blends greater than 10 volume percent are not permitted to be used in nonroad gasoline
engines.20 To calculate the nonroad fuel supply, we replaced the onroad marketshare of ethanol
blends above E10 (E15, E85) with E10 use, by using the fuel wizard to convert the E15 fuel to
E10. Additionally, the spatial allocation of the nonroad fuel supply (gasoline, diesel, CNG, and
LPG) was changed from county to fuel region, consistent with the onroad fuel supply.
In the nonroad fuel supply, there are two types of diesel: nonroad diesel (fuelTypelD 23), and
marine diesel (fuelTypelD 24). The only difference in the fuel properties between nonroad diesel
and marine diesel is sulfur content. The diesel sulfur content for these fuels used in MOVES2014
and MOVES2014a is shown in Table 9 below. The sulfur content was updated in MOVES2014a
to be consistent with the sulfur levels used in NONROAD2008 (See Table 2 of the referenced
document)22
In MOVES2014, the sulfur level of nonroad CNG and LPG was 16 ppm for all years. However,
in MOVES2014a, the CNG and LPG sulfur levels were updated to be 7.6 ppm for all years,
consistent with the onroad CNG sulfur level as discussed in Section 5.
17
-------
Table 9. Nonroad diesel sulfur content in MOVES2014 and MOVES2014a.
MOVES2014
MOVES2014a
Year
Nonroad
Marine
Nonroad
Marine
1999 and earlier
2284
2640
2284
2640
2000
2284
2637
2284
2640
2001
2284
2637
2284
2635
2002
2284
2637
2284
2637
2003
2284
2637
2284
2637
2004
2284
2637
2284
2637
2005
2284
2637
2284
2637
2006
2284
2637
2242
2588
2007
351
435
1139
1332
2008
351
435
351
435
2009
351
435
351
435
2010
351
435
165
319
2011
32
124
32
236
2012
32
124
32
124
2013
32
124
32
44
2014
11
55
20
52
2015
11
55
11
56
2016
11
55
11
56
2017
11
55
11
56
2018+
11
55
11
55
As expected, there was a significant increase in the marketshare of E10 in the nonroad fuel
supply between MOVES2014 and MOVES2014a. For the years 2015-2030, the E10 marketshare
increased from 79% of the total nonroad gasoline consumption in MOVES2014 to 100% of the
nonroad gasoline consumption. With the fuel supply change in MOVES2014a, a typical urban
county in the years 2015-2030, nonroad emissions were reduced by 1.5%-4.4% for total
hydrocarbon emissions (THC), 2.4%-7.6% for carbon monoxide (CO) emissions, and increased
by 0.4-2.3%) for NOx emissions. There were negligible changes in PM emissions consistent with
small changes in sulfur levels between the two fuel supplies. These changes are consistent with
the nonroad fuel effects in MOVES2014, where increasing the gasoline oxygenate level
decreases THC and CO, decreases NOx emissions.21 In MOVES2014, the diesel fuel sulfur
impacts nonroad sulfate emissions, which contributes a small percentage of the PM2.5 exhaust
emissions.22
MOVES2014a added the capability to calculate speciated organic emissions, including volatile
organic compound (VOC) emissions from nonroad THC emissions. VOC emissions are
calculated as a ratio to non-methane hydrocarbon emissions (NMHC) emissions. For gasoline
nonroad engines, the VOC/NMHC ratio is a function of the ethanol level as documented in the
Nonroad Speciation Report.23 Thus, changes in the nonroad fuel supply in MOVES2014a also
impact the speciated organic gaseous emissions from nonroad gasoline equipment.
18
-------
8 Fuel Wizard
In order to more easily facilitate the analysis of potential fuel control programs for state or local
programs, EPA has created a new tool in MOVES2014a designed to more easily create fuels that
are not represented by the default onroad fuel supply. Changes to fuel properties do not happen
independently; by changing a single property such as sulfur level or RVP, other fuel properties
change as well, such as aromatics or the distillation properties (T50, T90, etc). The fuel wizard
enables the end users to take advantage of the refinery modeling done by EPA as part of the Tier
3 rulemaking, capturing these secondary fuel property changes in a way that does not require
significant effort outside of the MOVES model. This allows for the full impact of proposed fuel
changes (as part of state or local programs) to be taken into account, including the subsequent
effects of non-regulated fuel property changes on emissions.
The adjustment factors used in the fuel wizard are the same as those used in the creation of the
default fuel supply (see Table 4, Table 5, and Table 6). The fuel wizard contains adjustment
factors for the three properties we believe are the most commonly analyzed for state and local
programs: ethanol, sulfur, and RVP. The fuel wizard is currently capable of creating fuels with
ethanol variations between EO - El5, sulfur from 5 ppm to 80 ppm, and RVP from 5 psi to 14
psi.
As part of the MOVES2014a update from MOVES2014, we corrected the fuel wizard to repair
several errors (primarily in the fuel adjustments for RVP and sulfur) that led to erroneous fuel
property adjustments.24'27 These errors only existed in the fuel wizard tool provided for MOVES
users, and did not impact the default fuel properties in the MOVES fuelFormulation table.
In MOVES2014a, we also updated the fuel wizard to not allow users to enter ethanol levels
greater than El 5. We do not recommend the use of the fuel wizard adjustment factors for ethanol
levels greater than El 5 because any results above El 5 are extrapolated, due to constraints in the
EPA refinery modeling and because the MOVES2014 fuel effects model is not capable of
modeling 'mid-level' ethanol blends (blends between El5 and E70).
The fuel wizard is used in conjunction with the county data manager in the MOVES graphical
user interface (GUI). Guidance on when users should use the fuel wizard is provided in our
technical guidance.25 After selecting a fuel contained in the default database that most closely
matches the fuel to be analyzed, the end user then invokes the fuel wizard to complete the
desired changes in fuel properties to the selected fuel. Please note that if multiple fuel property
changes are desired (i.e., a change in both sulfur and RVP level), it is possible that some
secondary fuel properties may be affected by multiple adjustment factors. Therefore, it is
suggested that the fuel wizard be used by changing properties in order of least to most significant
for the desired analysis.
We have also released a patch to MOVES2014a, the "MOVES2014a November 2016 patch",
which included an additional fix to the fuel wizard. The fuel wizard in MOVES2014a correctly
calculated the changes in fuel properties when the ethanol volume of a fuel was changed from EO
to E10, or from E10 to EO for an area with the 1 psi RVP waiver. However, the fuel wizard did
not produce correct fuel properties when a user in the 1 psi RVP waiver area changed the ethanol
19
-------
volume of a fuel from E10 to E15, or from E15 to E10. The November 2016 patch to the fuel
wizard corrected this issue.
For areas without the 1 psi waiver, the fuel wizard in MOVES2014a produced inaccurate fuel
properties when the ethanol volume of a fuel was changed from E0 to E10, E10 to E0, E10 to
E15, and E15 to E10. The step-by-step instructions for avoiding this bug are provided in the User
Interface Reference Manual.26 We plan to fix this bug permanently in the next version of
MOVES.
The emissions impacts of the fuel wizard bug in MOVES2014a without the patch is summarized
in Table 10.
Table 10. Percent differences in emissions between the correct fuel based on refinery modeling and the
incorrect fuel predicted by MOVES2014a fuel wizard for two sample counties in July 2015t
Ethanol
Content
Areas with 1
psi RVP wavier
Areas without 1 psi RVP wavier
E0 to E10
E10 to E0
E10 to E15
E15 to E10
E0 to E10
E10 to E0
E10 to E15
E15 to E10
NOx
0%
0%
-0.6%
0.5%
0.2%
-0.2%
0.1%
-0.3%
VOC
0%
0%
-3.2%
2.9%
-0.2%
0.4%
2.6%
-3.2%
pm25
0%
0%
-0.5%
0.5%
-0.3%
0.3%
0.6%
-0.8%
CO
0%
0%
-1.7%
1.9%
0.7%
0%
1.1%
-1.3%
"i" The results may vary by region, calendar year, and month.
As new refinery modeling information is generated (either as part of agency rulemakings or from
external modeling efforts1"13 27), we plan on expanding the functionality of the fuel wizard to both
incorporate a wider selection of properties that can be adjusted, as well as capture more complex
fuel property interactions
J For example, we are evaluating data shared to us on the variable impact of ethanol on T50 provided in the CRC
Review of MOVES2014.1327
20
-------
9 References
1 US Energy Information Administration. Annual Energy Outlook 2014 (Early Release). DOE/EIA-0383ER (2014).
December 2013.
2 US Energy Information Administration. Annual Energy Outlook 2014. DOE/EIA-0383(2014). April 2014.
3 USEPA Office of Transportation and Air Quality. Control of Air Pollution from Motor Vehicles: Tier 3 Motor
Vehicle Emission and Fuel Standards; Final Rule. Federal Register. Vol. 79, No. 81. April 28, 2014.
4 US EPA. Development of Gasoline Fuel Effects in the Motor Vehicle Emissions Simulator (MOVES2009). EPA-
420-P-09-004 August 2009.
5 US EPA. Control of Air Pollution from Motor Vehicles: Tier 3 Motor Vehicle Emission and Fuel Standards Final
Rule Regulatory Impact Analysis. Chapter 7.1 Impacts of the Rule on Emissions and Air Quality-Criteria and Toxic
Pollutant Emission Impacts. EPA-420-R-14-005. March 2014.
6 USEPA (2016). Fuel Effects on Exhaust Emissions from On-road Vehicles in MOVES2014. EPA-420-R-16-001.
Assessment and Standards Division. Office of Transportation and Air Quality. US Environmental Protection
Agency. Ann Arbor, MI. 2016. https://www.epa.gov/moves/moves-technical-reports.
7 OPIS/STALSBY Petroleum Terminal Encyclopedia. OPIS/STALSBY Wall, NJ: 2013.
8 US Energy Information Administration. US Energy Mapping System, http://www.eia.gov/state/maps.cfm. 2016.
9 USEPA (2015). MOVES2014a Module Reference. Office of Transportation and Air Quality. US Environmental
Protection Agency. Ann Arbor, MI. October, 2015. https://www.epa.gov/moves/moves2014a-latest-version-motor-
vehicle-emission-simulator-moves.
10 USEPA (2016). Gasoline Reid Vapor Pressure, https://www.epa.gov/gasoline-standards/gasoline-reid-vapor-
pressure.
11 Alliance of Automobile Manufacturers. Alliance North American Fuel Survey. January 2015.
12 US EPA. Conventional Gasoline Parameters by Reporting Year (1997-2005) and Reformulated Gasoline
Parameters by Reporting Year, https://www3.epa.gov/otaa/fuels 1/rfgreport.htm. February 2016.
13 USEPA (2016). U.S. EPA Response to CRC Project No. E-101, Review of EPA's MOVES2014 Model. 420-R-16-
012. Office of Transportation and Air Quality. US Environmental Protection Agency. Ann Arbor, MI. October 2016.
https://www.epa.gov/moves/moves-technical-reports.
14 USEPA (2015). Speciation of Total Organic Gas and Particulate Matter Emissions from On-road Vehicles in
MOVES2014. EPA-420-R-15-022. Office of Transportation and Air Quality. US Environmental Protection Agency.
Ann Arbor, MI. October 2014. https://www.epa.gov/moves/moves-technical-reports.
15 USEPA (2016). Heavy-Duty Highway Compression-Ignition Engines and Urban Buses: Exhaust Emission
Standards. EPA-420-B-16-018. Office of Transportation and Air Quality. US Environmental Protection Agency.
Ann Arbor, MI. March 2016. https://nepis.epa.gov/Exe/ZyPDF.cgi?Dockey=P 10009ZZ.pdf
16 USEPA (2015). Greenhouse Gas and Energy Consumption Rates for On-road Vehicles: Updates for
MOVES2014. EPA-420-R-15-003. Assessment and Standards Division. Office of Transportation and Air Quality.
US Environmental Protection Agency. Ann Arbor, MI. October, 2015. https://www.epa.gov/moves/moves-
technical-reports.
17 US Energy Information Administration. Annual Energy Outlook 2013. DOE/EIA-0383(2013). July 2013.
21
-------
18 Sonntag, D., "Impact of MOVES2014a and Plans for the Next Version of MOVES," 26th Coordinating Research
Council On-Road Vehicle Emissions Workshop (presentation), Newport Beach, CA: 13-16 March 2016,
http://www.crcao.org/workshops/index.html.
19 USEPA (2005). EPA's National Inventory Model (NMIM), A Consolidated Emissions Modeling System for
MOBILE6 andNONROAD. 420-R-05-024. Assessment and Standards Division. Office of Transportation and Air
Quality. US Environmental Protection Agency. Ann Arbor, MI. December, 2005.
https://www.epa.gov/otaq/nmim.htm.
20USEPA (2011). Fact Sheet: EPA Announces El 5 partial Waiver Decision. EPA-420-F-11-003. Office of
Transportation and Air Quality. US Environmental Protection Agency. Ann Arbor, MI. January, 2011.
https://www.epa.gov/gasoline-standards/el5-fuel-partial-waivers.
21 USEPA (2005). Exhaust Emission Effects of Fuel Sulfur and Oxygen on Gasoline Nonroad Engines. NR-
003cEPA-420-R-05-016. Office of Transportation and Air Quality. US Environmental Protection Agency. Ann
Arbor, MI. December, 2005. http://www3.epa.gov/otaq/nonrdmdl.htm.
22 USEPA (2010). Exhaust and Crankcase Emission Factors for Nonroad Engine Modeling -Compression-Ignition.
NR-009dEPA-420-R-10-018. Office of Transportation and Air Quality. US Environmental Protection Agency. Ann
Arbor, MI. July, 2010. http://www3.epa.gov/otaq/nonrdmdl.htm.
23 USEPA (2015). Speciation Profiles and Toxic Emission Factors for Nonroad Engines. EPA-420-R-15-019.
Office of Transportation and Air Quality. US Environmental Protection Agency. Ann Arbor, MI. November 2015.
https://clpub. epa.gov/si/si_public_record_report. cfm?dirEntryId=309339.
24 USEPA (2015). MOVES2014a Questions and Answers. EPA-420-F-15-046. Office of Transportation and Air
Quality. US Environmental Protection Agency. Ann Arbor, MI. November, 2015.
https://www.epa.gov/moves/moves2014a-latest-version-motor-vehicle-emission-simulator-moves.
25 USEPA (2015). MOVES2014 andMOVES2014a Technical Guidance: Using MOVES to Prepare Emission
Inventories for State Implementation Plans and Transportation Conformity. 420-B-15-093. Assessment and
Standards Division. Office of Transportation and Air Quality. US Environmental Protection Agency. Ann Arbor,
MI. November, 2015. https://www.epa.gov/moves/moves2014a-latest-version-motor-vehicle-emission-simulator-
moves.
26 USEPA (2016). MOVES2014a User Interface Manual. EPA-420-B-16-085. Office of Transportation and Air
Quality. US Environmental Protection Agency. Ann Arbor, MI. November, 2016.
https://www.epa.gov/moves/moves2014a-latest-version-motor-vehicle-emission-simulator-moves.
27 Heiken, J. G., M. Hixsonand J. Lyons (2016). Review ofEPA'sMOVES2014Model. CRC Report No. E-101.
August 11, 2016. https://crcao.org/reports/recentstudies2016/E-101/FINAL%20E101%20Report%20SR-
20160810%20w%20CRC%20Cover%20and%20 Appendices.pdf.
22
------- |